August 16, 2007 
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31
August 16, 2007 [00:00:10] *** derchris has joined #opensolaris [00:00:48] <jbk> evening [00:01:03] *** triplah_ has joined #opensolaris [00:01:36] <cyborg--> so...i am going to test this, but as a compromise for now, i am hoping that i can replace a 250GB device with a 500GB device but have 250GB more in the zpool [00:01:45] <CIA-17> jg: PSARC 2007/423 remdrv -C, 6525121 rem_drv of nexus driver should not revert /devices permissions below [00:03:28] *** sommerfeld has quit IRC [00:07:14] <FrostCS> e^ipi, you might try a 1024-9C02-00-SA or ATC-P2KX-10 [00:14:15] *** Megaf has joined #OpenSolaris [00:15:13] *** msmiffy has joined #opensolaris [00:15:21] *** msmiffy has left #opensolaris [00:18:56] *** Mdx4 has joined #opensolaris [00:21:15] *** sparc-kly has joined #opensolaris [00:21:16] *** ChanServ sets mode: +o sparc-kly [00:21:31] *** laca is now known as lacaAFK [00:22:35] *** onlinebacon has joined #opensolaris [00:25:28] *** purserj[sol] has quit IRC [00:27:31] *** Me[-_-]gaf has quit IRC [00:29:58] *** sparc-kly has quit IRC [00:30:27] *** sparc-kly has joined #opensolaris [00:30:28] *** ChanServ sets mode: +o sparc-kly [00:31:59] <delewis> yay, AMD unveils today what SPARC has had for years (hardware counters for cache misses).. http://hardware.slashdot.org/article.pl?sid=07/08/15/1925222&from=rss [00:33:16] *** Lurken1 has quit IRC [00:33:17] *** sstallion has quit IRC [00:35:29] <brendang> delewis: hoo-ray [00:35:34] <jbk> and that deserves an article? :) [00:35:46] <brendang> delewis: and do they have stall-cycles for cache misses too? :) [00:36:09] <jbk> 'omg, my pc can now branch too' [00:36:56] *** sstallion has joined #opensolaris [00:37:57] <jbk> cool.. got a mercurial repo setup for libdisasm... after cleaning things up a bit later tonight, i'll push what i currently have and see what people think.. [00:38:12] <jmcp> jbk: cool. please let me know [00:38:58] *** sparc-kly has quit IRC [00:39:09] *** Megaf has quit IRC [00:39:21] *** sparc-kly has joined #opensolaris [00:39:22] *** ChanServ sets mode: +o sparc-kly [00:41:32] <tsp> finally! got my sparc to suspend [00:41:50] <tsp> or at least it should [00:42:46] *** pramz has left #opensolaris [00:44:03] <trede> where can i read about ETA for next SXCE ? [00:44:26] <FrostCS> Christmas, think Christmas [00:44:27] *** sstallion has quit IRC [00:44:32] <delewis> here's at Intel has (re: hardware counters): http://oprofile.sourceforge.net/docs/intel-P4-events.php3 [00:44:35] <jmcp> http://www.opensolaris.org/os/community/onnv/onnv_schedule.txt, for a guesstimate [00:44:42] *** sstallion has joined #opensolaris [00:44:51] <delewis> doesn't look remotely impressive compared to the number of counters an UltraSPARC-III has. [00:45:41] <trede> jmcp: thanks but those dates are overdue since long [00:46:32] <movement> jbk: awesome... [00:46:58] <jmcp> trede: best I can do, sorry [00:47:39] *** triplah_ has quit IRC [00:48:12] *** jpdrawneek has joined #opensolaris [00:48:45] <trede> jmcp: thanks for trying to help. [00:51:27] *** sparc-kly has quit IRC [00:51:55] *** sparc-kly has joined #opensolaris [00:51:56] *** ChanServ sets mode: +o sparc-kly [00:54:50] *** yippi has joined #opensolaris [00:56:02] *** _hsilva is now known as hsilva [00:56:19] <reflect> hm.. on sparc, when I wanted to copy a disk layout, I just went into format, named a layout to something, changed disk and then used 'select' to apply it [00:56:48] <reflect> however, when I try it now, it seems to forget the named table when I change disks? [00:57:39] <axisys> we are trying to save money on power in our organization.. we are unix shop.. mostly solaris(sparc,x86), freebsd(sun x86) and very little linux.. we need a general idea for future purchase of CPU/Mem and Storage.. [00:58:08] *** Tpenta has joined #opensolaris [00:58:10] <movement> delewis: AMD has had them for years too [00:58:24] <movement> delewis: this is something on top (that SPARC doesn't AFAIK have at all) [00:58:25] <axisys> so going with T2k and X4500 should be our main focus? [00:58:34] *** sstallion has quit IRC [00:58:53] <movement> LWP is actually relatively interesting. [00:59:40] *** sstallion has joined #opensolaris [01:00:10] *** ChanServ sets mode: +o Tpenta [01:00:12] <seanmcg> reflect: prtvtoc and fmthard can be used together. [01:00:19] <CIA-17> jasmith: 6216103 assertion failed: curdata->ex_seccnt > 0 seen again [01:00:46] <reflect> seanmcg: thanks for the alternative method [01:00:55] <reflect> but, is this behaviour normal? [01:02:07] <seanmcg> unsure right now, haven't used format like that in years. Always used prtvtoc/fmthard :) Though if it disagrees with the manpage its probably a bug. [01:05:00] <reflect> was a long time ago since I used format myself, I'm gonna go read up on how it should work [01:05:10] *** EchoBinary has quit IRC [01:05:55] <axisys> Is there some url where I can get some guidance on how to pick a system that perform best for the price based on the functionality of a system? I am trying to get a basic idea if for example to run 4 zones with one of them running oracle should I go with a X4200 or T2k ? [01:06:35] *** sommerfeld has joined #opensolaris [01:06:36] *** ChanServ sets mode: +o sommerfeld [01:06:37] <tsoome> 4200 is pc:P [01:06:58] <tsoome> that means crappy hw [01:07:09] <jmcp> tsoome: bs [01:07:15] <tsoome> yea? [01:07:18] <jmcp> yeah [01:07:35] <jmcp> choose the right hw to do the task which you need done. [01:07:42] <tsoome> well, then why the hell upgradinf management card fw will down the system? [01:07:52] * jmcp shrugs [01:08:19] <jmcp> did you read the release notes? [01:08:27] <tsoome> ofc [01:08:44] <jmcp> having spent *many* years supporting customers, I never make that assumption [01:08:45] <axisys> jmcp: so what is the right hw for the job? is there some cheat sheet I can go by? [01:08:56] <axisys> jmcp: i have few hundred servers to choose from [01:08:59] <delewis> SPARC all around. [01:09:08] <axisys> delewis: i like that.. [01:09:26] <jmcp> axisys: you have to figure that out for yourself. figure out what high level requirements you have, and then do pros/cons on the various options which fit those requirements [01:09:28] <axisys> delewis: is it because of the obp or something else? [01:09:31] <jmcp> geez, it's not like it's rocket science [01:09:59] <tsoome> axisys: what job this DB will do? if its very GHz bound.... [01:10:04] <axisys> jmcp: hehe.. i am getting ready to fig out.. just wanted to see if there is some cheat sheet to get a start [01:10:05] *** vmlemon has quit IRC [01:10:12] <jmcp> axisys: no, no cheat sheet. [01:10:25] *** yippi has quit IRC [01:10:30] <jmcp> unless your org already has stuff like that (and a lot do), then you'll have to go it alone [01:10:43] *** leal has joined #opensolaris [01:10:46] <tsoome> our services/outs department is using T2k for oracle in zones quite happily [01:10:47] <axisys> jmcp: alone in that :P [01:10:48] <leal> hello.. [01:10:53] <jmcp> hi leal [01:11:00] <axisys> tsoome: sounds good [01:11:18] <jmcp> axisys: so what is the front end for your oracle instance going to be doing? [01:11:28] <tsoome> but it really depends how well the actual sql is written... [01:11:31] <jmcp> do you have minimum requirements for connections/sec? query time? [01:11:34] <leal> somebody knows about a "link" that explains threads and lwp solaris vs. linux? [01:11:41] *** nrubsig has joined #opensolaris [01:11:41] *** ChanServ sets mode: +o nrubsig [01:11:53] <tsoome> leal what solaris? [01:12:04] <nrubsig> Greetings, monsters, chicken, terrorists and Solaris users... [01:12:05] <tsoome> s10? [01:12:08] <leal> yes [01:12:24] *** bengtf has quit IRC [01:12:34] <tsoome> in s10, thread and lwp is in 1-1 relation [01:12:42] <jmcp> leal: purchase "Solaris Internals, 2nd edition" [01:12:52] <leal> i wish to understand why some apps "are" multi-threaded but dont use multi-core systems for example... [01:13:02] <tsoome> ? [01:13:09] <nrubsig> leal: ?! [01:13:10] <jmcp> leal: example, please [01:13:36] <leal> jmcp: that's for sure... but i can't find it here in Brazil... i will need to buy from amazon.. :) [01:13:39] <nrubsig> leal: do you mean applicatoins which use |setcontext()|&co. ? [01:13:46] <tsoome> this can only happen if multicore is not supported at all, or if kernel has no idea about threads in app [01:13:57] <jmcp> leal: or prentice hall, direct. either way, it's a very handy book to have [01:14:02] *** dduvall has quit IRC [01:14:10] <leal> sometime ago we have some problems with zope for example... [01:14:13] <nrubsig> dmarker: ping! [01:14:18] <jmcp> leal: this top level url might be handy for you http://developers.sun.com/solaris/learning/new2solaris/ [01:14:29] *** sparc-kly has quit IRC [01:14:39] <jmcp> leal: and http://www.solarisinternals.com/wiki/index.php/Solaris_Internals_and_Performance_FAQ [01:15:01] <leal> we have 10 threads running but because of a "issue" with zope/python, all the threads were running in one processor. [01:15:31] *** sparc-kly has joined #opensolaris [01:15:32] *** ChanServ sets mode: +o sparc-kly [01:16:02] <leal> I did read a article about apache and multi-core systems, and the performance in a mono processor was the same (actually, better in mono) [01:16:55] <leal> i did read something about kernel|user threads.... too. [01:17:38] <tsoome> how you know they are running in one cpu? [01:18:01] <dmarker> nrubsig: 'sup? [01:18:18] <leal> tsoome: i got a totally idle processor. [01:18:36] <tsoome> thats in solaris? [01:18:51] <leal> tsoome: no... sorry.. linux. [01:19:00] <tsoome> well.... [01:19:19] <leal> tsoome: that's why i want to know the differences... [01:19:25] <tsoome> suppose there is nothing more to add:D [01:19:38] <tsoome> run it in solaris;) [01:19:52] <leal> but i did a search and the problem was not linux, but "how" python/zope use threads. [01:20:17] <tsoome> no, it depends on thread model in use [01:21:31] <tsoome> maybe you can swith the thread lib.... but i have really no idea what alternatives there are in linux [01:22:26] <tsoome> but your case sounds like the kernel has no idea there are threads in app [01:23:17] <leal> maybe, but i need to read something about it. :) [01:23:21] <tsoome> or perhaps userland threads are deliberately bound to just one cpu or kernel "thread" [01:23:38] *** NikolaVeber has quit IRC [01:23:39] <leal> thanks anyway, i will look the links that you point me. [01:24:51] <leal> i did remember that in freebsd, the threads are showed like just one process... and in linux i have saw like pid's.... [01:25:13] <leal> i think that information was fake... and i want to know how it is in solaris. [01:25:25] <tsoome> well yes, thats the one way to make them visible:) [01:25:53] <sommerfeld> solaris 10 binds each userspace thread to a kernel lwp; multiple lwp's sharing a userspace address space make up a process. [01:26:17] <leal> tsoome: visible, but they will be bound to just one cpu. :) [01:26:31] <sommerfeld> prstat -L shows each lwp separately [01:26:46] <tsoome> older solaris systems used to use n-m model, but it proved to be overcomplicated [01:26:53] <sommerfeld> and multiple lwps of a single process can run simultaneously on multiple cpus [01:27:10] <sommerfeld> if you pbind a process, all threads are bound to a single cpu and won't run concurrently [01:27:45] <sommerfeld> pbind can bind individual lwps to individual cpus [01:28:41] <leal> i think that some softwares that say they are multi-threaded... are not. [01:29:49] <tsoome> depends, ofc you can write MT app, where only minor functionality is implemented in parallel, and major part is still singlethreaded [01:30:05] <sommerfeld> there's thread-safe (creates multiple threads) and then there's thread-hot (multiple threads can effectively use multiple cpus) [01:30:16] <leal> i guess make a program really multi-thread is not simple... so, some apps just use some "features" of mt... [01:30:41] <leal> tsoome: something like that. :) [01:31:43] <leal> sommerfeld: so, thread-safe apps just creates multiple threads but bound to a single cpu? [01:32:10] *** mritun_ has joined #opensolaris [01:32:34] <leal> is that a definition? [01:33:32] <mritun_> any dtrace experts around ? [01:33:35] <FrostCS> hrm, if my expenses are correct, I spent more in the last month being jobless, then I did in the previous year of employment. [01:33:43] <sommerfeld> not necessarily bound to single cpu, but only having at most one runnable thread at a time.. [01:33:54] *** pmh has quit IRC [01:34:06] *** bengtf__ has joined #opensolaris [01:34:07] *** bengtf__ is now known as bengtf [01:34:19] <mritun_> CDE is hung after login (perpetual hourglass) [01:34:28] <mritun_> no CPU usage and no IO [01:34:38] <FrostCS> easy fix, don't use CDE. [01:35:05] <mritun_> hmm, considered, but GNOME is a pig with lipstick [01:35:21] <FrostCS> CLI is where it's at. [01:35:24] <mritun_> especially on this VM with 512MB ram [01:35:50] <mritun_> looking at some way to diagnose what is stuck and where [01:35:53] <leal> sommerfeld: ok, so that's the diference: thread-safe and thread-hot? [01:35:59] <mritun_> (which is the fun part) [01:36:10] <nrubsig> mritun_: check whether $HOME is writeable or whether ~/.tt* files are corrupted [01:36:50] <nrubsig> mritun_: I mean ~/.TT* like ~/.TTauthority [01:37:11] <nrubsig> mritun_: and check /var/adm/messages [01:37:39] <mritun_> hmm, .TTauthority seems to be ok... [01:37:55] <mritun_> though it contains the last DHCP IP in the magic cookie [01:37:59] <mritun_> should be ok, right ? [01:38:13] <nrubsig> mritun_: right. [01:38:25] <mritun_> there are some binary chars also [01:38:41] <nrubsig> mritun_: but deleting ~/.TTauthority may help if the cookie is corrupted or invalid for some reason [01:39:03] <nrubsig> mritun_: ~/.TTauthority is like ~/.Xauthority [01:39:05] <FrostCS> is this Solaris 10 or SX? [01:39:06] <mritun_> yeps... just nuked it [01:39:18] <mritun_> status-quo [01:39:21] <mritun_> :( [01:39:30] <mritun_> SX [01:39:31] <nrubsig> mritun_: check whether $HOME is writeable by the intended user [01:39:32] <mritun_> b60 [01:39:39] <mritun_> yeps, I'm root [01:39:44] <nrubsig> mritun_: and check /var/adm/messages [01:39:47] <nrubsig> erm [01:39:50] <mritun_> mode 700 [01:40:01] <nrubsig> you use "root" to login into a GUI ? [01:40:13] <mritun_> aah [01:40:14] <nrubsig> oh dear... ;-( [01:40:26] *** et_alia[1fk] has quit IRC [01:40:27] <mritun_> don't worry, $HOME -> /root [01:40:34] <nrubsig> Oh [01:40:35] <nrubsig> ok [01:40:37] <nrubsig> :-) [01:40:37] <mritun_> ah there we go [01:40:42] <mritun_> copy pasting [01:40:48] <nrubsig> mritun_: what happened ? [01:41:04] <cyborg--> this is becoming very frustrating trying to boost the MTU on e1000g1 [01:41:08] <mritun_> Aug 16 05:08:38 flare dtsession[752]: [ID 907239 user.error] libtt[752]: _Tt_stream_socket::accept(): t_listen(): System error: Interrupted system call [01:41:09] <mritun_> Aug 16 05:08:38 flare dtsession[752]: [ID 907239 user.error] libtt[752]: _Tt_stream_socket::accept(): t_listen(): System error: Interrupted system call [01:41:20] <cyborg--> i modified the maxframesize in /kernel/drv/e1000g.conf but no luck... [01:42:00] *** hohum has quit IRC [01:42:06] <mritun_> looks like it's failing to listen on some socket [01:42:19] <Tempt> Mornin'. [01:42:23] *** nachox has joined #opensolaris [01:42:53] <nrubsig> mritun_: try http://svn.genunix.org/repos/on/branches/ksh93/gisburn/scripts/shnote.ksh next time... :-) [01:43:07] <nachox> evening [01:43:38] <mritun_> nrubsig: hehe... I saw it yesterday itself ;) neato [01:43:43] <seanmcg> g'night [01:44:13] <nrubsig> mritun_: unfortunately it needs ksh93... ;-/ [01:44:14] <Tempt> Hey, Java update on my windows box at work is trying to sell me google toolbar. [01:44:15] <mritun_> hmm, localhost entry was pointing to ipv6 interface [01:44:16] <FrostCS> cyborg-- ifconfig command isn't working? [01:44:30] <mritun_> nrubsig: yeps... too bad it isn't in SX yet [01:44:37] <FrostCS> turn ipv6 off, that's an old bug [01:44:54] <mritun_> ah [01:45:08] <FrostCS> cde won't load with ipv6 enabled [01:45:09] <mritun_> now I remember... I was playing around with IPv6... [01:45:12] <FrostCS> and it won't be fixed [01:45:15] <mritun_> and then that happened [01:45:20] <FrostCS> that would be why [01:45:28] *** leal has quit IRC [01:45:42] <FrostCS> cde is dying off.. so you might want to also get to know xfce or another window manager. [01:45:46] <nachox> Tempt, part of the google pack deal perhaps? :P [01:45:50] <mritun_> well, it does work with ipv6 enabled [01:46:16] <FrostCS> it will work with it enabled, but won't load with it enabled. [01:46:17] <mritun_> which is enabled by installer [01:46:38] <alanc> huh? I've used CDE with IPv6 enabled on many machines [01:46:43] <FrostCS> I am not sure if it's a definate problem, but it will certainly reoccur. [01:46:44] <mritun_> FrostCS: want to try out xfce actually ;) [01:46:49] <mritun_> neat WM... [01:47:01] <cyborg--> well, ifconfig e1000g1 mtu 9000 doesnt work [01:47:10] <cyborg--> i tried the other set values mentioned for the driver too [01:47:20] <FrostCS> because your missing something cyborg [01:47:36] <FrostCS> ifconfig e1000g1 mtu 9000 up [01:47:42] <mritun_> rebooting... [01:47:45] <tsoome> lol [01:47:49] <mritun_> loghost was also borked [01:47:57] <FrostCS> alanc, I can dig up the old bug if you want [01:48:23] *** kito has quit IRC [01:48:29] <alanc> I won't fix it even if you find it 8-) [01:48:57] <mritun_> ah [01:49:00] <mritun_> there it is... [01:49:03] <FrostCS> I know, that's why I was persuading him against cde. :-) [01:49:08] <mritun_> CDE in all its glory :D [01:49:16] <mritun_> Thanks FrostCS :) [01:49:34] <mritun_> Thanks nrubsig [01:50:27] <FrostCS> alanc, http://blogs.sun.com/gfaden/ see top entry [01:50:38] <alanc> it's been almost 7.5 years since I officially worked on CDE...though I've never been able to fully escape (damn you marketing and their pretty login screen desires!) [01:50:50] <cyborg--> FrostCS: ifconfig: setifmtu: SIOCSLIFMTU: e1000g1: Invalid argument [01:50:53] <bubbva> I love CDE! [01:51:19] *** dduvall has joined #opensolaris [01:51:36] <alanc> oh, I don't use Trusted much, so I wouldn't have noticed the trusted problem [01:52:12] <alanc> hmm, the TX desktop regression I caused isn't listed there... [01:52:17] <FrostCS> cyborg--, I'll try to dig it up.. SX right? [01:52:28] <cyborg--> this machine is sxce [01:52:33] <mritun_> alanc: Thanks for maintaining CDE for us lusers ;) [01:52:36] <FrostCS> alanc, better update the regression listing then! [01:52:43] <cyborg--> i am able to set mtu 1500 no error anything over 1500 fails [01:52:57] <cyborg--> i'ved edited the /kernel/drv/e1000g.conf to specify maxframe up to 16k [01:54:27] <mritun_> alanc: CDE is the most productive desktop/WM that I've ever come across [01:54:39] <alanc> though I'm partially responsible for the MIT-SHM/zone regression - I introduced the MIT-SHM code a couple years ago that does the check that got broken by the recent zone change [01:55:13] <seanmcg> [v]twm... [01:55:39] <FrostCS> oh, cyborg-- try unplumbing the interface first [01:56:30] <nachox> e is nice too when it comes to productivity [01:56:44] <cyborg--> so while the interface is umplumbed i cannot use ifconfig right? [01:56:59] <FrostCS> you can [01:57:18] <FrostCS> you bring it up with the same command i posted earlier [01:57:36] *** freakazoid0223 has quit IRC [01:57:43] <cyborg--> ifconfig: setifmtu: SIOCSLIFMTU: e1000g1: no such interface [01:57:57] <mritun_> ifconfig e1000g plumb [01:58:19] <cyborg--> yes i am aware plumb will bring it back, but frostcs was saying unplumb it first [01:59:54] *** theRealballchalk has joined #opensolaris [02:00:43] *** freakazoid0223 has joined #opensolaris [02:00:46] <bubbva> mritun_ I concur. CDE just works. It doesn't suck up all my memory doing nonsense, and doesn't randomly lose my config files & prevent me from logging on (though I've been told Gnome doesn't do that anymore) [02:00:50] <FrostCS> have you done an update_drv or rebooted since you changed the maxframesize cyborg--? [02:01:22] *** xsh has quit IRC [02:02:05] <alanc> hmm, forgot to include the Virtual Console ARC case in today's open agenda [02:02:30] <Tempt> alanc: You know, you could develop a new pretty login screen. [02:02:35] <alanc> oh well, it's approved now [02:02:39] <Tempt> alanc: With fancy features and compiz style eyecandy. [02:02:42] <cyborg--> I have rebooted yes [02:02:57] <FrostCS> cyborg--, and default_mtu in nge.conf is set to 9000? [02:03:08] <elektronkind> we need CDE+compiz [02:03:25] <sommerfeld> alanc: was marked as "closed" apparently by default/mistake [02:03:29] <alanc> Tempt: been there, done that, twice now (though xdm changes were much smaller scale than dtlogin) [02:04:12] <alanc> yeah [02:04:13] <cyborg--> FrostCS: nge.conf does not contain default_mtu , i can put it in, but this may result in the server becoming an island? [02:04:14] <Tempt> Naah, something really flashy. [02:04:28] <alanc> thats what gdm is for [02:04:52] <alanc> tell yippi he needs to put compiz support into gdm next time he's around [02:04:54] <FrostCS> cyborg--, if you ndd - get /dev/e1000g1 max_frame_size you should be able to make sure max frames is set properly atleast.. (after it's replumbed) [02:05:00] <elektronkind> not even the macos x login screen is flashy [02:05:29] <alanc> if you mistype your password, it will melt the screen down [02:05:34] <gdamore> FrostCS: hopefully one of these days brussels will make configuring jumbo frames sane... so most NICs will use the same method. [02:05:59] <FrostCS> gdamore, do you know what I am missing? I am sure you'd be better help then I would in this subject :-) [02:06:01] <gdamore> alanc: better the screen melting down than the user melting down.... [02:06:10] <Tempt> I want to see a MovieOS style login. Numbers streaming by, flashy text, flashy effects when password is incorrect ... [02:06:11] <gdamore> i haven't followed.... [02:06:22] <alanc> ooh, should have made VT switching rotate the cube like compiz....damn, too late to TCR their ARC case for that [02:06:24] <elektronkind> god I hope so. Right now it's a trip through wonderland just to figure out if one's particular version of bge or e1000g supports jumbo grames [02:06:32] <FrostCS> cyborg-- however, I doubt setting default_mtu property would hurt it :-) [02:06:52] <gdamore> generally you have to *reboot* (or unload and reload the driver) after changing the maximum MTUs. [02:07:11] <FrostCS> gdamore, he is trying to set up mtu for his e1000g, and is getting an error, and yes, he said he has rebooted after changing the maximum mtu [02:07:14] <gdamore> most drivers don't support tweaking it other than at start-of-day. kind of lame... [02:07:15] <alanc> though having an ARC case require eye candy would probably end up in the ARC melting down [02:07:55] <Tempt> Heh. [02:08:25] <gdamore> nah, just having joerg attend ARC would have the same result. [02:08:28] <FrostCS> gdamore, he is getting a type of that "ifconfig: setifmtu: SIOCSLIFMTU" error... [02:08:34] <cyborg--> ndd says max frame size is 1514 [02:08:48] <alanc> add -/ flags to every utility that there exists a schily version of! [02:09:00] <FrostCS> ah, then it looks as if the max_frame hasn't taken effect... [02:09:01] <alanc> make -/, sh -/, tar -/.... [02:09:08] <cyborg--> And when i try and set it, i get permission denied, the interface is unplumbed while i am trying this [02:09:25] <Tempt> Oh dear. [02:09:37] <alanc> though I still think -/ looks more like the old console spinner than a command flag [02:09:42] <Tempt> Why not demand VMS style syntax next? [02:09:44] <gdamore> what kind of e1000g is it, many dont' support jumbo frames [02:09:55] <cyborg--> this is an ibm x345 [02:10:08] <gdamore> onboard nic? [02:10:08] <Gman> alanc, one more command option you'll never use [02:10:17] <gdamore> if its an ich8, it doesn't do jumbo frames. [02:10:41] <cyborg--> onboard to x345 yeah [02:10:47] <gdamore> ICH8? [02:11:18] <alanc> just as long as we don't get to adding -\ flags for DOS pathname compatibility and have to explain to people how to quote that in their shells properly [02:11:21] <gdamore> yeah, looks like that. [02:12:31] <gdamore> well, i'm not sure now. [02:12:37] <Triskelios> are the older SXCE install images archived by Sun? [02:12:40] <gdamore> put the prtconf -vp to pastebin.ca. [02:13:03] <gdamore> Triskelios: yes. internally they are online. [02:13:14] * gdamore keeps his own copies around as well. [02:14:08] <cyborg--> http://www.pastebin.ca/658979 [02:14:11] <Triskelios> okay, a particular system wants a b63 x86 *CD* and I'm not sure if I should use the one from snv_69 [02:14:27] <nrubsig> alanc: do you have any good suggestion for another ksh93 demo (last one was shnote.ksh) ? [02:14:34] <Triskelios> (to install packages from an snv_63 DVD) [02:14:41] <gdamore> cyborg: just a sec. [02:15:25] <alanc> nrubsig: not off hand [02:15:57] <Tempt> nrubsig: How about a cheery text based front-end to an mp3 player? [02:16:08] <Tempt> nrubsig: That could use some of those multithreading features. [02:16:10] <gdamore> hmmm... i82546EB apparently [02:16:22] <cyborg--> which should not support it? [02:16:33] <Tempt> nrubsig: Or a menued frontend to zonecfg/zoneadm? [02:16:40] *** rmorse has left #opensolaris [02:16:40] *** rmorse has quit IRC [02:16:42] <gdamore> stilll figuring that part out [02:17:10] <cyborg--> i have one on my desk here, the chip says RC82545GM i have a case of these ones , i could go to the DC and install it, this is really just for testing [02:17:18] <cyborg--> but i want to get a feel for performance now [02:17:41] <nrubsig> Tempt: multithreading is still in prototype stage. [02:17:43] * alanc almost said "ksh93 on Rails", but that's not really a demo, and quite scary to actually think about [02:17:50] <nrubsig> Tempt: I mean multithreading in ksh93 [02:17:51] <nachox> lol [02:18:05] <gdamore> looks like it *should* support jumbo frames [02:18:08] <nrubsig> alanc: what would "ksh93 on rails" do ? [02:18:18] <Tempt> nrubsig: What about something that demonstrates the superior performance of ksh93? [02:18:21] <FrostCS> yea, google is saying it should be supported also.. [02:18:24] <gdamore> i need to leave now, but i'll be back in about 3 hours. [02:18:25] <Tempt> nrubsig: Does something flashy without sucking the CPU dry? [02:18:37] <FrostCS> I'll try to help if I can.. heh [02:18:46] * delewis has witnessed the superior performance of ksh93 [02:18:47] <alanc> database-backed ajax-y websites, like Ruby on Rails [02:18:59] <nrubsig> Tempt: we have the mandelbrot1.ksh demo as example for a multicpu parallel application. [02:19:08] <FrostCS> cyborg-- plumb your interface again, and edit your .conf again, attempting to set the Max frames again. [02:19:14] <nrubsig> delewis: ?! [02:19:15] <delewis> I converted a friend's cgi-scripts last week, which were shell scripts from bash to ksh93, and just from changing #!/usr/bin/bash to #!/usr/bin/ksh there was a 30% performance gain. [02:19:29] <delewis> (that wasn't even changing his usage of expr, sed, etc. to ksh93 builtins) [02:19:32] <cyborg--> FrostCS in /kernel/drv/e1000g.conf i have 3,3,3,3 [02:19:33] *** jpdrawneek has quit IRC [02:19:44] *** onlinebacon has quit IRC [02:19:44] <nachox> nrubsig, how do you actually spawn threads in ksh? [02:20:15] <nrubsig> nachox: additional builtin command [02:20:16] <Tempt> Actually, a replacement for "dialog" would be quite handy. [02:20:20] <FrostCS> cyborg--, gimme a sec, going to check something quick here. [02:20:21] <Tempt> That utility is getting old and crufty. [02:20:32] <brendang> with DTrace probes in both ksh88 and ksh93, showing superior performance will be easy [02:20:35] <nachox> nrubsig, aka completely unportable right? [02:20:46] <nrubsig> nachox: for now, yes [02:21:11] <nrubsig> nachox: in fact no other shell can implement threads anyway since you need subshells without |fork()| for each subshell first [02:21:15] <nachox> there is an RFE for bash or something like that to add threads? [02:21:20] <alanc> is anyone going to bother putting them into ksh88 though? [02:21:24] <nrubsig> nachox: AFAIK no. [02:21:50] <brendang> alanc: I hope so [02:21:53] <nrubsig> alanc: threads ? [02:22:01] <alanc> nrubsig: DTrace probes [02:22:38] [02:22:38] <brendang> alanc: it isn't a lot of effort, for potentially a lot of gain (I've seen a bit of ksh88 out there - which will take a while to get moved to ksh93) [02:22:44] <Tempt> nrubsig: Did you know that IBM ships ksh93 with AIX? [02:22:45] <alanc> I know Tpenta did them for /bin/sh, and there's been talk for a while in your project of doing them for ksh93 [02:23:16] <nachox> nrubsig, ouch i tend to use ksh scripts that are portable to bash just by changing the #! shell, no threads for me [02:23:19] <Tpenta> i want to settle down the interface efore I take on any other shells [02:23:21] <nrubsig> Tempt: yes, and I wish they would UPDATE their damn version from time to time... [02:23:28] <delewis> the ksh93 in AIX is ancient. [02:23:31] <alanc> brendang: yes, but the reason ksh88 got into the state it was in was because it was shipped and no one wanted to touch it again unless they had to, and since the source isn't open.... [02:23:35] <delewis> its either 'g' or 'h' IIRC [02:24:06] <Tpenta> alanc: can you see any problems if I wanted to "do the work" for the probes at the same time i was looking at other shells? [02:24:14] <brendang> alanc: yep. well. hopefully a little dtrace debug code to ksh88 won't hurt anything. [02:24:36] <Tpenta> well given that probably a third of the scripts in ON are ksh88, ... [02:24:39] <alanc> Tpenta: nope [02:24:56] <FrostCS> cyborg--, alright, can we try setting that MaxFrameSize again, what were you setting it to the last time? [02:25:03] <Tpenta> i did the numbers the other day in a blog, if memory serves, probably 70-80% of scripts in ON are sh and ksh88 [02:25:11] <Tpenta> almost all the rest are incarnations of perl [02:25:44] <nachox> maybe you can use dtrace to fix nightly <wink> [02:26:06] <cyborg--> # ndd -set /dev/e1000g1 MaxFrameSize 9000 [02:26:06] <cyborg--> operation failed: Invalid argument [02:26:16] <FrostCS> cyborg-- oh, you can't set by ndd [02:26:17] <alanc> now if you could sneak into DevPro land and put dtrace probes into make... [02:26:28] <FrostCS> cyborg--, it has to be edited manually [02:26:33] *** Mdx4 has quit IRC [02:27:15] <delewis> Tpenta: how difficult was it adding USDT probes to /sbin/sh? [02:27:22] <nachox> alanc, and cc? :) [02:27:41] <cyborg--> Okay... well, i have 3's in the /kernel/drv/e1000g.conf file under framesize, where else do i modify this [02:27:56] <Tpenta> not very, the most tricky part was the line number probes and a few things to do with command-* [02:28:00] <Tpenta> and subshell-* [02:28:01] *** linux_user400354 has joined #opensolaris [02:28:07] <delewis> ah. [02:28:12] <FrostCS> MaxFrameSize right cyborg--? [02:28:38] <brendang> so, just so everyone is clear on this, there is a *chance* that ksh88 probes will be completed before ksh93 [02:28:43] <delewis> I have a few applications that adding USDT probes to could prove beneficial for. [02:28:48] <cyborg--> MaxFrameSize=3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3; [02:29:05] <FrostCS> cyborg, alright, should work, can you try a reboot? [02:29:11] * Gman likes vance's latest article about the vmware ipo [02:29:18] <cyborg--> i've reboot since it has been set to that :| [02:29:25] <cyborg--> i could reboot again, but i havent changed anything [02:29:36] <FrostCS> cyborg--, I know.. it's redundant, but I am just trying to make sure things took [02:29:49] <cyborg--> Okay its rebooting [02:30:05] <cyborg--> (thank you for the attention on this minor issue) [02:30:24] *** quobecha has joined #opensolaris [02:30:43] <Tpenta> I wrote a blog on it about 2 years ago and that is pretty much the only source of information I used to add the probes, lemme find the link (or perhaps brendang has it handy) [02:31:04] <FrostCS> cyborg--, np, hopefully I can help you get it going.. [02:31:24] *** pitty has joined #opensolaris [02:31:32] *** quobecha has left #opensolaris [02:32:20] <pitty> anyone know if cut has an equivalent of "awk '{ print $NF }' " ? [02:32:56] *** tsp has quit IRC [02:32:56] <nachox> pitty, that prints the number of fields in every line? [02:34:43] <cyborg--> okay its rebooted [02:35:48] <cyborg--> so far no luck [02:35:48] <FrostCS> cyborg--, have you already added the default_mtu = 9000 line in nge.conf? or not yet? [02:35:52] <cyborg--> no [02:36:02] <FrostCS> can you add that please? [02:36:22] *** linux_user400354 has quit IRC [02:37:58] <cyborg--> okay i've added it [02:38:01] *** lacaAFK is now known as laca [02:40:07] *** bubbva has quit IRC [02:40:19] <FrostCS> cyborg-- in /etc/system is jumbo enabled? [02:41:00] <cyborg--> grep jumbo /etc/system turns up nothing [02:42:42] <FrostCS> ah, shouldn't need on nge it looks like [02:42:50] <FrostCS> can you try another annoying reboot now? [02:42:51] <axisys> jmcp, tsoome sorry I had to step away for few mins.. [02:43:03] <axisys> i am looking for some rule of thumbs.. [02:43:09] <cyborg--> k [02:43:30] <axisys> if it is a multithreaded app it will work best on T2k.. something like that [02:43:54] <axisys> so if app is cpu intensive then faster cpu is better for the app [02:44:26] <axisys> some recommended practices are what I am looking.. will help me to find the right server for right ap [02:44:29] <axisys> app* [02:47:25] <cyborg--> FrostCS Okay its rebooted now, when i do ifconfig -a still 1500 on the e1000g [02:47:38] <cyborg--> lo0 is 8232 but it always was [02:49:04] <FrostCS> ndd -get /dev/e1000g1 max_frame_size [02:49:04] <reflect> right, I tried that prtvtoc | fmthard thing to copy the disk tables.. it worked on one disk, but not the other (x86 system) .. so, I've been trying to fiddle with fdisk to make sure they look the same.. is there a way to copy the fdisk information too? [02:49:10] <FrostCS> still gives the same cyborg--? [02:49:17] <cyborg--> 1514 [02:49:20] <cyborg--> yep [02:49:39] <FrostCS> and e1000g is plumbed, correct? [02:49:52] <cyborg--> well its online [02:50:26] *** Kush- has joined #opensolaris [02:51:16] <FrostCS> what do you get with modinfo | grep nge ? [02:52:14] <cyborg--> Nothing [02:53:08] <Gman> heh, joerg didn't get his wishes [02:54:32] <FrostCS> hmm [02:56:05] *** yongsun has joined #opensolaris [02:56:24] <FrostCS> cyborg--, have you already posted to opensolaris.org ? [02:56:32] <cyborg--> No [02:56:46] <cyborg--> maybe i will go stick this other intel card in see if it works [02:56:52] <cyborg--> MT server adapter [02:57:08] <FrostCS> yea, unless you want to wait a few hours and see if gdamore can help when you get back.. [02:57:19] <cyborg--> well, the datacenter is below me so [02:57:35] <cyborg--> and... this is really for testing i will be using something better if i go ahead with this anyway.. i am gonna run down pop this card in [02:57:36] <FrostCS> I am far from an expert in networking, but for some reason your mtu is just refusing to change [02:57:47] <FrostCS> alright :-) [02:58:14] *** BatonT has joined #opensolaris [03:00:17] <mritun_> umm, xfce is complaining about xrender and xdamage [03:00:24] <jbk> uugh.. this old 802.11b card is awful.. damn you broadcom for not releasing your specs *shakes fist* [03:00:32] <mritun_> can I just download and install them, or is it hopeless ? [03:01:32] *** karrotx_ has joined #opensolaris [03:06:01] <FrostCS> mritun_ if none of the xfce self proclaimed specialists are here now, you can ask in #xfce [03:07:52] *** stevel has quit IRC [03:08:56] <jbk> ok.. http://hg.genunix.org/libdisasm.hg for all interested (I included a copy of libdisasm*.ch from ON so you don't need to grab the entire ON source to build the library) [03:11:44] <mritun_> FrostCS: I'm afraid of linux zealots :-( [03:13:18] *** wirwzd has joined #opensolaris [03:13:58] <FrostCS> mritun_, are you working from the xfce project page at opensolaris or compiling from scratch? [03:14:38] *** tsp has joined #opensolaris [03:15:04] *** simford has joined #opensolaris [03:15:29] <mritun_> compiling from scratch [03:15:30] <tsp> How do I get Solaris to detect my scsi card? I tried boot -r, but I'm at a loss [03:15:48] <mritun_> xfce on opensolaris is from blastwave repo [03:16:03] <mritun_> and would pull down whole gnome and gtk [03:16:16] <mritun_> doubling the size [03:16:17] <FrostCS> no, the opensolaris project isn't from blastwave [03:16:24] <mritun_> ah [03:16:43] <FrostCS> http://www.opensolaris.org/os/project/xfce/ [03:16:52] *** lloy0076 has joined #opensolaris [03:17:21] <lloy0076> Is it possible to run SXCE b69 under a Solaris XEN DOM-0 domain? [03:17:27] <jmcp> tsp: which card is it? [03:17:52] <FrostCS> lloy0076, I think that's how it was set up since the lastest Xen bits, right? [03:17:55] <tsp> jmcp: no clue [03:18:10] <tsp> I'm trying to find something that'll tell me [03:18:15] <nachox> scanpci [03:18:24] <nachox> or prtconf [03:18:29] <lloy0076> FrostCS: I have no idea :P [03:19:07] <FrostCS> lloy0076, give me a second or two to find it.. [03:19:40] <jmcp> tsp: prtpicl -v | less and search for pciclass,01 [03:19:41] <lloy0076> It appers that one could, but I'm not sure. [03:19:50] <FrostCS> http://blogs.sun.com/levon/entry/solaris_xen_update [03:20:12] <tsp> ah this sounds like it from prtconf: ADPT,2940UW (driver not attached) [03:20:25] <jmcp> aac, or adpu320 [03:20:28] <jmcp> I'm picking adpu320 [03:20:52] <jmcp> tsp: here's my glm card - http://pastebin.ca/659022 [03:21:04] <FrostCS> lloy0076, looks to be that it's an opensolaris domu :-) [03:21:05] <jmcp> look for the "compatible" property [03:21:25] * jmcp aways [03:21:43] <FrostCS> dom0 .. evil keyboard [03:21:58] <lloy0076> FrostCS: Ah, it seems that one can. [03:22:01] <lloy0076> heh [03:22:04] <mritun_> FrostCS: wow !!! thanks a bunch [03:22:25] <mritun_> I was half way through... will have to clean the crap from /usr/local [03:23:24] <FrostCS> I am sure you could always get help via desktop-discuss if you needed on any xfce problems, whether it be compiling, or using the project page though. [03:23:31] <FrostCS> just incase you had any problems. [03:25:31] <cyborg--> No luck, same with the other card... [03:25:56] <FrostCS> which version of sxce again cyborg--? [03:26:47] <cyborg--> b69 ? [03:26:51] <FrostCS> ah hrm [03:26:58] <cyborg--> i just downloaded it this week [03:27:14] <FrostCS> yea, then it's b69 (been there for a few weeks) [03:27:28] <FrostCS> I wish I was more of a networking expert.. [03:28:06] <FrostCS> well, you can always hang around and see if gdamore returns, or post on the mailing list about it. [03:29:20] <tsp> apparently I have a mac compatible scsi adapter, not going to help me much over here. [03:29:54] <tsp> if :driver,aapl,macos,powerpc is anything to go by [03:31:18] <Tempt> What the hell is gcj? [03:31:48] <alanc> isn't that the GNU java compiler? [03:31:58] *** SYS64738 has quit IRC [03:32:09] <Tempt> You're kidding, right? [03:32:33] <Tempt> Man, I give up. You'd think people would manage to build *java* apps that weren't overloaded with GNUisms [03:33:03] <alanc> GCJ: The GNU Compiler for Java: http://gcc.gnu.org/java/ [03:33:23] <Tempt> If I can't get it to build without I'm not going to bother. [03:34:06] <tsp> heh [03:34:39] <Tempt> Bloody GNU can go die in a fire. [03:34:56] <Tempt> I think I'm going to re-write the few GNU tools I actually use so I can boycott the entire project. [03:35:04] <tsp> gnu made some good utils though [03:35:05] <tsp> like grep [03:35:10] <tsp> well, gnu grep [03:35:13] <Tempt> And I'll license the tools under CDDL as a further raised finger. [03:36:13] <Tempt> tsp: Which particular feature makes GNU grep your favourite? [03:36:27] <nachox> recursive? [03:36:28] <tsp> yup [03:36:39] <nachox> use find! [03:36:42] <tsp> forgot what else, but I don't think solaris grep has that unless I missed it [03:37:11] <tsp> oh, find . -exec grep pattern {} \;? forgot about that one [03:37:32] <brendang> \; is bad. use + [03:37:57] <tsp> +? never heard of that one [03:37:57] <nachox> you probably want find only to show regular files though [03:38:04] <brendang> tsp: try it [03:38:13] <nachox> i never heard about + either [03:38:18] <Bartman007> I'll stick with gnu grep, I'd prefer to not spawn a few thousand grep processes :-P [03:38:21] <brendang> it is in the man page [03:38:31] <Tempt> Adding shit like -r is not in the UNIX spirit of small simple tools that do their job well [03:38:38] <nachox> in fact all the exec examples i've seen were using \; [03:38:41] <Tempt> The GNU spirit is large complex tools that break things in creative ways [03:38:44] <brendang> boyd taught me about the + a long time ago.... [03:38:46] <Tempt> (eg, emacs) [03:39:04] <Bartman007> Tempt: most of the gnu userland is a far cry from emacs.... [03:39:09] <tsp> emacs is weird. gnu ed on the other hand kicks solaris's ed around like a basketball [03:39:48] <Tempt> Bartman007: It's all a bit wierd and tries deliberately to be *different* to UNIX. [03:39:55] <Tempt> and rolls in crap that isn't required. [03:40:20] <nachox> some stuff is actually nice, they did a good job with sed [03:40:31] <Tempt> Besides, grep regex $(find ...) [03:40:34] <nachox> long options are a pain in the ass [03:40:40] *** sparc-kly has quit IRC [03:40:41] <Bartman007> Tempt: of course it tries to be different than unix, remember what it stands for? [03:41:08] *** sparc-kly has joined #opensolaris [03:41:09] *** ChanServ sets mode: +o sparc-kly [03:43:05] <bda> I wonder if it's hilarious if /advocacy/immigrants/ links to benr's Sol10 intro, which uses the phrase "people who know their shit." [03:43:29] *** yippi has joined #opensolaris [03:43:34] <Bartman007> IIRC, alot (most?) of the GNU userland that deviates from the exact POSIX standard have a POSIX mode. [03:43:49] <brendang> GNU sed, grep good? well, ok, they are nice enhancements from AT&T's Unix tools. But how many GNU tools are original, innovative? [03:44:10] <nachox> i didnt say they were innovative [03:44:16] <nachox> i said they were good [03:44:20] <Tempt> GNU hello world [03:44:24] <tsp> lol [03:44:27] <Tempt> That's a mighty impressive innovation [03:44:30] <Tempt> Over 5000 lines of code [03:44:34] <brendang> ok, they are good. [03:44:46] <elektronkind> wget and screen come to mind. [03:44:54] <nachox> other gnu stuff sucks [03:44:57] <Tempt> I don't think either actually started as GNU projects. [03:45:15] <nachox> ls, find, tar come into my mind [03:45:32] <tsp> tar xzf, makes our lives easier [03:45:33] <Tempt> Discussion of GNU tar is forbidden under the not_today_schily act of 2005 [03:45:41] <nachox> lol [03:45:47] <Tempt> UNIX find works fine [03:46:05] <Tempt> GNU ls I'll pay, because I do like colour ls [03:46:08] <Tempt> is pretty! [03:46:17] <nachox> please! [03:46:35] <nachox> the first thing i ever do when i start a redhat console is unalias ls [03:46:39] <bda> pft. -F should be good enough for anyone! [03:46:49] <Tempt> Hmm, since we have 256 colour terminals these likes, perhaps we could *enhance* gnuls colour support [03:46:50] <brendang> Tempt: you want colours? I did write "gwhiz" - which can colourise anything, not just ls.... [03:46:59] <Tempt> The older the file, the darker the colour [03:47:09] <Tempt> That way you can see the new updates by looking for the sparklies [03:47:16] <Tempt> brendang: Oh, from your *special* toolkit. [03:47:23] <tsp> nachox: heh, unalias ls, rm, cp, mv, set editor/visual to ed, that usually fixes most of the problems linux throws at me [03:47:23] <Bartman007> Tempt: new files have to BLINK! [03:47:37] <nachox> tsp, hahah [03:47:43] <richlowe> brendang: but does your ed syntax highlight? [03:47:45] <richlowe> brendang: and if not, why not? [03:47:48] <nachox> tsp, rm is not aliased unless youre root [03:48:13] <tsp> nachox: I always type rm -rf out of habbit now. I don't even know why I need the f, but its always there :) [03:48:27] <brendang> richlowe: that would be useful. so long as it highlighted COBOL, SNOBOL, APL, FORTRAN-77, DCL... [03:48:28] <nachox> force? [03:48:51] <tsp> heh [03:49:04] * tsp goes to check just what it forces that -r won't already get [03:49:05] <Tempt> We've all had colour terminals for a long time now, and I really don't mind interactive-only tools being colourful. I always call /bin/ls directly in scripts to ensure no GNUisms (in the rare cases where one needs ls in a script) [03:49:06] <Gman> colordtrace! [03:49:26] <nachox> tsp, r is recursive [03:49:40] <tsp> gotta love pipes [03:49:50] <tsp> ls -1 | sort -n | tail -1, or ls -1 | wc -l [03:49:56] <brendang> Gman: I've used colour in dtrace scripts a number of times now. usually for demos, putting different layers of the software stack in different colours (red == kernel, blue == lib*, etc) [03:50:09] <Gman> brendang, yeah, but we need a #pragma ;) [03:50:23] <brendang> Gman: ahh [03:50:30] <brendang> Gman: #pragma D option color [03:50:40] <Gman> you know, for when colorls becomes the default ;) [03:50:40] <Tempt> When I installed Gentoo Linux on my SunPCi card, I noticed it ships with a colour *grep* [03:50:43] <Tempt> That's wrong ... [03:50:47] <mritun_> ok [03:50:49] <mritun_> not good [03:50:54] <LeftWing> hah [03:50:55] <mritun_> xfce started [03:51:06] <mritun_> but "tips & tricks" window is blank :-/ [03:51:06] <richlowe> uh, GNU grep is colorized. [03:51:10] <nachox> Tempt, err, gnu grap does colors too, and gentoo sucks... :P [03:51:11] <richlowe> grep --color, same opts as ls [03:51:19] <nachox> *grep [03:51:33] <tsp> why would you want a color grep? one color for the filename, another for the text? [03:51:44] * mritun_ thought grap ==> GNUfied crap [03:51:45] <richlowe> tsp: it highlights the match, I guess. [03:51:46] <flyingparchment> tsp: colouring the matched text i assume [03:51:47] <Tempt> "./src/gpmisc.c", line 97: undefined symbol: fdopen64 [03:51:56] <Tempt> Is fdopen64 a GNUism/glibcism? [03:51:59] <LeftWing> Hah, match highlighting. That's neat. [03:52:00] <flyingparchment> i don't see why you'd need colours though, VMS search did it with just bold [03:52:05] <tsp> heh, emacspeak and c-s tried to do that [03:52:09] <Tempt> bold is effectively a colour [03:52:14] <tsp> with different voices - it gets really annoying really fast [03:52:29] <Gman> alanc, [slap head time, huh?] [03:52:41] <LeftWing> bold/colour/underlining/whatever -- they're all highlighting mechanisms. [03:53:37] <flyingparchment> Tempt: i see no fdopen64 on my linux system [03:54:14] <Tempt> Aah, I'll dig around a bit more [03:54:20] <Tempt> compiling ghostscript [03:54:26] <flyingparchment> don't see it on solaris either - but i suppose it would make sense to exist.. [03:54:41] <flyingparchment> (lf64(5) doesn't mention it) [03:55:22] <elektronkind> [daleg@solaris1]~$ nm /usr/lib/libc.so |grep fdopen64 ; echo $? [03:55:23] <elektronkind> 1 [03:55:29] <elektronkind> nooop. not on solaris [03:55:58] <Triskelios> Tempt: fdopen64 is not in glibc, fwiw... [03:56:15] <Triskelios> maybe a BSDism [03:56:28] <tsp> why do people insist on putting newly quoted email text at the bottom of everything that went before? [03:56:28] <flyingparchment> google suggests it's a Cygwinism [03:56:39] <nachox> why dont you look for that function in the source browser? [03:56:43] <flyingparchment> or maybe newlib, i think that's what cygwin uses as libc [03:56:51] <Triskelios> flyingparchment: interesting [03:57:40] <alanc> Gman: you can choose fully-compatible or available today - but not yet both - give us a little time [03:58:07] *** trede has quit IRC [03:59:11] <Gman> alanc, yeah, completely understand - was just miffed at how stupid i was for pointing out iconv and not realizing about X ;) [03:59:36] <alanc> see item #1 - have to recompile GNOME because libXext is incompatible 8-) [03:59:51] <Gman> :) [04:00:20] <alanc> though using GNOME libiconv seems overkill when there's a perfectly good iconv() in Solaris libc [04:00:21] <CIA-17> mh27603: 6589662 Error messages from cpu_acpi on install boot w/snv_70, 6589155 assertion failed: domp, file: ../../common/io/ppm/ppm.c [04:00:29] <alanc> err, GNU libiconv [04:01:06] <alanc> oh, except that's probably part of the stuff that Emanicipation Project needs to replace [04:01:22] <Gman> yep [04:01:42] <Bartman007> tsp: many mailing lists get quite pissed if you "top post" [04:02:10] <tsp> Bartman007: why? You can instantly get the replies when your whipping through a large thread [04:02:38] <alanc> guess I get my own head-slapping moment now [04:02:48] <Bartman007> tsp: because most people want a bit of context. [04:03:03] <tsp> ah [04:03:12] <Tempt> [11:56] <LeftWing> COLOURED GREP IS THE WIN. [04:03:19] <tsp> damn vmware is killing itself every time it tries to boot solaris [04:03:20] <Bartman007> tsp: I realize it is a huge pain for you, since you can't scroll through as easily, but that's the way it is. [04:03:32] *** hohum has joined #OpenSolaris [04:03:44] <boyd> jmcp: Yes, it is depressing. The thing is that I can often see the point, but shake my head at the way it's presented. [04:03:51] <tsp> once I write a better ed, I could probably take care of that, since I sometimes just do ed mbox and forget mutt altogether [04:04:26] <boyd> tsp: brendang has a ged written in java. Has sound effects and all :) [04:04:45] <tsp> neat [04:04:47] *** apersson has quit IRC [04:05:08] <tsp> windows cannot stay up for more than 5 days without crashing, or else grinding to a halt [04:05:30] <tsp> and unfortunately I'm going to need it for nfs-mounting the rest of my /home :) [04:06:12] *** sommerfeld has quit IRC [04:06:28] <boyd> I'm late to this... you need windows for nfs with ~ why? [04:06:41] <tsp> boyd: I can't hook this scsi disk up to my U5, incompatible scsi card [04:06:52] <Gman> hrm, curious, having site permissions to opensolaris.org doesn't give you the ability of editing any page :/ [04:06:53] <tsp> and I have a soalris vm on windows managing a 100gb ide zfs disk [04:07:18] <tsp> so I figure adding the 50gb scsi disk won't kill it - if I can get solaris to boot again without crashing vmware, that is [04:08:20] <Tempt> anyway, lunchytime [04:09:50] <jbk> ok, that was a long email [04:10:02] *** bnitz has quit IRC [04:10:37] *** logic has joined #opensolaris [04:11:08] <FrostCS> mritun_ you might need to compile up libgamin looks like the package was missing [04:16:23] <mritun_> hehe [04:16:27] <mritun_> just discovered ;) [04:16:40] <mritun_> the file explorer crashed... so looked it up [04:16:58] <mritun_> I doubt if gamin would compile though... looks like linux specific thing [04:17:02] <mritun_> trying though [04:18:01] *** Shiv__ has joined #opensolaris [04:18:38] *** jafari has joined #opensolaris [04:20:03] *** halton has joined #opensolaris [04:22:48] *** logic__ has quit IRC [04:25:02] *** nachox has quit IRC [04:28:10] <Chipdancer> jmcp: pingo! [04:28:21] <jmcp> acko [04:28:28] <Chipdancer> my solaris dom0 is working! [04:28:29] <jmcp> ... and Bingo was his name-o [04:28:32] <jmcp> w0000t! [04:29:09] <Chipdancer> there was a bug whereby image files had to be exactly on a 512b boundary and some of my image files gained a byte on copying it seems [04:29:15] <Chipdancer> a few other issues too, but that was the show stopper [04:29:42] <jmcp> interesting [04:29:51] <jmcp> did you file those issues @ b.o.o ? [04:30:07] *** alanc is now known as alanc-away [04:30:22] <Chipdancer> no, they were known issues, but ones that were hard to determine [04:30:27] <jmcp> oh, ok [04:30:36] * Chipdancer praises movement, dme and mrj for their help [04:32:38] *** jamesd has joined #opensolaris [04:32:45] *** jamesd_ has joined #opensolaris [04:34:08] *** halton has left #opensolaris [04:34:33] <mritun_> gcc is picking up /usr/ccs/bin/ld [04:34:39] <mritun_> which borks up libtool [04:34:59] <mritun_> how to tell it to pick up /usr/sfw/bin/gld ? [04:38:49] <nrubsig> next version of http://svn.genunix.org/repos/on/branches/ksh93/gisburn/scripts/shnote.ksh is out. [04:39:11] <nrubsig> can anyone please look at the "usage" text and check whether there is a way to make it less confusing ? [04:39:54] *** pitty has quit IRC [04:41:27] <nrubsig> jmcp: ping! [04:41:32] <jmcp> ack [04:41:39] <jmcp> nrubsig: checking that url now [04:41:56] <nrubsig> jmcp: do you have any idea how an arc case for /usr/bin/shnote would look like ? [04:42:04] <jmcp> short [04:42:07] <jmcp> a fasttrack [04:42:17] <jmcp> shouldn't be too controversial [04:42:25] <nrubsig> jmcp: erm... the problem is more that pastebin.ca is not a stable interface [04:42:42] <jmcp> you should include an option to choose a specific pastebin server [04:42:50] <jmcp> eg, -s http://rafb.net/paste [04:43:06] <nrubsig> jmcp: yeah, I was thinkit about adding a "provider" option and add support for Google notebook later [04:43:18] <Triskelios> mritun_: libtool normally does the right thing for ld... maybe the package has weird LDFLAGS? [04:43:34] <nrubsig> jmcp: devils question: Would you sponsor such a fasttrack ? [04:43:39] <mritun_> Triskelios: its trying to pass on version script [04:43:50] <Triskelios> mritun_: what software? [04:43:59] <jmcp> not me, I don't have sufficient standing in PSARC - I'm not a Member, Licensee or Intern [04:44:03] <mritun_> using --version-script, which the tradition ld in ccs/bin doesn't support [04:44:10] <jmcp> sommerfeld is a Member, Tpenta is an Intern [04:44:23] <mritun_> gamin - the async file notification plugin for gnome [04:44:31] *** eboutilier_ has joined #opensolaris [04:44:38] [04:44:39] <Gman> hey eboutilier_ [04:44:44] <Triskelios> mritun_: I suggest the SFE package [04:44:45] <nrubsig> Is eboutilier_ a ARC member ? [04:44:47] <eboutilier_> hey glynn [04:44:54] <eboutilier_> heh. [04:45:07] <Gman> mritun_, the jds dudes have been working on such a package [04:45:12] <Gman> mritun_, you might want to sync up with them [04:45:17] <nrubsig> eboutilier_: are you an ARC member ? [04:45:41] <eboutilier_> I used to be a member of the Automation and Research Computing Dept at a gov. agency I worked at... Does that count? [04:45:53] <nrubsig> groan [04:46:05] <Gman> they've put gamin through ARC (LSARC 2007/398) [04:46:16] <eboutilier_> That was 15 years ago, so I suppose not. :-P [04:47:05] <nrubsig> eboutillier_: I am looking for a victim which sponsors an ARC case for this little tool: [04:47:08] <nrubsig> $ shnote put "foo bar" [04:47:10] <nrubsig> SUCCESS: http://pastebin.ca/657301 [04:47:11] <nrubsig> $ shnote get 657301 [04:47:13] <nrubsig> foo bar [04:47:30] <mritun_> Gman: ah... long road then :( [04:47:44] <nrubsig> eboutilier_: e.g. "put" and "get" from/to internet clipboards like pastebin.ca and Google Notebook etc. [04:47:51] <mritun_> I can fix the makefile... but it's too much of horse work right now [04:48:13] <Gman> mritun_, not at all, there should be unstable packages available now [04:48:18] <mritun_> nrubsig: a suggestion, the arguments and usage text can be improved [04:48:27] <mritun_> Gman: where ? [04:48:31] <nrubsig> mritun_: how ? [04:48:37] <Gman> looking [04:49:07] <mritun_> nrubsig: pastebin -p - << take from stdin [04:49:21] <mritun_> nrubsig: pastebin -g 1234 OR [04:49:34] <mritun_> nrubsig: pastebin -g http://foo.com/1234 [04:49:39] <Gman> nrubsig, needs --put, --get and obviously -/ ;) [04:49:58] <mritun_> because a lot of times, you'd have the URL directly... much easier to just copy and paste the whole thing [04:50:07] <Gman> mritun_, http://dlc.sun.com/osol/jds/downloads/current/ [04:50:13] <Gman> but i'm not 100% sure those packages contain gamin [04:50:18] <Gman> trying to verify [04:50:31] <nrubsig> mritun_: put using files requires more tricky encoding of the request, I have to ask in the networking-discuss@ mailinglist for help with that. [04:50:34] <Gman> it's certainly gnome 2.19.x [04:50:48] *** kohju has left #opensolaris [04:50:49] <nrubsig> mritun_: and "get" using pastebin.ca URL works in the new version [04:51:05] *** jafari has quit IRC [04:51:06] <nrubsig> mritun_: see http://svn.genunix.org/repos/on/branches/ksh93/gisburn/scripts/shnote.ksh , follow string "recordname" [04:51:16] <Gman> mritun_, hrm, looks like he hasn't landed yet [04:51:25] <mritun_> Gman: no use for me... b60 over here [04:51:32] <Triskelios> mritun_: come on, SFEgamin =P [04:52:09] <Gman> Triskelios, bizarre, it's not there either [04:52:26] <Triskelios> um, it is... I have it installed [04:52:37] * nrubsig realises that all ARC sponsors are madly hiding from him... ;-((((((((((((( [04:52:43] * Gman svn ups [04:53:08] <Triskelios> wait, no I don't o_O [04:53:31] <Triskelios> oh, it's SUNWgamin there ;) [04:53:47] <Triskelios> that's convenient... [04:53:55] *** lloy0076 has left #opensolaris [04:54:38] <Gman> hrm, i don't have that package [04:54:40] <Gman> strange [04:54:43] <bda> http://bwitz.com/leibniz_newton.jpg # Ha! [04:55:03] <eboutilier_> Gman: Saw your message about website admin privs... [04:55:12] <Gman> eboutilier_, ah, yes? [04:55:50] <eboutilier_> What you can do is make yourself a "leader" temporarily, then you can edit the page. [04:56:10] <richlowe> Yup. [04:56:17] <Gman> eboutilier_, sick. [04:56:22] <richlowe> Gman: (which is why stevel seems to appear and disappear in various places...) [04:56:25] <Gman> but entirely doable ;) [04:56:32] <Triskelios> laca created SUNWgamin, apparently. don't know why it's under SUNW, maybe it's referenced in the ARC proposal [04:56:54] <Gman> Triskelios, doesn't appear to be in trunk for me [04:57:28] <Triskelios> Gman: it's still in SFE, not jds [04:57:37] <laca> Triskelios: i didn't [04:57:42] <richlowe> maybe the SUNW name is just a typo? [04:57:47] <Gman> eboutilier_, where can i add leaders? [04:57:53] *** trede has joined #opensolaris [04:58:12] <Gman> eboutilier_, i don't seem to have an option in my site admin menu [04:58:14] <laca> hmm... actually, ChangeLog says halton did, %changelog says it was doug [04:58:48] * richlowe wonders what Gman is trying to accomplish [04:59:04] <Gman> richlowe, just fixing a random bug in the san diego page [04:59:18] <Gman> richlowe, plus it's good to know where things are [04:59:26] <eboutilier_> Gman: That's 'cause you're a site admin, not a leader, only leaders can add leaders. Hee hee. Seriously. However! ... [04:59:47] * Gman smacks eboutilier_ :) [04:59:49] <eboutilier_> You're a site admin so you can change user profiles, so change yours to be a leader of the project you wan to edit. [05:00:22] <eboutilier_> Logical in a twisted way. [05:00:30] <eboutilier_> A way only Java programmers would think of :-) [05:00:54] <Gman> so evil :) [05:02:36] <eboutilier_> Oh, by the way, once you're a leader of that project you'll have access to add/delete leaders from it, including yourself when you're done! [05:02:57] * eboutilier_ lets out an evil laugh [05:04:08] * nrubsig prepares some holy water and looks at eboutilier_ [05:04:19] * jmcp goes for the garlic [05:04:21] <nrubsig> Evil shall not prevail! [05:04:26] <richlowe> flaming torches and pitchforks. [05:04:27] * eboutilier_ starts hearing tubular bells [05:04:31] <Gman> eboutilier_, updating users is pretty slow [05:04:33] <jmcp> it's ok, joerg isn't here right now [05:04:40] <Gman> eboutilier_, for my next trick i need to get mysql access :) [05:04:49] <eboutilier_> Gman: You wanted speed? [05:05:00] <nrubsig> eboutilier_: Embda grama hindam geriam mutasba gerra devillibis mutare! [05:05:19] * nrubsig tries to remove satan's influence from eboutilier_ [05:05:50] <nrubsig> richlowe: I failed. [05:05:53] <nrubsig> richlowe: use forks! [05:06:06] <eboutilier_> Gman: I have mysql access, it's called stevel. :-) [05:06:12] <nrubsig> heh [05:06:15] <Gman> heh [05:06:27] <nrubsig> !seen stevel [05:06:28] <Drone> stevel (stevel!n=stevel at 192 dot 18.43.225) was last seen in #opensolaris on Wed 15 Aug 2007 18:54 GMT, saying 'just verified - the password still works'. [05:06:46] <nrubsig> !seen satan [05:06:47] <Drone> satan (satan!n=chatzill at dslb-084-058-238-163 dot pools.arcor-ip.net) was last seen in #opensolaris on Tue 12 Jun 2007 00:51 GMT, saying 'Grrrrrrrraaaaaaaaaaaahhhhh'. [05:07:22] *** wirwzd has quit IRC [05:08:09] *** jamesd_ has quit IRC [05:08:16] *** lisppaste3 has joined #opensolaris [05:08:40] *** jamesd has quit IRC [05:09:40] *** alfism has quit IRC [05:11:48] <nrubsig> Gman: what's the current time in your timezone ? [05:12:25] <Gman> 3:12pm [05:13:10] <nrubsig> east coast or west coast ? [05:13:16] <gdamore> i'm back now... is cyborg- still here? [05:13:42] <Gman> nrubsig, very east, or very west - .nz [05:13:55] <nrubsig> Gman: oh [05:14:06] *** yippi has quit IRC [05:14:49] <nrubsig> Gman: that may explain why this channel went that quiet. US is asleep... [05:15:01] <richlowe> it's not that late in much of the US [05:15:20] <nrubsig> richlowe: 20:15h [05:15:21] <gdamore> 8pm here on the west coast. later elsewhere... [05:15:23] <jmcp> ssshhh don't wake them [05:15:24] * brendang isn't asleep [05:15:33] <nrubsig> don't wake them ? [05:15:38] <Tempt> brendang: That's because you *live* at work. [05:15:39] <gdamore> WAKE UP YOU LAZY LAY-ABOUTS! [05:15:49] <brendang> Tempt: I do the night shift [05:15:50] <nrubsig> BAAAAANGGG BAMBAMBAM WAKEY WAKEY, GET UP US!!! [05:16:05] <Tempt> brendang: Does that night shift last more than 14 hours? [05:16:06] <nrubsig> BRRRRRRRRRRAM BAMBAM DING! [05:16:26] * brendang puts on headphones [05:18:36] <brendang> gee, these headphones work well [05:19:01] *** richlowe has quit IRC [05:19:03] * Tempt slams down a hammer on a railway detonator next to brendang [05:19:40] <brendang> Tempt: aww, come on. like StateRail would leave detonators around for the public ot steal.. [05:19:57] *** richlowe has joined #opensolaris [05:20:00] <Tempt> brendang: Of course. shittyrail would never leave boxes of them lying around. [05:20:58] <brendang> Tempt: if you hang off the front of a train, you can scoop them up before the train hits them. of course, you have to be pretty fast. [05:21:21] <Tempt> Or you could use a magnet! [05:21:45] <Tempt> Of course, if you go near a shittyrail train with a magnet it'll probably come flying off the rails. [05:21:51] <Tempt> Actually, it'll probably do that without a magnet. [05:23:03] *** jamesd_ has joined #opensolaris [05:23:17] *** Shiv__ has quit IRC [05:23:49] *** jamesd has joined #opensolaris [05:23:57] <brendang> Tempt: noo.. surely they have a perfect track record ... http://en.wikipedia.org/wiki/Railway_accidents_in_New_South_Wales ... [05:24:23] <Tempt> brendang: I got a most amusing set of MP3s from LeftWing mocking them ... I gather you've heard them? [05:24:28] *** sioraiocht has quit IRC [05:24:33] *** sioraiocht has joined #opensolaris [05:24:46] <brendang> Tempt: no? [05:26:07] <Tempt> Let me dig it up. [05:27:04] *** ChanServ sets mode: +o jamesd [05:27:22] *** bda has quit IRC [05:29:57] *** laca has quit IRC [05:30:00] *** laca_ has quit IRC [05:35:46] *** halton has joined #opensolaris [05:36:19] *** tsp has quit IRC [05:36:57] *** whaq has quit IRC [05:36:58] <theRealballchalk> i'm trying to burn a dvd and it tells me /usr/bin/cdrecord.bin: Drive does not support TAO recording. and Illegal write mode for this drive. [05:37:03] <theRealballchalk> what does that mean? [05:37:22] <theRealballchalk> i did 'cdrecord -v dev=1,0,0 blah.iso' [05:38:11] *** jamesd_ has quit IRC [05:38:28] *** jamesd_ has joined #opensolaris [05:38:59] <theRealballchalk> it's a DVD+R btw [05:39:40] *** jamesd has quit IRC [05:40:08] *** jamesd__ has joined #opensolaris [05:40:09] <Tempt> add the option -dao [05:40:10] *** cyborg-- has quit IRC [05:42:48] *** sstallion has quit IRC [05:44:11] <theRealballchalk> Tempt: i just found out that my device won't write DVD+R medias [05:44:19] <Tempt> +R [05:44:20] <Tempt> Woah [05:44:34] <Tempt> Does anyone use +R anymore? [05:45:14] <theRealballchalk> Tempt: dunno but Best Buy had it at a discount price of $12.00 for a pack of 50 lol [05:45:18] <theRealballchalk> hey thanks [05:45:19] <theRealballchalk> brb [05:45:23] *** theRealballchalk has left #opensolaris [05:49:13] *** trede has quit IRC [05:50:00] *** rootard has left #opensolaris [05:52:22] *** Megaf has joined #OpenSolaris [06:03:05] *** Gman has quit IRC [06:03:56] *** kohju has joined #opensolaris [06:04:46] <elektronkind> quick, what's a good way to panic a box from within mdb? [06:05:13] <elektronkind> ISTR a way to get a box to crash by changing /'s inode [06:06:36] <jmcp> elektronkind: I use 0>rip :c :c [06:06:51] *** jmcp has quit IRC [06:07:05] <richlowe> apparently he does, too [06:07:10] <richlowe> elektronkind: $systemdump [06:07:17] <richlowe> it does what jmcp said, but with a little more magic. [06:07:22] <boyd> !reboot -d :) [06:07:29] <richlowe> boyd: that isn't always effective. [06:07:34] <boyd> true [06:07:35] <richlowe> and sometimes, is worse. [06:07:43] <richlowe> reboot -dqn may be firm enough. [06:07:46] <boyd> It was a jike anyway [06:07:49] <boyd> and a joke too [06:07:50] <elektronkind> I just did a dtrace -w -n 'BEGIN{panic();}' [06:07:53] <richlowe> but 0'ing %pc always works [06:07:57] <richlowe> (or it better) [06:08:04] <boyd> heh [06:08:20] <elektronkind> finally I might get a core that'll explain these zfs hangs [06:08:29] <richlowe> elektronkind: sparc or x86? [06:08:43] * boyd remembers then there were "drop to debugger" buttons. (like MACSBUG) [06:09:00] <richlowe> (not that it matters hang-wise, I just want to see the scat stuff actually be useful) :) [06:09:08] <elektronkind> x86... s10u3 fully patched. hangs during a clone destroy [06:09:09] *** bda has joined #opensolaris [06:09:17] <elektronkind> 64bit, x4100m2 [06:09:40] <elektronkind> I got a core it seems. lets see how useful it is when the box comes back up. [06:10:03] *** jmcp has joined #opensolaris [06:10:21] * jmcp growls @ ctrl-alt-backspace [06:10:52] <richlowe> jmcp: how'd you do it? [06:11:20] <jmcp> the hand is quicker than the eye [06:11:27] <jmcp> and I think I've got some sticky keys on this kbd [06:11:30] <boyd> Hah [06:12:50] <Tempt> panic a box? You mean STOP-A sync? [06:13:25] <jmcp> or you could run panicsys>rip :c :c [06:13:34] <richlowe> $systemdump, as above. [06:13:38] <jmcp> or, my favourite - $<systemdump [06:13:44] <richlowe> Yeah, that one. [06:13:56] <boyd> Does anyone know if the SIL Open Font License would be Ok for distribution with OpenSolaris? There are several nice high-qual fonts that we could use. [06:14:09] <jmcp> boyd: got a pointer to the license? [06:14:55] <boyd> http://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&id=OFL [06:15:26] *** lloy0076 has joined #opensolaris [06:15:53] <jmcp> boyd: looks ok to me [06:15:59] <Tempt> boyd: URL for fonts? [06:16:12] <boyd> Fonts: http://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&item_id=OFL_fonts [06:16:23] <boyd> I will mention it in -discuss [06:17:16] <jmcp> and documentation community [06:18:40] <boyd> ? [06:18:55] <boyd> You mean so they can point people a them to install themself? [06:19:07] <richlowe> or so they can bake them into their materials, if they wanted... [06:19:30] <richlowe> plus, anything involving "license" and -discuss, is de facto a horrible idea. [06:19:34] <richlowe> :) [06:19:41] <jmcp> boyd: I reckon the G11n/I18n folks would be interested [06:19:56] <lloy0076> What does http://www.pastebin.ca/659177 actually mean? Does that mean that the Linux linux-pool has 2 processors but could use a third, and the pool_default has 2 processors and could use the other 2? [06:20:12] <boyd> jmcp: probably... [06:20:14] <lloy0076> Or does it mean that linux-pool and pool_default get 2 processors and that this is set in stone. [06:20:28] <bda> http://nostarch.com/frameset.php?startat=irc # 437 pages. [06:20:29] <bda> What? [06:21:15] <Tempt> boyd: What's got you poking at fonts, anyway? [06:21:16] <boyd> lloy0076: I feel like it should be the former... but I'm not sure [06:21:50] <boyd> Tempt: something like this: http://xkcd.com/214/ [06:22:58] <Tempt> That's a cartoon taking the shit out of wikipedia. [06:23:13] <Tempt> Amusing as it is ... it isn't clickin' [06:23:16] <boyd> It's a cartoon taking the shit out of me. That's how I got to fonts. [06:24:00] <boyd> I can't remember where I started... 3 days ago. [06:24:37] <lloy0076> boyd: It appears that "objectives" can affect it; given that it seems that the objectives are wt-load, it should be able to manipulate the resources. [06:24:51] <boyd> W00t. I got autoconf to do my bidding.. now I can do the real work that I'm trying to achieve [06:24:56] <richlowe> boyd: you'll never find a decent mono font. [06:24:58] <Tempt> aaaaaaaaaaaah [06:25:01] <richlowe> also, autoconf never does your bidding. [06:25:08] <richlowe> it just pretends to, to lull you into a false sense of security. [06:25:08] <boyd> lloy0076: Yeah, that's kinda the idea I have in minds [06:25:13] <boyd> Heh [06:25:49] <bda> Monaco is a decent mono font. [06:25:50] <boyd> I'm found "lucida sans mono" or something like that to be OK. [06:25:56] <Tempt> ./configure --more-fucking-keystrokes-worth-of-options-than-it-would-take-to-edit-a-fucking-make-file=what-a-great-big-bloody-joke --prefix=ignore-me-and-write-over-usr-bin-and-everything-else --suck --suck --suck [06:26:13] <lloy0076> boyd: My glitch at the moment is trying to get enough stuff to load the damned thing enough to cause it to use more than 2 processors :P [06:26:24] <boyd> Tempt: all I was doing was getting ./configure --enable-dtrace to do something useful. [06:26:31] <boyd> Heh [06:26:36] <boyd> yes > /dev/null ? [06:26:59] <boyd> Tempt: and by useful I mean "include my code in the build" [06:27:11] <richlowe> boyd: which shell? [06:27:13] <Tempt> lloy0076: You're trying to load a machine up? Just find something that parallel compiles well and do a tight loop with gmake -j 999 ; gmake clean [06:27:26] <boyd> richlowe: zsh ATM [06:27:34] <Tempt> richlowe: I have an excellent monospaced font. [06:28:06] <richlowe> and you're about to say 'fixed' or something, aren't you? [06:28:12] <lloy0076> I assume that "size" in poolstat means "the number of CPUs in the pset" - but it doesn't really say... [06:29:15] *** pitty has joined #opensolaris [06:29:42] * boyd remembers using "10x20" a lot in the old days [06:31:06] <Tempt> Tempt: VGA consolefont [06:31:17] <Tempt> richlowe: VGA consolefont, now that I'm done talking to myself [06:31:24] <Tempt> ripped from a Tseng Labs ET6000 [06:32:17] <boyd> Not bad.. I had an a pair of Apple ][ fonts in truetype for a while (low res 40 col and highres 80 col) [06:33:29] <boyd> Actually I don't mind whatever this default font for gnome-terminal is. [06:34:49] <Tempt> boyd: Ooh, nifty. [06:35:12] <Tempt> vga.pcf is full of win, although you need to use the magical terminfo if you want line draw characters to work properly. [06:35:21] <Tempt> I really should put both the terminfo and the font online somewhere [06:35:41] *** eboutilier_ has quit IRC [06:38:01] <lloy0076> Damn. [06:38:17] <lloy0076> gmake[6]: warning: jobserver unavailable: using -j1. Add `+' to parent make rule :: I presume that means gmake won't run parallel? [06:38:54] <jmcp> lloy0076: yes, because you're using the wrong OS [06:39:01] * jmcp hides behind a convenient tree [06:39:58] <dlg> heh [06:40:10] <lloy0076> What's wrong with OpenSolaris :P ? [06:40:51] <jmcp> sorry, I thought you were still talking about a linux instance [06:41:04] <jmcp> lloy0076: dmake >> gmake, any day [06:41:17] <richlowe> but gmake has no many interesting, uh, things. [06:41:22] <jmcp> hrh [06:41:25] <richlowe> I guess. [06:41:28] <jmcp> that was my Beavis impression [06:41:33] <jmcp> heheh is what I meant [06:41:40] <richlowe> though, actually, I guess dmake's hatred of # is one way gmake is in theory better. [06:41:51] <richlowe> "I'll get you yet, POUND_SIGN!" [06:42:00] <jmcp> yeah, it's a damned HASH [06:45:53] <elektronkind> weird. so my server got all partially hung up while executing a 'zfs set readonly=on' on a filesystem, which in my case is something that's done right after a clone of the same name was destroyed [06:46:00] <richlowe> jmcp: not quite what I meant, but hell yeah. [06:46:12] *** Megaf is now known as _Megaf [06:46:19] *** jHoNDoE has joined #opensolaris [06:46:38] <elektronkind> lots of zfs_delete_threads() sticking around, doing nothing it seems. wedged, perhaps [06:47:55] *** _Megaf is now known as _Megaf|Sleep [06:48:12] <_Megaf|Sleep> Good Night All [06:48:54] *** delphi1000 has joined #opensolaris [06:49:20] <delphi1000> hi everybody [06:52:43] *** mritun_ has left #opensolaris [06:56:44] <BatonT> ive done a zfs replace on smaller drives in my array (one by one) to larger ones, except the zpool hasnt increased in size once i removed all the smaller drives, im sure ive done this before... [06:58:52] <delphi1000> what? [06:58:59] <delphi1000> what is your question? [06:59:15] <BatonT> why am i not seeing the aditional space? [07:00:07] <delphi1000> the drivers was installed correctly? [07:01:01] <BatonT> yes.. my zpool contains 4 x 500gb drives, but i still only have the space of 4 x 400gb [07:02:01] <jmcp> elektronkind: you need to get a crash dump if at all possible [07:02:12] <elektronkind> oh yeah, I got one [07:02:20] <jmcp> or at least the output of echo "::threadlist -v" |mdb -k [07:02:28] <jmcp> then email details to zfs-discuss [07:02:33] <elektronkind> http://pastebin.ca/659205 [07:03:14] <elektronkind> it's s10u3, so I'll post it there and run the core up to sunsolve with a open case for it [07:03:42] <jmcp> elektronkind: the thing to do is to find the oldest thread, and figure out what it's waiting on [07:04:12] *** sparc-kly has quit IRC [07:04:15] <elektronkind> it would have the lowest thread id ? [07:04:19] <jmcp> not necessarily [07:06:39] <jmcp> can't recall exactly, but I think you'd be looking for the kthread_t's t_stime element [07:06:56] <richlowe> sounds right. [07:07:44] <jmcp> elektronkind: you've also got a thread waiting for reclock() to return, and some afs threads [07:08:08] <BatonT> OK my zpool issue where it wasnt getting larger after replacing small drives with big ones... well it is now seeing extra space after a reboot! [07:08:15] <jmcp> elektronkind: logging a call is your best option [07:10:23] <elektronkind> yeah, I will do so [07:10:34] <elektronkind> jesus hp laserjet christ [07:10:39] <libkeiser> the afs kernel thread backtraces look normal. doubt afs is involved since elektronkind is probably using ufs for cache, and namei if there's a server [07:10:48] <elektronkind> a threadlist -v busted my 10k line scrollback buffer [07:11:09] <elektronkind> yeah, the afs callabck listener is normal there [07:12:40] <richlowe> elektronkind: what's ffffffffa70d7ca0 sleeping on? [07:13:16] *** jamesd_ has quit IRC [07:13:39] *** jamesd__ has quit IRC [07:17:15] <elektronkind> running typegraph now [07:20:55] <richlowe> Now that's gonna take you a while. [07:21:03] <richlowe> it's awesome, when it finishes though. :) [07:21:30] <elektronkind> typegraph: maximum nodes => 5946684 [07:21:32] <elektronkind> :/ [07:24:04] *** tsoome has quit IRC [07:27:03] *** sunmaster has joined #opensolaris [07:33:38] *** delphi1000 has quit IRC [07:36:19] * elektronkind makes like a thread and sleep [07:36:26] <elektronkind> gnight, thanks for the help all [07:36:37] *** Doc has quit IRC [07:36:42] *** dlynes has joined #opensolaris [07:37:29] *** dlynes has quit IRC [07:37:37] *** dlynes has joined #opensolaris [07:39:29] *** Berny_ has joined #opensolaris [07:40:09] *** Berny has quit IRC [07:45:31] *** Berny__ has joined #opensolaris [07:45:31] *** Berny_ has quit IRC [07:45:32] *** Doc has joined #opensolaris [07:48:38] <boyd> Tempt: ping [07:48:43] <Tempt> plonk [07:48:53] <boyd> That's what I was gonna ask about [07:48:58] <Tempt> ooh [07:49:01] <Tempt> good idea [07:49:12] <Tempt> Where? [07:49:20] *** tsoome has joined #opensolaris [07:49:54] <boyd> No preference... the default seems to be spleen [07:50:22] <boyd> Anyway, chat in ~20 mins [07:51:23] <Tempt> okay [07:51:26] <Tempt> spleen seems fine to me [07:51:29] <Tempt> But I won't be out before 5:00 [07:58:36] *** baijiutong has joined #opensolaris [07:59:12] *** tsp has joined #opensolaris [07:59:36] <tsp> Anyone have the command that will display detailed disk info? I can't find it anywhere, but need to check that this new disk is the right size [07:59:46] <flyingparchment> tsp: format? [08:00:19] <Tempt> iostat -E [08:01:24] <g4lt-mordant> prtvtoc [08:01:27] <tsp> it was some command that gave me a big list with my drives serial number and stuff [08:01:33] <Tempt> iostat -E [08:01:35] <Tempt> iostat -E [08:01:40] <Tempt> I've got you three times now ... [08:01:45] <tsp> oh, -E [08:01:47] <tsp> I had -e [08:01:59] <tsp> thanks! there it is [08:02:00] <bda> :E [08:02:02] <bda> Hairlip. [08:02:06] <bda> Harelip? [08:02:12] <bda> Fatlip. [08:02:13] * bda dances. [08:03:08] <tsp> ack! good thing I didn't try to mount that as zfs, that was the capacity clipping jumper :) [08:04:41] <Bartman007> tsp: just wondering, how are you able to clarify that it was a capital E and not a lowercase e? [08:04:46] *** estibi has joined #opensolaris [08:04:51] <tsp> I have a way to read character by cahracter if I need to [08:05:00] <Bartman007> ah. [08:05:01] <tsp> and the e was in a higher pitch [08:06:32] <tsp> If I hear corrupt magic number one more time... :) [08:07:41] <tsp> damn. it was trying to tell me I had the wrong drive, so I guess it was justified [08:09:55] *** estibi has quit IRC [08:09:59] *** yongsun has quit IRC [08:10:29] *** timely has joined #opensolaris [08:10:53] <e^ipi> ahoy [08:12:06] <e^ipi> !seen alanc [08:12:13] <Drone> alanc (alanc!n=alanc at 192 dot 18.43.225) was last seen in #opensolaris on Thu 16 Aug 2007 02:02 GMT, saying 'guess I get my own head-slapping moment now'. [08:12:41] <sbahra> heh [08:12:43] <g4lt-mordant> !seen useful_license_discussion [08:12:46] <Drone> I've never seen useful_license_discussion talk in #opensolaris. [08:12:55] <sbahra> !seen one two [08:12:57] <Drone> I've never seen one two talk in #opensolaris. [08:13:05] <sbahra> !seen useful license discussion [08:13:08] <Drone> I've never seen useful license discussion talk in #opensolaris. [08:13:11] <sbahra> g4lt-mordant, :-D [08:16:10] <e^ipi> my question for alanc is technical and not religious, i assure you [08:17:10] <g4lt-mordant> I just got reminded of drone by your comment ;) [08:17:20] <richlowe> e^ipi: what is it? [08:17:48] <e^ipi> richlowe: wondering if I can cram an apple PPC video card in to my sparc & expect it to work [08:18:24] <e^ipi> because Sun's framebuffers don't use vesa sync on 16:10 resolutions, only on 4:3 [08:18:52] *** Drone has quit IRC [08:19:03] <oxygene> e^ipi: if you can live with it not providing an obp console (and thus, probably solaris console), then likely yes [08:19:20] <e^ipi> I see.... [08:19:30] <oxygene> it _might_ work even with console [08:19:31] *** timelyx has quit IRC [08:19:46] <tsp> noooo, S10U3 zfs and B66zfs are incompatible [08:19:48] <e^ipi> could I use the FFB2 as the console, and setup X to use the radeon ? [08:19:54] <oxygene> should work [08:19:55] <g4lt-mordant> ITYM "16x9", HTH [08:20:00] <e^ipi> fantastic. [08:20:06] <richlowe> g4lt-mordant: I don't think he did. [08:20:10] <e^ipi> g4lt-mordant: what? [08:20:11] <richlowe> I think he meant 8:5 ;) [08:20:14] <oxygene> but apple does some weird stuff in their fcode (ppc binary blob with driver for the OS), and I think they leave out some other things [08:20:15] <tsp> I swear, this day is getting worse and worse [08:20:21] <e^ipi> no, i meant 16:10 [08:20:27] <g4lt-mordant> HD is 16:9, by the specs [08:20:33] <e^ipi> the native res is 1600x1000 [08:20:36] <oxygene> g4lt-mordant: 16:10 is "a bit more" for the media controls [08:20:49] <e^ipi> and sun's cards can do 16:10 [08:20:58] <e^ipi> but evidently not in VESA mode [08:20:59] <tsp> Will Sun care if I download every version of OpenSolaris as its released? [08:21:28] <g4lt-mordant> right, wouldn't want to be exactly like ATSC, that'd be too easy to make multifunction equipment [08:23:02] <g4lt-mordant> e^ipi, which radeon, might I ask? [08:23:11] <e^ipi> mac radeon 7000 [08:23:35] <e^ipi> they're pretty common [08:23:58] <e^ipi> I dunno if 3d will work, I suspect not [08:24:03] <g4lt-mordant> in that case, the proper question is "can I cram an XVR-100 clone into my sparc" [08:24:14] <e^ipi> but i'm more interested in a 2d desktop that doesn't look like ass than I am about 3d [08:24:19] <g4lt-mordant> mac radeon 7000 == XVR100 [08:24:45] <e^ipi> can the xvr100 do 1600x1000 with VESA and not SOG sync? [08:24:59] <g4lt-mordant> second question, why not just get a full-on XVR100? [08:25:20] <g4lt-mordant> it's the same card, the only differencee is the firmware [08:26:30] <e^ipi> because I need a specific resolution with vesa sync [08:26:37] <e^ipi> and none of the other Sun cards can do it [08:27:32] <g4lt-mordant> well, like I said the XVR100 is a mac radeon 7000 with some firmware changes to make it look nice for obp [08:28:49] <e^ipi> http://www.sun.com/desktop/products/graphics/xvr1000/specs.html <-- under "resolutions supported", if I can find a spec sheet like that for xvr-100, i'll consider it [08:29:07] <Tempt> Get a 600 [08:29:13] <Tempt> much faster [08:29:24] <g4lt-mordant> it used to be on sunsolve before they raped it [08:29:36] <e^ipi> Tempt: no... it leaves me with the same problem I have now [08:29:39] <g4lt-mordant> hang on, let me find it on the old mirror [08:29:44] <e^ipi> my 16:10 monitor is unsupported by the card [08:29:49] <Tempt> aaaaaaah. [08:30:08] <e^ipi> because the card is brain-dead, and won't use VESA sync on anything but 4:3 resolutions [08:30:29] <Tempt> XVR-600 has DVI and your problems go away [08:30:35] <Tempt> Unless you bought a display with no DVIV. [08:30:36] <Tempt> DVI [08:30:38] <Tempt> In which case ... [08:30:39] <Tempt> pwned [08:30:49] <toblun> tsp: No, why sould they? [08:31:15] <e^ipi> my monitor has DVI [08:31:33] <Tempt> Well, there you go. [08:31:47] <Tempt> Get a framebuffer that does DVI and connect by DVI and DVI your way to DVI happiness [08:32:11] <e^ipi> does DVI still not require syncing? [08:32:33] <e^ipi> also, look at the sheet again... 1600x1000 is unsupported on anything but 13w3 [08:35:54] *** WickedWicky has joined #opensolaris [08:36:01] <WickedWicky> good morning/evening al [08:36:02] <WickedWicky> l [08:36:54] *** Drone has joined #opensolaris [08:37:12] <Tempt> xvr-100 doesn't have 13w3 [08:38:08] *** slowhog has left #opensolaris [08:38:51] <sunmaster> I am reading this word "framebuffer" again and again on forums/chats/MLs of Unix-near communities: Can someone, please, explain me the difference between framebuffer and graphics card? [08:39:02] <Tempt> Possible values are: 1600x1000x66, 1600x1000x76. [08:39:11] <Tempt> that's from fbconfig on my xvr-600 enabled machine [08:39:18] <tsp> toblun: don't they have to pay for all the bandwidth we use to download it? [08:39:20] <Tempt> so not ideal, no. [08:39:21] <WickedWicky> heya Tempt [08:39:27] <Tempt> Heya WW [08:39:47] <oxygene> sunmaster: graphics card implies that it's a plugin card - framebuffer is an older term, which merely describes the functionality [08:40:10] <toblun> tsp: Sure they have to. But still, I dont think they mind... [08:40:24] <tsp> toblun: I guess if they did, they'd set up rsync servers [08:40:53] <sunmaster> oxygene, Thanks. [08:43:08] <Triskelios> sunmaster: http://en.wikipedia.org/wiki/Framebuffer also describes the evolution of graphics chipsets from framebuffers [08:43:11] <g4lt-mordant> e^ipi, oh, hey, cnet agreees with me about the XVR100, but what the fuck do I know http://reviews.cnet.com/graphics-cards/sun-xvr-100-graphics/4507-8902_7-32508129.html?tag=nav [08:44:41] <Triskelios> for the XVR 100, the tables linked from http://www.sun.com/products-n-solutions/hardware/docs/html/819-3290-12/chap1.intro.html#pgfId-999203 seems sort of authoritative [08:45:30] *** nostoi has joined #opensolaris [08:46:23] *** Cyrille has joined #opensolaris [08:47:06] *** Gman has joined #opensolaris [08:49:52] * WickedWicky sheds a tear. I have an uptime of more than 12 hours after compiling onnv-gate [08:50:00] <WickedWicky> xcuse me while I have an emo moment [08:51:28] <e^ipi> Triskelios: notice the table which states that the only VESA resolution is 1600x1200... which is a 4:3 res., and not a 16:9/16:10 [08:51:46] <e^ipi> making it quite useless to me [08:53:12] <Triskelios> yes, you're SOL. 1600x1000 probably won't turn out so well, either [08:53:44] <WickedWicky> you wanna connect a widescreen to a framebuffer? [08:53:57] *** FunkyELF has joined #opensolaris [08:54:14] <FunkyELF> seriously....how stable is zfs right now on opensolaris express edition? [08:54:18] <e^ipi> WickedWicky: i want to connect a widescreen to a SPARC framebuffer & have it use some sort of 16:9 or 16:10 resolution [08:54:40] <WickedWicky> I wasnt aware there were VESA standards for widescreen [08:54:51] <Triskelios> FunkyELF: good enough for small sites, at least [08:54:51] <g4lt-mordant> FunkyELF, I'm using it quite happily on both my laptop and my U60 [08:55:00] <WickedWicky> or that widescreen existed when they made those standards [08:55:06] <e^ipi> there are vesa standards for syncing, and sun standards for syncing... [08:55:15] <boyd> Tempt: spleen good.. still got people here [08:55:24] <FunkyELF> I'm thinking about running a server with samba and apache (with php and mysql) [08:55:33] <e^ipi> they are not the same thing, and my monitor only supports vesa (as it is not a sun product) [08:55:46] <FunkyELF> I was going to do the software raid thing, but then I read about ZFS and how it was better than raid 5 [08:56:23] *** phimic has joined #opensolaris [08:56:32] <g4lt-mordant> yeah, on my U60, I use it to make a JBOD a single pool, and it works well. it even survives a flaky disk now [08:56:34] <Triskelios> e^ipi: that is what sync on screen adapters are for [08:56:58] <FunkyELF> g4lt-mordant, you don't use raid-z? [08:57:07] <g4lt-mordant> ust zpool [08:57:29] <FunkyELF> ust? [08:57:31] <FunkyELF> just? [08:57:35] <bobbyz> e^ipi: as WickedWicky pointed out, 1600x1000 is not a standard vesa resolution. [08:57:47] <nrubsig> does anyone here have a T1/T2-based machine ? [08:58:07] <FunkyELF> so zpool can't survive a disk failure but raidz can and is better than raid5? [08:58:11] <Triskelios> bobbyz: neither is 1600x1050 afaik, but that works with most video cards [08:58:18] <Triskelios> FunkyELF: correct [08:58:23] <g4lt-mordant> maybe when I get the round tuit to make the JBOD into a useful size, I may make it a raidz, but right now it's just a 711 stuffed with 18's, no point in raiding it [08:58:25] <e^ipi> DVI may solve the problem, but I'm not willing to drop $200 on a sun framebuffer to test out a theory [08:59:02] <g4lt-mordant> well, zpool can survive disk failure, FSVO failure [08:59:17] <e^ipi> i'd rather spend $15-20 on an ancient mac framebuffer that works 90% of the way ( console excluded ) which I know can drive the monitor properly [09:00:02] <FunkyELF> what happens when there is a bug found in ZFS?....is there something that comes out and traverses the filesystem and fixes things? [09:00:17] *** gnut has joined #opensolaris [09:00:20] <gnut> hi all [09:00:29] <gnut> has anyone gotten the wpi0 to wotrk on their laptop? [09:00:45] <WickedWicky> nope [09:00:57] <Triskelios> gnut: been working fine since snv_62 [09:01:00] <WickedWicky> it timesout on loading the firmware [09:01:09] *** linux_user400354 has joined #opensolaris [09:01:10] <gnut> WickedWicky: yeah. i get that issue of timing out when trying to connect [09:01:15] <WickedWicky> HP laptop? [09:01:20] <gnut> lenovo x60 [09:01:34] *** linux_user400354 has quit IRC [09:01:59] <WickedWicky> for me it's a bug in the kernel, where it finds the PCI-E2PCI bridge but doesnt look behind it. So the device broadcasts presence but the kernel cant address it properly [09:02:03] <gnut> i found a hack posted on the forums to edit the firmware so that the pcie bridge can forward memory along?... so now i can plumb the device [09:02:06] <gnut> but i can't conncet [09:02:11] <WickedWicky> there is apparently some hack around it to address the memory range manualy [09:02:17] <WickedWicky> yea tha [09:02:19] <gnut> yeah [09:02:22] <WickedWicky> I didnt try that though [09:02:30] <gnut> it's pretty easy to do [09:02:33] <gnut> so i can plumb it now [09:02:36] <WickedWicky> I am using a PCMCIA card now, D-link , works fine [09:02:41] <WickedWicky> oh really? [09:02:44] <gnut> but dladm connect-wifi doesn't succeed [09:02:59] <WickedWicky> hmm [09:03:02] <Triskelios> FunkyELF: zpool scrub will do an integrity check on the data on-demand... not sure about other corruption, but that almost never happens and is usually caught [09:03:11] <gnut> WickedWicky: yeah... just go to the command line [09:03:16] <gnut> run mdb -KF [09:03:29] <gnut> type 2-3 lines of commands (which i've memorized now :) ) [09:03:32] <gnut> and then that's it. [09:03:41] <WickedWicky> I'll give it a spin [09:04:04] <gnut> https://www.opensolaris.org/jive/thread.jspa?threadID=32222&tstart=15 [09:04:28] <gnut> last post [09:04:37] <gnut> Triskelios: got yours to work? [09:04:55] <Triskelios> gnut: it has worked since the initial code drop of wpi [09:05:02] <gnut> which laptop do you have? [09:05:07] <gnut> i'm on snv_65 [09:05:15] <gnut> but i guess it's not the witreless driver issue [09:05:18] <Triskelios> VAIO VGN SZ72B/B [09:05:21] <gnut> it's the pcie [09:06:28] <Triskelios> I think I've used snv_62, snv_63, snv_66 [09:07:06] <Triskelios> pci8086,27d2, instance #2 (driver name: pcie_pci) is my pcie/pci bridge [09:07:26] <gnut> hmm [09:07:29] <gnut> that's what I have, too. [09:07:41] <gnut> pci8086,27e2 instance #1 [09:07:50] <gnut> i meant to type pci8086,27d2, instance #1 [09:08:00] *** karrotx has quit IRC [09:08:48] <boyd> Tempt: ping [09:10:29] <gnut> dladm show-linkprop wpi0 shows a question mark under the value of "radio" [09:10:32] <gnut> not on nor off [09:11:00] <Triskelios> that's normal [09:11:17] <Triskelios> powermode is also ? for me [09:11:38] <Tempt> boyd: Pongarez [09:11:38] *** Berny__ has quit IRC [09:11:45] <gnut> Triskelios: yeah. same here. [09:12:24] *** Berny__ has joined #opensolaris [09:12:25] *** Berny__ has quit IRC [09:12:26] <FunkyELF> booting this opensolaris express community editiion....I can boot open solaris or opensolaris developer edition [09:12:28] <boyd> Tempt: going Tapas? :) [09:12:29] <FunkyELF> whats the difference? [09:12:31] <gnut> dladm connect-wifi -e mykey -e essid wpi0 just keeps scrolling... [09:12:46] <Triskelios> gnut: does scan-wifi work? [09:12:55] <gnut> yup [09:12:55] <Triskelios> I've occasionally had it fail to load the firmware (maybe one out of every 10-20 boots) but a reboot fixes that [09:13:03] <Triskelios> interesting [09:13:04] <gnut> scanning works fine [09:13:19] <gnut> and i know i'm transmitting since my router sees the MAC address of my laptop get deposited there [09:13:20] <e^ipi> FunkyELF: just pedantry, but it's solaris express: community edition [09:13:25] <e^ipi> opensolaris is just a pile of code [09:13:26] <gnut> so i've enabled access to that MAC address [09:13:47] <e^ipi> SX:CE is an opensolaris distro [09:13:56] <gnut> hmmm [09:13:58] <gnut> let me check something [09:14:04] <Triskelios> try a non-encrypted AP? [09:15:22] <gnut> yeah. i was going to do that [09:15:35] <Triskelios> oh wait [09:15:51] <Triskelios> you're not specifying "mykey" as the actual key are you? [09:15:58] <gnut> no [09:16:12] <Triskelios> it should be the name of the secure object (create-secobj, etc.) [09:16:17] <gnut> i did the dladm create-secobj -c wep -f file_of_key mykey [09:16:23] <gnut> as in the man page [09:16:26] <FunkyELF> I'm trying to run this DVD iso image in VirtualBox in Linux...it had a bunch of dots (....) going , then it just reboots [09:16:27] <Triskelios> okay [09:16:55] <bobbyz> FunkyELF: the current VirtualBox version doesn't handle solaris properly [09:17:05] <FunkyELF> bobbyz, damn [09:17:15] <bobbyz> qemu was choking for me too. I had to use vmware-player [09:17:22] <FunkyELF> maybe I should burn this thing and install it? [09:17:39] <bobbyz> apparently there was somesort of a regression from 1.38 to 1.40 with regard to solaris [09:19:24] <bobbyz> if you don't mind vmware-player, you can use easyvmx.com to generate a nice blank vmx/vmdk for use with solaris [09:19:41] *** dunc has quit IRC [09:19:43] <WickedWicky> I dont think Solaris will work out of the box in qemu either [09:19:47] <asyd> \_o< [09:20:03] <WickedWicky> I have seen some fora posts where peple managed to do it by using a debuger during the loading of the kernel [09:20:03] <asyd> hmm citrix want to buy xensource, sounds strange [09:20:19] <FunkyELF> damn, seems like a hassle [09:20:21] <WickedWicky> _____/\_____\o/ ack ack shark attack [09:20:30] <gnut> hmm.. Netgear must be a foreign company since the english in their router configuration menus is pretty bad [09:20:36] <WickedWicky> morning asyd [09:21:09] <g4lt-mordant> have I mentioned the non-wiseneess of running bzcat SunStudio12-solaris-x86-200705-pkg.tar.bz2| tar xvf - while you have any time constraints? [09:21:21] <WickedWicky> nope [09:21:39] <WickedWicky> why run bzcat anyway? [09:21:39] <WickedWicky> bunzip2 -c [09:21:40] <g4lt-mordant> okay, it's not too wise [09:22:03] * FunkyELF goes to bed and thinks about actually burning this iso instead of trying to get it working in virtualbox [09:22:05] <g4lt-mordant> WickedWicky, which does precisely what that bzcat doesn't do? [09:22:06] <bobbyz> is sun studio 12 open to the public for download as 11 was? [09:22:09] <WickedWicky> bunzip2 -c SunStudio12-solaris-x86-200705-pkg.tar.bz2 | tar xvf - [09:22:17] <WickedWicky> yes but that _is_ wise and it works [09:22:33] <WickedWicky> bzcat does not I presume, other wise it'd be a wise thing to do [09:22:43] <g4lt-mordant> WickedWicky, it does xactly the same thing in exactly the same timframe [09:22:54] <WickedWicky> then what is your problem? [09:22:54] <jmcp> g4lt-mordant: perhaps you should add the "n" flag to tar [09:23:02] <WickedWicky> it's not like it takes half an hour [09:23:13] *** mikefut has joined #opensolaris [09:23:14] <Triskelios> bobbyz: has been for a few months [09:23:18] <WickedWicky> unless you do it over a serial console [09:23:23] <WickedWicky> then you should ditch the v [09:23:24] <bobbyz> Triskelios: sweet! I had no idea. [09:23:33] <WickedWicky> hey jmcp [09:23:43] <Triskelios> bobbyz: it's also on the SXCE DVDs [09:23:47] <g4lt-mordant> jmcp, just finger memory, I'd guess [09:24:27] <bobbyz> Triskelios: wonderful. I'll have to grab it :) [09:24:37] <jmcp> g4lt-mordant: yeah, it gets me too [09:24:40] <jmcp> WickedWicky: hi there [09:24:45] <WickedWicky> you might be able to speed up things a bit by doing 'tar xf -' instead of 'tar xvf'. I had cases where writing to stdout slowed things down with 20 to 30 percent [09:25:07] <jmcp> especially over a serial console [09:25:10] <WickedWicky> yes [09:25:30] <WickedWicky> or slow framebuffers for that matter [09:25:32] * g4lt-mordant has ben using the ?zcat | tar xvf so long by now that my fingers would revolt if I didn't do it that way [09:25:58] <WickedWicky> it's just a suggestion :) [09:26:18] <bobbyz> I like good ol' gtar jxvf :) [09:26:42] * Triskelios always does tar tf first so and so is never compelled to use 'v' [09:27:00] <bobbyz> that's probably smart [09:27:30] <WickedWicky> I find it pretty useless to see how tar processes SUNWspro/whatnot/in/some/path [09:27:34] <WickedWicky> errors will go to stderr anyway [09:28:12] * g4lt-mordant throws a large muskox at bobbyz [09:28:47] <bobbyz> hehe [09:29:23] <sbahra> Gah [09:29:26] <WickedWicky> seems england is getting closer and closer to us every year... we even have their summers now [09:29:30] <sbahra> I wanted to install Solaris on my x86 box ;[ [09:29:39] <WickedWicky> what's letting you? [09:29:44] <sbahra> Not. [09:29:54] <sbahra> Well, I let a friend borrow my CDs. [09:30:04] <nrubsig> sbahra: mistake #1 [09:30:07] <sbahra> haha [09:30:10] <sbahra> Yes [09:30:19] <nrubsig> sbahra: or better.... [09:30:34] <nrubsig> sbahra: mistake #1 - mistake #Inf+ [09:30:40] *** sunmaster has quit IRC [09:31:14] <nrubsig> sbahra: float i ; for ((i=0 ; i < Inf+ ; i+=1)) ; do sbahra_do_mistake ; done [09:31:40] <sbahra> nrubsig, overflow :D [09:31:41] <WickedWicky> I lent someone my Nevermore CD once... since then I either make copies for them or I tell them to get it somewhere else [09:31:46] <nrubsig> sbahra: no. [09:31:51] <nrubsig> sbahra: floating point math [09:32:52] <sbahra> nrubsig, didn't see the float ;] [09:32:58] <nrubsig> sbahra: in fact it's an endless loop since the minimum resolution will be greater than 1 at some higher value of 'i', making the "i+=1" a no-op [09:33:23] <nrubsig> sbahra: welcome to the new shell [09:33:28] <g4lt-mordant> dd if=/dev/dsk/<insert cdrom here> of=foo.iso;cdrw -vC foo;echo "here's your CDs" [09:33:32] <sbahra> nrubsig, heh [09:33:57] <nrubsig> Shinden_: http://svn.genunix.org/repos/on/branches/ksh93/gisburn/prototype005/usr/src/lib/libshell/common/fun/mandelbrotset1 [09:34:00] <sbahra> Man [09:34:05] <sbahra> GWU deleted all my code. [09:34:36] <nrubsig> sbahra: http://www.opensolaris.org/os/project/ksh93-integration/screenshots/ksh93_mandelbrot_output001.txt [09:34:41] <nrubsig> erm [09:34:53] <nrubsig> sbahra:: http://svn.genunix.org/repos/on/branches/ksh93/gisburn/prototype005/usr/src/lib/libshell/common/fun/mandelbrotset1 [09:35:11] * nrubsig didn't intend to harass poor Shinden_ with that stuff [09:35:25] <sbahra> 1:32:83 :D [09:35:54] <sbahra> oh wow [09:35:59] <sbahra> You were serious ;] [09:36:04] <sbahra> That's pretty funny/cool. [09:36:05] <nrubsig> sbahra: minutes. On an Ultra1/140MHz [09:36:13] <gnut> Triskelios: it works! [09:36:26] <gnut> i guess a reboot does help [09:37:19] <sbahra> nrubsig, so, integer/float are actual ksh types? [09:37:34] <sbahra> nrubsig, as in, standard...? or what's going on? [09:37:41] <nrubsig> sbahra: no, they are alias for $ typeset <options> ... [09:38:47] <nrubsig> sbahra: $ alias | /usr/bin/egrep "int|flo" [09:38:48] <nrubsig> float='typeset -lE' [09:38:50] <nrubsig> integer='typeset -li' [09:39:50] <nrubsig> sbahra: and http://svn.genunix.org/repos/on/branches/ksh93/gisburn/prototype005/usr/src/lib/libshell/common/fun/termclock uses sin()/cos() functions [09:40:08] <gnut> hmm [09:40:09] <gnut> okay. [09:40:13] <gnut> now on to the next one? [09:40:28] <gnut> anyone got the broadcom 1490 (from a dell d620) wireless working? [09:41:04] <sbahra> nrubsig, aha, I see. [09:41:11] <sbahra> nrubsig, cool. Thanks for elaborating. [09:41:31] <Triskelios> gnut: cool [09:41:40] <g4lt-mordant> ahh, I now know why my bzcat|tar was taking so fucking long, my swap file was 11M :( [09:41:43] <WickedWicky> be right back folks, gotta make a walk to the office [09:41:52] *** WickedWicky has quit IRC [09:43:13] <gnut> okay. this dell one seems like it'l lbe pretty hard [09:43:14] <gnut> sleep [09:43:15] <gnut> night [09:43:16] *** gnut has left #opensolaris [09:44:07] *** linux_user400354 has joined #opensolaris [09:48:52] <g4lt-mordant> I think I've finally figured out why SXDE requires so much memory. it's the craptacular installer it uses instead of pkgadd for the SPRO parts [09:48:56] *** KermitTheFragger has joined #opensolaris [09:49:47] <g4lt-mordant> if the installer is any indication, PLEASE stay with susV packaging [09:50:03] <richlowe> there's pkg* under there. [09:50:04] <richlowe> (I think) [09:50:19] <richlowe> but the current SDX installer is a hack for SDX [09:50:29] <richlowe> snv_70 should be using Dwarf caiman, when it appears, which may or may not suck less. [09:50:54] <Triskelios> caiman is being integrated already?! [09:51:05] * g4lt-mordant is taking bets [09:51:41] <richlowe> dwarf caiman, pretty much just the GUI, I think. [09:52:26] <g4lt-mordant> richlowe, BTW, it's not just a hack for SDX, it's used by the actual SS12 stuff as weell [09:52:52] <Triskelios> same installer as Studio11 I think... [09:54:17] <g4lt-mordant> god save us [09:54:35] <g4lt-mordant> prodrg FTL [09:54:50] <richlowe> g4lt-mordant: the studio installer isn't, no. [09:54:56] <richlowe> g4lt-mordant: the dumbed down solaris+studio install is. [09:54:58] <lloy0076> Errk [09:55:10] <lloy0076> Dwarf Caiman - Existing User Data will not be preserved :P [09:55:13] <lloy0076> http://www.opensolaris.org/os/project/caiman/Dwarf/ [09:55:25] <Triskelios> g4lt-mordant: er? dwarf caiman has little to do with the current installers [09:55:26] <richlowe> g4lt-mordant: don't use the DE installer, install studio afterward. [09:55:49] <Triskelios> lloy0076: check out the demo [09:56:10] <g4lt-mordant> richlowe, that's what i'm attempting [09:56:30] <richlowe> g4lt-mordant: depending on your needs, using the CBE studio is probably nicer, too [09:56:34] <richlowe> given it's just a tarball. [09:56:37] <richlowe> opensolaris.org/os/downloads/on [09:56:53] <richlowe> slightly old (compared to the CBE), very old (compared to 'new and shiny'). [09:57:26] <trygvis> http://www.e-thug.net/pics/lollerskates.gif [09:57:50] *** peteh has joined #opensolaris [09:59:26] *** lloy0076 has left #opensolaris [10:00:14] *** dmarker has quit IRC [10:00:25] *** dmarker has joined #opensolaris [10:07:14] *** nrubsig has left #opensolaris [10:08:49] <toblun> humm, I want to run OpenSolaris on my macbook. What is the way to go, vmware fusion or parallel desktops? [10:12:17] <Chipdancer> toblun: 6 of one [10:12:36] *** MattMan has joined #opensolaris [10:14:36] *** jfndi has joined #opensolaris [10:20:47] <trygvis> toblun: http://www.good-for-nothing.org/page.php?17 [10:21:29] <asyd> S1 [10:21:33] <asyd> raa [10:23:48] <toblun> trygvis: tnx. but i need myt trackpad. =/ [10:29:05] <g4lt-mordant> toblun, what are you going on about? dude just said to try either, and you whinge about your trackpad? if you're that concerned about your trackpad and you can't use it, why are you using either? [10:29:47] <Chipdancer> how do you check whether the system time (hwclock) is UTC or localtime? [10:31:38] <Chipdancer> don't worry [10:31:43] <Chipdancer> discoveret date's -u flag [10:31:56] <Chipdancer> damn xen + localtime in domU's [10:32:15] <Chipdancer> (just trying to wrok out how to get xen to tell domU's the time in UTC, not localtime) [10:35:53] *** nostoi has quit IRC [10:37:56] <Atomdrache> This isn't a Solaris question so much as an old SPARC question. I typed setenv boot-device disk3 on a SPARCstation IPX and told it to boot. disk3 refers to sd@0,0, which is what I want. Previously boot disk3 worked fine. Well, now the display won't come up when I turn the machine on. What could cause this? [10:38:38] *** longman has joined #opensolaris [10:38:44] *** mikefut has quit IRC [10:38:47] <longman> dobry [10:39:39] <longman> ups! [10:39:46] <longman> morning! [10:39:49] <longman> :) [10:40:49] <g4lt-mordant> Atomdrache, IIRC, the IPX used a really old version of openprom, are you sue that syntax is right? [10:41:02] <Atomdrache> It didn't give me an error when I entered that. [10:41:15] <Atomdrache> Just said boot-device = disk3. [10:41:29] <Atomdrache> OBP is at v2.4 on it, I think. [10:41:40] <g4lt-mordant> that high? [10:41:58] <Atomdrache> Yeah, maybe I read it wrong. Or maybe it's been updated by whoever owned it before. [10:42:01] <Atomdrache> But why would the display refuse to come up? [10:44:41] <g4lt-mordant> have you tried otehr boot devices and they worked to bring up the screen? [10:45:14] *** linux_user400354 has quit IRC [10:45:53] <Atomdrache> I can't try anything right now because I can't get any kind of display. [10:45:58] <Atomdrache> I'm about to try a serial terminal. [10:46:17] <Atomdrache> Like, I turn it on and it just stays black. [10:46:24] <Atomdrache> It's been sitting there for like ten minutes. [10:46:33] <Atomdrache> No apparent keyboard error codes from POST. [10:46:40] <Atomdrache> I guess I'll see what the serial terminal tells me. [10:46:53] *** nikinana has joined #opensolaris [10:49:45] <Atomdrache> Crap. Nothing on that tip connection. I'll try a different serial terminal. [10:53:21] *** xsh has joined #opensolaris [10:55:15] *** Berny has joined #opensolaris [10:55:20] <Berny> morning [11:03:37] <Atomdrache> Okay, says it isn't detecting the hard drive and somehow the output device was set to "sc..". I have no idea what it did. [11:05:56] *** migi has joined #opensolaris [11:07:24] *** estibi has joined #opensolaris [11:13:01] <Atomdrache> It can't see anything on the SCSI chain, and it can't see anything on the sbus. [11:18:53] <Atomdrache> Does anybody know what might cause this? [11:19:18] <Atomdrache> Like, did some part of my motherboard just relinquish the magic smoke, or is something hosed in the NVRAM that I should fix via OBP? [11:19:53] <g4lt-mordant> what's the mac showing, numbers or 00.00.00.00.00.00 [11:20:15] <Atomdrache> Numbers. [11:20:22] <Atomdrache> Hostname and ethernet address seem to be fine. [11:20:48] <Atomdrache> I suppose by "is something hosed in the NVRAM" I mean "did I do something strange to an environmental variable or some such thing in OBP". [11:20:57] <g4lt-mordant> well, you can always stop-n to set to defaults [11:21:23] <Atomdrache> Rather than do that when there may be a problem elsewhere, I'd like to know exactly why it can't see anything on its SBUS or SCSI chain. [11:21:32] <Atomdrache> Does anythign come to mind which might cause that? [11:21:39] *** yarihm has joined #OpenSolaris [11:22:09] <Atomdrache> I exited the operating system, did setenv boot-device disk3, and typed boot. Now I can only talk to it with a serial terminal and it can't see SBUS or SCSI devices. [11:23:14] <g4lt-mordant> with an IPX, there's damn little on the SCSI chain inside. heck, there's damn little on the sbus [11:24:00] <Atomdrache> Except for a hard drive and a QFE card. [11:24:05] <Atomdrache> It saw them before. [11:24:08] <Atomdrache> Now it sees nothing. [11:24:24] <Atomdrache> I guarantee that this is not normal, [11:24:34] <Atomdrache> unless my hard drive and QFE both just simultaneously caught fire. [11:24:45] <g4lt-mordant> QFE? on a IPX? talk about pulling a race car with a moped [11:25:20] <Atomdrache> That isn't the problem, else it would have manifested last month. [11:25:48] <Atomdrache> I'm asking if you know of anything that would just make all of my SCSI and sbus devices seem to disappear and screw up my input and output device settings. [11:26:39] <g4lt-mordant> no, it should work even then, FSVO "work". TBH here, I'd say stop-n and rebuild your aliases rather than worry about what part is b0rked [11:26:50] <Atomdrache> input_device, meanwhile, has been replaced with something like ".. ...................". [11:28:01] *** WickedWicky has joined #opensolaris [11:28:08] <Atomdrache> Also, I don't know what stop-n is on this thing. [11:28:08] <g4lt-mordant> that's the second alias that's essentially garbage. I think you overflowed a memory location [11:28:24] <Atomdrache> I have to use a Macintosh as a terminal because tip wasn't working for some reason. [11:28:38] <Atomdrache> Overflowed a memory location? That'd explain a few things. [11:29:02] <g4lt-mordant> go back to screen and keyboard, since stop-n doesn't really require a working framebuffer [11:29:33] <Atomdrache> Will that require me to go rewrite all that stuff with the ethernet address and whatnot? [11:30:01] <Atomdrache> I'll do it if I have to, but this is really annoying. [11:30:19] *** Snake007uk has joined #opensolaris [11:30:25] <g4lt-mordant> no, just the aliases. only time I evere had to manually write the ethernet address was when my NVRAM battery failed [11:30:54] <Atomdrache> I did that a while ago, which is why I ask :3 [11:31:07] <Atomdrache> (hopefully the only time I need to use a hacksaw on an integrated circuit package) [11:31:26] <g4lt-mordant> Atomdrache, how much more time are you going to waste troubleshooting when you can have prbably built all th ealiases back in the time you're doing this? [11:32:38] <Atomdrache> Oi, I'm umplugging and moving it. Give me a moment. [11:33:06] <Atomdrache> I've never done that before and frankly don't know anything about resetting all the aliases, hence my hesitation. [11:33:15] *** bunker has joined #opensolaris [11:35:05] <Atomdrache> Okay, let's see if I can get a display this time now that I've fixed output-device and input-device. [11:35:18] <Atomdrache> Honestly, something overflowing and flooding OBP with crap would explain a lot of this. [11:35:43] <ofu> so U4 will be out in 9 hours? [11:36:36] *** longman has quit IRC [11:36:49] <Atomdrache> Oops, forgot setenv diag-switch? false. This'll take a little while. [11:37:17] <Atomdrache> (and I still don't know that fixing the input/output devices will let the screen come up or let me send it commands from the keyboard.) [11:37:18] <quasi> ofu: and ready for running on power 6 tommorw - http://www.theregister.co.uk/2007/08/16/software_surprise_sun_ibm/ ? ;) [11:38:35] <ofu> hooray [11:38:49] * g4lt-mordant bets heavily against that, given the status of the PPC port of opensolaris [11:39:19] <tsoome> most probably its x86 related news;) [11:39:27] <oxygene> g4lt-mordant: the ppc port is "somewhere inside sunlabs" - so maybe they actually made it work since their last interaction with the outside world? [11:39:54] * g4lt-mordant knew he could re-derail this back toward opensolaris [11:39:59] <tsoome> oxygene: what use is for os if you dont have apps [11:40:20] <oxygene> tsoome: "java solves it all" [11:40:23] <tsoome> most sw vendors do not accept even solaris x86 [11:40:38] <oxygene> though I honestly guess ibm starts selling N2s [11:43:10] <g4lt-mordant> maybe they're selling PI to IBM for a lame donkey and a can of spam [11:44:00] *** longman has joined #opensolaris [11:44:23] <Atomdrache> It's not coming up. Back to the serial console. [11:46:21] <g4lt-mordant> oxygene, a long time has passed since the last interaction of sunlabs's PPC interaction with the outside world, they could hav redesigned the PPC since then four-five times over [11:48:04] *** Pietro_S has joined #opensolaris [11:48:24] <g4lt-mordant> okay, given taht rambled even more than usual, that's natures way of telling me to GO TO BED, YOU MORON [11:49:37] *** simford has quit IRC [11:53:53] <Atomdrache> Hmm...might've been the diagnostic switch. I forgot just how slow this thing is to test the memory. [11:56:23] <Atomdrache> (Still not optimistic I can get the screen to come up so I can apply stop+N. Feeling a strong urge to go find my hammer.) [11:59:41] <Atomdrache> Meanwhile, sbus probe list has been set from 0123 (default) to nothing. Yeah, that looks just a whit suspicious. Restoring that to default. [12:00:01] <Atomdrache> I guess the way to apply stop+n with this Mac terminal is to make the right column look like the left :3 [12:00:02] *** derchris has quit IRC [12:00:09] *** derchris has joined #opensolaris [12:00:18] <Atomdrache> And there's my qfe and framebuffer. [12:00:35] *** dunc_ has joined #opensolaris [12:00:36] <Atomdrache> And there's my scsi chain. [12:00:50] <dlg> mmmscsi [12:00:59] *** dunc_ has quit IRC [12:01:00] <Atomdrache> I now know of a very vexing thing to do to somebody's OBP settings that would surely piss them off for hours. [12:12:37] *** jpdrawneek has joined #opensolaris [12:13:19] *** mikefut has joined #opensolaris [12:14:11] *** triplah_ has joined #opensolaris [12:29:35] *** estibi has quit IRC [12:31:45] *** mikefut has quit IRC [12:32:30] *** mikefut has joined #opensolaris [12:34:37] *** sparc-kly has joined #opensolaris [12:34:38] *** ChanServ sets mode: +o sparc-kly [12:36:30] *** mikefut has quit IRC [12:36:41] *** mikefut has joined #opensolaris [12:39:30] *** jpdrawneek has quit IRC [12:40:48] *** longman has quit IRC [12:48:01] *** trede has joined #opensolaris [12:57:54] *** axisys has quit IRC [12:58:20] *** kjetilho has joined #opensolaris [12:58:53] <kjetilho> hey guys, I have a problem with Jumpstart on x86 -- it can't find the mirrored root disk after installation [12:59:11] <trygvis> morn [12:59:23] <kjetilho> hi, trygvis :) [12:59:57] *** jfndi has quit IRC [13:00:07] <kjetilho> in failsafe mode, I can mount one of the sides of the mirror, and the files look fine to me. but "bootadm list-archive -R /mnt" doesn't list neither /kernel/drv/md.conf nor /etc/lvm/mddb.cf [13:00:14] *** salamanders has joined #opensolaris [13:04:18] *** deather_ has joined #opensolaris [13:10:02] *** lloy0076 has joined #opensolaris [13:10:36] <lloy0076> I want to remove all packages that match CSW* (eg by doing pkginfo | grep CSW) - is there any way I can pass pkgrm a glob to do this? [13:10:57] <asyd> pkginfo | grep | xargs ? [13:11:37] <lloy0076> *hmmm* [13:11:57] <lloy0076> I know "pkginfo [13:11:58] <lloy0076> I know that pkginfo | grep CSW | cut ??? | xargs ... would work. [13:12:10] *** sparc-kly has quit IRC [13:12:20] <lloy0076> However I forget how to tell "cut" to get the second field and the delimeter's a space [13:12:34] <lloy0076> Never mind me. [13:12:58] <lloy0076> I forgot the shell would eat the space to the -d arg ;P [13:13:05] <asyd> or awk [13:13:06] *** bengtf has quit IRC [13:13:13] <asyd> oups, too late [13:15:31] <lloy0076> asyd: The offer to help is appreciated anyway. [13:15:39] <lloy0076> And thanks oxygene too. [13:16:24] *** __william__ has joined #opensolaris [13:16:41] *** andyshack has joined #opensolaris [13:17:19] <lloy0076> Is it usual to have, presumably, ZFS squatting on about 1.5-2gb of memory? [13:17:40] *** victor__ has left #opensolaris [13:17:54] *** kaiwai has joined #opensolaris [13:18:13] <lloy0076> I have been shunting data around the place between UFS and ZFS so it's possible I've plonked that much data as a candidate for ZFS' memory caches, but it strikes me as oddly high. [13:18:52] <kaiwai> B71 is nice and stable :) [13:19:23] <__william__> hi all [13:19:42] <lloy0076> __william__: Hi ther. [13:19:47] *** __william__ has quit IRC [13:20:15] <oxygene> lloy0076: zfs' memory management works that way, yes [13:20:17] *** __william__ has joined #opensolaris [13:21:07] *** deather has quit IRC [13:21:55] *** __william__ is now known as _william_ [13:22:18] *** _william_ has left #opensolaris [13:22:38] <kaiwai> the disappearing william [13:22:58] <kaiwai> someone says hello and he retreats - very french of him [13:23:22] <andyshack> hah [13:23:37] *** jamesd has joined #opensolaris [13:23:39] *** jamesd_ has joined #opensolaris [13:25:30] *** Berny_ has joined #opensolaris [13:25:30] *** Berny has quit IRC [13:26:59] *** Berny__ has joined #opensolaris [13:26:59] *** Berny_ has quit IRC [13:27:13] *** bunker has quit IRC [13:33:13] <kaiwai> hmm, it would be nice if Sun compiled Solaris/StarOffice/whole damn thing pre-binding so that things loaded quicker [13:33:29] <dlg> solaris has prebinding? [13:34:10] *** bengtf__ has joined #opensolaris [13:34:13] *** bengtf__ is now known as bengtf [13:34:35] <asyd> there is an equivalent iirc [13:36:27] <oxygene> prebinding is mostly a matter of mangling ELF files the right way - what special support does that need? [13:37:29] <quasi> why not just do the old emacs trick of loading the proggy and dumping the memory and reloading whenever you need it? ;) [13:38:16] <kjetilho> nice thing about prebinding is that you can randomise addresses [13:38:18] <oxygene> quasi: hard to do with shared libraries, inefficient to do if you copy the shared libraries into the dump [13:38:36] <kjetilho> which makes it harder to make an exploit which works everywhere [13:39:00] *** Tpenta has quit IRC [13:39:17] <sneaker> it sounds like something gentoo would do [13:39:21] <tsoome> damn gnome should be dumped thats all [13:39:27] <oxygene> dragonfly supports checkpointing of applications.. basically a controlled core dump, with a tool to jump into it later [13:39:42] <quasi> oxygene: good thing I'm generally ok with just vi then ;) [13:39:46] <tsoome> its insane if your app will load like 200 libraries, ofc it will be slow like a hell [13:39:51] *** Kernel86_ has joined #OpenSolaris [13:39:59] <oxygene> also "resident" - load app, resolve all symbols, and right before jumping into main(), break out of it (but keep the thing in ram) [13:40:14] <oxygene> that way, when you load it again, no linking is necessary [13:40:34] *** Kernel86 has quit IRC [13:40:54] *** cypromis has joined #opensolaris [13:42:57] *** lloy0076 has left #opensolaris [13:44:49] *** apuc has joined #opensolaris [13:45:37] <kaiwai> hmm, IIRC its called direct binding [13:45:40] <kaiwai> ld -B direct [13:47:25] <quasi> oxygene: and hey presto - you've got widows all over again ;) [13:48:12] *** jfndi has joined #opensolaris [13:50:32] *** Kernel86_ has quit IRC [13:51:15] *** gobbler has joined #opensolaris [13:51:37] <kaiwai> tsoome: there are alot of applications that use heaps of files [13:52:14] <kaiwai> tsoome: one has to take into account that the application calls libraries and those libraries call libraries etc. etc. [13:52:17] *** pablomh has quit IRC [13:52:25] <kaiwai> xchat alone is bound to 20+ libraries according to ldd [13:52:42] *** Kernel86 has joined #OpenSolaris [13:53:21] <PerterB> And how much of the startup times of those applications do you think is ld binding, compared to other initialisations? Did you try measuring with dtrace? [13:53:52] *** jamesd has quit IRC [13:54:09] <Tempt> Evenin' all [13:54:14] <seanmcg> pldd `pgrep -x xchat` | wc -l == 42 :( [13:54:28] <seanmcg> afternoon all [13:55:15] *** jamesd_ has quit IRC [13:55:42] <andyshack> i am so lost. ive got a fileserver and a webserver. the home dirs are on the file server and automount? on the webserver. should i be researching the path from the webserver to the fileserver home to add a user that exists on the file server to the webserver to have their home dir automount ? [13:56:42] <Cyrille> probably yes. [13:56:57] <andyshack> thanks Cyrille [13:57:10] <Cyrille> but I'm not sure I understood the question. [13:57:23] <PerterB> me either :) [13:57:27] <andyshack> yeah neither do i as im short on terminoligy [13:57:58] *** Kernel86_ has joined #OpenSolaris [13:58:06] <Cyrille> say you have fileserver:/export/home/<people> automounted on your webserver at /home/<people> and you want to add a user to the webserver, then its home should be specified as /home/someone. [13:58:18] <andyshack> 2 computers, need to add a user. 1 is a dedicated fileserver, the other a dedicated webserver. [13:58:19] <Cyrille> at least that's what I think I understood you were asking. [13:58:32] <andyshack> the file server holds use /home [13:59:05] <andyshack> the webserver "automounts" (is that correct?) eg /home/andy from fileserver:/home/andy when i log in automatically [13:59:17] *** Hystrix has joined #opensolaris [14:00:01] <andyshack> or : whats it called when my home dir follows me around different machines ? [14:00:17] <Cyrille> needy [14:00:22] <Cyrille> ;-) [14:00:28] <andyshack> most deff [14:01:11] <andyshack> its already setup and working, im jsut trying to work out how it works so i can replicate users about the place. [14:01:28] <andyshack> can you point me to some related keyords that can get me on track with searching ? [14:02:34] <kjetilho> andyshack: the automounting is not related to logging in, but that someone tries to access the directory [14:03:00] <kjetilho> so others can do "cd /home/andy" and it will automount if needed [14:03:23] <andyshack> so is this simply working by having the home dir on the file server available for export and adding a user on the webserver with its home dir pointing to /net/fileserver ? [14:03:39] <richlowe> /net mounts happen via the automounter. [14:03:54] <richlowe> I think you want automountd(1M) as far as docs go. [14:04:09] <andyshack> ok cheers rich [14:04:12] *** Kernel86_ has quit IRC [14:05:04] *** Kernel86_ has joined #OpenSolaris [14:06:10] *** lgodard has joined #opensolaris [14:06:13] <lgodard> hi all [14:06:22] <lgodard> new in your world [14:06:35] *** pablomh has joined #OpenSolaris [14:06:41] <lgodard> i've a a SunBlade 100 that i want to reinstall with opensolaris [14:06:56] <lgodard> it does not boot on the dvd [14:07:26] <lgodard> i retreived the image and burned the iso on a rw dvd disk [14:07:44] <lgodard> any hint to boot on it and install opensolaris ? [14:09:55] <asyd> probe ide-all ; boot cdrom ? [14:10:04] <hile_> it's entirely possible that the DVD can't read the media [14:10:13] *** master_of_master has joined #opensolaris [14:10:55] <lgodard> asyd: i don't know how to check if it tries to boot on cd/dvd [14:10:55] *** Kernel86 has quit IRC [14:11:10] <lgodard> hile_: because of a +RW ? [14:11:13] <hile_> did you try boot cdrom at OBP [14:11:21] <hile_> no, because they're old-ass drives that are sometimes finicky [14:11:33] <lgodard> hile_: OBP ? [14:11:48] <hile_> OpenBoot PROM [14:12:29] <lgodard> you mean, turn on with the dvd ? yes I tried [14:13:12] *** Gropi has joined #opensolaris [14:14:36] *** Gman has quit IRC [14:14:49] <hile_> no.. [14:14:56] <hile_> i mean instruct the machine to boot from that device [14:15:39] <lgodard> hile_: it is a new amchine for me. How can i instruct the machine to boot from it ? [14:15:52] <lgodard> hile_: hit a special key at startup ? [14:16:03] *** Kernel86_ has quit IRC [14:16:33] <Pietro_S> lgodard: doesn't have blade 100 only cdroms? [14:17:26] <lgodard> it is written DVDrom in the front [14:17:37] <lgodard> but cdrom on screen [14:18:46] *** lloy0076 has joined #opensolaris [14:19:20] *** lloy0076 has left #opensolaris [14:19:22] *** Yamazaki-kun has joined #opensolaris [14:19:56] <lgodard> Pietro_S: probing /pci@1f,0 Device d ide disk cdrom [14:20:39] <lgodard> Pietro_S: but is is written dvdrom on the machine [14:21:26] <lgodard> btw, i can try with a cd instead of a dvd but, do i really need to burn the 6 CD that are on the site ? any netinstall available as on linux ? [14:23:01] *** master_o1_master has quit IRC [14:24:42] *** vmlemon has joined #opensolaris [14:24:55] *** kaiwai has quit IRC [14:25:08] <Pietro_S> there is jumpstart [14:25:18] *** kaiwai has joined #opensolaris [14:25:38] *** Gman has joined #opensolaris [14:25:54] *** davidsl has joined #opensolaris [14:26:13] <Hystrix> you'll need a running solaris installation for jumpstarting [14:26:47] *** laca has joined #opensolaris [14:26:49] <Pietro_S> lgodard: did you download sparc version of dvd? [14:26:53] <lgodard> it is a gift from Sun Grenoble (France) for the OpenOffice.org project (QA team) [14:27:14] <kaiwai> Grenoble? [14:27:18] <lgodard> Pietro_S: yes sparc version - but does not seem even to try booting on it [14:27:43] <asyd> lgodard: /join #opensolaris-fr :p [14:27:54] <lgodard> there is a solaris install on it but do not have any login/password and need to reinstall the machine [14:28:02] <lgodard> asyd: good shoot [14:28:14] <asyd> you're member of OOo qa team ? with Sophie ? [14:28:17] <kaiwai> lgodard: you need some help with solaris? [14:28:29] <davidsl> Well, it seems that blastwave xchat and JDS don't thump each other. [14:28:31] <davidsl> heh [14:28:42] <Pietro_S> I would try to download fist cd and try if it will boot (if you don't care about bandwith) [14:28:49] <kaiwai> davidsl: meh, I prefer to compile my own rather than use blastwave [14:28:53] <Pietro_S> davidsl: try xchat from SFE repository [14:29:07] <davidsl> (btw, my other nic is lloy0076 :P) [14:29:17] <lgodard> Pietro_S: yes, i thought on this - do not have any cd right here but will do [14:29:36] <lgodard> asyd: yes, with Sophie [14:29:44] <davidsl> kaiwai, I don't mind compiling my own, but sometimes I can't be bothered. [14:29:53] <lgodard> but, more for automated tests and tinderbox [14:30:21] <kaiwai> davidsl: true, the problem is if you download blastwave there is the dependencies and crap that get pulled down as well [14:30:30] <davidsl> kaiwai, Oh, and I'm an ex-Debian GNU/Linux user... [14:30:46] <davidsl> kaiwai, So downloading the earth doesn't phase me. [14:30:55] <kaiwai> davidsl: I'm an ex Amiga/FreeBSD/Mac user :) [14:31:03] <kaiwai> oops, forgot BBC Micro :P [14:31:05] <Pietro_S> lgodard: try to to hit stop-a when booting and then boot cdrom [14:31:19] <hile_> but does having to maintain multiple copies of libraries phase you, davidsl? [14:31:29] <davidsl> kaiwai, The funniest one I had though was on FreeBSD 4.X - I wanted a TEXT ONLY chat client, so did "make mutt" in /usr/ports/.../mutt only to find it wanted to compile XFree86. [14:31:34] <davidsl> hile_, Not really. [14:31:47] <Pietro_S> blastwaves are coward they use gcc on evverything ;-) [14:31:53] <davidsl> hile_, On a production system, I'd compile things with the same compilers if I had to. [14:32:06] <trygvis> Pietro_S: eh, wtf? no, they don't. gcc is the exception [14:32:08] <davidsl> hile_, But this is my home system and if it falls in a heap, it's like, meh. Fix it. [14:32:09] <kaiwai> davidsl: ah, never used it, I prefer using XBitchX [14:32:25] <davidsl> kaiwai, Sorry, that was an e-mail client. [14:32:34] <kaiwai> Pietro_S: I'd love to see Studio everywhere, but everything it compiles, things break quicker than my virginity was [14:32:41] * davidsl since when has mutt been a chat client - I'm losing my mind [14:33:02] <davidsl> kaiwai, Well, hopefully we're not all like you're virginity!!! [14:33:04] <kaiwai> davidsl: I thought it was kinda strange mentioning mutt - does it have Xorg frontend? [14:33:05] * davidsl ducks for cover [14:33:17] <kaiwai> lol [14:33:18] <davidsl> kaiwai, It had some silly dependency in it. [14:33:25] <Pietro_S> trygvis: really? I didn't check it for long time, but I saw lot's of gcc binaries [14:33:46] <kaiwai> davidsl: one thing I did like was the ports - heaps of software compileable and up to date [14:33:59] <davidsl> kaiwai, True, I didn't mind that. [14:34:22] <davidsl> kaiwai, Though sometimes it's painful to have to wait for something to compile. I started using ports when things took ages to compile (unlike these days). [14:34:49] <lgodard> Pietro_S: FATAL: OpenBoot initilization sequence prematurely terminated - FATAL: system is not bootable, boot command is disabled [14:35:06] <kaiwai> lol, reminds me of my PIII 550 compiling KDE on my FreeBSD 4.x machine which took over 2 days to compile from scratch :) [14:35:21] <kaiwai> then once compiled I found that there was an update to a bug that stopped me from mounting removable media :P [14:35:28] <davidsl> lol [14:35:35] <trygvis> Pietro_S: binaries as in *built* with gcc or gcc packages? [14:35:48] *** hohum has quit IRC [14:35:50] <kaiwai> trygvis: built with gcc [14:35:53] *** halton has quit IRC [14:36:05] <kaiwai> damn, when is sun going to update their gcc included with solaris? [14:36:09] <Pietro_S> built by gcc [14:36:19] <davidsl> kaiwai, You could compile it yerself :P [14:36:59] <kaiwai> davidsl: tried that and it died [14:36:59] * davidsl brings the channel to a standstill [14:37:04] <davidsl> heh [14:37:14] <davidsl> You mean Sun Studio wouldn't compile GCC? [14:37:16] <kaiwai> well, it compiled but the problem is that it can't find the dependent *.so's [14:37:17] <davidsl> The sacrilege! [14:37:29] <kaiwai> and I refused to use LD_LIBRARY_PATH [14:37:34] <Pietro_S> trygvis: I thought that everything build by gcc is the reason why blastwave has special copy of gnome libraries [14:37:45] <kaiwai> nope, compiled it using gcc 3.4.3 included with solaris [14:37:53] <oxygene> kaiwai: compile with LDFLAGS="-R$rightpath" in configure's environment [14:38:02] <davidsl> kaiwai: I generally have to fiddle the default library paths to make sure blastwave doesn't fall in a heap. [14:38:03] <kaiwai> thats what I used [14:38:39] <Pietro_S> lgodard: sry I won't help you much I don't have around any sparc machine to check right command :-( [14:38:47] *** axisys has joined #opensolaris [14:38:49] <kaiwai> oxygene: yeap, I had it setup, -L/usr/local/lib -R/usr/local/lib (destination for where the *.so would eventually end up) [14:38:52] <kaiwai> it failed to work [14:39:01] <lgodard> ok, no problem - thanks a lot Pietro_S [14:39:27] <oxygene> kaiwai: hmm.. gcc has the speciality that it bootstraps compilers - are you sure that they were still in the build environment of the final build? [14:39:37] <oxygene> kaiwai: I don't know off-hand how to modify the later passes [14:39:51] <kjetilho> I wish ld had an option which said, add to rpath all -L paths which were actually used [14:40:01] <Tempt> gcc has the speciality that it should go away. [14:40:18] <Tempt> Nearly all compilers are self hosting. [14:40:23] <kaiwai> oxygene: the main reason why I'm compiling 4.2.1 is so I can compile KDE which isn't compiling with the gcc thats included with solaris [14:40:41] <kaiwai> I mean, if there were precompiled KDE 3.5.7 binaries I wouldn't try to compile it myself [14:40:50] <davidsl> kaiwai: Aren't there packages for that (although not official) [14:40:52] <davidsl> ?? [14:40:57] <oxygene> kaiwai: hmm.. the one in pmpkg (the complete one is in the gnustep hierarchy) worked the last time I tried it [14:41:00] <Tempt> You think a different gcc will make any real difference? [14:41:10] <kaiwai> pmpkg? [14:41:14] <oxygene> kaiwai: and I can't think of any special flags I set [14:41:17] <kaiwai> tempt: I'll give anaything a go [14:41:22] <trygvis> Pietro_S: no, that is because it can't depend on what's on the solaris box [14:41:28] <Pietro_S> kaiwai: newer gcc is also in SFE [14:41:29] <oxygene> kaiwai: my ports-alike project [14:41:35] <kaiwai> Tempt: unfortunately when one compiles on Solaris, things break [14:42:14] <oxygene> though if I really leave solaris (and currently it looks a lot like it), it might bitrot over time if no-one picks it up (and I wouldn't bet on that) *shrug* [14:42:15] <davidsl> kaiwai: Don't you mean, when you compile GNU specific stuff, it tends to fall apart on Solaris... [14:42:39] <oxygene> davidsl: gnu projects tend to be portable - the non-gnu opensource stuff has much more problems on non-linux [14:42:40] <davidsl> oxygene: You're thinking about leaving? [14:42:46] <Pietro_S> trygvis: thanks for the explanation, and sorry that I mis-acuse blastwave [14:42:49] <kaiwai> davidsl: na, more "anything" fails to compile; wine 0.9.43 as a prime example [14:42:54] <kaiwai> ntdll not compiling [14:43:12] <trygvis> http://theforrester.wordpress.com/2007/08/13/the-100-oldest-domains-on-the-internet/ [14:43:13] <oxygene> for 0.9.43 I have a (tiny) patch in the tree, too ;) [14:43:42] *** apuc has quit IRC [14:43:49] *** apuc has joined #opensolaris [14:44:02] <oxygene> davidsl: right now I desktop is setup like this: solaris, failsafe xterm X session, qemu with dragonfly in it, ssh -X into it, startkde - so kde now manages the qemu window it's emulated in [14:44:19] <oxygene> davidsl: and I have to say this setup works better for me in several regards than solaris proper [14:44:28] <Pietro_S> trygvis: is there any way how can I see patches which blastwave use to compile with studio? [14:44:29] <kaiwai> oxygene: who ever released 0.9.43 without checking whether it compiles needs to have an eletric cattle prod up the jaxy [14:45:06] <davidsl> hmmm [14:45:07] <oxygene> kaiwai: it compiles.. this time it's the fault of solaris: too many headers with various symbols polluted (without _ prefix) [14:45:29] <oxygene> davidsl: "your mileage may vary" of course [14:45:46] <davidsl> oxygene: You mean http://www.dragonflybsd.org/index.shtml [14:45:52] <oxygene> davidsl: yes [14:45:59] <kaiwai> oxygene: thats usual [14:46:03] <kaiwai> and Xft headers missing [14:46:21] <kaiwai> and some dick deciding to litter directories from hear to eternity with bloody links and crap [14:46:22] <trygvis> Pietro_S: ask on #blastwave, but I assume they use the latest patches for sun studio 11 [14:46:26] <kaiwai> *here [14:46:34] <oxygene> this time wine stumbled over solaris defining a struct called "version" [14:46:43] <kaiwai> yeap, thats the one [14:46:59] <kaiwai> oh, and sound is screwed up too - play something using it, and find that sound is buggered till the next reboot [14:47:05] <oxygene> kaiwai: LDFLAGS="-R$TARGET/lib -L/usr/sfw/lib -R/usr/sfw/lib -L/usr/X11R6/lib -R/usr/X11R6/lib" CPPFLAGS="-I/usr/sfw/include -I/usr/X11R6/include" [14:47:15] <oxygene> kaiwai: with that, wine finds practically everything here [14:47:40] <kaiwai> thats ok, got that end all sorted, but the Xft headers missing is annoying [14:48:11] <kaiwai> when B69 is installed the link points to a non-existant directory [14:49:55] <Pietro_S> trygvis: no I meant patches for other software, which they use to force to compile with suncc [14:51:08] <kaiwai> in an ideal world, everytime something doesn't work with suncc the maintain of that said programmer is electricuted [14:51:13] <kaiwai> *maintainer [14:51:44] <oxygene> uh, so where's the complete guide to solaris and suncc bugs? [14:52:11] <kaiwai> well, they're not necessarily bugs, more...features, yes, features :) [14:52:27] <oxygene> like struct version? [14:52:50] <Gekkko[PDA]> hai [14:52:56] *** davidsl has quit IRC [14:52:57] <oxygene> or backtrace() [14:53:00] <kaiwai> yeap, of course there are loads of stupid tihings in solaris [14:53:16] <kaiwai> like I said, look at all the f*cking stupid links everywhere for instance [14:53:17] <oxygene> or #define SS [14:53:50] <trygvis> Pietro_S: for most of the packages you can see the patches in the files/ directory of the package at svn.blastwave.org [14:53:53] <trygvis> Pietro_S: http://svn.blastwave.org/trac/browser/csw/trunk [14:54:05] *** andyshack has quit IRC [15:06:54] *** lloy0076 has joined #opensolaris [15:10:41] *** lloy0076 has left #opensolaris [15:12:27] *** pablomh has quit IRC [15:22:54] *** cypromis has quit IRC [15:26:59] *** hohum has joined #OpenSolaris [15:28:48] *** lgodard has left #opensolaris [15:30:53] *** cypromis has joined #opensolaris [15:33:17] <kaiwai> damn, whose running the dlc.sun.com/osol? [15:33:22] <kaiwai> a blind, pissed midget? [15:38:21] *** halton has joined #opensolaris [15:40:19] <tsp> I hope replacing my U5's psu made it more reliable, even if said PSU is from a junker :) [15:43:14] *** triplah_ has quit IRC [15:44:06] *** Yamazaki-kun has quit IRC [15:44:17] *** apuc has quit IRC [15:45:47] *** laca is now known as lacaAFK [15:46:09] <Gekkko[PDA]> firewire or usb? [15:46:10] <FrostCS> tsp, simply means there is now a higher chance of a redundant failure. [15:46:38] <FrostCS> tunafish or chicken? [15:47:22] <Gekkko[PDA]> tuna chicken sa/lad [15:49:12] <kaiwai> interesting to see the linux/vmware jihad occuring over at osnews.com [15:49:18] <kaiwai> http://www.osnews.com/comment.php?news_id=18462 [15:49:49] <FrostCS> oh wait.. were in #opensolaris [15:49:56] <FrostCS> it's alright, we all make mistakes kaiwai [15:50:49] <kaiwai> FrostCS: no, just seeing the crap in the Linux community and thinking "thank god this commity doesn't have people like that" [15:50:54] <kaiwai> then you popped up [15:51:05] <oxygene> heh [15:51:55] <FrostCS> we aren't a linux chat, and we aren't a vmware chat.. [15:52:31] <FrostCS> therefore, it's irrelevant ... [15:53:20] <oxygene> kaiwai: nasty boy, you just killed all opensolaris discussions in here by your mention of linux and vmware! [15:53:31] <oxygene> .o0O oh, the talk before was about tuna and chicken [15:54:08] <kaiwai> hmm, my cat likes both [15:54:16] <FrostCS> well, there was an non-descriptive question, so I simply followed it up with another :-) [15:54:51] <oxygene> that non-descriptive question didn't have anything to do with opensolaris, either :-) [15:55:51] <kjetilho> anyone know how mirrored root works with GRUB? [15:56:14] <oxygene> kjetilho: simply choose one or the other in grub? [15:56:32] <kjetilho> no, I mean, my install can't find the md device. [15:56:33] <FrostCS> oxygene, of course it did, it's a question regarding which is better for lunch while working with opensolaris. [15:56:47] <kjetilho> and I wonder if reason is that the boot-archive has no MD configuration? [15:57:33] <oxygene> might be - did you update it? [15:57:43] *** sartek has joined #opensolaris [15:57:45] <kjetilho> yes, I tried that [15:58:12] *** sommerfeld has joined #opensolaris [15:58:13] *** ChanServ sets mode: +o sommerfeld [16:00:53] *** splunk has joined #opensolaris [16:01:24] <splunk> Hi All...I'm running Solaris 10 and I'm having serious issues mounting my usb hard drive...I was able to mount my usb thumb drive no problem and its even formatted with FAT 32....the system sees the device but no go...when I try to mount it says mount /dev/dsk/c7t0d0p0:c is not a DOS file system [16:04:21] <splunk> any takers?? [16:06:31] <kjetilho> hrm... why is boot_archive needed, anyway? [16:07:25] <kjetilho> I guess it's due to it being easier to switch the root when it's a pure RAM disk? ie. converting from mounting one side of the mirror to the md device is hard? [16:07:51] <splunk> um [16:07:58] <kjetilho> if so, the MD configuration needs to be inside it, though? [16:08:19] <splunk> why is this so difficult to be honest....I was able to fdisk my thumb drive..and mounted just as I noted before just a different disk and it worked fine [16:09:36] *** logic has quit IRC [16:09:40] *** logic_ has joined #opensolaris [16:09:53] <Berny__> splunk: try the next partition on that drive? [16:09:56] <splunk> i did [16:09:57] <splunk> haha [16:10:08] <splunk> theres only one partition though [16:10:14] <splunk> i only created one...i can go back and create two [16:10:18] <splunk> FAT 32 and say SOl12 [16:10:24] <splunk> or maybe Unix = UFS ? [16:10:38] <Berny__> well if you created just one, then there is probably just one [16:11:00] <Berny__> did you partition and format that drive under solaris? [16:11:03] <FrostCS> you might want to take a look at http://blogs.sun.com/SteveJay/date/20041130 [16:11:06] *** Berny__ is now known as Berny [16:11:08] <FrostCS> it's a similar issue [16:11:14] <splunk> yes I formated and FDISK under Solaris [16:11:51] <splunk> hmm [16:12:12] <FrostCS> if I don't answer the phone once.. why do people keep calling? jeez [16:12:36] <LeftWing> FrostCS: Perhaps you're the Go-To Person? [16:12:41] <FrostCS> I hate people who ignore my voicemail [16:13:14] <FrostCS> well now they can only talk to my voicemail since I perma-forward them to my vm. [16:13:56] <kaiwai> hmm, is it possible to tarball a complete OS, throw it onto a dvd, create a zfsroot boot then copy it back? [16:13:56] <kjetilho> anyone here doing mirrored root on Solaris/x86? [16:14:17] <Berny> kjetilho: yes [16:14:29] <FrostCS> probably a lot of people kjetilho [16:14:45] <kjetilho> Berny: can you check what bootadm list-archive says? [16:14:52] <kjetilho> FrostCS: I would imagine that, too [16:15:09] <kjetilho> Berny: e.g., is /etc/lvm/mddb.cf listed? [16:15:16] <Berny> works fine as long nope [16:15:19] <Berny> not listetd [16:15:25] <kjetilho> and not /kernel/drv/md.conf either? [16:15:34] <Berny> nope [16:15:36] <kjetilho> hmm, how does this thing work, then? [16:15:40] <Berny> just kernel [16:15:54] <Berny> dunno if that includes the whole dir tree under kernel though [16:16:20] <splunk> no go [16:16:26] <kjetilho> hmm. [16:16:49] <splunk> i've formated the drive and fdisk using [16:16:52] <splunk> format -e [16:16:56] <Berny> but where it boots from is also in /boot/solaris/bootenv.rc [16:17:21] *** sstallion has joined #opensolaris [16:17:25] <kjetilho> ah, good point [16:17:41] <renihs> gruml, netapp fas 270 and no default passwords around ...great [16:17:56] <splunk> why am I able to mount my usb thumb drive so easily [16:18:08] <kjetilho> Berny: did you do the install using Jumpstart, or did you mirror root after the fact? [16:18:19] <Berny> mirrored after install [16:18:53] <elektronkind> kjetilho: boot archive is like initrd in linux [16:18:59] <elektronkind> serves the same funciton [16:19:00] <Berny> borked it with liveupgrade, cause /sbin/biosdev fscked up the identical disks and so grub failed to get it right [16:19:04] <FrostCS> splunk, I can't find any details on it for opensolaris, it looks like it's a solaris-ish bug only.. [16:19:05] <kjetilho> I couldn't get mirrored root to work out of the box on SPARC either, but there I found a workaround for the postinstall script :/ [16:19:38] <Berny> so i fixed that by editing bootenv.rc, booted the fresh update, mirrored again [16:19:40] <elektronkind> kjetilho: are you running metaroot after creating the mirror for /? [16:19:42] <kjetilho> elektronkind: I guess this means it only needs /kernel/drv/md.conf and not /etc/lvm/mddb.cf to start up MD sufficiently, then? [16:19:50] <Berny> haven't tried lu again yet [16:19:51] *** jHoNDoE has quit IRC [16:19:52] <FrostCS> isn't mirroring a root like a simple command with SVM? [16:19:55] <kjetilho> elektronkind: the work is done by Jumpstart [16:20:19] <kjetilho> but yes, it runs metaroot or something like it [16:20:24] <Berny> FrostCS: 4 or 5 commands iirc [16:20:29] *** EchoBinary has joined #opensolaris [16:20:51] <Berny> metainit twice, lockfs, metaroot it should be [16:21:03] <splunk> yea [16:21:05] <kaiwai> anyone used zfs for root/boot? [16:21:07] <splunk> its really pissing me off [16:21:28] <kjetilho> Berny: so you're suggesting to do it in my S99firstboot script instead? :-/ [16:21:29] <Berny> splunk: whats rmformat -l telling you? [16:21:51] <Berny> kjetilho: i would do it manually if its just one box [16:22:05] *** Tpenta has joined #opensolaris [16:22:10] <splunk> um it sees the device [16:22:19] <splunk> thats not a problem [16:22:35] <splunk> just can't mount it [16:22:45] <splunk> so pissed [16:22:52] <Berny> ah wait [16:22:57] <asyd> are you sure it's not mounted on /media? [16:22:58] <Berny> thats sol10 you said? [16:22:59] <kjetilho> Berny: well... I might end up having to do that, but it's potentially a common occurence to reinstall the boxes [16:23:04] <splunk> yea [16:23:27] <kaiwai> does the installer detect zfs and able to install on there? [16:23:34] *** ChanServ sets mode: +o Tpenta [16:23:56] <Berny> i vaguely remember an issue where the header in the fat32 partition was "wrong" which let sol assume it's not a fat32 fs on there [16:24:14] <splunk> mount: /dev/dsk/c7t0d0p0:c is not a DOS filesystem [16:24:15] <Berny> kaiwai: not yet as far as i know [16:24:27] <kaiwai> :-( [16:24:32] <splunk> what i'm doing is ....mount -F pcfs /dev/dsk/c7t0d0p0:c /dsk1 [16:24:36] <kaiwai> trying to find a way to move from UFS to ZFS [16:24:57] <splunk> but my usb thumb drive is FAT 32 [16:24:58] <FrostCS> you can LU fine through ZFS [16:24:59] <tsoome> its a bit early:) [16:25:10] <splunk> and I mounted that not problem [16:25:39] <FrostCS> splunk usb memory sticks, and hard drives are a bit different [16:25:46] <splunk> yea [16:25:48] <FrostCS> most usb memory sticks are also automounted. [16:25:51] <splunk> yup [16:25:58] <splunk> that too [16:26:10] <FrostCS> yet you've said you have deleted the partition in solaris, and reformatted it? [16:26:21] <kaiwai> FrostCS: so if I created a ZFS partition, loaded up the installer, it would detect it? [16:26:36] *** baijiutong has quit IRC [16:26:39] <FrostCS> kaiwai, LU doesn't use the installer [16:26:43] <splunk> even when I trun a cfgadm it sees it on usb4/6 [16:27:00] <Berny> splunk: opensolaris.org/jive/thread.jspa?messageID=95115 [16:27:03] <kaiwai> hmm, ok [16:27:09] <Berny> that could be your issue? [16:27:15] <splunk> let me take a look [16:27:57] <splunk> haha and I do have a seagate..let me keep reading [16:30:00] <kaiwai> FrostCS: I assume with LU you need to have Solaris already installed on it :( [16:30:29] <FrostCS> kaiwai, well that's usually why they call it LIVE [16:30:30] *** vmlemon_ has joined #opensolaris [16:30:53] <kaiwai> so I wouldn't be able to use it as a way to install solaris onto a zfs partition? [16:31:06] *** Hystrix has quit IRC [16:31:09] <splunk> hmm [16:31:12] <splunk> i've got no idea [16:31:18] <splunk> i'm not a SOlaris exper [16:31:19] <splunk> tt [16:31:31] *** vmlemon has quit IRC [16:31:33] *** vmlemon_ is now known as vmlemon [16:31:53] <FrostCS> kaiwai, you need to read up on LU procedures, you are walking around in circles. [16:32:04] <kaiwai> ok [16:32:07] <kaiwai> I'll try it tomorrow [16:32:14] <FrostCS> splunk, you've already formatted the drive under solaris, and it still won't read? is that correct? [16:32:37] <Berny> splunk: if there are no data on this disk just flat it over again... [16:32:44] <FrostCS> I'd be interested to see if that would still be the case under SX, since there is a lot of new drivers included. [16:33:03] <FrostCS> what is the make/model of the drive also? [16:33:41] *** Chihan has joined #OpenSolaris [16:33:47] * Berny has a 300gb usb drive... fdisk'd under snv, two partitions... one fat32 and one ufs... both mount fine all by themselves [16:34:05] <renihs> hmm whats the syntax for ipf to dump/show the rules? no access to a ipf man currently? :( [16:34:14] *** jfndi has quit IRC [16:34:19] <FrostCS> I'll add to Berny's, that if your reformatting it, be sure to delete all partitions prior to the reformat. [16:35:05] *** axisys has quit IRC [16:35:15] *** axisys has joined #opensolaris [16:35:27] *** Chihan has quit IRC [16:35:46] <Berny> prolly be worth to zero the first few blocks of the disk with dd :-) [16:35:54] <Berny> just to make sure :-) [16:37:50] *** EchoBinary has joined #opensolaris [16:38:26] <splunk> ????? [16:38:37] <splunk> Berny: I've formated and fdisked [16:38:42] <splunk> here i'll bring up the tables agin [16:39:20] <splunk> right now theres a WIN95 FAT parition 60% and a DOS 16 paritition 20% [16:39:52] <FrostCS> why? heh [16:40:05] <splunk> because thats what I was playing with at the time [16:40:11] <splunk> what FS should I try instead? [16:40:14] <splunk> UNIX? [16:40:18] *** hile_ has quit IRC [16:40:18] <splunk> UFS? [16:40:56] <FrostCS> I thought you had it all as FAT32? [16:41:01] <splunk> now if I try to mount the drive to /mnt it says "is already mounted or /mnt is busy..i guess I can't moult multiple devices to a directory containing mounted devices [16:41:08] <splunk> I did I just changed it out recently [16:41:11] <splunk> i can make it all FAT 32 again.... [16:41:15] *** hile_ has joined #Opensolaris [16:41:33] <FrostCS> wipe those ones off, then format to fat32 [16:41:41] <splunk> ok done [16:41:49] <splunk> ACTIVE WIN95 FAT 32 100% [16:42:08] <kjetilho> okay, I verified that /kernel/drv/md.conf inside the root_archive looks healthy. [16:42:25] <FrostCS> and are you going to say which make/model this drive is splunk? [16:42:51] <splunk> yup.... [16:43:41] <kjetilho> what on earth is making it panic? :-/ [16:43:44] <kjetilho> panic[cpu0]/thread=fffffffffbc22dc0: cannot mount root path /pseudo/md@0:0,0,blk [16:43:52] <kaiwai> hmm [16:43:58] <kaiwai> must be me ner the machine :) [16:44:05] <Berny> kjetilho: it's telling you it can't mount the root device [16:44:06] <kaiwai> *kaiwai moves away* [16:44:07] <splunk> cfgadm -l -s cols=ap_id:infot [16:44:14] <kjetilho> Berny: :-) [16:44:16] <splunk> MF seaate [16:44:24] <splunk> MF: Seagate....FREEAGENT PRO [16:44:28] <splunk> 700 GIGS [16:44:42] <splunk> NCONFIGS: 1 [16:45:14] <Berny> kjetilho: boot off the first mirror half... (fix bootenv.rc from failsafe and update bootarchive is what i did) [16:46:18] <kjetilho> well yeah, but I want Jumpstart to work. it's freshly installed, no data to worry about [16:46:55] <splunk> same thing [16:47:07] <Berny> well then give it another shot... haven't done mirror with jumpstart yet... [16:47:17] <splunk> i'm about to implode [16:47:34] <kjetilho> Berny: I've tried many times, adding more debug output each time [16:47:40] *** _Megaf|Sleep is now known as _Megaf [16:47:51] <Berny> maybe fat32 doesn't like 700gigs ... [16:48:13] <Berny> wasn't there a magical border where even windooze refused to create a fat32 fs? [16:48:20] <kjetilho> Berny: could you do me another favour and do: [16:48:21] <kjetilho> /boot/solaris/bin/root_archive unpack /mnt/platform/i86pc/boot_archive /tmp/T [16:48:24] <oxygene> I think fat32 can be up to 2tb - but not all implementations support it [16:48:38] <FrostCS> Seagate wouldn't have been dumb enough to prevent the first block from being rewritten would they? [16:48:41] <kjetilho> grep bootpath /tmp/T/boot/solaris/bootenv.rc [16:48:48] <splunk> who knows [16:48:58] <splunk> solaris says its formatted [16:49:24] <FrostCS> you can use dd to output the firstblock to a bin file, to make sure it has the FAT32 signature.. [16:49:28] <kjetilho> Berny: eh, remove the /mnt in the first command [16:49:30] *** phimic has quit IRC [16:50:01] <FrostCS> you could always give ufs a shot if you want also.. [16:50:54] <FrostCS> It is starting to sound like it's a seagate issue, I know the drive would work flawlessly internally, which is what is annoying [16:51:14] <splunk> i don't kno whow to format it as ufs [16:51:19] <splunk> i think I had an issue with that [16:51:34] <Berny> setprop bootpath /pci@0,0/pci-ide@5/ide@0/cmdk@0,0:a [16:51:37] <FrostCS> or it could be that none of the pcfs fixes have been implemented for Solaris 10 yet.. [16:51:53] <kjetilho> Berny: ok, good. the raw primary side of the mirror [16:52:05] <Berny> yepp [16:52:18] <Berny> splunk: format, create a slice and newfs [16:53:09] <splunk> newfs /dev/rdsk/c7t0d0p0 [16:53:20] <splunk> Can not determine pairtition size: invalid argument [16:53:32] *** yippi has joined #opensolaris [16:53:39] <Berny> did i say format and create a slice first? :-) [16:53:55] <Berny> though i think that would need a solaris partition [16:54:13] <splunk> not a Solaris guru...create a slice...create a paritition you mean? [16:54:21] <Berny> not quite [16:54:26] <splunk> ha [16:54:29] <splunk> yea didnt think so [16:54:34] <Berny> with fdisk you create a partition on x86 [16:54:41] <splunk> correct [16:55:00] <Berny> inside that partition you create slices with format (assuming you have made your partition type solaris) [16:55:14] <Berny> that slice you would then newfs [16:55:22] <splunk> i'm using format -e [16:55:27] <splunk> then I pick the drive [16:55:32] <splunk> then it gives me a list of choices [16:55:45] <splunk> "Format Menu" [16:56:40] <Berny> press p [16:56:58] <Berny> and p again to see what you have [16:56:58] <splunk> "Please run fdisk first" [16:57:18] <Berny> you have no valid partitions i'd say :-) [16:57:38] <splunk> alright so lets blow away the current FAT 32 patition [16:57:52] <splunk> and create a Unix partition 100% [16:57:53] <_Megaf> Hello all, good morning [16:58:25] <Berny> either that 700g is too much... or for whatever reason your box didn't re-read the partition table after modifying it with fdisk [16:58:39] <splunk> pl [16:58:51] <splunk> ok...I made a Sol partition [16:58:53] <splunk> I Pressed P.... [16:59:06] <splunk> now which area am I looked for [16:59:11] <splunk> theres 9 rows [16:59:21] <Berny> yes [16:59:22] <splunk> mostly unassigned...backup and a boot [16:59:35] <Berny> 0 for the first [16:59:41] <splunk> 0-9 [16:59:45] <splunk> so 10 ha [16:59:53] <Berny> type 0 [17:00:07] <splunk> 4.24 gb [17:00:11] <splunk> flag : wm [17:00:18] <splunk> unassigned [17:00:38] <splunk> cylinders 68-621 [17:01:06] <Berny> type 0 [17:01:09] <Berny> to modify it [17:01:21] <Berny> id tag: unassigned [17:01:26] <Berny> flags wm [17:01:35] <Berny> starting cyl 0 [17:01:58] <splunk> ok [17:02:07] <splunk> part 0 is now 0-553 [17:02:07] *** Shinden_ is now known as Shinden [17:02:18] <splunk> flag still wm and unassigned [17:02:20] <Berny> size <take the end cyl from backup, add one [17:02:34] <Berny> enter that number followed by a c [17:02:49] <splunk> 9723 [17:02:55] <splunk> backup is currently 0-9722 [17:03:00] <Berny> press p again, you should now have one big slice 0 [17:03:01] <splunk> 74.48gb [17:03:15] <splunk> wait [17:03:17] <splunk> uh [17:03:32] *** tsoome has quit IRC [17:03:47] <splunk> you lost me on the backup part [17:03:50] <splunk> part 0 is done [17:03:56] <splunk> backup is part 2 [17:03:58] <Berny> leave backup alone [17:04:01] <splunk> ok [17:04:10] <Berny> thats just info for you how big your disk is [17:04:20] <Berny> all the other set to size 0 [17:04:29] <splunk> negative [17:04:34] <splunk> part 0 is still only 4.24gb [17:04:48] <splunk> part 7 which is 622-9721 is 69 gb [17:04:54] <splunk> theres a Boot Tag [17:04:57] <splunk> which 7mbs [17:05:06] <splunk> Part 2 which is backup is 74 gigs [17:05:11] <Berny> ok again, what does the backup line say for cylinders? [17:05:19] <splunk> 0-9722 [17:05:24] <Berny> ok, type 0 [17:05:26] *** apuc has joined #opensolaris [17:05:31] <splunk> k [17:05:33] <Berny> id tag: unassigned [17:05:39] <Berny> flags wm [17:05:44] <Berny> starting cyl 0 [17:05:51] <splunk> enter partition size [17:05:52] <splunk> ? [17:05:52] <Berny> size 9723c [17:06:14] <splunk> ok [17:06:20] <splunk> part 0 is now 74 gigs [17:06:27] *** RaD|Tz has joined #opensolaris [17:06:39] <Berny> for slice 1,3,4,5,6,7,8,9 you do the same, but enter 0 for start cyl and 0c for size [17:06:54] <kjetilho> splunk: clearly the disklabel is not spanning the whole disk [17:07:22] <splunk> yea i was wondering what th ehell [17:07:25] <splunk> why is that [17:07:25] <Berny> kjetilho: oh yeah right, didn't notice this before [17:07:26] *** wamty has joined #opensolaris [17:07:30] <wamty> anyne has a u20 ? [17:07:32] <splunk> "sarcasm" [17:07:36] <wamty> or better? [17:07:38] <wamty> :) [17:08:02] * Berny has a v880 and a sf4800 a few blade 1500 etc [17:08:33] <FrostCS> or better. [17:08:36] *** sartek has quit IRC [17:08:41] <Berny> splunk: did format or fdisk complain about not being able to determine the size or geometry of the drive? [17:08:54] <wamty> FrostCS, what you got? [17:08:57] <wamty> Berny, cool! [17:09:06] <splunk> nope [17:09:17] <splunk> ok now I hit p..... everything is cleared out....except of 0 and backup [17:09:25] <splunk> 0 is 0-9722 and 2 is 0-9722 [17:09:26] <Berny> type label [17:09:34] <splunk> both are 74.48 gigs [17:09:36] <splunk> ok [17:09:40] <wamty> Berny, v880 better than U20? [17:09:40] <FrostCS> wamty, it's easier to type "or better" then list things out. [17:09:45] <splunk> SMI or EFI? [17:09:51] <Berny> smi [17:09:55] <wamty> yea.. but could you show me please [17:10:03] <splunk> Warning: no backup labels [17:10:12] <Berny> wamty: look at suns system handbook :-) [17:10:21] <Berny> splunk: thats fine [17:10:34] <splunk> ok [17:10:37] <Berny> now newfs /dev/rdsk/c7t0d0s0 [17:11:06] <FrostCS> wamty, look through sun's entry and mid-range line for the past 10 years.. and just figure most of it. [17:11:19] <splunk> hmmm looks like its formatting [17:11:36] <Berny> hehe FrostCS you got a copy of my setup? :-) [17:11:39] <wamty> Berny, fascinating [17:11:40] *** salamanders has quit IRC [17:11:43] <splunk> real quick though , why did it only show in the partition table only two parts with only 74 gigs..all the other parts except boot were 0 [17:12:00] <wamty> Berny, mind showing me a uname please [17:12:07] <FrostCS> Berny, I like to think of it as an Antique shop, with a few new goodies. [17:12:08] <wamty> I would love to see this uname -m [17:12:08] <splunk> this is a 700 gig drive, i guess after it formats it should recreate the cylinders [17:12:18] *** alphakiller has joined #opensolaris [17:12:22] <Berny> splunk: cause your fdisk partition might only be 74gig? [17:12:25] *** cmihai has joined #OpenSolaris [17:12:27] <cmihai> hi [17:12:37] <splunk> i made it 100% though [17:12:43] <splunk> ok thats done now [17:12:51] <Berny> now try to mount it [17:13:06] <wamty> Berny ? [17:13:11] *** jlc has quit IRC [17:13:21] <splunk> mount -F ufs ? or pcfs? [17:13:26] <Berny> wamty: it's just saying SunOS mars 5.1 sun4u sparc SUNW,SunFire [17:13:27] <splunk> or just do mount -F [17:13:39] <Berny> mount /dev/dsk/c7t0d0s0 /mnt [17:13:52] <Berny> if /mnt isn't busy [17:14:12] *** stevel has joined #opensolaris [17:14:13] *** ChanServ sets mode: +o stevel [17:14:16] <Berny> hi steve [17:14:23] <stevel> 'morning berny [17:14:36] <alphakiller> I was thinking on porting an OS to playstation 3 [17:14:41] <splunk> mount: /dev/dsk/c7t0d0p0 is not this fstype [17:14:42] <Berny> .oO(other side of the world...) [17:14:49] <alphakiller> opensolaris may be a good candidate [17:14:51] <cmihai> alphakiller: PS3 runs Linux [17:14:54] <Berny> mount -F ufs /dev/dsk/c7t0d0s0 /mnt [17:14:56] <cmihai> Start there [17:14:56] <alphakiller> yeah I kbow [17:14:56] *** WickedWicky has quit IRC [17:14:57] <splunk> k [17:15:07] <alphakiller> *know [17:15:11] <alphakiller> and ? [17:15:14] <cmihai> alphakiller: NetBSD PS2 uses Linux PS2 kit / loader. [17:15:30] <alphakiller> PS3 [17:15:31] <alphakiller> ... [17:15:32] <alphakiller> not ps2 [17:15:43] <alphakiller> ps3 is far different from ps2 [17:15:45] <cmihai> I know, I was saying you'll need to use Linux regardless. [17:15:50] <FrostCS> alphakiller, see the powerpc mailing list at opensolaris.org [17:15:53] *** mega has joined #opensolaris [17:16:01] <alphakiller> and seems pretty easy to port to powerpc .... [17:16:06] <alphakiller> because the hypervisor stuff ... [17:16:14] <FrostCS> SPUFS still needs to be ported over to opensolaris codebase. [17:16:20] <alphakiller> it seems pretty well documented ... [17:16:23] <cmihai> Besides, porting OSOL to the CPU architecture is a PITA, there was an old Solaris that ran PowerPC.. and there are efforts now... but I can't even get a kernel to boot :P [17:16:42] <alphakiller> FrostCS, I've coded a simple PS3 Hypervisor Library [17:16:43] <splunk> why s0 ? [17:16:59] <splunk> c7t0d0p0 or c7t0d0s0 ? [17:17:05] <alphakiller> dunno if that might help guys [17:17:14] <cmihai> See the Polaris blokes [17:17:22] <alphakiller> with such Hypervisor, it's pretty easy to mount SPU FS [17:17:24] <FrostCS> s0 is sparc, p0 is x86 [17:17:35] <splunk> yea [17:17:36] <alphakiller> map SPE [17:17:37] <splunk> roge [17:17:38] <splunk> rha [17:17:38] <alphakiller> etc ... [17:17:53] <Berny> splunk: if that doesn't work somethings wrong with that disk... either solaris can't handle it because it get the geometry wrong or the drive just sucks 8-)s0 [17:17:56] <Berny> s0 [17:18:04] <alphakiller> I'm getting involved in so many projects ... [17:18:09] *** cmihai has left #OpenSolaris [17:18:14] <Berny> p0 is for the fdisk partitions s0 is for the slice [17:18:17] <alphakiller> dunno wheter to get involved in opensolaris or not [17:18:27] <Berny> .oO(partitions inside a solaris partition) [17:18:28] <alphakiller> and if would be interesting on getting it working under ps3 [17:18:35] <FrostCS> alphakiller, if you can integrate spufs into the code repository, you shouldn't have any troubles, most of the code for powerpc is available on opensolaris.org, and last I knew it would load on the ps3 [17:19:36] <alphakiller> FrostCS: main problems are in ps3buf code [17:19:38] <alphakiller> etc [17:19:39] <splunk> mount -F ufs /dev/dsk/c7t0d0p0 is not this fstype [17:19:42] <alphakiller> fb ... [17:19:47] *** hohum has quit IRC [17:19:54] <FrostCS> I talked with Arnd Bergmann about what would need to be done a while ago, but sadly I have no time for it at the moment. [17:21:08] <FrostCS> there is a #cell channel on oftc.net you might be interested in joining.. between that and the powerpc community (lists) you should be able to get some good information on the subject. [17:21:11] <Berny> splunk: c7t0ds0! [17:21:24] <Berny> splunk: c7t0d0s0! [17:21:48] <splunk> that worked [17:21:49] <splunk> :) [17:21:51] <splunk> yaya [17:21:54] <Berny> see [17:21:58] <splunk> THE FUNK man [17:22:12] <Berny> now go and find out why it only reports 74gigs [17:22:12] <alphakiller> hehe [17:22:13] <alphakiller> thanks ! [17:22:16] <splunk> lol [17:22:17] <splunk> hahahh [17:22:24] <Berny> na seriously [17:22:35] <FrostCS> 74gigs is a decent compromise. [17:22:41] <Berny> you sure you didn't miss a 0 when entering the size in fdisk? [17:22:56] <splunk> 100% i type [17:22:58] <splunk> lol [17:23:13] *** alphakiller is now known as alphakiller|away [17:23:23] <splunk> i don't understand why all the cylinders are funked up [17:23:24] <Berny> just checking :-) [17:24:44] <splunk> hmmm in fdisk it shows Type Solaris2 start 2 cyclinders en 91199 legnth 91198 [17:24:46] <splunk> 100% [17:24:55] <splunk> something doenst seem right [17:25:28] *** tsoome has joined #opensolaris [17:25:42] <Berny> hmm [17:26:05] *** baijiutong has joined #opensolaris [17:26:46] <splunk> hmm [17:26:52] <splunk> looking at the partition table now [17:26:54] *** jlc has joined #opensolaris [17:26:57] <splunk> backup is 698 gigs [17:27:11] <splunk> so I have to change part 0 to 0-91193 [17:27:16] <splunk> cuz its only 9722 [17:27:22] <Berny> hmm, so slice zero should be the same size [17:27:23] <splunk> its only reading the first slice??? [17:27:26] <splunk> yea [17:27:46] <Berny> does start end end in format tell you the same for 0 and 2? [17:28:33] <splunk> part 0 ..0-9723 part 2 0-01193 [17:28:39] *** sommerfeld has quit IRC [17:28:40] <splunk> *91193 [17:28:45] <Berny> HA! [17:28:54] <Berny> make that the same for slice 0 [17:29:01] <Berny> then newfs again [17:29:06] <Berny> but umount /mnt before [17:29:20] <splunk> um [17:29:22] *** axle_512 has left #opensolaris [17:29:29] <splunk> i'm in partition > [17:29:35] <splunk> i select part 0 [17:29:44] <splunk> unassigned [17:29:47] <splunk> wm [17:29:52] <splunk> 0 [17:29:57] <splunk> now it wants partition size [17:30:05] <Berny> 91193c [17:30:30] <splunk> BOOM [17:30:35] <Berny> label [17:30:38] <FrostCS> it exploded? [17:30:40] <Berny> exit [17:30:42] <FrostCS> oh noes.. [17:30:43] <Berny> newfs [17:30:55] <Shinden> what dyndns client is fros OS ? [17:31:02] <Shinden> for [17:32:06] <FrostCS> last I knew, DynDNS wasn't dependent on an OS... [17:32:24] * Berny lets his router do the dyndns 8-) [17:32:31] *** sartek has joined #opensolaris [17:32:32] <stevel> there are perl scripts too [17:33:20] *** jlc has quit IRC [17:33:27] <splunk> i'm all about IOS [17:33:31] <splunk> working on my CCSP [17:33:39] <splunk> decided to go the security track [17:33:47] <splunk> CCNA was actually kinda tough but I nailed it [17:33:56] <FrostCS> I am all about R&R [17:34:22] <FrostCS> I thought lounging around all day would be murder.. but I seem to be pulling it off alright. [17:34:50] <splunk> so newfs automatically creates the UFS fs on the disk huh? [17:34:51] <Berny> rest and recreation? :> [17:35:20] <Berny> yeah newfs is nice wrapper for mkfs [17:35:41] <Berny> just a little easier to remember the options :> [17:35:44] <splunk> man the last two days i've been working on this [17:35:51] <splunk> couldnt get this garbage to work for the life of me [17:35:58] <splunk> i was pulling my pubes out [17:36:09] <splunk> i was so mad [17:37:33] *** mega_ has joined #opensolaris [17:38:26] *** hohum has joined #OpenSolaris [17:40:20] <splunk> Berny: I really appreciate your help [17:42:16] *** hile_ has quit IRC [17:42:18] <g4lt-mordant> Berny, for us, it's more rm -rf /* and relaxation ;P [17:42:44] <the-decider> g4lt-mordant: that'll miss all your .* directories -- you don't wnat to leave anything laying about ;) [17:43:06] <Berny> :> [17:43:25] <g4lt-mordant> the-decider, right, but somply rm -rf /doesn't work, so you have to do it in two swipes [17:43:47] <the-decider> just make sure you don't remove /lib/libc.so until you're done. [17:44:06] <the-decider> so remove the .* directories first ;) [17:44:38] *** WickedWicky has joined #opensolaris [17:44:53] <the-decider> once your /bin/rm -fr /* is running, libc is already loaded and won't really disappear until it exists... [17:45:00] <the-decider> er exits. [17:45:06] <the-decider> (yup, done that before.) [17:47:55] *** lacaAFK is now known as laca [17:47:59] <WickedWicky> heya laca [17:48:07] <laca> hi WickedWicky [17:48:24] <splunk> now i'm trying to understand the difference between a slice and a partition [17:49:57] *** nikinana has quit IRC [17:50:48] *** xmas has joined #opensolaris [17:50:50] <WickedWicky> Windows and Linux put a file system on a partition. (open)solaris puts file systems on pieces (slices) of this partition. [17:51:01] <PerterB> splunk: the fact that "format" calls slices partitions doesn't help ;) [17:51:17] <WickedWicky> so on solaris you can have more than one file system within a partition where on linux you have a filesystem on the partition [17:52:08] <PerterB> WickedWicky: right, and you can only really have one solaris partition per disk [17:52:30] <WickedWicky> yeps [17:53:25] <splunk> hmm [17:53:58] <splunk> i wonder now that the disk is UFS it will get automounted next time [17:54:11] <splunk> i think i have to edit some kernal.usb.conf file or something [17:54:25] <WickedWicky> |---------------------| <-- disk [17:54:26] <WickedWicky> |---------|-----------| <---- two partitions [17:54:26] <WickedWicky> |--|--|---|-----------| <-- three slices in the first partition [17:54:33] <WickedWicky> something like that [17:54:56] <splunk> ahh i see [17:55:03] *** mega has quit IRC [17:55:23] <splunk> within in FDISK tho it shows just one partition .... [17:55:24] <wamty> Berny, you own a v880 ? [17:55:30] <wamty> why would you need such a beast, berny? [17:55:33] <elektronkind> look out! WickedWicky is laying down the ascii art knowledge! [17:55:40] <WickedWicky> lol, no [17:56:07] *** sommerfeld has joined #opensolaris [17:56:08] *** ChanServ sets mode: +o sommerfeld [17:56:16] <WickedWicky> heya sommerfeld! [17:56:54] <splunk> im gonan try firewire see if I run into any problems cuz i have 4 of these external 700 gig drives [17:56:59] <splunk> its almost like a mini data center [17:56:59] <splunk> lol [17:57:07] <splunk> "sneakernet data center " :) [17:57:12] <wamty> Berny ? [17:57:16] <WickedWicky> memory problems are solved. The real problem was 1 DDR400 dimm and 2x pc3200 giving timing issues. It's just that they started to play up after installing Opensolaris. I guess Linux never had to address more than 2GB of memory or something [17:57:47] <splunk> whats the primary different between Solaris and opensolaris [17:58:58] <WickedWicky> that one I wont touch [17:59:02] <Berny> wamty: i never said i owned them [17:59:37] <splunk> hahaha [17:59:47] <g4lt-mordant> WickedWicky, given that it's in the opeensolaris FAQ, good call [17:59:47] <splunk> i was just messin i'll do my own research on that [17:59:49] <PerterB> splunk: be careful with firewire... on some versions of solaris there's a known bug that can hang your machine (it was fixed around b54 I think) [17:59:59] <splunk> hmmm [18:00:18] *** _william_ has joined #opensolaris [18:00:28] <splunk> i'm in a test environment so I have no choice just to be sure, thanks for the heads up [18:00:34] *** _Megaf has quit IRC [18:01:05] <PerterB> http://bugs.opensolaris.org/view_bug.do?bug_id=6445725 [18:01:38] <wamty> Berny, what do you use them for? [18:01:41] <wamty> what do they run please? :) [18:02:02] <g4lt-mordant> hmm, since I'm reinstalling anyways, do I need all the KDC and that crud to use kerberos security on localhost only? [18:02:07] <Berny> sol10 as i said earlier [18:02:19] <Berny> keeping all our precious data safe... [18:03:16] <wamty> nice [18:03:20] <splunk> Berny: I forgot to ask, was there a reason why I left the backup slice a lone? [18:03:30] <splunk> on the disk [18:03:44] <Berny> backup is the "pointer" to the whole disk [18:03:48] *** sfire||mouse has joined #opensolaris [18:03:57] <splunk> w/o that it won't see anything I assume? [18:04:12] <splunk> i can reduce the size of it though? [18:04:18] <Berny> no [18:04:36] <Berny> it's more a virtual slice so the os knows the whole size of the disk [18:04:41] <splunk> yea [18:04:45] <Berny> doesn't take up any real space [18:04:47] <PerterB> just leave slice 2 alone... various things assume it overlaps the whole disk ("backup" slice is slightly misleading) [18:05:15] <splunk> gotcha [18:05:18] <splunk> thanks [18:05:24] <Berny> write something to it and you f*ck up all other data on the disk :-) [18:05:27] <splunk> theres a boot slice too [18:05:43] <splunk> only if I wanted to boot the disk [18:05:46] <splunk> which may bee neede din the future [18:05:47] <Berny> that's only needed if you intend to boot from it [18:05:52] <splunk> ya [18:05:55] *** apuc has quit IRC [18:06:10] * Berny moves himself home [18:08:41] *** vmlemon has quit IRC [18:08:56] *** gobbler has quit IRC [18:09:22] *** hile_ has joined #opensolaris [18:14:11] *** cmihai has joined #OpenSolaris [18:14:22] *** cmihai has left #OpenSolaris [18:15:47] *** apuc has joined #opensolaris [18:18:53] *** RaD|Tz has quit IRC [18:20:09] *** gdamore has quit IRC [18:20:24] *** StylusEater_Work has joined #opensolaris [18:20:40] *** gdamore has joined #opensolaris [18:20:47] *** gdamor2 has joined #opensolaris [18:21:27] *** Cyrille has quit IRC [18:21:48] <the-decider> df [18:21:50] *** gdamor2 has left #opensolaris [18:21:54] <the-decider> (oopz) [18:22:15] *** yippi has quit IRC [18:22:22] *** mikefut has quit IRC [18:22:52] <g4lt-mordant> da' fail ;P [18:23:18] *** gdamor2 has joined #opensolaris [18:23:23] *** gdamore has left #opensolaris [18:23:24] *** gdamor2 has left #opensolaris [18:23:26] *** gdamor2 has joined #opensolaris [18:23:41] *** gdamor2 has left #opensolaris [18:23:45] <seanmcg> gdamor2: still with the name confusion ?-) [18:24:49] *** comay has joined #opensolaris [18:27:36] *** halton has quit IRC [18:27:56] *** peteh has quit IRC [18:33:44] *** gdamore has joined #opensolaris [18:37:11] *** mega_ has quit IRC [18:37:57] *** WickedWicky has quit IRC [18:43:09] *** sfire||mouse has quit IRC [18:45:47] *** vmlemon has joined #opensolaris [18:47:24] *** RaD|Tz has joined #opensolaris [18:51:21] *** EchoBinary has quit IRC [18:51:53] *** MattMan has quit IRC [18:53:03] *** bengtf has quit IRC [19:00:30] *** bubbva has joined #opensolaris [19:02:24] *** Snake007uk has quit IRC [19:04:51] *** sartek has quit IRC [19:07:34] *** Shiv__ has joined #opensolaris [19:07:51] *** yarihm has quit IRC [19:09:07] *** apersson has joined #opensolaris [19:12:05] *** Megaf has joined #OpenSolaris [19:13:51] *** wamty has quit IRC [19:14:32] *** bengtf__ has joined #opensolaris [19:14:33] *** bengtf__ is now known as bengtf [19:16:18] *** apuc has quit IRC [19:18:02] *** migi has quit IRC [19:18:40] *** mrdeviant has joined #opensolaris [19:24:18] *** cypromis has quit IRC [19:25:09] *** Shiv__ has quit IRC [19:25:33] *** jfndi has joined #opensolaris [19:27:08] *** jfndi has quit IRC [19:27:51] *** KermitTheFragger has quit IRC [19:28:06] *** Megaf is now known as _Megaf [19:28:23] *** Megaf has joined #OpenSolaris [19:29:17] *** _Megaf has quit IRC [19:30:41] *** Jondice has quit IRC [19:32:07] *** xmas has quit IRC [19:32:53] *** Megaf has quit IRC [19:36:09] * quasi sits back and starts waiting for http://www.sun.com/aboutsun/media/presskits/2007-0816/ [19:36:28] *** cub- has joined #opensolaris [19:37:02] <cub-> how do I view the memory usage in sol 10 x86 ? [19:37:50] <StylusEater_Work> free? [19:37:53] <StylusEater_Work> guessing [19:37:59] <quasi> echo ::memstat |mdb -k [19:38:26] <quasi> StylusEater_Work: wrong [19:38:38] <StylusEater_Work> quasi: I can read [19:38:43] <seanmcg> by process ? prstat -s size is handy [19:38:55] <quasi> other options are vmstat and top [19:39:06] <seanmcg> don't use top. [19:39:09] <StylusEater_Work> free ... just never ported? [19:39:23] <quasi> StylusEater_Work: no need [19:39:27] <seanmcg> not on solaris, its doesn't do things right..... [19:39:35] <seanmcg> top that it. [19:39:57] *** NikolaVeber has joined #opensolaris [19:40:04] <quasi> seanmcg: it is a decent approximation when you're in a hurry [19:40:31] <cub-> top is the most user friendly :\ [19:41:15] <seanmcg> quasi: in a hurry prstat is there, top may not be. [19:41:31] <seanmcg> userfriendly != correct (at times)... [19:41:51] <quasi> seanmcg: true - prstat just isn't all that good at showing system totals [19:42:31] <seanmcg> neither is top - the values it show may not be true... [19:42:35] * quasi really, really doesn't want to defend top [19:42:47] <seanmcg> ;-) [19:43:34] *** WickedWicky has joined #opensolaris [19:44:06] <WickedWicky> Aloa [19:49:33] *** cypromis has joined #opensolaris [19:51:12] *** vmlemon_ has joined #opensolaris [19:52:44] *** vmlemon has quit IRC [19:52:49] *** vmlemon_ is now known as vmlemon [19:53:25] *** sz has joined #opensolaris [19:56:51] <Pietro_S> laca: can you check if you have ctags-01-destdir.diff [19:57:02] <Pietro_S> in the SFE patche dir? [19:57:15] * laca svn updates [19:57:22] <laca> nope [19:57:24] <Pietro_S> looks like doug forgot to upload patch [19:57:47] <laca> yes [19:57:53] <laca> email him [19:58:40] <Pietro_S> directly to him or to desktop-discuss? [19:58:56] <laca> directly to him, /me thinks [20:00:17] <CIA-17> as145665: 6589627 Need to create inc.flg or reg.flg to list files that do not reside in the current directory [20:04:03] *** perlmonk has quit IRC [20:09:41] *** Megaf_ has joined #OpenSolaris [20:10:21] *** WickedWicky has quit IRC [20:11:47] *** WickedWicky has joined #opensolaris [20:11:53] *** dunc has joined #opensolaris [20:13:35] *** sfire||mouse has joined #opensolaris [20:13:56] *** kaiwai has quit IRC [20:14:19] <splunk> hmmm [20:15:13] *** alfism has joined #opensolaris [20:15:32] <timeless> could someone glance at man find.1 for me? [20:15:55] <timeless> do you see -typec (note the lack of a space) [20:16:00] <timeless> and is that intentional? [20:16:34] *** stevel changes topic to "Latest SXCE 69 | Latest ON 71 | Starter kits: http://get.opensolaris.org" [20:16:41] <stevel> onnv_71 is available [20:16:46] * timeless pokes stevel [20:16:51] *** alanc-away is now known as alanc [20:16:54] * stevel jabs timeless [20:16:56] <timeless> or alanc :) [20:17:14] <stevel> seems like a typo [20:17:24] <timeless> ok, good [20:18:31] <timeless> ok, stupid question, how does one search bugs.opensolaris for this? [20:18:32] *** deather_ is now known as deather [20:18:36] <timeless> i'm trying kewords: man 1 find type [20:18:45] <timeless> but that's not exactly returning a small set (1270 results) [20:19:40] <alanc> the source does have -type in bold, c in italic, just looks like the space between the two got lost [20:20:03] <timeless> yeah, i'm sure i'd have figured that part out [20:20:11] <timeless> but i needed to be sure it was still buggy in a current build [20:20:12] <splunk> I guess now I have to figure out to have my usb devices automounted post reboot [20:20:15] <timeless> since i'm using some old version [20:20:26] <timeless> SunOS swift 5.11 snv_53 i86pc i386 i86pc [20:20:30] <stevel> 6593043 [20:20:34] <timeless> otoh, it isn't 20 releases old yet :) [20:20:48] <stevel> 'There should be blank space between "-type" option and value on find(1) manpage' [20:20:49] *** vmlemon has quit IRC [20:21:08] <timeless> stevel: um, is that bug a secret? [20:21:26] <timeless> http://bugs.opensolaris.org/search.do;jsessionid=6fe4f311948f244ec6272b67fc900?process=1&type=&sortBy=relevance&bugStatus=&perPage=10&bugId=6593043&keyword=&textSearch=&category=&subcategory=&since= [20:21:32] <timeless> Your search returned 0 results. [20:21:38] <stevel> timeless: it contains highly confidential information about our secret NSA backdoor via the find.1 manpage [20:21:54] <stevel> most likely solaris/manpage isn't one of the published categories [20:21:55] <timeless> stevel: yeah well [20:21:56] <stevel> which is odd [20:22:03] <wesolows> Those wacky NSA guys, always putting stuff in the last place you'd ever look! [20:22:16] <stevel> weird. b.o.o. lists solaris/manpage as a category [20:22:36] <timeless> i've even found bugs about man find in the search [20:22:49] <stevel> ohh [20:22:51] <timeless> stevel: fwiw, oo isn't the only group w/ secrets [20:22:52] <stevel> it was only filed yesterday [20:22:56] <stevel> it probably hasn't been synced up to b.o.o. yet [20:23:10] <stevel> b.o.o. only syncs up once a day [20:23:11] <timeless> i got a bug filed in our public tracker w/ a private flag [20:23:14] <timeless> so i can't see it [20:23:24] <timeless> the bug's about something not remotely confidential [20:23:30] <timeless> and i'd sooner wontfix it [20:23:33] <timeless> but i can't see it, so i can't [20:24:05] <timeless> stevel: heh, that's amusing, what are the odds that two people would notice w/in days [20:24:12] * the-decider thought that was the find . -type f -contents secret -sendto gonzales command. [20:24:17] <stevel> timeless: 6593043 has been accepted (today) - so it should be fixed soon [20:24:26] <stevel> timeless: yeah nice timing :) [20:24:42] * timeless is like that [20:24:55] <wesolows> the-decider: No, the bug is that you have to type that for those things to happen. [20:25:22] <the-decider> ah cool, so they've got the autoreporting kernel module in place. [20:25:23] <splunk> anyone know which file in solaris 10 I need to edit to automount usb devices post reboot? [20:25:38] * timeless tries to remember how trof syntax works [20:26:04] <stevel> cat /dev/random | mailx askdoj at usdoj dot gov [20:26:10] <stevel> though i guess cat will never exit [20:26:18] <stevel> might want to do 'dd' instead :-P [20:26:24] <holcomb> in a for loop [20:26:28] * stevel nods [20:26:31] <WickedWicky> it will be one heck of a long mail you'll be making there :P [20:26:42] <stevel> seriously - i'm gonna need a whole lotta tubes for that mail [20:27:08] <holcomb> did you receive large emails consisting of what seemd to be random characters? [20:27:11] <holcomb> ag: i don't recall [20:28:56] *** vmlemon has joined #opensolaris [20:30:53] <timeless> stevel/alanc: so, back to man find.1 [20:31:00] <timeless> look at Complex Expressions [20:31:03] <timeless> do you have: [20:31:07] <timeless> 1)(expression) [20:31:11] <timeless> 2)!expression [20:31:13] <Pietro_S> splunk: I don't know how it works in Solaris 10, but in opensolaris you just insert and see usb devive on your favourite desktop and in /media/your_usb_disk [20:31:17] <timeless> 3) expression[-a] expression [20:31:21] <timeless> 4) expression-oexpression [20:31:43] <timeless> note the lack of spaces between 1)(, 2)!, v. the space after 3) and 4) [20:34:02] * timeless reads man ksh.1 and tries to figure out why there's a line running through the middle of a few parageaphs [20:34:37] *** nrubsig has joined #opensolaris [20:34:38] *** ChanServ sets mode: +o nrubsig [20:34:50] <splunk> hmmm [20:34:55] <timeless> also, lex.1, search for intyyless(int n) [20:35:04] <timeless> compare int yymore(void) [20:35:16] * nrubsig looks around.... [20:35:26] <nrubsig> Did anyone saw kupfer today ? [20:35:27] <timeless> nrubsig: you missed a summon :) [20:35:28] <splunk> well took me probably 16 hours to figure out how to mount my external usb hard drive...had to reformat the bitch as ufs and redo all the cyclinders and slices it was a nightmere [20:35:36] <nrubsig> timeless: when ? [20:35:51] <timeless> about 2 moments before you joined :) [20:35:51] <nrubsig> !summon -history .*gisburn.* [20:36:00] <timeless> i was asking about the old ksh.1 man page :) [20:36:49] * timeless eyes mailx.1 [20:36:57] <nrubsig> timeless: which problems do you have ? [20:37:04] <splunk> well on a side not, i Just installed Firefox...and i am forced to initate the app in CLI which is fine but i want to build an icon on the desktop [20:37:09] <timeless> why would that have a space before +? [20:37:11] <splunk> so I'm lookin at the docs... [20:37:23] <timeless> splunk: eh?? [20:37:37] <splunk> and it wants me to add a desktop file to the system [20:37:40] <splunk> yea... [20:37:46] * timeless reads man ksh.1 and tries to figure out why there's a line running through the middle of a few paragraphs [20:37:49] *** Jondice has joined #opensolaris [20:38:03] <nrubsig> timeless: bad processing maybe ? [20:38:06] *** RaD|Tz has quit IRC [20:38:13] <splunk> for one I have to fix the display issue first [20:38:16] <timeless> nrubsig: i'm mostly wondering at the intent [20:38:36] *** saguti has joined #opensolaris [20:38:47] <splunk> ..../opt/sfw/lib/firefox/ ./fireox GTK-warning **: Cannot open display [20:38:47] <nrubsig> timeless: ksh.1 is processed using a hair-raising style from (AFAIK) SolBook---->troff--->ascii [20:39:11] <timeless> sounds like fun [20:39:12] *** [R]itchie has quit IRC [20:39:14] <splunk> so then I'm forced to do a DISPLAY:0 export DISPLAY echo $DISDPLAY [20:39:19] <splunk> and I have no idea whys [20:39:24] <nrubsig> splunk: did you pay the firefox license fee for closed-os installations ? [20:39:24] <timeless> splunk: did you forget to enable xforwarding when you used ssh? [20:39:50] *** yarihm has joined #OpenSolaris [20:39:53] <splunk> wasn't aware of any fees..I just downloaded it [20:39:56] <splunk> and installed [20:40:01] * nrubsig giggles madly [20:40:04] <splunk> but running the app is tough [20:40:17] * timeless chuckles [20:40:19] *** [R]itchie has joined #opensolaris [20:40:29] <nrubsig> splunk: you now own mozilla.org your firstborn and left arm!! [20:40:43] <splunk> ok [20:40:48] <splunk> i'll give them a rain check for now [20:40:58] <splunk> but flag it for follow up:) [20:41:01] <timeless> nrubsig =~ s/own/owe/ [20:41:12] <nrubsig> *tilt* [20:41:15] <splunk> i thought it was free [20:41:23] <nrubsig> gisburn ----> english ----> garbage [20:41:26] <splunk> lol [20:41:35] <timeless> VERIFIED [20:41:37] <splunk> any idea why i get this cannot open display issue when I try to run the app from CLI [20:41:41] <nrubsig> splunk: yes, it's free. I am just trying to scare you with little success. [20:41:47] <splunk> yea [20:41:48] <splunk> lol [20:41:54] <timeless> splunk: most apps want a DISPLAY [20:42:07] <timeless> what kind of cli is this? [20:42:08] <splunk> theres my conceptual boundary [20:42:16] <splunk> im just running it in terminal [20:42:17] <splunk> lol [20:42:17] <timeless> note that firefox is fully capable of remoting [20:42:18] <nrubsig> splunk: but... maybe... human baby meat... would be tasty [20:42:25] <timeless> e.g. i'll run firefox from solaris to windows [20:42:25] <saguti> Hello. What is the righ procedure to participate in solving a bug? [20:42:34] <nrubsig> saguti: which kind of bug ? [20:42:37] <timeless> saguti: do you mean writing a fix? [20:42:47] <timeless> there's a checkbox when you file a bug saying you want to workon it :) [20:42:50] <Pietro_S> hmm, will ever be build 70 released or is that a secret plan how to force people to use bfu? [20:43:18] <timeless> pietro: you have something against 71? [20:43:54] <Pietro_S> I meant sxce, I don't want to install ON alone ... [20:44:08] <splunk> so I guess those no quick fix for this display problem huh? [20:44:12] <tsp> wow, suspend on a sparc really isn't that useful :) [20:44:21] <timeless> splunk: i still don't understand [20:44:28] <timeless> is this a solaris console login [20:44:31] <timeless> or an xterm [20:44:34] <timeless> or xrvt [20:44:41] <timeless> or gnome-terminal, or konsole or ... [20:44:45] <splunk> I installed firefox....and I"m forced to run it in the terminal [20:44:48] <nrubsig> splunk: erm... the DISPLAY variable is needed to teach the GUI apps. where it should display the windows [20:44:57] <splunk> yea how do you go about doing that? [20:45:08] <timeless> what app is this CLI?? [20:45:21] <splunk> im just in Solaris 10's terminal [20:45:23] <splunk> shell [20:45:30] <nrubsig> splunk: do you have X11 running ? [20:45:35] <splunk> i dunno [20:45:39] <splunk> :( [20:45:42] <nrubsig> groan [20:45:43] <timeless> this is a shell in X? [20:45:46] <splunk> sry [20:45:50] <splunk> how do I find out [20:45:57] <timeless> can you move it w/ you rmouse? [20:46:03] <splunk> hah yea [20:46:14] <splunk> i've got jds running [20:46:14] <nrubsig> splunk: try $ ps -ef | fgrep X # and post the output, please... [20:46:14] <splunk> lol [20:46:47] <timeless> personally, i'd try running another terminal from the java button [20:46:56] <timeless> and see if export lists DISPLAY [20:46:57] <splunk> its X11 [20:47:18] <splunk> usr/x11/bin/XOR :0 -depth 24 -nobanner -auth /var/dt/A:0-apaWZa [20:47:22] <splunk> root' [20:47:28] <splunk> then a few more things [20:47:31] <splunk> do you need it all? [20:49:57] <nrubsig> splunk: my interest was whether the Xserver is running at id 0 or something else. [20:50:07] <nrubsig> splunk: ok, screw it... [20:50:12] <nrubsig> splunk: .... try: [20:50:16] <saguti> I want begin to develop for Solaris, I would like to take the advice starting by trying to solve the oss-bit-size classified bugs which are listed on Bugs database. [20:50:35] <splunk> ... [20:50:38] <nrubsig> splunk: $ cd dir_where_you_unpackaged_firefox ; (DISPLAY=:0.0 ./firefox) [20:50:44] <saguti> On the page says that one self should contact the responsible engineer, but I am not sure.... [20:51:05] <saguti> how to do it, because I could not find the email address. [20:51:21] <timeless> what url is this? [20:51:23] <nrubsig> saguti: which bugid do you like to fix ? [20:51:32] <splunk> then "export DISPLAY" [20:51:34] <splunk> ./firefox [20:51:34] *** bobbyz has quit IRC [20:51:38] <splunk> it loads the app [20:51:46] <splunk> but how do I tweak it so I don't have to do that every time? [20:51:53] <splunk> what is that DISPLAY stuff do? [20:52:06] <nrubsig> splunk: in theory it should work out of the box. [20:52:11] *** Plazma_Work has joined #opensolaris [20:52:12] <saguti> nrubsig: I would like to work in this one: 6490715 [20:52:22] <nrubsig> splunk: ask in desktop-discuss at opensolaris dot org for help. [20:52:27] <Plazma_Work> isnt opensolairs compaitble with elf/linux binaries? or do i need to add support in [20:52:38] <timeless> splunk: try closing the terminal and opening another one [20:52:43] <splunk> ok [20:52:50] <timeless> see if export lists DISPLAY there [20:52:54] <saguti> nrubsig: I tried guessing the email address of the respponsible engineer listed, but I am not sure if that is the right procedure. [20:52:55] <timeless> plazma: no [20:53:07] <Plazma_Work> mm [20:53:08] <splunk> loads fine [20:53:09] <oxygene> Plazma_Work: in linux branded zones it is [20:53:14] <oxygene> Plazma_Work: elsewhere it's not [20:53:15] <splunk> but when I reboot or something I have to set the DISPLAY again [20:53:16] <timeless> plazma: you can use an lx zone if you need to [20:53:22] <Plazma_Work> ahh [20:53:31] <timeless> that offers a 2.4 kernel [20:53:38] <splunk> so its annoying [20:53:39] <nrubsig> splunk: that's why I said please ask in desktop-discuss at opensolaris dot org [20:53:42] <Plazma_Work> trying to get a game to work in opensolaris, for testing purposes [20:53:43] <splunk> ok [20:53:55] <timeless> plazma: if you have sources, build :) [20:54:00] <timeless> if you don't have sources, you're probably toast [20:54:10] <Plazma_Work> yea [20:54:11] <nrubsig> splunk: I don't use JDS often and I can't help with diagnosing errors within that environment. [20:54:19] <splunk> ok, thanks anyway [20:54:21] <timeless> at some point whatever it is won't support the 2.4 kernel [20:54:42] <timeless> splunk: it seems odd that you'd get an xterm of any kind w/o the DISPLAY var set [20:54:49] <timeless> and that's not a bug in firefox [20:55:02] <timeless> it needs to use DISPLAY so that you could tell it to display somewhere else [20:55:07] <timeless> e.g. on my n800 [20:55:36] <Pietro_S> Plazma_Work: what game? just curious [20:55:43] <splunk> yea [20:55:50] <splunk> i can try the other display environment [20:55:59] <Plazma_Work> tremulous [20:56:06] <splunk> the other pre packaged display [20:56:09] <Plazma_Work> its # 1 on opensolaris's site so.. [20:56:10] <splunk> i wish Solaris had KDE [20:57:31] <trygvis> blastwave has kde [20:57:36] *** nrubsig was kicked by nrubsig (self-made-suicide) [20:57:37] <splunk> lets try common desktop environment [20:58:33] <splunk> cde SUCKS [20:59:05] <Plazma_Work> no it doesnt [20:59:07] <Plazma_Work> its sexy [20:59:38] <splunk> lolnope same thing in CDE.......cannot open display [20:59:39] <Pietro_S> splunk: try xfce [21:00:06] <splunk> ? [21:00:20] <CIA-17> ml29623: 6568476 IPsec Tx performance is bottlenecked by only one nxge HW ring being used, 6544706 ndd wildcard query of /dev/nxge0 panics the system with an assertion failure, 6571370 Atlas: data corruption with TCP_Corrupt while runing IO stress tests [21:00:27] *** Mdx4 has joined #opensolaris [21:00:30] *** Gursikh has joined #opensolaris [21:00:42] <splunk> jde is sexy [21:00:45] <kjetilho> stupid format(1m) -- since when did Unix stop me from shooting myself in the foot? [21:00:51] *** bigjohnto has quit IRC [21:01:42] <Pietro_S> xfce can be called as sucessor of CBE, because fist version of xfce were inspired by CBE [21:04:32] <Pietro_S> Plazma_Work: take look into archives of game-discuss at opensolaris dot org, there can be fouund info how to compile it or link to package [21:04:37] *** steleman has quit IRC [21:06:39] *** steleman has joined #opensolaris [21:07:33] <Plazma_Work> mm in the discussino boards right? only 3 pages [21:12:33] <lasseoe> hrm, seems sunsolve is bolloxed again [21:12:46] <lasseoe> can't download patchdiag.xref [21:12:51] <holcomb> retry [21:12:54] <holcomb> just happened to me too [21:13:11] <holcomb> patches are sometimes downloading, sometimes not [21:13:38] *** risc_solaris has joined #opensolaris [21:14:23] *** Khalsa has joined #opensolaris [21:14:25] <risc_solaris> Hi. Sorry for such a low-level question, but while attempting to boot Solaris 10 or OpenSolaris, the system constantly halts at "configuring /dev" [21:15:01] <risc_solaris> the hardware is a bit old, but prolific: Intel 440BX, Pentium III, Nvidia GeForce FX 5200, SiI 3512 SATA [21:15:04] <ofu> it just takes some time [21:15:07] <Plazma_Work> sounds like a problem with pseduo devices [21:15:22] <Plazma_Work> well.. i take that back.. if tha twas the case it probably wouldnt get that far [21:15:35] <ofu> what does boot -v tell you? [21:16:01] <risc_solaris> i actually didn't realize what my options were as far as a verbose boot [21:16:02] *** xsh has quit IRC [21:16:09] <risc_solaris> if i invoke that from grub, it will display the entire process? [21:16:56] <Plazma_Work> boot -v is on sparc [21:17:33] <ofu> but you could also use -v as a grub option [21:17:34] <splunk> anyoone use wireshark on their solaris boxes? [21:17:46] <risc_solaris> my only idea is that the SATA controller is unsupported [21:17:55] <dunc> ooh risc_solaris i had that [21:18:00] <dunc> with a sata controller [21:18:03] <dunc> what type? [21:18:03] <risc_solaris> ah [21:18:06] <trygvis> it still wouldn't hang the boot process [21:18:10] <dunc> it did mine [21:18:22] <risc_solaris> SiI 3512 [21:18:30] <dunc> with -v enabled i'd see where, but still didn't work [21:18:49] <risc_solaris> did you ever find a resolution? [21:18:59] <dunc> i got a mobo with a better sata controller :) [21:19:27] <risc_solaris> which controller was giving you the problem? [21:19:28] <dunc> s/better/worky/ one maybe [21:19:46] <dunc> i can't remember, have lent it to a mate, but bear with, i might be able to find the order in old emails [21:19:50] <dunc> [hourglass] [21:20:16] *** sarah has joined #opensolaris [21:20:20] <dunc> oh hang on, i mailed storagediscuss about it [21:20:57] <risc_solaris> oh, it's alright... don't go out of your way to find it [21:21:00] <dunc> ah, Adaptec 1205SA [21:21:01] <splunk> I'm blastwave.org and i'm trying to download wireshark, their website design is horrible [21:21:05] <risc_solaris> oh, wow... an adaptec [21:21:10] <splunk> i can't seem to find where to download the whole package [21:21:14] <splunk> it just lists files [21:21:27] <sstallion> splunk: you are supposed to use pkg-get [21:21:41] <splunk> well.. [21:21:50] <sstallion> splunk: RTFM: http://www.blastwave.org/howto.html [21:21:59] <splunk> i can't seem to find NIC drivers for my Dell box [21:22:12] <splunk> Solaris isnt picking up my NIC so its a stand alone :( [21:22:14] <Plazma_Work> hmm... what hte crap.. gmkdir not found.. wierd [21:22:15] <splunk> any ideas? [21:22:21] <risc_solaris> so it looks like it's the SATA controller, then... I guess I'll cram in the old AHA2940 to play with it [21:22:26] <ofu> how useful is wireshark without a nic driver? [21:22:27] <sstallion> splunk: hrmm. what device is it? [21:22:29] <risc_solaris> thanks [21:22:32] *** risc_solaris has quit IRC [21:22:49] <splunk> Broadcom 5752 gig ethernet [21:22:59] <splunk> I saw drivers for 5751 [21:23:03] <splunk> or 53 [21:23:04] <splunk> but no 52 [21:23:10] <splunk> on the broadcome website for solaris [21:23:20] <sstallion> one sec let me see what i have [21:23:22] <sstallion> its a bge [21:23:32] <splunk> sweet [21:23:33] <splunk> thanks [21:26:04] <Plazma_Work> hmm wtf [21:26:11] <Plazma_Work> gmkdir not found [21:26:34] *** _william_ has left #opensolaris [21:26:57] <sstallion> splunk: have you checked broadcom's site? [21:27:10] <splunk> yea of course.... [21:27:47] <sstallion> bcm5704? [21:28:23] <splunk> Dell's website says BCM5752 [21:28:36] *** tsp has quit IRC [21:28:45] <splunk> iv got solaris running now in vmware on the mahine right now and the host is running ubuntu [21:28:51] <splunk> but solaris obviously picks up the virtual nic [21:29:01] *** Gursikh has quit IRC [21:29:07] <splunk> but solaris on my standalone machine isn't picking up the BCM 5752 [21:29:08] *** Khalsa has left #opensolaris [21:29:17] <sstallion> im on a BCM5721 [21:30:17] <splunk> http://www.sun.com/bigadmin/hcl/data/components/details/815.html [21:30:20] <splunk> closes I can find [21:30:24] <splunk> but i havnt tried it yet [21:30:37] <splunk> think I should? [21:31:00] <sstallion> you could try this one: http://www.broadcom.com/support/ethernet_nic/driver-sla.php?driver=570x-Solaris [21:31:17] <sstallion> it just depends on how well the chipset handles [21:31:49] <sstallion> hrmm i wonder if that driver supports jumbo frames [21:32:15] <seanmcg> splunk: whats the device id of the card ? /usr/X11/bin/scanpci can tell you. The 5752 should be working for you with the bge driver [21:32:42] <splunk> 14e4,1600 bge Broadcom Corporation Broadcom Corporation NetXtreme BCM5752 Gigabit Ethernet PCI Express [21:32:49] <splunk> ok let me look [21:33:19] <seanmcg> Ya, that should be ok. what version of solaris have you ? [21:33:25] <splunk> `0 [21:33:26] <seanmcg> s/version/build/ [21:33:27] <splunk> 10 [21:33:43] <sstallion> splunk: does dladm show anything? [21:33:47] <seanmcg> what update of 10 then ? Support for that went into s10u3. [21:33:54] <seanmcg> cat /etc/release [21:34:02] <sstallion> (dladm show-dev) [21:34:27] <splunk> 11/06 [21:34:34] <splunk> s10x_u3wos_10 [21:35:19] <splunk> um dldm wh?? [21:35:21] <splunk> oh [21:35:23] <splunk> hold [21:35:38] <splunk> bge0 [21:35:43] <splunk> link:uknown speed0 [21:35:50] <splunk> so it does see it? [21:35:55] <splunk> but i only see the loop back int [21:36:00] <splunk> i don't see the bge0 [21:36:07] <splunk> int [21:36:08] <sstallion> issue: ifconfig bge0 plumb [21:36:25] <splunk> nothing came back [21:36:30] <sstallion> now issue ifconfig -a [21:36:32] <splunk> i don't have a link to the NIC yet [21:36:45] <splunk> shit [21:36:46] <splunk> ha [21:36:51] <splunk> i see the bge0 [21:36:54] <seanmcg> ;-) [21:36:58] <sstallion> make sure you have a /etc/hostname.bge0 ;) [21:37:03] <splunk> how the hell [21:37:08] <sstallion> otherwise a device will not get plumbed at boot [21:37:16] <splunk> what does plumb mean? [21:37:19] <sstallion> welcome to solaris :) [21:37:23] <splunk> man I gotta catch up on my solaris skills [21:37:28] <splunk> only been doing for about 1.5 years [21:38:22] <sstallion> man ifconfig and search for plumb... it will be your first hit [21:38:29] <elektronkind> oooh [21:38:31] <elektronkind> IBM (NYSE: IBM) and Sun Microsystems, Inc. (Nasdaq: SUNW) will hold a teleconference to discuss an operating system agreement. Jonathan Schwartz, CEO and president of Sun, and Bill Zeitler, senior vice president and group executive, IBM Systems and Technology Group, will host the call for press and analysts. [21:38:37] <splunk> ok [21:38:46] <elektronkind> official IBM support for Solaris on their x86 boxen, perhaps? [21:38:48] <sstallion> elektronkind: its up on sun.com now... i wonder whats going on [21:39:17] <sstallion> elektronkind: IBM sSeries :P [21:40:02] <kjetilho> that's very interesting [21:40:40] <kjetilho> ah, IBM will distribute Solaris [21:41:02] <Tpenta> http://biz.yahoo.com/ap/070816/ibm_sun_microsystems.html?.v=2 [21:41:06] <Tpenta> AND sell support [21:42:02] <elektronkind> IBM Expands Support for the Solaris OS on x86 Systems [21:42:02] <elektronkind> ARMONK, NY SANTA CLARA, Calif. August 16, 2007 IBM (NYSE: IBM) and Sun (NASDAQ: SUNW) today announced that IBM will distribute the Solaris Operating System (OS) and Solaris Subscriptions for select x86-based IBM System x servers and BladeCenter servers to clients through IBM's routes to market. [21:42:02] <elektronkind> there it is [21:42:03] <elektronkind> http://www.sun.com/aboutsun/pr/2007-08/sunflash.20070816.1.xml [21:42:23] <splunk> i've got a lot accomplished today [21:43:18] <elektronkind> Andrew Morton *really* hates us now [21:43:28] <splunk> now that I have an internet connection the solaris box maybe I can pull some much needed updates [21:43:33] <Tpenta> hmmm? Why do you say that? [21:43:59] <saguti> Hello. What is the right procedure to be involved in fixing a bug? [21:46:01] <elektronkind> Tpenta: well, he is Mr. Solaris is an abortion, after all, and IBM is perceived to be one of the biggest vendors of Linux-based stuff. I wouldn't put increased ad hominem attacks towards Solaris past him [21:46:59] *** Mdx4 has quit IRC [21:47:29] * Tpenta wouldn't put words in his mouth; he's good at doig that on his own :) [21:50:00] <seanmcg> saguti: you've read http://www.opensolaris.org/os/communities/participation ? [21:50:08] <kjetilho> Tpenta: and his foot, too [21:50:24] <splunk> i'm assumming bge0 is set to DHCP can I force it to pull an IP or that should happen automatically? [21:50:44] <splunk> oh wait [21:50:53] <splunk> ifconfig bge0 auto-dhcp [21:50:59] <splunk> got it [21:52:14] *** Plazma_Work has left #opensolaris [21:52:29] <seanmcg> splunk: you'll need to touch /etc/dhcp.bge0 for it to survive reboots. [21:52:52] <splunk> hmm its barely alive right now [21:52:59] <splunk> nothing is going out [21:53:03] <splunk> but packets are comin gin [21:53:11] <splunk> it was able to pull an IP from the dhcp server [21:53:12] <saguti> seanmcg: I wrote to the Responsible engineer telling that I am interested in working in a specific bug fix... Is that right? [21:53:35] <splunk> how can I confirm DNS settings? [21:53:56] <ofu> /etc/resolv.conf and /etc/nsswitch.conf [21:54:17] <kjetilho> doh! all my problems with Jumpstart and mirroring is a known bug! [21:54:22] <kjetilho> http://www.webservertalk.com/archive100-2007-1-1778920.html [21:54:24] <seanmcg> saguti: not really.. that page will show you how. Need a sponsor first. [21:55:03] <splunk> hmmm [21:55:14] <saguti> seanmcg: Ah, ok... but, the bug has already an sponsor... [21:56:11] <splunk> something is up I can't get out [21:56:16] <seanmcg> hmm, so someone is already working on your bug ? [21:56:21] <splunk> of the broadcast domain [21:56:34] <ofu> splunk: missing default route? [21:56:44] <splunk> nah router is fine [21:56:46] <saguti> seanmcg: How can I find it out? [21:56:50] <splunk> i have another machine on the LAN [21:56:51] <splunk> working fine [21:57:02] <seanmcg> saguti: whats the bug ? [21:57:09] <splunk> i can ping the gateway [21:57:15] <saguti> seanmcg: It is bug 6490715. [21:57:19] <splunk> im thinking its the NIC or something isnt enabled [21:57:23] <saguti> seanmcg: Something related to touch command. [21:57:28] <splunk> how can I verify more network settings of the NIC? [21:57:42] *** Darth_Wa1er has joined #opensolaris [21:58:34] <Pietro_S> #include <rak/string_manip.h> [21:58:55] <Pietro_S> how such include can include files from main dir of package? [21:59:21] <splunk> usually its the other way around [21:59:27] *** Tpenta has left #opensolaris [21:59:28] <splunk> maybe the NIC is set to simplex [21:59:28] <splunk> haha [21:59:37] *** Tpenta has joined #opensolaris [21:59:38] *** ChanServ sets mode: +o Tpenta [21:59:38] *** Gropi has quit IRC [21:59:38] <dunc> kjetilho, i did some jumstarting to mirrors once. basically had the finish script (or whatever it's called) create an init script that did the mirroring, deleted itself, and rebooted [21:59:38] <seanmcg> saguti: I'd still start with that webpage, and say you're interested in that bug... [21:59:48] <dunc> kjetilho, bit messy, but did the job [22:00:14] <ofu> splunk: ifconfig -a, netstat -nr, dladm show-dev should tell you everything [22:00:24] <CIA-17> josephb: 6593326 dtrace test suite failures [22:00:48] <saguti> seamcg: Ok... but, silly me, How can I "say" I am interested in the bug? [22:00:55] *** bengtf_ has joined #opensolaris [22:01:03] *** Plazma_Work has joined #opensolaris [22:01:08] <seanmcg> read the page, it tells you who/what to email.. [22:01:09] <Plazma_Work> does opensolaris support opengl? [22:01:25] <Pietro_S> Plazma_Work: yes, if you have nvidia [22:01:32] <kjetilho> dunc: yeah, but it seems "wrong" :) [22:01:34] <Plazma_Work> crap [22:01:35] *** The-spiki has quit IRC [22:01:38] <seanmcg> and software via mesa [22:01:39] <dunc> it does yeah [22:01:43] <kjetilho> and it's an extra reboot :) [22:01:45] <Plazma_Work> crappy built in intel [22:01:48] <Plazma_Work> gm 950 [22:02:38] <oxygene> oh, intel support is upcoming, it seems [22:02:45] *** cub- has quit IRC [22:03:03] <saguti> seanmcg: Thanks a lot! [22:03:18] <seanmcg> np [22:04:03] <dunc> i wouldn't worry about an extra reboot though [22:04:50] *** The-spiki has joined #opensolaris [22:05:18] <kjetilho> no, that was tongue in cheek [22:05:41] <kjetilho> it's more that it's not too confidence inspiring when stuff like this isn't working [22:05:51] <kjetilho> I mean -- it should be the default setup, right? [22:06:31] *** Darth_Wader has quit IRC [22:06:41] <kjetilho> Jumpstart doesn't support it too well, though, since it's kind of painful to find a disk to use as a mirror automatically [22:06:56] *** estibi has joined #opensolaris [22:07:00] <kjetilho> so the profile has to be created dynamically to solve it [22:07:38] <splunk> i keep doing ifconfig bge0 netstat -a and it keeps coming back ifconfig netstat: bad address [22:08:19] <kjetilho> finally found the bugid: 6510504 -- but it's not available. quite annoying [22:09:11] <splunk> hmm [22:09:42] *** Plazma_Work has left #opensolaris [22:11:47] <elektronkind> alright, why is it that I keep getting a permission denied when trying to mount a NFS share within a local zone? The Permission Denied error references the mount point, not the share... which is even more odd. [22:12:16] <elektronkind> [root@svn]/# mount /local/svn-repos [22:12:17] <elektronkind> nfs mount: mount: /local/svn-repos: Permission denied [22:12:20] <elektronkind> so peculiar. [22:12:33] <delewis> elektronkind: /etc/default/nfs, make sure DOMAINNAME is the same. [22:12:50] <delewis> er, NFSMAPID_DOMAIN [22:13:03] <delewis> you'll need to restart nfsmapid, and probably nfs/client to be safe. [22:13:23] <elektronkind> ah right, that thing [22:13:25] <splunk> ofu: can I configure my NIC using the GUI within Solaris [22:13:28] <elektronkind> vive nfsv4 [22:13:33] <splunk> since this isnt IOS i'm almost screwed [22:13:34] <splunk> lol [22:15:06] * delewis recently ran into a NFSMAPID_DOMAIN issue [22:15:22] <delewis> any NFS export mounted in a local zone would make all the uid's and gid's map to nobody. [22:16:26] <Pietro_S> laca: what variable of spec is set to dir where source is extracted and make is run? [22:16:45] *** Gekkko[PDA] has quit IRC [22:17:19] *** Murmuria has joined #opensolaris [22:18:46] <networkdump> is there a reason for having 2 log directory (/var/adm + /var/log) ? [22:18:52] <seanmcg> splunk: read http://docs.sun.com/app/docs/doc/816-4554 - and you can come back here to answer questions :) [22:19:12] <laca> Pietro_S: that's the -n option of %setup [22:19:17] <kjetilho> networkdump: /var/log is a "new" invention [22:19:19] <laca> %setup -n name_of_dir [22:19:29] <kjetilho> but many people think it is a better name ... [22:19:31] <laca> add -c if you want pkgbuild to create that dir [22:19:45] *** d-s-d has joined #opensolaris [22:19:46] <d-s-d> hi [22:19:50] <networkdump> and why /var/svc/log? [22:20:07] <kjetilho> good question [22:20:21] <networkdump> this is anoying [22:20:38] <kjetilho> actually, we used to NFS-mount /var/log so we had a common log directory [22:20:49] <kjetilho> that would totally break SMF :-) [22:21:07] *** gnut has joined #opensolaris [22:21:08] <kjetilho> not saying it was a clever thing to do, but that was our setup [22:21:22] <quasi> very not clever ;) [22:21:52] *** tombhadAC has joined #opensolaris [22:21:53] <kjetilho> obviously not suitable for high traffic logs [22:22:10] <networkdump> use udp or syslog-ng [22:22:30] <kjetilho> sure, but not all apps support syslog [22:22:44] <splunk> o [22:22:46] <kjetilho> and the namespace is pretty cramped, too. local0..7, you know [22:23:06] <gnut> hi [22:23:20] <gnut> anyone have issues running iconv to convert .inf files for use with ndiswrapper? [22:23:27] <Pietro_S> laca: I need that dir as variable so I can add another CXXFLAG exactly -I<main_dir_of_package> [22:23:54] <d-s-d> I intend to install a new server at home, but I couldn't yet decide what operating system to use. Solaris is getting interesting to me because of its support for ZFS. I've also been thinking of OpenBSD, 'cause i might also use it as the internet router/Firewall. I've got an old Ultra 10 with 1 Gig of RAM which I might use for that purpose. So, what do you think? I don't have that much experience concerning Solaris. Maybe, somone might give me tip whether Sola [22:24:22] <kjetilho> d-s-d: truncated after "tip whether Sola" [22:24:41] <d-s-d> Last scentence reads: Maybe, somone might give me tip whether Solaris is a good choice in my case...? [22:24:47] <splunk> hmm [22:24:51] <kjetilho> I'd say go for Solaris if you want a large fileserver. ie. hundreds of gigs or more [22:24:53] *** wisnios has joined #opensolaris [22:25:31] <kjetilho> not sure I would recommend the Ultra 10, though. most scrapheap PC-s are faster, and you really want SATA for the disks anyway [22:26:55] <trygvis> kjetilho: dude, you're on #opensolaris, older stuff than that is often recommended here [22:27:09] <kjetilho> it's still crap [22:27:11] <laca> Pietro_S: ah, okay, you can do sometinhg like %_builddir/%name-%version [22:27:17] <trygvis> yeah [22:27:24] <bengtf_> d-s-d what speed of ultra10 ? thats the question if it is the fast one use sol10, I do that and a slow one as firewall ;) [22:27:30] <kjetilho> useful to practice on OpenBoot, but not much else [22:28:11] <d-s-d> bengtf_: I think it's an UltraSparc IIe at about 440 MHz. [22:28:30] <Pietro_S> laca: thanks that's what I need, by the way did you get my PM? [22:29:52] <bengtf_> 440 is good enough to run sol10, just get a large enough disk ;) and some scsi cards [22:29:55] *** nikinana has joined #opensolaris [22:29:56] <gnut> aha. got iconv to work [22:30:07] <gnut> now i'm stuck on make ndis [22:30:21] <kjetilho> bengtf_: it's going to be much more expensive than an old PC with SATA [22:30:28] <d-s-d> bengtf_: What kind of scsi-cards are supported by sol10? [22:30:43] *** cypromis has quit IRC [22:30:50] <wesolows> plenty of them. Virtually all the LSI stuff is [22:30:57] <wesolows> Which is what everyone uses. [22:31:21] <d-s-d> kjetilho: That's a good point... I do have some SCSI-HDs around and various Adaptec-SCSI-Cards... but large scsi-hds are expensive. [22:31:24] <bengtf_> got u10 for free and some scsi cards, had to buy seagate 120G ide disk though pretty cheap [22:32:49] <d-s-d> On the other hand, i could also use ide hard drives (two 500 gig drives should do it). [22:32:55] <bengtf_> nope [22:33:01] <bengtf_> 120G max [22:33:06] <nikinana> ON71 can be downloading now? [22:33:08] <d-s-d> why's that? [22:33:21] <bengtf_> ata100 only [22:33:33] <d-s-d> there's no way around that? [22:33:45] <bengtf_> perhaps sata card maybe [22:34:00] <kjetilho> bengtf_: it's not ATA100, but the chipset on the Ultra 10 [22:34:05] <laca> Pietro_S: no, i didn't, i should probably identify with nickserv [22:34:53] <bengtf_> kjetilho: yep, also many ata100 pcs had the same max limit [22:35:20] <d-s-d> Well, there's also another machine i could use. It's much more powerfull, but it's amd64 and i'm not sure how well (open)solaris runs on intel-hardware. [22:35:29] <kjetilho> it runs very well [22:35:30] <gnut> i'm getting lots of "conflicting types for 'signal' " errors when trying to compile ndis [22:36:01] <d-s-d> kjetilho: Is there an official intel-version of sol10 by sun? [22:36:06] <bengtf_> d-s-d make sure you use geforce cards [22:36:20] <sstallion> d-s-d: ... sol10 [22:36:23] <bengtf_> d-s-d yep just download it [22:36:43] <d-s-d> bengtf_: you're talking about graphics... i don't need any graphics on a server, do I...? :) [22:37:46] <reflect> do I need to fill in an RFE to get ZFS root mirror to also mirror grub data.. or is that already planned? [22:37:57] <bengtf_> d-s-d missed that hehe, server or no server sol10 on amd64 x2 runs pretty good, my main desktop is such a beast [22:38:25] *** slowhog has joined #opensolaris [22:39:15] *** sfire||mouse has quit IRC [22:39:24] *** saguti has quit IRC [22:39:34] <the-decider> yar, register article on the s86/ibm deal -- great quote: [22:39:38] <the-decider> Solaris x86 enjoys a relatively small but devoted developer community. The Linux crowd seems to have taken notice of this with zealots bashing Solaris x86 at every chance during recent open source trade shows. Such concern is understandable given that Solaris offers a number of high-end features not found with Linux. (Note to self: Start flame war.) [22:39:49] <the-decider> http://www.theregister.co.uk/2007/08/16/sun_ibm_solarisx86/ [22:40:02] *** Plazma_Work has joined #opensolaris [22:41:15] <elektronkind> In a real shocker, Zeitler also noted that he "would like to see" some Solaris on System P work, which would have Solaris compete head-to-head with AIX. [22:41:25] <elektronkind> ahaha, the shocker! [22:41:53] * elektronkind wants to tape a printout of the shocker to both of the p670's downstairs [22:42:24] <the-decider> we need some Sun logo stickers ;) [22:43:34] <d-s-d> hmm... i could also configure my amd64 x2 to be server and workstation... i just need two more sata-drives, little bit more ram... hmm, ideas are comming up... :) [22:44:04] <d-s-d> So, what about routing capabilities? [22:44:15] <bengtf_> and a geforce, to get get full opengl support ;) [22:45:33] <bengtf_> d-s-d 4G mem 4.6Ghz amd64x2 runs pretty smooth [22:45:41] *** cmihai has joined #OpenSolaris [22:45:46] <cmihai> OMG. NVIDIA sucks ass. [22:45:51] *** Tpenta has quit IRC [22:45:55] <cmihai> They make cards but provide NO drivers! [22:46:03] <reflect> thanks for sharing, cmihai :) [22:46:04] <bengtf_> for sol10 they do [22:46:15] <cmihai> No they don't/ [22:46:23] <cmihai> Not for their mobile series. Laptop cards. [22:46:24] <alanc> for their graphics cards they do [22:46:31] <d-s-d> bengtf_: i don't know what graphics card i have... it's onboard thing that uses the main memory... ;) [22:46:36] <cmihai> Got a GeForce 8600m GT, they have NO drivers what so ever. [22:47:01] <alanc> hmm, thought that was in the latest drivers from nvidia.com [22:47:07] <Pietro_S> cmihai: did you check latest driver on their web? [22:47:10] <cmihai> Nope, none of their stuff installs. [22:47:18] <cmihai> It's for the desktop version of their stuff. Not even the betas. [22:47:21] <reflect> do I need to fill in an RFE to get ZFS root mirror to also mirror grub data.. or is that already planned? [22:47:25] <cmihai> Their response was: go ask your laptop manufacturer. [22:47:37] <cmihai> Mine was: go fornicate yourselves with an iron stick. [22:48:00] <Bartman007> cmihai: typically because mobile cards are tweaked by the OEMs (often underclocked because they dont' have adequate cooling) so nVidia leaves it up to the OEMs to give you drivers. [22:48:12] <cmihai> Riiiight. [22:48:27] *** Tpenta has joined #opensolaris [22:49:07] * Bartman007 has run into more than one mobile videocard that the vendor flashed some customize firmware to... [22:49:22] <bengtf_> d-s-d get another cards if thats the configuration, get a card that doesnt steal memory ;) [22:50:03] <d-s-d> bengtf_: I've just look it up: The whole thing is an ATI-chipset... ;-( [22:50:20] <cmihai> Well, not that it matters now, does it? [22:50:32] <cmihai> Unless they expect me to reverse engineer the thing and write a driver. [22:50:53] <Plazma_Work> when building a package.. what does "problem: mode is not numeric" mean/refer to [22:50:56] <cmihai> Should have just went with the crappy onboard intels. [22:50:58] <bengtf_> d-s-d on a server it doesnt matter if you dont use X [22:51:22] <Gman> hooray it's friday! [22:51:39] <bengtf_> Gman: no its not , well not yet anyway [22:51:46] *** jpdrawneek has joined #opensolaris [22:51:59] *** wisnios has quit IRC [22:52:48] *** Triskelios has quit IRC [22:53:28] *** Teknix has quit IRC [22:53:50] <reflect> dlg: you awake? [22:54:04] <reflect> or any other sun employee, for that matter.. [22:54:32] *** Plazma_Work has quit IRC [22:54:32] <jmcp> dlg isn't a Sun employee [22:54:38] <jmcp> reflect: what questions do you have? [22:54:42] <WickedWicky> call Sun Support and open a case when you need them [22:54:45] <WickedWicky> ; -) [22:54:57] <reflect> I have a nick that seems to be a sun employee [22:55:06] <reflect> and I need to get in touch with him/her [22:55:11] <jmcp> reflect: who? [22:55:18] <reflect> "inaddy" [22:55:36] <jmcp> google? [22:55:52] <wesolows> !seen inaddy [22:55:58] <Drone> inaddy (inaddy!i=inaddy at c9513a4c dot virtua.com.br) was last seen in #opensolaris on Fri 15 Jun 2007 09:01 GMT, saying ';)'. [22:56:22] <reflect> he or she has registered a channel that.. I would like to discuss a change of ownership to [22:56:43] <jmcp> reflect: https://blueprints.launchpad.net/nexenta/+spec/nexenta-org-irc-channel [22:56:50] <wesolows> Why not leave a message with services? [22:56:52] <WickedWicky> well, Sun has an office in Brasilia/Distrito Federal [22:57:04] <reflect> wesolows: he or she hasn't been on for atleast a month? [22:57:08] *** StylusEater_Work has left #opensolaris [22:57:09] <WickedWicky> Brasilia being the capital city of Brazil and Distrito Federal being the state [22:57:30] <wesolows> reflect: Not in this channel. [22:57:50] <reflect> yeah, even more here [22:58:11] <reflect> I was just hoping that perhaps someone knew who it was [22:58:23] <wesolows> If he or she works for Sun, I don't know him or her. [22:58:56] <reflect> Last Seen Address: i=inaddy@nat/sun/x-11f931f1c6e57503 <-- can you get that without working for sun, btw? [22:59:09] <WickedWicky> I almost got hired by the office in Brasil... but then I got told the office there would be reduced to 1/3 of what it was [22:59:43] <WickedWicky> so I decided to work for an ISP there [22:59:53] <wesolows> How should I know? That's freenode translating some address into something it thinks has meaning. [23:00:11] <wesolows> If it really was a SWAN address, then you'd probably have to be at least a contractor. [23:00:37] <reflect> I know very little of freenode, I just thought that perhaps it was required to do something before getting that kind of address [23:00:42] *** nostoi has joined #opensolaris [23:02:22] <jpdrawneek> got a quick question for sun guys [23:02:30] <d-s-d> Nexenta seems to be the most popular community based OpenSolaris System. Are there any reasons why one should choose one over the other? [23:02:52] <d-s-d> ... i mean, besides open solaris itself. [23:03:04] *** ChanServ sets mode: +o Tpenta [23:03:05] <reflect> d-s-d: for me, it was the package management system [23:03:22] <jpdrawneek> how quick is the finance side? Been chaseing an invoice and hear nothing from them for a month [23:03:39] <d-s-d> What about hardware support? Do they both support the same amount of hardware? [23:04:00] <reflect> d-s-d: nexenta usually lags a couple of releases behind opensolaris [23:04:00] <WickedWicky> jpdrawneek: talk to your accountant? I dont think this is the place for that [23:04:01] <wesolows> jpdrawneek: I really wouldn't know, and even if I did it would be improper to comment. [23:04:35] <jpdrawneek> k - just spent weeks ringing [23:04:42] <WickedWicky> when I have a question about invoices/quotes I am suposed to contact the sun account manager we have.. I think this is the normal/regular way to go [23:04:48] <jpdrawneek> just want to know if thats normal [23:05:18] <jpdrawneek> we don't have an account manager - just a bill [23:05:49] <cmihai> Hah, I've tricked NVIDIA :D [23:05:56] <d-s-d> What about software support... I'm currently using ubuntu, so i'm used to apt. But i'm not afraid to switch if another package management system offers me the applications i need. [23:06:07] <WickedWicky> cmihai: hax0r [23:06:27] <cmihai> Changed the version in their drivers to use 8600m GT instead of 8600 and it works perfectly. [23:06:29] <cmihai> Bastards. [23:06:36] <WickedWicky> ... [23:06:49] <WickedWicky> you have an Nvidia 8600.... [23:06:58] <cmihai> Yeah, 8600mobile GT [23:06:59] <WickedWicky> to do what? render xterm in 0.000picoseconds? [23:07:20] <richlowe> the xscreensaver hack with the cow, perhaps? [23:07:21] <cmihai> Play games on Solaris duh [23:07:24] <reflect> WickedWicky: dude, it's atleast 0.003picoseconds [23:07:26] <d-s-d> WickedWicky: well, you could attach some 3d-effects to your xterm... ;-) [23:07:38] <cmihai> http://unixsadm.blogspot.com/2007/08/gaming-on-solaris.html - tons of games! :-D [23:07:48] <WickedWicky> I cant really say anything, I have a radeon 9700 [23:07:52] <cmihai> Sparc or x86 there's always good old Quake 3 :D [23:07:54] <Pietro_S> there is also gaming comunity ;-) [23:08:00] <WickedWicky> 256MB video memory for Eterm, hooray :P [23:08:12] <WickedWicky> haha, that's so true [23:08:16] <cmihai> Don't forget the 1GB Hypermemory :D [23:08:36] <cmihai> Those eTerms are so resource hungry :D [23:08:37] * WickedWicky is kinda happy his machine is stable now [23:08:44] <WickedWicky> I'll have to buy a new strip of memory though [23:08:52] <cmihai> Yeah, this piece of crap only has 2GB [23:09:06] <cmihai> And I can't find 2GB SODIMMS, it only has 2 slots ;-( [23:09:16] <WickedWicky> I had 3GB. But then I installed opensolaris and my machine paniced at random. [23:09:23] <cmihai> cool [23:09:31] <WickedWicky> apparently I was mixing 2xpc3200 dimms with 1xDDR400 [23:09:44] <WickedWicky> timing issues ftw [23:09:47] <cmihai> Yep [23:09:50] <cmihai> Been there.. [23:10:01] <cmihai> Specs said CL 2.5, experience said 3 hehe [23:10:16] <cmihai> Especially if you have Dual Channel stuff... [23:10:19] <WickedWicky> it ran for 3 years though. Guess my other operating system never had to address more than 2GB or something.. or opensolaris manages my memory different [23:10:27] <cmihai> don't mix crap memory, it will fuck up [23:10:51] <cmihai> Oh, Windows can't access > 2GB RAM unless you start it from bootloader with > /3GB /PAE [23:11:01] <cmihai> And I'm guessing you never did... [23:11:10] <WickedWicky> it was Linux, but I am always afraid I'll get slaughtered for typing that word here [23:11:11] <cmihai> Sure it's detected, but what's the point if it's never addressed [23:11:15] *** Gursikh has joined #opensolaris [23:11:21] <the-decider> life is too short for crap RAM. [23:11:24] <Pietro_S> cmihai: or you aren't using bullshit with x64 appendix [23:11:30] <Bartman007> WickedWicky: pc3200 *is* DDR400 [23:11:44] <cmihai> Pietro_S, ever tried using x64 Windows? Hah [23:11:49] <cmihai> It's even worse then Win ME :D [23:11:50] *** Gursikh has left #opensolaris [23:11:52] <WickedWicky> serious? [23:11:54] <the-decider> life is also too short for the confusing terms used to describe various types of crap, or non crap, ram. [23:11:58] <Bartman007> WickedWicky: yes. [23:11:59] <cmihai> Duh, no drivers. For ANYTHING [23:12:05] <cmihai> Not even memory sticks ffs [23:12:18] <cmihai> Wanna make fun of someone? [23:12:23] <cmihai> Buy them a Windows Vista Ultimate. [23:12:26] <cmihai> 64 bit! [23:12:38] * reflect burps [23:12:41] <the-decider> let me guess -- no software runs on it? [23:12:42] <reflect> oh, sorry. [23:12:54] <WickedWicky> Bartman007: then I dunno, everything works fine when I try another memory module that says pc3200 [23:12:55] <Bartman007> WickedWicky: and ram is supposed to drop the clockspeed down to the frequency of the slowest stick (unless you are forcing it to a specific speed) [23:12:59] <Pietro_S> cmihai: I still have them on this machine as multi-boot, and I used them only for internet banking and sending batches to public assuurance companies [23:13:03] <cmihai> the-decider, hardware and software support is the same as for Windows NT... ALpha :D [23:13:14] *** alphakiller|away is now known as alphakiller [23:13:24] <WickedWicky> using the DIMM on its own without the other two modules works fine [23:13:41] <cmihai> WickedWicky, yes, memory using different timings doesn't like running in Dual Channel [23:13:49] <cmihai> Just move it around so it won't use that. [23:13:54] <cmihai> Different colored slots. [23:14:14] <cmihai> Sheesh, it's so x86. [23:14:18] <Bartman007> WickedWicky: that said, there are still plenty of things that could cause issues, it's not my machine so I don't know what's going on :-) [23:14:23] <cmihai> They do RAID-0 for the memory... I mean, wtf [23:14:27] <WickedWicky> this is an x86 :P (Athlon 2400XP+) [23:14:35] <cmihai> It's not bad enough that it's not ECC [23:14:36] *** nostoi has quit IRC [23:14:39] <cmihai> They put it in RAID-0 [23:14:57] <cmihai> Now look at IBM, they do ECC and ChipKill (RAID-1) for memory. You can hot swap the thing lol [23:15:00] <WickedWicky> Bartman007: we'll see. Memory is cheap these days. I am happy to see it running stable now [23:15:01] <Pietro_S> when I use -I<dir> in CXXFLAGGS, why <include rak/blabla.h> isn't found when it's in dir directory? [23:15:19] <WickedWicky> and I should have my sparcstation 20 soon to have a play with [23:15:27] <cmihai> That's always fun. Ebay? [23:15:49] <WickedWicky> no, a coworker who wanted to get rid of, he asked 5 bucks for it [23:15:57] <cmihai> Can't beat the price :-] [23:15:58] <WickedWicky> inclusive the monitor and the keyboard [23:16:24] <WickedWicky> I have to charter someone with a car to bring the stuff to my house [23:16:31] <WickedWicky> I am not gonna take that stuff with me in the train [23:16:32] <cmihai> Sure it's crap, but makes a nice OpenBSD firewall or something. [23:16:33] <Bartman007> cmihai: I thought chipkill was more like a raid-5 setup (distributed ECC data and spare chips) [23:16:58] <cmihai> Bartman007, eh, something like that. [23:17:18] <cmihai> Yeah, RAID-5, it's across all chips [23:17:35] <cmihai> Anywho, great stuff :-) [23:19:19] <sommerfeld> or, well, ecc codes tuned to cope with loss of specific contiguous sets of bits [23:20:32] *** mrdeviant has quit IRC [23:21:10] *** jpdrawnee1 has joined #opensolaris [23:21:28] *** rpm has joined #opensolaris [23:24:14] <Bartman007> sommerfeld: and the ability to recover from the loss of an entire chip (though I suppose that is a specific contiguous set of bits...) [23:24:51] <sommerfeld> right, the trick is laying out the bits across the chips so that the right thing happens if you lose a chip [23:25:52] <moazamraja> speaking of IBM.... [23:26:06] <moazamraja> wonder if they'll actually genuinely *try* to sell Solaris x86 on their x86 boxen [23:26:36] <rpm> i'd like to play a bit with opensolaris.. what should i be installing.. belenix, nextena, solaris express.. [23:26:58] <cmihai> moazamraja, yeah, I heard there was some meeting today or something about that. [23:27:06] <cmihai> You know, to pull some kind of HP [23:27:23] *** cypromis has joined #opensolaris [23:27:43] *** Jondice has quit IRC [23:28:45] *** jpdrawneek has quit IRC [23:29:21] <Pietro_S> oh, I don't know if I should damn or celebrate - the sf which I currently port released newer version - including patches which I posted to them today (but it needs more patches) [23:30:36] <moazamraja> cmihai: it's announced, whatevr it really means [23:30:49] <moazamraja> http://www.informationweek.com/hardware/showArticle.jhtml?articleID=201800583 [23:31:17] *** Auralis has quit IRC [23:31:26] *** Marv|Home has joined #opensolaris [23:31:48] *** Auralis has joined #opensolaris [23:33:34] <gnut> conflicting types for signal in proc.h and signal.h [23:33:35] <gnut> oh boy [23:34:19] *** cmihai has quit IRC [23:36:41] <sommerfeld> gnut: you compiling kernel or user code? [23:36:51] <gnut> sommerfeld: ndis [23:36:55] <gnut> sommerfeld: user code [23:37:03] <sommerfeld> why does it need proc.h ? [23:37:08] <gnut> i don't know [23:37:13] <gnut> but when i do make ndis [23:37:24] <gnut> it says proc.h and signal.h have conflicting types for signal [23:37:42] <gnut> /usr/include/sys/proc.h:628: error: conflicting types for 'signal' [23:37:42] <gnut> /usr/include/sys/signal.h:229: error: previous declaration of 'signal' was here [23:38:38] <WickedWicky> see ya later all [23:40:45] <movement> anyone got the most recent opensolaris SS11 compilers installed? [23:41:26] *** cmihai has joined #OpenSolaris [23:41:51] <sommerfeld> gnut: what version of solaris? [23:42:32] <elektronkind> _KERNEL isn't being defined by the compiler, is it? [23:43:04] <gnut> snv_65 [23:43:09] *** SlackDweber has joined #opensolaris [23:43:10] <gnut> snv_69.. sorry [23:43:48] <gnut> sommerfeld: it's snv_69 running on a dell latitude d620... in 32-bit mode.. the wireless card is the dell wireless 1490 [23:44:03] <gnut> using bcmwl5 Broadcom 4312 chipset i believe [23:44:22] *** Marv|LG has quit IRC [23:44:58] <sommerfeld> so, signal.h hasn't changed since 2006. unless emacs is counting lines wrong, line 229 doesn't have a declaration of signal() [23:45:15] *** alphakiller is now known as alphakiller|away [23:45:22] <sommerfeld> what's the output of "what /usr/include/sys/signal.h" ? [23:46:10] <reflect> I'm sorry, what? [23:46:40] <sommerfeld> the "what" command in /usr/ccs/bin [23:46:59] <reflect> oh, thanks [23:47:13] <gnut> /usr/include/sys/signal.h: [23:47:13] <gnut> signal.h 1.22 05/06/08 SMI [23:47:19] <gnut> i didn't know of the what command... cool [23:47:57] <gnut> this is line 229 of signal.h [23:48:00] <gnut> extern void (*signal(int, void (*)(int)))(int); [23:48:07] <reflect> d-s-d: the correct channel would be #gnusol [23:48:08] <gnut> #ifdef __cplusplus [23:48:08] <gnut> extern "C" SIG_PF signal(int, SIG_PF); [23:48:08] <gnut> #else [23:48:08] <gnut> extern void (*signal(int, void (*)(int)))(int); [23:48:08] <gnut> #endif [23:48:10] <gnut> extern int raise(int); [23:48:55] <gnut> and including signal.h is the third include of if_ndis.c [23:49:23] *** perlmonk has joined #opensolaris [23:50:20] <sommerfeld> WTF? signal.h 1.69 06/04/23 SMI [23:50:26] <sommerfeld> is what i'd expect [23:50:50] <gnut> hmmm [23:50:51] <gnut> heheh [23:50:54] <gnut> well... i have snv_69 [23:51:15] <gnut> anyone else have this ? [23:51:40] <gnut> woah! [23:51:42] <gnut> you're right [23:51:47] <gnut> i checked another snv_69 machine [23:51:53] <gnut> and it's 1.69 like yours.... [23:51:56] *** bengtf has quit IRC [23:52:00] *** yarihm has quit IRC [23:52:11] <sommerfeld> pkgchk SUNWhea [23:52:15] *** Megaf_ has joined #OpenSolaris [23:52:17] <seanmcg> that looks like its /usr/ucbinclude/sys/signal.h .. [23:52:52] <gnut> seanmcg: yeah... you're right [23:53:01] <gnut> so somehow, it got copied over to /usr/include/sys? [23:53:08] *** SlackDweber has quit IRC [23:53:56] <Megaf_> hi all [23:54:24] <gnut> hmm... [23:54:27] <gnut> this is really wweird [23:54:31] <gnut> let me check my third computer [23:54:34] *** Triskelios has joined #opensolaris [23:54:58] *** cypromis has quit IRC [23:55:46] <gnut> pkgchk SUNWhea [23:55:46] <gnut> ERROR: /usr/include/pwd.h [23:55:46] <gnut> modtime <07/09/07 11:49:19 PM> expected <08/13/07 02:18:58 PM> actual [23:55:46] <gnut> ERROR: /usr/include/signal.h [23:55:46] <gnut> pathname does not exist [23:55:47] <gnut> ERROR: /usr/include/sys/signal.h [23:55:48] <gnut> permissions <0644> expected <0600> actual [23:55:50] <gnut> group name <bin> expected <root> actual [23:55:52] <gnut> modtime <07/09/07 11:49:27 PM> expected <08/13/07 02:14:11 PM> actual [23:55:54] <gnut> file size <9773> expected <15516> actual [23:55:56] <gnut> file cksum <65528> expected <27285> actual [23:55:59] <gnut> ERROR: /usr/include/sys/timeb.h [23:56:02] <gnut> modtime <07/09/07 11:49:28 PM> expected <08/09/07 07:03:37 PM> actual [23:56:05] <gnut> ERROR: /usr/include/sys/types.h [23:56:06] <gnut> modtime <07/09/07 11:49:28 PM> expected <08/09/07 07:37:50 PM> actual [23:56:08] <gnut> ERROR: /usr/include/sys/wait.h [23:56:11] <gnut> modtime <07/09/07 11:49:28 PM> expected <08/09/07 07:31:48 PM> actual [23:56:14] <gnut> ERROR: /usr/include/wait.h [23:56:17] <gnut> modtime <07/09/07 11:49:20 PM> expected <08/09/07 07:30:21 PM> actual [23:56:18] <gnut> okay. ... it must be corrupted. [23:56:21] <gnut> where can i get a snv_69 version of this? [23:56:31] *** Megaf_ has quit IRC [23:56:52] *** _Megaf has joined #OpenSolaris [23:57:41] <seanmcg> Got the install image of snv_69 somewhere ? [23:57:46] <_Megaf> :) [23:57:52] <gnut> ah... the isos... yeah. let me look it up [23:57:53] *** JonathanW has joined #opensolaris [23:57:58] *** jfndi has joined #opensolaris [23:58:13] <FrostCS> pastebin.ca is lovely too [23:58:43] <gnut> seanmcg: yeah. i found the directory. i'll copy it over and pkgadd it. [23:59:12] <seanmcg> you may want to pkgchk your system...