[00:02:19] *** sparc has left #opensolaris [00:17:43] *** gm152 has joined #opensolaris [00:22:28] *** HolgerS has quit IRC [00:25:53] *** Gman has joined #opensolaris [00:25:53] *** ChanServ sets mode: +o Gman [00:26:33] <Gman> e^ipi: there [00:26:38] <Gman> ? [00:31:37] *** Yorlik has quit IRC [00:32:01] *** odix has quit IRC [00:33:21] *** derchris has quit IRC [00:33:26] *** derchris has joined #opensolaris [00:35:40] *** libkeiser has joined #opensolaris [00:36:39] *** Mazon has quit IRC [00:37:01] *** Mazon has joined #opensolaris [00:37:16] *** ninjaslim has quit IRC [00:45:39] *** stux|away has quit IRC [00:47:08] *** guru][ has joined #opensolaris [00:48:36] *** libkeise1 has quit IRC [01:03:35] <jmcp> CosmicDJ: thanks for that , fixed [01:03:38] *** jgracin has quit IRC [01:10:38] <guru][> jmcp: hi, how you doin' [01:11:13] <guru][> finally i got it works "the grub issue OpenSolaris+CentOS" [01:14:24] *** chubs_ has joined #opensolaris [01:19:52] *** art-vandelay has quit IRC [01:20:41] <coffman> hmpf [01:21:14] <coffman> xorg still leaks badass [01:21:31] <e^ipi> Gman: yo [01:21:43] *** stux has joined #opensolaris [01:21:46] <e^ipi> just got your email [01:21:54] <coffman> 371M cant be normal [01:21:56] <Gman> e^ipi: yeah, let me know and it'll take 5 mins to set up [01:22:29] <guru][> coffman: what are you talking about [01:22:33] <guru][> be more specific [01:22:41] <e^ipi> i sent a reply, but just emancipation-discuss is great [01:23:42] <coffman> guru][: well. after memory usage of xorg on my machine looks bad. it uses more and more memory by time [01:23:55] *** mega has joined #opensolaris [01:24:44] <guru][> you are using Gnome? [01:24:49] <guru][> coffman? [01:24:53] <coffman> guru][: yes [01:24:58] <Gman> e^ipi: done, details in your mailbox [01:25:06] <e^ipi> thanks [01:25:10] <guru][> then, migrate to xfce, of openbox or whatever you want [01:25:24] <guru][> low requirments and nice performance [01:25:29] <coffman> *sigh* [01:25:33] <jamesd> coffman, use dtrace to find out what is telling it to use more memory... there is the bad guy... Xorg is guilty enough because it mmaps the videoram so it appears to use 100's of MB of memory that isn't actually in your pool of ram [01:25:36] *** Gman is now known as GmanAFK [01:28:27] *** mega has quit IRC [01:28:51] *** mega has joined #opensolaris [01:28:57] *** pramz has quit IRC [01:29:12] <coffman> will do some time [01:29:33] <coffman> also firefox is a bitch on sxce [01:30:47] <jamesd> firefox is a bitch everywhere.... [01:30:54] <coffman> yeah [01:31:13] *** rrrand has quit IRC [01:34:51] *** mega has quit IRC [01:45:47] *** mega has joined #opensolaris [01:46:07] *** yarihm has quit IRC [01:46:11] *** Yorlik has joined #opensolaris [01:47:10] *** sah-work has joined #opensolaris [01:49:21] <asdxz> helloo guys [01:49:26] <asdxz> my nick was banned [01:49:52] <asdxz> cause i was running a script that replied to messages like hello [01:50:03] <asdxz> i deactivated [01:50:07] <asdxz> it [01:50:22] <asdxz> i wasn't aware that it was running in this server [01:52:27] <e^ipi> as i recall it also had auto-rejoin [01:52:38] <e^ipi> hence the ban [01:53:41] *** not-me-guv has joined #opensolaris [01:55:50] <asdxz> auto-join? [01:56:03] <e^ipi> yeah, i kicked the bot, and it came back [01:56:05] <asdxz> oh [01:56:05] <e^ipi> so i banned it [01:56:26] <asdxz> O.o [01:56:33] <asdxz> it wasn't a boa [01:56:34] <asdxz> bot [01:56:35] <asdxz> kkkkkk [01:56:47] <asdxz> i was learning the basic of irc scripting [01:56:50] <asdxz> and like every noob [01:56:57] <asdxz> i did noob stuff [01:56:58] <asdxz> hehe [01:56:59] <asdxz> :S [01:58:09] <asdxz> it's a command, that's when it was said hi, or hello i said hello (nick) [01:58:17] <e^ipi> and then you got to deal with the consequences. be thankful you weren't learning the basics of skydiving. [01:58:29] <asdxz> but it is off now [01:58:29] <asdxz> won't trouble again [01:58:29] <asdxz> :) [01:59:02] <e^ipi> no, because next time it's permanent [01:59:30] <e^ipi> not here, but i've banned china before... banning brazil isn't off the table of nuclear options. [02:00:02] *** AxeZ has quit IRC [02:00:22] <e^ipi> ( someone was hopping around and the only consistent bit of their hostmask was .cn... so i banned * at * dot cn ) [02:01:03] <e^ipi> i unbanned china shortly afterwards, but the problem was solved [02:01:52] <purserj> ah but could you ban the us? [02:02:02] <coffman> asdxz: scripting irc? wtf? [02:02:10] <coffman> asdxz: just dont [02:03:25] <e^ipi> +1 [02:09:29] *** psychonate has quit IRC [02:15:42] *** sailorvrz_ has quit IRC [02:24:07] *** sailorvrz_ has joined #opensolaris [02:26:26] <Yorlik> Hello. I am ELIZA. How can I help you? [02:26:29] <Yorlik> :P [02:29:54] <benley> Tell me more about your mother. [02:30:03] <coffman> ./kick Yorlik [02:30:21] <benley> right. So! Does sxce support zfsroot installs nowadays? [02:30:30] <benley> (I can't imagine it wouldn't, but it's been a long time since I did a fresh install...) [02:30:35] <Yorlik> benley: yes [02:30:47] <benley> Does liveupgrade support zfs yet? [02:31:26] <TomJ> has done for a while [02:31:28] <benley> or should I stick with ufs if I like using liveupgrade [02:33:25] <coffman> benley: you can get zfs root on sxce via console installer [02:33:44] <coffman> benley: and lu works (kind of) with zfs root [02:33:52] <benley> heh. how hacky? [02:34:35] * benley now looks for docs [02:34:36] *** mega has quit IRC [02:34:47] <coffman> i dunno if it is only my experience but zfs root,zones on zfs and lu is a bit trouble [02:35:04] <benley> I don't use zones on my sxce box currently [02:37:56] <coffman> benley: well, i would go for zfs root anyhow [02:38:10] <coffman> better then lu with ufs o0 [02:39:25] *** mega has joined #opensolaris [02:39:28] <benley> heh, ok [02:39:30] <benley> I'll give it a shot [02:39:38] <asdxz> coffman: it's a simple command i used in a channel on other network for a friend [02:39:38] * benley runs off to get food and will report back [02:40:01] <asdxz> i don't like or know much about scripting [02:40:43] <asdxz> luckly i was banned [02:40:48] <asdxz> or it would still be working [02:40:51] <asdxz> in all the networks [02:42:04] *** nachox has joined #opensolaris [02:42:12] <nachox> evening guys [02:42:38] <coffman> yo nachox [02:43:39] <asdxz> e^ipi: the nick i was using was the whole ip wasn't it? [02:44:30] <asdxz> i had a problem with a "vhost" i think that is the name, in another network, i think it don't interfere in the other networks [02:44:50] <asdxz> but got rid of it [02:45:27] *** ky-san_ is now known as ky-san [02:49:35] *** Odin- has quit IRC [02:49:43] *** sailorvrz_ has quit IRC [02:54:02] *** guru][ has quit IRC [02:58:14] *** sah-work has quit IRC [03:01:58] <asdxz> one doubt about the os [03:02:11] <asdxz> if the effects aren't activable [03:02:25] <asdxz> or are enabled and stop working immediatly [03:02:34] <TomJ> effects? [03:02:41] *** sah-work has joined #opensolaris [03:02:44] <asdxz> the problem is my vga? [03:02:48] <asdxz> TomJ [03:02:50] <asdxz> yes [03:02:58] <benley> you mean the compiz stuff? [03:02:59] <asdxz> the visual [03:03:00] <TomJ> yes, i expect the problem is your vga [03:03:01] <asdxz> ones [03:03:02] <asdxz> visual effects [03:03:07] <asdxz> that come with the os [03:03:13] <TomJ> try wiring it up differently, see how that goes [03:03:41] <benley> asdxz: if the effects manager thingie starts and then quits, it might just not support your video card [03:04:01] <asdxz> benley [03:04:03] <asdxz> yes [03:04:07] <asdxz> it's just like that [03:04:19] <asdxz> my vga is an ati 9550 [03:04:20] <asdxz> quite old [03:04:26] <asdxz> i think 6 yo [03:04:33] <TomJ> you, or the vga card? [03:04:46] <asdxz> hehe [03:04:48] <asdxz> the vga [03:04:51] <asdxz> :P [03:05:05] <asdxz> i think it was released in 2003 [03:05:35] <asdxz> no [03:05:37] <asdxz> 2004 [03:12:18] *** likaijun has joined #opensolaris [03:13:22] <e^ipi> ati is not supported. [03:17:19] *** chendy has joined #opensolaris [03:17:20] *** mega has quit IRC [03:18:10] *** sah-work has quit IRC [03:19:37] *** sah-work has joined #opensolaris [03:21:51] *** sah-work_ has joined #opensolaris [03:22:21] *** sah-work has quit IRC [03:23:06] *** sah-work_ has quit IRC [03:35:09] *** mega has joined #opensolaris [03:41:06] <jbk> hmm here's a question.. the bios on this thing doesn't support booting off a usb thumb drive, however, grub can see it -- if i stick an iso on there, is there a way using grub on my hd to have it then boot off the usb drive? [03:42:30] <ky-san> grub contains some USB stack? [03:44:32] <jbk> well i think the bios sees them, it just doesn't present the drive as an option for booting [03:44:43] <jbk> so you can access it with the bios interfaces [03:44:56] <jbk> jsut it only lets you boot from dvd or hd [03:46:04] *** ninjaslim has joined #opensolaris [03:50:48] *** asdxz has quit IRC [03:50:48] <ky-san> it seems that grub supports booting from iso9660 (iso9660_stage_1_5 stage2_eltorito) [03:50:50] *** Odin- has joined #opensolaris [03:51:19] <ky-san> why not to try? [03:51:20] <jbk> yeah, but you can't just do root(hdXX,Y) kernel ... module ... from the cdrom [03:54:55] *** not-me-guv has quit IRC [03:55:27] <ky-san> http://ubuntuforums.org/archive/index.php/t-345212.html [03:56:27] <jbk> ahh [03:59:12] *** Odin- has quit IRC [04:01:52] *** nitrile has quit IRC [04:16:31] *** balbirs has joined #opensolaris [04:16:33] *** ][guru has joined #opensolaris [04:17:15] <balbirs> I am just curisous, when was Solaris 10 was launched first, I mean realeased date ? [04:17:52] *** jcea has quit IRC [04:18:53] <][guru> i have a Q, does OpenSolaris support reiserfs? [04:20:18] <coffman> lul [04:20:31] <coffman> ][guru: no [04:20:44] <coffman> ][guru: maybe with fuse [04:20:47] <coffman> someday [04:21:12] *** stukag_ has quit IRC [04:21:14] <coffman> there isnt even ext3 support [04:21:31] <coffman> only hacked, read only [04:21:46] <][guru> you mean even ext3 no support niether [04:21:47] <][guru> :( [04:22:09] <][guru> ok, what about read only, is it by normal mounting? [04:22:39] <nachox> ][guru, there are projects in opensolaris to support both ext3 and jfs [04:23:08] * e^ipi doesn't really see the point [04:23:19] <][guru> nachox, and what time they gonna finished from that project :D [04:23:27] * nachox does, mirating from linux to solaris [04:23:30] <][guru> a couple of years, or century :) [04:23:44] <e^ipi> ][guru: i don't know, when are you going to submit patches? [04:23:53] <e^ipi> nachox: i guess, but NFS works pretty well for that too [04:23:54] <][guru> lol [04:24:13] <e^ipi> nachox: for that matter, you can run linux in a xen domU, give it the disk, export via NFS, and the copy over [04:24:33] <][guru> big jam [04:24:57] <nachox> ][guru, http://opensolaris.org/os/project/ext3/Documentation/quick_start/ [04:25:16] <][guru> got disappointed with OpenSolaris till now, don't know about what's comming, gonna cheer me up or not :) [04:25:22] <nachox> e^ipi, does it really have to be that hard? [04:25:38] <e^ipi> no, just sayin' [04:26:11] <][guru> thanks e^ipi [04:26:14] <][guru> nice link [04:26:23] <e^ipi> nachox. [04:26:28] <e^ipi> i didn't give you a link [04:26:41] <e^ipi> i questioned the premise of the endeavor [04:26:57] <nachox> hehe, e^ipi is scared of the link actually [04:27:05] <jbk> hmm this is going to be academic if i can't find my 8gb usb drive... [04:27:38] <coffman> ][guru: http://blogs.sun.com/pradhap/entry/mount_ntfs_ext2_ext3_in while you work on the ext3 project :P [04:28:33] <coffman> things like ext3/ntfs support are something where sun could easily throw out some money to external people [04:28:33] *** mega has quit IRC [04:28:49] *** mega has joined #opensolaris [04:28:59] <e^ipi> yeah, because they're known for doing that... [04:29:14] <][guru> but why they don't want to do that? [04:29:26] <coffman> but they give 20k$ out for porting a web stack :/ [04:29:29] <coffman> stupid [04:29:29] <][guru> bussinus tricks :) [04:29:43] <nachox> from the link above " Ext2 driver is currently very experimental. It should cause data loss and/or kernel panic. Please be very careful.", it SHOULD? what the hell? [04:29:54] <e^ipi> heh [04:30:05] <e^ipi> if it doesn't, file a bug [04:30:21] <][guru> lol [04:30:47] <e^ipi> srsly, sun has contests pretty frequently, a bounty would probably be a better way to go about it [04:31:07] <e^ipi> have everyone vote on things they want, and then put up a bounty for the first person to do it [04:31:13] <e^ipi> throw a little antagonism in to the mix [04:31:49] <nachox> "we want ZFS in linux"! [04:31:52] <nachox> er, i mean... [04:32:21] <e^ipi> linux can have ZFS, the design docs are out there, there's reference source out there [04:32:28] *** Yorlik has quit IRC [04:32:37] <e^ipi> there's even stubs of it that are license-compatible in GRUB [04:32:39] <e^ipi> they can start there [04:32:41] <nachox> parts of it are also gpl2 because of grub [04:32:43] <nachox> nod [04:33:05] <e^ipi> but that would involve the linux people actually doing any work, instead of relying on the dying leftovers of old UNIX companies ( *cough* sgi *cough* ) [04:33:06] <jbk> but why do that? btrfs is going to take over the world :) [04:33:21] <e^ipi> or swiping from BSD [04:33:45] <nachox> jbk, oracle is behind that particular fs [04:33:46] <jbk> of course coming from oracle, it'll require a separate install, take 4 hours to install, and if you ue an incorrect flag, it will require a 4 hour reinstall :) [04:33:53] <nachox> haha [04:34:11] <e^ipi> and then it will swap everything to the ground, and segfault [04:34:18] <jbk> if you exclude the rdbms, and stuff acquired [04:34:24] <jbk> and look what they've actually produced [04:34:27] <jbk> such as crs [04:34:55] *** Gekz has quit IRC [04:35:06] <jbk> let's just say i wouldn't be making any bets on it :) [04:35:13] <coffman> what i realy would love to see is zfs on windows :P [04:35:23] <e^ipi> coffman: me too, that'd be hilarious [04:35:27] <jbk> i've been tempted to but the $130 fs development kit to port it [04:35:34] <jbk> won't be able to boot off of it [04:35:45] <jbk> but would be nice for relatives [04:35:58] <nachox> well, ZFS has support for sids and windows attributes after all... [04:36:03] <e^ipi> jbk: either way... OSX has it, freebsd has it, solaris has it... if windows has it that'd cover pretty much everyone except linux [04:36:17] <coffman> heh [04:36:41] <coffman> ill see some one riping it with interix :P [04:37:54] <e^ipi> i'm sure plenty of windows people would be extatic to have ZFS [04:38:06] <nachox> ntfs is not so bad [04:38:07] <benley> if I understand correctly, one of the big problems with porting ZFS to linux natively is licensing. [04:38:14] <benley> GPL being what it is and all. [04:38:22] <e^ipi> benley: yes, that's the main reason [04:38:41] <e^ipi> there's also a question about vfs layer semantics, but mostly it's just the GPL being too restrictive [04:38:59] <coffman> its the revoking thing in cddl [04:39:00] <nachox> actually the problem is not only licensing, i think there was a pattent issue as well [04:39:15] <ninjaslim> if the Linux folks even port over ZFS it'll be a half-ass implementation to begin with [04:39:27] <e^ipi> like the rest of the kernel [04:39:30] <e^ipi> so no real change there [04:39:35] <ninjaslim> yup [04:39:52] <ninjaslim> seriously it overwhelms to see how much quality there is on the Unix side compared to a similar OS like Linux [04:39:53] <e^ipi> but with respect to patents... linux already violates a bunch of them, so they obviously are just using it as an excuse [04:40:05] <e^ipi> patents in tech are by in large defensive [04:40:22] <ninjaslim> i was astounded when I compared FreeBSD to Linux, then Mac OS X to Linux, and now that i've actually gotten used to Solaris somewhat i'm even more astounded [04:40:27] <coffman> btw, who is doing FAT on solaris? [04:40:41] <nachox> linux is ok for most workloads, specially if you have some kind of cluster, ignoring that is what put solaris where was 4 years ago imho [04:40:42] <benley> hasn't solaris had fat/fat32 support since forever? [04:41:02] <nachox> pcfs [04:41:08] <coffman> the problem ist, the fat support of solaris by the books [04:41:22] <coffman> thats a problem if you have a off the books device [04:41:27] <coffman> like a mp3 player [04:41:32] <coffman> that i have :( [04:41:37] <e^ipi> nachox: right, any turing machine can anything any other turing machine can run, but there's a continuum of better -> worse here [04:41:38] *** mega has quit IRC [04:41:54] *** mega has joined #opensolaris [04:41:57] <coffman> works fine as masstorage under windows, linux and mac os [04:42:07] <coffman> but not und solaris [04:42:33] <benley> is that a filesystem issue or a usb driver one? [04:42:38] <jbk> hmm i guess i'll just go to fry's tomorrow [04:42:44] <coffman> pre snv_9x i wasnt even able to mount it [04:43:02] *** stukag has joined #opensolaris [04:43:05] <coffman> now i write new files on it and they are gone the moment i turn on the player [04:43:27] <benley> heh [04:43:28] <e^ipi> coffman: flaky device? [04:43:41] <e^ipi> coffman: can you write a new FS to it via solaris and see if that works? [04:43:45] <benley> I wish there was something less godawful than fat32 that many things supported [04:43:59] * benley burns a b98 dvd [04:44:04] <nachox> e^ipi, i agree there, you dont need to talk me into believing solaris is better than linux, i already know that, i stopped using linux because of solaris, i tought shell scripting in solaris and gave away belenix cds to the class, but i'm also trying to be unbiased, linux has its merits [04:44:22] <coffman> e^ipi: will try, just got to my mind again [04:44:42] <coffman> im back to sxce on the laptop since b95 o0 [04:45:36] *** cchapman has joined #opensolaris [04:50:43] *** mega has quit IRC [04:56:54] <ninjaslim> is anyone here running solaris in virtualbox, have you gotten additions working [05:00:07] <jbk> yeah [05:01:31] <ninjaslim> jbk: how, the additions in sxce don't work, they install fine though [05:02:03] <jbk> hmm i didn't have any issues [05:03:12] <ninjaslim> jbk: for me i install it but the window doesn't resize dynamically which is one way i know it's not working [05:03:32] *** freakazoid0223 has quit IRC [05:03:49] *** phrost has joined #opensolaris [05:04:48] <phrost> does snv_98 sxce build have support for zfs root? one of the earlier builds had an option during partitioning if memory serves me right. [05:04:55] *** coffman_ has joined #opensolaris [05:05:42] <benley> phrost: I was told that it supports it now when I asked earlier [05:05:48] <benley> phrost: I'm about to find out for sure :) [05:06:34] <palowoda> phrost: Just remember to use the 'text' console install option for installation with zfs root. [05:07:37] <jbk> hmm patchadd doesn't seem to like zfs root [05:08:00] * jbk really hopes he isn't gonna have to install a s10 image, just to get the right ss12 image [05:10:19] <ninjaslim> jbk: can you dynamically switch between full screen and normal in solaris with additions installed [05:10:42] <jbk> i have it running at work, not here [05:10:46] <jbk> i'd have to get back to you [05:11:11] <ninjaslim> jbk: cause i realize that it might be working but that the additions for solaris may not be the samea s the ones for linux [05:11:21] <phrost> palowoda: thanks for the info, that's probably what i did wrong,i was running it in GUI install [05:18:07] *** coffman has quit IRC [05:19:51] *** RohanRNS has joined #opensolaris [05:22:09] *** Yorlik has joined #opensolaris [05:31:13] *** coffman_ is now known as coffman [05:34:29] *** chubs_ has quit IRC [05:35:09] *** nachox has quit IRC [05:37:42] *** ][guru has quit IRC [05:40:32] *** gm152 has quit IRC [05:41:02] <coffman> e^ipi: one side note, it worked with solaris 10 back in the days... [05:42:49] *** Gekz has joined #OpenSolaris [05:43:28] <e^ipi> what did? [05:43:36] <benley> his fat mp3 player thing [05:44:55] <coffman> oh dear, i hope i dont sound like that one guy with the ipod o_0 [05:49:07] *** dwc-_ has quit IRC [05:50:29] *** chendy has quit IRC [05:53:51] *** stukag has quit IRC [05:58:03] *** nitrile has joined #opensolaris [06:00:28] *** fry___ has joined #opensolaris [06:00:55] *** fry___ has left #opensolaris [06:06:10] *** asarch has joined #opensolaris [06:08:36] * sstallion yawns [06:09:31] *** bourgois has joined #opensolaris [06:14:11] *** fr4g has joined #opensolaris [06:26:24] *** ShadowHntr has joined #opensolaris [06:28:11] *** RohanRNS has quit IRC [06:28:18] <coffman> hm [06:28:38] <coffman> remind me to not use rdsk while capture a image with dd from a dvd o0 [06:30:29] *** stux is now known as stux|away [06:32:19] *** delphi1000 has joined #opensolaris [06:32:29] *** delphi1000 has left #opensolaris [06:40:45] *** axisys has joined #opensolaris [06:51:32] <sstallion> e^ipi: okay, package support is added to emanc. drivers now [06:51:51] <sstallion> it functions just like ON... cd to drivers/pkgdefs, and issue a make all install [06:52:18] <sstallion> packages are placed in PKGARCHIVE like ON (defaults to CODEMGR_WS/packages/$(MARCH)) [06:53:58] <phrost> any idea for dns resolution failure? can ping the dns server fine, resolv.conf is proper, nsswitch.conf has files dns for host lookup but every DNS query times out [06:54:07] <phrost> tried both local nameservers and public ones [06:58:12] <e^ipi> sstallion: sweet [06:58:19] <e^ipi> sstallion: by the way, we have a mailing list now [06:58:22] <sstallion> oh yeah ? [06:58:35] *** asarch has quit IRC [06:58:38] <e^ipi> yep [06:58:39] <e^ipi> http://mail.opensolaris.org/mailman/listinfo/emancipation-discuss [06:58:40] <sstallion> working on adding some automagic for compressing the packages for dist. [06:58:43] <sstallion> ahh good [06:58:48] <sstallion> i'll subscribe tonite [06:59:04] <sstallion> _dsw seems like an interesting chap ;P [06:59:22] <e^ipi> heh [06:59:24] <jbk> me too [06:59:34] *** delphi1000 has joined #opensolaris [06:59:47] *** delphi1000 has left #opensolaris [07:02:47] <sstallion> jbk: e^ipi: any reason why we wouldnt want to dist using bz2 ? [07:02:56] <e^ipi> i can't think of any [07:03:22] <jbk> same here [07:04:12] <sstallion> *nod* [07:11:14] <sstallion> OSOLre-11.11,REV=2008.09.26.22.47-sparc.tar.bz2 [07:11:22] <sstallion> comments for or against ? [07:12:09] <sstallion> (i'm tempted not to add targets for this since most people should not be using them) [07:12:58] <phrost> can you turn off udp/tcp segment offloading for a nic? [07:13:07] <sstallion> phrost: depends on the nic [07:13:20] *** axisys has quit IRC [07:14:07] <phrost> its a pos rge0, i think thats what the problem is [07:14:14] <phrost> about the only thing working right is ICMP and ARP [07:15:26] <sstallion> I'd check your rge.conf and the output of dladm show-linkprop [07:15:37] <sstallion> otherwise, check the driver source [07:16:05] <phrost> yeah i already went through show-linkprop, noting there relavent to TSO [07:17:06] *** chris_unix_dude has joined #opensolaris [07:18:09] <sstallion> i'm checking the rge source now ... one sec [07:18:46] <sstallion> there isn't an option to disable it [07:18:56] <sstallion> but hchksum is supported on some (but not all) devices [07:19:34] <phrost> i'm assuming it's a cheap onboard that they included a cheap non-supporting RTL nic and that's why TCP and UDP aren't working at all [07:19:56] <phrost> the box can UDP traceroute fine- but that's ICMP on the return trip- i think UDP/TCP TSO is causing the box to not get any inbound TCP/UDP at all [07:20:22] <sstallion> that seems odd... the rge driver is one of the nicer drivers out there [07:20:38] <sstallion> have you isolated everything (i.e. no firewalls, connected directly, etc.) ? [07:20:59] <phrost> its a blank sxce 98 install- i assume ipfilter is disabled by default [07:21:06] <phrost> didn't see it at all in `svcs | grep ipf` [07:21:25] <sstallion> have you directly connected to the nic with another machine and tried testing that way ? [07:21:55] <phrost> no, but i tried 3 ports on the same vlan on the cisco catalyst [07:23:25] <sstallion> isolate the device [07:23:32] <phrost> lemme grab the laptop and head back down to the solaris box [07:23:32] <sstallion> its completely possible there is an issue with the cat [07:23:42] <phrost> brb [07:32:23] <phrost> ok, if i hook directly in to it with the laptop and run snoop on rge0 i'm only seeing outbound- no inbound TCP or UDP at all [07:34:25] <phrost> RTL8168 chipset [07:35:59] <phrost> testing a fix from the forums [07:37:40] <sstallion> looks like hcksum is supported on that chip [07:37:55] <sstallion> its an easy hack if you have the source handy [07:38:09] <phrost> sure enough, thats the problem [07:38:11] <sstallion> any chance you are on a sparc ? [07:38:29] <phrost> setting 'set ip:dohwchksum = 0' in /etc/system solved it [07:38:35] <sstallion> ahh neic [07:38:37] <phrost> there's already a bug report filed by someone else [07:38:40] <sstallion> err nice [07:38:53] <sstallion> I didnt realize there was a global knob for that [07:39:06] <phrost> neither did i till i went digging through the forums [07:40:22] *** sfire||mouse has quit IRC [07:50:32] *** swa has quit IRC [07:52:16] *** Tilt has joined #opensolaris [07:56:26] *** chendy has joined #opensolaris [08:01:41] *** fr4g has quit IRC [08:12:24] *** netj has joined #opensolaris [08:14:17] *** gehmehgeh has joined #opensolaris [08:14:46] *** ky-san has quit IRC [08:17:49] *** ePax has quit IRC [08:17:53] *** ky-san has joined #opensolaris [08:21:09] *** netj has quit IRC [08:28:16] *** Cookie` has quit IRC [08:31:52] *** Tilt has quit IRC [08:33:34] <e^ipi> GmanAFK: ping? [08:36:47] *** dnm has joined #opensolaris [08:38:37] <jbk> hmm [08:40:34] <jbk> 'To complicate matters, part of the NFS code is in the closed repository' [08:40:38] <jbk> looks like lockd [08:40:46] <jbk> hopefully nothing new is in there [08:41:43] *** balbirs has left #opensolaris [08:44:51] <e^ipi> usr/closed can die in a fire [08:44:52] <e^ipi> ... [08:44:56] *** lkthomas has quit IRC [08:50:13] <sstallion> heh [08:50:19] <sstallion> okay guys, bz2's are now generated by default ;) [08:51:23] <sstallion> http://rafb.net/p/0eyiRd23.html [08:52:58] <sstallion> either of you guys on parallels at the moment? [08:53:18] <jbk> there's a way where you can convert a package so that all the actual files are put into a single bz file, and then uncompressed during pkgadd [08:53:26] <jbk> but i don't know how to generate it [08:53:53] <sstallion> well, I did it this way so that when it comes time to integrate with ON, we wont have to change anything [08:54:24] <sstallion> the Makefiles are damned near identical [08:54:47] <sstallion> there is just a new bz2 target which is tacked on as a dependency on the install target [08:56:38] <jbk> i think the stuff to do it is probably part of the install consolidation that hasn't been opened [08:56:41] <jbk> amusingly [08:57:01] <jbk> the utility to do it is documented in the application packaging guide, but apparently not available to anyone outside of sun [08:57:54] <sstallion> figures :/ [08:59:14] <phrost> any way you can have the zones not run the dtlogin by default? [09:00:02] <e^ipi> disable it? [09:00:28] <sstallion> phrost: log into the zone and disable the service (i.e. svcadm disable cde-login) [09:00:52] <phrost> but you can't script that at create time for a non-global zone? [09:01:03] <sstallion> phrost: create a template zone [09:01:15] <sstallion> normally on a box with several zones, I create a single template zone for each brand [09:01:26] <sstallion> in it, I customize all the usual crap and disable all but the bare minimum services [09:01:36] <sstallion> then rather than just create a new zone each time, I just clone the template [09:02:22] <e^ipi> yeah, i make a clone of my DNS server zone [09:02:41] <e^ipi> it's got minimal added on to it aside from my /etc/dns.foo [09:03:29] <sstallion> phrost: http://rafb.net/p/c4172F73.html [09:03:43] <sstallion> thats my main opteron box at home [09:04:05] <sstallion> (running 10u5) [09:04:59] <phrost> web stuff in lx? [09:05:22] <e^ipi> why would you do that? [09:05:31] <e^ipi> we have a web stack [09:05:34] <e^ipi> several, actually [09:05:35] <sstallion> e^ipi: I have a user which wants linux *shrug* [09:05:41] <sstallion> all of my stuff sits in solaris (web01) [09:05:58] <e^ipi> i meant the < phrost> web stuff in lx? comment [09:06:06] <e^ipi> i know why you would have a linux zone in general [09:06:08] <phrost> e^ipi: in his zone list he has a web instance running on lx [09:06:09] <phrost> lol [09:07:13] <e^ipi> I thought of buying a niagra and cutting it up, but unless you're rackspace/etc i don't think there's a lot of money to be made in the hosting business [09:07:14] <sstallion> i have a pretty fair amount of hardware here in the rack, so I have several users which vpn in to use it... that was a case where a user wanted a linux runtime [09:07:32] <e^ipi> sstallion: does your user know that he's running on solaris? [09:07:38] <sstallion> of course ;) [09:07:44] <sstallion> he doesn't notice it though [09:08:02] <sstallion> running rhel3u8 IIRC [09:08:18] <phrost> do zones plumb lo0 ipv6 by default? [09:08:31] <sstallion> if you define an entry for it yes [09:08:52] <sstallion> man inet_type [09:09:02] *** not-me-guv has joined #opensolaris [09:09:33] <sstallion> he does quite a bit on the sparc box, and its running snv_89 [09:09:44] <sstallion> so I'm slowly weaning him off of the linux b.s. [09:11:13] * phrost is trying to convert some rhel boxen [09:11:25] <sstallion> actually now that I think about it, i'm tempted to drop the svn zone and convert everything over to hg [09:12:27] *** Cookie` has joined #opensolaris [09:13:13] <phrost> hrm i created inet_default and set DEFAULT_IP=BOTH but the zone still wont bring up ipv6 on lo0 and is consequently causing sendmail to fail [09:13:57] *** ttmrichter has joined #opensolaris [09:14:09] <sstallion> check /etc/hosts [09:14:14] <sstallion> make sure there is an entry for ::1 [09:14:22] <phrost> there is [09:14:30] <phrost> it looks like another known bug [09:14:43] <sstallion> if you don't like sendmail's complaints, just disable IPV6 support in sendmail.cf (I do... I f'n *detest* that change) [09:15:07] <sstallion> i'll adopt ipv6 as soon as everyone else does :P [09:15:11] <phrost> i dont know if it's complaining- i think it might be fatal [09:15:20] <phrost> well maybe not [09:15:22] <sstallion> disable the option [09:15:28] <sstallion> open up the sendmail.cf [09:15:33] <sstallion> search for 'V6' [09:15:36] <phrost> pfft sendmail [09:15:41] <sstallion> there are two spots... comment out each line [09:17:34] <sstallion> oh well time for bed [09:17:35] <sstallion> nite all [09:19:10] <phrost> gn [09:26:32] *** chendy has quit IRC [09:26:41] *** chendy has joined #opensolaris [09:37:41] *** xRaich[o]2x has joined #opensolaris [09:41:06] *** cchapman has quit IRC [09:41:42] *** Gekz has quit IRC [09:41:55] *** Gekz has joined #OpenSolaris [09:53:56] *** bb3 has joined #opensolaris [09:57:44] <e^ipi> anyone happen to know anything about the status of zfs on osx? [09:58:07] <e^ipi> still totally crap and somewhat dangerous, or actually useful? [10:01:45] *** sartek has joined #opensolaris [10:14:11] *** harukomoto has quit IRC [10:19:22] *** Gekz has quit IRC [10:19:35] *** Gekz has joined #OpenSolaris [10:20:13] *** bb3 has quit IRC [10:22:39] *** gottadoit has quit IRC [10:29:26] *** sailorvrz_ has joined #opensolaris [10:29:54] *** anilg has joined #opensolaris [10:33:36] *** Fish has joined #opensolaris [10:34:19] *** Gekz has quit IRC [10:36:00] <codestr0m> e^ipi: where can I find out more details on the continued work of emancipation gate? [10:36:13] <e^ipi> we have a mailing list now [10:36:34] <e^ipi> http://mail.opensolaris.org/mailman/listinfo/emancipation-discuss [10:36:34] <codestr0m> ok. I'll sub.. I'm not sure I saw that info on your blog [10:36:58] <e^ipi> i have a very short attention span [10:37:04] <e^ipi> hence, i twitter much more than I blog [10:37:34] <codestr0m> I'll read the archives.. I'm pretty sure I got most things to compile yesterday.. I've not checked the kernel to see if it boots or validated anything else though [10:37:51] <codestr0m> and wasn't sure if you're pulling from onnv-gate and merging the changes with your work or how that's being done [10:38:07] <codestr0m> new list with no posts yet [10:38:22] <codestr0m> I'll let others know about the project as I think some may be interested in this [10:39:01] <e^ipi> there aren't archives [10:39:13] <e^ipi> as the mailing list has existed for ~ 6 hours [10:39:29] <e^ipi> also, the kernel can't boot [10:39:31] <codestr0m> ok. that would explain it [10:39:34] <e^ipi> since the kernel can't build [10:39:39] <e^ipi> since the gate can't build [10:39:41] <e^ipi> because I broke it. [10:39:45] <e^ipi> on purpose [10:39:56] <codestr0m> since when.. I played with some things and got it to build yesterday [10:40:01] <e^ipi> o_O [10:40:05] <e^ipi> i doubt that [10:40:14] <codestr0m> I'll upload the nightly log [10:40:36] <codestr0m> ==== Nightly distributed build completed: Sat Sep 27 19:24:30 CEST 2008 ==== [10:40:36] <codestr0m> ==== Total build time ==== [10:40:36] <codestr0m> real 1.9932825:59:35 [10:40:50] <codestr0m> I confirmed some of the stuff for sanity, but not a lot [10:42:42] <e^ipi> did you merge with onnv-gate ? [10:43:54] <codestr0m> well. since there was no documentation.. I just started hacking [10:44:20] <codestr0m> I wasn't even sure I should ask you about this... I know you're pretty busy [10:44:40] <codestr0m> how often do you want to merge.. is anyone interested in doing this.. etc.. [10:44:50] <e^ipi> merging is the last of my concerns [10:45:23] <codestr0m> when was the last merge you did and from what hg changeset did it start.. I think I can figure most of this out though [10:45:25] <e^ipi> considering merging implies a bunch of work to un-fuck the emancipation gate [10:46:12] *** ninjaslim has quit IRC [10:46:21] <e^ipi> work that can be better spent actually being productive [10:46:29] <codestr0m> ok. well. don't know what your plans are, but thanks [10:46:39] <e^ipi> my two changesets are 1f22392696c1 and 5413315b90a4 [10:46:57] <e^ipi> if 1f22392696c1 isn't tip, it's not emancipation-gate [10:47:45] <codestr0m> that guy erast was around he should be interested in this, but frankly I'm not sure I care to even tell him about this [10:48:00] *** slash^ has quit IRC [10:48:08] *** yarihm has joined #opensolaris [10:48:10] <e^ipi> whatever, more hands are more hands [10:48:20] <codestr0m> yeah, but I doubt he'll contribute back [10:48:32] <e^ipi> well, you don't know that [10:48:36] <e^ipi> unfair to pre-judge [10:48:42] <codestr0m> he's supposed to have done work for nexenta about the bootstrap and has scripts for O/N but won't release it [10:49:11] <e^ipi> *shrug* [10:49:35] <codestr0m> exactly.. I'll focus on being productive and let other people know who are much better coders than me [10:49:49] <e^ipi> "unproductive" isn't a big deal so long as nobody's stalled on it [10:50:11] <codestr0m> well. I had to change the nightly script to remove checks for the binary bits [10:50:20] <codestr0m> then had to change some paths for my funked dev env [10:50:24] <codestr0m> and a few other things [10:50:27] <e^ipi> umm... nightly doesn't check [10:50:29] <e^ipi> the makefiles check [10:51:03] <e^ipi> go check the emancipation pages again, i threw some more stuff up [10:51:07] <codestr0m> check_closed_tree() { [10:51:07] <codestr0m> 1012 if [ -z "$CLOSED_IS_PRESENT" ]; then [10:51:24] <e^ipi> yeah, that checks if you're using closed_bins or if you work at Sun [10:51:26] <codestr0m> ok. sec. I'll find it again [10:51:55] <e^ipi> in either case, the long and short of it is that emancipation-gate still needs closed_bins. you strip libc_i18n.a out of them though [10:52:02] <codestr0m> well. I set opensolaris.sh correctly and in the end just changed the check and things started progressing [10:52:14] <e^ipi> and then it fails to bulid, since libc's quite broken at that point [10:52:26] <e^ipi> libc will also bail entirely if you don't fix it's makefiles [10:52:40] <e^ipi> the hg changesets that i posted, and the emancipation gate both reflect that [10:52:47] <e^ipi> they've already got broken makefiles [10:53:00] <e^ipi> ( and a fixed mapfile-vers ) [10:53:05] *** slash^ has joined #opensolaris [10:56:10] <codestr0m> changeset: 5841:9940b9a8ec65 [10:56:10] <codestr0m> tag: tip [10:56:16] <codestr0m> I guess I pulled something different [10:56:20] <e^ipi> not my gate [10:56:35] <codestr0m> hg clone ssh://anon at hg dot opensolaris.org/hg/emancipation/emancipation-gate [10:57:08] <codestr0m> what's the correct one so I can sync [10:57:46] <e^ipi> ssh://hg.opensolaris.org/hg/emancipation/i18n-gate [10:58:03] <e^ipi> i think you'll probably need to start from a fresh clone [10:58:12] <e^ipi> i dumped emancipation-gate a while back [10:58:34] <e^ipi> no way to delete it off the website entirely, but i've removed references to it as best I could [10:58:41] <e^ipi> I apologize [10:59:01] <codestr0m> does i18n-gate == full O/N and if I may ask why you dumped emancipation gate? [10:59:13] <codestr0m> sorry if I'm digging up deprecated work [10:59:15] <e^ipi> because my entire approach was wrong [10:59:33] *** shaftyy has joined #opensolaris [10:59:35] <e^ipi> what I did was look for vaguely related freebsd code and imported that [10:59:49] <codestr0m> you commented on this before and I saved the notes, but any more details and I'll save as well [10:59:55] <codestr0m> yeah. that's in your blog [11:00:03] <codestr0m> I thought you were making good progress with it? [11:00:12] <e^ipi> what I should have done ( and what I resolved to do with the new gate ) is implement bare-bones functionality to support POSIX/C only, so that we have /something/ that works [11:00:33] <e^ipi> one can continue to make progress on a dead end road... [11:00:51] <codestr0m> the other approach didn't work as the fbsd code was that borked? [11:01:13] <e^ipi> it wasn't borked so much as freebsd's libc is quite a bit different from ours [11:02:19] <codestr0m> ok. and that's the work in emancipation-gate [11:02:25] <codestr0m> I'm cloning the new work [11:02:48] <codestr0m> and also if you want to update your blog that may prevent more curious people stumbling upon it in the future [11:04:39] *** Gekz has joined #OpenSolaris [11:05:43] <codestr0m> "Note that this gate does not build cleanly currently. It will fail, and you'll get a list of missing symbols that the linker couldn't find. Those are the ones we're after" [11:05:50] <codestr0m> k. that's a very clear todo list then [11:10:04] *** thana has joined #opensolaris [11:10:21] <thana> hi [11:13:17] <e^ipi> i already posted a todo list [11:13:29] <e^ipi> http://opensolaris.org/os/project/emancipation/tasks/libc_i18n/i18n_symbols/ [11:13:39] *** Gekz has joined #OpenSolaris [11:17:38] <thana> zfs is really nice... but it would be even nicer if snapshots would work... [11:17:54] *** saablover has joined #opensolaris [11:18:11] *** anilg has quit IRC [11:19:42] <saablover> hi folks [11:19:48] <saablover> anyone awake + [11:19:49] <saablover> ? [11:23:00] *** bb3 has joined #opensolaris [11:23:35] <e^ipi> thana: they do work. [11:23:39] <e^ipi> saablover: no. [11:24:25] *** relling1 has joined #opensolaris [11:25:32] <saablover> I am playing with beadm on a separate zpool [11:25:35] <saablover> but it fails all the time [11:26:12] <e^ipi> search the lists, file a bug [11:26:17] <thana> e^ipi: actually not. just tried the 5th time or so to get xvm installed and every time it drove my system anbootable. so i decided to rollback the snapshot i took after installation and now... well its even worse. before the rollback i got the chance to use a rescue console. now it talks about unknown symbols in the boot archive -- how can it miss anything? after installation wehen i took the snapshot it was there... [11:27:07] <e^ipi> get xvm installed? [11:27:15] <e^ipi> you mean press the down arrow when grub boots? [11:27:31] <tsoome> boot from net/cd and rebuild the boot archive [11:27:56] <thana> e^ipi: no. my 2008.05 doesn't know anything about xvm [11:29:26] *** duck4 has joined #opensolaris [11:30:44] *** duck4 has quit IRC [11:31:13] *** pjfloyd has joined #opensolaris [11:31:54] <palowoda> Maybe Indiana wasn't a good choice for xvm (aka Xen) testing. [11:32:18] <saablover> it should work [11:32:22] <saablover> xen on indiana [11:32:27] <saablover> install these packages [11:32:36] <saablover> SUNWxvmhvm [11:32:40] <saablover> SUNWvirtinst [11:32:43] <saablover> SUNWlibvirt [11:32:47] <saablover> SUNWurlgrabber [11:32:52] <saablover> and it should do it's thing [11:33:03] <palowoda> There you go problem solved. [11:33:09] <saablover> look at this url [11:33:17] <saablover> http://malsserver.blogspot.com/ [11:33:19] <palowoda> By the way which version of Indiana? [11:33:33] <thana> trying to do this it threw some errors and destroyed the system. even the rollback of a snapshot taken before doesn't after installing these packages... [11:33:35] <saablover> upgrade to the latest one via pkg image-update first :) [11:34:22] <palowoda> thana: Did you upgrade to the version that saablover indicated? [11:34:45] <saablover> kernel build 98 should work fine [11:35:11] *** relling__ has quit IRC [11:36:03] *** likaijun has quit IRC [11:37:37] <thana> palowoda: i did. and as expected pkg wasn't able to fully execute it... [11:37:57] <saablover> then I would reinstall my box [11:38:04] <saablover> I think you have destroyed something [11:38:04] <thana> at least after that image-update the rollback worked :-/ [11:39:11] <thana> i reinstalled it several times. it is either quite unstable or the manpages doesn't tell anything about this or that option/program kann destroy everything... [11:39:41] <palowoda> Did you do a clean install of Indiana build 98? [11:39:58] <palowoda> So many builds to choose from you know. [11:40:12] <saablover> no :) [11:40:17] <saablover> 98 is the latest one [11:40:23] <saablover> just do pkg image-update [11:40:30] <saablover> and it should update to the latest bits [11:40:35] <palowoda> Heh [11:40:39] <thana> palowoda: i think so, yes [11:40:51] <palowoda> maybe [11:41:19] <palowoda> ips update has a few bug-a-boo's you know. [11:41:30] <saablover> hehe I am a Sun engineer :) It's my mission to say it works :) [11:42:01] <thana> :) [11:42:09] <palowoda> I'm an ex Sun engineer. Continue on your mission. [11:42:31] <saablover> ex ? left the ship ? [11:42:59] <palowoda> Are you saying I should have sold more of my stock? [11:43:19] <thana> i'm a frustrated sun customer it's my mission to mkae your mission fail ;) [11:43:28] <saablover> hehe [11:43:38] <palowoda> Damn my mission is fishing. [11:44:27] <palowoda> You can tell the little fishies not to bite the hook if it makes you feel better. :) [11:46:01] <thana> so you say i should reinstall the system an perform an pkg image-update after installing? [11:46:03] <saablover> why doesn't this work :) [11:46:10] <saablover> thana: yes [11:47:02] <e^ipi> saablover: my guess is that you did an image-upgrade without reading the instructions [11:47:12] <e^ipi> the first release had a nasty IPS bug that bricked systems [11:47:16] <palowoda> maybe a bingo [11:47:20] *** pjfloyd has left #opensolaris [11:47:21] <saablover> no no [11:47:24] <sailorvrz_> yay my serial lom cable works :D [11:47:41] <e^ipi> err, image-update [11:47:42] <saablover> my zpool is on another disk [11:47:56] <e^ipi> don't really know how that's relevant [11:48:01] <saablover> the updating of the grub menu lst get's confused I guess [11:48:15] <palowoda> Just install build 98 of indiana [11:48:15] <saablover> I don't have problems with pkg image-update :) [11:48:37] <saablover> I am trying to install a second boot environment to do research stuff [11:48:42] <saablover> on another disk [11:48:48] <saablover> and that fails [11:48:48] <e^ipi> man beadm ? [11:48:56] <saablover> look what I try to do [11:49:06] <saablover> zpool: zfsroot is on another disk [11:49:16] <saablover> I do: beadm create -p zfsroot embedded [11:49:18] <e^ipi> it's also possible that the functionality is just simply not supported yet, to which my reply will be "file a bug" [11:49:34] <saablover> where embedded is the name of my new environment [11:49:39] <saablover> but it does not work :) [11:49:50] <e^ipi> do remember that most of the technology surrounding that product is quite new and quite untested [11:50:02] <saablover> that's why I like to test it :) [11:50:14] <saablover> this is quite amazing when it works :) [11:50:16] <e^ipi> and file bugs when you run in to stuff like this [11:50:20] <palowoda> And you should log a bug. [11:50:25] <saablover> I will [11:50:34] <e^ipi> defect.opensolaris.org [11:51:22] * e^ipi should really put his money where his mouth is and start importing some of the d.o.o stuff in to bugster [11:51:54] *** comay has quit IRC [11:52:10] <palowoda> Good lord the bugster d.o.o stuff needs something to be desired. [11:52:21] *** comay has joined #opensolaris [11:52:27] <e^ipi> bugs.os.o you mean? [11:53:20] <palowoda> The whole bugster <-> defects database sync maintance is just a pain. [11:53:34] <e^ipi> d.o.o doesn't have a sync [11:53:45] <e^ipi> the sync is "engineer reads d.o.o, files bugster bug" [11:53:48] *** relling__ has joined #opensolaris [11:53:54] <palowoda> Yeah I know. [11:54:08] <e^ipi> there ought to be some sort of bridge that way though, yeah [11:54:09] <palowoda> It can't go on forever. [11:54:54] <saablover> hmm is this related to install or the packaging ? [11:55:46] <palowoda> saablover: Most likely camian installer. [11:55:54] <palowoda> A guess though. [11:56:05] <saablover> I think packaging [11:56:13] <palowoda> Because? [11:56:24] <saablover> nah [11:56:26] <saablover> it's installer [11:57:35] <saablover> I will try with a file based zpool [11:57:38] <saablover> on the same disk [11:57:46] <saablover> let's see what this does [11:58:59] <palowoda> I don't understand what a "file based zpool" means. [11:59:14] <saablover> hehe [11:59:19] <saablover> you can use zfs on a file too [11:59:36] <saablover> mkfile 6G testfile [11:59:44] <saablover> where testfile is 6 gigabyte big [11:59:53] <saablover> zpool create testing testfile [11:59:57] <saablover> voila :) [12:00:02] <saablover> this is a very neat feature [12:00:08] <saablover> but I need to eat [12:02:51] *** relling1 has quit IRC [12:04:20] <palowoda> Enjoy the mushrooms and report back the d.o.o bugid. [12:07:34] *** art-vandelay has joined #opensolaris [12:08:21] *** art-vandelay has quit IRC [12:08:27] *** bb3 has quit IRC [12:10:19] *** art-vandelay has joined #opensolaris [12:12:42] *** AxeZ has joined #opensolaris [12:34:39] <kimc> installed the COMSTAR iSCSI port provider as described in benr's blog: http://www.cuddletech.com/blog/pivot/entry.php?id=968 [12:35:15] *** revlo has joined #opensolaris [12:36:06] <kimc> works great.. provides iSCSI with a 750GB ZFS pool [12:37:40] <kimc> for test purposes have now connected a win vista machine to the iscsi target and formatted the space with ntfs [12:38:15] <kimc> then changed the media center video recorder config to record to that space [12:39:23] *** chendy has quit IRC [12:39:42] <trygvis> neat [12:41:00] <kimc> this yields ~700GB of useable recording space.. 82 hours in HD mode [12:41:56] *** gehmehgeh has quit IRC [12:46:53] <kimc> when the iscsi target (non-comstar type) was announced several months ago the reliability of the win media center video recording was less than it is today [12:47:48] <kimc> a few bugs have been fixed in the target and also some windows iscsi compatibility problems worked-around [12:48:44] <kimc> it would be interesting to see how the reliability of the non-comstar iscsi target compares today.. probably much better [12:50:41] *** sparc has joined #opensolaris [12:52:10] <kimc> the comstar iscsi target is: online 17:50:27 svc:/network/iscsi/target:default [12:52:59] <kimc> the non-comstart iscsi target is: disabled 17:50:06 svc:/system/iscsitgt:default [12:55:36] <kimc> [root@solster ~]> itadm list-target [12:55:36] <kimc> TARGET NAME STATE SESSIONS [12:55:36] <kimc> iqn.1986-03.com.sun:02:70560336-6324-4c20-ba60-d2a9bb40ec38 online 1 [12:56:22] *** derchris has quit IRC [12:56:24] <trygvis> though, how is this different from the normal zfs+iscsi support? [12:56:26] *** derchris has joined #opensolaris [12:59:58] <kimc> with comstar the design is completely modular: http://www.opensolaris.org/os/project/comstar/ [13:00:09] <trygvis> right, I see that part [13:01:09] <kimc> i dont think the original zfs+iscsi target ever implemented mpxio ..this does [13:04:42] <kimc> says here (on the page linked) that comstar is available 'by default after you install the latest sxce' [13:05:17] <kimc> that doesn't appear to be the case at the moment [13:05:25] <IvanR_> Plus port providers can be built on top of comstar, so more than iscsi can be built on top of it, like FC, FCoE, SAS. [13:08:33] <kimc> sure [13:08:41] <IvanR_> Take a thumper, put it in the rack, plug it into FC, and start allocating lun^H^H^Hzvols from the array^H^H^H^H^Hpool on your SAN^H^Holaris system to your server hosts. [13:10:51] <kimc> with the prices of 10 GBit ethernet hardware coming down you might not want FC [13:11:39] <IvanR_> I imagine that's why there's FCoE too. [13:16:49] *** nemski is now known as SovietNemski [13:17:43] *** SovietNemski is now known as NaziNemski [13:18:44] <ShadowHntr> lol [13:18:55] *** NaziNemski is now known as ComradNemski [13:19:46] *** psychonate has joined #opensolaris [13:25:27] <jmcp> kimc: mpxio is a separate layer to zfs and iscsi [13:25:36] <jmcp> iscsi is a transport which sits below zfs [13:25:58] <jmcp> and as far as I'm aware, iscsi on OpenSolaris uses mpxio by default [13:27:32] *** lucianno has joined #opensolaris [13:36:59] *** sparc has left #opensolaris [13:42:13] <Pietro_S> is there any mailing list /gug reporting for suncc compiler? [13:42:33] <kimc> i didn't think mpxio was working on OpenSolaris with the original iscsitgt [13:44:22] <kimc> i'm aware its a seperate layer.. i seem to remember some talk on one of the lists suggesting it has only been working recently no? [13:46:46] <kimc> lets see here.. why FCoE: http://www.opensolaris.org/os/project/fcoe/ [13:47:43] <kimc> heh.. reduced cooling requirements [13:48:35] <kimc> renewable Fibre Channel [13:48:57] <jmcp> kimc: I wasn't aware of any problems with mpxio and iscsitgtd [13:49:16] <jmcp> my work is closely related to the MPxIO team, so I'd expect to hear if there had been any [13:49:25] <kimc> ok very good.. prolly my confusion ) [13:49:44] <jmcp> nothing to worry about hear, move along to the next conversation point please.... [13:49:44] <jmcp> :) [13:50:27] <kimc> what is the difference between MPxIO and multipathing?> [13:50:37] <jmcp> naming? [13:50:42] <kimc> ahh.. [13:51:01] <jmcp> Sun's implementation of multipathing is called, technically, the Solaris Multipathed I/O feature [13:51:28] <jmcp> the initial project name was "mpxio" [13:51:54] <jmcp> when Marketing got involved, it got changed - officially - to StorEdge Traffic Manager Software [13:51:56] <jmcp> hence "stmsboot" [13:52:11] <jmcp> just like "zones" became "N1 Grid Containers" [13:52:20] <jmcp> and everybody still calls them zones, and they use "mpxio" [13:52:22] <jmcp> :-) [13:52:26] <kimc> multiple ethernet connections: 2x gig ethernet ~= 2 gig fibre ? [13:52:34] <TomJ> I've never heard N1 Grid before [13:52:43] <jmcp> TomJ: lucky you [13:52:51] <Tempt> 2 x gig-E != 2Gbit FC [13:52:58] <jmcp> kimc: I dunno. I'm not sure how much overhead the tcp/ip stack has [13:53:09] <jmcp> kimc: Tempt would know some practical figures, I'm sure [13:53:20] <TomJ> the 2GB FC would be faster, right? [13:53:27] <Tempt> Well, it's more that an aggregated connection isn't just like having double the bandwidth [13:53:31] <kimc> but the overhead is the factor reducing from fully 2x [13:53:49] <Tempt> It isn't just IP stack overhead, it's the way link aggregation works. [13:53:57] <kimc> ahh ok [13:53:58] <_mary_kate_> also, FC is not a connection-oriented protocol like TCP, is it? [13:54:04] <_mary_kate_> (it just sends scsi packets) [13:54:27] <jmcp> _mary_kate_: FC encapsulates scsi packets in FC frames [13:56:43] <Tempt> jmcp: And did you accuse me of practicality? My word. [13:56:51] <jmcp> heh [13:59:18] <Tempt> Anyway, it's still the weekend, I'd rather not think about the endless hassles of storage. [14:02:41] <jmcp> yeah [14:02:49] <jmcp> how about ranting on IB instead? [14:03:44] <Tempt> Actually, I've managed to ignore anything involving UNIX & rants all weekend [14:04:14] <jmcp> good for you! [14:04:25] <Tempt> I've had new toy distractions. [14:04:33] <jmcp> :) [14:04:40] <trygvis> like what? AIX? [14:04:48] <Tempt> trygvis: No. [14:05:00] <trygvis> or like a girlfriend? [14:05:08] <Tempt> jmcp: What about your weekend? [14:05:27] <jmcp> gardening, pool cleanup... [14:05:28] <Tempt> trygvis: Still the same one. [14:05:38] <jmcp> a rather non-IT weekend [14:05:42] <Tempt> No pool cleaning slave? [14:05:51] <jmcp> Kreepy Krauly [14:06:11] <jmcp> only a few bits that needed an extra scrub [14:06:25] <jmcp> J can't help cos the impending arrival is really making life hard [14:06:35] <jmcp> btw, *never* ask me to prune a shrub for you [14:06:52] <Tempt> Aah, how long before your family adds another member? [14:06:58] <jmcp> early Jan [14:07:41] <Tempt> Not that long then. [14:07:46] <jmcp> not long at all [14:07:50] <Tempt> And at the hottest time of the year, too. [14:08:14] <jmcp> nope, that's more February round these parts [14:08:19] <jmcp> but still, it'll be hot [14:08:23] <jmcp> and we'll enjoy the pool [14:08:38] <Tempt> Been quite warm this weekend down here. [14:08:41] <Tempt> Was 29 yesterday. [14:09:27] <thana> should't the system set up a grub entry for xvm automatically installing the xvm packages? [14:09:49] <jmcp> Tempt: we were expecting 30+ here today, didn't happen :( [14:09:59] <jmcp> thana: are you not seeing any grub entries for xVM ? [14:10:07] <thana> true [14:10:18] <Tempt> jmcp: The longer is takes to cross that 30 barrier, the better. [14:10:23] <Tempt> jmcp: I don't need to be running the air con yet. [14:11:05] *** ComradNemski is now known as ChancellorNemski [14:11:19] <Tempt> Although for the first time in living memory I'm actually looking forward to daylight saving. [14:11:42] <jmcp> heh [14:13:00] <Tempt> No patching hundreds of machines, and I'll appreciate the chance to be outside taking pictures before the sun goes down. [14:15:35] *** spiki has quit IRC [14:23:20] <Yorlik> Sorry for the n00b question: How do I assign a variable in a shell script a value derived from a command execution in backticks ? Seems like variable=`sed somestuff` doesnt assign the rresult, but I can echo it. [14:23:41] *** nbkr has joined #opensolaris [14:23:59] <Tempt> pastebin what you've done and what you get [14:24:07] <Yorlik> oki [14:25:08] <TomJ> hehe, the Signapore grand prix is running at night and using 3.6 megawatts of power for lights. Global warming, they've heard of it. [14:26:01] <Tempt> *shrug* [14:26:13] <Tempt> Not that much and not for very long. [14:26:31] <ballChalk> is it coming from a coal plant? [14:26:39] <TomJ> hamsters [14:26:53] *** ShadowHntr has quit IRC [14:27:09] *** gottadoit has joined #opensolaris [14:27:48] <jmcp> gnite all [14:27:48] * jmcp sleeps [14:29:35] <IvanR_> thana: Did you run bootadm -m upgrade ? [14:30:04] *** derchris has quit IRC [14:39:13] *** CosmicDJ has quit IRC [14:39:59] *** CosmicDJ has joined #OpenSolaris [14:40:57] *** stukag has joined #opensolaris [14:42:18] <thana> IvanR_: no. didn't knew this tool yet. thanks [14:44:58] *** derchris has joined #opensolaris [14:45:41] <thana> IvanR_: the manual doesn't say anything about a switch -m or a subcommand upgrade... [14:47:36] <thana> it's not the problem to manually update grubs menu.lst but i wonder if anything went wrong since i read grub was automaticalle updated [14:48:42] <IvanR_> I got it from here: http://opensolaris.org/os/community/xen/docs/setupsoldom0/ [14:49:30] <IvanR_> Though I haven't spent much time with xvm, besides booting up for a little while the other day on my sxce machine. [14:49:58] *** freakazoid0223 has joined #opensolaris [14:50:07] *** not-me-guv has quit IRC [14:56:07] *** houst0n- has joined #opensolaris [14:59:09] *** Odin- has joined #opensolaris [15:02:06] *** relling1 has joined #opensolaris [15:05:29] *** nbkr has quit IRC [15:05:55] *** dclarke_away is now known as dclarke [15:07:14] *** evocallaghan has joined #opensolaris [15:09:49] *** ChancellorNemski is now known as nemski [15:12:16] *** relling__ has quit IRC [15:16:38] <lucianno> hi guys [15:17:56] <lucianno> A partition of 10 mb is enough to install Solaris 10? [15:18:06] <_mary_kate_> uh... no [15:18:06] <TomJ> do you mean 10gb? [15:18:52] <lucianno> oh, yeah! [15:19:02] <TomJ> a full install of solaris 10 uses about 5GB of disk [15:19:05] <TomJ> so 10 gb is possible, though tight [15:19:16] <TomJ> of course there is swap too [15:19:28] <TomJ> if you're just trying it out then 10gb is ok, but you'll want more for real usage [15:21:49] <lucianno> I want use it soon, but for now want to test it [15:23:15] <lucianno> TomJ : thanks much [15:34:06] *** dclarke is now known as dclarke_away [15:34:42] <Yorlik> I have an error in a shell script I can't resolve. ANy helpe for some newbie scripter ?: http://pastebin.com/d66a401c5 [15:35:05] *** Fish has quit IRC [15:35:09] <Yorlik> I don't understand why the conditional doesn't work. [15:35:19] <TomJ> Yorlik: spaces around [[ ]] [15:35:22] <TomJ> line 15 [15:35:29] <Yorlik> OH CRAP ! [15:35:36] <TomJ> you dont need ; at the end of every line [15:35:36] <Yorlik> I hate this stuff ... thanks ! [15:35:38] <TomJ> (thoug hit's harmless) [15:35:48] <TomJ> you use ` ` in one place and $() in another [15:35:54] <TomJ> standardise on $() unless you're writing for /bin/sh [15:37:00] <TomJ> you can do the two seds in one with : sed 's/blah/foo/;s/bar/foo/' [15:38:12] <TomJ> and you can avoid sed completely with: file=${each/libexec\/amd64\//} ; file=${file/.so/} [15:38:57] <TomJ> oh, in the [[ you want spaces around the = too [15:39:11] <TomJ> if [[ $test = 'mod_' ]] ; then .. [15:40:50] <TomJ> you have something weird going on with your quotes in line 17 [15:41:00] <TomJ> not sure why the .c isn't in quotes too [15:42:59] *** MrBIOS_ has joined #OpenSolaris [15:43:24] *** dclarke_away is now known as dclarke [15:43:37] <Yorlik> Now the condition is always true ... even if test<>mod_ [15:44:39] <Yorlik> I wonder if it is comparing the lebgth od the strings and not the content [15:45:20] <TomJ> pastebin [15:50:03] *** gerard13 has joined #opensolaris [15:50:11] <Yorlik> Here: http://pastebin.com/d3e2b4a46 [15:50:57] <Yorlik> not to say i removed all *.conf before the run .. [15:52:23] * Yorlik gets bashed by Sir Bash-a-lot [15:53:49] <Dominic> Yorlik: as TomJ said, you need spaces around the '=' [15:54:07] <Yorlik> Argh ... put the spaces in the wrong place .. thanks! [15:54:11] <Dominic> each part is an argument to test [15:55:24] *** joes has quit IRC [15:56:31] <Yorlik> Doh -- now it was all false, but [[ $test = 'mod_' ]]finally ficed it. Thanks a lot ! [15:57:18] <Yorlik> If you don't do scripting every day it can really become a pain in the b... [15:59:34] <lucianno> TomJ : in installation the system determines the size of the slice automatically for me? [16:02:38] *** shaftyy has quit IRC [16:05:38] *** coffman has quit IRC [16:07:35] *** lucianno has quit IRC [16:10:17] *** sponix has joined #opensolaris [16:11:03] *** sponix has quit IRC [16:15:19] *** shaftyy has joined #opensolaris [16:32:36] *** f0cus has joined #opensolaris [16:32:46] <f0cus> hi [16:33:33] *** jfndi_ has joined #opensolaris [16:36:34] *** gerard13 has quit IRC [16:36:39] <f0cus> how to install perl on Solaris 9 sparc machine? [16:39:48] <Stric> anything wrong with /usr/bin/perl ? [16:40:01] <jamesd> f0cus, it should be there allready, or sunfreeware.com or blastwave.org [16:41:55] *** dclarke is now known as dclarke_away [16:43:45] *** luc^ has joined #opensolaris [16:44:04] *** c00p has quit IRC [16:44:15] *** c00p has joined #opensolaris [16:49:10] *** user_ has joined #Opensolaris [16:49:35] <user_> hi from spain [16:50:15] *** sartek has quit IRC [16:54:01] *** karrotx has quit IRC [16:55:07] *** user_ has left #Opensolaris [16:56:50] *** pogma has joined #opensolaris [16:57:18] *** lucianno has joined #opensolaris [16:58:10] <pogma> Hi, I thought defect.opensolaris.org was the preferred bug tracker? Is it not? We opened http://defect.opensolaris.org/bz/show_bug.cgi?id=3450, but someone just copied the bug into bugs.opensolaris.org and closed the one on defect.opensolaris.org?! [16:58:28] <pogma> (pretty annoying) [16:59:08] <lucianno> why i dont see the windows partition in hte list instalation? [16:59:11] *** karrotx has joined #opensolaris [17:00:03] *** Odin- has quit IRC [17:03:54] <Yorlik> How can i connect to a lost PTY, when ssh crashed and I didn't use screen ? [17:04:25] <Yorlik> PTY/pts [17:04:54] <Pietro_S> is there c++ lint in sstudio?> [17:08:25] <ZOP> well now this is interesting...whatever opensolaris b97 installer does to the boot blocks my BIOS on my tyan S2937 really doesn't like it. [17:08:37] <ZOP> Yorlik: you can't. [17:08:58] <Yorlik> Okay .. thanks .. time to get screen ... [17:09:04] <ZOP> Yorlik: the apps connected to it received SIGHUP and should no longer be running. [17:09:14] <ZOP> (if they behave themselves) [17:10:00] <Yorlik> is screen or a similar tool in sxce ? [17:12:26] <ZOP> idk, i've been running nexenta. which it looks like i'll continue to have to do since sxce and opensol eitehr don't boot or have broken e1000 drivers. [17:16:13] *** chonan has joined #OpenSolaris [17:16:31] <ZOP> (couldn't get resolver happy during install and now post install my BIOS locks up during init and detect of bootable devs) [17:20:02] *** wurlitzer has joined #opensolaris [17:20:24] <IvanR_> Yorlik: If the process is still running, you can use nohup -p to redirect its output to a file. [17:21:03] <Yorlik> It was just a mcedit i did on a file. [17:21:47] <IvanR_> Then it's likely gone, nohup wouldn't help with anything that's trying to read input. [17:27:25] *** lucianno has quit IRC [17:28:57] *** ttmrichter has quit IRC [17:32:09] *** medar has quit IRC [17:32:40] *** DaRtH_VaDeR has joined #opensolaris [17:33:44] <seanmcg> Yorlik, screen isn't yet in there, theres a psarc case though: PSARC/2008/413 [17:34:02] <Yorlik> found it in SFE in the meantime [17:34:20] * Yorlik likes SFE [17:35:49] *** ninjaslim has joined #opensolaris [17:37:04] *** mikl has joined #opensolaris [17:42:00] *** evocallaghan has left #opensolaris [17:45:47] *** medar has joined #opensolaris [17:45:54] *** Fish has joined #opensolaris [17:47:40] <Yorlik> I gave my secondary root the roles "Primary Administrator" and "Software Installation" and "All" with usermod -p '<roles>' but still cannot use zlogin. what else would be needed? [17:50:14] <CosmicDJ> Yorlik: man ppriv [17:50:21] <seanmcg> Zone Management [17:50:30] <Yorlik> Thanks [17:51:40] <seanmcg> found via: grep zlogin /etc/security/exec_attr [17:52:45] *** outworlder has joined #opensolaris [17:52:59] <outworlder> how do I update the nvidia driver shipped with opensolaris? [17:53:29] <seanmcg> that driver in opensolaris is usualy very upto date already [17:54:16] <seanmcg> though there are the downloadable ones from the nvidia website, which also have the directions to install/upgrade them iff they are updated versions [17:55:04] <codestr0m> I need to look at the nightly flags, but I'm trying to disable lint from running and speed up my O/N builds [17:56:25] <codestr0m> can I just drop the l NIGHTLY_OPTIONS="-FNnaCDlmr" and other suggestions anyone? [17:59:09] <seanmcg> I use '-MNazmni', though they're a tad old now, it doesn't build the debug binaries either. [17:59:36] <seanmcg> If you don't want to test all compiling with gcc then modify the shadow compiler flags too [18:00:01] *** outworlder has quit IRC [18:08:04] *** DaRtH_VaDeR is now known as DaRtHo|aWay [18:11:12] <saablover> how can you list the loaded modules to installed packages ? [18:11:35] <saablover> modinfo and then pkg contents ? [18:16:55] *** karrotx has left #opensolaris [18:18:41] <saablover> anyone still alive and kicking + [18:18:43] <saablover> ? [18:19:17] *** MrBIOS_ has quit IRC [18:19:25] *** f0cus has quit IRC [18:21:15] <thana> i just pluged of the sata cable of one of the root-discs to check that one of them can fail and the system is still bootable. it is :) but after plugging in the cable again, ho can i make the system recognizing the disc without rebooting again? [18:21:51] *** mikl has quit IRC [18:22:02] *** art-vandelay has quit IRC [18:22:21] <CosmicDJ> thana: is your hw hot-swapable? [18:22:57] *** bug_L has joined #opensolaris [18:23:00] *** libkeise1 has joined #opensolaris [18:24:11] *** art-vandelay has joined #opensolaris [18:24:12] <thana> CosmicDJ: aren't all sata dicsc supposed to be? [18:24:25] *** art-vandelay has quit IRC [18:28:44] *** sartek has joined #opensolaris [18:28:51] *** chris_unix_dude has quit IRC [18:31:02] <sartek> i saw today a fu-si notebook with opensolaris "preloaded" [18:31:20] *** jfndi_ has quit IRC [18:31:42] <Pietro_S> sartek: do you have a link for that easter egg? ;-)_ [18:32:01] <sartek> Pietro_S: at a super market thingy [18:32:14] *** _luc^ has joined #opensolaris [18:34:42] *** luc^ has quit IRC [18:36:00] *** sparkleytone has quit IRC [18:36:21] *** libkeiser has quit IRC [18:44:54] *** mikl has joined #opensolaris [18:45:23] <tomww> sollte das SATA im alten IDE-Emulationsmodus laufen, also der Treiber "cmd" uim Devicepfad auftauchen, dann haste vmtl. Pech gehabt [18:45:29] <tomww> sorry :-) [18:51:37] <codestr0m> seanmcg: the wording on this double negative is strange.. "-F do _not_ do a non-DEBUG build" that means -F will do a DEBUG build and you normally don't.. I assume that a debug build will be slower than normal.. after I get this thing to compile I want to dig into the ACPI bits [18:52:01] <bug_L> codestr0m : hi [18:52:08] <codestr0m> bug_L: Hi [18:52:46] <bug_L> why i dont see my win32 partition in installation b98 [18:53:23] <codestr0m> bug_L: I don't know anything about win32 [18:53:43] <codestr0m> I assume if it's ntfs partition solaris doesn't support this, but I could be totally wrong [18:54:25] <bug_L> codestr0m : sorry, but I only want install dual boot with xp [18:55:10] <codestr0m> bug_L: I wish I could help you.. I'm just not windows support.. googling may help you.. I'm pretty sure there are howto's on this and if not report back here [18:55:44] <bug_L> =\,ok...thks much [18:55:55] <codestr0m> yeah. sorry I can't be more help [18:57:09] <oxygene> bug_L: you may look for fuse for solaris, and ntfs3g on top of that. no idea what's the status, but I think there are people working on it [19:00:36] <bug_L> oxygene: cool, but i dont to share the partition between two systems, I want install the opensolaris after windows xp, in dual boot, but the partition list no show the partition win, and I cannot to miss it =) [19:00:36] *** shaftyyy has joined #opensolaris [19:01:42] <codestr0m> bug_L: your partition is taking up the whole drive or you already divided things? [19:02:42] <codestr0m> seanmcg: thanks for the CW_NO_SHADOW= ; export CW_NO_SHADOW tip.. I'm also a bit unsure if I want make clobber or not.. I assume clobber just overwrites everything before, but would it otherwise only overwrite the changed bits? [19:02:44] <bug_L> I already divided things, =), [19:02:57] *** shaftyyy has quit IRC [19:03:23] <codestr0m> bug_L: well.. if it sees the disk and doesn't recognize your partitions tables.. try google.. these things can be *that* uncommon [19:03:42] <oxygene> hmm.. solaris doesn't know extended partitions [19:03:53] *** TheK_ has joined #opensolaris [19:04:09] * TheK_ the troll enters with his tail behind his legs looking mighty humble. [19:04:25] <bug_L> oxygene: hhmm....maybe it [19:06:08] <TheK_> /usr/bin/gcc + /usr/bin/ld... I keep getting troubles using gdb because of missing symbol tables. i though I corrected this issue by making sure that /opt/gnu/bin/ld was renamed gld. [19:06:37] <TheK_> What is the issue when mixing gnu ld and solaris ld? Can it be determined by looking at run paths and stuff? [19:08:55] <oxygene> TheK_: they have different sets of options. gcc is specifically built for one or the other [19:09:34] <oxygene> TheK_: you could set a PATH like /usr/ccs/bin:/usr/bin:..., to avoid a gnu ld interfering [19:11:04] <TheK_> oxygen: ok.. but if for some reason I have a rough library in my /usr/lib path which is killing gdb, what can I do to locate it and fix it? [19:11:36] * TheK_ is wildly assuming it _is_ a rough library killing his gdb. [19:16:03] *** bug_L has quit IRC [19:16:39] *** mikl has quit IRC [19:17:09] *** shaftyy has quit IRC [19:17:21] <Pietro_S> TheK_: or you can use dbx ;-) [19:18:38] <TheK_> Pietro_S: yes, maybe I should try that. [19:23:04] *** cypromis has joined #opensolaris [19:23:09] *** fexy has joined #opensolaris [19:23:11] *** cypromis has quit IRC [19:29:43] *** guru][ has joined #opensolaris [19:33:52] *** guru][ has quit IRC [19:39:15] <benley> if I'm installing sxce with zfsroot, is there any compelling reason why I'd want to put /var on a separate dataset? [19:39:24] <benley> Or are the reasons for doing that exactly the same as if I wasn't using zfs? [19:41:05] <saablover> I am experimenting with slimming down opensolaris, I created a new boot env but I deleted my beadm package :( [19:41:18] <saablover> can't go back to my orig situation :) [19:41:23] <benley> heh, oops [19:42:28] <saablover> can someone do a pkg contents and lookup the beadm file ? [19:42:31] <saablover> hehe [19:42:46] *** fr4g has joined #opensolaris [19:46:12] *** mega has joined #opensolaris [19:47:12] <codestr0m> saablover: you mean basename file sbin/beadm pkg:/SUNWgui-install at 0 dot 5.11-0.96 [19:47:12] *** cky_ has joined #opensolaris [19:47:17] <codestr0m> or you need the contents of that pkg? [19:47:49] <saablover> no that's the one :) [19:48:12] <codestr0m> pkg search -r beadm I think is what you were looking for [19:48:16] <saablover> I will see what does selecting the other be in grub does [19:48:19] <Yorlik> Did I get it right lx branded zones are bound to 32 bit even on 64 bit sxce [19:48:28] <saablover> where be = boot environment [19:48:36] <saablover> Yorlik: yes [19:48:42] <Yorlik> thanks [19:48:53] <saablover> Yorlik: I don't know about the 2.6 things they are putting in place [19:49:29] <saablover> aha selecting the other boot environment in grub does put me back to the original env [19:49:31] <saablover> cool [19:49:54] *** mega has quit IRC [19:50:17] *** mega has joined #opensolaris [19:50:33] *** cky has quit IRC [19:50:53] <Yorlik> Is it possible to give someone access to a zone console when the zone is booting from a CD image to allow him install (E.G. gentoo) [19:51:06] <Yorlik> I want to give a friend an lx branded zone [19:51:18] <Yorlik> and he likes gentoo [19:51:52] <saablover> you don't want to give him access to your global zone ? [19:52:11] <Yorlik> Yes. I would do if necessary, but I'd like to circumvent it [19:52:40] <Yorlik> Of course I could give him a sudo on zlogin [19:52:59] <Yorlik> But it interests me for theoretical reasonc [19:54:12] <saablover> you can play with rbac [19:54:22] <saablover> give him a limited root account [19:54:36] <TomJ> I'm not sure yuo can delegate control of a specific zone, just all zlogin rights? [19:54:41] <CosmicDJ> the root account inside the zone should be limited :) [19:54:57] <saablover> don't know the profile for zlogin [19:55:04] <saablover> I should look at it [19:55:06] <Yorlik> zone admin ... [19:55:18] *** guru][ has joined #opensolaris [19:55:22] <CosmicDJ> why zlogin? you can only do that from the global zone... give him ssh access to the zone and you're fine [19:55:36] <saablover> he wants to see the machine boot [19:55:37] <Yorlik> I want him to install the zone from a cd image [19:55:50] <Yorlik> Like a serial console [19:55:53] <Yorlik> not more [19:55:54] <CosmicDJ> ah, that's possible? [19:56:00] <Yorlik> Thats what I asked ... [19:56:03] <saablover> zlogin )C [19:56:05] <saablover> oops [19:56:09] <saablover> zlogin -C [19:56:16] <saablover> but never played with brands [19:56:21] <Yorlik> Yes - but he can do it to every zone I have then [19:56:39] <CosmicDJ> well I heard that it's possible to zfs clone + transfer zones... [19:56:54] <Yorlik> I don't have lx zones to clone ... [19:57:05] <saablover> brandz that is [19:58:07] <Yorlik> Maybe I just give him a standard brandz and he can do his gentoo install from there with chroting and stuff [19:58:25] <saablover> nah spend some time with rbac :) [19:58:29] <CosmicDJ> IIRC gentoo isn't supported; you're on your own trying that [19:58:31] <saablover> it's a VERY funny thing :) [19:59:06] <Yorlik> I remeber time when I locked myself our from stuff when playing with rbac [20:00:26] <saablover> always play on a virtualbox with these things :) [20:00:34] <saablover> you can revert to snapshots :) [20:00:38] *** evocallaghan has joined #opensolaris [20:01:08] *** mega_ has joined #opensolaris [20:02:00] <saablover> hmmm a large bank is failing here in Belgium [20:02:03] <Yorlik> esxi is ma friend [20:02:14] <evocallaghan> banks do that .. [20:02:33] <saablover> nice to know that the gov injects millions [20:02:34] <evocallaghan> Do you mean there failing at covering it up ? [20:02:39] <saablover> yeah :) [20:02:47] <Yorlik> BTW: I tried to use the mysql driver for apches auth_dbd module and it said I don't have the driver ... isn't it in SXCE ??? [20:02:56] <Yorlik> apache2.2 that is [20:02:58] <evocallaghan> Sound like one of the better banks to me :p [20:03:16] <saablover> I hate this principle, they can waste enormous amounts of money and no problem, the gov helps :) [20:03:43] <evocallaghan> Yea well do something about it [20:03:44] *** mega has quit IRC [20:05:11] <codestr0m> e^ipi: not sure if you care about errors like "cc: No input file specified, no output generated " on your gate.. I've been messing with my nighlty options and may have incompatible options though http://rafb.net/p/R8ERFa12.html [20:05:24] *** mega has joined #opensolaris [20:07:38] *** codestr0m has left #opensolaris [20:08:31] *** codestr0m has joined #opensolaris [20:12:24] <evocallaghan> Gezz, quantum mechanics has a few hard things.. [20:12:48] <e^ipi> codestr0m: i have no idea what you've done to your gate but everything you've said so far indicates that you're doing something wildly different from the rest of us. If you want to run off and do your own thing that's fine, but I can't help you or offer advice unless you do roughly the same thing as us [20:13:30] *** ChanServ sets mode: +o comay [20:13:56] *** mega_ has quit IRC [20:16:53] <seanmcg> codestr0m, was away there. actually a debug build alone (i.e. not building the non-debug bits - _only_ the debug bits) may be faster since compiling with optimisation is generally slower.. [20:17:45] <e^ipi> CW_NO_SHADOW will also speed things up immensely [20:17:45] *** mega has quit IRC [20:18:03] *** mega has joined #opensolaris [20:18:11] *** art-vandelay has joined #opensolaris [20:18:31] <seanmcg> those nightly flags of mine just build the non-debug bits, and just a bfu - no packages, no debug bits etc. My builds are primarily for performance testing so don't need the debug bits. [20:18:55] <seanmcg> yarihm, e^ipi, codestr0m spotted that already :) [20:18:56] *** mega has quit IRC [20:19:25] *** mega has joined #opensolaris [20:20:03] <codestr0m> e^ipi: you are less than helpful at times and I suspect it's your lack of patience which drives you to be as annoying as those who annoy you [20:20:39] <seanmcg> s/yarihm/ya/, damn tab trigger key [20:23:19] <evocallaghan> Any one here a skilled mathmation ? [20:23:19] *** mega has quit IRC [20:23:26] <evocallaghan> Or know a good IRC chan ? [20:23:34] <_mary_kate_> efnet #math [20:23:40] <evocallaghan> thanks [20:23:42] <e^ipi> codestr0m: i apologize, but ultimately you really do need to do what we're doing ; i suspect the errors your getting is because you've got CLOSED_IS_PRESENT set somewhere, or you've tricked nightly in to doing the same thing [20:23:43] *** mega has joined #opensolaris [20:23:53] <e^ipi> and since you're not @ Sun, closed /isn't/ present [20:24:01] *** mega has quit IRC [20:24:15] <e^ipi> my suggestion is to wipe out your nightly script, and start from scratch [20:24:19] *** mega has joined #opensolaris [20:24:35] <e^ipi> I use an unaltered tools/env/opensolaris.sh to build [20:24:47] <e^ipi> (aside from pointing it at the correct path, and changing the mailto ) [20:24:52] <e^ipi> please use that. [20:25:25] <e^ipi> if you've altered /opt/onbld/bin/* directly, remove SUNWonbld and install a new one. [20:25:35] * bda adjusts e^ipi's tie. [20:25:36] *** mega has quit IRC [20:26:04] <codestr0m> e^ipi: you're half right.. try to be a little more patient with those in the community.. seriously.. even as helpful as you can be. this is unneeded. My changes are merely for the closed bin check [20:26:21] <e^ipi> codestr0m: and i'm telling you that the closed bin check needs to remain. [20:26:23] <codestr0m> I'm not behind the firewall and even if I was it wouldn't change my end goal [20:26:42] <sstallion> afternoon all [20:26:51] <e^ipi> codestr0m: the closed bins are still needed for building the gate, with the exception of wiping out libc_i18n.a [20:26:53] <codestr0m> k. well. I'll either manage on my own, you'll help or others will. I won't address you on this directly anymore as your attitude is clear [20:27:19] <e^ipi> codestr0m: my attitude is i'm trying to tell you what you need to do to get on the same page as us [20:27:26] <e^ipi> and you keep ignoring me [20:27:46] <codestr0m> e^ipi: my page may be different than yours.. I'm not entirely focused on libc [20:28:12] <seanmcg> codestr0m, what is your goal here ? (curious) [20:28:29] <e^ipi> codestr0m: then your best bet is to once again, leave nightly there, and use the actual individual makefiles to accomplish it [20:28:32] <codestr0m> seanmcg: compile gate w/o closed bins.. and yes to help with libc when that comes up [20:28:39] <e^ipi> that is, change usr/src/cmd/Makefile to remove the checks [20:28:42] <e^ipi> not nightly [20:28:52] <codestr0m> e^ipi: gotcha. ok. that I can do [20:29:13] <e^ipi> look at my changeset for usr/src/lib/libc/Makefile and friends [20:29:29] <sstallion> codestr0m: you cannot use Makefiles bundled with ON if you are not going to be using the closed bits [20:29:30] <e^ipi> the Makefile checks for closed bins, you need to remove those sorts of things [20:29:32] <seanmcg> not having delved tooo much into not needing libc during compile, can you ?-) [20:29:40] <codestr0m> seanmcg: once I also have a better understanding of this process I'm going to also create individual packages for all cmds and such in the gate [20:30:07] <seanmcg> which libc will they be linked against if not the libc in closed ? [20:30:10] <sstallion> codestr0m: its already there. look in usr/src/pkgdefs [20:30:22] <e^ipi> seanmcg: the ones we're writing [20:30:40] <seanmcg> emancipation proj ? [20:30:41] <codestr0m> sstallion: I'll convert those away from pkgdef format [20:30:45] <e^ipi> seanmcg: yeah [20:30:52] <sstallion> codestr0m: why ? [20:30:56] * seanmcg gets the bulb on now. [20:31:00] <seanmcg> thanks [20:31:12] *** evocallaghan has left #opensolaris [20:31:18] <codestr0m> seanmcg: yeah. sorry for not being explicit about not using main onnv-gate [20:31:54] <e^ipi> codestr0m: but you're going to need to leave the closed check in [20:32:03] <e^ipi> otherwise the makefiles will look for usr/closed, which you don't have [20:32:22] <codestr0m> e^ipi: I'll take a look at modifying the Makefiles as the alternative you suggested [20:32:45] <sstallion> codestr0m: out of curiosity what are you trying to accomplish ? [20:32:55] <sstallion> typically you should be working *with* ON, not against it... [20:34:01] <codestr0m> sstallion: 1) I see O/N and most of the building of OpenSolaris to be 20 years of nasty 2) I'm going to be building IPS packages directly.. no SVR4 middle.. I have some personal hacking I want to do and lastly. I want to remove any closed source bins [20:34:22] <codestr0m> which will also include and work has been started to get things like OpenJDK6 compiling and nvidia drivers [20:34:32] <sstallion> codestr0m: well, something to consider is that much of this effort is already under way [20:34:45] <codestr0m> sstallion: is it public or not? [20:34:46] <sstallion> it would probably be more realistic to work with others rather than trying to do everything yourself [20:34:51] <e^ipi> i'm certain that the IPS guys are already trying to turn ON in to ips packages [20:34:53] <sstallion> of course its public [20:35:13] <e^ipi> they're pretty good about working in the open [20:35:19] <sstallion> honestly, I would focus on the closed bits first and let the packaging gurus deal with IPS [20:35:24] <codestr0m> e^ipi: IPS is only handling the front-end with no plans to create a coherent backend build system.. I asked them [20:35:46] <sstallion> but before insulting e^ipi, I would consider the fact that he leads up the emancipation project... [20:35:59] <codestr0m> sstallion: insulting.. show me where I insulted him [20:36:26] <e^ipi> codestr0m: here's the deal... both are a lot of work. I suggest you focus on one, and then the other, but not both [20:36:39] <Yorlik> hmm ... anyone here familiar with apr_dbd_mysql.so - the apache2.2 mysql driver ? It seems not to be insxce and I can't find a solaris version anywhere ... [20:36:44] <e^ipi> focussing on both, as a single contributor, will likely take years [20:37:08] <codestr0m> e^ipi: to package it I must dive in and understand.. that's the step I'm at now.. from there I'll take the most gain route [20:37:13] <e^ipi> meanwhile, sun's already going to be changing ON aronud to build IPS packages, since the rumor goes that nevada's going IPS soon enough anyways [20:37:50] <codestr0m> k.. I see your point.. I just see how this is currently [20:39:37] <e^ipi> emancipated libc alone is already a few years worth of work for a single contributor [20:40:19] <sstallion> e^ipi: wasnt libc opened up ? [20:40:54] <e^ipi> sstallion: not the i18n bits of it [20:41:07] <codestr0m> If this process is easily reproducible and documented. (which I plan to help with) then others who are doing asinine things like porting glibc don't have an excuse not to help [20:41:26] <sstallion> e^ipi: *nod* [20:41:33] <e^ipi> and i've been working on that [20:42:06] <codestr0m> as a community member people ask.. and the information is out there, but it's sporadic and terse [20:42:34] <e^ipi> <@e^ipi> and i've been working on that [20:42:46] *** nitrile has quit IRC [20:43:09] <codestr0m> @<@e^ipi> and i've been working on that@ and for that I thank you [20:45:05] *** Odin- has joined #opensolaris [20:45:29] *** mega has joined #opensolaris [20:46:31] *** yarihm has quit IRC [20:46:37] *** art-vandelay has quit IRC [20:46:59] <comay> codestr0m, e^ipi is correct; there are folks in ON already looking at populating a repository directly out of a build [20:47:56] <codestr0m> comay: this solves the front-end, but not the backend of it being an easier to reproduce build process.. as in.. find a bug. .download only the source I want.. build.. test.. ect.. [20:48:49] <comay> ok, didn't realize that's what you were looking for [20:48:50] <codestr0m> or dl the source I want and magically pull all the deps I need and not just the full O/N tree.. anyway.. [20:49:05] <codestr0m> comay: sure np.. I'm possibly not clear on my goals [20:49:10] <seanmcg> you want to modularise ON ? [20:49:23] <comay> sounds like it [20:49:47] <e^ipi> that's a whole different kettle of fish [20:50:04] <comay> there are parts of ON that should be separated out (many of the legacy utilities for one) but sadly, i don't think those contribute much build time [20:50:08] *** saablover has quit IRC [20:50:28] <e^ipi> no, perl does though [20:50:59] <e^ipi> ON has become unfortunately quite inter-dependent and i'm not sure that you really /can/ modularize it easily [20:51:01] *** MeP3aBeu has joined #opensolaris [20:51:05] <codestr0m> comay: I'm not so much worried about build time long run.. just I want to get to a point where I can reproduce a very clear error from the ground up.. [20:51:14] <seanmcg> speaking of which, anyone fancy bumping up the perl version in ON ? (ducks) [20:52:15] <e^ipi> codestr0m: also take a look at freebsd's sources, they've got a bunch of stuff in there as well (not nearly as convoluted as ON) [20:52:26] <codestr0m> I'd also like to be able to turn on a bit more verbosity in the logs [20:52:29] <e^ipi> you can't simply grab a portion of it either, and that's fine... it's a whole system [20:52:37] *** jgracin has joined #opensolaris [20:52:43] *** sailorvrz_ has quit IRC [20:52:49] <e^ipi> similarly with ON [20:52:59] <codestr0m> e^ipi: I'm very familiar with ports. and the obsd sqlports project along with about 10 other build systems and their benefits.. [20:53:10] <e^ipi> i'm not talking about ports [20:53:26] <e^ipi> kernel && world [20:53:51] <comay> seanmcg, we need to find someone interested in owning perl :-) [20:54:09] <e^ipi> you can't really mix & match them on freebsd, and you shouldn't be able to mix & match them on solaris either [20:54:12] <seanmcg> comay, so I've heard :) [20:54:35] <e^ipi> comay: that sounds like a... fun... job [20:54:40] <codestr0m> e^ipi: you're trying to define what an end-user should do. which is different than what a developer or package maintainer would want to do [20:55:00] <e^ipi> codestr0m: not sure I follow [20:55:01] <comay> i don't know the differences but i would expect updating it to the latest 5.8.x branch to be fairly easy [20:55:28] <e^ipi> codestr0m: developers can't mix & match kernel & world either [20:55:42] <benley> argh, why doesn't the installer ask me for a hostname? [20:56:06] <e^ipi> benley: because you picked DHCP [20:56:08] <codestr0m> e^ipi: well.. in a packages lifecycle.. I see things as broken between a backend and a front-end.. ports, portage and friends.. all lump the whole thing together.. IPS handles the front-end and pkgtool/pkgbuild handles the backend.. more coming.. [20:56:19] <benley> e^ipi: I guess that's what I figured [20:56:25] <benley> e^ipi: if only dhcp had worked! :-P [20:56:28] <e^ipi> codestr0m: what's that got to do with ON ? [20:56:33] <seanmcg> comay, its all the bits that are dependant on perl, all that testing perhaps. [20:57:06] <e^ipi> codestr0m: ON is the core. it is different than the peripheral software [20:57:08] <comay> yes, that would likely be the most difficult part [20:57:32] <seanmcg> though python could end up in a similar way [20:57:46] <e^ipi> comay: even freebsd treats it different... they have kernel, world, and ports... kernel and world can't be used except together... ports, doesn't matter [20:58:06] <codestr0m> e^ipi: on is the core, but it's made into different parts which still start with a single dependency and all work together. for example.. a version bump in perl.. why shouldn't that be easier [20:58:11] <comay> sorry e^ipi, i was talking about perl & testing :-) [20:58:33] <e^ipi> comay: sorry, i meant codestr0m [20:58:51] <codestr0m> putting everything in a single tree makes sense in a way because you know it will all work together, but from maintenance is can be a bugger [20:59:14] <e^ipi> codestr0m: it's not a tree, it's a graph [20:59:18] <e^ipi> codestr0m: and that's the problem. [20:59:19] <sstallion> codestr0m: I think you are really oversimplifying [20:59:37] <e^ipi> things are inseperable from each other in quite a number of places [20:59:47] <sstallion> net is a fantastic example [20:59:56] <benley> e^ipi: actually, dhcp did work. shouldn't it have set its hostname based on a reverse lookup of its assigend ip? [21:00:12] <e^ipi> benley: dhcp is voodoo to me [21:00:15] <benley> heh [21:00:27] <benley> perhaps I should turn it off, it's not that important at home [21:00:28] <seanmcg> codestr0m, bumping perl means also making sure (really sure0 that anything in ON that depends on perl still works in all their myraid of ways [21:00:35] <sstallion> benley: it depends. do you have something defined in /etc/nodename ? [21:00:41] <benley> sstallion: nope [21:00:56] <sstallion> also are you requesting a hostname in /etc/default/dhcpagent? [21:00:58] <benley> sstallion: if I set that and leave /etc/dhcp.<ifname> in place, will things just generally work? [21:01:13] <benley> sstallion: this is a fresh install, I haven't set anything yet [21:01:14] <sstallion> it should be an empty file, but yes [21:01:27] <sstallion> assuming your dhcp server sends a hostname [21:01:30] *** Rarok has joined #opensolaris [21:01:34] <sstallion> otherwise you will get unknown ;) [21:01:37] *** Rarok has quit IRC [21:01:39] *** bourgois has quit IRC [21:01:40] <codestr0m> seanmcg: sure I understand that.. and to test that you have to 1) commit the new version locally 2) compile world 3) test apps.. however. to really test apps you need to share your changes.. and how is doing that in the current process? [21:01:45] <benley> ah, I might not be sending a hostname from my dhcp server. [21:01:55] <sstallion> if you arent, that would explain it [21:02:01] <sstallion> I generally prefer to set /etc/nodename anyway [21:02:04] *** Rarok has joined #opensolaris [21:02:21] * benley sets it [21:02:31] <e^ipi> codestr0m: it's fine [21:02:40] <e^ipi> codestr0m: you publish your hg tree, or the changeset from it [21:02:58] <e^ipi> codestr0m: /opt/onbld/bin/webrev does it in a really neat way [21:02:58] *** mega has quit IRC [21:03:10] *** Rarok has left #opensolaris [21:03:14] <codestr0m> e^ipi: and then the person who gets your hg tree has to build it? [21:03:17] *** mega has joined #opensolaris [21:03:20] <e^ipi> codestr0m: yes [21:03:22] <seanmcg> codestr0m, so.. theres two versions of perl in sxce, two versions of java (there was three at one point i think),... [21:03:28] <e^ipi> codestr0m: nightly -i doesn't take that long [21:04:28] <codestr0m> I'll take notes on this.. and when I have more time try to document my thoughts. a person doing QA and a person doing package maintenance shouldn't have to compile/build a nightly imho. it's a different role [21:04:44] <e^ipi> why would they have to do it? [21:04:57] <e^ipi> building a package is entirely unrelated to ON [21:05:06] <e^ipi> similarly with QA [21:05:15] <seanmcg> folks would build their ON version (a bfu say) and submit that for testing. [21:05:58] <seanmcg> one can do that now via test.opensolaris.org [21:06:03] <codestr0m> seanmcg: ok. that can make sense.. publishing a tree for some QA guy to verify the version bump of perl app foo works isn't very clean as I was understanding it [21:07:39] <seanmcg> for the most part the QA guy (teams really) don't (and shouldn't) care too much of the contents, they monitor for failures/regressions. Knowledge of every bug fix being tried out isn't viable [21:08:20] <codestr0m> QA teams just have a checklist/procedure they normally run through [21:08:29] <codestr0m> at least that's my experience [21:08:35] <seanmcg> and even then, the QA guy doesn't see this, its automatic. See test.opensolaris.org for examples. [21:08:46] <codestr0m> I was just looking there [21:09:35] <codestr0m> I think there was a proposal to expand test.o.o out to allow vpn in so people could test desktop apps as well.. I forget now [21:09:52] <seanmcg> in sun anyways the larger QA teams are about fully automatic. [21:10:05] <codestr0m> yes of course.. manual testing is crazy [21:10:07] <seanmcg> Yes, vpn in, Jim G ( I think) is working on that [21:10:15] *** derchris^ has joined #opensolaris [21:10:21] <seanmcg> and compile farms [21:10:23] <e^ipi> also, if you break ON and the auto tests don't catch it... Sun's workforce feels it [21:11:00] <sstallion> seanmcg: you know, if test.os.o supported NICDRV, it would make my life much easier ;) [21:11:49] *** derchris^ has quit IRC [21:11:59] *** derchris^ has joined #opensolaris [21:12:07] <codestr0m> oh. I know you guys will love this question.. is there a way to turn on more verbosity for nightly.log? [21:12:08] <codestr0m> install -s -m 644 -f /usr/local/src/i18n-gate/proto/root_i386/usr/include/ast "tmpastinclude/wordexp.h" fuzzy --> 1 job fuzzy --> 2 jobs [21:12:18] *** derchris has quit IRC [21:12:33] <sstallion> codestr0m: heh, pass -dd to dmake [21:12:40] <sstallion> that will give you all the noise you could ever want [21:12:55] *** _luc^ has quit IRC [21:13:04] <seanmcg> how verbose do you need it ? The log is huge as it is :) [21:13:09] *** art-vandelay has joined #opensolaris [21:13:17] *** luc^ has joined #opensolaris [21:13:46] <codestr0m> well. it's been doing "something" for a while.. I dunno. I suppose I'm interested in what it's compiling at least [21:14:06] *** GHReyes has joined #opensolaris [21:14:22] <e^ipi> codestr0m: ps -ef [21:14:23] <GHReyes> hi all [21:14:37] <e^ipi> or ptree for a fancy tree of it [21:15:29] <seanmcg> sstallion, I'm not sure. could ask on the test mailing list ? Jim is quite responsive [21:15:53] <sstallion> seanmcg: it wouldnt be feasible... it would require root access on 2 machines with specific hardware [21:15:54] <codestr0m> http://rafb.net/p/Z1enuo82.html [21:15:56] <seanmcg> codestr0m, tail -f the log ? [21:16:17] <codestr0m> seanmcg: yeah. I'm tailing the log and it's been too quiet. that's why I asked [21:16:49] <seanmcg> oholiks, then ptree as e^ipi suggests and execsnoop from the DTT [21:17:17] <seanmcg> s/oholiks/oh/, the trigger key strikes again, bad trigger key. [21:18:44] *** XYOX has joined #opensolaris [21:19:54] *** palowoda has quit IRC [21:21:15] <codestr0m> seanmcg: *THANKS* I've yet to read more on DTT, but just found execsnoop in DTraceScripts/TopMostUseful/ and that's very handy [21:21:48] <seanmcg> codestr0m, if its too quiet for too long then perhaps something is wrong, running out of mem, disk i/o, network (of building over nfs), etc... [21:24:20] <codestr0m> I should really be doing this on the 2nd laptop, but haven't setup the dev env over there.. execsnoop is showing it's working http://rafb.net/p/7kzubD30.html (small snip) [21:27:41] <seanmcg> my group has a script that bundles whats needed into a file backed zpool, so we're able to move that to whatever machine is available, it then builds a bunch of bfus.. [21:28:45] <codestr0m> that's quite clever.. [21:28:45] *** mega has quit IRC [21:28:56] *** swa has joined #opensolaris [21:29:04] *** mega has joined #opensolaris [21:29:10] <codestr0m> I'm still not taking advantage of all the benefits of DTT/zfs/smf.. zones. etc [21:30:38] <benley> hooray, I got my zfsroot system up and running [21:30:46] <seanmcg> zfs makes us work more... no longer do we stand around watching the fsck take 2 days on the server after a power outage :) [21:30:51] <benley> and it imported my existing raidz pool with minimal fighting [21:31:06] <benley> seanmcg: to be fair there are other filesystems that don't make you watch fsck for two days [21:31:33] <seanmcg> sure, but when youve only got ufs and (shudder) vxfs.. [21:31:44] *** hircus has joined #opensolaris [21:33:19] *** XYOX has left #opensolaris [21:33:55] <benley> even ufs has logging [21:34:09] <benley> though I wouldn't necessarily trust its data integrity like I would zfs :) [21:34:10] *** mega has quit IRC [21:34:39] *** mega has joined #opensolaris [21:34:59] <victori_> seanmcg: nice troll, journaled fs? [21:35:20] <seanmcg> benley, I agree ufs logging does help for the most part. I still remeber waiting two days a few times though :) [21:35:35] <seanmcg> victori_, journaled, logging, ya.. [21:48:50] *** lucianno has joined #opensolaris [21:49:23] <lucianno> gracias senhores, I am onboard!!! [21:50:17] *** GHReyes has quit IRC [21:52:19] *** sailorvrz has joined #opensolaris [21:55:59] *** art-vandelay has quit IRC [21:57:24] *** art-vandelay has joined #opensolaris [21:58:13] *** lucianno has left #opensolaris [21:58:23] *** nitrile has joined #opensolaris [21:59:14] <Yorlik> Has anone an idea how to authenticate users in apacha2.2 against a mysql database ? it seems the mysql driver for mod_authn_dbd is not in the distribution. [21:59:26] <Yorlik> s/apacha/apache [22:01:06] *** mega has quit IRC [22:01:27] *** mega has joined #opensolaris [22:04:19] <tomww> s/mod_authn_dbd/ldap/ as a more generic approach if you want [22:05:23] <Yorlik> I'm a complete ldap noob ... and I have software like forums , CMS and such which use mysql ...I want to have only one point of authentication. [22:05:37] *** lucianno has joined #opensolaris [22:05:55] *** mega has quit IRC [22:06:03] <Yorlik> Its phpBB3, Mediwiki and Drupal. [22:06:09] <Yorlik> ans svn [22:06:14] <Yorlik> and svn [22:06:24] *** mega has joined #opensolaris [22:07:02] <Yorlik> I only want the unix accounts stay in files. [22:07:17] *** guru][ has quit IRC [22:08:21] *** art-vandelay has quit IRC [22:14:15] *** asarch has joined #opensolaris [22:14:21] *** art-vandelay has joined #opensolaris [22:14:22] *** cky_ is now known as cky [22:14:22] *** mega has quit IRC [22:14:59] *** mega has joined #opensolaris [22:16:02] *** relling1 has quit IRC [22:16:35] *** mega has quit IRC [22:16:40] *** mikl has joined #opensolaris [22:16:57] *** mega has joined #opensolaris [22:18:04] *** asarch has quit IRC [22:19:31] *** seagull_7 has joined #opensolaris [22:20:57] <jbk> afternoon [22:20:59] <sstallion> anybody around which is using parallels at the moment ? [22:21:03] <sstallion> heya jbk [22:21:14] *** mega has quit IRC [22:21:36] *** mega has joined #opensolaris [22:23:09] <bda> sstallion: Not just at the moment, but I do, sometimes. [22:23:16] <bda> (mainly for Jumpstarting) [22:23:32] <sstallion> bda: gotcha. probably best if i do an install here... what I was going to ask could be destructive :) [22:23:53] *** Fish has quit IRC [22:24:00] <bda> ruh roh? :P [22:25:27] *** sah-work has joined #opensolaris [22:30:36] *** mega has quit IRC [22:30:48] *** jgracin has quit IRC [22:30:53] <sstallion> bda: well, I wanted to check to be sure that some of the chip init on the re driver works correctly in parallels [22:31:04] <bda> aha [22:31:10] <sstallion> but that involves uninstalling the existing parallels driver :) [22:31:37] <bda> sstallion: If you want help with that, I'm game. [22:31:41] <sstallion> sure [22:31:46] <sstallion> its mostly just curiosity :) [22:31:50] <bda> Anything would be better than the Parallel's driver. [22:31:51] <bda> nod [22:31:55] <sstallion> mind you [22:31:58] <sstallion> it doesnt function :) [22:32:02] <bda> ha [22:32:06] <bda> Well, not *anything*, then. ;) [22:32:08] <sstallion> all you can do is plumb the interface... do anything else and you'll likely panic [22:32:09] <sstallion> haha [22:32:30] <bda> ha. [22:32:40] <sstallion> it should be done fairly soon [22:32:54] <bda> Cool beans. I'm available for testing. :) [22:33:07] <bda> Being able to Jumpstart multiple boxes would be pretty nice. [22:34:44] *** GHReyes has joined #opensolaris [22:35:10] <sstallion> what issues are you having with the parallels driver? [22:35:26] <bda> It falls over under load. [22:35:39] <sstallion> ouch [22:35:50] <bda> e.g., NFS traffic + watching prstat via ssh. :) [22:36:03] <sstallion> the chipset is pretty slow, but there is no reason for it to fail under those kinds of conditions [22:36:23] <bda> Yeah. [22:37:29] <sstallion> jbk: do you have a list of device and vendor id's for cge yet ? [22:38:34] <jbk> no i figured probably need to get a barebones driver going first, then can start to worry about that [22:38:42] <sstallion> okay [22:38:47] <sstallion> i'll just stub out the pkgdef then [22:39:24] <sstallion> been playing language lawyer this morning... re-licensed the makefiles and drver conf's [22:39:54] <sstallion> was doing some digging through on, and drivers under the BSD don't use the BSD licenses in Makefiles or driver conf's, so I swapped ours to match [22:42:07] *** stukag has quit IRC [22:45:41] <GHReyes> hi guys [22:46:49] <GHReyes> i have a snv_79a working in a C300 series Compaq [22:48:05] <GHReyes> all fine (except audio) [22:48:32] <GHReyes> i solved wifi with my previous experience [22:48:49] <GHReyes> but i hace a new issue [22:49:08] <GHReyes> The size of the letters in the post-login windows are huge, once the desktop everything seems normal again. [22:49:25] *** jub has joined #opensolaris [22:50:02] *** mega has joined #opensolaris [22:50:26] <GHReyes> after pass login window and before desktop is functional... [22:51:29] *** tilt has joined #OpenSolaris [22:51:58] <GHReyes> while the panel is starting, the messages are so big [22:52:10] <tilt> i'm trying to compile the openmp test program with sun studio express and i'm getting an error i don't think i fully understand [22:52:51] <tilt> i can compile openmp programs in the ide, but when in a console i get undefined symbols [22:53:00] <tilt> think it could be a linker problem ? [22:53:24] *** jbasse has joined #opensolaris [22:54:16] <e^ipi> you need to actually pass the compiler the flags for it [22:54:54] <tilt> i am [22:55:02] <tilt> and don't forget -O3 [22:55:04] <tilt> ;) [22:55:29] <tilt> soon openmp won't set a default optimization level [22:55:36] <tilt> /nod [22:55:38] <e^ipi> cc != gcc [22:56:05] <jmcp> POSIX specifies compiler command line options? since when? [22:56:08] * jmcp surprised [22:56:16] <tilt> but regardless of xO3 or O3 i still get this link error [22:56:16] *** mega has quit IRC [22:56:27] <_mary_kate_> jmcp: for the 'c99' command [22:56:28] <smtms> jmcp, it's an option of the c99 utility [22:56:35] <jmcp> h, right [22:56:47] <jmcp> I should go and read the standard [22:56:59] <_mary_kate_> c99 [-c][-D name[=value]]...[-E][-g][-I directory] ... [-L directory] ... [-o outfile][-Ooptlevel][-s][-U name]... operand ... [22:57:27] *** JoergB has quit IRC [22:57:51] <e^ipi> tilt: which symbols are undefined? [22:57:55] * jmcp wonders whether the c99 part of POSIX was written by GNU [22:58:18] <_mary_kate_> jmcp: well, -O is a standard option in every 'cc' [22:58:45] <jmcp> hmmmph [22:58:51] <jbk> couldn't that require then c99 --define name=value --debug --include directory --link directory --outfile .. --optimize=... [22:58:58] <jbk> if gnu did it? [22:59:58] <smtms> I wonder if a future POSIX standard will have long options [23:00:28] <e^ipi> long options are an abomination. [23:00:42] <smtms> e^ipi, '--' means end of options [23:00:46] <e^ipi> exactly. [23:00:54] <GHReyes> Hi guys, the size of the letters in the post-login windows are huge, once the desktop everything seems normal again. After pass login window and before desktop become functional (while the panel is starting), the messages font are so big. Any idea? [23:00:55] <smtms> e^ipi, '--xyz' is not the same as '--' [23:01:11] <jmcp> GHReyes: you're still running snv_79? [23:01:18] <GHReyes> yes! [23:01:48] <GHReyes> i am a classic man [23:01:54] <GHReyes> =) [23:02:15] *** hspaans has joined #opensolaris [23:02:44] <GHReyes> jmcp similar issue sucedded in Ubuntu 8.06 but in login windows with user/pwd font [23:04:39] *** noyb has joined #opensolaris [23:08:58] <tilt> i don't think the c compiler is passing the right arguments to the linker [23:09:02] *** jbasse has quit IRC [23:09:36] *** xRaich[o]2x has left #opensolaris [23:09:49] <sstallion> jbk: looks like cge breaks using the ON makefiles [23:10:00] <sstallion> want me to push out the changes and let you sort them out ? [23:10:00] <jbk> in what way? [23:10:00] <jmcp> GHReyes: snv_79 is nearly 12 months old at this point, and you'd be *much* better off using a newer build [23:10:26] <sstallion> jbk: not exactly sure to be honest... I see no reason for the compile to fail; especially when the original built flawlessly [23:10:28] <jmcp> GHReyes: that said, have you looked at the Font aka Appearance settings? [23:10:36] <GHReyes> newer is not better, jeje [23:10:54] <GHReyes> today, i cant upgrade... [23:11:01] <GHReyes> maybe in december [23:11:18] <jmcp> actually, newer *is* better [23:11:22] <lucianno> hi guys [23:11:33] <lucianno> this channel is only developers? [23:11:40] <jmcp> GHReyes: I've used 79, and subsequent builds .... newer is most definitely better [23:11:44] <jmcp> lucianno: definitely not! [23:11:45] <jbk> well honestly, i was planning with the current gate to get things working before bringin in the rest of ON, as it just makes building a pain in the ass [23:11:57] <lucianno> ho thanks god [23:12:02] <jmcp> lucianno: we just don't like license trolls :) [23:12:03] <jbk> that's why I brought over the nemo include files [23:12:10] <GHReyes> jmcp, i pass from sxde 5/07 and this issue appears [23:12:21] <sstallion> jbk: *nod* [23:12:22] <jbk> so you could compile it outside of ON.. then I figured once it was working, to worry about getting it integrated into the ON tree [23:12:30] <jmcp> GHReyes: is there any reason stopping you from installing a newer release? [23:12:32] <GHReyes> jmcp, i upgrade OSS driver and crash my notebook [23:12:46] <sstallion> it looks like the suncc build fails and its going secondary and gcc of course, borks [23:12:47] <sstallion> http://rafb.net/p/xyNGAk13.html [23:12:58] <tilt> i have to use those old audio drivers, and my audio randomly messes up [23:13:06] <lucianno> jmcp: trolls?? what meaning ~trolls [23:13:30] <GHReyes> jmcp, i am working in critical proyect, i cant spent any time with upgrades... [23:13:46] <jmcp> lucianno: people who try to "convert" us, who say that GPLv2 is the only possible Open and Free license. [23:13:54] <jmcp> lucianno: those people can go away, frankly [23:14:02] <jmcp> GHReyes: have you heard of Live Upgrade? [23:14:14] <smtms> jmcp, it's GPLv3! :-P [23:14:17] <tilt> audiovia686 (via 686 audio driver) <- i sooo need a new computer :( [23:14:22] <tilt> accepting donations! [23:14:36] <GHReyes> any works fine... except this BIG BIG font at beginning of desltop... "The Panel" , etc [23:14:40] <jmcp> smtms: ah, I think the GPLv2 adherents would call you a heretic :) [23:14:43] <e^ipi> tilt: local school board's IT office... go dig in their trash [23:15:01] <e^ipi> tilt: when I was in jr. high that's what I did [23:15:07] <lucianno> ah yeah, [23:15:07] <tilt> e: last time i did that i got arrested :( [23:15:07] * hspaans gives tilt a penny [23:15:19] <tilt> no no, pay pal onlys! [23:15:27] <e^ipi> i never got hassled *shrug* trash is public property [23:15:29] <GHReyes> jmcp i heard about too many problems with live upgrade. [23:15:46] *** mega has joined #opensolaris [23:15:47] <jmcp> GHReyes: please pull down a newer build and try it again. [23:15:49] <GHReyes> i want no risk at this time [23:15:57] <lucianno> jmcp: thanks [23:15:58] <jmcp> GHReyes: *everything* has risk [23:16:13] <jmcp> GHReyes: sometimes you have to trust what other people are suggesting [23:16:13] * hspaans lives for the risk [23:16:33] <jmcp> I'm not willing to try to debug a problem you're seeing on a build which is around 12 months old. There have been *massive* changes in that time [23:16:38] <tilt> wanna see something cool, add a tumbdrive to your root pool and then just yank out the thumb drive [23:16:49] <lucianno> and for all too, I think I am the youngest user [23:17:07] <jmcp> there was somebody in here last week who claimed to be a 13yr old [23:17:17] <tilt> not only will your machine no longer boot, but everytime you plug the thumbdrive into a solaris machine it'll bitch about not being able to mount rpool... no matter how many times you format it [23:17:23] <smtms> lucianno, only 12? [23:17:38] <GHReyes> the guys of local community are working in issues related with live upgrade, no thanks until december [23:17:40] <jmcp> given the well-documented cases of FBI agents pretending to be 12 or 13 .... [23:18:11] <hspaans> tilt: create a zpool on usb with macosx and put in a solaris 10 box. it used to give some fireworks ;-) [23:18:20] <tilt> wooo [23:18:24] <GHReyes> my problem is a issue with GDM [23:18:24] <lucianno> smtms: hhm?? I dont understand, =/ [23:18:29] <tilt> wish my macbook wasn't in virginia [23:18:40] <jbk> sstallion: i suspect it's not liking something cge.h or cgereg.h that's parsing correctly but not doing something quite right [23:18:52] <lucianno> smtms: but, see you bye [23:19:02] <tilt> the wrird thing is after the thumbdrive bitches about not being able to mount rpool... it still automounts the fs and the thumbdrive works fine [23:19:06] <sstallion> jbk: *nod* i'll dig into it [23:19:07] *** lucianno has left #opensolaris [23:19:10] <sstallion> almost have the cge pkgdef done [23:19:37] <tilt> i can't figure out why sol things it's a member of "rpool" and always tries to mount it] [23:20:11] <jbk> i seem to recall some talk about rearranging some of the brussels stuff, so the DLD errors might have been victim of that [23:20:26] <jbk> and just needs another header file [23:21:16] <jbk> though strangely enough [23:21:21] <jbk> i just did a pull [23:21:28] *** [jbasse] has joined #opensolaris [23:21:32] <jbk> and it still compiles cleanly with my original makefile [23:21:54] * hspaans wonders if Sun will follow Gnome and start to deliver Empathy in the future [23:22:28] <jbk> are you using ss12? [23:22:54] <sstallion> jbk: i'm on an older pull using ss11 [23:23:17] <sstallion> let me check when it was made [23:23:20] <jbk> oh hmm.. [23:23:46] <sstallion> ahh looks like i pulled recently... last tag is snv_97 [23:24:20] <sstallion> same exact makefiles are used for re... no issues there [23:26:06] <jbk> what revision is cge.c? [23:26:31] <jbk> changeset: 21:5e696953417e [23:26:34] <jbk> that's what i have [23:28:00] *** [jbasse] has quit IRC [23:28:12] <sstallion> same here [23:28:21] <jbk> strange [23:28:23] <sstallion> I think you are right... cge was building just fine [23:28:40] <sstallion> when I dropped it into the new structure to link against ON it started throwing shits [23:28:59] <jbk> it might be a matter of dealing with header files [23:28:59] *** mega has quit IRC [23:29:03] <jbk> since it's a bit annoying [23:29:04] <sstallion> yeah [23:29:07] <jbk> in ON [23:29:18] *** mega has joined #opensolaris [23:29:23] <sstallion> well, I have a local commit... do you want to pull from that and play with it, or should I just push the changes up ? [23:30:08] <jbk> go ahead, but not sure when i'll be able to get to it.. [23:30:18] <jbk> well maybe next weekend if i bring a laptop [23:30:33] <sstallion> I can just throw it on a branch too [23:30:41] <TheK_> Anyone got a working jinput package for opensolaris? [23:31:04] <jbk> or if not, i'm going to netapp training in a few weeks in LA, so i'll have some free time that week after class [23:31:18] <sstallion> okay [23:31:21] <sstallion> i'll push for now [23:32:03] <sstallion> changeset 75 [23:32:41] <sstallion> meh i really should use my local repo more [23:35:39] <seagull_7> TheK_: have you seen this thread http://opensolaris.org/jive/thread.jspa?threadID=30920&tstart=30 [23:35:42] *** cypromis has joined #opensolaris [23:36:00] *** MeP3aBeu has quit IRC [23:36:32] <TheK_> seagull_7: Yes, I just read that actually. [23:36:34] <TheK_> thanks [23:38:55] *** TheK_ has quit IRC [23:41:00] *** mikl has quit IRC [23:41:10] *** asiudhxcz has joined #opensolaris [23:41:21] *** AxeZ has quit IRC [23:41:28] <asiudhxcz> can my nick be unban? [23:41:36] <asiudhxcz> :\ [23:44:18] *** jbasse has joined #opensolaris [23:54:54] *** stukag has joined #opensolaris [23:55:18] *** solaris2 has joined #opensolaris [23:55:37] <solaris2> Can i only install O.S from live cd .. or is there another installation media ? [23:56:16] <Stric> there is a net install preview too [23:56:58] *** art-vandelay has quit IRC [23:57:38] <solaris2> Mhmmm i installed O.S with liveCD before and the internet seemed to work when it wanted [23:57:53] <solaris2> i'd boot up and it worked, then boot up and it didn't =/ [23:57:57] *** mega has quit IRC [23:58:53] <hspaans> it didn't do what? [23:58:58] *** asidzx has joined #opensolaris [23:59:19] <solaris2> browser wouldn't load sites [23:59:34] *** art-vandelay has joined #opensolaris [23:59:51] *** asidzx is now known as lolmac