January 30, 2007 
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31
January 30, 2007 [00:00:09] <dclarke> anyways .. I have to call up Tom Goguen .. he and I have been chatting a lot lately [00:00:28] <dclarke> the idea of "Software as a Service" is one hell of a big leap and it needs to be made [00:00:33] <dclarke> but we need better package management tools [00:00:59] <dclarke> and we may need to write from scratch .. with backward compatibility in place ti the SVR4 specs and hell .. an ARC case would kill the process [00:01:07] <delewis> dclarke, not as long as you want to maintain SVID compliance. [00:01:24] <coffman> and on air traffic controll, i got realy good insight whats going on in germany at the dfs - "deutsche flug sicherung" (german airtraffic controll) - they are mostly linux now, no cde/motif is left there [00:01:26] <dclarke> delewis : I happen to believe in compliance [00:01:37] <delewis> coffman, Linux is there, yes. [00:01:42] <delewis> CDE/Motif is still there. [00:01:47] <delewis> XiG offers CDE for Linux. [00:01:53] <delewis> (hence why they're alive) [00:02:08] <Gman> dclarke, there are plenty of those types of people inside sun, i'm sure most of the sun engineers in this channel come across those people on a daily basis [00:02:09] <dclarke> Gman : I was not picking a fight .. I was really boldly loudly saying that you live in JDS/GNOME land and how would you see what I see ? really ? [00:02:15] <delewis> XiG's largest customer base is supposedly ATC. [00:02:40] <Gman> dclarke, i see what's going on inside sun on a daily basis, i have a pretty good idea of what conversations are happening [00:03:00] <dclarke> Gman : from what I hear .. its scary [00:03:17] <Gman> it's a relatively large company, not everyone buys into the idea of open source [00:03:22] <Gman> not totally surprising [00:03:39] <dclarke> like I said .. its HUGE transition in the last two years [00:04:00] <dclarke> there was this guy, a Sun exec type , flew up here over and over and we had lunch and talked and talked [00:04:16] <dclarke> he could not for the life of him figure out how open source was of any damn value to the comapny [00:04:25] <dclarke> and he was 2 levels below Schwartz [00:04:44] <dclarke> scary .. and I got a free lunch over and over and some Sun hats .. wearing one now actually [00:05:17] <Gman> people at the bottom of the chain get it, people right at the top of the chain get it, it just gets fuzzy in the middle [00:05:22] <dclarke> on a plus side .. he left the Solaris software sales side and is safely positioned elsewhere now .. [00:05:25] <Gman> and those people are the ones trying to make money [00:05:34] <delewis> *cough* SPARC graphics team *cough* [00:05:35] <dclarke> Gman : yep .. and it scary as hell [00:06:02] <dclarke> Gman : then when I show them these stats ( and they ask me for data ) that show 40,000 downloads peak .. per day [00:06:17] <dclarke> Gman : they think .. wow man .. put a price on that and retire tomorrow [00:06:26] <dclarke> Gman : they have *no clue* [00:06:58] <dclarke> Gman : all I ever wanted to do with Blastwave was ensure that everyone on both x86 and Sparc has access to everything up to date for free for ever [00:07:08] <dclarke> Gman : I thought .. it was a "good thing" to do [00:08:50] <Tpenta> [00:09:00] * dclarke watches a tumble weed roll by [00:09:14] <Tpenta> Coolies, looks like s10 got it's common criterea certification http://blogs.sun.com/jimlaurent/entry/solaris_10_common_criteria_security [00:09:18] <dclarke> hello Tpenta [00:09:30] <coffman> dclarke: since u did not answered me, why does blastwave not provide the patches that they use or there build scripts? [00:09:31] <Gman> dclarke, i agree with you, your ideals, blastwave [00:09:34] <Gman> you're preaching to the crowd [00:09:43] <dclarke> sorry [00:09:48] <Theoden-Nexenta> Man - most folks wouldn't watch weed tumble by - they'd light that sucker! [00:10:01] <dclarke> I often need to preech to Linux people and in IT departments and they argue like crazy [00:10:28] <Tpenta> hi dennis [00:10:31] <dclarke> the worst was last week when a guy told me that Solaris x86 does not exist [00:10:38] <Theoden-Nexenta> dclarke: Linux users are like right-wing conservatives - religious zealots. [00:10:39] <dclarke> and I blame Sun marketing for that fight [00:10:58] <Tpenta> dennis, that's when you pull out your solaris notebook [00:10:59] <dclarke> Tpenta : should I update my non-debug build article ? [00:10:59] <Tpenta> :) [00:11:24] <Tpenta> ummmmm i havent looked at it in a bit. does it still talk about having to modify usr/src/Makefile? [00:11:27] <dclarke> Tpenta : I'd rather just show them the Sun website and then say gee .. click there [00:11:38] <dclarke> Tpenta : yeah .. its a hack [00:11:46] <Tpenta> or hand them a nexenta boot cd :) [00:11:56] <Tpenta> the live cd i mean [00:12:10] <dclarke> Tpenta : so .. I was thinking that today I need to fix the Blastwave HOWTO .. update the build article [00:12:13] <Odin-LAP> Nexenta has two major flaws, IMO. [00:12:20] <dclarke> solve world hunger .. get some sleep [00:12:26] <Tpenta> :) [00:12:29] <jmcp> dclarke: check your inbox ... the opensolaris lists are now being harvested by the 419ers [00:12:36] <Odin-LAP> It's in fairly early stages, and it's mostly juts Ubuntu [00:12:39] <Odin-LAP> Bah. [00:12:39] * Tpenta is finally able to relax a bit. kids went back to school today [00:12:44] <dclarke> Tpenta : what is a 419er ? [00:12:50] <Tpenta> ? [00:12:50] <Odin-LAP> just Ubuntu with OpenSolaris kernel. [00:12:53] <Tpenta> context? [00:13:01] <jmcp> dclarke: Nigerian scammer [00:13:02] <Odin-LAP> dclarke: Nigerian 419 scammers, I suppose. [00:13:09] * dclarke oh hell [00:13:10] <Tpenta> that could be it [00:13:31] <jmcp> http://en.wikipedia.org/wiki/Nigerian_419 [00:13:35] <jmcp> "advance fee fraud" [00:13:44] <dclarke> wow .. Blastwave website is a tad busy ... [00:13:45] <dinolinux> Odin-LAP: i agree with you there, when i tried nexenta, it was buggy as well [00:13:54] <dclarke> Generating summary report [00:13:55] <dclarke> Saving history information... [00:13:56] <dclarke> 2173064 records (9 bad) in 391.35 seconds, 5552/sec [00:14:00] <Tpenta> dclarke: as long as it's not us pinging it [00:14:03] <dclarke> 2 million hits [00:14:13] <steleman> yay. my new and improved DSL is up. [00:14:23] <dclarke> Tpenta : nah .. its the article in ComputerWorld making the rounds [00:14:24] <Odin-LAP> dinolinux: I find it quite usable, though. But yeah, buggy. [00:14:34] <Tpenta> that was a good interview dennis. well done [00:14:36] <steleman> 7 down 1.5 up. [00:14:39] <Odin-LAP> dinolinux: That's part of the "early stages" thing. ;) [00:14:44] <dinolinux> yeah [00:15:04] <dinolinux> it was completely starting-stage when i tried it [00:15:16] <dinolinux> it was a bit over a year ago [00:15:19] <dclarke> Tpenta : the guy cut me short .. I was saying a LOT of things but really I was trying to get across that OpenSolaris is real UNIX and it could easily run everywhere and that means on IBM hardware too [00:15:45] *** __william__ has joined #opensolaris [00:16:03] <Odin-LAP> dinolinux: I've been using it on my desktop for around half a year now. [00:16:04] *** b3stbuddy has left #opensolaris [00:16:09] <Tpenta> i think you still came over pretty well. there was a definite solaris on power push type spin :) [00:16:52] <Theoden-Nexenta> the latest verion of nexenta is pretty good IMHO. [00:16:59] <dclarke> I think that I do what I can .. and the idea that Solaris can be everywhere was what he was after .. that IBM has reasons to get involved .. or be scared [00:17:04] <Odin-LAP> dclarke: The "real UNIX" thing spooks some people, methinks. Especially Linux types. (Of which I was one. Solaris machismo *really* looks offputting from that viewpoint.) [00:17:13] <Tpenta> a quotable quote my friend [00:17:32] <jmcp> dclarke: got an url? [00:17:36] * dclarke blush [00:17:40] <dclarke> Thank You .. [00:18:08] <dclarke> let me get a URL [00:18:21] <jmcp> ta [00:18:22] <dclarke> its in syndacation now so .. a google for Solaris and Dennis Clarke will work [00:18:42] <jmcp> ta [00:19:16] <dclarke> just a sec .. its now on a IBM website [00:19:23] <Tpenta> rofl [00:19:32] * Odin-LAP ponders what the URL should be to... [00:19:35] <jmcp> dclarke: "computerworld solaris dennis clarke" [00:19:39] <jmcp> http://www.computerworld.com.au/index.php/id;443780237;fp;16;fpid;0 [00:19:45] <dclarke> http://www-03.ibm.com/developerworks/blogs/page/powerarchitecture [00:19:59] <dclarke> its "readers digest" summary is there also [00:20:31] <jmcp> ah [00:20:32] <jmcp> cool [00:20:33] <jmcp> thanks [00:21:12] *** sartek has quit IRC [00:21:37] <dclarke> http://www.computerworld.com/action/article.do?command=viewArticleBasic&articleId=9009122&intsrc=news_ts_head [00:24:03] <Odin-LAP> Huh. [00:24:04] * dclarke reads http://www.blastwave.org/man/init_1m.html [00:24:14] <dclarke> huh ? [00:24:15] <Odin-LAP> On that subject, how far along is the PPC port? [00:24:28] <dclarke> we are fighting towards a shell prompt [00:24:36] <dclarke> not as easy as you would think [00:24:51] <dclarke> we need to write our own memory management [00:24:55] <Odin-LAP> dclarke: A generic interested interjection. [00:24:56] *** _william_ has quit IRC [00:24:59] <Odin-LAP> Hm. I see. [00:25:01] <dclarke> vmhat and ... [00:27:19] <dclarke> be right back ... [00:28:59] *** __william__ is now known as _william_ [00:33:09] *** postwait_ has quit IRC [00:33:45] *** sartek has joined #opensolaris [00:34:31] <dclarke> WARNING: add_spec: No major number for fp oops http://rafb.net/p/pmVNAx77.html [00:35:11] <dclarke> this machine has no fibre devices .. so .. whats up with that error ? [00:40:39] *** _william_ has quit IRC [00:41:56] *** jmcp has quit IRC [00:42:32] *** McBofhnyah has quit IRC [00:42:40] *** nachox has joined #opensolaris [00:42:54] *** McBofhnyah has joined #opensolaris [00:43:48] *** jmcp has joined #opensolaris [00:44:16] *** coffman has quit IRC [00:44:19] <dclarke> now I'm confused .. why would my first zone have zone number 3 ? [00:44:29] <dclarke> # zoneadm list -vc [00:44:30] <dclarke> ID NAME STATUS PATH BRAND [00:44:32] <dclarke> 0 global running / native [00:44:33] <dclarke> 3 ora10g001 running /zone/ora10g001 native [00:45:54] <richlowe> you had two other zones running at the time you booted that zone? [00:46:58] <dclarke> nope [00:47:02] <dclarke> fresh install [00:47:06] <dclarke> first zone [00:47:14] <dclarke> snv_55b [00:47:19] <dclarke> weird eh ? [00:47:42] <nachox> you had other zones configured? [00:47:56] *** tsoome has quit IRC [00:48:49] <richlowe> nachox: zones don't have zoneids until they boot. [00:49:13] <dclarke> no .. just a global zone [00:49:18] <dclarke> never had anything else [00:49:21] <nachox> right [00:49:31] <dclarke> # uptime [00:49:33] <dclarke> 6:50pm up 7:26, 2 users, load average: 0.00, 0.01, 0.01 [00:49:48] <dclarke> # ls /etc/zones [00:49:50] <dclarke> SUNWblank.xml SUNWlx.xml ora10g001.xml [00:49:52] <dclarke> SUNWdefault.xml index [00:50:01] <dclarke> so .. I was expecting a zoneid of 1 [00:50:02] *** yarihm has quit IRC [00:50:17] <comay> dclarke: your zone went through a few reboots the first time (once after the sysidtools ran, for example) [00:50:30] <dclarke> yes .. definately [00:50:50] <dclarke> so .. if I set the zone autoboot to false and reboot this entire box [00:51:00] <dclarke> then I will get a zoneid of 1 perhaps [00:51:22] <comay> actually, if you just reboot the box the new zone will end up with a zoneid of 1 [00:51:23] <dclarke> not that its a variable of any value to any application .. maybe its available in kstats .. not sure [00:53:00] <axisys> i am still trying to find out if i can take a mirror disk from 450 and boot a netra t1 with that [00:53:35] <axisys> i am getting /usr partition bad error [00:53:40] *** uebayasi has joined #opensolaris [00:53:45] <dclarke> axisys : ? no way [00:53:46] <axisys> but i fsck'd that couple times w/ clean output [00:53:50] <dclarke> axisys : ? no way wwould that boot [00:54:08] <dclarke> axisys : boot the CDROM in single user mode .. then mount the filesystems on that disk [00:54:09] <axisys> dclarke: please shed some light to my ignorance [00:54:17] <dclarke> axisys : I'll try [00:54:25] <axisys> dclarke: :-) [00:54:28] <richlowe> dclarke: only as an implementation detail of id_space, the canonical way to refer to a zone is its name. [00:54:31] <dclarke> axisys : the disk was part of a mirror right ? SVM ? [00:54:43] <axisys> dclarke: right.. also sol 8 that is [00:54:45] *** postwait_ has joined #opensolaris [00:55:04] <dclarke> richlowe : I'm still not sure of the various function calls to make to gather zone info from the global zone [00:55:09] <axisys> i cleaned the vfstab and removed the entries from systems file [00:55:13] <dclarke> richlowe : its bit of a learning curve [00:55:33] <dclarke> richlowe : and I'd rather enumerate zones based on numbers .. not ASCII strings [00:55:39] <axisys> for the disk i am playinh w/ after i detached it [00:56:09] <dclarke> axisys : well .. this is a toughie .. to pull a disk from one machine and take it to another and think .. all will be well [00:56:12] <sommerfeld> axisys: the supported way of doing that involves using flash archives; it does fixups for device-space related stuff [00:56:14] *** gm152 has joined #opensolaris [00:56:38] <dclarke> sommerfeld : I was thinking devfsadm with the root pointed to somewhere else [00:56:58] <axisys> sommerfeld: i got the idea of yours.. [00:57:05] <sommerfeld> the trick is getting the root mounted read-write. [00:57:22] <sommerfeld> .. when the correct device node doesn't already exist. [00:57:25] *** jsubl2 has joined #opensolaris [00:57:26] <dclarke> ya know .. boot in single user mode to the CDROM/DVD then /usr/sbin/devfsadm -v -C -r /mnt/root_of_disk [00:57:41] <dclarke> that should fill in the device paths [00:57:57] <axisys> dclarke: let me try that [00:58:23] <dclarke> axisys : just be sure to mount everything from that disk in the proper place [00:58:28] *** polk__ has quit IRC [00:58:45] <axisys> dclarke: ok [00:58:48] <dclarke> axisys : really the root filesystem should be what matters but if /var is in there too .. don't know if its needed [00:59:35] *** polk__ has joined #opensolaris [00:59:42] <axisys> axisys: booting again from sol 10 cdrom [00:59:53] <axisys> dclarke: oops that was for u.. damn tab compl [01:00:07] <dclarke> no prob .. I'm hear too [01:00:10] <dclarke> here [01:00:13] <dclarke> er ! [01:00:22] <axisys> dclarke: i made sure the vfstab points to c1t0d0 [01:00:48] <dclarke> is that what you expect on the Netra ? [01:00:55] <axisys> 450: c0t1d0 (second disk) --> Netra t1: c1t0d0 (root disk) [01:00:57] <dclarke> I wwould think that it would be c0t0d0s0 [01:01:14] <axisys> dclarke: i boot from cdrom and ran format to verify [01:01:21] <dclarke> okay [01:01:22] <axisys> on n t1 [01:01:32] <dclarke> Netra T1 .. what model ? [01:01:40] <axisys> then i will try your trick the devfsadm that is [01:01:51] <axisys> Netra T1 200 (UltraSPARC-IIe 500MHz) [01:01:57] <dclarke> UltraSPARC-IIi-cEngine [01:02:00] <sommerfeld> controller assignments also depend on /etc/path_to_inst [01:02:10] <dclarke> no no .. its UltraSPARC-IIe [01:02:26] <axisys> sommerfeld: should i remove it /etc/path_to_inst [01:02:31] <dclarke> yeah .. I was thinking that the controlelr path cna be reset with a boot -srv [01:02:44] <sommerfeld> and so when booting with the path_to_inst of a different system, the controller numbers will be .. different. [01:03:05] <axisys> so boot -srv will be done after the devfsadm ..correct? [01:03:07] <sommerfeld> axisys: I'm not sure. [01:03:21] <sommerfeld> axisys: you're deep into "if it breaks, you get to keep both broken pieces" terrain. [01:03:38] <dclarke> yep .. I was thinking something more .. colorful [01:03:46] <axisys> sommerfeld: netra t1 is not in prod yet [01:04:00] <axisys> so not worry if it breaks [01:04:13] <axisys> i will put it back into 450 and remirror if i have to [01:04:16] * dclarke I'm thinking it should never be in prod at this rate ... [01:04:31] <axisys> ok i booted from cdrom now [01:04:37] <axisys> ready for the devfsadm [01:04:44] <dclarke> no no [01:04:53] <dclarke> did you mount the disk ? [01:05:16] <dclarke> mount the root slice from that disk onto /mnt [01:05:56] <axisys> not yet... damn it.. i just realized i put that disk back into 450 already .. before i left work [01:06:01] <axisys> shit [01:06:24] <dclarke> oh hell .. you need the new improved quantum mechanical devfsadm that bends time and spce [01:06:25] <axisys> i guess i can go ahead remirror it anyways [01:06:30] *** McBofhnyah has quit IRC [01:06:32] <axisys> dclarke: hehe [01:06:43] <axisys> i will try it tomorrow.. [01:06:48] <dclarke> yep [01:06:49] <sommerfeld> definitely need to adapt those storagetek autochanger robots to handle disks.... [01:06:50] <axisys> dclarke, sommerfeld thanks guys [01:07:05] * dclarke thinks .. gee .. didn't get anything done [01:07:06] *** McBofhnyah has joined #opensolaris [01:07:39] <dclarke> open question : wwhere do we stand on kernel parameters for things like Oracle ? I think that kernel parameters are evil and should be avoided but am I shafted by Oracle on this ? [01:08:02] <dclarke> like set semsys:seminfo_semmni=100 poo in /etc/system [01:08:24] <dclarke> or use resource control projects and hope for the best ? [01:09:01] <nachox> hmm, interesting mail from joerg about a little sparc for appliances [01:09:42] <dclarke> Tpenta : hello ? you working on ND bits for 20070122 ? [01:10:35] <Tpenta> I could. I kinda dropped back to delivering for specials and named, is there a reason to do this one? [01:10:47] <brendang_> I haven't done a recent Oracle install, but I was hoping that Oracle did it's own process resource control requests; up to the privileged limits anyway... [01:10:50] <dclarke> not really [01:10:51] <Tpenta> 32 bit sparc might get "interesting" [01:11:03] *** Theoden-Nexenta has quit IRC [01:11:14] <dclarke> I have no idea what er even bother with 32-bit Sparc [01:11:41] *** Theoden-Nexenta has joined #opensolaris [01:11:42] <dclarke> Tpenta : I was going to kick off a build but .. non-debug is the only way I fly anymore [01:12:07] <nachox> i thought there were no 32 bit sparc kernels left in solaris though [01:12:34] <jmcp> nachox: x86 lives [01:12:43] <dclarke> nachox : I see no reason for them to exist at all anymore [01:13:02] <nachox> dclarke: appliances do not need 64 bits [01:13:07] <dclarke> nachox : given that its a fight to get some people to even look at slow 1.6 GHz UltraSparc [01:13:19] <nachox> so i do see a good reason for them to exist [01:13:20] <dclarke> slow .. thats what they call that [01:13:32] * dclarke shrug [01:13:52] <dclarke> I figure ARM or PowerPC is the way to go [01:13:59] <dclarke> lower power all the way [01:14:19] <dclarke> heck .. even the V40z has a PowerPC chip in it for the LOM [01:14:29] <sommerfeld> dclarke: I thought a bunch of the oracle-originated semaphore, etc., tunable voodoo was obsoleted by s10 [01:14:45] <dclarke> sommerfeld : God I hope so [01:14:51] <sommerfeld> all that became dynamically allocated at runtime with no upper limit other than available memory [01:15:10] <dclarke> sommerfeld : the Oracle installer may barf at me however [01:16:07] * dclarke reads http://download-uk.oracle.com/docs/cd/B19306_01/install.102/b15697/toc.htm [01:17:04] *** jesse-jads has joined #opensolaris [01:18:59] <delewis> sommerfeld, you still have to tune one or two things. [01:19:14] <delewis> the Oracle docs cover that, though. [01:19:38] <elektronkind> dclarke: use projects for the oracle semaphore/shared memory config [01:20:06] <elektronkind> user.oracle:100::::process.max-sem-nsems=(priv,256,deny);project.max-sem-ids=(priv,128,deny);project.max-shm-ids=(priv,128,deny);project.max-shm-memory=(priv,4294967296,deny) [01:20:15] <dclarke> prctl -n project.max-shm-memory -i project oracle [01:20:33] <sommerfeld> the architecture is that solaris should to the greatest extent possible self-tune at runtime. [01:20:38] <sommerfeld> but we don't always get there [01:20:54] <dclarke> I figure with snv_55b I want to test that theory [01:21:09] <delewis> dclarke, just make sure you change your JAVA_HOME [01:21:17] <dclarke> oh ? [01:21:24] <delewis> snv_55b ships JSE 1.6, which breaks pretty much anything and everything. [01:21:26] <dclarke> to some instance in /usr/foo ? [01:21:34] <dclarke> okay .. good one [01:21:36] <delewis> dclarke, yes, you want Java 1.5 [01:21:37] <dclarke> thanks [01:21:57] <dclarke> I guess I better set that in /etc/default/login [01:22:11] <dclarke> as well as the .profile for the various users/accounts [01:22:40] *** polk__ has quit IRC [01:22:48] <dclarke> elektronkind : those are the entries in your /etc/project ? [01:22:58] <elektronkind> dclarke: yah [01:22:59] *** swa_mobil has joined #opensolaris [01:23:33] <dclarke> I'll go for jdk1.5.0_10 [01:23:40] *** polk_ has joined #opensolaris [01:24:04] <elektronkind> user.oracle:100:Oracle Database:oracle:dba,oper,oinstall:process.max-sem-nsems=( [01:24:06] <elektronkind> priv,256,deny);project.max-sem-ids=(priv,100,deny);project.max-shm-ids=(priv,100 [01:24:06] <elektronkind> ,deny);project.max-shm-memory=(priv,4294967296,deny) [01:24:11] <elektronkind> there's a "better" one to use [01:24:14] <dclarke> so JAVA_HOME should be /usr/jdk/instances/jdk1.5.0 [01:24:37] <delewis> yes [01:24:47] <dclarke> perfect .. I'll set that [01:26:24] <delewis> this time next week I'll be doing an Oracle install on my laptop once it arrives [01:26:36] <elektronkind> dclarke: as the oracle user, you can verify with 'id -p' and 'projects -l user.oracle' [01:27:19] <dclarke> one sec .. trying to catch up as I document this also [01:27:26] <dclarke> I'm a total Oracle noob [01:27:37] <elektronkind> same here. I struggle. [01:27:44] <dclarke> okay . .cool [01:27:55] <dclarke> I just realized .. I'm doing this wrong [01:27:58] <delewis> its not so bad. come your 10th Oracle install, you'll be de-sensitized. [01:28:08] <elektronkind> I'm dangerous enough to know how to apply patches and what 'shutdown immediate' does. [01:28:09] <dclarke> damn .. I need to do this in the zone .. not in Global zone [01:28:10] <elektronkind> heh [01:28:22] <dclarke> uadmin 2 0 [01:28:45] *** flyinglunatic has left #opensolaris [01:30:16] <delewis> zones+ZFS make a nice Oracle testing environment. [01:30:40] <delewis> if the install or database creation process goes bad, just rollback. [01:30:43] <dclarke> I'm doing a Zone plus UFS [01:31:27] <delewis> I've got two sets of snapshots: one for pre-Oracle install and one for pre-database creation. [01:32:03] <dclarke> that is an excellent way to do things [01:32:13] <delewis> makes life easy :-) [01:32:20] <dclarke> but I'll start with UFS and then come up with a way to migrate [01:32:21] <delewis> Oracle is a PITA to remove from a system. [01:32:40] <delewis> removing a database has gotten much better with 10g, but I still prefer to make sure its really, really gone :-) [01:32:48] <dclarke> should be easy .. run the disks ove a large oscillating magnet [01:32:49] <delewis> pre-10g you didn't even have a drop database command [01:33:10] * dclarke chuckles at http://www.opensolaris.org/os/community/mdb/architecture/ [01:33:17] <elektronkind> any suggestions for 10g on ZFS? [01:33:31] <delewis> which meant you had to go around and find all the datafiles, etc. [01:33:38] <elektronkind> I'm trying to figure out the best layout, what to do with block sizs and all [01:34:11] <delewis> elektronkind, www.solarisinternals.com/wiki/index.php/ZFS_Best_Practices_Guide [01:34:24] <delewis> solarisinternals.com seems to be down, though :-( [01:35:10] <delewis> my Oracle installs are just for development/evaluation, so I'm just using RAID-Z and no tuning at all. Far from optimal :-) [01:35:28] <elektronkind> gotcha [01:35:36] <dclarke> here is something that should be a bug .. imo .. if I create a zone which uses a network interface . .say bge1 then the Global zone ought to be able to boot and not give me heartache about bge1 not being plumbed [01:36:11] <delewis> hmm, never seen that. [01:36:18] <delewis> I take it bge1 actually exists? [01:36:27] <dclarke> yep [01:36:30] <delewis> weird. [01:36:43] <delewis> I've never had to plumb interfaces my zones use. [01:36:58] <dclarke> must be a snv_55b *thing* [01:37:09] <dclarke> thales console login: Jan 29 19:34:02 thales zoneadmd[565]: [zone 'ora10g001'] WARNING: skipping interface 'bge1' which may not be present/plumbed in the global zone.: No such device or address [01:37:26] <dclarke> # ls /etc/hostn* [01:37:28] <dclarke> /etc/hostname.bge0 [01:38:27] <dclarke> http://rafb.net/p/S2pv1S82.html [01:44:13] <comay> dclarke: you see to have bge0 plumbed in the global zone but you've specified bge1 for the non-global zone [01:44:24] <dclarke> yes [01:44:30] <comay> that's how it's always worked :-) [01:44:33] <dclarke> exactly [01:44:41] <dclarke> what ? [01:44:52] <comay> until IP instances, the global zone must have each interface you want to use plumbed [01:44:53] <dclarke> I don't want the zone to have anything to do with bge0 [01:45:07] <dclarke> the zone uses bge1 [01:45:16] <dclarke> and only bge1 [01:45:24] <dclarke> the global zone does not use bge1 [01:45:33] <comay> understood; but until IP instances is available, you're forced to plumb bge1 in the global zone [01:45:43] <dclarke> oh .. [01:45:50] <comay> it doesn't have to be up & running, but it needs to be plumbed [01:45:55] <dclarke> I guess I have to stick in a ifconfig in a script somewhere [01:46:38] <dclarke> or maybe touch /etc/hostname.bge1 [01:49:33] <dclarke> yep .. that works perfectly [01:50:11] <dclarke> reboot set the zoneid to 1 also .. cool [01:50:14] <tg> hey [01:50:17] <dclarke> # zoneadm list -vc [01:50:19] <dclarke> ID NAME STATUS PATH BRAND [01:50:20] <dclarke> 0 global running / native [01:50:22] <dclarke> 1 ora10g001 running /zone/ora10g001 native [01:50:35] <dclarke> ho [01:50:38] <tg> anyone tried zfsboot? [01:50:57] *** comay has quit IRC [01:51:11] <jmcp> tg: it's not available, properly, yet [01:51:32] <tg> i found some source and howto on opensolaris.org [01:52:09] <tg> but its for b44 i think [01:52:10] <jmcp> so what I meant was, yes, it's around, but you have to be (a) confident with bfu, (b) not scared of killing your system, and (c) willing to file bug reports [01:52:32] <dclarke> well .. at least (a) and (c) [01:52:33] <jmcp> tg: I was using it before I got a new system. ... it was still a bit of a kludge though [01:52:47] <dclarke> always fear (b) [01:52:57] <nachox> speaking of weird features, does anyone know what happened to the virtual consoles project? [01:53:06] <jmcp> dclarke: yeah ... well, the chances of killing your system (try sys-unconfig on a bfu'd zfsroot box) are somewhat greater [01:53:12] <dclarke> they were taken out back and shot [01:53:26] <tg> i got some new disks [01:53:37] <tg> and trying to install solaris now [01:53:50] <tg> so i'm not afraid of killing it;) [01:53:51] <dclarke> jmcp : once upon a time long long ago it was great fun to NFS share out something from a zone and then mount it in the Global zone .. boom [01:54:12] <nachox> hehe [01:54:35] <jmcp> :) [01:55:35] <nachox> try creating a few files, export then using lofiadm, create a mirrored pool with zfs and overwrite the files :) that used to kill my old nevadas [01:56:21] <dclarke> so many kewl ways to blow up a server [01:57:42] *** nachox has quit IRC [01:58:10] *** nachox has joined #opensolaris [02:00:12] <_syphilis_> why is the SUN(SM) ALERT WEEKLY SUMMARY REPORT delivered in uppercase now? [02:00:49] <jmcp> _syphilis_: SO YOU CAN BE PROPERLY ALARMED BY IT [02:00:50] <dclarke> because you weren't bloody listening when it was softly spoken ! [02:02:58] <nachox> ... [02:04:43] *** benr has quit IRC [02:08:49] <Auralis> because spam went back to lower case subjects [02:10:09] * jmcp lunches [02:13:45] *** DebolazY has joined #opensolaris [02:15:10] *** sommerfeld has quit IRC [02:15:30] *** alanc_away has quit IRC [02:15:30] *** doof has quit IRC [02:15:30] *** mikaeld has quit IRC [02:15:54] *** alanc_away has joined #opensolaris [02:15:54] *** doof has joined #opensolaris [02:15:54] *** mikaeld has joined #opensolaris [02:22:33] *** rachel_ has quit IRC [02:23:03] *** rachel_ has joined #opensolaris [02:26:54] *** whaq_ has joined #opensolaris [02:27:50] *** yongsun_ has joined #opensolaris [02:28:01] *** whaq has quit IRC [02:30:25] *** DebolazX has quit IRC [02:32:24] * Tpenta swears. I just remembered I left a 400g chocolate in the car, ... it's liquid. [02:32:49] *** schily_ has joined #opensolaris [02:33:02] <dclarke> oh hell [02:33:06] <dclarke> at 44 C [02:36:23] <Tpenta> just laid flat in the fridge and crossed fingers [02:37:08] *** Odin-LAP has quit IRC [02:37:24] <dclarke> oh .. but .. [02:37:43] <dclarke> but it has already been liquified and now you hope to solidify it .. [02:37:52] <dclarke> it never tastes quite the same [02:38:07] *** lloy0076 has joined #opensolaris [02:38:08] <dclarke> damn .. thanks .. now I have to go searching for chocolate in the house [02:39:46] *** freakazoid0223 has joined #opensolaris [02:40:54] <steleman> nutella. [02:41:05] <steleman> i must. [02:41:22] *** schily has quit IRC [02:50:33] *** CosmicDJ_ has quit IRC [02:50:43] *** MikeTLive has joined #opensolaris [02:52:31] <Tpenta> dclarke, I'm really upset abt that, I saw that the 400g block was only 50c more expensive than the 250, ... now i got gloop. [02:52:38] <Tpenta> i was looking forward to tha [02:52:40] <Tpenta> t [02:52:57] <dclarke> perhaps this is serendipity talking to the waistline ? [02:53:23] <Tpenta> probably [02:53:34] * dclarke look for the silver lining [02:53:41] * dclarke finds gloop [02:53:45] <dclarke> sorry [02:53:48] <Tpenta> the sil;ver lining is tarnished [02:57:16] <dclarke> this is odd [02:57:24] <dclarke> # grep staroffice8 /var/sadm/install/contents returns nothing [02:57:35] <dclarke> I pkgrm'ed all startoffice packages [02:57:36] *** stevel has quit IRC [02:57:47] <dclarke> and yet /opt/staroffice8 remains [03:00:07] <richlowe> I don't see a package with a 'd' entry for /opt/staroffice8 [03:00:23] <richlowe> ... which goes back to my prior complaint about that only being a warning, rather than error as far as pkgmk is concerned. :) [03:03:10] *** yippi has joined #opensolaris [03:03:25] <dclarke> yep .. as a guy that looks at package proto files .. a lot .. that would stop me from a release right there [03:04:31] <dclarke> be back in a bit .. [03:06:33] *** adlpaf has joined #opensolaris [03:10:34] *** halton has joined #opensolaris [03:11:52] *** simford has joined #opensolaris [03:13:42] *** comay has joined #opensolaris [03:19:51] *** phips has quit IRC [03:31:25] *** nbkk6fo has joined #opensolaris [03:31:30] *** nbkk6fo is now known as rodrick-brown [03:41:17] *** nachox has quit IRC [03:42:58] *** Gman is now known as GmanAFK [03:49:22] *** adlpaf has quit IRC [03:50:06] *** mrdeviant has joined #opensolaris [03:53:49] *** McBofhnyah has quit IRC [03:58:18] *** jmcp has quit IRC [04:07:29] *** hile_ has joined #opensolaris [04:14:42] *** Peanut has quit IRC [04:14:48] *** Peanut has joined #opensolaris [04:16:59] *** jbk has quit IRC [04:17:05] *** jbk has joined #opensolaris [04:26:13] *** ofu_ has joined #opensolaris [04:28:37] *** mrdeviant has quit IRC [04:31:10] *** ofu has quit IRC [04:54:22] *** hell has joined #opensolaris [04:55:13] <hell> anyone know what the deal is with the recent versions of jdk install always say corrupt file [04:55:42] <hell> Unpacking... [04:55:42] <hell> Checksumming... [04:55:42] <hell> The download file appears to be corrupted. [04:56:01] *** uebayasi has quit IRC [05:01:04] *** maram has joined #opensolaris [05:03:35] *** uebayasi has joined #opensolaris [05:04:23] *** laca has quit IRC [05:04:46] *** jsubl2 has quit IRC [05:05:41] *** bank__[1] has quit IRC [05:08:22] <elektronkind> yawn [05:14:28] *** jmcp has joined #opensolaris [05:17:01] <elektronkind> heye james [05:18:25] <jmcp> hi elektronkind [05:18:44] <jmcp> had some dsl issues ... telco bloke came out ..... problem is in the pit down the road from me [05:21:59] <elektronkind> corroded x-connect I bet. that got me several times when I had dsl [05:22:52] <elektronkind> jmcp: off the top of your head, you wouldn't know of a way to see how many pending I/Os there are in ZFS's IO scheduler, would you? [05:23:06] <jmcp> sorry, drawing a blank on that [05:23:17] <jmcp> "gel-affected" is the term that the linesman used [05:23:17] <elektronkind> hmm, k [05:23:26] <elektronkind> heh [05:24:13] <richlowe> Hm. [05:27:17] <elektronkind> gel-affected. must be some local term :) [05:29:38] <jmcp> yeah, they pump some sort of gel into connection boxes to keep moisture out .... but according to a Senate inquiry that gel reacts with moisture and corrodes the copper! [05:31:09] <dclarke> hello [05:31:12] <richlowe> possibly vdev_t->vdev_queue? [05:31:29] <elektronkind> I guess that stuff matters more in that semi-tropical, oceanic venue of yours [05:31:44] <elektronkind> richlowe: hmmm, I'll check that out [05:31:58] <dclarke> up here in Canada we get near perfect conductivity via sub-zero temps [05:32:12] <dclarke> moister is the stuff frozen everywhere [05:32:17] <jmcp> elektronkind: ha :) [05:32:33] <elektronkind> up where you are I bet it gets cold enough to turn a ordinary sidewalk into a superconductor [05:32:48] <dclarke> not too bad tonight .. -12C [05:32:53] * jmcp lol [05:32:59] <dclarke> I was just out getting gas [05:33:07] *** McBofhnyah has joined #opensolaris [05:33:18] *** McBofhnyah is now known as McBofh [05:33:19] <elektronkind> what, you buy gas by the frozen cube there? [05:33:20] <dclarke> thankfully everyone with brains up here swiwtches to pure synthetic oil [05:33:39] <dclarke> that reminds me of a story [05:33:48] <dclarke> if anyone wants to hear it [05:33:50] <rodrick-brown> man can Sun like finally kill 8 or something plssssss [05:34:01] <rodrick-brown> like release new hardware that dont support 8 :-P [05:34:18] <dclarke> rodrick-brown : the T1000 and T2000 don't [05:34:27] <rodrick-brown> we dont use many of those :(I [05:34:28] <jbk> us-iv+ don't [05:34:33] <elektronkind> I'm up for a dclarke story [05:34:44] <jbk> the 1.5ghz + us-iv don't either i don't believe [05:34:47] <rodrick-brown> jbk yeah they do [05:34:53] <dclarke> when I was in the ARMY there was this listening station up North [05:34:56] <jbk> maybe it's the 1.8 then [05:35:05] <jbk> i thought it was pretty much down to like the v240s [05:35:05] <dclarke> thats North with a capital N .. so .. near the pole [05:35:29] <dclarke> the temp up there is -60 C or so [05:35:36] <dclarke> then it cools down at night [05:35:45] <dclarke> which lasts six months .. more or less [05:36:02] <dclarke> a C130 Herc transport carrier was flying in there [05:36:09] <dclarke> to drop supplies [05:36:23] <dclarke> the SOP standing order at the staion is that you never shut off the engines [05:36:30] <elektronkind> hopefully beer and strippers were on the manifest [05:36:35] *** yongsun_ has quit IRC [05:36:49] <dclarke> anyways .. the rookie pilot was flying in too fast and not taking into consideration that the airfield was just snow and ice [05:37:16] <dclarke> the fact that it was a "she" had nothing to do with it [05:37:40] <dclarke> anyways .. she landed and pushed the engines into revers thrust like mad but that was not enough to stop the aircraft [05:37:55] <dclarke> it went skidding into the end of the airfield and into deep snow [05:38:35] <dclarke> the radio controllers got her on the phone to count bodies and to check for injuries etc etc but failed to mention to not turn off the engines [05:38:42] <dclarke> visibility was 30 feet or less [05:38:47] <dclarke> pitch black [05:39:03] <dclarke> at this point a distance of 500 m or more was a rescue mission [05:39:13] <dclarke> because exposed flash freezes solid in seconds [05:39:32] <dclarke> with the engines off the aircraft would quickly turn into a big freezer [05:40:06] <dclarke> so all the skidoos and tractors on site were called into action to get out to the aircraft with medical gear etc etc and [05:40:21] <dclarke> the hope was to get there with a "start cart" for the aircraft [05:40:43] <dclarke> a start cart was just a big mobile pack of batteries and a deisel moter that ran a heater [05:40:58] <dclarke> anyways .. by the time they got there the oil was starting to gell [05:41:20] <dclarke> the first priority was the air crew that were not dressed for -60 weather [05:41:37] <dclarke> three guesses what the second priority was [05:41:45] <dclarke> anyone take a guess ? [05:42:01] <elektronkind> um, get the engines going? [05:42:17] <dclarke> you would think so .. but no [05:42:34] <SunTzuTech> put out the fire [05:42:37] <elektronkind> a frozen herc is a awfully large lawn ornament [05:42:38] <dclarke> thats a common guess even in the artic courses I took [05:42:50] <dclarke> yep .. it is .. but its not a priority [05:43:06] <elektronkind> ah the cargo... keep it from freezing [05:43:11] <dclarke> the ! [05:43:14] <dclarke> bingo [05:43:15] <dclarke> yes [05:43:18] <elektronkind> people, cargo, plane [05:43:28] <dclarke> when you are miles from the pole on the other side its not easy to get supplies up there [05:43:48] <dclarke> get the damn cargo .. the stuff that can not freez quickly [05:44:02] <dclarke> like radio parts and supplies like batteries [05:44:07] <dclarke> blood packs [05:44:12] <dclarke> medical gear [05:44:34] <dclarke> this entire incident became a textbook example of what goes wrong and the things that needed to be fixed [05:44:49] <dclarke> like .. where to load gear in a plane in case it fails like this [05:45:11] <dclarke> medical gear located behind fifteen tons of engine parts and tractor gear is not a very smart move [05:45:34] <dclarke> top make a long story short [05:45:36] <elektronkind> sheesh [05:45:45] <dclarke> the aircraft was scrapped on site [05:46:00] <dclarke> regardless of the fact that it was not really damaged [05:46:18] <dclarke> it just isn't feasible to fart around with outside in those conditions [05:46:33] <dclarke> unless in a state of war then losing lives or body parts is not an option [05:46:38] <dclarke> for a bloody aircraft [05:46:47] <richlowe> elektronkind: vdev->vdev_queue->vq_pending_tree, I *think* [05:46:55] <elektronkind> one young pilot's career was probably poo-poo'd [05:46:58] <dclarke> the act of touching metal up there wil lcost you your hand or at least your fingers [05:47:16] <dclarke> that young pilot may still be peeling potatoes somewhere [05:47:33] <dclarke> another loaded aircraft was dispatched right away [05:47:46] <dclarke> the oil frozen solid in the other one [05:47:47] <elektronkind> richlowe: I can fetch the contents of that with dtrace, eh? [05:48:04] *** yongsun_ has joined #opensolaris [05:48:27] <dclarke> the place .. if anyone wants to look it up .. is CFS Alert [05:48:51] <dclarke> Canadian Forces Station Alert .. its on the DEW Line .. the Norad listening station [05:49:06] <elektronkind> thule? [05:49:40] <elektronkind> wait, that's greenland [05:50:05] <elektronkind> oh damn, ellessmere [05:50:06] <richlowe> elektronkind: probably, I'm just hitting <up><enter> in mdb, really really fast. [05:50:11] <dclarke> anyways .. a six month tour up there got you a spec pay grade upgrade for life [05:50:13] <richlowe> it's the height of technology ;) [05:50:44] <elektronkind> pray tell, what's the mdb mojo you're using? [05:50:53] <dclarke> speaking of the height of technology .. I have just net booted snv_55b for Sparc [05:50:54] *** maram has quit IRC [05:50:56] <elektronkind> I'm a complete newb beyond the mdb basics [05:51:11] * dclarke just uses it for crash dump stuff [05:51:17] <dclarke> its deep voodoo [05:51:25] <dclarke> on purpose I think [05:51:28] <delewis> its better than if mdb did not exist :-) [05:51:34] <elektronkind> http://en.wikipedia.org/wiki/CFS_Alert [05:51:34] *** Theoden-Nexenta has left #opensolaris [05:51:34] <delewis> *cough* Linux *cough* [05:51:41] <richlowe> <address of vdev>::print vdev_t vdev_queue.vq_pending_tree | ::walk avl ! wc -l [05:51:45] <richlowe> though that should probably be [05:51:47] <elektronkind> dclarke: that crash appears to be mentioned [05:51:56] <richlowe> <address of vdev>::print vdev_t vdev_queue.vq_pending_tree.avl_numnodes [05:52:11] <dclarke> no way .. that was all classified in the day [05:52:34] <elektronkind> it's there in the wikipedia article [05:52:38] <dclarke> I got into crypto because of that posting [05:53:05] <dclarke> made for a sweet way to transfer into cool projects at Pratt & Whitney during the Gulf War [05:54:10] <dclarke> On October 30, 1991 an AIRCOM CC-130 Hercules ?? [05:54:15] <dclarke> nope .. thats not mine [05:54:26] <dclarke> no one died in 1989 [05:54:41] <elektronkind> must be a different crash then [05:54:47] <dclarke> The pilot apparently was flying by sight rather than relying on instruments. ! [05:54:52] <dclarke> stupid bastard [05:55:15] <SunTzuTech> that's a hell of a miss. 18 miles [05:55:18] <dclarke> pardon my language but that is like working with explosives while smoking [05:55:39] <dclarke> which reminds me of another story [05:55:47] <dclarke> this one is real real short [05:56:11] <dclarke> it was a miserable horrible day in CFB Petawawa .. home of the Airborne Regiment up here [05:56:18] <yongsun_> guys, I am trying to build the latest emacs-xft on solaris with sunstudio 11, but failed, do you have any hint? [05:56:19] <dclarke> where ice freezes in July [05:56:31] <dclarke> yongsun_ : pkg-get -i emacs [05:56:34] <SunTzuTech> ironic [05:56:50] <dclarke> anyways .. I'm on duty along with this other guy [05:56:56] <dclarke> its drizzilng [05:57:01] <dclarke> cold as hell [05:57:04] <dclarke> middle of the night [05:57:26] <dclarke> anyways .. we were just manning a damn radio point anyways [05:57:29] <yongsun_> dclarke, the emacs on blastwave is 21.4, [05:57:41] <dclarke> 21.4 .. what rev are you looking into ? [05:58:02] <yongsun_> dclarke, I am looking for 23.xx [05:58:14] <delewis> yongsun_, 23.xx hasn't been released. [05:58:19] *** yippi has quit IRC [05:58:20] <dclarke> Latest stable release: 21.4 (Feb 6, 2005) [05:58:21] <yongsun_> dclarke, refer to http://www.genunix.org/wiki/index.php/Emacs-xft, [05:58:40] <yongsun_> dclarke, I use the instructions for gcc, but still failed [05:59:00] *** deather_ has joined #opensolaris [05:59:03] <dclarke> drop me an email at dclarke at blastwave dot org and I talk with the emacs guys [05:59:15] <dclarke> see if we can snap it together for you pronto [05:59:19] <delewis> 23.xx is only available from CVS because it has not been released. [05:59:25] <dclarke> soundds like something we should have [05:59:30] <yongsun_> dclarke, thank you very much! [05:59:33] *** whaq_ has quit IRC [05:59:36] <dclarke> well hell .. thats what I thought [05:59:42] <dclarke> can't hurt .. ya know [06:00:05] <delewis> I recall having stability issues with the CVS sources (no surprise) [06:00:16] <delewis> so I just stick with the Sun Studio-supplied XEmacs, nowadays. [06:00:31] <dclarke> I'm certain that the emacs at Blastwave is hellish stable [06:00:46] <dclarke> anyways .. so I was standing in the rain freezing in the middle of the night [06:01:08] <dclarke> and I turn to my buddy and said .. screw this .. let's go sit in the ammo tent [06:01:09] <delewis> I've sort of found myself regressing back to vi for development work lately, though :-) [06:01:20] * dclarke uses vi all the time [06:01:23] <Tpenta> vi is not a regression [06:01:31] <delewis> vi, :set number, and I'm good to go. [06:01:34] <dclarke> the ammo tent is just that .. a tent .. a small one [06:01:35] <yongsun_> delewis, dclarke, emacs with out xft support is painful [06:02:21] <dclarke> but in this case it had 20 rounds of HE ARTY and WP ( white Phosphorus ) rounds [06:02:26] <yongsun_> dclarke, and I can build the latest cvs source on ubuntu with gcc successfully [06:02:29] <dclarke> plus small stuff [06:02:38] <delewis> yongsun_, that's not saying a lot :-) [06:02:39] <dclarke> yongsun_ : good to hear [06:02:42] *** gm152 has quit IRC [06:03:01] <dclarke> so anyways .. is anyone following these drinking stories ? [06:03:10] <dclarke> damn I was funny as hell in a bar [06:03:19] <e^ipi> everyone always is [06:03:20] <delewis> 21.xx had some serious compilation issues on Solaris that was later patched in CVS [06:03:24] <dclarke> anyways .. its pitch black ... like .. really dark [06:03:51] <dclarke> I am sitting on ammo crates in this tent and then buddy flicks open his lighter and lights a smoke [06:04:12] <delewis> *poof* [06:04:23] <dclarke> man .. talk about how fast I reached out and clawed that cig and lighter out of his face [06:04:37] <Tpenta> poof? Orientation/preference switch? [06:04:52] <dclarke> ammo is pretty damn stable ya know .. but thats just stoopid [06:05:02] <Tpenta> asking for trouble [06:05:23] <dclarke> asking for wings and a hrp but God will need to pick up the pieces first [06:05:28] <dclarke> harp [06:05:35] <dclarke> thats an old army joke [06:06:00] <dclarke> not exactly my closest call .. but that was scary as hell [06:06:03] *** deather has quit IRC [06:06:14] <dclarke> and then the guy .. for a sec .. a split sec .. got pissed at me .. then silence [06:06:21] <Tpenta> some people just dont know how to use that grey matter inside their skull [06:06:25] <dclarke> he didn't say a damn word for at least an hour [06:06:38] <dclarke> there is a lot of that in the army [06:06:59] <dclarke> which is why they use picture flip charts to train people hoe to open a ration pack [06:07:07] *** Vratha has joined #opensolaris [06:07:39] * Tpenta recalls a friend in the reserves over here in Oz. They were being visited by a group of marines and running some wargames around glenrock lagoon, comments like "look a walking tree" [06:07:59] <Tpenta> they were so down tha they had their asses handed to them by "guardsmen", engineers no less [06:08:23] <dclarke> ha ha .. I had the pleasure of working with some ozzie Artillery guys [06:08:35] <dclarke> and the ozzies were hellish sharp [06:09:02] <Tpenta> the aussie sas is right up there with the brits; theyu are scary [06:09:42] <dclarke> yep .. I worked with the SAS also .. [06:09:53] <dclarke> once upon a time .. I was insane [06:09:58] <dclarke> I'm much better now [06:10:04] <dclarke> hence the name .. Blastwave [06:10:13] <dclarke> but I have this permanent ringing in my ears [06:11:09] <dclarke> I actually had a riot of a time drinking with a pile of Aussie officers in CFB Gagetown one summer [06:11:23] <dclarke> geez ... the women up here went nuts because of the damn accent [06:11:28] <dclarke> it was shameful [06:11:41] <dclarke> so .. I guess I drank a lot on the sidelines watching that display [06:13:23] <dclarke> Blastwave : I used to work with very close proximity to explosive ordinance .. shaped charges, and clearance events .. like .. move that bridge [06:13:56] <dclarke> hrmmm .. I am at the stage here where snv_55b is asking for locale etc etc .. [06:14:10] <dclarke> will this smoothly upgrade my snv_52 system ? [06:14:19] <dclarke> its a little .. iffy to me [06:16:19] <dclarke> anyone file a bug report for the installer that asks me if I want to install extra value software ? specifically the #pragma idnet " at (#)SunVTS dot name1.5 06/08/04 SMI" Sun Validation Test Suite 6.3 70.0 MB [06:19:27] *** alanc is now known as alanc-away [06:19:32] <jbk> hmm.. so apparently you can't mount stuff on cvxfs filesytems.. makes sense i suppose [06:20:22] *** _syphilis_ has quit IRC [06:20:45] <jbk> the fact i was trying to mount a dataset from a mainframe didn't help diagnose the issue either :) [06:21:01] <dclarke> ummm .. am I missing something here [06:21:22] <jbk> i've just been working on stuff cursing until i figured out the problem [06:21:25] <jbk> and am rambling now :) [06:21:30] <dclarke> I get to that point where I am asked if I want to preserve data [06:22:21] <SunTzuTech> did you back it up to a write only device? [06:24:14] <dclarke> SunTzuTech : nope [06:24:51] <SunTzuTech> so you want to preserve your data? [06:24:54] <e^ipi> neat, HP i2000 on ebay for $100 [06:25:16] <dclarke> SunTzuTech : I'll halt the process and hit STOP-A [06:25:31] <dclarke> SunTzuTech : its not worth taking the chances [06:25:34] <e^ipi> openvms runs on itanic, right? [06:26:17] <SunTzuTech> I use preserve all the time. But then, I usually run with a single slice root, one or two LU partation and a big export which I preserve. The LU slices may or may not be preserved depending on what's in them [06:26:24] <jbk> e^ipi: i think so [06:28:15] <e^ipi> though i'm sure if i asked nice enough HP'd probably give me one [06:28:36] <jbk> probably.. not like they're really selling many of them [06:29:50] <jbk> i remember when the first itanics came out from hp.. it was quite amusing [06:30:03] <jbk> the 'desktop' was slightly smaller than a e450 [06:30:14] <jbk> required a 700W PS (this was in like 2000) [06:30:38] <jbk> couldn't run X w/ HP-UX.. had to use redhat or a beta copy of windows [06:30:49] <jbk> the sun admins weren't impressed :) [06:32:06] <e^ipi> i wonder who i'd contact about that sort of thing [06:33:40] <e^ipi> ultimately i'm just bored waiting for sun labs to throw more PPC code over the wall i think [06:35:38] <e^ipi> I don't want to do anything for fear of stepping over anyone's toes at the moment, since they've apparantly got some code pending for us [06:36:17] <lloy0076> Gawd I hate it when something just WORKS [06:36:18] *** yydzero has joined #opensolaris [06:36:27] <e^ipi> ? [06:37:24] <jmcp> lloy0076: no opportunity to delve? [06:37:45] <lloy0076> It normally means that my regression tests and testing have missed something. [06:38:21] *** swa_mobil has quit IRC [06:41:23] <lloy0076> This time, though, it meant I spent a heap of time coding something because I thought learning "betwixt" would be "difficult" only to find out that I could have done it using "betwixt" (from Jakarta Commons) woud have been much easier and probably more robust. [06:43:00] *** trygvis_ has joined #opensolaris [06:43:01] *** trygvis has quit IRC [06:44:31] *** bank__ has joined #opensolaris [06:49:40] *** dclarke is now known as dclarke_ZZZzzz [06:49:49] *** bank__ has quit IRC [06:55:16] *** galt has joined #opensolaris [06:55:20] *** galt has quit IRC [07:06:06] *** slowhog has left #opensolaris [07:13:39] *** lloy0076 has quit IRC [07:26:16] *** lopa has joined #opensolaris [07:26:29] <lopa> vista has like 20000 editions , now they think they are ADOBE and starting to release some Expression suite :) [07:26:48] *** andyshack has joined #opensolaris [07:28:15] *** yongsun__ has joined #opensolaris [07:29:27] *** yongsun_ has quit IRC [07:34:19] *** asimismo has joined #opensolaris [07:42:30] *** asimismo has left #opensolaris [08:00:10] *** logic has joined #opensolaris [08:14:04] *** lopa has quit IRC [08:18:43] *** Cyrille has joined #opensolaris [08:21:02] *** benignbala has joined #opensolaris [08:21:31] <benignbala> hello all, I am installing opensolaris in a lab. [08:21:46] <benignbala> can anybody help in how to create a flash archive? [08:23:11] <benignbala> I need some tutorials other than the one in sun. I cannot fully understand that. [08:30:25] *** boxix has quit IRC [08:33:10] <DerJoern> flarcreate -n "WebServer" -c -S -R / /export/install/flash [08:40:42] *** Dar_HOME has quit IRC [08:52:18] <trygvis_> is it not possible to run dtrace without root? [08:58:26] *** kawazu has joined #opensolaris [08:59:13] *** qdk has quit IRC [08:59:19] <kawazu> hi all... someone around capable of helping debugging an OpenSolaris system lockup? [09:05:58] *** kawazu has quit IRC [09:07:05] <CIA-22> vi117747: 6473179 SCTP assoc. should be resilient to addrs. being un-plumbed and re-plumbed on a diff. interface [09:11:23] *** Dar has joined #opensolaris [09:12:28] *** dunc has quit IRC [09:19:16] *** qdk has joined #opensolaris [09:21:16] *** mazon is now known as Mazon [09:22:52] *** tsoome has joined #opensolaris [09:27:28] *** pikapika has quit IRC [09:29:01] *** LordKing has joined #opensolaris [09:29:39] *** Triskelios has quit IRC [09:32:05] *** cmihai has joined #OpenSolaris [09:38:53] *** benignbala has quit IRC [09:39:37] *** deather_ is now known as deather [09:43:09] *** yydzero has quit IRC [09:44:21] <Gr|ffous> how is everyone this evening? [09:44:58] <Vratha> good; you? [09:46:17] *** trygvis_ is now known as trygvis [09:46:48] <tsoome> .oO it's morning here:P [09:47:33] <Vratha> yup yup [09:48:20] <Gr|ffous> ready for another hard nights abusing my compiler/linker :) [09:48:46] * Gr|ffous reads up on RPATHs more [09:48:59] <Vratha> i'm about finished with night abusing my compiler/linker/hardware [09:49:08] *** karrotx has quit IRC [09:49:14] <Vratha> i'm going to finish testing my device driver framework tomorrow i think... it's pretty much done now [09:49:27] <Vratha> for minix though... not opensolaris... i'll start on that in a few months [09:49:47] <Gr|ffous> heh, minix still has it's uses eh? [09:50:14] <Gr|ffous> is that a bit like being forced to learn latin in school, to 'expand ones' mind', even though you'll NEVER use it anywhere else? [09:50:21] <e^ipi> i bought tannenbaum's other book [09:50:37] <e^ipi> not the minix one [09:51:02] <Vratha> his minix book is ass [09:51:07] <Vratha> so which one did you get? [09:51:12] <e^ipi> modern operating systems [09:51:17] <Vratha> oh, that's a good one [09:51:21] <e^ipi> yeah [09:51:32] <Vratha> his distributed systems: principles and paradigms is also good [09:51:33] * Gr|ffous giggles slightly [09:51:37] <e^ipi> Gr|ffous: ? [09:51:56] <Vratha> e^ipi: uh, i was not forced to use this; this is part of my thesis [09:52:02] <Vratha> a userspace device driver framework [09:52:04] *** dunc has joined #opensolaris [09:52:07] <Vratha> i mean gri [09:52:10] <Vratha> Gr|ffous: [09:52:31] <Gr|ffous> It just seems funny that a guy that wrote an OS that well, never made it primetime, back in the early 90's, is writing about modern OSes. probably a bit unfair of me to laugh... it seemed funny though before I thought it through [09:52:32] <Vratha> you just watch when i put it on solaris :) [09:52:54] *** bzcrib has joined #opensolaris [09:53:01] <Vratha> he knows a lot about OS... just because your OS doesn't become mainstream doesn't mean you're stupid about modern OS design [09:53:03] <e^ipi> Gr|ffous: he's an academic [09:53:06] <e^ipi> not a marketing guy [09:53:52] <Gr|ffous> is minix the mach based one? [09:53:53] <Vratha> and his Modern OS book hardly touts microkernels as "the one true way", which is nice... he shows what's actually used and some further design concepts [09:54:02] <Vratha> no, minix has never been and never will be based on mach [09:54:12] <Vratha> you're thinking of the hurd [09:54:41] <Gr|ffous> hmm... ok clue me up, there is monolithic or microkernel as the two main types right. Where does mach fit in? [09:54:56] <Vratha> mach is a hybrid [09:55:08] <Vratha> started as a microkernel, but it proved too slow [09:55:15] <Gr|ffous> fear the cylons! [09:55:18] <Vratha> windows and VMS are also based on mach [09:55:23] <Vratha> and OS X has a mach subsystem [09:55:44] <Gr|ffous> windows even today isn't monolithic? [09:55:47] <e^ipi> and it's also slow as crap [09:55:52] <e^ipi> ( XNU, i mean ) [09:55:53] <Vratha> OS X isn't slow [09:56:07] <Gr|ffous> it's threading is meant to be [09:56:10] *** zaurus has joined #opensolaris [09:56:14] <Vratha> oh, i see.. well the WWDC keynote shows OS X slamming windows with WoW using new parallel APIs :) [09:56:30] <zaurus> I create a zone named "myzone" [09:56:36] <Cyrille> congratulations [09:56:37] <zaurus> then when running ps -ef [09:56:39] <Gr|ffous> you go zaurus! [09:56:44] <zaurus> it hangs [09:56:45] <zaurus> why [09:56:48] <zaurus> I mean [09:56:54] <zaurus> ps -ef from the global zone [09:57:05] <zaurus> it displays some lines and hangs [09:57:19] <zaurus> and prior to creating the zone , ps -ef was ok [09:57:22] <zaurus> any ideas? [09:57:27] <Gr|ffous> that's a neat trick... I seem to recall doing something similar under linux, but I don't recall what the cause or fix was sorry [09:57:38] <e^ipi> last i checked OSX's malloc() was about 3x slower than solaris [09:57:43] <e^ipi> fork was about twice as slow [09:57:58] <Gr|ffous> zaurus, is the zone initialise and running? [09:58:09] <zaurus> yep [09:58:28] <Vratha> e^ipi: well, luckily most modern designs won't use those horrendously :) [09:58:33] <zaurus> zoneadm list -cv [09:58:36] <zaurus> shows: 1 myzone running /export/home/myzone native [09:58:36] <Gr|ffous> zaurus, are you using nfs? [09:58:41] <Cyrille> you could try and truss the ps to find out on what operation in particular it's hanging or what resource it's waiting for. [09:58:54] <Vratha> compilers and their dynamic on-stack arrays pretty much get rid of most malloc() needs [09:58:57] <zaurus> yes, but not from the zone config [09:59:01] <Vratha> s/arrays/memory/ [09:59:20] <zaurus> /export/home is mounted as nfs from the non global zone [09:59:27] <zaurus> could this cause problems? [09:59:58] <Gr|ffous> well it shouldn't but unexplained hangs on an otherwise working system often trace back to an nfs issue, I've found [10:00:09] <Gr|ffous> might be somewhere to start? truss is a good idea too [10:00:20] <zaurus> I will try stopping the nfs [10:00:25] <LeftWing> Why would you NFS-mount /export/home from the global zone? [10:00:50] <Cyrille> he said from the non global zone. [10:00:55] <Cyrille> but that's even more confusing. [10:00:56] <zaurus> I don't have direct access to that machine throgh a terminal [10:01:02] <LeftWing> ... that's what I meant, the non-global. =P [10:01:17] <zaurus> i WAS WRONG [10:01:18] <Gr|ffous> welcome LeftWing :) [10:01:25] <zaurus> Sorry from the GLOBAL zone [10:01:27] <LeftWing> Greetings, Gr|ffous. [10:01:34] <Cyrille> how about we actually type what we mean, I find that this usually helps with communication accuracy ;-) [10:01:42] * Gr|ffous grins [10:01:55] <LeftWing> Cyrille: Why don't you tell me what you really think. ;) [10:02:10] <Cyrille> not printable in this family channel ;-) [10:02:15] <LeftWing> haha [10:02:16] <LeftWing> Oh noes. [10:02:30] <Vratha> f**|< was probably somewhere in there [10:02:33] <Vratha> just a guess [10:02:43] * LeftWing runs away to move some parts into an Ultra 5. [10:03:12] *** karrotx has joined #opensolaris [10:04:46] *** bengtf has quit IRC [10:06:55] <Gr|ffous> linking questions for the more knowledgable in here; I am trying to compile amarok, and I want it to link against the /usr/lib/ version of libz. amarok, requires taglib, which in turn in linked to the /opt/csw/lib version of libz. Where does this leave amarok when it's trying to find libz. Must it use /opt/csw/lib/libz because one of the apps that it's linking against is using it? [10:08:36] <tsoome> mm is amarok using tablib directly? [10:08:49] <tsoome> taglib, that is.... [10:09:24] <tsoome> ah stupid question [10:09:34] <tsoome> too little of coffe.... [10:09:58] <tsoome> you may still use whatever libz you like [10:10:05] *** alanc_away has quit IRC [10:10:06] *** alanc-away has quit IRC [10:10:27] *** alanc-away has joined #opensolaris [10:10:31] *** alanc_away has joined #opensolaris [10:10:32] <tsoome> but if taglib already links /opt/csw version, You may consider to use it as well to share the same code [10:11:13] <Gr|ffous> hmm [10:11:28] <Gr|ffous> ok, so there is no dependancy 'inheritance' as such? [10:11:43] <tsoome> ? [10:13:36] *** bengtf_ has quit IRC [10:13:47] <Gr|ffous> just confirming in my own words. Even though the taglib bin links to one library, another app linking with taglib's headers can still use a different lib, _of the same name_... IE amarok has taglib as a 'child', but it doesn't inherit the reference to libz that taglib has [10:14:20] <tsoome> nope [10:14:42] <Gr|ffous> sorry, my terminologies are very lacking. It's all very new still for me [10:15:10] <tsoome> you got the linker and libraries guide from docs.sun.com? [10:15:20] <tsoome> it's very good reading [10:15:51] <raph_ael> hello [10:16:27] <Gr|ffous> I've essentially traced my problem with amarok down to taglib. I have the csw version of taglib which uses the csw one funnily enough, and one I made myself, which grabbed the /usr/lib ver. If I point amarok to the /usr/lib linked version of taglib, it compiles fine and links to /usr/lib/libz correctly. If I point it at the csw taglib, I get libz.so.1 (SUNW_1.1) => (version not found) [10:16:36] <Gr|ffous> good reading if you understand it all! ;) [10:16:45] <Gr|ffous> I'm trying to work my way through it [10:16:53] <Gr|ffous> hi raph_ael [10:17:10] *** bunker_work has joined #opensolaris [10:17:11] <raph_ael> :) [10:18:44] <Cyrille> if you point it to the opt/csw taglib at link time, you need to add a -R to specify the run time lookup path so that it grabs libz from csw. [10:18:45] <Gr|ffous> tsoome, if I then use LD_LIBRARY_PATH, to allow the 'bad' amarok bin to see /usr/lib, it's happy again... [10:19:16] *** karrotx_ has joined #opensolaris [10:19:27] <Cyrille> you need to be consistent between how you resolve symbols at link time and how you load them at run time. [10:19:50] <Gr|ffous> like this? http://rafb.net/p/mD4ocM60.html [10:19:50] *** karrotx has quit IRC [10:20:37] <Gr|ffous> actually that is a bad example, the -L is in the wrong order. I have done it in the right order with the same result [10:21:10] <Cyrille> well if you put both -L/-R /opt/csw/lib and /usr/lib, it's anybody (who's read the linker's docs)'s guess as to which'll get picked... [10:21:21] <Gr|ffous> but I am doing it right though? -R for <when you start, look here for your libs> and -L <when compiling/linker, find yer libs here>? [10:21:38] <Cyrille> yup [10:21:39] <Gr|ffous> is that not a search order? [10:21:49] <Cyrille> it is [10:22:16] <Cyrille> it's just that I couldn't tell you offhand what the resulting order is (i.e first in line first looked up or the other way around) [10:22:23] <Gr|ffous> oh.. [10:22:56] <Gr|ffous> ok, so if I make no mention of /usr/lib, should I it (it didn't), or does that still fail because /usr/lib is in the system's search path? [10:22:57] <Cyrille> but your runpath from the dump is in the left to right lookup order (that I know) [10:23:06] <Gr|ffous> *should it work [10:23:30] *** dunc_ has joined #opensolaris [10:23:52] <Cyrille> I'd say if you link with only opt/csw and put only opt/csw in your runpath that should work. [10:24:13] <Cyrille> but I could be wrong and I don't know what exactly is in the opt/csw/libz. [10:24:27] <Cyrille> and yes /usr/lib is in the default search path. [10:24:32] *** dunc_ has quit IRC [10:26:50] <Gr|ffous> ooh! I think I found the problem!! ldd -s /path/to/amarokbin: http://rafb.net/p/6J4ETC83.html [10:27:24] <Gr|ffous> It seems that it's libpng that is causing the issue. [10:27:41] <Gr|ffous> so, is this a second level dependancy? [10:28:02] <Gr|ffous> amarok links against libpng, which in turn links against libz (but is broken) [10:28:55] *** MattMan has joined #opensolaris [10:29:21] <Gr|ffous> yeah, looking through the full ldd -s output, there are two version not found errors, and both are libpng libs looking for libz [10:29:28] <Gr|ffous> am I on the right track? [10:31:31] <Cyrille> Gr|ffous, it seems the issue is that these libs want a versioned library when you only have a non-versioned shared object in your run path. [10:31:40] <Cyrille> man ln should help heaps ;-) [10:32:12] <Cyrille> if you only have libz.so around and these are desperate for libz.so.1, just give them one. [10:33:31] *** bengtf_ has joined #opensolaris [10:33:32] *** bengtf_ is now known as bengtf [10:33:52] <Gr|ffous> oh there is a libz.so.1, but it's not in the RPATH [10:34:01] <Cyrille> also libpng from that last ldd seems to have a weirdo runpath which doesn't include /opt/csw/lib (which is where your libz is, presumably) [10:34:13] <Gr|ffous> yeah, that's what I'm getting at [10:34:29] <Gr|ffous> what is $ISALIST exactly, my shell doesn't have it defined... should it? [10:34:30] <Vanuatoo> Is solaris express 1/07 going to be released? [10:35:00] <trygvis> if so they better hurry up :) [10:35:12] <Cyrille> if you mean "ever" I think there's a good chance of that, though maybe not with the same name ;-) [10:35:36] <Vanuatoo> For instance 2/07? [10:36:42] <Cyrille> honestly, I have no idea (because I don't know anything about that). [10:37:03] <Vanuatoo> Can anyone from sun confirm or deny that> [10:37:29] <Vanuatoo> I'm seriously thinking using sxcr builds [10:37:42] <Vanuatoo> The only thing that worries me is how stable they are [10:41:01] *** peteh has joined #opensolaris [10:45:05] *** nostoi has joined #opensolaris [10:47:26] <zaurus> I tried zlogin to a zone on a machine which does not run nfs and still ps -ef hangs [10:47:37] <zaurus> but this happens ONLY if I run zlogin [10:47:48] <zaurus> without zlogin, ps -ef is OK [10:47:51] <zaurus> any ideas? [10:56:07] <tomww> zaurus: first-time zlogin -C zonename was done to initialize the zone? [10:57:36] *** calumb has joined #opensolaris [10:59:15] <cmihai> Vanuatoo: SXCR builds are fine for home use, testing and development. If you don't need the support for SX, go SXCR. Besides, if you're usign LiveUpgrades you'll be fine. If a new build is "unstable" go back to your old enviroment :) [11:01:00] <Vanuatoo> I'm developing java apps [11:01:17] *** uebayasi has quit IRC [11:01:24] <Vanuatoo> is anyone taking care of opensolaris forums? [11:01:27] <Vanuatoo> http://www.opensolaris.org/jive/forum.jspa?forumID=39 [11:01:35] <Vanuatoo> Some exceptions are displayed [11:01:51] <Vanuatoo> yesterday and the day before yesterday this topic was empty [11:02:52] <cmihai> Vanuatoo: well, if you're looking for a stable development platform... or a devel platform.. well, it's really up to you. But SXCR 55b is pretty nice. Comes with devel tools also (NetBeans, SUNWpro, etc). Quite nice. [11:03:31] <Vanuatoo> Actually the only thing I'm worried not to loose my files :) [11:03:38] *** calumb is now known as calMTG [11:03:55] <Vanuatoo> I already got used to the fact that everything is slow and memory hog :) [11:03:58] *** halton has left #opensolaris [11:04:09] *** yippi has joined #opensolaris [11:04:12] <cmihai> Can't see how that could happen without hardware failure or PEBKAC involved :) [11:04:22] <cmihai> Besides, you should have regular backups in the first place. [11:04:24] <Vanuatoo> I love that word [11:04:33] <cmihai> Hell, at least of your work... [11:04:36] <Vanuatoo> I do everyday backups [11:04:47] <Vanuatoo> of my netbeans projects [11:05:02] <cmihai> Then you're good to go :) [11:05:45] <Vanuatoo> Now I'm upgrading to every sx release, is it possible with sxcr stuff or it's recommended to install from the scratch every time [11:06:04] <cmihai> No, why would you do that? [11:06:07] <cmihai> Use LiveUpgrades. [11:06:13] <cmihai> Easiest way to do it. Safest too. [11:06:28] <Vanuatoo> I'm lazy person to read the manual :) [11:06:33] <cmihai> If something goes wrong with the upgrade, go back to the older version. [11:06:36] <Vanuatoo> I just put the dvd and tell it to upgrade [11:06:45] <cmihai> I don't even burn the DVD. [11:06:59] <cmihai> I mount the iso, issue "luupgrade" and reboot. As easy as 123 [11:07:00] <Cyrille> you just slightly warm it? [11:07:13] <Vanuatoo> cmihai: wow cool :) [11:07:32] <Vanuatoo> mount iso with lofiadm? [11:08:14] <cmihai> Yup [11:08:47] <Vanuatoo> I'm definitely going to use sxcr builds [11:08:51] <tsoome> damit, myx / ufs has gone bad.... so mutch about logging..... [11:09:58] <cmihai> LiveUpgrade is pretty much lofiadm -a cd.iso, mount -F hsfs /dev/lofi/1 /dvd "lucreate -c "sol1" -m /:/dev/dsk/YOURDISKHERE:ufs -n "sol2"", "luupgrade -u -n sol2 -s /dvd; luactivate sol2; init 6 :) [11:10:15] <cmihai> tsoome: that's odd. You sure it's fs corruption and not hardware errors? [11:10:47] <Vanuatoo> is init 6 the same as shutdown -i6 -g0 -y? [11:10:58] <tsoome> adaptec scsi + scsi hdd [11:11:17] <tsoome> but I think it the result of some power outages.... [11:11:18] <cmihai> Adaptec sucks, but that isn't the issue anyway (I guess). [11:11:37] <cmihai> Vanuatoo: nah, always use init 6 with luupgrades, or it won't upgrade your boot archive [11:11:39] <tsoome> well, it's sun W2100z workstation;) [11:11:44] <cmihai> Heh [11:11:48] *** lplatypus has joined #opensolaris [11:12:23] <tsoome> looks like some dir(s) got lost and there is a some count of unref'd files... [11:12:36] <Vanuatoo> cmihai: YOURDISKHERE placeholder stands for something like c0d0so? [11:12:49] <cmihai> It's a placeholder for your altroot [11:13:05] <cmihai> c0d0s3 or whatever [11:13:11] <cmihai> format should tell you. [11:13:20] <Vanuatoo> I've got two hdd-s so I can use any of them as alternative root holder, right>? [11:13:38] <cmihai> /dev/dsk/c1d0s0 is in use for live upgrade /. Please see ludelete(1M). [11:13:55] <cmihai> Vanuatoo: well, you can either use 2 disks or use 1 disk with a different partition [11:14:06] <cmihai> Personally I'd rather use the RAID and diff partition approach, but it's really up to you. [11:14:28] <Vanuatoo> how much space is needed approx for developer config [11:14:39] <cmihai> Uh.. dunno, 10GB? [11:14:48] <Vanuatoo> I'm little low on space [11:14:59] <cmihai> Bah, how low are you? [11:15:14] <cmihai> Just make those roots and altroots at least 15GB and you're good to go. [11:15:23] <cmihai> /dev/dsk/c1d0s3 21G 8.0G 13G 39% / [11:15:34] <Vanuatoo> I'm got 6.6gb free [11:15:53] <cmihai> Get a new disk. [11:15:56] <Vanuatoo> :) [11:16:07] <cmihai> At like 1$/GB... [11:16:10] <Vanuatoo> I'm going to upgrade to conroe config in about a month [11:16:31] <Vanuatoo> and will have 150gb wd raptor drive [11:17:16] <Vanuatoo> Little offtopic: Workstations come with ECC SDRAM memory. Is it possible to turn off ECC? [11:18:43] *** bunker_work has quit IRC [11:18:52] <tsoome> why do you want to turn it off? [11:19:33] <e^ipi> maybe he wants uncorrected errors [11:19:42] <e^ipi> testing? [11:19:49] <Vanuatoo> I want faster memory :) [11:20:00] <tsoome> ? [11:20:05] <tsoome> buy a better machine [11:20:08] <e^ipi> yay for micro-optimizations [11:20:15] <e^ipi> just overclock the machine [11:20:26] <Vanuatoo> but it will not speed up RAM [11:20:37] <Vanuatoo> Workstations come with DDR2 667 ECC RAM [11:21:00] <Vanuatoo> I guess it's a little bit slower than non ECC DDR2 800 [11:21:01] <tsoome> write a better programs [11:21:06] <tsoome> you will gain more.... [11:21:10] <Vanuatoo> :) [11:21:19] <andersmo> Vanuatoo: What's the performance cost of ECC memory? Got any hard numbers? [11:21:31] <Vanuatoo> That are the words I say in my mind to solaris developers [11:21:48] <Vanuatoo> andersmo: I read some article about it [11:21:59] <Vanuatoo> and that was about 10-15% slower [11:22:00] <tsoome> like? [11:22:01] <e^ipi> how delightfully vague [11:22:12] <e^ipi> "i read it... in some stuff" [11:22:19] <e^ipi> "you should check it out" [11:22:33] <Vanuatoo> e^ipi: Do you want link? [11:23:22] <e^ipi> i believe you that it's marginally slower [11:23:29] <tsoome> Vanuatoo: so, silent bad memory and corrupted data is preferred? [11:23:39] <Vanuatoo> ECC is doing error checking and correction. I'm not doing rocket science and from my experience non ECC memory is doing fine [11:23:48] <Vanuatoo> If it's not spoiled in the first place [11:24:21] <e^ipi> computers are fast enough these days that you''d be hard pressed to notice a difference, particularly on a workstation [11:24:32] <e^ipi> you can bench out a difference, but _notice_, i doubt it [11:24:46] <Vanuatoo> e^ipi: sounds promising [11:24:57] <e^ipi> i can't tell the difference between a 1.5ghz and a 2ghz machine [11:25:18] <Vanuatoo> Realy don't understand why workstaions are so overpriced [11:25:37] <Vanuatoo> For example If I buy top ultra20 it will cost about 2.8 [11:25:38] <andersmo> you can probably notice it if you count in the placebo effect, though. Hey, for a decent fee I'll happily "turn off" ECC on your memory, placebo style. ;) [11:25:51] <Vanuatoo> andersmo: lol [11:25:58] <Vanuatoo> It's proven by benchmarks [11:26:07] <Vanuatoo> so not only placebo takes place [11:26:18] <tsoome> you can prove *anything* with benchmarks [11:26:23] <Vanuatoo> tsoome: lol [11:26:24] <andersmo> Placebo is a big, big market - just look at hi fi audio... =) [11:26:40] <tsoome> you can prove your laptop is faster than SF25k [11:26:41] <e^ipi> again, there's a difference between marginal benchmark differences, and actually being able to _NOTICE_ a difference [11:26:53] <tsoome> but it does not make SF25k useless;) [11:26:58] <e^ipi> tsoome: single-threaded CPU-bound task? [11:27:04] <tsoome> for instance [11:27:05] <Vanuatoo> so you're saying it does not matter if you buy ddr2 667,800,or 1066 memory? [11:27:19] <e^ipi> if you want a killer gaming rig, build one out of parts [11:27:29] <e^ipi> or buy dell's " i have more money than sense" brand [11:27:40] <e^ipi> xpr or xor or something [11:28:22] <Vanuatoo> I've built custom config that will outperform (I bet 100%) by 2 times ultra20 in every benchmark :D [11:28:36] <Vanuatoo> and it will cost me 1900$ [11:28:43] <e^ipi> so do it [11:28:51] <Vanuatoo> e^ipi: I'm already :) [11:28:52] *** damienc has joined #opensolaris [11:29:20] <Vanuatoo> And I don't understand the policy to charge 250$ for seagate 80gb hdd [11:29:32] <Vanuatoo> and 900$ for 19TFT monitor [11:29:35] *** triplah_ has quit IRC [11:29:52] <Vanuatoo> and what a beautiful phrase sun workstation for the price of P4 [11:29:59] <Vanuatoo> they are laughing at us [11:30:20] <e^ipi> "they" ? [11:30:27] <Vanuatoo> Sun marketing :) [11:30:44] <Vanuatoo> They want me to smoke what they smoke [11:30:49] <tsoome> you don't *have to* buy a monitor from sun;) [11:31:00] <Vanuatoo> and what about 80gb hdd? [11:31:01] <e^ipi> for that matter, you don't have to buy a machine from sun [11:31:04] <Vanuatoo> for 250$ [11:31:13] <e^ipi> solaris runs on commodity junk these days [11:31:29] <Vanuatoo> and what about opteron 1220SE that costs 450 on the market and 1100 from sun? [11:31:56] <e^ipi> if what sun sells doesn't work for you, don't buy it [11:31:58] <Vanuatoo> What is the point in workstations? support? [11:32:09] <e^ipi> certified platform, my guess [11:32:48] <Vanuatoo> It's like sony. +10% because it's sony [11:32:59] <Vanuatoo> sony effect :) [11:33:14] <e^ipi> go price out a POWER5 workstation [11:33:22] <e^ipi> or an HPPA [11:33:27] <Vanuatoo> I'm not interested in them [11:33:36] <Vanuatoo> I'm talking about commodity hard [11:33:46] <e^ipi> every UNIX vendor sells expensive kit [11:33:50] <Vanuatoo> And sun is not using any unique parts in his workstaion [11:33:55] <Vanuatoo> Just the case I guess [11:33:58] <Vanuatoo> Mother is Tayan [11:34:01] <Vanuatoo> Tyan [11:34:05] <Vanuatoo> CPU is AMD [11:34:10] <Vanuatoo> HDD is hitachi [11:34:13] <Vanuatoo> or something [11:34:33] <Vanuatoo> So I guess one pays 2800 for the name sun and support [11:35:02] <e^ipi> *shrug* and fair enough [11:40:58] <e^ipi> neat, $20 vax on teh ebayz [11:48:55] *** MaGre has joined #opensolaris [11:48:58] *** freakazoid0223 has quit IRC [11:49:10] *** Tpenta has quit IRC [11:54:31] *** Peanut has quit IRC [11:56:59] *** calMTG is now known as calumb [11:58:32] *** coffman has joined #opensolaris [11:59:47] <coffman> morning [12:01:25] *** yippi has quit IRC [12:08:45] *** mlh has quit IRC [12:10:42] *** _357181_ has left #opensolaris [12:10:42] *** _357181_ has joined #opensolaris [12:18:28] *** lplatypus has quit IRC [12:26:33] *** kloczek has quit IRC [12:27:38] *** kloczek has joined #opensolaris [12:27:58] *** mlh has joined #opensolaris [12:28:05] *** whaq has joined #opensolaris [12:31:30] *** DivineEntity has quit IRC [12:34:39] *** mlh has quit IRC [12:36:54] *** mlh has joined #opensolaris [12:54:43] *** svoboda_ has joined #opensolaris [12:55:24] *** mlh has quit IRC [12:55:43] *** mlh has joined #opensolaris [13:10:23] *** bunker_work has joined #opensolaris [13:15:35] *** andyshack has quit IRC [13:23:29] <tsoome> damit, multiple fsck's and still ufdump is complaining that directory at inode 27 vanished .... [13:24:00] <Vanuatoo> tsoome: can you confirm that www.nvidia.com gives some strange stuff? [13:24:13] <tsoome> ? [13:24:22] <tsoome> like broken web? [13:24:33] <Vanuatoo> some page is displayed [13:24:46] <Vanuatoo> but it's almost empty [13:25:03] <Vanuatoo> Som rss stuff [13:25:09] <tsoome> it redirects to page/home.html [13:25:25] <Vanuatoo> exactly [13:25:30] <tsoome> and it's empty - an white and grey [13:25:37] <Vanuatoo> yes [13:25:53] <Vanuatoo> I can't beleive. It;s like this already for 5-6 hours [13:26:07] <Vanuatoo> Their admin must have been dead [13:29:24] *** nostoi has quit IRC [13:39:12] <Cyrille> or asleep [13:40:53] *** bzcrib has quit IRC [13:41:38] *** merlin_ has joined #opensolaris [13:51:52] <coffman> nice one :) [13:53:41] *** Mazon is now known as mazon [13:54:04] *** Vratha has quit IRC [13:54:19] *** yongsun__ has quit IRC [13:59:53] *** Odin-LAP has joined #opensolaris [14:03:38] *** merlin_ has quit IRC [14:10:17] *** nostoi has joined #opensolaris [14:11:46] *** bzcrib has joined #opensolaris [14:13:33] *** nachox has joined #opensolaris [14:13:38] *** _syphilis_ has joined #opensolaris [14:14:23] <nachox> morning [14:15:36] <dojtoll> Hello [14:15:53] *** Tpenta has joined #opensolaris [14:16:14] *** MikeTLive has left #opensolaris [14:18:01] <Berny_> bugger [14:18:25] <Berny_> i hate hard disks [14:20:04] <Berny_> why the heck does a fc/al attached disk just disappear? [14:21:24] <cap_> storage sucks, we should give it up [14:21:24] <Stric> had to go to the toilet? [14:21:53] <Berny_> yeah should be back by now... [14:24:21] *** nbkk6fo_ has joined #opensolaris [14:24:22] *** rodrick-brown has quit IRC [14:37:46] *** Peanut has joined #opensolaris [14:41:19] *** DMJC-L has joined #opensolaris [14:41:21] *** dunc_ has joined #opensolaris [14:41:32] <DMJC-L> what are the pros/cons of opensolaris vs solaris? [14:41:41] <DMJC-L> I have never installed either [14:41:50] <tsoome> solaris is stable [14:41:55] <cmihai> Meh [14:42:10] <DMJC-L> so solaris 10 with zfs is stable? [14:42:15] <lasseoe> of course [14:42:18] *** calumb is now known as calAFK [14:42:23] <DMJC-L> but the desktop environment is older? [14:42:24] <lasseoe> This ain't not Linux [14:42:25] <tsoome> opensolaris is development snapshot of next solaris [14:42:39] <DMJC-L> hehehe lasseoe that's what I'm hopign for [14:42:48] <DMJC-L> also, what's raid 5 like under solaris? [14:42:51] <Cyrille> solaris can be supported [14:42:51] <lasseoe> DMJC-L: that's a good start :) [14:42:59] <DMJC-L> I recently had a raid 5 die on it's ass under linux [14:43:05] <nachox> recent nevadas have a new gnome, solaris still has an old gnome [14:43:06] <DMJC-L> and it was unrecoverable [14:43:17] <DMJC-L> what about kde? [14:43:22] <coffman> DMJC-L: raid 5 is for the kids [14:43:23] <tsoome> install it and use [14:43:28] <PerterB> raid5 under SVM and raidz under zfs are both pretty reliable [14:43:32] <tsoome> it's up to you [14:43:37] <coffman> DMJC-L: we got zraid [14:43:44] <DMJC-L> what is zraid? [14:43:49] <DMJC-L> got info about it? [14:43:50] *** bzcrib has quit IRC [14:44:14] <lasseoe> RAIDZ [14:44:30] <DMJC-L> ok [14:44:35] <lasseoe> it's ZFS's way of doing RAID-5, so to speak [14:45:22] <PerterB> http://blogs.sun.com/bonwick/entry/raid_z [14:45:30] <DMJC-L> is wikipedia correct when it says adding new drives is a pain? [14:45:55] <tsoome> to raid5? [14:46:32] <DMJC-L> raidz [14:46:49] <DMJC-L> http://en.wikipedia.org/wiki/ZFS#Limitations [14:46:56] *** bzcrib has joined #opensolaris [14:47:05] <tsoome> usual way to expand a raid5 is to create new parity group - this means the minimum disk count to be added is 3 [14:48:19] <DMJC-L> I've got 3x320gb hdds which will be my initial install [14:48:27] <tsoome> this wikipedia sounds like BS [14:48:39] <DMJC-L> possibly is [14:48:55] <tsoome> xisting slices, whether they be mirror, raidz, or raidz2 cannot be grown by adding disks. [14:49:01] <tsoome> but you can [14:49:26] <andersmo> I think you actually can extend a mirror? [14:49:35] <andersmo> (from two to three-way) [14:49:44] <tsoome> ? [14:50:01] <tsoome> no, you can expand the storage [14:50:14] <tsoome> add next pair of mirrored disks [14:50:27] <tsoome> and all io will be spread over all disks [14:52:06] <coffman> zfs does auto stripe over it [14:53:14] <tsoome> someone from sun zfs groups should put real information to this wikipedia:P [14:53:23] <DMJC-L> hmm [14:53:38] <DMJC-L> basically I'd want a raid 5 style setup using 3 disks of 320gb each [14:53:47] <DMJC-L> and I'd be looking at later adding another 2 320gb hdds [14:53:49] <tsoome> so do it? [14:53:57] <tsoome> another 2? [14:54:00] <DMJC-L> just wanting to know if it'd work [14:54:02] <DMJC-L> yeah [14:54:11] <andersmo> you'll have to add them as a mirror, but yeah, it'd work. =) [14:54:17] <andersmo> (or as single disks) [14:54:18] <DMJC-L> as a mirror? [14:54:21] <tsoome> sure [14:54:28] <tsoome> r5 need minimum 3 disks [14:54:38] <DMJC-L> it'd have 3 initially [14:54:43] <tsoome> and it's not reasonable to rebuild an entire r5 [14:54:52] *** nostoi has quit IRC [14:55:06] <tsoome> you do know what the R5 is? [14:55:23] <DMJC-L> afaik raid 5 is usually 2-3 disks with an extra disk as a parity drive [14:55:26] <cmihai> In other words you can't add disk in a dynamic way to RAIDZ [14:55:27] <tsoome> how the data and parity is layd on disks etc [14:55:34] <cmihai> You can add the to the pool though [14:55:52] <andersmo> extending a raid5/raidz would require a complete honkin' rebuild of the whole raid5/z set. Doing that in a manner that guarantees survival even if the system crashes halfway through the restructuring is... a lot of work. =) [14:58:08] <DMJC-L> hmm [14:59:30] <tsoome> only practical way to expand R5 is to add next similar set of disks - if you have started with 2D+1P R5, then expand it with set of 3 disks [14:59:57] <tsoome> thais will keep same perfomance and availability to whole pool [15:01:12] <tsoome> even if you can add mirror or single disk, there is a little point of doing this, since you will get only a mess, where some data is protected by R5, some by mirror and some data is not protected at all [15:01:27] <tsoome> and this is obviously not good [15:01:48] <DMJC-L> so there's no way to have zfs rewrite all 5 disks on the fly to redistribute the parity data? [15:02:10] <andersmo> nope, sorry. [15:02:16] <andersmo> At least not at the present time. [15:02:31] <Berny_> get 5 new disks, do a zfs send/receive, send the old (now spare disks) back [15:02:33] <Berny_> :-) [15:02:34] <andersmo> If you're a file system wizard I'm sure you'd be welcome to implement it. It'd be very cool to have. =) [15:02:58] <DMJC-L> doh [15:02:59] <DMJC-L> heh [15:03:21] <DMJC-L> although by the time I can afford another 2 320gb hdds 400gb hdds could be cheaper heh [15:03:33] <tsoome> and then there is a question, how big R5 set you want to have? [15:04:01] <andersmo> I think veritas volume manager can do it... but I'm not really sure. Anyhow, I think you can get a few 320GB disks for the price of a vxvm license. =) [15:04:08] <DMJC-L> upto 7hdds [15:04:29] <DMJC-L> 3 pata 4 sata [15:04:36] <tsoome> do you really want to have say, 20 data disks protected by single parity? [15:05:01] <DMJC-L> it wouldn't be 20 [15:05:04] <DMJC-L> it's for home use [15:05:08] <DMJC-L> not for a data center [15:05:49] <tsoome> hm, but why not start with 5 disks right now? [15:06:22] <DMJC-L> I'm broke, and unemployed atm? [15:06:30] <tsoome> I see:) [15:06:35] <tsoome> it happens.... [15:06:43] <DMJC-L> (poor student syndrome) [15:07:20] <Berny_> being broke happens to an awful lot of married guys with pretty decent jobs too... [15:07:37] <DMJC-L> heh yeah [15:07:44] <tsoome> .oO I can't afford 320GB hdd's :P [15:07:55] <tsoome> but I'm only a poor skydiver;) [15:08:02] <DMJC-L> lol [15:08:09] * Berny_ just got 8 73gb pata drives :-) [15:08:19] <tsoome> good gor you;) [15:08:28] <andersmo> tsoome: Now skydiving is a money sink, yeah... =) [15:08:33] <Berny_> 5years old, and about to die :-) [15:08:37] <andersmo> Fun, though. =) [15:08:57] <Berny_> plus a broken raid controller, that only works as scsi id 0 [15:09:11] <Berny_> and you cannot power the bugger down and restart [15:09:28] <Berny_> gotta wait a few days until it forgets the data from the nvram or so [15:12:01] <DMJC-L> is OSS the only sound system on solaris? [15:12:10] <_syphilis_> no, solaris has a native audio api [15:12:32] <DMJC-L> list of supported hardware? [15:13:24] <Berny_> http://www.sun.com/bigadmin/hcl/ [15:14:28] <DMJC-L> thanks [15:16:07] *** Pazza has joined #opensolaris [15:16:56] <jbk> morning [15:24:27] *** nbkk6fo_ has quit IRC [15:27:30] *** Odin-LAP has quit IRC [15:29:09] <jbk> strange... [15:29:40] <jbk> i sent off 16 explorers last night, and the only response i got was for one of them (which was one of the last ones to be sent) [15:33:34] *** dunc_ has quit IRC [15:34:01] *** bondolo has quit IRC [15:35:32] *** bank__ has joined #opensolaris [15:35:39] <bank__> hi XD [15:37:48] *** postwait_ has quit IRC [15:38:40] *** Pazza has quit IRC [15:39:11] *** zaurus has quit IRC [15:45:05] *** Cyrille has quit IRC [15:46:21] *** logic has quit IRC [15:46:24] *** logic_ has joined #opensolaris [15:46:46] *** tsoome has quit IRC [15:54:23] *** calAFK is now known as calumb [15:54:58] *** MaGre has quit IRC [15:55:49] *** _tsoome_ has joined #opensolaris [15:56:27] <Vanuatoo> Is it hard to create new locale for solaris? [15:56:45] <Vanuatoo> I just want to type Georgian (ka_GE) characters [15:57:46] *** MaGre has joined #opensolaris [15:58:23] *** Cyrille has joined #opensolaris [15:58:41] *** bzcrib has quit IRC [16:00:57] *** calumb is now known as calAFK [16:01:51] *** laca has joined #opensolaris [16:05:15] *** mrdeviant has joined #opensolaris [16:17:25] *** coffman has quit IRC [16:29:13] *** calAFK is now known as calumb [16:30:57] *** MaGre has quit IRC [16:31:00] *** zaurus has joined #opensolaris [16:31:58] <zaurus> How I know the cmdline of a process from /proc/pid ? (what in linux is done by cat /proc/pid/cmdline)? [16:32:19] <trygvis> pargs [16:33:15] <zaurus> ok , "pargs pid" [16:33:31] <zaurus> but is there aside from it also something under /proc? [16:34:16] <lasseoe> no [16:34:29] <jteo> proc does not spit out text. [16:34:33] <mrdeviant> zaurus, the /proc filesystem on solaris is very different from /proc on linux. on solaris, /proc is a group of binary files that the ptools (see proc(1)) operate on [16:34:39] * quasi *mumbles* solaris is not linux [16:34:52] <lasseoe> SNL [16:36:05] <jteo> quasi: shocking. aren't all Linuxes the same? <sarcasm/> [16:36:37] *** benluo has joined #opensolaris [16:36:40] <benluo> hi [16:36:41] <lasseoe> haha [16:36:58] <lasseoe> Solaris; the different Linux [16:37:31] <quasi> jteo: an awful lot of people turning up here lately seems to think that [16:37:34] <elektronkind> zaurus: /proc on solaris is not something that's populated by files of random type and format as they are on linux [16:38:33] <benluo> can i install solaris 10 64bit version in AMD turion 64 x2 notebook [16:38:45] <elektronkind> zaurus: Like Linux, /proc on Solaris is a user-land interface to process-related stuff in the kernel, but it's designed to be programmed against instead of assuming only cat(1) will do the job. [16:39:24] <elektronkind> zaurus: hence the existence of the "p-commands" such as pargs, pfiles, pcred, and so on [16:39:43] <bank__> hi elektronkind [16:39:51] <elektronkind> heya bank__ [16:40:02] <bank__> :) [16:40:52] <_tsoome_> I really "like" the linux approach. nice text formaated content and not easy to read not parse with programs.... really cool feature. [16:41:12] <_tsoome_> :P [16:41:39] <mrdeviant> it would be nice if /proc/cpuinfo and /proc/meminfo were even remotely similar on different platforms [16:41:48] <elektronkind> it wouldn't be so bad if everything in Linux's procfs was in a XML format at least [16:41:55] <mrdeviant> but that assumes that linux dev's value consistency.... [16:41:58] <_tsoome_> jesus! [16:42:03] <_tsoome_> xml format:P [16:42:06] <_tsoome_> :) [16:42:11] <elektronkind> it's better than nothing [16:42:24] <elektronkind> right now it's whatever the linux dev felt like doing on that morning [16:42:34] <jteo> it's OSS! Fix it yerself! [16:42:37] <jteo> [sic] [16:42:38] <elektronkind> heh [16:43:07] <elektronkind> I'll fix it by booting kernel/unix, k thx [16:43:47] <Marv|LG> most linux /proc can be parsed with awk [16:43:56] *** RaD|Tz has joined #opensolaris [16:44:20] *** Odin-LAP has joined #opensolaris [16:44:37] <elektronkind> I remember my first time dealing with LVM in Linux for some mirrors I was trying to set up, and looking for the Linux equivalent of the metastat command [16:44:59] <elektronkind> spent hours [16:45:18] <elektronkind> but then discovered that the "command" I was looking for was "cat /proc/mdstat" [16:45:28] <elektronkind> wtf [16:45:46] *** gustav3d has joined #opensolaris [16:46:14] <cap_> elektronkind, fwiw, mdstat has nothing to do with lvm [16:46:54] <elektronkind> well that just goes to show you how much I pay attention to that OS I guess [16:47:28] <elektronkind> either way, it's annoyingly bass ackwards in highly visible places [16:51:24] *** mazon is now known as Mazon [16:53:22] *** LordKing has quit IRC [16:54:37] <bank__> how 're you elektronkind [16:57:07] *** flyinglunatic has joined #opensolaris [16:59:38] *** Mazon is now known as mazon [17:00:05] *** RaD|Tz has quit IRC [17:01:23] <axisys> anyone can help me with this ufsrestore procedure http://rafb.net/p/cg0gHz24.html ? [17:01:32] <axisys> i am not sure what volume it is talking about [17:03:30] *** RaD|Tz has joined #opensolaris [17:06:12] *** Dar is now known as Dar_HOME [17:06:59] <benluo> who install 64bit solaris? [17:07:10] <benluo> where to download it? [17:07:33] <_syphilis_> ben: there is no special 64-bit version, the normal solaris has 64-bit support [17:08:02] <richlowe> 10 and above *only* has 64bit support on sparc, on x86 the installer runs 32bit, but after that it'll do the right thing. [17:08:11] <axisys> i am doing ufsrestore w/ rvf since the dump was level 0.. seems to be working now [17:08:36] <richlowe> uname -a will still show i86pc/i386 etc. [17:08:42] <richlowe> use isainfo -b if you want a quick check. [17:09:38] *** Symmetria has joined #opensolaris [17:12:16] *** nikorc has joined #opensolaris [17:13:12] *** calumb is now known as calMTG [17:14:02] *** bank__ has quit IRC [17:16:48] <benluo> richlowe: thanx [17:23:05] *** LeftWing has quit IRC [17:23:15] *** _tsoome_ has quit IRC [17:23:51] <rodrickbrown> does gzip support largefiles? [17:26:12] *** benluo has quit IRC [17:26:39] <mrdeviant> yes, /usr/bin/gzip is largefile aware [17:26:53] <mrdeviant> a list of largefile-aware utils is in largefile(5) [17:27:11] *** compukid_ has quit IRC [17:27:31] *** bunker_work has quit IRC [17:27:32] *** compukid_ has joined #opensolaris [17:27:39] *** yarihm has joined #OpenSolaris [17:27:45] <rodrickbrown> humm [17:27:46] <rodrickbrown> bash-3.00# nm -D $(which gzip)|egrep '(open|lseek|stat)64$' [17:27:46] <rodrickbrown> [54] | 188104| 0|FUNC |GLOB |0 |UNDEF |fstat64 [17:27:46] <rodrickbrown> [117] | 188200| 0|FUNC |GLOB |0 |UNDEF |lseek64 [17:27:46] <rodrickbrown> [143] | 188140| 0|FUNC |GLOB |0 |UNDEF |lstat64 [17:27:47] <rodrickbrown> [87] | 188116| 0|FUNC |GLOB |0 |UNDEF |open64 [17:27:49] <rodrickbrown> [195] | 188152| 0|FUNC |GLOB |0 |UNDEF |stat64 [17:27:52] <rodrickbrown> why are all those undef? [17:28:38] *** Triskelios has joined #opensolaris [17:28:42] <_syphilis_> because they come from libc [17:28:48] *** karrotx_ has quit IRC [17:29:11] <rodrickbrown> what do you mean? [17:29:16] <rodrickbrown> well ofcourse they came from libc [17:29:23] <_syphilis_> what do you expect them to show as? [17:29:43] <rodrickbrown> nm my head is in the clouds today [17:29:46] <mrdeviant> rodrickbrown, if you do a ldd on /usr/bin/gzip, you'll see what libraries it links against. those libs will satisfy the undefined symbols in the gzip binary [17:29:47] <rodrickbrown> hehe [17:29:50] <richlowe> rodrickbrown: nah, nm is right. [17:29:54] <richlowe> rodrickbrown: just your head that's in the clouds ;) [17:31:59] <jteo> ENOCAFFEINE? (gawd i miss that stuff) [17:37:25] *** qdk has quit IRC [17:38:22] *** nostoi has joined #opensolaris [17:41:28] *** princeXh4 has joined #opensolaris [17:43:52] *** qdk has joined #opensolaris [17:44:00] *** estibi has joined #opensolaris [17:44:23] *** princeXh4 is now known as pseudoXh4 [17:45:27] *** onlinebacon has joined #opensolaris [17:49:12] *** nostoi has quit IRC [17:54:57] *** Theoden-Nexenta has joined #opensolaris [17:55:26] *** Gman_ has joined #opensolaris [17:55:48] *** GmanAFK has quit IRC [18:02:11] *** pseudoXh4 has quit IRC [18:04:23] *** _tsoome_ has joined #opensolaris [18:05:51] *** pseudoXh4 has joined #opensolaris [18:06:23] *** bengtf has quit IRC [18:08:52] *** boyd_ has joined #opensolaris [18:09:03] *** mazon is now known as Mazon [18:09:14] *** Theoden-Nexenta has left #opensolaris [18:09:43] *** jopi has quit IRC [18:10:18] *** jopi has joined #opensolaris [18:12:47] *** wamty has joined #opensolaris [18:12:50] <wamty> I just noticed that vim can do sftp [18:14:40] <wamty> From what I can tell, it doesn't perform locking though [18:24:16] *** boyd has quit IRC [18:25:08] *** onlinebacon has quit IRC [18:26:57] *** steleman has quit IRC [18:27:49] *** webar7 has joined #opensolaris [18:28:04] *** broadcast has quit IRC [18:28:05] *** Tpenta has quit IRC [18:28:05] *** peteh has quit IRC [18:29:21] <webar7> is it true that myspace is a massive WinXP Server farm ?! :-\ would be nice for opensolaris (and sun) if it was behind some popular sites [18:29:33] <webar7> instead of xp :( [18:30:43] <hile_> who knows [18:30:46] <dinolinux> ebay runs on sun stuff, i think [18:30:47] <rodrickbrown> myspace runs Windows 2003 server on hp hardware [18:31:00] <rodrickbrown> webar7 many sites are powered by Sun [18:31:35] <webar7> dinolinux, kind of ... [18:33:49] <calMTG> ebay.ie and ebay.co.uk used to have a nice "Powered by Sun" logo on the front page :) [18:34:00] <wamty> some day there will be a vimshell. I see it :) [18:34:35] <calMTG> ebay.com had "Powered by HP" though :/ [18:35:04] <Auralis> the backend ebay server is a e10k [18:35:33] <rodrickbrown> Auralis not anymore [18:35:40] <hile_> morning auralis [18:35:45] <rodrickbrown> e10k's are not worth the power they consume :-P [18:36:18] <richlowe> Hm. [18:36:27] <Auralis> rodrickbrown: well, there was a report a few months ago about ebays serverpark, the guy from ebay mention the e10k [18:36:55] <rodrickbrown> webar7 do you use nvidia products? [18:37:13] <rodrickbrown> http://www.rodrickbrown.com/docs/nvidia-engineroom_download.wmv [18:37:18] <rodrickbrown> might bring a joy to your eye [18:37:19] <webar7> cool [18:37:27] <hile_> wmv??? [18:37:47] *** _tsoome_ has quit IRC [18:37:49] <rodrickbrown> i'm too lazy to conver it to anything eelse :) [18:37:52] <rodrickbrown> its how I found it [18:38:12] <webar7> fluendo! :) [18:40:17] <webar7> sun makes the biggest contribution to opensource ; to the environment (cool threads); to filesystems and future of network infrastructure (ZFS), and has the best hardware but apple's in the black and sun's in the red ... grrrrr [18:40:26] *** steleman has joined #opensolaris [18:40:55] <calMTG> maybe we should start selling phones... [18:41:05] <webar7> I betcha sun's engineers are smarter but possibly not as well dressed ... [18:41:28] <webar7> I hope iphone tanks :) [18:41:31] <estibi> i have just got free solaris media kit :))) [18:41:34] <calMTG> half the usability folk I know here worked at Apple first :) [18:41:42] <webar7> openmoko.com! [18:41:45] *** wamty has quit IRC [18:41:50] *** damienc has quit IRC [18:41:51] <webar7> heheh [18:42:15] <dinolinux> estibi: nice, i'm still waiting for mine :/ [18:42:37] <webar7> calMTG, i wonder if apple's mythical great leap/lead in usability is declining ... [18:43:02] * Auralis never found appls especialy good at usability [18:43:37] <gdamore> the usability of MacOS circa 1991 was vastly superior to anything Microsoft or UNIX had to offer. [18:43:59] <gdamore> granted, that was before it really qualified as an "OS". (No memory protection, etc.) [18:44:12] *** gdamore has left #opensolaris [18:44:21] <webar7> that's cause it copied Sun :) [18:44:24] *** slowhog has joined #opensolaris [18:44:31] <calMTG> webar7: I guess OS X Leopard will give us some idea soon enough... most of the coolest stuff in that seems to be built on Sun technology though :) [18:44:34] <Auralis> well, amiga was better then macos imo [18:44:54] <webar7> ha [18:44:55] * calMTG still has two Amigas at home... [18:45:07] <webar7> in terms of mass market gadgets that 100$ laptop is neato for kidz but I want a 400$ sun laptop :) [18:58:43] *** webar7 has quit IRC [18:59:08] *** logic_ has quit IRC [18:59:15] *** calMTG has quit IRC [19:01:16] *** MattMan has quit IRC [19:03:15] *** Camus_ has joined #opensolaris [19:03:48] <Camus_> would the opensolaris release have better support for a specific Firewire chipset than the solaris 10 version on sun.com / [19:03:48] <Camus_> ? [19:05:47] *** RaD|Tz has quit IRC [19:06:23] <Camus_> anyone? [19:06:50] <Auralis> it is possible [19:07:11] <comay> Camus_:it might. best to ask on device-discuss at opensolaris dot org. nevada/opensolaris has lots of improvements over s10 but not sure about firewire [19:07:27] <Camus_> hmm [19:07:34] <Camus_> woul du guys konw how to troubleshoot device drivers? [19:11:41] *** sartek has quit IRC [19:16:31] <CIA-22> Rod Evans: 6494228 pclose() error when an audit library calls popen() and the main target is being run under ldd; 6518480 ld -melf_i386 doesn't complain [19:17:48] *** logic_ has joined #opensolaris [19:18:00] *** yarihm has quit IRC [19:21:50] *** sartek has joined #opensolaris [19:22:57] *** gdamore has joined #opensolaris [19:22:58] <Camus_> no? [19:22:59] <Camus_> not really [19:23:00] <Camus_> i see ;D [19:23:08] <Camus_> wel damnit i just wanna beat up thi ssun box right now [19:23:15] <Camus_> i want my damn 500gb firewire 800 drive to work [19:23:18] <Camus_> can't test oracle rac ;D [19:24:02] <gdamore> heh. ieee1394 on sun boxes is pretty much just for webcams. :-) [19:24:39] <richlowe> wasn't Auralis using (trying to use?) firewire attached storage at one point? [19:25:00] <richlowe> I seem to recall conversation about a consistent panic on boot related to external disks from a year or so back. [19:25:16] <Auralis> firewire dvdwriter it was [19:27:33] *** dunc has quit IRC [19:27:55] <Auralis> it killed the kernel when the vold was accessing it during login screen start [19:27:59] <Camus_> oh wow [19:28:12] <Camus_> i unplug the device while in the system [19:28:15] <Camus_> and the system panics [19:28:18] <Camus_> coredump ;( [19:28:59] <richlowe> Auralis: oddly, I managed to do similar with a flaky IDE DVD a few days ago. [19:29:06] <richlowe> I'm yet to look into the how and why though. [19:29:13] <gdamore> yeah, ieee1394 might not be hotplug safe. :-) [19:29:17] <richlowe> (not least because the damn thing is working right now, so the machine won't panic again) [19:29:27] <gdamore> you might be able to cfgadm remove it from the system first, though. [19:29:32] <Camus_> gdamore [19:29:41] <Camus_> cfgadm won't even show me hte device is connected [19:29:45] <Auralis> i can plug my fw dvd writer just fine, once my desktop is running [19:30:03] <gdamore> plug it in, or remove it? [19:30:12] <Camus_> its strange though, scanpci shows that the Texas Instrumjents controller is there, and modinfo shows me that hci1394 is loaded but no go in terms of gettin ghtis external disk going [19:30:18] <richlowe> spew of sd (via atapi) errors from it on boot, ran rmformat (foolishly), and it hit a kmem debug check. (overran part of the redzone) [19:30:49] *** RaD|Tz has joined #opensolaris [19:31:07] <gdamore> ouch. what version of Solaris? [19:31:12] *** nachox has left #opensolaris [19:31:14] <gdamore> (was it a boot drive?) [19:31:32] <richlowe> gdamore: me? no, flaky a flaky DVD drive. (see above) :) [19:31:46] <Auralis> however, since writing over fw to a dvd writer is positivly absolutly not working at all, i use the usb2 ports on my writer without a problem [19:31:52] <gdamore> then it should definitely not have panic'd. [19:31:55] <richlowe> and as for version "onnv-gate as of about last friday. [19:32:34] <richlowe> and no it shouldn't, when I've had a chance to look into the dump more I intend to find out why it did. :) [19:33:04] <gdamore> good. [19:33:39] <gdamore> some of the error paths in these drivers are amongst the least well debugged. [19:34:46] <gdamore> Camus_: if you can use usb2 instead of firewire, do so. i wasn't even aware that firewire storage was usable at all. [19:34:55] <Camus_> i dont think so either [19:35:38] <Camus_> forget solaris [19:35:40] <Camus_> ill do it linux [19:36:27] *** compukid_ has quit IRC [19:37:19] <gdamore> Camus_: hmm there is a scsa1394 driver. Did you try doing devfsadm -i scsa1394? [19:41:16] <Camus_> failed to aqttach [19:42:53] <gdamore> hmm... that sounds like a problem. Does your device conform to SBP-2? (Do you know?) [19:43:04] *** postwait_ has joined #opensolaris [19:43:28] <gdamore> you could also look at prtconf -vp and search for firewire. storage should be firwewire00609e,010483 [19:44:09] <Camus_> i grep'd for firewire [19:44:10] <Camus_> and nothing [19:45:12] <gdamore> hang on, let me try something on my sunblade 100 [19:45:21] *** compukid_ has joined #opensolaris [19:45:27] <gdamore> try prtconf -v (no p) [19:45:47] <gdamore> the p was indicating a prom property. i don't think firewire shows up in the prom [19:47:22] <Camus_> hmm [19:47:25] <Camus_> firewire0000 came up [19:47:26] <Camus_> nice [19:48:19] <gdamore> okay it sounds like your device is not exposing the right vendor/device ids [19:48:48] <Camus_> is that bad [19:48:51] <Camus_> can it be fixed/ [19:48:54] <Camus_> did i do something / [19:49:07] <Camus_> s/\//? [19:49:07] <gdamore> dunno. but you could try using update_drv to force it to recognize for now. [19:49:11] *** pseudoXh4 has quit IRC [19:49:59] <Camus_> gda [19:50:02] <Camus_> i love you [19:50:04] <Camus_> check this out.. [19:50:25] <Camus_> unit (driver not attached) [19:50:25] <Camus_> Hardware properties: [19:50:25] <Camus_> name='unit-dir-offset' type=int items=1 [19:50:25] <Camus_> value=00000012 [19:50:25] <Camus_> name='reg' type=int items=6 [19:50:25] <Camus_> value=00010800.0000d766.00000000.00000000.00000000.0 [19:50:27] <Camus_> 0000000 [19:50:29] <Camus_> name='compatible' type=string items=1 [19:50:32] <Camus_> value='firewire080020,000001' [19:50:33] <Camus_> sorry for the flood guys [19:51:07] <gdamore> use the compatible property, firewire080020,000001 listed. [19:51:13] *** RaD|Tz has quit IRC [19:51:37] <Camus_> what do i do with that? [19:51:40] <Camus_> update_drv firewire080020,000001 ? [19:52:33] <gdamore> update_drv -a -i 'firewire080020,000001' scsa1394 [19:52:58] *** Darwin has quit IRC [19:53:09] <Camus_> where should i send my g/f ? [19:53:10] *** Darwin has joined #opensolaris [19:53:14] <Camus_> to you office or home? [19:53:24] <Camus_> the scsa1394 failed to attach :( [19:53:44] <gdamore> bummer. [19:53:45] <tomww> does xmms use gnome virtual filesystem? this would solve a temporary problem with xmms playing music on a SMB_share [19:53:57] <Camus_> so the scsa1394 wont recognize it huh/ [19:53:59] <gdamore> look in the system logs, /var/adm/messages or dmesg, maybe there is some useful output there [19:54:27] <Camus_> nada [19:54:59] <gdamore> looking at source now, just a sec [19:55:14] <axisys> is there a way I can download sol 8 from somewhere? [19:55:30] <Auralis> from sun [19:55:32] <gdamore> yes. but i don't recall the exact url. [19:56:12] <Auralis> you have 17 days left to figure out how, then it stops shipping [19:56:35] *** hile_ has quit IRC [19:56:52] <axisys> i do have a contract .. but i like to download it instead of waiting for a shipment [19:57:33] <gdamore> Camus_: I don't see any obvious checks for vendor/device id in the driver itself, but its spread across a bunch of files, so i could be mistaken. (It checks for certain firmware revs though.) [19:58:02] <jmcp> axisys: *why* do you want to get sol8? are you a masochist? [19:58:11] *** hile_ has joined #opensolaris [19:58:17] <jmcp> hi hile_ [19:58:20] <CIA-22> Roger Faulkner: 6518780 deadlock due to fork and suspend thread [19:58:22] <axisys> one of our box has sol 8 and i need to do some fix on it [19:58:26] <Camus_> well gda - if there were any revisions to the 1394 implementation in opensolaris [19:58:36] <Camus_> than i'd have to find out by installing, i'm still using the solaris 10 instal from sun.COM [19:58:50] <axisys> i tried to image it to a disk and boot from that disk on a diff system.. and no luck [19:58:53] <gdamore> which update? [19:59:03] <axisys> 07/03 or higher [19:59:04] <Camus_> i'm not sure - it was installed befor eme but i cannot check [19:59:12] <gdamore> cat /etc/release [19:59:13] <Camus_> theres not /etc/VERSION or anything ;D [19:59:15] <Camus_> ahh [19:59:20] <Camus_> 11/04 [19:59:21] <Camus_> 11/06 [19:59:34] <gdamore> 11/06 is the latest. that's good. [19:59:43] <Camus_> i've seen there is patch clusters already [19:59:51] <gdamore> but i've still no clue why your drive won't attach. [19:59:52] <axisys> jmcp: i need that to make an identical system [19:59:53] <Camus_> i do not have a sun support contract so.. u know.. [19:59:54] <jmcp> Camus_: btw, there doesn't appear to be a cfgadm plugin for scsa1394 [20:00:04] <jmcp> axisys: I'll send you a sympathy packet [20:00:05] <gdamore> note that updates are just a bunch of patch updates bundled together. :-) [20:00:13] <Auralis> http://www.sun.com/software/solaris/8/index.xml [20:00:16] <Camus_> yeah [20:00:20] <jmcp> gdamore: as well as the odd new feature, too [20:00:21] <Camus_> beats downloading them one by one ;D [20:00:45] <gdamore> the odd new feature is still delivered as a "patch" (feature patch) :-) [20:00:56] <Camus_> symping ? [20:00:59] <Camus_> sympathy packets ;D [20:01:00] <Camus_> ? [20:01:21] <jmcp> Camus_: sent occasionally when it's the easiest thing to send ..... some people prefer them to getting flowers :) [20:01:25] <jmcp> gdamore: semantics, mostly [20:01:37] <Camus_> lol [20:02:42] <Camus_> these boxes will be replaced by the x4200's we just ordered [20:02:46] <gdamore> Camus_: I'm out of ideas. Sorry. :-( [20:02:51] <Camus_> and a real array, a 6140.. unfortunately this test thing iddn't work [20:02:58] <Camus_> gdamore - thanks though.. i learned a thing or two from u [20:03:04] <Camus_> that i will use to raise my pay grade :D [20:03:11] <gdamore> np. :-) [20:03:55] * gdamore wonders if he could learn a thing or two to raise his own pay grade.... hmm.... [20:04:01] *** RaD|Tz has joined #opensolaris [20:06:06] <Camus_> so where we all located [20:06:09] <axisys> Auralis: thnx [20:06:10] <Camus_> new york, ny [20:06:55] <gdamore> so. cal. 60-70 miles from any where with tech employers. ;-) [20:08:01] <Camus_> i'll be in NM in a few days - want to drive to California [20:08:03] <axisys> now anyone knows why a mirrored disk off of 450 cannot not be used to boot a netra t1? [20:08:42] <gdamore> its a fair bit of a drive. you have to get across arizona, which takes a while. [20:09:03] <Camus_> yeah drove across there already [20:09:37] <Camus_> live in tucson [20:09:39] <Camus_> drove to florida [20:09:45] <jmcp> axisys: first thing that comes to mind is that the device paths are different [20:11:05] <axisys> jmcp: i ran this devfsadm -v -C -r /mnt where /mnt is s0 my root [20:11:10] <axisys> jmcp: from cdrom [20:11:37] <jmcp> axisys: mirrored with svm? [20:11:40] <Auralis> hrmm, the sun discplay configuation manager is neat, it lets you create custom settings for Xsun [20:11:55] * jmcp emphasises that axisys is trying a *completely unsupported by Sun* operation [20:12:09] <axisys> jmcp: yes [20:12:27] <axisys> jmcp: i changed the vfstab to point to c1t1d0 [20:12:28] <jmcp> axisys: then I would expect that you would need to get the metadbs fixed up too [20:12:42] <jmcp> did you try a boot -a, as well? [20:12:47] <axisys> jmcp: just one disk.. i removed metadbs [20:12:55] <axisys> jmcp: i tried boot -srv [20:13:08] <axisys> jmcp: becasue i cleared /ect/system [20:13:11] <jmcp> try the -a [20:13:22] <axisys> jmcp: so no ref to metadevice [20:13:25] <axisys> jmcp: ok [20:14:12] <axisys> jmcp: i get the error message exactly like this http://rafb.net/p/Kf2BIj20.html [20:14:26] <axisys> it sees the os.. but not the /usr partition [20:14:45] <axisys> i tried fsck the /usr partition couple times .. it is always clean [20:14:57] *** pjd has quit IRC [20:15:00] <axisys> /dev/dsk/c1t1d0s4 -> /usr [20:15:01] *** pjd has joined #opensolaris [20:15:05] <axisys> which is correct [20:15:16] <axisys> not sure why it says no such file or dir [20:16:31] <jmcp> axisys: is that c1t1d0 from teh e450's point of view, or from the netra's? [20:16:44] <axisys> netra [20:16:50] <jmcp> really? [20:16:56] <axisys> on 450 it was c0t1d0 [20:17:25] <jmcp> oh, yeah .. on the netra you've got a cd or dvd as c0, right? [20:17:32] <axisys> jmcp: let me boot from cdrom again [20:17:51] <axisys> cdrom -> /pci@1f,0/pci@1,1/ide@d/cdrom@0,0:f [20:18:48] <jmcp> are you sure that you've got c1t1 rather c1t0 ? [20:19:58] <axisys> jmcp: i am booting form disk1 ... c1t0 does not have any OS.. so i should not get that far [20:20:21] <axisys> jmcp: plus it sees the hostname of 450.. so it is reading the right disk [20:22:45] <jmcp> axisys: don't make assumptions, check facts [20:27:09] <Cyrille> much funnier to speculate wildly... [20:27:38] <jmcp> axisys: and what does vfstab say for / on this disk? [20:28:04] <axisys> jmcp: http://rafb.net/p/OmrDcq27.html [20:29:04] <jmcp> axisys: if the vfstab from your transferred disk, what device is used for "/" ? [20:29:58] <axisys> jmcp: http://rafb.net/p/dX8y1k71.html [20:30:21] <axisys> which is disk1 on netra [20:31:11] <CIA-22> Michael Speer: 6506840 NIU Code Clean up related to code review comments; 6513043 Jumbo frames stress tests caused RDC stopped receiving packets; 6513045 dladm shows link state standby after creating an aggregation; 6518450 nxge: Fix provided for CR 6489683 incomplete [20:31:13] *** chuck_unix has joined #opensolaris [20:31:32] <axisys> jmcp: i yanked out disk0 a while ago [20:31:36] <axisys> jmcp: http://rafb.net/p/ne3LA092.html [20:32:32] <axisys> jmcp: let me go back to ok prompt and check the devalias one more time [20:33:02] <Symmetria> UGH [20:33:07] <Symmetria> it seems my solaris box just died [20:33:59] <axisys> jmcp: http://rafb.net/p/1jBMWB52.html [20:36:48] <jmcp> axisys: go and check that the path /dev/dsk/c1t1d0s4 actually points to /pci@1f,0/pci@1/scsi@8/disk@1,0:d [20:38:03] <Symmetria> http://www.pastebin.ca/333224 <=== anyone got any idea what coulda caused that [20:38:26] <Symmetria> that happened right before my machine completely locked up for 15 minutes [20:38:35] <Symmetria> eventually it came right but had a load average of 80+ [20:40:18] <axisys> jmcp: /dev/dsk/c1t1d0s4 -> ../../devices/pci@1f,0/pci@1/scsi@8/sd@1,0:e [20:40:58] <axisys> s4 to e looks right [20:41:41] <jmcp> ok, yeah [20:43:24] <axisys> jmcp: filesystem looks clean too http://rafb.net/p/3Wwn2v12.html [20:43:48] <jmcp> axisys: you've got something else going on [20:43:57] <jmcp> axisys: sorry, I can't help any more - work happening [20:44:08] <axisys> jmcp: thnx tho [20:48:11] *** stevel has joined #opensolaris [20:48:11] *** ChanServ sets mode: +o stevel [20:48:54] <Symmetria> hrm [20:49:00] *** Gman_ is now known as Gman [20:50:33] <richlowe> stevel, Gman. [20:50:40] <Gman> hey rich [20:50:46] <stevel> 'afternoon rich, glynn [20:51:35] *** CosmicDJ has joined #OpenSolaris [20:52:36] *** RaD|Tz has quit IRC [20:53:34] <rodrickbrown> wow this is freaking awesome [20:53:36] <rodrickbrown> http://opensolaris.org/os/project/crossbow/ [20:58:54] *** tsoome has joined #opensolaris [20:59:17] * bubbva is away: away [21:00:56] <Gman> hi stevel [21:05:26] *** Tpenta has joined #opensolaris [21:06:56] <Gman> hey Tpenta [21:07:09] <Symmetria> URGHHHHHHHHHH [21:07:27] <Symmetria> someone is attacking one of my servers [21:07:46] <Tpenta> hi gman [21:08:48] <Tpenta> hey, when i upgraded to b56 on the u60, I lost nearly all of my desktop shortcut keys, not only the key map, but the key action, i have no entries for things like front, copy, screen right and left [21:08:48] <Gman> stop them quick, pull the plugs! [21:09:40] <Gman> Tpenta, did you check the gconf database? [21:10:26] *** Camus_ has quit IRC [21:10:54] <Tpenta> not yet, how does one do that? [21:11:19] <Gman> best to open up gconf-editor from the commandline [21:12:00] <Gman> changes should be under /desktop/gnome i think [21:12:30] <Gman> in peripherals/keyboard [21:12:35] <Tpenta> i'll look at it when i get into the ofice today; for now I'd better go and start organising things for the kids to go to school [21:13:19] <Gman> or in /apps/gnome_settings_daemon [21:13:30] <Gman> or in /apps/metacity ;) [21:14:16] <Tpenta> telya something, not having right, left and font is a real pita [21:15:11] * Gman never used those keys [21:16:08] <Tpenta> ctrl-alt-right and ctrl-alt-left; use them all the time [21:16:15] <Tpenta> and front too [21:17:13] <Gman> alt-esc is another good one to know [21:18:31] *** sommerfeld has joined #opensolaris [21:18:32] *** ChanServ sets mode: +o sommerfeld [21:19:22] <Tpenta> morning bill [21:20:15] <Gman> man, it's so fricken cool that jonathan continues to blog [21:21:17] <Tpenta> oh yea [21:21:20] <Cyrille> Tpenta, it's apps/metacity (and then *keybindings) which contain that sort of shortcuts. [21:21:33] <Tpenta> even cooler thath he continues to comment on blogs [21:21:53] <Tpenta> he wished me a happy bday on mine and commented on the hair growing back [21:22:01] <Tpenta> back on jan 9 [21:22:02] *** tsoome1 has joined #opensolaris [21:22:30] <sommerfeld> good afternoon. i think whatever cold is making the rounds is starting to affect me. [21:22:59] <Tpenta> hope you kick that one quickly bill [21:23:02] <Gman> Tpenta, that's pretty cool [21:23:19] <richlowe> if it's the same cold I have, it's a drawn out bugger of a thing. [21:24:04] <Gman> richlowe, whisky, water, cloves, lemon, sugar - perfect remedy :) [21:24:33] <quasi> sommerfeld: do some heavy drinking and see if it helps (and if it doesn't, at least you had fun trying ;) [21:25:00] *** rcheli has joined #opensolaris [21:27:13] *** Cyrille has quit IRC [21:27:14] <Gman> can anyone confirm or deny the rumour that gosling shouted the last beer bust? [21:27:35] <Odin-LAP> ... what? [21:27:56] <sommerfeld> i heard rumors of a beer bust in mpk but nothing more than that [21:28:09] <Odin-LAP> "beer bust"? [21:28:15] <Gman> Odin-LAP, read jonathan's blog - free beer basically [21:28:15] <jmcp> Tpenta: we should have a beer bust in Sydney [21:28:24] <Odin-LAP> Gman: Ah. [21:28:27] <Gman> sommerfeld, i thought there was beer busts every 2nd week in mpk [21:28:30] *** tsoome has quit IRC [21:28:51] * Odin-LAP wonders if those could be introduced where he works. [21:29:00] *** tsoome1 is now known as tsoome [21:29:03] <Odin-LAP> (A national liquor monopoly. :) [21:30:08] <rcheli> With the change in Daylight savings time issue, am I okay with the solaris 10 servers ? Any additional patches needed ? [21:30:57] <quasi> rcheli: there's a patch or two for the changes [21:30:58] <sommerfeld> Odin-LAP: in the US, some of that is at state level in some states, leading to absurdities like "New Hampshire State Liquor Store And Safety Rest Stop" on one of the main highways... [21:31:18] <Odin-LAP> Heh. [21:31:32] <rcheli> quasi: if I do smpatch update -L on all of my solaris boxes today, I should be covered, right ? [21:31:39] <rcheli> or there are more patches coming for this ? [21:31:50] <Tpenta> jmcp didnt you see the note about friday? [21:32:03] * Odin-LAP contends that there's greater choice and lower prices as it is than would ever be feasible without the state-run monopoly. [21:32:03] <quasi> rcheli: should be fine if you're up to date [21:32:23] <sommerfeld> Gman: so, I've been to smallish beer busts on the 3rd floor of MPK17. But I hear the one last Friday was in the cafeteria. [21:32:31] <Gman> ahh [21:32:34] <Odin-LAP> (Especially due to the state-run part of it.) [21:33:38] <jmcp> Tpenta: as a regular thing, I mean :) [21:33:40] <rcheli> quasi: some patches don't come when I do smpatch update -L, will the patches for daylight saving issue come when i do smpatch ? [21:34:07] <Tpenta> we have steak wednesday [21:34:26] <quasi> rcheli: ah, you've not got a contract? [21:34:45] <rcheli> quasi: only for some systems,not all [21:35:22] <quasi> rcheli: pca checks all patches even if you have no contract, so that might be worth a try [21:36:52] <g4lt-U60> sommerfeld, I fail to see how a toll road can be called a "main highway" [21:38:21] <rcheli> quasi: is it good practice to use pca for all systems including the one that has contract ? [21:39:10] <sommerfeld> g4lt-U60: I think there's more than one such combined facility, and I know at least one of them isn't on a toll road [21:39:40] <quasi> rcheli: that's debatable [21:40:05] <quasi> rcheli: but it works well for finding patches when you don't have a contract [21:41:52] <rcheli> quasi: cuz so far I use smpatch update -L for all systems including the one with contract [21:42:13] <rcheli> what do you use to update the patches on a system with contract ? [21:42:29] <quasi> smpatch [21:42:37] <rcheli> hah, ok [21:43:18] <quasi> except patches for studio 11 on boxes with zones - there pca works well [21:44:41] <g4lt-U60> sommerfeld, sort of. it straddles a toll road and a secondary freeway [21:45:14] *** qbit has joined #opensolaris [21:47:49] *** _william_ has joined #opensolaris [21:47:58] <rodrickbrown> does sun provide local patch server like interface for packages? [21:48:03] <axisys> my jumpstart dir shows this Copyright JDS-THIRDPARTYLICENSEREADME Solaris_10 installer [21:48:22] <axisys> how do i tell which release sol 10 is this? [21:48:42] <_william_> hi all [21:48:51] <mrdeviant> axisys, cat /etc/release [21:49:05] *** nadir has joined #opensolaris [21:49:15] <axisys> mrdeviant: nawp.. i am talking about jumpstart dir [21:49:21] <axisys> not the OS itself [21:49:41] <axisys> I have a jumpstart dir built from a OS CD [21:49:57] <axisys> how do i tell which release the OS is from the jumpstart dir? [21:50:36] <axisys> here is jumpstart dir looks like [21:50:39] <axisys> Copyright JDS-THIRDPARTYLICENSEREADME Solaris_10 installer [21:51:01] <axisys> how do i tell which version Solaris_10 that is? [21:51:43] <charlieS> look in the Product dir, and grep 118833 [21:51:51] <charlieS> what's the -##? [21:52:28] <charlieS> oh, that won't be there. [21:52:31] <charlieS> nevermind me :P [21:52:49] <axisys> grep 118833 * none :P [21:53:08] <CosmicDJ> 118855 for x86 [21:53:36] <charlieS> xWhat? pfft. :) [21:55:30] <mrdeviant> um, you could look in Solaris_10/Product/SUNWsolnm/reloc/etc/release :P [21:56:11] *** pikapika has joined #opensolaris [21:56:30] <charlieS> nice :) [21:56:45] <pikapika> hello [21:56:53] <axisys> mrdeviant: u da man [22:00:17] <Symmetria> ARGH [22:00:22] *** maldous has joined #opensolaris [22:00:24] <Symmetria> dammit someone is hitting my server *again* [22:00:42] <quasi> hit them back? [22:00:43] <maldous> moo. [22:01:29] <maldous> may I ask 11/06 questions here without being shunned? [22:02:03] <Symmetria> quasi err no [22:02:03] <Auralis> sure [22:02:09] <Symmetria> dont hit things back like that [22:02:13] <Symmetria> it never helps to escaulate with kids [22:02:21] *** compukid__ has joined #opensolaris [22:02:31] *** compukid_ has quit IRC [22:02:39] <maldous> i have an e1000g aggregate configure as a vlan. it only works, when I have snoop running (presuming promisicuous required). [22:02:48] <Tpenta> you could dare them to break into your host at 127.0.0.1 [22:03:04] <Symmetria> Tpenta right now they are syn flooding a major server of mine on its http port [22:03:07] <myrkraverk> what programming languages do the drow use? [22:03:26] <maldous> anyone experienced this? [22:03:41] <quasi> Symmetria: ah, flooding - I thought it was physical violence ;) [22:03:59] <jbk> ok.. should anything but a setfacl command be changing the mask? [22:04:03] <maldous> I found someone on javaforum who mentioned an update to the ce driver resolved this, but i can't see any e1000g patches.. [22:04:30] <maldous> i'm fully patched up and have all the sun support I need, just want to kick this now. [22:04:41] *** mrdeviant has left #opensolaris [22:04:53] <charlieS> jbk: chmod can. [22:05:07] <maldous> aggr vlan only works with snoop. i'll listen for a bit.. [22:08:36] <maldous> anyone working with aggregates or vlans? [22:08:45] *** jesse-jads has quit IRC [22:11:44] <Symmetria> maldous I have plenty of 802.1Q vlans under solaris 10 x86 [22:13:21] <jbk> hmm.. i'd be curious to know the thought process behind that.. [22:13:29] <jbk> doesn't seem intuitive [22:13:41] <jbk> although i suppose it's somewhat moot with zfs (if we ever get there) [22:13:58] <maldous> i have used sun trunking for a while, and have today tried to aggregate and trunk without luck. i want a new ge driver i guess. [22:14:18] <maldous> or, what is a snoop command that has no output? ;) [22:14:19] <Symmetria> hrm, Im using e1000g for straight 802.1q vlans and its working fine [22:14:53] <maldous> find a box with two nics, aggregate and try then ;) [22:15:44] * Symmetria sighs at this syn flood [22:16:35] <Symmetria> 11:16pm up 18 day(s), 22:41, 2 users, load average: 173.95, 344.63, 203.41 [22:16:53] <e^ipi> whoever's doing that is totally 1337 and i have nothing but the utmost respect for his m4d skillz [22:17:18] <Symmetria> wierd I have absolutely no idea what actually caused that [22:17:22] <Symmetria> nothing in dmesg [22:17:27] <Symmetria> nothing in netflow that I can see [22:21:58] *** laca has quit IRC [22:22:27] *** laca has joined #opensolaris [22:23:51] *** freakazoid0223 has joined #opensolaris [22:24:07] <rcheli> hmm...where is the sendmail directory on sol 10 ? [22:24:21] <rodrickbrown> config or bin ? [22:24:23] <tsoome> a /etc/mail [22:24:40] <rcheli> ahh,i was trying /etc/default/mail [22:24:43] <rcheli> thanks tsoome [22:25:35] <Symmetria> Im wondering if Im not having problems on one of my san's and writing to it is lagging behind real badly or something and its causing it to think its not there for a minute and causing huge spikes in load as the thing waits for it to come back [22:31:15] *** hile_ has quit IRC [22:31:48] *** dlynes_laptop has joined #opensolaris [22:34:10] *** phipsbook has joined #opensolaris [22:35:04] *** qbit has left #opensolaris [22:36:01] *** phipsbook has quit IRC [22:36:15] *** phipsbook has joined #opensolaris [22:36:42] *** LeftWing has joined #OpenSolaris [22:39:23] *** steleman_wor1 has joined #opensolaris [22:41:39] *** yarihm has joined #OpenSolaris [22:44:29] *** zrb has joined #opensolaris [22:44:55] *** kloczek has quit IRC [22:49:03] *** steleman_wor1 has quit IRC [22:53:28] *** maldous has left #opensolaris [22:55:14] *** gisburn has joined #opensolaris [23:00:16] *** gisburn has quit IRC [23:05:33] * bubbva is back (gone 02:06:16) [23:05:55] *** pseudoXh4 has joined #opensolaris [23:15:44] *** hile_ has joined #opensolaris [23:19:53] *** gisburn has joined #opensolaris [23:24:07] <gisburn> !seen alhopper [23:24:16] *** sartek has quit IRC [23:25:21] <gisburn> !seen kupfer [23:25:22] <Drone> AlHopper (AlHopper!n=chatzill at logical dot logical-approach.com) was last seen in #opensolaris on Wed 14 Jun 2006 20:17 GMT, saying 'stefanp: it's becoming an increasing issue ... some Northbridges are already dissipating 20Watts ... and rising!'. [23:25:30] <Drone> kupfer (kupfer!i=kupfer@nat/sun/x-307d735aa42be95d) was last seen in #opensolaris on Thu 11 Jan 2007 00:16 GMT, saying 'later, folks.'. [23:25:41] <gisburn> grumpf [23:26:13] *** gisburn has left #opensolaris [23:30:33] *** stevel changes topic to "Latest builds: SXCR: 56 | ON: 56 | OGB needs you! : http://tinyurl.com/yvj8ot" [23:32:39] *** sommerfeld has quit IRC [23:32:46] *** sartek has joined #opensolaris [23:35:25] *** stevel changes topic to "Latest builds: SXCE: 56 | ON: 56 | OGB needs you! : http://tinyurl.com/yvj8ot" [23:35:36] <stevel> R becomes E. yaaaaaaaay marketing [23:35:52] <gdamore> wtf does SXCE mean? [23:36:13] <stevel> release => edition [23:36:22] <tsoome> solaris express customer edition? [23:36:37] <gdamore> i dunno what peeves me more. [23:36:38] <phipsbook> 'community' [23:36:50] <tsoome> ah yes [23:36:52] <gdamore> annoyances like this name change? [23:37:15] <gdamore> or the fact that some marketdroid is probably making a good salary for making these changes? [23:37:41] <gdamore> (i.e. more salary than the folks who actually "contribute" to solaris. :-) [23:38:19] * gdamore still has unhealed bruises from the Solaris 2.6 -> Solaris 7 name change. [23:39:10] <jbk> at least they didn't go to Solaris '99 or such :) [23:39:48] * gdamore smacks jbk for even suggesting the idea in a forum where a marketdroid might lurk. [23:40:05] <jbk> you think a marketdroid knows what irc is? [23:40:14] <gdamore> good point. [23:40:31] *** Mazon is now known as mazon [23:40:37] <gdamore> hopefully nothing here is archived where google can find it. :-) [23:40:57] <jbk> i think the only think more obscure would be some sort of MU* [23:42:21] <AbeFroman> web page still points to 55 [23:42:41] <CosmicDJ> gdamore: I hope the java 1.5 -> 6 change didn't kill you ;) [23:43:05] <gdamore> nah. i'd left Sun by then, and I hardly touch Java anymore. :-) [23:43:23] <Gman> stevel, edition? how lame. [23:43:44] * gdamore still needs to finish the rewrite of the Tadpole wifi tool in GTK. (First version was in Java) [23:43:47] <Gman> i assume it's to highlight it's an unsupported product [23:45:22] <richlowe> Nah. [23:47:18] *** triplah_ has joined #opensolaris [23:50:31] <jbk> christ.. this daylight's savings has people running around like chickens with their head cut off :)