October 23, 2006 
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
October 23, 2006 [00:00:44] *** bank__ has quit IRC [00:01:37] <dunstan> There was a site listing Sun's blog entries related with technical aspects of Solaris. Does anyone have a link to that site? I've lost it from my notes. [00:02:00] *** ahashim has left #opensolaris [00:02:41] *** bank has joined #opensolaris [00:02:51] <bank> oh no! [00:02:57] <bank> I lost my server again!!! [00:03:36] <bank> what wrong with my zfs!? [00:05:36] <hile_> the first thing that comes to mind is your lack of backups and perhaps your lack of a HA config :) [00:06:01] <bank> I didn't do any special [00:06:12] <bank> I just refresh whole system with b50 [00:06:17] <bank> :( I lost my server agaiN!!! [00:06:31] *** nwf has joined #opensolaris [00:06:46] <bank> let's take a look http://paste.lisp.org/display/28516 [00:07:36] <stevel> dunstan: opensolaris.org/os/blogs ? or planetsolaris.org? [00:08:10] <bank> :( what i am wrong with zfs!! [00:08:23] <bank> madness. I just refresh the whole system. [00:11:35] <Gman> http://mail.opensolaris.org/pipermail/opensolaris-discuss/2006-October/021145.html [00:11:39] <Gman> opensolaris weekly news #34 [00:11:44] <Gman> later dudes :) [00:11:46] *** Gman has quit IRC [00:12:28] <OnkelSchorsch> yay [00:12:31] <dunstan> stevel: Aha! Thanks. [00:13:52] *** nexrafa has joined #opensolaris [00:18:10] *** Mazon is now known as mazon [00:22:15] *** Jiko_ is now known as Jiko [00:25:08] *** Pr0fMikey is now known as ProfMikey [00:27:50] *** dunstan has quit IRC [00:35:27] *** coffman has quit IRC [00:46:23] *** piwi has joined #opensolaris [00:51:58] *** Tpenta has joined #opensolaris [00:53:47] *** klepparii is now known as kleppari [00:57:46] *** esaxe has joined #opensolaris [00:58:00] *** lloy0076 has joined #opensolaris [00:58:34] <Stric> http://research.sun.com/techrep/2006/abstract-156.html .. isn't this java thing going a bit far? :) [00:59:26] <Auralis> a bit? [01:00:17] <boyd> Good grief. We must stop them before they kill us all! [01:01:48] <Error_404> lol [01:01:54] <Error_404> that's fantastic [01:02:25] <Error_404> how would you write directly to memory in java is what I wanna know [01:02:30] <Error_404> esp. since endianess matters [01:10:14] <Andrew____> Well, the driver they wrote was the RAM disk driver, a pseudo driver (no hardware to drive). That probably made things significantly simpler than chosing a driver with extensive and perculiar requirements to access the hardware. [01:10:21] *** gisburn has joined #opensolaris [01:10:28] <gisburn> !seen kupfer [01:10:30] <Drone> kupfer (kupfer!i=kupfer@nat/sun/x-3d845599e2560142) was last seen in #opensolaris on Mon 16 Oct 2006 20:31 GMT, saying 'well, if they're both using the same v4 domain, I don't know what the problem is.'. [01:10:42] <gisburn> f*ck [01:10:49] <gisburn> x!!xx@@@!! [01:10:58] <stevel> it's a weekend. people aren't at work [01:11:14] <gisburn> thats no excuse [01:11:56] <gisburn> stevel: I've been busy with more demo applications, including http://www.opensolaris.org/os/project/ksh93-integration/screenshots/ksh93_mandelbrot001.png and a rSS reader [01:12:12] <boyd> I'm at work [01:12:25] *** fet-kviga has joined #opensolaris [01:12:30] <gisburn> stevel: see ? boyd is working, too! [01:12:41] <boyd> It's 9:13AM monday [01:13:11] * gisburn suggests xntpd for boyd - you're 28secs to early [01:13:19] <rodrickbrown> Error_404, int[] foo = new int[2]; foo[0] = 10; [01:13:20] <rodrickbrown> woot [01:13:53] <Error_404> ? [01:13:53] <gisburn> stevel: do you have any suggestions for more sample applications ? [01:14:01] *** lloy0076 has left #opensolaris [01:14:02] <fet-kviga> the C compiler that come swith Sun Studio/x86 comments its asm code by linenumber while the SPARC version comments it by the actual C code; is there a way to get the SPARC version's behavior on x86? or do we have to wait for an `update' ? [01:14:25] <gisburn> fet-kviga: which compiler versionn ? [01:14:39] <fet-kviga> the one that comes with sun studio 10 [01:15:10] <fet-kviga> i think sun studio 11 is the same as well [01:18:56] *** tek7 has joined #opensolaris [01:19:45] * stevel is working too [01:19:54] <gisburn> stevel: working on... what ? [01:19:55] <stevel> and it's sunday afternoon here :-P [01:20:02] <stevel> catching up on work [01:20:08] <tek7> Hi guys, i just installed SXCR on my ultra10 and the graphics look horriffic, been looking around for hours on google and sun.com but can't seem to figure out how to change the res on Xsun - at the moment it's 800x600x60/8 [01:20:18] <gisburn> stevel: do you have any ideas for /usr/demo/ksh/fun/ ? [01:20:24] <tek7> Can you point me in the direction of a man page or such? [01:20:24] <stevel> nope, sorry [01:20:25] * stevel isn't fun [01:20:32] <Auralis> tek7: fbconfig is yourf riend [01:20:48] <tek7> yeah i discovered that and m64config [01:20:53] <tek7> but i get an error when i run that [01:20:56] <gisburn> stevel: so far we have pacman, a mandelbrot generator and a RSS reader... [01:20:57] *** koolniczka has quit IRC [01:22:41] * gisburn needs better ideas [01:26:25] *** gm152 has quit IRC [01:35:27] *** bengtf has quit IRC [01:41:55] *** astinus has joined #opensolaris [01:44:14] *** Tzoa has quit IRC [01:46:53] <astinus> alanc: I've got problems that're specific to OpenSolaris with nVidia drivers; should I file a bug report? To test I installed Ubuntu Dapper onto the system, installed nVidia drivers, configured X.org and it works fine .. same xorg.conf under OpenSolaris with nVidia driver fails. [01:47:32] <Error_404> what precisely do you think will come of that? [01:47:50] <Triskelios> shouldn't the bug be reported to nvidia? [01:48:07] <Error_404> Triskelios: i assumed he meant "file a bug with nVidia" [01:49:12] <astinus> File a bug with either - there's clearly something amiss. Driver versions (nVidia) under Ubuntu / Solaris are identical - both cards are detected under both operating systems too, I just get weird fscked up errors about GLX module loading in OpenSolaris [01:51:05] <astinus> I'm not at all familiar with how drivers / modules under Solaris work, I'm trying to learn all this stuff, but 5+ years experience with Linux gives me a lot of familiarity there (and a lot of catching up elsewhere) .. My question is -should- I file a bug report, and if so, with whom? [01:53:36] <Triskelios> astinus: um, a "GLX module" would be an X server module [01:54:20] <astinus> nvidia-glx [01:56:42] *** sahafeez has joined #opensolaris [01:57:22] <Triskelios> astinus: GLX is an X extension, the module is just a X server module, not a kernel module. the driver has a different kernel module component like DRI has [01:58:52] <astinus> Triskelios: Okay, so my issue lies with the X server module provided by nVidia to facilitate GLX while using their cards? Still not sure how I can fix it - I really need both graphics cards and all three displays to work :/ [02:00:04] *** OnkelSchorsch has quit IRC [02:06:23] *** piwi has quit IRC [02:09:23] *** gnu2it2 has quit IRC [02:09:26] <Triskelios> astinus: having the error text would help, if it's from the generic XFree86-style module loading facilities, or ask the nvidia people if it's from their driver... [02:10:06] <astinus> Triskelios: Okay, I'm actually loading SX back onto the system now (since Ubuntu lacks ZFS, which I desperately want) so I'll have error messages in Pastebin shortly I expect :) [02:11:30] *** uebayasi has joined #opensolaris [02:13:46] *** stevel has quit IRC [02:19:05] *** sahafeez has quit IRC [02:19:38] *** tek7 has quit IRC [02:32:40] *** gisburn has left #opensolaris [02:33:07] *** Risky has quit IRC [02:33:18] *** esaxe has quit IRC [02:34:17] *** Risky has joined #opensolaris [02:36:26] <kimc> dclarke: are you around ? [02:41:39] *** Risky has quit IRC [02:43:19] *** gm152 has joined #opensolaris [02:54:21] <dclarke> I just got in [02:54:27] <dclarke> hello [02:56:24] <jamesd_> dclarke, how many more years in university do you have till your qualified to do every canadian's job [02:56:29] <jamesd_> zamboni driver [02:56:45] <jamesd_> ;-p [02:56:54] *** yongsun has joined #opensolaris [02:57:33] <dclarke> I'm not a doctor yet [02:57:47] <dclarke> but I can sharpen pencils [02:57:54] <dclarke> and sweep floors [02:58:07] <dclarke> but not at the same time [02:58:27] <jamesd_> dclarke, it was a joke :-) [02:58:42] <dclarke> oh [02:59:36] <dclarke> # isainfo -v [02:59:36] <dclarke> 32-bit sparc applications [02:59:49] <dclarke> I can't recall how to force the 64-bit kernel to boot [02:59:56] <dclarke> I think its in a boot.conf somewhere [02:59:58] <jamesd_> dclarke, man boot [03:00:07] <dclarke> or /etc/system [03:00:16] <dclarke> no no .. the OBP can forst that too [03:00:21] <jamesd_> i think its boot.conf [03:00:29] <dclarke> I'll set the 64-bit kernel to boot by default [03:02:17] <dclarke> on the other hand .. this is a good machine for testing with [03:02:24] <dclarke> its a 32-bit Solarsi 8 box [03:02:36] <dclarke> sort of the lowest common denominator [03:03:38] <dclarke> power failuer here [03:03:41] <dclarke> damn [03:04:08] <dclarke> freaking Canadian power systems [03:04:25] <dclarke> this .. this is why we buy 300 pound UPS battery packs [03:04:36] <astinus> lol [03:04:43] <jamesd_> and $200 snow shovels [03:05:23] <astinus> and $20,000 heated toilet seats for un-freezing your arse, which froze the moment your pants dropped. [03:06:21] <dclarke> real Canadians can step out to an out house in February and still admire the stars and norther sky .. at 50 below [03:06:52] <jamesd_> and do so in their underwear without a single shiver or chill [03:07:08] <astinus> real canadians sound like they need a room with soft walls, and a person in white clothing to feed them brightly coloured blue and orange sweeties [03:07:11] <dclarke> barefoot [03:07:24] <boyd> All work and no play.... [03:07:25] *** yongsun has quit IRC [03:07:34] <dclarke> I think I'll go roll out that picture of my front yard again [03:07:46] <dclarke> that generally sets the record straight [03:08:18] <boyd> dclarke: You sound like a swede. [03:08:39] <dclarke> http://www.blastwave.org/dclarke/early_snow_640x480.jpg [03:08:50] <dclarke> the forest line is about 200 feet away [03:08:57] <dclarke> that was taken at 5 AM or so [03:09:07] <boyd> Looks nice. Kinda Surreal [03:09:30] <astinus> That's the sorta place you need to drop one 'o them new blackboxes [03:09:32] <jamesd_> http://photos1.blogger.com/blogger/6362/604/1600/IMG_0519.0.jpg <-- my friend clearing off the sat dish to watch a hockey game on t.v. [03:09:41] <dclarke> http://www.blastwave.org/dclarke/cold_feet.jpg [03:09:52] <dclarke> I took those photos .. barefoot at 30 below [03:09:53] <astinus> Sod hooking up water, you'll need to add heating to your mobile datacentre or the processors will get frostbite :S [03:10:19] <astinus> dclarke: Do you still have toes? [03:10:20] <boyd> That's just nuts, dclarke :) [03:10:29] *** yongsun has joined #opensolaris [03:10:37] <astinus> (and .. I need to ask this .. but is that fake snow and the picture some kinda stunt? :P) [03:11:08] <dclarke> no .. its the real thing [03:11:13] <astinus> :o [03:11:16] <dclarke> I just wandered out to take a picture [03:11:25] <boyd> If I took it it would be fake:) [03:11:25] <astinus> dang :S [03:11:32] <dclarke> and sorta .. looked dow and thought .. hey .. let's take a picture of my toes [03:11:44] <boyd> Of course you did. [03:11:45] <dclarke> then I went in to get coffee [03:11:46] <astinus> dclarke: btw, I really liked your "how to build non-debug opensolaris" article - thank you. [03:11:50] * astinus just read it [03:11:53] <dclarke> it helps ? [03:12:00] <dclarke> nice and clear ? [03:12:05] <astinus> well .. I'm new to Solaris, but I understood it :) [03:12:14] <astinus> going to have to test it out soon. [03:12:16] <dclarke> perfect !! [03:12:28] <dclarke> I am in process of writing a SAMP document also [03:12:43] <dclarke> Solaris + Apache + MySQL + PHP5 [03:12:44] <kimc> dclarke: output of the test is at http://pastebin.co.uk/4511 [03:12:50] <astinus> be nice if you had a section covering 'live upgrade' type foo .. so perhaps getting your build onto a different partition, booting off that, so if your BFU goes wrong you're not fscked [03:12:53] <dclarke> but I will add in PostgreSQL [03:13:06] <dclarke> good point .. [03:13:07] <astinus> dclarke: Could you add in Subversion + dav_svn too? :P [03:13:12] * dclarke reaches for a post it note [03:13:23] * dclarke scribbles [03:13:36] <astinus> dclarke: and .. I dunno if Blastwave has Trac packages [03:13:43] <astinus> but there's 6bn dependencies involved with Trac [03:13:59] <dclarke> no problem [03:14:15] <dclarke> $ uptime [03:14:15] <dclarke> 9:12pm up 3:08, 1 user, load average: 0.00, 0.03, 0.21 [03:14:16] <astinus> ... I'm a developer and can't live without my Trac + SVN .. I'm moving from Linux -> Solaris so I have ZFS and other nice stuff .. it'd be cool if I could 'easily' work out how to do Apache, PHP, SQL, Subversion + Trac [03:14:17] <dclarke> dammit [03:14:39] <dclarke> astinus : I need feedback from you .. so please .. drop me an email ? [03:14:48] <kimc> dclarke: the crucible test is up eh ? [03:14:49] <boyd> Interesting. ~3 weeks after Sun employ the JRuby guys, MS are employing The RubyCLR guy [03:14:51] <dclarke> to dclarke at blastwave dot org [03:14:52] <astinus> (ie: if that lot worked, I've got time to learn the rest slowly .. if that lot doesn't work, I can't do my job and am screwed) [03:14:55] *** solaris-user has joined #opensolaris [03:14:57] <astinus> dclarke: Sure :D [03:15:07] <dclarke> thanks man [03:15:16] *** solaris-user is now known as pablo-costa [03:15:18] <dclarke> my Domino server just packed it in .. down she goes [03:15:19] <astinus> you're welcome, those kinda guides save my ass in more ways than one [03:15:22] <dclarke> dammit ! [03:15:33] <kimc> cu all.. /leave [03:15:38] *** kimc has left #opensolaris [03:16:27] <astinus> dclarke: Just an idea, but gentoo-wiki.org is a really useful resource (I'm a gentoo dev, but .. y'know .. can't know everything) .. some kinda OpenSolaris wiki type thing would be nice .. I don't see one already (at least, not one that's good) and if your guides were wiki'fied people could update/improve them [03:16:52] <dclarke> I'm working on it [03:17:01] <astinus> cool :) [03:17:03] <twincest> astinus: genunix.org/wiki/ [03:17:06] <dclarke> today is a big day here [03:17:17] <dclarke> its the Blastwave birthday [03:17:18] <astinus> twincest: ooer, thanks. looking now [03:17:28] <dclarke> and of course .. to rain on my parade .. power failure [03:17:36] <twincest> i don't know how useful it is right now, but it's there at least :) [03:17:39] <boyd> Happy Birthday Blastwave! [03:17:56] <dclarke> 22 Oct 2002 Blastwave.org went live [03:17:59] <astinus> Aye, Happy Birthday Blastwave :) [03:18:01] <dclarke> still here [03:18:20] <dclarke> despite all manner of problems [03:18:30] <astinus> dclarke: Are you, per chance, familiar with Subversion? (if so, this might be an easy question) [03:19:25] <dclarke> seen svn.blastwave.org ? [03:19:28] <astinus> dclarke: the university here has a pair of V240's driving Rays .. I need a Subversion client to be available to 9-10 users, I don't admin the V240s .. Presumably I could drop the Blastwave stuff into ~/bin for those users, edit their PATH, and be done with it? [03:19:34] <dclarke> or polaris.blastwave.org ? [03:19:46] <astinus> <3 Trac!!!! [03:19:52] <astinus> dclarke++ [03:19:54] <dclarke> ta da ! [03:20:08] <dclarke> like dude ! I love Trac + subversion [03:21:13] <astinus> Okay .. well I'm a student at University of Wales, Aberystwyth (www.aber.ac.uk) and they say we gotta use 'group fileshare' for a big project in our 2nd year (CS degree) which rather blows .. no accountability, they don't use ZFS, and their incremental backups -> SAN are only every 1hr [03:21:53] <boyd> You could put an SVN or HG repos on the group fileshare :) [03:22:02] <twincest> or a zfs pool ;-) [03:22:09] <dclarke> sorry .. I'd love to hang out here .. but I need to get my power situation sorted out [03:22:18] <boyd> Sure. [03:22:26] <astinus> twincest: Hm? ZFS pool on the fileshare? [03:22:33] <dclarke> then .. I announce the Blastwave birthday on the site and start the 2006 - 2007 fund raiser [03:22:35] * astinus is curious [03:22:45] <twincest> astinus: zpools can be backed by normal files [03:22:54] <astinus> twincest: and you don't need any extra roles to make a pool? [03:23:08] <twincest> oh, you don't have root? then that won't work [03:23:11] * astinus shakes his head [03:23:16] <astinus> I don't have root on the V240s, nope [03:23:34] <astinus> Rock <-- astinus --> Hard Place [03:23:52] <boyd> HG, dude. [03:23:59] * boyd goes back to teaching [03:24:14] <astinus> boyd: I can't get anything new loaded on the V240s, best I can do is something in ~/bin [03:24:33] <twincest> when i use other people's systems i usually end up with a large ~/local :) [03:24:37] <astinus> boyd: Academic year has started so they freeze the software on the servers [03:25:00] <boyd> That'd be fine for hg [03:25:35] * astinus thinks he needs to read up on Hg [03:26:35] <astinus> oh, did I mention... I need whatever version control system we use to have a Windows client, preferably with GUI, that can be installed/run without admin rights :S [03:31:35] <Error_404> bzr works too [03:31:45] <Error_404> as for gui, write one [03:32:00] <Error_404> bzr is *VERY* slow, though [03:33:46] <Error_404> as in "set up a cronjob to push/pull changes for midnight, maybe it'll be done when you come in to work at 9" slow [03:34:04] <astinus> owch [03:34:11] <astinus> that bad? [03:34:26] <Error_404> on occasion, yeah [03:34:36] <Error_404> usually it's just a 1 or 2 hour job, ime [03:34:42] <twincest> for how large of a repository? [03:35:02] <Error_404> 100 megs or so [03:35:25] *** gm152 has quit IRC [03:35:40] <Stric> eh, sounds like it has serious issues.. [03:36:32] * delewis considers converting his personal repository from CVS to Hg [03:50:55] *** echelog has joined #opensolaris [03:53:21] *** azr|el has quit IRC [04:04:03] <clee> delewis: hg is neat, as is bzr, as is git [04:04:08] * clee has been playing with all of them [04:04:15] <clee> CVS is so last-century. :) [04:11:07] *** silly_girl22 has joined #opensolaris [04:12:38] *** Kernel86 has joined #OpenSolaris [04:24:38] *** pablo-costa has quit IRC [04:32:27] <rodrickbrown> does 11 do iscsi initiator & target yet? [04:33:13] *** gm152 has joined #opensolaris [04:48:08] *** laca has quit IRC [04:53:14] <delewis> rodrickbrown: it has for awhile [04:53:35] <delewis> you can even do iscsi target on 10 if you install the package [04:53:42] <delewis> (though, I highly doubt it's supported) :-) [04:57:08] <rodrickbrown> delewis, I've been wondering how feesible it might be to use iscsi for a simulated san? [04:57:17] <rodrickbrown> to get VCS here at home w/o all the HW [04:57:38] *** kilohertz has joined #opensolaris [04:57:45] <rodrickbrown> basically have 2 hosts mount share an iscsi lun [04:57:48] <rodrickbrown> and mount it :-) [04:58:16] <rodrickbrown> not sure if its even possible but would save me so much time/effort/hw/electricity if It could be done [04:59:13] <delewis> rodrickbrown: that's exactly what I did to play around with VxVM [04:59:22] <delewis> it's certainly possible [04:59:24] *** kilohertz has quit IRC [05:00:48] <rodrickbrown> so you've done this? [05:01:00] <delewis> yes, I also played around with ZFS and the AIX LVM [05:01:19] <delewis> at the time I had terrible performance with iSCSI target in Solaris, so I had to use a Linux target and Solaris/AIX initiator [05:01:33] <delewis> and the performance was still terrible, as I only have a 10/100 LAN [05:01:45] <rodrickbrown> wow really [05:01:48] *** sahafeez has joined #opensolaris [05:02:01] <rodrickbrown> i'm shocked that linux had better performance :-) [05:02:02] <astinus> dclarke: Hm, gonna bounce you some feedback in a moment. After I'm more comfy with Solaris, we should sit down and talk, I'd be interested in maintaining some packages for blastwave.org :) [05:02:15] <rodrickbrown> I guess they've been getting free R&D from NetApp [05:02:27] <delewis> rodrickbrown: Linux has had iSCSI target support *much* longer (at least since the 2.4 days), and there's been a lot of corporate backing [05:02:30] <delewis> namely by Cisco [05:03:13] *** silly_girl22 has quit IRC [05:06:34] <sahafeez> strange question - i am using a sun box that lets you use a pc keyboard/mouse on a old style (type5/6) system. i take it there would be no way to get my scroll wheel working on the mouse. [05:07:23] <rodrickbrown> delewis, can a single host be iniitator and target ? [05:07:33] <rodrickbrown> thinking if I can just use 2 nodes for this vcs cluster [05:08:55] <rodrickbrown> nm [05:09:06] <delewis> rodrickbrown: I don't see why not [05:09:09] <rodrickbrown> if I failed the box hosting the shared storage everything ill go lol [05:09:19] <rodrickbrown> so ill need 3 boxes minimum [05:09:59] <delewis> sahafeez: there's really no reason why the scrollwheel wouldn't work [05:10:06] <delewis> you just need to re-configure your X11 server [05:10:43] <delewis> a signal is a signal is a signal from a button from a button pressed on the mouse that the computer receives, whether that be the first, second, third, or nth mouse button. [05:10:50] <sahafeez> well, the thing is that i am using a pc mouse thru a device that lets the sun think it is a sun crossbow or type5 mouse so.. [05:11:32] <delewis> sahafeez: the Type7 mouse has a scrollwheel -- would this not work with that device (USB/whatever is irrelevant)? [05:11:57] <sahafeez> ok, did the type 7 just work out of the box with the wheel. [05:12:18] <delewis> sahafeez: Sun wouldn't include a scrollwheel if it didn't work :-) [05:14:16] <sahafeez> ok, then i think that it will not work becuase if it should work out of the box and does not then the device is not sending the signales thru. heck, did the old style din port even support the signalling > [05:14:21] <delewis> sahafeez: the only problem I can see is if that device didn't map signals 4 and 5 [05:14:30] <delewis> which is scroll up and scroll down (respectively) [05:14:46] <delewis> signal 3 would be mapped (as this is the middle button) [05:15:02] <sahafeez> ah, ok. how do i see what is on the bus at this point when i scroll. maybe the adapter is doing something funny. [05:15:18] <delewis> sahafeez: no clue, but I think it'll end up being some debugger-fu [05:15:55] <sahafeez> no letmeseethestuffonthekeyboardbus command then. ... ;) [05:16:04] <delewis> doubtful :-) [05:16:15] <delewis> you'll need to access the structures in the mouse driver (whatever that may be) [05:16:24] <delewis> I'm being overly general here, but I think the generalization is valid [05:17:35] <delewis> or you could try and use DTrace to detect the interrupt that's generated when the nth mouse button is pressed [05:17:45] <sahafeez> ok. to much trouble for now. i have to get my obsd box a new kernel with raid support - dumbing down my office to 2 boxes - ultra 80 and obsd firewall/nas [05:18:09] <astinus> Damnit :X [05:18:10] <sahafeez> hum. dtrace on an irq. how? or is there a url. that sounds cool. [05:18:11] <astinus> (EE) NVIDIA(1): Failed to initialize the NVIDIA graphics device! [05:18:11] <astinus> (EE) NVIDIA(1): *** Aborting *** [05:18:24] <delewis> sahafeez: are you familiar with intrstat? [05:18:25] <astinus> Getting that for NVIDIA(1) and NVIDIA(2) which means no multi-head :/ [05:18:28] <delewis> it's just a DTrace consumer [05:18:34] <sahafeez> no. sorry. not really a prger [05:18:37] <delewis> (it simply detects the number of interrupts) [05:18:48] <delewis> similar to trapstat [05:19:00] <delewis> which detects the number of traps (and which traps) that are being generated [05:19:05] <delewis> again, just a simple DTrace consumer [05:19:43] <sahafeez> cool. gives me a place to start. i will ask some of the prger here later...thanks! [05:21:40] <astinus> Weird .. nVidia is suggesting I issue a 'reboot -- -r' for some reason [05:22:11] <boyd> because they don't know about touch /reconfigure? [05:22:34] * boyd wonders why brandz targetted RHEL3, not 4 [05:22:51] <astinus> boyd: touch /reconfigure, then reboot? [05:22:55] <twincest> astinus: have you done a reconfiguration boot since you installed the driver? [05:23:00] <astinus> twincest: I have not. [05:23:02] <boyd> astinus: Yeah [05:23:18] <twincest> you need to (although the driver should really tell you to do that when you install it) [05:23:37] <delewis> wouldn't a simple cfgadm do the trick without rebooting? [05:23:52] <delewis> (which is all touch /configure or -r do anyway, I think) [05:23:53] <twincest> i'm not sure how much cfgadm does on x86 [05:23:53] <boyd> Never worked for me with Nvidia [05:23:57] <boyd> (cfgadm) [05:23:59] <delewis> boyd: ah [05:24:00] <astinus> Okay, that didn't fix my issue [05:24:02] <delewis> there must be more to it then [05:24:15] <twincest> astinus: do you know the pci id of the card? [05:24:32] <astinus> twincest: Yeah, its specified in the xorg.conf [05:24:38] <astinus> twincest: I've got Xorg.0.log here too [05:24:54] <twincest> find it in prtconf, does it say '(driver not attached)'? [05:25:29] <astinus> ooh, I don't know -that- ID of the card [05:25:44] <astinus> one of my GPUs is PCI:1:0:0, the other is PCI:5:0:0 [05:25:45] <twincest> /usr/X11/bin/scanpci will show it [05:25:58] <twincest> or is it /usr/X/bin? one of those [05:26:15] <astinus> its only showing one card [05:26:30] <astinus> actualy... wait [05:26:46] <astinus> pci bus 0x0001 cardnum 0x00 function 0x00: vendor 0x10de device 0x0163 [05:26:46] <astinus> nVidia Corporation GeForce 6200 LE [05:26:58] <astinus> pci bus 0x0005 cardnum 0x00 function 0x00: vendor 0x10de device 0x0161 [05:26:58] <astinus> nVidia Corporation GeForce 6200 TurboCache(TM) [05:27:01] <astinus> there we go :) [05:27:11] <twincest> okay, so what does prtconf say about those? [05:27:50] <astinus> twincest: the numbers are quite different, ie: pci15d9,5680 (driver not attached) [05:27:59] <twincest> that's not the same device then [05:28:08] <twincest> there's no entry for pci10de,0161 ? [05:28:15] <astinus> looking [05:28:24] <twincest> sorry, it'll be ,161 [05:28:26] <boyd> don't forget compatibles... [05:28:45] <twincest> fwiw, older versions of the nvidia driver did not add non-quadro PCI IDs to driver_aliases [05:28:50] <twincest> but i thought they'd changed that [05:29:00] <boyd> So did I... but I had to do it myself recently [05:29:03] <astinus> I don't see anything in prtconf [05:29:21] <twincest> astinus: grep 10de,161 /etc/driver_aliases [05:29:34] <astinus> nvidia "pci10de,161" [05:29:42] <twincest> hmm. [05:29:54] <twincest> prtconf|grep 10de,161 shows nothing? [05:30:07] <astinus> nvidia "pci10de,161" [05:30:21] <astinus> second doesn't return anything... [05:30:32] <twincest> prtconf|grep pci10de [05:30:47] <astinus> twincest: returns nothing [05:30:52] <twincest> hm that's odd [05:31:11] <astinus> http://www.rafb.net/paste/results/YgMjK720.html [05:31:14] <astinus> prtconf [05:31:35] <twincest> well try this anyway: add_drv -i '"pci10de,163"' nvidia [05:31:36] <astinus> from the grep of driver_aliases, it looks like my second card lacks an entry [05:31:39] <twincest> (note the odd quotes) [05:31:55] <astinus> (nvidia) already in use as a driver or alias. [05:32:04] <astinus> (can't it be used for multiple devices? [05:32:20] <astinus> i need the nvidia driver to work for two cards :S [05:33:00] <twincest> oh, add_drv -f [05:33:05] <twincest> (-f -i ...) [05:33:16] <astinus> same error [05:33:19] <astinus> already in use :/ [05:33:44] <astinus> same [05:33:48] <twincest> hm [05:33:53] <twincest> that's not what the manual page says :P [05:34:16] <astinus> (nvidia) already in use as a driver or alias. [05:34:17] <twincest> you could try editing driver_aliases manually and rebooting but it might break something [05:34:24] <sahafeez> rhythmbox broke in 49/vermillion50? just dies on startup here.. [05:34:39] <sahafeez> ls [05:35:10] <astinus> twincest: rebooting now [05:35:32] <twincest> you'll want boot -r [05:35:39] <astinus> ack, already sent it down [05:35:49] <astinus> I'll grab GRUB on reboot [05:36:42] <astinus> this problem = weird :( [05:37:09] <astinus> so in addition to the kernel/module lines, I want a boot -r on the end? [05:39:33] * delewis thinks onnv-notify is out of sync [05:39:58] <delewis> just got a putback message for "lx zone install sometimes hangs (fix lint)", but accorind to hg log, that was putback on the 21st [05:40:44] <astinus> twincest: *sigh* same errors [05:44:08] <astinus> trying one more thing... [05:46:21] <astinus> SON OF A! [05:46:26] * astinus sends twincest beer [05:46:31] * astinus sends twincest women [05:46:37] <astinus> you are a STAR! [05:47:07] *** sahafeez has quit IRC [05:47:12] <astinus> manually editing the file to tell it to use the driver, then doing 'reboot -- -r' worked [05:47:27] * astinus cheers [05:47:39] <astinus> omg, the 3-day ordeal of having one display device is OVER [05:49:15] *** uebayasi has quit IRC [05:51:33] <astinus> now to write to nVidia and request that get added to some kinda FAQ [05:56:52] <astinus> twincest: If you're ever in Wales, come see me, beer awaits you. [05:57:12] <twincest> hmm, i'm close, but not quite enough even with free beer :( [05:57:26] <astinus> UK? [05:57:31] <twincest> oxford [05:57:39] <astinus> oh cool, another Brit :) [05:57:44] <astinus> I'm in Aberystwyth [05:58:39] *** sahafeez has joined #opensolaris [05:59:32] <sahafeez> question - i am on 49 and want to go to 50. my dvd burner is on the fritz. simplest way to do it without burning anything? [05:59:40] <twincest> live upgrade [05:59:53] <astinus> thats what I need to learn about next :) [05:59:57] <sahafeez> does it require an unused disk [06:00:10] <twincest> you need a new root slice, doesn't have to be a whole disk [06:00:19] <sahafeez> hum. that sucks. not setup for that. [06:00:25] <twincest> LU is essential for SX users unless you're rich or really like burning dvds :) [06:00:36] <bank> hello [06:00:47] <twincest> saha: network install? [06:01:00] <sahafeez> hum. interesting. no jumpstart box. [06:01:22] <astinus> twincest: I have a 73GB scsi drive, which is mirrored by my Adaptec controller with another .. it appears to be native to Solaris, so I just see the once disk .. I set aside 20GB (each) for two root slices, 4GB for swap, and the remainder is /export/home -- presumably thats LU compatible? [06:01:51] <twincest> astinus: i would ditch the hardware raid, use SVM and temporarily detach the mirror to upgrade [06:01:58] <bank> I lost my connection with server second time. [06:02:01] <twincest> but some people dislike that in case the drive breaks during the upgrade [06:02:07] <astinus> twincest: *nod* [06:02:28] <bank> after refreshing whole system with b50. I didn't do anything except zpool create and zfs create [06:02:40] <sahafeez> anyone here running vermillion 50. can you tell me if rhythmbox breaks on startup for you? [06:03:01] <bank> this is the summmary what I am done .. http://paste.lisp.org/display/28516 [06:03:17] <sahafeez> 7hr and 18mins for parity re-write. fun. [06:03:39] <twincest> astinus: if you don't want to do that, then what you have seems fine [06:04:30] *** gm152 has quit IRC [06:05:06] <bank> does the problem realy come from zfs.? [06:05:13] <twincest> maybe some day there'll be a "zfs swap" command for live upgrade :) [06:06:06] <bank> ... [06:06:46] <Error_404> twincest: my swap is on zfs [06:06:47] <sahafeez> so if i want to be upgrade proof i should / and /altroot on the boot drive? [06:06:56] <twincest> error: no, i mean a command to swap two filesystems [06:07:01] <Error_404> oh [06:07:08] <twincest> error: so you could upgrade into /a, 'zfs swap / /a', reboot, zfs destroy /a [06:07:12] <Error_404> well, you could always just change mountpoints [06:07:27] <twincest> does that work with zfs root? [06:07:42] <Error_404> considering zfs doesn't boot without some hacking, no [06:07:43] <sahafeez> root zfs in build 50?!?! [06:07:50] <Error_404> sahafeez: heh, you wish [06:08:31] <bank> Error_404: yesterday I refresh the system after the grub with blink cursor. but after I do te zpool , zfs create then restart the machine. then I lost the connection to server again. [06:08:50] <bank> Does I done anything wrong with pool ,zfs creation ? > http://paste.lisp.org/display/28516 [06:09:03] <Error_404> bank: dunno what to tell you man... not without physical access to the machine [06:09:37] <bank> Error_404: Did you take a look at http://paste.lisp.org/display/28516 ? [06:09:44] <bank> Do you think I done anything wrong? [06:09:48] *** sahafeez has left #opensolaris [06:10:13] <Error_404> beats me, it's only a screen dump [06:10:16] <bank> I quite sure . because I didn't do anything except. useradd . and umount , zpool create , zfs create. [06:10:20] <Error_404> why are you using "zfs add" anyways? [06:10:53] <bank> I create severals slices. [06:10:56] <Error_404> instead of just zpool create foo c1d0s4 c1d0s5 [06:10:59] <bank> and I make sure it is not root. [06:11:04] <bank> oh. [06:11:05] <Error_404> get it all in one shit [06:11:06] *** ShadowHntr has quit IRC [06:11:07] <Error_404> shot [06:11:08] <Error_404> heh [06:11:09] <bank> oh. [06:11:31] <bank> that great. [06:12:12] <bank> by the way,. [06:12:29] <bank> from the summary .... you don' think i done anything wrong right? [06:12:38] <bank> :( I fear to use zfs again. [06:12:49] <Error_404> unless you gave your swap slice to zfs, i dunno [06:12:57] <Error_404> hard to tell what's wrong with the machine [06:13:28] <Error_404> actually, for that matter... why do you have 2 partitions on the same drive in a zpool? why not just make the partition bigger? [06:13:42] <astinus> twincest: Okay, I have a 8-port SATA controller [06:13:51] <astinus> twincest: assume disks are connected to ports 1,3,7,8 [06:14:14] <astinus> twincest: If I put them into a raidz, then the ports get changed, presumably ZFS is intelligent enough to not scream and lose its marbles? [06:14:34] <twincest> yes, it uses the disk id to locate devices, not bus id [06:14:38] <bank> I zfs presentation show we can create pool seemlessly from several disk , slices so I would like to test. [06:14:49] <bank> another is . when I do lucreate , luupgrade in future. [06:15:18] <bank> :( I dunno why , It's very bad after 1 hr of installation. [06:15:30] <bank> and back home, fire some zfs , zpool command then reboot [06:15:34] <bank> and lost connection. [06:16:00] <Error_404> oh, right... and why do you rent server space for a test server? why not just have it at your feet? [06:16:21] <Error_404> that one's been bugging me since last night [06:17:58] <bank> it's hard to tell. I never do something like this. I just would like to try. and I don't think ... there will be an error [06:18:26] <bank> I still don't know what's wrong [06:18:39] <Error_404> i really have no idea [06:18:41] <bank> will the sysstem can boot with useradd? I guess no [06:18:49] <Error_404> ? [06:19:46] <bank> I think I remember that I use only several commands. useradd , df - h , zpool create , umount -f , vi /etc/vfstab , zfs create [06:19:57] <bank> and then after all . I test by "reboot" [06:20:00] <bank> reboot command [06:20:28] <bank> :( testing on vmware , will have a same result? [06:20:36] <Error_404> well, if it's just sitting @ the grub prompt obviously it can't find your kernel [06:20:45] <Error_404> or it's own config files [06:22:22] <Error_404> so blame (in this order): the harddrive, something to do with your partition table, the harddrive controller, the machine's BIOS [06:24:16] <bank> zfs :( [06:24:32] <bank> I hurt. [06:24:33] <Error_404> unlikely [06:24:58] <Error_404> an isolated incident with such a hugely distributed piece of code? it's more likely your machine [06:26:23] <bank> ^o) [06:30:36] <Error_404> speaking of colo space, i wonder when mod3's gonna get back to me with pricing & configuration advice [06:32:46] <bank> error_404 [06:32:52] <Error_404> ? [06:33:02] <bank> insert belenix cd , at root@belenix [06:33:09] <bank> then what should I inspect. [06:33:31] <bank> I didn't want to recovery . just would like to have an information to tell you. [06:33:39] <Error_404> 'mount' to see if it's mounted your root directory [06:33:47] <bank> then I refresh, and never touch zfs anymore. [06:33:48] <Error_404> if so, check your grub.conf [06:34:03] <Error_404> it's got nothing to do with zfs... [06:34:23] <bank> 'mount' ? instead of df -h ? [06:34:26] <Error_404> if it was zfs svc:/system/filesystem/local would spit up, but the machine'd boot at least past grub.conf [06:35:14] <Error_404> 'mount' will tell you everything that's mounted [06:35:24] <Error_404> df -h is a lot quieter [06:36:53] <bank> if so check grub.conf [06:37:14] <bank> if it was zfs svc:/system/filesystem/local IN GRUB.CONF? [06:37:23] <Error_404> what? no... [06:37:45] <bank> what do you mean ? please describe it a little more. [06:37:57] *** Xh4 has joined #opensolaris [06:38:06] <bank> after at root@belenix, then I 'mount' to see root [06:38:38] <bank> to see if it's mounted my root directory [06:38:43] <bank> if so [06:38:52] <Error_404> i mean look at if your root filesystem mounts, if so, go to it's mount point, and check boot/grub/menu.lst [06:39:28] <bank> check it? [06:39:33] <bank> how? .. vi ? [06:39:40] <Error_404> or less [06:39:41] <Error_404> or cat [06:39:54] <bank> then what I going to do withing this file [06:39:57] <bank> *within [06:40:01] <Xh4> vi <3 [06:40:07] <bank> what I am looking for? [06:40:43] <Error_404> man grub & man bootadm [06:40:47] <Error_404> they'll tell you what to do [06:41:00] <Error_404> you just need to make sure everything's kosher [06:42:37] <bank> ... [06:44:56] <bank> I guess it advance topic [06:45:19] <Error_404> no, i'm still convinced it's hardware [06:45:42] <bank> hardware?! [06:45:47] <bank> x2100!? [06:46:03] <Error_404> yes, even quality hardware sometimes flakes out [06:46:08] <Error_404> call sun [06:46:15] <Error_404> it's probably a bad drive or somethign [06:46:24] <bank> how could I sure that it is hardware problem. [06:47:53] <bank> :( everything begun hard after throwing window life. [06:48:05] <bank> I will go there [06:48:06] <bank> bye [06:48:33] *** bank has quit IRC [06:51:28] <astinus> its probably an X2100 (not an M2) [06:51:35] <astinus> i thik those had quite a few issues [06:53:41] <Error_404> doesn't suprise me considering they were literally just over $700 [06:54:09] <astinus> my mate works in a Sun Lab in the UK [06:54:13] <astinus> ends up servicing some of the faulty stuff [06:54:22] <astinus> hah, he used to whine so much about the pre-M2 chassis [06:56:21] *** sahafeez has joined #opensolaris [06:58:25] <sahafeez> ok i am re-doing my system so that i can do live updates. i just need 2 roots right? [06:58:50] <astinus> swap + two root slices, at a minimum for LU [06:58:51] <sahafeez> so if i setup / /atlroot /usr /opt /tmp /var swap i am ok for doing a live update? [07:00:29] *** sparc-kly_ has joined #opensolaris [07:01:23] <astinus> Error_404: you fairly familiar with ZFS by any chance? [07:01:33] *** bank__ has joined #opensolaris [07:01:45] <Error_404> define "fairly familiar"? [07:01:46] <bank__> Error404: [07:01:52] <bank__> it is x2100 bios! [07:02:07] <bank__> I search on google. [07:02:14] <bank__> from your opinion it might right. [07:02:32] <bank__> x86: BIOS Hangs on Reboot After Using ZFS on Sun Ultra 20 or Sun Fire X2100 (6363449) [07:02:39] <Error_404> there you go [07:02:44] <Error_404> i told you it was hardware [07:02:46] <bank__> Sun FireTM X2100 server might hang on reboot if disks connected these systems to contain a ZFS storage pool. [07:02:55] <bank__> this is exactly me? [07:03:19] <astinus> Error_404: Okay, I have five * 250GB drives currently on a hardware RAID controller and running ext3. I have two more 250GB drives arriving later today - is there any way I can define a degraded RAID-Z array using the two 250GB drives, so I've got a bit of extra space/slack to help with the migration, then add the 5 drives ~6 hours later? [07:03:22] <Error_404> yes [07:03:49] <Error_404> astinus: no [07:03:52] <bank__> Error404: he said : Disconnect the disks that are used by ZFS before rebooting the system until the minimum BIOS level that supports ZFS on these systems can be installed [07:04:01] <bank__> so I have to pull SATA hdd out right? [07:04:02] <astinus> So I need minimum three drives to start with? [07:04:14] <Error_404> astinus: for now, and for raidz yes [07:04:18] * astinus nods [07:04:19] <astinus> Okay, thanks :) [07:04:25] <bank__> then I update bios.... [07:04:44] <Error_404> bank__: no, since you don't care about the machine, you have a choice: pull the drive, or reformat the machine [07:04:45] <bank__> does that sentence mean to pull SATA HDD out from machine? [07:05:01] <Error_404> yes [07:05:04] <bank__> no no , he told me we can fix. [07:05:09] <bank__> http://docs.sun.com/app/docs/doc/819-5947/6n80cmc6l?a=view [07:05:22] <bank__> sound like pull HDD out update BIOS. [07:05:54] <bank__> can I across Supplemental 1.4 to Supplemental 1.5? [07:06:16] <Error_404> bank__: you can remove the HDD from the machine. but since it's a fresh install, you can also just reformat the machine, run the update, and then play with zfs [07:06:43] <Error_404> up to you [07:07:15] <bank__> I guess I update bios is not too hard. [07:07:27] <bank__> I mean, I hope update bios is not hard. [07:07:35] <Error_404> it wont be [07:08:14] <bank__> if I know it is error specifically for x2100 then . I will not refresh the system :( :( I should search google and found this [07:08:18] <bank__> :'( very bad [07:08:46] <bank__> :| good course. [07:08:49] <Error_404> it happens [07:09:00] *** sparc-kly has quit IRC [07:09:24] <Error_404> at least sun fixes the problem [07:10:08] <Error_404> my motherboard manufacturer doesn't... my machine is pretty fragile at the moment... i have to set the BIOS to not scan the SATA controller for drives or else it won't boot [07:11:55] <bank__> I will run away from hardware and OS problem. [07:12:25] <bank__> :P [07:14:04] *** djgregor has joined #opensolaris [07:17:37] *** bougie has joined #opensolaris [07:19:23] *** bank has joined #opensolaris [07:21:35] <bank> Error_404: How're you doing? [07:21:46] <Error_404> s'alright [07:21:57] <Error_404> got linear algebra in the morning [07:22:02] <Error_404> at the moment, teaching myself java [07:22:02] *** aiki_kid has joined #opensolaris [07:22:07] <bank> :D [07:22:09] <bank> java [07:22:10] <astinus> masochist [07:22:26] <Error_404> astinus: regardless, people will pay me money if i know java [07:22:38] <aiki_kid> how can i release the audio device in solaris? [07:22:41] <astinus> i know :P we have to use Java on my CS course [07:22:59] <Error_404> we use C++ on solaris9 at my school [07:23:11] <astinus> weird combo [07:23:25] <bank> C++ on unix is hard. [07:23:34] <bank> no Vector , no String. [07:23:45] <Error_404> bank: ? [07:23:49] <Error_404> char * foo; [07:24:37] <bank> Last friday, I try to convert c++.net 2003 proj to something like .. make (I not sure what is that) [07:24:49] <bank> the old code compose of Vector and String everywhere. [07:25:01] <bank> :( [07:25:02] <Error_404> then your old code used specialized libraries [07:25:11] <Error_404> has nothing to do with windows and everything to do with shitty code [07:25:11] *** jamesd_ has quit IRC [07:25:16] <bank> yes I guess .net [07:25:28] *** jamesd_ has joined #opensolaris [07:25:35] *** aiki_kid has quit IRC [07:25:47] <bank> Error_404: Are there any replacement to Vector? [07:25:52] <bank> Vector<String> [07:25:57] <Error_404> yes, char* [07:25:58] <astinus> eugh [07:25:59] <astinus> generics [07:26:04] <Error_404> just a simple pointer [07:26:08] <bank> oh.. [07:26:22] <bank> make existing bad code more complicated. [07:26:38] <Error_404> solution: don't write bad code in the first place [07:27:08] <astinus> lol [07:27:13] <astinus> easier said than done [07:27:17] <bank> Error_404:May I ask one question? , corba or web services to integrate legacy c++ to java application. [07:27:20] <astinus> esp. when you're learning [07:27:33] <bank> astinus:exactly [07:27:38] <Error_404> astinus: that's why a lot of schools teach java... [07:27:40] <bank> painful. [07:27:45] <Error_404> it's slightly harder to write crap code [07:27:58] <Error_404> or rather, your crap code doesn't explode so spectacularly [07:27:59] <astinus> Error_404: hah, you've not seen some of our first years with Swing :P [07:28:29] <astinus> one week you teach them recursion [07:28:32] <bank> let them use kind of ve or matisse [07:28:36] <astinus> the next week is swing fundamentals [07:28:36] <Error_404> though it leaves you with the problem of a JVM that eats most of your ram [07:28:48] <astinus> suprising how many try to combine, resulting in really screwed up code [07:28:59] <Error_404> a properly written J2EE application is currently taking up 512M of my ram [07:29:09] <astinus> bank: the point isn't to teach them to use an IDE, its to teach them how to code, and GUIs do that quite well [07:29:10] <Error_404> and it's nothing more complicated than a blog [07:29:27] <bank> astinus: :D [07:29:30] <astinus> (as in, writing Swing teaches good object orientation) [07:29:42] <bank> I know. I hate blue unmodifiable code. [07:29:58] <bank> blog? [07:30:07] *** sahafeez has quit IRC [07:30:07] <bank> why note roller. [07:30:12] <bank> why not roller. [07:30:25] <Error_404> *shrug* [07:30:42] <Error_404> pebble's easier to use [07:30:47] <Error_404> & works out of the box [07:30:53] <bank> really. it is hard to manage propertychange and fireproperty event that flow around one panel to another panel [07:31:04] <bank> pebble? [07:31:21] <bank> wow [07:31:29] <bank> never know that before. [07:31:30] <Error_404> http://pebble.sourceforge.net/ [07:31:33] <Error_404> it's nice, i like it [07:31:59] *** sahafeez has joined #opensolaris [07:33:15] *** bank__ has quit IRC [07:33:38] <bank> wow [07:33:48] <bank> it's look good. [07:34:16] <bank> thank you. [07:34:21] <Error_404> np [07:34:45] <bank> i think roller don't have image captcha and knowledge cloud [07:35:00] <Error_404> roller's ugly most of the time [07:35:04] <bank> lol [07:35:09] <bank> it is use by ibm and sun? [07:35:12] <bank> I ever heard. [07:35:19] <Error_404> i've not played enough with it to see if you can make it better, but so far, it's ugly [07:35:20] <bank> I felt it. [07:35:32] <astinus> *shrug* ibm and sun can afford to customize it [07:35:37] <bank> what is the ugly things. [07:35:43] <Error_404> even jon's blog looks very... 1998 [07:35:48] <bank> lol [07:35:49] <Error_404> and he's the CEO [07:35:52] <bank> lol [07:36:07] <astinus> Error_404: it got a makeover about 4 months ago tho [07:36:32] <Error_404> i'm referring to the current look [07:38:03] <bank> does jon = jonathan. [07:38:05] *** Xh4 has quit IRC [07:38:24] <bank> who wear eyeglasses and long hair. [07:38:37] <Error_404> bank: yes [07:38:45] <Error_404> the CEO of sun microsystems [07:39:13] <bank> I ever see him in sun sdn podcast. he speak very fast and very marketing man [07:39:29] <Error_404> i didn't get that feeling from him [07:39:39] <Error_404> not as much as steve jobs anyways [07:39:46] <Error_404> or gates [07:39:53] <astinus> *grIN* [07:39:58] <bank> no no marketing sentences. [07:39:58] <astinus> i'd say.. [07:40:03] <bank> I like it. [07:40:04] <astinus> you have to be a marketing person, to be CEO of sun [07:40:13] <bank> :P [07:40:22] <astinus> after all, you need to meet with big clients and sell them stuff, no? [07:40:35] <Error_404> yeah [07:41:37] <bank> I agree 1998 with this picture , http://blogs.sun.com/roller/resources/jonathan/HP_Big_Medium.JPG [07:41:51] <astinus> hah, thats -funny- [07:41:52] <Error_404> bank: i meant his blog layout [07:41:54] <Error_404> looks very old [07:42:02] <Error_404> like a website from 1998 [07:42:04] <astinus> giving HP a new home :P [07:42:24] <bank> what do you expected from 2007 [07:42:36] <Error_404> http://web.archive.org/web/19981206003227/www.wired.com/news/ [07:42:46] <bank> web 2 jsf and ajax. [07:42:46] <Error_404> that's what wired.com looked like in 1998 [07:42:53] <Error_404> http://blogs.sun.com/jonathan [07:43:02] <Error_404> that's what jon's blog looks like right now [07:44:09] <bank> I agree. wired.com [07:44:37] <bank> oh not wired.com just archieve.org [07:44:45] <Error_404> yes, it's archive.com [07:44:58] <Error_404> that's a snapshot of wired.com from 1998 [07:45:27] <bank> lol I got it. [07:46:39] <bank> I cheer for opensolaris podcast. [07:47:13] <bank> it will be great. [07:48:28] <bank> also java irc chat. [07:48:57] <Error_404> on itunes? [07:49:22] <astinus> java IRC chat != new [07:49:26] <delewis> bank: Freenode provides a Java IRC client. [07:49:34] <delewis> it is not the responsibility of the channel to provide one [07:49:37] <delewis> only the network [07:49:38] <astinus> delewis: we do? [07:49:43] <delewis> astinus: if not, they should. [07:49:46] <delewis> s/we/they/ [07:49:56] *** tsoome has quit IRC [07:50:10] <astinus> heh, verbal slip. up until last month I was staff :P [07:50:19] <delewis> ah :-) [07:50:46] <bank> no no i mean channel [07:50:47] <delewis> though, for some reason, I can't see the majority of Freenode users being interested in a Java IRC client, given the demographics, though, I could be wrong. [07:50:57] <bank> on sun website like opensolaris. [07:51:00] <bank> may be java.net [07:51:03] <astinus> delewis: freenode never provided one because it was always open to abuse [07:51:13] <delewis> bank: and I said, it is not the responsibility of an individual Freenode channel to provide a Java IRC client. [07:51:38] <astinus> delewis: It was left down to individual groups/channels, on the basis that should it be abused, they're in a position to deal with that - or we can simply ban their CGI:IRC gateway... only affects them, not the network [07:51:48] <delewis> astinus: ugh. [07:52:11] <astinus> seriously, the amount of kiddies who wreak havoc on here every day .. push the responsibility down the chain where possible :P [07:52:29] *** bougie has quit IRC [07:52:54] <bank> delewis: how do you think about opensolaris podcast [07:52:58] <delewis> well, it stands to reason that anyone using OpenSolaris should be able to build an IRC client (or at the very least install a pre-build package) [07:53:10] <delewis> bank: I don't like the term "podcast" :-) [07:53:45] <astinus> delewis: I concur, and I think your assumption about the general freenode user was correct - by and large, if you're using Linux, you can apt-get, yum install or rpm -ivh a client of some sort :P [07:53:58] <delewis> though, it remains to be seen what subject material a "podcast" could be created to promote/demonstrate. [07:54:13] <astinus> i would say ZFS, except thats more a general Solaris thing [07:54:23] <delewis> lots of things are fairly documented, or users are taking it upon themselves to document/podcast, etc. [07:54:36] <Error_404> delewis: is the opensolaris netcast (heh, like that) gonna be on itunes? [07:55:19] <delewis> well, I think the appropriate thing to do (which would also require the least work on behalf of the OpenSolaris community) is to put together a centralized repository of links to user-created podcasts/HOWTOs [07:55:29] <delewis> the genunix.org wiki would be perfect for such a thing [07:55:44] <astinus> anyone see the thumper promotion? [07:56:07] <astinus> make a funny video, win a 12TB thumper [07:56:15] <delewis> heh [07:56:21] <delewis> I wonder how "funny" it has to be ;-) [07:56:35] <bank> sometime a formal english is hard to understand. [07:56:41] <astinus> for a 30k thumper, I'd say pretty f*cking funny :P [07:57:38] <delewis> bank: hey, the one language I speak, I at least speak it somewhat fluently, though I envy those who can speak who are familiar with multiple languages (and speak at least one or more fluently) :-) [07:58:08] <delewis> and I'm in dire need of coffee this late :-( [07:58:38] <delewis> sifting through the UltraSPARC-I/-II MMU specifications with no caffiene at midnight isn't very exhilerating. [07:58:44] <astinus> lol :P [07:59:50] <delewis> brb, time to reboot for the new BFU [07:59:52] *** delewis has quit IRC [08:01:00] <bank> hard to describe. but I don't think there are a english course that appropriate for me in this time. [08:01:12] <bank> *at this time. [08:02:45] *** bank has quit IRC [08:04:15] *** qbit has joined #opensolaris [08:07:15] *** tsoome has joined #opensolaris [08:19:38] *** gnu2it2 has joined #opensolaris [08:20:17] *** gnu2it2 has quit IRC [08:29:34] *** delewis has joined #opensolaris [08:29:43] <delewis> $ uname -a [08:29:44] <delewis> SunOS galileo.network.lan 5.11 on_build_20061022 sun4u sparc SUNW,Sun-Blade-1000 [08:29:49] <astinus> :D [08:29:55] <delewis> $ echo ${.sh.version} [08:29:55] <delewis> Version M 1993-12-28 s- [08:29:59] <delewis> gisburn would be proud :-) [08:31:07] * delewis wonders if later JDS versions than in nv48 detect removable media properly after the removal of vold for Tamarack [08:31:28] <delewis> $ echo ${.sh.version} [08:31:36] <delewis> er, annoying paste buffer. [08:32:01] *** PosixCompliant has joined #opensolaris [08:32:04] <astinus> my lovely fileserver migration goes well :D [08:32:24] <astinus> currently NFS'ing stuff about my network [08:32:29] <astinus> getting about 450Mbit which is reasonable I guess. [08:32:32] <delewis> hey, I did one of those a couple of months ago ;-) [08:32:38] <delewis> when ZFS hit Solaris 10 (6/06) [08:32:49] <astinus> i'm getting rid of Loonix off my fileservers [08:32:56] <astinus> in favour of ZFS :P [08:33:01] <delewis> I didn't have that much data to migrate only 45GB, but I still had to cpio it all up, and copy it to another system with enough diskspace [08:33:08] <delewis> I was using UFS [08:33:11] <astinus> I have 1.8TB [08:33:15] <delewis> ouch. [08:33:23] <astinus> so juggling the disks around is a right f*cker [08:33:29] <delewis> see, all my media is burnt to CD/DVDs :-) [08:33:44] <astinus> ever try burning >2TB (i deleted a bunch) to DVD? :P [08:33:55] <delewis> the only thing I have on there really is my music and download repository [08:33:57] *** gnu2it2 has joined #opensolaris [08:34:21] <astinus> music, tv, movies, my photo collection, work, subversion repos, wwwroot [08:34:22] <delewis> I have this nasty habit of keeping everything I download [08:34:32] <astinus> I used to do that, then I hit 4.3TB [08:34:39] <astinus> and couldn't afford more disks, so I learnt to not do it. [08:34:49] * astinus is a student [08:34:51] <delewis> $ ls /imports/downloads/software | wc -l [08:34:52] <delewis> 116 [08:35:12] <astinus> of 1.8TB, about 1.1TB is TV [08:35:20] *** gnu2it2 has quit IRC [08:35:29] <astinus> so ZFS compression doesn't do jack. :? [08:35:39] <delewis> which includes directories like 'azureus', 'cvs', 'postgresql', 'wmaker', AIX updates, TSM updates, 'gnupg', etc. etc. etc. (basically, lots of tarballs) [08:35:42] <delewis> astinus: sure it does [08:35:53] <astinus> does it take time to compress? ie: do it in the background? [08:35:55] <delewis> $ zfs get all export | grep ratio [08:35:55] <delewis> export compressratio 1.09x - [08:36:02] <astinus> because at the moment, my compressratio is 1.00x [08:36:03] <delewis> astinus: it compresses as you're writing [08:36:08] *** gnu2it2 has joined #opensolaris [08:36:18] <delewis> astinus: it's not *really* 1.00x [08:36:21] <delewis> there's round-off error [08:36:26] <delewis> and if you have lots of binary files [08:36:30] <delewis> that's always going to be approaching one [08:36:43] <astinus> these are very tightly encoded TV rips [08:36:54] <delewis> well, of course, you can't compress those very well ;-) [08:37:03] <astinus> 300-350MB per 45 minute show [08:37:16] <astinus> which is four-pass ffMpeg [08:37:49] <astinus> my rip/encode box has 3 TV cards :) [08:38:01] <astinus> dualie dual-core opteron .. works wonders :D [08:38:07] <delewis> ZFS compression is also not ad hoc, meaning, if you start writing to ZFS pool without compression and you later enable compression, what you've already written won't be compressed. [08:38:37] <astinus> Ah, I enabled it from the start [08:38:46] <delewis> astinus: envy. I've been re-encoding my Star Trek DVD collection on my SB1000 (2x750MHz UltraSPARC-IIIs with 8MB cache/processor) and it takes forever. [08:38:57] <delewis> FFmpeg doesn't have a lot of SPARC optimisation [08:38:59] <astinus> I've got 5 * 200GB drives in a pool, no RAID Z for now [08:39:08] <astinus> thats for backing up stuff [08:39:17] <astinus> I've ordered two new 250GBs [08:39:30] <astinus> hoping to be done backing up my larger array today... [08:39:38] <astinus> thats 6 * 250GB in RAID-5 [08:40:02] <astinus> then I can drop all 8 * 250GB into the new fileserver, RAID Z it .. and move most of the stuff off the 200s [08:40:30] <astinus> then I need to move my smaller array of 200GB drives, thats the remaining "issue" [08:40:36] <Triskelios> that's way more storage than our lab... [08:40:59] <astinus> should be able to yank the 550-600GB off and put them on these 200's I'm using right now... [08:41:15] <astinus> Then I'm "hoping" to somehow put 8 * 200GB into the chassis, and RAID Z that too :X [08:41:21] * astinus snickers [08:41:47] <astinus> Which -should- give me about 3.2TB usable, split across 16 drives and two arrays. [08:42:49] <LeftWing> delewis: What series have you got? [08:42:57] <astinus> delewis: If I had a fatter upstream I'd give you some FTP access, I have 250GB of Star Trek .. thats TNG S1-7, VOY S1-7, DS9 S1-7, ENT S1-4 [08:43:44] * astinus sighs [08:44:07] *** mazon is now known as Mazon [08:44:09] <LeftWing> I ordered TNG S1-7 the other day, they were AU$36/season. =D [08:44:25] <tsoome> hm, basically all star trek? (besides original first one) [08:44:25] <astinus> "Get that boy off my bridge!" [08:44:41] <LeftWing> Frigging Wesley. [08:44:44] <astinus> tsoome: Yeah, I have TOS on DVD - been meaning to rip it. [08:45:47] <tsoome> I had to remove them to make some space:( [08:45:48] *** Risky has joined #opensolaris [08:46:16] <astinus> I'm hopeful that with ZFS and RAID-Z, I should be "safe" for another year or two [08:46:32] <astinus> I know at some point I really need to get a whole new array, but thats so expensive [08:47:40] [08:48:12] <delewis> astinus: I already have all of those :-) [08:48:24] <delewis> in fact, I've got ENTS1-5 *and* Babylon 5 S1-S5 [08:48:33] <delewis> they're all burnt to DVD, though [08:48:38] <astinus> ENT S5? [08:48:39] <astinus> :o [08:48:43] <astinus> I wasn't aware that existed. [08:48:47] <delewis> oh :-) [08:48:56] <delewis> 4 seasons seems too short for some reason, but you're right. [08:48:57] <astinus> omfg, amazon .. where are jooooooo .. I need DVDs *now*! [08:49:17] <astinus> ooer [08:49:18] * LeftWing diverts power to commerce. [08:49:20] <astinus> Stargate SG1 [08:49:28] * delewis isn't a Stargate fan [08:49:29] <astinus> they're on S10 now :o [08:49:35] <delewis> I've been re-watching TNG episodes the last few days [08:49:39] * astinus nods [08:49:42] <delewis> particuarly S7 [08:49:43] <astinus> know that feeling :) I'm on DS9 [08:49:54] <delewis> I've seen every DS9 episode about 10 times :-( [08:49:55] <astinus> sisko nailing jem hadar in the Defiant is good watching :P [08:50:00] <delewis> especially, the Dominion War battles -- at least 15. [08:50:13] <astinus> hah, I still marvel at the graphics of the time [08:50:18] <delewis> yeah :-( [08:50:20] <astinus> some of those jem hadar cruisers were exquisite [08:50:24] <delewis> I remember the first large Dominion War battle [08:50:32] <astinus> when about 400 ships closed on DS9? [08:50:37] <delewis> you know, the one where the Klingons don't sign-on at first, and they're trying to re-take DS9 [08:50:40] <astinus> yeah [08:50:42] <delewis> then 3/4s of the way through [08:50:45] <astinus> and they decloak [08:50:47] <delewis> the Klingons just come blasting through [08:50:50] <delewis> blowing shit away [08:50:51] <astinus> and its like "Honey, we're HOOOOOOOOOOOME!" [08:50:56] <astinus> heheheh [08:51:02] <delewis> god, I was in awe [08:51:07] <astinus> aye, that brings back memories [08:51:12] <astinus> time to fire up that episode methinks [08:51:18] <delewis> hehe [08:51:22] * astinus flicks on his projector [08:52:00] <astinus> delewis: If you're ever in Wales, come see me, ever watched it in 220" wall sized with surround sound and battle-rumble coming off a 12" subwoofer? :P [08:52:13] <delewis> nice, however, I'm a bit far from Wales :-) [08:52:20] <astinus> :( shame [08:52:23] <delewis> Jackson, TN, USA :-( [08:52:46] <astinus> I don't know USA very well.. I have friends in Mountain View, CA and Kansas [08:52:58] <sahafeez> why the fuck does every other version of sxcr break rhythmbox. [08:53:03] <g4lt-mordant> toto, I don't think we're in kansas anymore....' [08:53:10] <delewis> sahafeez: because it can :-) [08:53:16] <delewis> SXCR is meant to break things [08:53:42] <g4lt-mordant> and fortunately, breaking rhythmbox isn't hard [08:53:48] <delewis> that's one of the down-sides of being on the cutting edge, I suppose [08:53:49] <sahafeez> it is sick. it is the one app that i really need besides the web and terminal. every fuck version. and it breaks on start. [08:53:58] <delewis> though, on the upside, that gives *you* plenty of stuff to fix :-) [08:54:04] <delewis> in hopes of getting Hg commit access :-) [08:54:09] <astinus> sahafeez: Grab it from Blastwave [08:54:45] <sahafeez> does the blastwave version support aac [08:54:56] <astinus> probably, but I don't know for sure [08:55:01] <astinus> TIAS [08:55:01] <delewis> every package in Blastwave supports everything. [08:55:11] <astinus> I fixed my nVidia graphics issue earlier. [08:55:19] <astinus> twincest is a f*cking GENIUS [08:55:23] <sahafeez> oh, does anyone else have the icon for startoffice go awol from the menu in every version of jds after the 1st login [08:55:23] <delewis> if Rhythmbox can support AAC, I can assure you that they've built it with that support. [08:55:34] <delewis> sahafeez: interesting [08:55:38] <delewis> I just noticed that myself :-) [08:55:47] * delewis still wishes .Trash worked on Solaris/SPARC [08:56:10] <delewis> I think the JDS team has decided that I'm crazy for pointing out that it works fine on Solaris/x86 but not Solaris/SPARC [08:56:20] <astinus> RAWR! So much for giving R/W access to buddies with Windows computers [08:56:25] <sahafeez> i am telling you as soon as they get xorg fucking working on sparc again (breaks every day) i an thinking about going back to linux [08:56:30] <astinus> my directories are full of *RETARDED* thumbs.db files [08:56:36] <delewis> sahafeez: evil. [08:56:41] <delewis> SPARC on Linux? [08:56:45] <delewis> are you crazy? :-) [08:57:01] <delewis> astinus: don't forget desktop.ini's :-) [08:57:04] <astinus> Gentoo/SPARC :P [08:57:09] * delewis chokes [08:57:10] <Error_404> heh, do it man... graphics will be the least of your problems [08:57:13] <sahafeez> gentoo-sparc baby. if you can get xorg working and a type 5 then it is great. good luck getting it working [08:57:14] <boyd> Haha [08:57:28] <delewis> sahafeez: I tried doing a Gentoo install on my SB1000 time [08:57:38] <delewis> I concluded I would be dead before I was able to get it working for my particular tastes. [08:57:44] <astinus> delewis: a mate of mine went through manually deleting about 500 thumbs.db and desktop.ini files on his box [08:57:47] <boyd> hehe [08:57:50] <delewis> install Solaris, *poof*, it just works on my SB1000 [08:57:55] <sahafeez> i am used to solaris just working you know. 2.6...2.9 but 10 is fucking insane as a desktop [08:57:56] <astinus> delewis: he was *gutted* when I showed him a find one-liner [08:58:00] <delewis> with Gentoo, you've got to build a kernel (with USB support, I might add for the keyboard and mouse) [08:58:05] <delewis> install 50,000 packages [08:58:06] <delewis> to get it working [08:58:09] <delewis> and pray shit doesn't break. [08:58:12] * astinus coughs and points at his cloak [08:58:22] <delewis> and then you realize that you have *0* observability tools [08:58:26] <delewis> a piss-poor development environment [08:58:28] <Error_404> gentoo is such obnoxious garbage [08:58:30] <sahafeez> i get it working just fine on every sparc i have - it is just that x is broked [08:58:32] <delewis> and every single thing has been compiled with gcc [08:58:36] <delewis> which implies [08:58:43] * dclarke coughs and continues work on his Solarsi Desktop [08:58:45] <delewis> every single thing has sub-optimal performance [08:58:46] <astinus> a tonne of security issues. [08:58:49] <delewis> need I say more? [08:58:57] <delewis> or did I touch all the high points? [08:58:58] <g4lt-mordant> deather_, you frogot about writing off the smartcard.... [08:59:06] <g4lt-mordant> delewis, ^^^^^^^^^6 [08:59:08] <delewis> g4lt-mordant: oh, that's impossible! :-) [08:59:12] <astinus> delewis: You missed the fact that emerge --sync takes 2-3 hours to regen metadata sometimes :P [08:59:19] <delewis> *sigh* [08:59:22] <sahafeez> security --- whatever. i have an openbsd box infront of everything. no issues. [08:59:25] <delewis> things work so nicely with Solaris [08:59:45] <astinus> If Solaris would run on my laptop, I'd remove Gentoo, until then I'm stuffed. [08:59:54] <sahafeez> with execption of the desktop and everything else you need to run for a desktop. ;) [08:59:56] * delewis hopes Solaris runs on his Ferrari ;-) [09:00:06] <delewis> sahafeez: Solaris works fine on my desktop [09:00:09] <delewis> even on my SB1000 [09:00:23] <astinus> I also know Gentoo very well .. been a package maintainer for ~2 years .. better the devil you know 'n all that, I prefer the flexibility to something like Red Hat [09:00:31] <delewis> then again, I'm also the MPlayer/Solaris maintainer, so I *make* things work :-) [09:00:38] <g4lt-mordant> delewis, I'd hope so, since they used ferraris to develop sol/X86 [09:00:38] <sahafeez> make rhythmbox work on my u80 for me. oh, and make it so i do not have to tell the sdtaudiotool to use the line out every boot [09:00:43] <delewis> g4lt-mordant: well [09:00:47] <delewis> the Ferrari 5000 is fairly new [09:00:54] <twincest> i heard the newer ferraris aren't quite so well supported [09:00:55] <delewis> most of the developers have Ferrari 3000/4000s [09:00:58] <twincest> but i imagine it's being worked on [09:01:04] <sahafeez> still...i am a solaris guy for the most part ;) [09:01:09] <delewis> if it isn't, that just gives me a reason for a couple of putbacks :-) [09:01:13] *** bengtf__ has joined #opensolaris [09:01:16] <sahafeez> just a bit pissed off right now. [09:01:17] *** bengtf__ is now known as bengtf [09:02:00] * delewis still wishes he could've gone the SPARC portable route [09:02:13] <Error_404> $3000 later... [09:02:17] <delewis> yeah :-) [09:02:33] <boyd> And a 2.3 minute battery life.... :) [09:02:34] <delewis> of course, this just gives me a reason to get acquainted with AMD64. [09:02:37] <sahafeez> can someone tell me why /etc/dt/config/Xservers does not work in 49 [09:02:38] *** uebayasi has joined #opensolaris [09:02:45] <delewis> which actually has a very SPARC-like virtual memory lay-out. [09:03:05] <delewis> boyd: and you were right about the hole in the AMD64 VM space -- it's only 48 bits. [09:03:16] <boyd> Wow... [09:03:28] <boyd> I'd given up on my memory :) [09:03:39] <delewis> well, I was pretty surprised myself [09:03:44] <delewis> SI2 didn't mention anything about it [09:03:50] <sahafeez> do you have to totally kill dt to have it read it if you just created it? log out in does not seem to work [09:03:54] <delewis> nor did the AMD64 manual I happened to be reading [09:03:58] <astinus> delewis: fwiw, all my boxes are AMD64 or EM64T and they're generally nice .. I used to have a pair of U60s and a SB1000 too .. while the SPARCs were good, they can't hold a candle to a dual dual-core 2.6GHz opteron [09:03:59] <delewis> though, it just defines the AMD64 spec [09:04:02] <delewis> and not an implementation [09:04:05] <boyd> sahafeez: I'd think so [09:04:12] * boyd nods [09:04:25] <delewis> astinus: true, but SPARC is far more elegant :-) [09:04:38] <delewis> especially in regards to register windows and MMU design [09:04:41] <astinus> delewis: I'm a break down the door and beat shit up guy :P [09:04:48] <delewis> x86 is an ugly beast in that regard with segment registers, etc. etc. [09:04:52] <delewis> and all sorts of other nasty things [09:04:58] * astinus nods [09:04:59] <sahafeez> is dt started via svc [09:05:02] <boyd> nasty nasty... [09:05:09] <astinus> sahafeez: yes [09:05:26] <boyd> sahafeez: /usr/dt/bin/dtconfig -reset May do it [09:05:29] <astinus> delewis: I -really- like svcadm under Solaris [09:05:32] <astinus> delewis: its what Linux *should* have [09:05:42] <delewis> and all the 286 compatibility that's been incorporated sickens me [09:05:48] <boyd> I'm told that initng does much of the same type of thing [09:05:51] <delewis> especially regarding the LDT/GDT descriptors [09:05:53] <sahafeez> hum. anyway to get a list off all processes startup/down under svc [09:05:54] <boyd> (as smf) [09:06:00] <astinus> boyd: pfft, its immature and breaks a lot [09:06:01] <Error_404> astinus: if solaris offers all the features you wish linux has... [09:06:06] <delewis> astinus: it's what every UNIX should have [09:06:07] <delewis> well [09:06:08] <astinus> boyd: upstart is mildly decent, but thats Ubuntu specific [09:06:09] <Error_404> wouldn't the solution be... use solaris instead? [09:06:12] <delewis> to be fair, AIX had something similar years ago. [09:06:20] <astinus> Error_404: what do you think I'm doing to my fileserver :P [09:06:27] <delewis> it wasn't quite like smf [09:06:28] <astinus> Error_404: and wishing I could do to el laptop [09:06:30] <sahafeez> that worked boyd thanks. [09:06:34] <delewis> and only solved a few of the problems with init [09:06:44] <delewis> namely dependencies [09:06:47] <delewis> but smf does more than that [09:06:53] <Error_404> astinus: can't help you... works fine on my lappy [09:06:58] <delewis> so the comparison between what AIX does and smf isn't quite so valid. [09:07:03] <astinus> I have a Samsung Q35 [09:07:12] <sahafeez> i like gentoo's start things. simpile to understand. rc-config add bla level [09:07:13] <Error_404> i have an acer $400 special [09:07:23] <astinus> one of the nasty things about my laptop, ipw3945 wifi [09:07:30] <astinus> which definitely isn't supported under Solaris yet [09:07:32] <sahafeez> one of the issues with going to this xml stuff is it is hard to read in vi [09:07:35] <boyd> I like some of the launchd features.... scheduled sevices (a la cron), per-user services when they log on/off, and services that run on file/directory change [09:07:39] <delewis> sahafeez: that's just a wrapper around the grotesque init scripts [09:07:44] <boyd> sahafeez: You got that right [09:07:47] <delewis> it still doesn't take care of dependency management [09:07:48] <tsoome> and it's not really fair to compare latest amd/intel with some 5-7 year old sparc.... [09:08:01] <delewis> all rc-add does is copy (or symlink) the init.d script to a particular run-level [09:08:06] <delewis> nothing you couldn't accomplish with a single command. [09:08:13] <Error_404> sahafeez: vim... [09:08:15] <Error_404> :syn on [09:08:16] <sahafeez> delewis: yes it does. there is a define for each items depenency IIRC [09:08:19] <astinus> delewis: not quite [09:08:19] <Error_404> *nod* [09:08:32] <astinus> delewis: rc-update add *does* symlink stuff [09:08:34] <sahafeez> Error_404 thanks i will try that [09:08:36] <delewis> tsoome: regardless of speed, what's ironic is SPARCv8 is still far more elegant than AMD64 (which still has all the compatibility cruft) [09:08:45] <astinus> delewis: but the init scripts themselves are very intelligent, and dependency aware [09:08:47] <delewis> and they *finally* got the idea to drop hardware segment management [09:08:48] <sahafeez> launchd is nice [09:08:50] <delewis> after.. [09:08:55] <delewis> 20 years [09:09:12] <LeftWing> astinus: They're only as intelligent as their creators. ;P [09:09:23] <tsoome> delewis: thats true;) [09:09:38] <sahafeez> ugh. hate this type 6. need to find a new type 5 on ebay [09:09:40] <delewis> long mode on AMD64 leaves segment management up to the OS [09:09:41] <astinus> LeftWing: http://www.gentoo-portage.com/AJAX/Ebuild/34013/View [09:09:46] <astinus> LeftWing: there's an ebuild :S [09:09:48] <delewis> (which is the way it should've been) [09:09:49] <delewis> :-) [09:09:57] * astinus notes those should have been XML'd already [09:10:52] * delewis wants a Type 7 [09:10:57] * astinus too [09:10:57] * LeftWing does too. [09:11:06] <astinus> preferably cheap/free [09:11:11] <delewis> I don't the Type 6 is *that* bad [09:11:13] <delewis> it could be better [09:11:14] <astinus> may try to blag one at LWE [09:11:16] <astinus> they sometimes give freebies [09:11:23] <Error_404> astinus: trust me, gentoo's init scripts are *NOT* intelligent [09:11:24] <delewis> think* [09:11:24] <LeftWing> I have ... four Type 6's I believe. [09:11:31] <delewis> LeftWing: :-) [09:11:33] <boyd> My type6 spits on the other POS that I had before it [09:11:38] <astinus> Error_404: more so than slackwares :P [09:11:44] * delewis has one for his only SPARC system (well, minus the SPARCstation 5 sitting in the closet) [09:11:46] <Error_404> last time i used gentoo, it wouldn't associate with a wireless network [09:12:05] <tsoome> init scripts can't be intelligent by definition:D [09:12:05] <Error_404> because of some retarded bug that still exists & is documented but nobody'll get off their ass & fix it [09:12:21] <Error_404> because it's linux, so bugs don't get fixed [09:12:24] <astinus> Error_404: whats the bug with? Gentoo or wpa_supplicant? [09:12:25] <boyd> What's so much better about a type 7? OLED keycaps or something? :) [09:12:26] <Error_404> it's not sexy to fix bugs [09:12:36] <Error_404> astinus: gentoo. [09:12:40] <delewis> boyd: it's basically a Type 5 USB [09:12:41] <delewis> from what I know [09:12:46] <astinus> Error_404: find me a bit of info about the bug, and I'll fix it. [09:13:04] <delewis> and the Type 7 mouse has a scrollwheel [09:13:07] <delewis> which I *hate* [09:13:15] * delewis wonders when Sun became a peecee mouse vendor [09:13:21] <sahafeez> type 5 and ibm type m. the only real keyboards [09:13:24] <astinus> I can't live without a scrollwheel [09:13:25] <delewis> UNIX workstations should have a three-button mouse, period. [09:13:32] <delewis> astinus: evil! :-) [09:13:58] <astinus> delewis: I'm partially disabled, my Logitech MX-1000 makes a world of difference to me, and even with a type7 I'd only want the keyboard [09:14:02] <LeftWing> Scrollwheels are handy, but I enjoy lauding the three-button rhetoric over Windows admins I work with. =) [09:14:03] <Error_404> astinus: bug #143698 [09:14:04] <sahafeez> i wonder if i stuck an usb card in my u80 (there are pc ones that work) and plugged in t7 mouse if it would work [09:14:23] <delewis> I just use pgup and pgdn for scroll capability :-) [09:14:42] *** Gman has joined #opensolaris [09:14:42] <boyd> Ok, I'm having brainfade... remind me what is the diff type6 -> type5? [09:14:49] <delewis> I often find that most scrollwheels have very poorly implemented mouse wheels that make using button 3 very difficult (as the wheel slides, easily) [09:15:02] <delewis> boyd: Type 6 is the plastic (very plastic) USB keyboard [09:15:03] *** Cyrille has joined #opensolaris [09:15:18] <sahafeez> i use my ibook 95% of the time. one button...as steve intended it.. [09:15:19] <boyd> delewis: Not *all* usb... this one isn't [09:15:22] <astinus> Error_404: hrm, ok [09:15:35] <delewis> Type 5 is also plastic (but higher quality, and uses actual springs for keys) but uses the regular Sun keyboard-style connector [09:15:36] <sahafeez> hate this type 6. going to buy a type 5 tommorrow on ebay [09:15:37] <LeftWing> sahafeez: As Xerox PARC intended it. [09:15:39] <delewis> boyd: really? I thougt they were [09:15:44] <delewis> Type 6s were only sold with USB systems [09:15:51] <sahafeez> no, parc was 3 button [09:15:52] <delewis> SB100, SB150, SB1000 [09:15:55] <sahafeez> or was it 2 [09:15:58] <sahafeez> but it was not 1 [09:16:00] <delewis> which didn't have the Sun keyboard/mouse connector [09:16:04] <LeftWing> sahafeez: I believe at least one of their designs was one button. [09:16:16] * boyd turns his kb over. 83tarduhsvsv Model: Type 6 .... plugged into an ultra 10. [09:16:21] <LeftWing> sahafeez: After all, they had craploads of money to throw at building stuff. [09:16:30] <delewis> boyd: ugh. [09:16:40] <delewis> I thought Sun had confined that keyboard only to those of us who were unfortunate enough to have USB only systems [09:16:47] <boyd> It's not USB... it daisychains the mouse [09:16:51] <delewis> though, the Type 7 takes care of that, as it is basically a USB Type 5 [09:16:57] <delewis> boyd: yeah. [09:17:10] [09:17:13] <astinus> delewis: ouch :P [09:17:18] <boyd> hmm... so you're saying you liked the square-looking thing? [09:17:23] <delewis> boyd: well [09:17:36] <boyd> (not for the look but for the feel?) [09:17:36] <delewis> more precisely, I liked the *feel* of the square-looking thing. [09:17:43] <delewis> the Type 6 uses rubber feedbacks under the keys [09:17:44] * boyd nods [09:17:45] <sahafeez> lots of type 5s on ebay. not w/mice tho.. [09:17:50] <sahafeez> $2.99 + 12 shipping [09:17:57] <delewis> which just aren't solid enough for me, given I still use an IBM Model M "clicky" [09:18:09] <delewis> the Type 5 had spring feedbacks [09:18:23] <tsoome> I really cant get why people like those clicks.... ;) [09:18:25] <boyd> I'll have to try the type7 out when my 13 U20M2s and 13 U45s arrive [09:18:34] <delewis> tsoome: it makes you feel more productive :-) [09:18:41] <LeftWing> boyd: Trade you a Type 6 for a Type 7? =P [09:18:52] * delewis stabs LeftWing [09:18:53] <delewis> :-) [09:18:56] <sahafeez> i have a type m and was using it 5 mins ago with my u80. switch back to the type 5 becasue of the converter i had did not work well as far as the 3 button trackball [09:18:59] <delewis> you con. [09:19:12] <LeftWing> delewis: If he's got 26 surely he can spare one. ;D [09:19:21] <boyd> Hehe [09:19:24] <delewis> oh [09:19:25] <boyd> Not mine to give :( [09:19:27] <Error_404> model M = best keyboard evar [09:19:28] <delewis> hey, send me one, too ;-) [09:19:42] <delewis> Error_404: my only complaint about the Model M is that isn't UNIX-style [09:19:47] <delewis> and not just the control key [09:19:48] <boyd> Hey, what's zone state "down" mean? [09:19:52] <delewis> you're missing 10 keys :-) [09:19:53] <sahafeez> yah. i do not know if i said this but when i worked at ibm is was a $110 part. [09:20:07] <sahafeez> but it was green tag which ment you did not have to account for it. [09:20:08] <delewis> sahafeez: along with the $300 terminal. [09:20:20] * delewis still has his 3153 terminal [09:20:32] <delewis> best. serial console. ever. [09:20:33] <sahafeez> hum that would not be a type m keyboard. [09:20:42] <sahafeez> but still spring key [09:20:59] <astinus> my mate got given a crate of IBM keyboards, the old school ones [09:21:11] <astinus> works for them, said he could dispose of them [09:21:18] [09:21:27] <tsoome> solaris documentation quality is going down after they started to integrate freeware:( [09:21:35] <delewis> http://riemann.solnetworks.net/~dlewis/images/pics/infowindow3153/infowindow3153_2.jpg [09:22:12] <astinus> delewis: :D [09:22:35] <sahafeez> 15 pounds is cheap. have a look http://www.clickykeyboards.com/index.cfm/fa/categories.main/parentcat/9230 [09:22:46] <delewis> an IBM terminal using an IBM keyboard hooked to an IBM pSeries p640 :-) [09:22:56] <delewis> I've got the whole IBM setup going :-) [09:23:07] <astinus> p640 a current model? [09:23:19] <delewis> circa 1999-2003 [09:23:24] <astinus> I know they make some -real- fast Power5 stuff nowadays [09:23:32] * LeftWing wonders if he can get AU$600 for an E450 with 4*480, 4GB, 3PSU + 2 spare PSU, 2 drive backplanes [09:23:36] <delewis> http://riemann.solnetworks.net/~dlewis/inventory/7026-B80.html [09:24:31] <astinus> nice! [09:24:58] <astinus> LeftWing: $600aud? :o [09:25:03] <sahafeez> there. just bought a type 5 key/mouse for $27 shipped. [09:25:08] [09:25:15] <LeftWing> astinus: 500 what? [09:25:19] <astinus> GBP [09:25:24] <LeftWing> Isn't that like [09:25:27] <LeftWing> 5.3 billion AU? [09:25:28] <LeftWing> =P [09:25:36] <astinus> about 1400AUD i think [09:25:43] <LeftWing> Sweet mother of I need to sell that. [09:26:07] <astinus> ie: I saw a SB2000 go on ebay for ~1550 yesterday [09:26:20] <LeftWing> 1550 GBP? [09:26:23] <astinus> and a SB1000 for ~700 [09:26:24] <astinus> yeah [09:26:25] <sahafeez> fuck just buy it on ebay.com and pay the shipping. [09:26:42] <LeftWing> I don't want to buy it, I want to sell it. =P [09:26:52] <astinus> LeftWing: advertise on ebay.co.uk :P [09:26:53] <sahafeez> i just had some guy in the EU pay me $15 for a elite3d and $26 to ship [09:27:11] <astinus> UNIX stuff is a ripoff over here (for the most part) [09:27:32] [09:27:40] <LeftWing> The only thing stopping me from selling it is the effort it would take to organise packing and shipping. [09:29:28] <Error_404> LeftWing: craigslist it [09:29:49] <LeftWing> I think I'd prefer to eBay it. [09:30:03] <sahafeez> hum .. i sold my u60 to my ex-girlfiend for $200 [09:30:49] <astinus> you had a girlfriend who's a unix geek? :o [09:31:00] <LeftWing> And you let her get away? ;P [09:31:02] <astinus> (wait. .. mental overload ... can not compute) [09:31:31] <astinus> okay, i dont normally ask personal questions, but I have this need to ask this one... [09:31:34] <astinus> was she hot? [09:31:41] <sahafeez> profiles.yahoo.com/uf0girl she is a solaris cert sysadmin. [09:31:44] <astinus> if so, I'm officially hunting you down [09:31:48] <sahafeez> she is just a nut job [09:31:58] <Error_404> in my experience most geek girls are [09:31:59] <sahafeez> have a look. or myspace.com/uf0girl [09:32:01] <astinus> OMFG [09:32:08] * astinus hunts sahafeez down [09:32:08] <LeftWing> Pity. Nutjobs are irritating. [09:32:09] * Error_404 <3's his non-geek [09:32:18] <astinus> I can live with a nutjob [09:32:19] <sahafeez> i took the picture in 2000 in our place in san jose [09:32:26] <LeftWing> astinus: No, you can't. [09:32:47] <Error_404> you really can't [09:32:58] <astinus> LeftWing: sure you can .. you just need good quality rope/chain [09:33:03] * astinus snickers [09:33:06] <sahafeez> yes, but i have found that the crazier they are the better the sex. [09:33:14] <astinus> thats so true :/ [09:33:25] <astinus> its $deity taunting us [09:33:34] * LeftWing wonders which PostIt(tm) Note to attend to first. [09:33:41] <sahafeez> my wife is a nut job, but in a livable way, unlike kellie [09:33:42] <LeftWing> I really need a call management app. [09:34:09] <astinus> LeftWing: I'll write you one for that E450 :P [09:34:23] <sahafeez> one time she made a dr. app that i had been buggin her about forever. so we go to dinner (sushi) before hand. she get so trashed she passes out at the doctors [09:34:37] <sahafeez> this is the ex.. [09:34:38] <Error_404> abortion? [09:34:44] <sahafeez> no [09:34:49] <sahafeez> it was fun for a year. [09:35:08] *** deather_ is now known as deather [09:35:15] <LeftWing> I think I'll go with the workstation that fails the mersenne prime test. [09:35:34] <Error_404> LeftWing: do it! [09:36:19] <sahafeez> hum. useradd -d /export/home/sahafeez should make the dir right? [09:36:40] <astinus> useradd -d directory -s shell username [09:37:20] <sahafeez> did all that. useradd -c "Dumb Ass" -d /export/home/dumbass -u 501 dumbass [09:37:23] <sahafeez> no dir [09:37:48] <sahafeez> vipw shows /home/dumbass [09:38:45] <sahafeez> what i screw up [09:38:54] *** MattMan has joined #opensolaris [09:39:08] <astinus> i dunno, I have /home sym'd to /export/home anyway [09:39:21] <sahafeez> worked in 48 ;) [09:40:28] * LeftWing & [09:40:35] <sahafeez> ah, order [09:40:57] <sahafeez> useradd -d /export/home/bla -m -s /bin/bash -c "Bla Bla" -u 501 bla [09:41:31] <sahafeez> i have not slept in like 2 days.. [09:41:35] <sahafeez> i should do that .. [09:43:25] <astinus> lol [09:43:30] *** bank has joined #opensolaris [09:43:57] <sahafeez> started a project to rid myself of the 10 boxes of computers and get down to the U80 and an openbsd NAS box/FW [09:44:00] *** MattMan has quit IRC [09:44:04] <sahafeez> then my 3 year old got a cold... [09:45:07] <sahafeez> 3 hours left on my raid parity on the obsd box. wish they did zfs [09:47:02] <sahafeez> ok url for a live upgrade anyone? [09:50:25] <sickness> Http://www.sickness.it/liveupgrade.txt [09:50:44] <sahafeez> thanks. will i hate myself after doing this? [09:51:31] <sahafeez> i created an slice called /altroot last install so i can do this.. [09:51:50] <sahafeez> but /usr et, al. are other mounts. does that mater? [09:52:05] *** nyati has joined #opensolaris [09:52:58] <sickness> i really love liveupgrade and it's not difficult to master, but usual advice is backup your data and try on a test box or vm :) [09:53:49] <sahafeez> hum. only one. box. just installed with 49. would get 50 but my dvd burner took a shit. so i setup the /altroot cuz i was told i need it for lu [09:54:04] <sahafeez> does rhythmbox work under 50? anyone? [09:55:10] <sahafeez> if not i will wait for 51 [09:56:08] <sickness> mmm so you'd better read the official docs... It's surely doable but i've never tried... [09:57:12] *** Dar_HOME is now known as Dar [09:58:17] <bank> hi sickness. [09:58:39] <sahafeez> sickness: was that an answer to something i asked [09:59:56] <sahafeez> a while back someone pointed me to a cmdline tool for setting lineout/speaker. anyone remember what is was called. [10:10:03] *** Gman has quit IRC [10:10:06] *** alanc_away has quit IRC [10:10:54] *** alanc_away has joined #opensolaris [10:11:00] *** Gman has joined #opensolaris [10:11:59] *** damienc has joined #opensolaris [10:15:12] *** Tpenta has quit IRC [10:18:50] <sickness> hi bank [10:19:07] <bank> sickness: [10:19:15] <bank> sickness: it is bios problem. [10:19:30] *** sahafeez has quit IRC [10:20:30] <sickness> sahafeez: i've that command at home... Now from mobile... [10:20:33] <raph_ael> hello [10:20:54] <bank> sicknesss:x86: BIOS Hangs on Reboot After Using ZFS on Sun Ultra 20 or Sun Fire X2100 (6363449) [10:20:59] <bank> sickness:http://docs.sun.com/app/docs/doc/819-5947/6n80cmc6l?a=view [10:21:05] <bank> :| [10:21:35] <bank> I will going to update bios. [10:21:45] <astinus> lol, ship it back to Sun, maybe they'll replace it with an M2 :P [10:22:38] <bank> great. [10:22:45] <bank> sun great company! [10:22:46] <astinus> stranger things have happened with 1U servers from Sun :P [10:22:49] <bank> :( dream [10:22:56] * astinus fondly remembers Cobalt RaQ's ;) [10:23:22] <bank> thank you sun , x2100 and zfs. I already refresh the whole system [10:26:02] <bank> boring monday evening. [10:28:04] *** vigos has quit IRC [10:29:29] *** bengtf has quit IRC [10:50:12] *** jamesd__ has joined #opensolaris [10:51:32] *** jamesd_ has quit IRC [10:52:43] *** sparc-kly__ has joined #opensolaris [10:52:54] *** nyati has quit IRC [10:55:42] *** bank has quit IRC [11:01:07] *** sparc-kly_ has quit IRC [11:20:15] *** sylvain has joined #opensolaris [11:20:17] *** lplatypus has joined #opensolaris [11:52:56] *** astinus has quit IRC [11:57:40] *** kimc has joined #opensolaris [11:58:06] <kimc> good morning [12:11:14] *** lplatypus has quit IRC [12:13:49] *** calumb has joined #opensolaris [12:14:28] *** lplatypus has joined #opensolaris [12:23:53] *** lplatypus_ has joined #opensolaris [12:24:17] *** yongsun has quit IRC [12:26:07] *** lplatypus_ has quit IRC [12:27:04] *** lplatypus_ has joined #opensolaris [12:27:43] *** lplatypus_ has left #opensolaris [12:32:21] *** lplatypus has quit IRC [12:33:37] *** kimc has left #opensolaris [12:33:40] <jteo> hello all. [12:35:43] *** nexrafa has quit IRC [12:36:54] *** lplatypus has joined #opensolaris [12:37:21] <lplatypus> what's the ipf module (module id 195)? it's hanging my nexenta alpha6 installcd boot. [12:37:31] <trygvis> ip filter [12:40:27] <ofu_> how can i find out if one dual-core or 2 single-cores are running in a x4200? Sol10U1, no prtdiag [12:40:37] <jteo> psrinfo -vp [12:40:52] <ofu_> tells me that i have 2 processors [12:41:10] <jteo> then 2 single cores. [12:41:13] <ofu_> aah, thx [12:41:14] <jteo> virtual processors are cores. [12:41:27] <jteo> so you have "2 physical processors" eh? [12:41:34] <ofu_> yes [12:42:34] <jteo> lucky yew [12:42:37] <ofu_> aah, on a dual-core it says The physical processor has 2 virtual processors (0, 1) and Dual Core AMD Opteron(tm) Processor 275 [12:43:45] <lplatypus> oh maybe ipf is just the last kernel module loaded, and the hang happens later... when I boot the nexenta cd choosing the single-user option rather than the install option, i do see the "installing ipf" message but then it prompts me for a password and i'm successfully booted to single user mode [12:44:56] *** salamanders has joined #opensolaris [12:48:22] <lplatypus> has anyone else here done a fresh install of nexenta alpha 6? [12:52:13] *** jmcp has left #opensolaris [12:53:35] *** Gman is now known as GmanAFK [12:57:12] *** bank has joined #opensolaris [12:57:23] <bank> hi [12:57:44] <bank> I have some question about legacy mountpoint with zfs and non global zone [13:03:00] *** bunker_work has joined #opensolaris [13:04:57] *** uebayasi has quit IRC [13:08:26] *** Cyrille has quit IRC [13:08:32] *** PosixCompliant has quit IRC [13:09:29] *** bank has quit IRC [13:10:31] *** Cyrille has joined #opensolaris [13:14:02] *** mega has joined #opensolaris [13:18:21] *** lplatypus has quit IRC [13:18:39] <sickness> just back from work :) [13:18:58] *** lplatypus has joined #opensolaris [13:23:35] *** lplatypus has quit IRC [13:24:13] *** lplatypus has joined #opensolaris [13:25:07] *** npt-laptop has quit IRC [13:27:10] *** ProfMikey has quit IRC [13:33:10] *** lplatypus has quit IRC [13:35:41] *** gm152 has joined #opensolaris [13:37:54] *** Cyrille has quit IRC [13:38:35] *** Cyrille has joined #opensolaris [13:47:19] *** bank has joined #opensolaris [13:50:18] *** laca has joined #opensolaris [13:55:57] *** sparc-kly has joined #opensolaris [13:55:58] *** ChanServ sets mode: +o sparc-kly [14:01:14] *** calumb is now known as calLNCH [14:02:17] *** rodrickbrown has quit IRC [14:07:44] *** gm152 has quit IRC [14:11:51] *** Tzoa has joined #opensolaris [14:12:25] *** Mazon is now known as mazon [14:18:32] *** sparc-kly__ has quit IRC [14:39:30] *** nyati has joined #opensolaris [14:42:09] *** bsharitt has joined #opensolaris [14:43:21] <bsharitt> I'm planning to setup a fileserver using opensolaris. does anyone here have any experience with real life usage of raid-z? [14:45:10] *** _schily__ has quit IRC [14:45:40] *** coffman has joined #opensolaris [14:48:33] *** bsharitt has quit IRC [14:51:31] <jteo> why do some people think IRC is like the freakin' eight ball? [14:56:18] <icon> eight ball? [14:57:45] *** lin has joined #opensolaris [15:00:17] *** trs81 has quit IRC [15:00:41] *** mega has quit IRC [15:03:22] *** sparc-kly_ has joined #opensolaris [15:10:12] *** sparc-kly has quit IRC [15:14:13] *** calLNCH is now known as calumb [15:14:40] *** pablo-costa has joined #opensolaris [15:15:05] *** pablo-costa is now known as solaris-user [15:16:34] <solaris-user> anyone knows if ultra-30 support ultrasparcII 300+ processors?????] [15:21:08] <bank> eight ball? [15:24:31] *** qbit has left #opensolaris [15:26:46] *** nwf has quit IRC [15:27:11] *** trs81 has joined #opensolaris [15:29:11] *** sparc-kly__ has joined #opensolaris [15:33:12] *** karrotx has quit IRC [15:34:40] *** mazon is now known as Mazon [15:36:54] *** Tpenta has joined #opensolaris [15:38:43] *** sparc-kly_ has quit IRC [15:40:09] *** mammoth has joined #opensolaris [15:40:23] *** lin has quit IRC [15:40:53] *** mammoth has quit IRC [15:44:39] *** aiki_kid has joined #opensolaris [15:49:41] *** jonkelle has joined #opensolaris [15:51:03] *** sparc-kly has joined #opensolaris [15:51:04] *** ChanServ sets mode: +o sparc-kly [15:51:13] *** DataStream is now known as AJ__Z0 [15:51:15] *** tsoome has quit IRC [15:51:29] *** calumb has quit IRC [15:51:37] *** AJ__Z0 is now known as DataStream [15:54:57] *** solaris-user has quit IRC [15:57:55] *** mrdeviant has joined #opensolaris [16:02:25] *** cormac has quit IRC [16:02:46] *** aiki_kid has quit IRC [16:10:40] *** loke has joined #opensolaris [16:12:23] *** glagasse has joined #opensolaris [16:13:37] *** sparc-kly__ has quit IRC [16:24:59] <jteo> is there a real need for growable raid-z vdevs? [16:25:35] <elektronkind> I suppose there is, for people who want to avoid the disk/space tradeoff [16:26:23] <Stric> jteo: I'd want it personally.. [16:27:52] <Error_404> i would just like the ability to turn a mirror in to a raidz [16:28:16] <Error_404> so that i can upgrade when i can afford to throw more harddrives in the machine [16:29:03] <jteo> even if it means waiting/offline conversion? [16:30:42] <AbeFroman> absolutely [16:30:45] <bank> hey, I reconfiguring BIND on my server . but some of it is working and some is not ... I change the header number but it seem like it doesn't update ... [16:31:00] <bank> Do I miss some whitespace .. ? http://pastebin.ca/217189 [16:31:01] *** sopor_aeternus is now known as zarathustra [16:31:04] *** kloczek has joined #opensolaris [16:31:17] <jteo> AbeFroman: intriguing. i didn't think anyone was patient enough. [16:31:32] <jengelh> whitespace is no issue in bind [16:31:48] <jengelh> though you have one too much [16:31:52] <jengelh> two even [16:31:58] <jengelh> you'd be better off using $ORIGIN anyway [16:32:19] <AbeFroman> i'm patient enough to migrate 23T off of ufs filesystems to convert them to zfs... [16:33:14] <jengelh> http://pastebin.ca/217191 [16:33:54] <jteo> AbeFroman: alrighto. yet another topic of..dissent on zfs-discuss. ;) [16:34:09] <AbeFroman> haha [16:38:30] *** stevel has joined #opensolaris [16:38:30] *** ChanServ sets mode: +o stevel [16:39:52] <jteo> AbeFroman: give an example of this raidz to mirror thingy. Say, you want to convert a 3 disk raid-z to a 3 disk mirror? [16:41:20] <AbeFroman> i was talking more about the adding a disk to a raidz [16:41:26] <AbeFroman> or raidz2 [16:41:46] <bank> jteo:? [16:42:08] <jteo> AbeFroman: ah. [16:42:14] <bank> I am not sure what I am wrong. May I change db serial number to next week? may be it update [16:42:21] *** jonkelle has quit IRC [16:42:24] <bank> I dunno it's not update or . it is wrong configure [16:42:25] <bank> :( [16:43:01] <oxygene> bank: you can set the serial number to whatever (as long as it's larger than the old numbers) - that date-like format is just convention [16:44:51] <jteo> AbeFroman: but most people don't want a large raid-z vdev anyway? [16:45:41] <Stric> is going from say 3->4 disks considered "large" ? [16:48:45] <bank> it is really unknow some of it work. some of it doesn't [16:49:37] <elektronkind> in the time of 500 and 750GB drives, wishing to grow a raidz vdev is quite reasonable [16:50:36] <bank> oh jteo [16:50:45] <bank> I just realize that you add $ORIGIN [16:50:48] <bank> sorry [16:51:38] <jengelh> hey i added that [16:54:30] *** GmanAFK has quit IRC [16:55:41] *** GmanAFK has joined #opensolaris [16:56:01] <bank> ooppss [16:56:06] <bank> sorry jengelh [16:56:19] <jengelh> tada [16:56:21] <bank> dependency require_all/none file://localhost/etc/named.conf (absent) [16:56:34] <jengelh> dependency wth? [16:57:12] <bank> oh /etc/named.conf [16:57:17] <bank> I del it. [16:57:19] *** AHG has joined #opensolaris [16:58:53] <bank> it is online now, let me ping. [17:00:39] <bank> it still an unknown host. [17:00:50] <bank> or no answer [17:01:39] <jengelh> then the zone did not get loaded [17:01:44] <jengelh> check/var/log/messages [17:01:50] <jteo> hmm. looks like someone external mentioned he would be working on adding disks to raidz [17:02:08] <oxygene> hurray.. 10u2 with latest patches (w/o contract) is incapable of creating zones - "zonecfg_add_index: symbol not found" on commit.. let's see which patch is responsible for that [17:02:18] <bank> it doesn't exists [17:02:25] <bank> the services is online. [17:02:53] <jengelh> well whereever named dumps its diagnostic output [17:04:50] <bank> what does host -a help? [17:05:09] <bank> it show result almost of them [17:05:19] <jengelh> try dig(1) [17:05:28] <jengelh> dig bank.com axfr [17:05:40] <jengelh> if all goes well (and localhost is allowed to do axfr) then the zone will be printed [17:05:55] <oxygene> ah, 122663.. okay.. [17:06:38] *** coffman has quit IRC [17:07:08] <bank> connection timed out; no servers could be reached [17:09:12] *** bunker_work has quit IRC [17:09:17] *** mrdeviant has left #opensolaris [17:22:59] <elektronkind> sunsolve REALLY needs to be more reliable [17:24:00] <jteo> elektronkind: down again? [17:24:23] <elektronkind> jteo: downloading patches is hanging right now :/ the rest of the site seems up though [17:25:08] <jteo> elektronkind: :( [17:29:11] <bank> it seem that new bind configuration didn't effect. [17:29:11] *** gnu2it2 has quit IRC [17:32:11] *** bank has quit IRC [17:32:11] *** gdamore has joined #opensolaris [17:32:22] <gdamore> hi * [17:32:23] <jteo> gdamore: greetngs. [17:32:39] *** laca is now known as lacaAFK [17:34:37] <delewis> elektronkind: use pca [17:34:44] <delewis> smpatch hanging is really nothing new [17:36:47] *** stevel has quit IRC [17:45:51] *** vmhobbes has joined #OpenSolaris [17:45:52] <gdamore> i'm thinking of hooking up an SATA controller into a 360MHz USIIi system, and using it as a NAS server (using ZFS on the disks). Am I insane? [17:48:10] *** _william_ has joined #opensolaris [17:48:41] <_william_> hi all [17:48:46] <loke> gdamore: perhaps [17:49:10] <jteo> gdamore: i don't think the SATA framework works for sparc. [17:49:11] <loke> gdamore: you you like to wear funny trousers and walk around singing happy birthday to strange people? [17:49:26] <loke> gdamore: if yes, then there is a chance you might be insane [17:50:34] *** salamanders has quit IRC [17:50:47] <gdamore> heh. we have some SATA drivers in our own source tree. [17:51:13] <jteo> gdamore: more of those goodies coming our way i hope once you get your SCA. [17:51:53] <postwait> I'm trying to get sol10 to ldap auth from a directory server and having a bitch of a time. [17:51:54] <gdamore> dunno about SATA. my preference would be so, but mgmt has to decide. [17:51:55] *** stevel has joined #opensolaris [17:51:56] *** ChanServ sets mode: +o stevel [17:52:31] *** gnu2it2 has joined #opensolaris [17:52:40] *** bank has joined #opensolaris [17:54:07] <bank> can I accelerate update dns? [17:54:16] <alanc_away> gdamore has worked on video drivers, there's not just a chance he's insane, but a strong certainty he must be [17:55:00] <bank> 10800 ; refresh every 3 hours to 60 [17:55:11] <gdamore> heh. [17:56:11] <gdamore> i've touched a lot of different drivers in Solaris. some are more insane than others. the video certain is on the "more" side. [17:57:01] <gdamore> i want a cheap NAS device that supports NFS and NIS. I tried NetBSD on a cobalt, but NFS "bugs" or design deficiencies make me not want to use it. [17:57:29] <quasi> so use solaris on a via [17:57:40] <quasi> then you zfs as well [17:58:08] <gdamore> i was thinking about that. does sx86 work reasonably well on a via class system? [17:58:29] <gdamore> i once tried sx86 (Solaris 8) on a Via C3 laptop but it wouldn't support the CPU [17:58:48] <quasi> s8 didn't support much of anything [17:59:36] <quasi> I know it worked on the good old M10000 - asyd might know some more about C7 support [18:00:16] *** andrei has joined #opensolaris [18:00:31] <ofu_> does bufhwm also work for zfs? [18:01:14] <gdamore> anyway, i already have a bunch of USIIi systems here. a U10, and about 3 other of our "Voyager IIi" systems. (a U10-ish system with cardbus, in a "shoebox" formfactor) [18:01:46] <postwait> So, NFS and ZFS don't like each other [18:01:53] <postwait> is that improved in u3? [18:01:54] <quasi> gdamore: sure, if you don't care about power consumption ;) [18:02:20] <gdamore> postwait: that was part of what I was worried about. If I can't use NFS with ZFS, then maybe I should go with hardware raid instead? [18:02:43] <quasi> postwait: benr posted something like that recently as well - and from his posts I don't think it has been fixed yet [18:02:44] <gdamore> i am willing to run OpenSolaris, if that helps. [18:03:29] <bank> :( [18:04:27] <oxygene> grr.. zones and packages is a really sad story (even with full zones, it seems) - try installing and uninstalling a package with "e" type files (the ones that write "Modifying $filename" after installing everything, eg. openoffice-desktop-integration) globally on a zoned system.. [18:05:09] <quasi> postwait: on the other hand, http://milek.blogspot.com/2006/09/zfs-in-high-availability-environments.html looks pretty promising [18:05:12] <postwait> quasi: FWIW I posted something like that too [18:05:41] <bank> I dunno why my BIND configure doesn't update :( [18:05:47] <postwait> I've played with that. postwait doesn't like beta version of clustering. [18:05:56] <postwait> non-beta versions have plenty of bugs to keep me busy. [18:06:55] <quasi> yeah, I keep asking sun when it stops being beta - however it has some really interesting features a certain postgres you know would like ;) [18:07:05] <gdamore> so what exactly is the problem with ZFS and NFS? [18:07:19] <jteo> gdamore: the fsync bug exists in U2. [18:07:39] <gdamore> bugid? [18:08:01] <quasi> gdamore: but something else seems to be at play as well from what benr posted [18:09:23] <jteo> gdamore: one sec [18:11:06] <bank> :( [18:13:13] <jteo> gdamore: http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6413510 [18:13:32] <gdamore> jteo: thanks. [18:14:44] <gdamore> fixed in b48, it says. so maybe not too much of a problem, but i have to use nevada. [18:17:23] <fet-kviga> how can i get xterm-color or even color for my xterm? i am using vim to do some coding, but i don't see any color highliting ;\ [18:18:16] <gdamore> i think blastwave has an xterm-color [18:18:35] *** bunker has joined #opensolaris [18:18:49] <fet-kviga> other than that, you cannot get color in your xterm? ;\ man.. that is really inconvenient [18:19:20] *** ahl has joined #opensolaris [18:19:27] <gdamore> just verified, xterm in S10 has xterm color. but you need to set TERM=xtermc (note the trailing c) [18:20:22] <fet-kviga> ahh.. bra [18:20:27] <fet-kviga> that works [18:20:32] <fet-kviga> thank you [18:22:08] *** bougie has joined #opensolaris [18:23:12] <alanc_away> Solaris got color in xterm in S9 [18:23:55] <gdamore> i kind of recall that. i don't have xtermc as TERM by default because i still work on OS' that don't have the terminfo info xtermc. [18:25:20] *** slowhog has joined #opensolaris [18:31:32] <bank> my bind configuration doesn't update. [18:33:13] <bank> :( [18:33:16] <gdamore> what do you mean by "doesn't update"? [18:33:33] <bank> I try ping , nslookup , host -a [18:33:47] <bank> it still point to the old location [18:34:06] <bank> I update the db serial number to 2008 .. and refresh time to 60 [18:34:24] <bank> enable / disable one hundread times. [18:34:31] <bank> it still point to the old location [18:34:32] <bank> :( [18:35:06] <gdamore> if the resolver has cached the settings with an _old_ ttl, then the old cache has expire first [18:35:47] <bank> I have to wait? [18:36:04] <gdamore> but i believe nslookup doesn't go thru the normal resolver cache. i could be mistaken, however. [18:36:48] *** nwf has joined #opensolaris [18:36:51] <gdamore> can you do a direct lookup against your master DNS server? [18:37:00] <bank> how could i do that? [18:41:00] <gdamore> when using nslookup, at the > prompt do "server <ipaddr>" where <ipaddr> is the IP address of your master DNS server [18:41:30] <bank> http://picasaweb.google.com/vichai.ar/Extras/photo#4989127019192582162 [18:43:09] <gdamore> your Minimum Time To Live value is still pretty high. [18:43:10] *** damienc has quit IRC [18:43:33] <bank> pardon? [18:43:41] <gdamore> as a consquence, caching servers will not notice updates for ~1day. [18:43:52] *** Cyrille has quit IRC [18:43:59] <gdamore> your secondaries should get timely updates though. [18:44:01] <bank> so I will change it to 60 [18:44:17] <gdamore> yes, but it will take ~1 day for that to get into effect. [18:44:28] <bank> oh why! [18:44:53] <bank> it up to last time setting? [18:44:53] <gdamore> when changing ttls and such, it is important to realize that the cached value is stored on the caching server. it won't notice new values (in this case a new minttl) until the _old_ ttl expires [18:45:04] <bank> ohhhhhhhhhhhhhhhh [18:45:13] <bank> so I will have to wait until tomorrow midnight [18:45:13] <sickness> re [18:45:14] <bank> :'( [18:45:26] <gdamore> yes. this is a "feature" of DNS. :-) [18:45:36] *** AHG has quit IRC [18:45:46] <bank> nslookup server <ipaddr> has no effect still point to old location. [18:46:29] <bank> gdamore [18:46:34] <gdamore> yes? [18:46:40] <noyb> dclarke: ping [18:46:42] <bank> gdamore: Do I done a correct zone file? [18:46:55] <gdamore> the zone file looks okay. [18:47:04] *** gnu2it2 has quit IRC [18:47:11] <bank> Can I change 2008 back to today date? :P [18:47:18] <bougie> hello [18:47:19] <gdamore> yes. [18:47:55] <bank> gdamore : why nslookup server <ipaddr> has no effect? [18:48:01] <gdamore> altough, i'm not sure if some dns server expect a strictly increasing serial number [18:48:05] <bank> I mean it doesn't show my new result. [18:48:25] <gdamore> i don't know. one possible concern, is, is your DNS server pointing to the right db files? [18:48:46] <bank> yes .. [18:48:51] <bank> why ? [18:49:02] <bank> there are no other file else [18:49:03] <gdamore> cause if you're editing the wrong files, it won't take effect. [18:49:08] <bank> lol [18:49:11] <gdamore> check /etc/named.conf [18:49:24] <richlowe> or if you never reloaded the zone. [18:49:35] <bank> file "db.xxxxx.com"; [18:49:50] <bank> so db.xxxx.com have to be in /var/named/db.xxxx.com [18:50:03] <bank> it is that, as you have seen. [18:50:08] <bank> reloaded the zone? [18:50:30] <gdamore> pkill -1 named ? [18:51:35] <bank> is that differ from svcadm disable and enable again? [18:52:08] <gdamore> possibly. disable doesn't stop the currently running instance by default [18:52:28] <bank> from prstat I see [18:52:29] <bank> 1723 root 5852K 3356K sleep 59 0 0:00:00 0.0% named/4 [18:52:46] <Sporq> anyone here know of any good books covering active directory and ldap interoperability? [18:52:54] <gdamore> svcadm restart would be better. [18:53:25] <bank> ok it is gone [18:53:29] <bank> so I will enable it again [18:54:21] *** nwf has quit IRC [18:54:32] *** ahl has left #opensolaris [18:54:52] *** nwf has joined #opensolaris [18:55:03] <bank> :( still old [18:55:11] <bank> I may need to wait until tomorrow [18:55:16] <bank> if it still the same. [18:55:19] <bank> that would be mad. [18:55:54] <gdamore> perhaps you should specifically provide the following: your IP address, the nslookup command you are using (in detail), the expected output, the actual output [18:56:30] <postwait> Are there any gotchas with using auto_home for users in ldap? [18:56:43] <postwait> I can do things like "id postwait" on my system. [18:56:50] <postwait> and I get my uid... [18:57:02] <postwait> but when I su - postwait, it says no home dir [18:57:11] <postwait> homedir in ldap is /home/postwait [18:57:20] <postwait> dir on the box is /export/home/postwait [18:57:31] <bank> http://pastebin.ca/217446 [18:57:36] <bank> replace xxx with kranular [18:57:39] <postwait> and I have my auto_home set up * localhost:/export/home/& (which works for local users) [18:57:41] <quasi> maybe hacking a bit at the auto home conf [18:59:13] <gdamore> the DNS server you are using is is not kranular.com, you are going to a secondary or caching server (619.19.245.51) [18:59:29] <gdamore> so you have caching in the way [19:00:19] <gdamore> try doing ">server 61.19.245.19" or whatever the NS record for kranular.com is (i closed the window that had your zone file in it) [19:00:56] <gdamore> found it: so "> server 61.19.242.219" in nslookup, and then do your lookup. [19:01:49] <gdamore> btw, i usually set my minttl to 3600, and my secondary refresh to around 1800. [19:02:14] <gdamore> that means that it takes an hour for updates to propogate, but it allows for reasonable caching still. [19:02:42] <bank> connection timed out; no servers could be reached [19:03:17] <gdamore> well, that sounds bad. is your master name server running on 61.19.242.19, and is it reachable? [19:03:30] <bank> zone files? http://picasaweb.google.com/vichai.ar/Extras/photo#4989127019192582162 [19:03:41] <bank> BIND is running on 61.19.242.19 [19:03:44] <bank> no [19:03:46] <bank> .18 [19:03:53] <bank> BIND is running on 61.19.242.18 [19:04:09] <gdamore> well, your "master" zone file indicates .219. [19:04:16] <bank> argg [19:04:19] <bank> arggg [19:04:24] <bank> root of evil [19:05:24] <bank> gdamore [19:05:39] <bank> I would like to point ns1, ns2 to .18 [19:07:01] <bank> no it already point to 18 [19:07:11] <bank> so I should delete it? [19:07:38] <gdamore> well, it won't matter. you still have to wait ~24 hours for caches to time out. [19:08:09] <bank> I am goint to delete IN NS www.... [19:08:14] <bank> all of IN NS [19:08:26] <bank> will that ok? [19:08:31] <gdamore> i dont think so. [19:08:43] <gdamore> deleting all NS records means _no_ DNS will work for your zone. [19:09:19] <gdamore> you need at least one NS record for your zone, I think. you don't need additionals if you don't want em [19:10:00] <bank> gdamore [19:10:05] <gdamore> and as i said, it won't solve the problem that you already have, with a 24 hour cache. unless you can flush the caches on the servers you care about, you can't do anything about it [19:10:18] <bank> do you understand that someone point ns1,ns2 to .18 for me [19:11:05] <bank> ok .... [19:11:10] <bank> I will sleep... [19:11:50] <gdamore> good luck. :-) [19:12:57] <bank> good night [19:12:59] *** bank has quit IRC [19:13:06] <jteo> hmm. i'm tempted to switch to HR for IT, just because of the idiots in it. [19:14:36] <gdamore> well, who do you think _hired_ the idiots? :-) [19:14:54] *** loke has quit IRC [19:15:22] <jteo> gdamore: :) ah well. -sigh- [19:16:33] <jteo> gdamore: is your HR department any better? [19:16:40] <jteo> or did i just make you cough up coffee? [19:17:08] <gdamore> i dunno. we didn't have an HR department when I was hired. our current HR department is someplace I've never been. (now that we are owned by General Dynamics) [19:17:37] <gdamore> now our IT department... :-) [19:18:58] <jteo> gdamore: i see. [19:19:59] * gdamore recognizes that the worst job in the world would probably be in IT at any company that he works at. because as a former sr. sysadmin and now developer, he is probably the worst person in the world to have to provide IT services for. [19:20:24] <gdamore> by the same token, the IT staff at Sun have an unenviable position. [19:20:43] <alanc> we solved that problem at Sun - we got rid of most of the IT staff [19:20:46] <gdamore> though from experience, most of the developers at Sun have very little if any sysadmin experience. [19:20:49] <alanc> unfortunately [19:20:53] <gdamore> heh. [19:21:34] <gdamore> i recall that most of the developers for the E10K tools had _no_ idea on how admins typically ran their systems. [19:22:08] <sickness> lol [19:22:24] <gdamore> and in one kernel class, i saw an engineer who knew how to break into a kernel and give himself root with kadb, but couldn't then make the change permanent by editing the password file. [19:23:36] *** stevel has quit IRC [19:23:47] <gdamore> on the other hand, when I was working in IT, I knew people who had left former developer jobs to become sysads. I think I'm almost the only person I know who transitioned the other direction. [19:24:23] <gdamore> s/almost// [19:24:31] <alanc> I did sysadmin and tech support before graduation - learned enough to not want to do either of those for a career [19:24:48] <jteo> encouraging news. :) [19:24:51] <alanc> and the guy who works on xscreensaver was a lab admin a few years ago [19:24:52] <gdamore> i couldnt' get a decent software position right out of school. [19:25:25] <gdamore> but qualcomm hired me right away to do sysadmin. i had experience doing both sysadmin and developer stuff on my resume. [19:25:50] <alanc> I was lucky - had a friend from school working in the desktop group at Sun - passed my resume to his boss who wanted an intern to do RE tools development, and from there was able to move into development [19:25:57] <gdamore> a couple years ago i finally removed all references to ever being a sysad from my resume. i was getting too many calls from headhunters who wanted to hire sysads. [19:26:50] <gdamore> i moved out of IT to work at Sun.... my resume was handed to Sun by the E10K support rep. :-) went to work on E10K SSP software. [19:27:15] <jteo> i'm taking notes. [19:27:19] <gdamore> heh. [19:27:45] *** vmhobbes has quit IRC [19:27:52] <gdamore> i would think Sun recruiters would be watching places like #opensolaris. :-) [19:28:00] *** odla has joined #opensolaris [19:28:12] <gdamore> i know I would watch opensolaris-code, if I were an HR person. :-) [19:28:16] <alanc> since I'd worked in Sun support, one of the interviewers searched for the bugs I'd filed there and liked what he saw enough to recommend me [19:28:23] <oxygene> gdamore: and that's why you aren't ;) [19:28:28] <odla> how is wireless support in opensolaris? i have an orinoco card, that uses the orinoco_cs driver in linux ...is this support in OS? [19:28:29] <gdamore> hahaha [19:28:46] <sickness> oxygene: lol :) [19:29:01] <alanc> I think they're relying on the developers already employed by Sun to submit referrals of community members they've had a chance to work with in depth [19:29:25] <gdamore> there is support for it in opensolaris on x86. but not on sparc, and not on stock S10 (even 06/06) [19:29:55] <Error_404> hmm... containers aren't what i'm looking for i think [19:29:56] <odla> gdamore: is that to me? [19:30:03] <odla> gdamore: i have an x86 [19:30:07] <Error_404> i'm looking to limit memory of a zone... suggestions? [19:30:08] <gdamore> odla: yes [19:30:18] <odla> what about the pcmcia slot? [19:30:21] <gdamore> odla: its also PCMCIA only [19:30:39] <gdamore> odla: pcmcia is supported on x86. but probably not for all chipsets. [19:31:04] * gdamore gdamore does not run Solaris on x86 hardware, except his desktop, which is a Sun Ultra 20. [19:31:19] *** john053 has joined #opensolaris [19:31:47] *** john053 has left #opensolaris [19:32:30] <gdamore> any idea what the reasonable memory minimum requirement is for SXCR 50 on an Ultra 20 (to be used strictly as an NFS server) [19:32:37] <odla> is requires the i82365 to be built into the linux kernel [19:32:40] <quasi> Error_404: rcapd [19:33:22] <gdamore> odla: Solaris is much more modular than Linux in that regard. but you'll have to try it out to find out if it works [19:33:26] <Error_404> drats, i was hoping you weren't gonna say that [19:33:39] <odla> i know it didn't about a year ago when i tried solaris [19:33:59] <oxygene> odla: wireless drivers appeared over the last year, so it might have changed [19:34:51] * gdamore has fixes for pcpw that will make them work properly with miniPCI cards. he hopes he gets a signed SCA soon so that he can contribute them. [19:35:18] <quasi> Error_404: alternatively you could do an evil dtrace to kill off evil procs [19:36:52] *** vmhobbes has joined #OpenSolaris [19:37:19] <Error_404> quasi: not an option [19:38:05] <Error_404> i'm trying to config myself a zone so i can test how much i'm gonna have to pay mod3 [19:38:36] <jteo> ? [19:38:50] <jteo> gdamore: you ever gonna tell me about your encounters with NetBSD politics? [19:39:11] *** sahafeez has joined #opensolaris [19:39:13] <odla> i know my pcmcia slot is a texas instrument slot and that it uses the yenta driver [19:39:18] <odla> it's a dell inspiron 1100 [19:41:39] *** nwf has quit IRC [19:42:57] *** deather_ has joined #opensolaris [19:46:36] <Error_404> oh.... neato [19:46:46] <Error_404> glassfish'll run (terribly) on 64meg [19:50:32] <gdamore> odla: i'm going to be adding some support for some additional pcmcia chipsets to the driver soon. [19:51:30] <gdamore> wow, Xorg and firefox are both ~750MB memory now. I think I need to log off and restart X. :-( [19:52:03] <gdamore> i really hate that closing windows in firefox doesn't seem to give memory back, unless I close _all_ of them. [19:52:04] <clee> hooray Firefox, leaking pixmaps in the X server [19:52:37] <Error_404> yay! [19:53:02] <gdamore> i'm surprised this problem isn't bothering anyone else enough that it has gotten fixed. [19:53:43] *** mage2 has joined #opensolaris [19:53:47] <mage2> hello all [19:53:52] <gdamore> anyone tried the latest firefox 2 rc yet? (is it any better?) [19:55:13] *** phalenor- has quit IRC [19:55:34] <mage2> any one here solaris network gurus? [19:55:49] <alanc> haven't tried a Unix build yet - the Windows build seemed much more like a 1.6, than a 2.0 - not a really big change from 1.5, just some new icons and a few new features [19:56:10] *** sickness has quit IRC [19:56:12] <gdamore> i'd be happy with it if it didn't suck up all the memory on my system. [19:56:31] <gdamore> i'd use opera, if it didn't crash so frequently. [19:56:43] * gdamore wants a decent light weight browser for Solaris. [19:57:17] *** phalenor- has joined #opensolaris [19:57:54] <Error_404> gdamore: isn't there a solaris version of opera? [19:57:55] <jamesd__> gdamore, lynx [19:58:03] <Error_404> actually, i'm certain there is [19:59:00] <gdamore> there is, but it crashes ~daily for me. see "decent". [19:59:06] <gdamore> lynx != decent. [19:59:28] <Error_404> hmm... i never had problems with opera [19:59:29] <Error_404> *shrug* [19:59:35] *** deather has quit IRC [20:00:02] *** sickness has joined #opensolaris [20:00:38] <sickness> i'm back [20:00:43] <Error_404> hi back [20:01:09] * delewis notes the acroread plugin does not work properly with Opera, either [20:01:46] <delewis> that's actually probably not a bad thing, as whenever Firefox uses it, it keeps acroread open even if you aren't looking at any PDFs. [20:02:14] <delewis> so, Firefox bloat + acroread bloat == bad. [20:02:31] <Error_404> hmm... i better uncap this zone's memory before it swaps my machine to death [20:03:15] <gdamore> delewis: yes: firefox bloat is my single biggest complaint with Solaris at the moment [20:03:44] <delewis> gdamore: what do you use a mailreader? [20:03:49] <gdamore> firefox is the reason I have to to logout and back in every week or thereabouts. because after a week, its too damn big [20:03:52] <gdamore> thunderbird. [20:03:59] <delewis> ah, that's bloated, too :-) [20:04:04] <delewis> not nearly as bad as Evolution, though [20:04:05] *** krozinov has joined #opensolaris [20:04:10] <gdamore> thunderbird works pretty well. it grows to 128M, but it _stays_ there. [20:04:21] <delewis> true [20:04:54] <gdamore> there is a keyboard bug with thunderbird though, where every few days i have to kill it off, because it won't accept new keystrokes. it could be an issue while running with xfce4 [20:05:16] <delewis> Thunderbird is *extremely* sluggish when editting reply messages where quotes exist [20:05:17] <gdamore> i tried opera mail too. its possible that the mailer functionality of opera is what was causing my opera crashes. i dunno [20:05:45] <delewis> gdamore: I've had one Opera crash thus far since I started using 9.x the day before yesterday (I took a long Firefox 'sabatical' [20:05:48] <delewis> ) [20:05:59] <gdamore> do you use it for reading mail? [20:06:04] <delewis> gdamore: nope [20:06:45] <gdamore> i had a crash about every day or so with opera. maybe i could go a little longer, but it got really really annoying when ~1 dozen windows just "closed". [20:07:07] <gdamore> statistically, delewis, your experiences match mine so far. :-) [20:07:14] <delewis> hmm, I never have that many tabs/windows open at once [20:07:18] <delewis> usually, I have at most 4 or 5 [20:07:25] <delewis> so the comparison may not be valid [20:07:28] <sickness> well at least opera keeps memory of your last tabs opened [20:07:35] <delewis> indeed :-) [20:07:36] <gdamore> i open a lot. [20:07:47] <sickness> if firefox crashes and you have 12 important things opened... you'd better bookmark them before :P [20:07:49] <delewis> I get claustrophobic when I open that many windows :-0 [20:07:52] <delewis> er :-) [20:07:57] * andersmo uses sessionsaver with firefox. [20:08:00] <sickness> eheh :) [20:08:20] <gdamore> some windows (e.g. stateful cookies, etc.) can't auto resume that way. [20:08:21] *** stevel has joined #opensolaris [20:08:21] *** ChanServ sets mode: +o stevel [20:08:24] <icon> i must be the last person on earth that doesnt have issues with ffox [20:08:28] <icon> (crashing) [20:08:29] <andersmo> (which remembers on your tabs - even across crashes. =) [20:08:56] <andersmo> For me FF dies pretty regularly on Solaris, but lives pretty long on Linux. [20:09:06] <gdamore> firefox almost never crashes on me. in that regard it is _much_ more stable than anything else. but it leaks like a seive. [20:09:15] <sickness> oxygene: quick question, again, do you confirm that the mplayer compiled with pmpkg is able to play win32 codecs? (provided the codecs are put in the right directory) [20:09:18] <gdamore> s/seive/sieve/ [20:09:26] <icon> i think my current sol session is over a week or two [20:09:38] <rydis> The main problem with firefox appears to be that most users (re)start it daily, so the effects of having it running for weeks isn't seen much by the developers. [20:09:43] <gdamore> i.e. i use ffox heavily, and i typically have it up for a week or two. until the memory consumption becomes unbearable, anyway. [20:09:55] * icon nods [20:10:03] <icon> weee! heap! [20:10:27] <andersmo> yumyum! [20:11:08] <gdamore> the worst part is, after killing off firefox, my Xorg server is still ~741 MB. [20:11:19] * delewis wonders when SXCR 51 will be released [20:11:40] <gdamore> at the moment all I have running is gaim, an xterm, and the xfce4 window manager. [20:11:42] <delewis> according to the onnv_schedule, onnv_51 was delivered to WOS today [20:11:51] <gdamore> and top, which shows Xorg ~741MB [20:12:02] <delewis> yes, Xorg is another one of those bloated beasts [20:12:07] <gdamore> (347MB resident) [20:12:09] <delewis> of course, that's probably because of Firefox storing pixmaps [20:12:15] <delewis> :-) [20:12:19] <gdamore> right. [20:12:39] * andersmo mumbles something about xrestop [20:13:25] <oxygene> sickness: yes [20:13:26] <rydis> Mozilla has been doing that pretty much forever. A bit bothersome on the NCD X terminals we used to use which had 8 megs of memory. [20:13:44] <alanc> Xorg just uses malloc/free, so when firefox releases the pixmaps, they just go back to the unused space in the Xorg heap [20:13:49] <delewis> yes, I fondly remember the pre-Mozilla-1.x days [20:13:50] <oxygene> sickness: though there were some bugs in the tree that I lacked a patch for [20:13:52] <delewis> where the pixmap storage was insane [20:14:03] <delewis> or at least more noticable on my workstation with 512MB of memory [20:14:28] <alanc> gisburn keeps wanting to get the Xsun code to mmap() pixmaps ported to Xorg, so that when they're freed the address space shrinks, but it hasn't happened yet [20:14:37] <sickness> oxygene: bugs? is mencoder built also? [20:14:40] <oxygene> sickness: I'm just working on mplayer 1.0rc1, which needs one last change to regain xv support (lost it underway) [20:14:43] <oxygene> sickness: yes [20:14:52] <delewis> oxygene: what!? [20:14:53] <sickness> k, tnx! :) [20:14:56] <delewis> Xvideo support is gone? [20:15:01] <gdamore> alanc: that would be nice. [20:15:15] * delewis should've tested last night before the release [20:15:16] <oxygene> delewis: they changed the x11 include/library paths detection in the configure script [20:15:16] <delewis> :-( [20:15:33] <delewis> oxygene: oh, that's just a simple --with-extraincdir/--with-extralibdir [20:15:34] <oxygene> delewis: so the pmpkg build is lacking it [20:15:35] <delewis> for a minute I was worried :-) [20:15:55] * delewis nudges oxygene to look in the AUTHORS file under the Solaris port section [20:16:00] <oxygene> delewis: I know, I know [20:16:20] <alanc> someone should bitch at Sun to fix 6444310: No xv.h under /usr/include/X11/extensions [20:16:27] <delewis> alanc: yes [20:16:44] <delewis> why customers have to be the ones to bitch, I'm not sure [20:16:56] *** lacaAFK is now known as laca [20:17:15] <alanc> because people look at me funny when I sit in my office yelling at myself to fix my bugs [20:17:27] <delewis> alanc: :-) [20:17:34] <alanc> (I was joking - I don't expect a customer to have to escalate) [20:17:37] <delewis> hey, that sounds like grounds for a promotion. [20:18:05] <alanc> I might get it in the Xorg 7.2 bundle of fixes if I have time [20:18:06] <oxygene> delewis: one thing's a regression in rc1 though: try running configure with /bin/sh (ie. default) - lots of $() crap in the directfb section - want a patch? [20:18:16] <delewis> oxygene: that should've been fixed [20:18:25] <oxygene> delewis: I'm using rc1 tarball [20:18:29] <delewis> a patch was comitted before rc1 should've been released [20:18:37] <delewis> ugh, I guess they forgot to throw it in the tag [20:18:40] <delewis> or branch or whatever [20:18:49] <delewis> oxygene: I just replaced the $() with ``s [20:18:54] <oxygene> did so [20:19:13] <delewis> ironically, /usr/xpg4/bin/sh does fine with it [20:19:30] <delewis> which leads me to conclude it's a /bin/sh-ism [20:19:47] <twincest> god forbid anyone assume /bin/sh is posix compliant :) [20:19:50] <alanc> that's because xpg4/bin/sh is a ksh variant [20:19:58] <oxygene> yes.. /bin/sh is just solaris compliant ;) [20:20:00] <alanc> /bin/sh is so ancient it doesn't have $() support [20:20:03] <delewis> alanc: yes, I know this :-) [20:20:25] * delewis prods Sun to 'cp /usr/bin/ksh93 /bin/sh' [20:20:32] <alanc> I am such a slacker: [20:20:34] <delewis> gisburn might like that too much, though [20:20:35] <alanc> Open X11 Group Defects [20:20:35] <alanc> Mon Oct 23 07:00:04 PDT 2006 [20:20:35] <alanc> RESPONSIBLE_ENGINEER P1 P2 P3 P4 P5 TOTAL [20:20:37] <alanc> alan.coopersmith at sun dot com 1 5 42 37 5 90 [20:20:46] <oxygene> delewis: /bin/sh is already a symlink - just retarget it [20:23:30] <oxygene> sickness: if you want, I can pack up the mplayer 1.0rc1 stuff for you to download once I see that it's fine [20:23:58] <delewis> I'm asking questions in #mplayerdev right now as to why it never made it into the branch :-) [20:24:10] <delewis> though, I highly doubt an rc1 tarball is going to be re-released. [20:24:35] <oxygene> unlikely [20:24:36] *** jcea has quit IRC [20:24:49] <delewis> ugh, this irritates me. [20:24:52] <oxygene> and well, it's a trivial fix [20:24:55] <delewis> pre8 built perfectly [20:25:27] <delewis> oxygene: you'll be surprised how many stumble onto 'mplayer-users' with the problem [20:25:36] <delewis> or rather contact me :-) [20:26:35] <delewis> actually, I had no idea we were going to do a release until last night (2 hours before the actual target) [20:26:47] <oxygene> delewis: point them to http://openbios.org/~oxygene/patch-20061023-1-bin-sh-compliance and hope for an rc2 soon with such a fix in it :) [20:26:56] <delewis> oxygene: :-) [20:27:13] <delewis> I guess I'll be committing something for the docs later, and I'll link to that [20:27:22] <delewis> well, there's really no point [20:27:42] <oxygene> delewis: oh, libavcodec/h264.c last_coeff_* (some array) is marked static, which the compiler/linker (gcc as of 10u2) doesn't like [20:27:50] *** jcea has joined #opensolaris [20:28:04] <delewis> oxygene: it just generated a warning, right? [20:28:13] <oxygene> delewis: "can't find symbol" on link time [20:28:13] *** nwf has joined #opensolaris [20:28:33] <delewis> oxygene: thanks [20:28:34] <oxygene> delewis: as it's marked local in the object file [20:28:48] <oxygene> delewis: patch-20061023-2-fix-static in the same directory [20:28:52] *** Jiko has quit IRC [20:32:06] <sahafeez> as i wait an hour for openbsd to format my 1tb raid 5, i wish again that solaris had pf [20:34:33] <AbeFroman> no shit [20:36:37] <sahafeez> there are 2 issues. one, i do not like ipf and i am do not know solaris well enough to feel secure in securing it as firewall [20:36:53] <gdamore> heh. this is a really fast crash in opera. [20:37:00] <delewis> sahafeez: what is there to secure? [20:37:11] <gdamore> try to download a copy of the b50 DVD for SXCR. [20:37:15] <delewis> you enable ipf, enable port forwarding, and *poof* [20:37:17] <oxygene> sickness: http://openbios.org/~oxygene/mplayer.tar.bz2 - updated pmpkg files for mplayer 1.0rc1, tested for win32 codecs and xv [20:37:51] <gdamore> afterwards, attempts to start opera fail: opera: Module initialization failure. Generic failure (-1) [20:38:16] <sahafeez> hum. take a default install of solaris and put it on the net. how will it do? [20:38:24] <delewis> sahafeez: perfectly fine [20:38:41] <sahafeez> and if i remember, i helped you fix your ipf issues ;) [20:38:56] <delewis> I've implemented firewalls for clients that have full installs of Solaris (though, they're serving as a bit more than just a firewall -- file server, backup medium, etc.) [20:39:02] <sahafeez> solaris has 2 many things that show up on a port scan on a default install. [20:39:15] <delewis> sahafeez: an ipf issue that did not effect the functionality of the firewall. [20:39:20] <delewis> sahafeez: and? [20:39:34] <delewis> there's a such thing as secure-by-default in the last few SX releaess [20:39:51] <delewis> and even so, you're perfectly able to disable those (or use a tool like JASS, which does that and a bit more for you) [20:39:52] <oxygene> sahafeez: I think that's being worked on.. apart from that: svccfg apply /var/svc/profile/generic_limited_net.xml [20:40:03] <delewis> secure-by-default is already implemented [20:40:25] <sahafeez> until it is tested for a year i will not feel right. [20:40:33] <AbeFroman> ipf has been around forever [20:40:38] <delewis> sahafeez: you think OpenBSD is tested for a year before each releaes? [20:40:44] <twincest> what's there to test about secure-by-default? [20:40:46] <oxygene> delewis: given its functionality, probably yes [20:40:59] <delewis> sorry, but even Solaris Express is more stable than most operating system releases. [20:41:04] <delewis> oxygene: ;-) [20:41:13] <delewis> "given its functionality," I like that. [20:41:45] <sahafeez> and to be honest like i said, my understanding of solaris is limited compaired to my understanding of openbsd. also i am still working on an understanding of the new startup setup as all my solaris is pre-10 (2.6-2.9) [20:42:12] <AbeFroman> pf is just a lot nicer [20:42:21] <sahafeez> delewis: yes. openbsd is 100% security each release. i would bet on it. 1 hole in 8 years. [20:42:34] <delewis> sahafeez: notice I was stability, not security. [20:42:41] <delewis> the two are mutually exclusive in most cases [20:42:44] *** myrkraverk has joined #opensolaris [20:42:49] <delewis> s/was/said/ [20:42:50] <oxygene> sahafeez: "1 hole in 8 years" in a system that can do about next to nothing without installing all the "untested" stuff.. well.. [20:43:04] <oxygene> sahafeez: I can give you 0 networked holes in 8 years with dos, too [20:43:08] <oxygene> more like 15 [20:43:09] <myrkraverk> oxygene: which system? [20:43:14] <sahafeez> bah, it this case it does 100% what i want it to do [20:43:31] <delewis> AbeFroman: I'm sure pf is nice, and there's absolutely why it cannot be implemented in OpenSolaris. [20:43:45] <myrkraverk> oxygene: I can also give you 0 network holes in my wrist watch ;) [20:43:57] <oxygene> myrkraverk: that's my point - about as usable :) [20:44:14] <delewis> Sun has decided that ipf is for them and implemented it -- perfectly acceptable choice, and anyone is able to implement pf, as OpenSolaris is more than just Solaris or Sun, now. [20:44:17] <oxygene> heh, assign darren reed with implementing pf in opensolaris *runs* [20:44:18] *** solarisjon has joined #opensolaris [20:44:21] <delewis> oxygene: :-) [20:44:24] <AbeFroman> haha [20:44:25] * delewis hears a gun shot [20:44:30] <myrkraverk> can someone point me to some info on built-in-modem support in solaris (for laptops) ? [20:44:33] <AbeFroman> i was about to ask if he ever comes in here [20:44:38] <AbeFroman> before i said pf is much nicer [20:44:48] <delewis> AbeFroman: he comes in here from time to time [20:44:54] <delewis> under the alias darrenm, I believe [20:44:58] <oxygene> AbeFroman: reading his sun.com blog, we seems more concerned with world politics than with ipf *shrug* [20:45:00] <myrkraverk> what is pf, btw? [20:45:06] <sahafeez> btw, i was not aware of the svccfg apply /var/svc/profile/generic_limited_net.xml - which i stated. my point was that under openbsd i would not need to run that at all. it starts out that way. (thanks - i am going to read about it now tho) [20:45:07] <oxygene> delewis: darrenm is someone else, or not? [20:45:20] <delewis> oxygene: right, I'm not thinking today [20:45:33] <sahafeez> myrkraverk: http://www.openbsd.org/faq/pf/ [20:45:38] <delewis> both are security guys, hence my confusion [20:45:46] * delewis needs sleep after studying for an exam last night [20:46:09] <myrkraverk> sahafeez: thanks [20:46:12] <oxygene> sahafeez: generic_limited_net is to become default - it's just about having "no surprises" on upgrades that probably caused a bit of a fight before that could become a new policy [20:46:12] <sahafeez> did you get an good grade? [20:46:40] *** ZeuSun has joined #opensolaris [20:47:09] <delewis> I certainly hope so [20:47:31] <sahafeez> like i said, i am looking forward to it. on the other had i know pf works so i put up with the 1 hour disk format and the 5 hour of raid checking.. i still have not gotten a good understanding of the new sol10 startup stuff so i do not feel like i can use it now. [20:47:34] <delewis> one of those Philosophy classes with an exam that contained 4 or 5 essay questions that you can just write freely on ;-) [20:47:41] <sahafeez> is there a good primer on it btw? [20:48:29] <sahafeez> bah, you should do fine. philosophy was an easy class because you can pretty much bullshit your way thru it - just make it your philosophy ;) [20:48:48] <delewis> sahafeez: precisely. [20:48:55] <AbeFroman> i'm moving my home router from obsd to sol10 this weekend. [20:49:10] <delewis> I've been running Solaris 9 on my router for about two years, now. [20:49:24] <sahafeez> AbeFroman: why? [20:49:45] <AbeFroman> the exact opposite reason you h ave [20:49:50] <AbeFroman> i'm more comfortable with solaris [20:50:04] *** sparc-kly_ has joined #opensolaris [20:50:18] <sahafeez> ah, ok. is it just a router or does it do something else? [20:50:23] <delewis> oxygene: I just did a build from the latest subversion sources and I never encountered the 'static' bug [20:50:57] <delewis> $ gcc --version [20:50:58] <delewis> gcc (GCC) 4.1.1 [20:51:03] <delewis> that might make a difference, though. [20:51:41] *** flynux has quit IRC [20:51:46] *** flynux has joined #opensolaris [20:52:48] <ZeuSun> does anybody know anything about zd1211 usb wireless driver on solaris ? there is an alpha project in sf.net but the author , Martin Rosenau, is no longer maintaining it [20:53:12] <sahafeez> AbeFroman: before you make that change have a look at these. start at the bottom. myrkraverk you should have a look to since you asked. it does stuff that ipf cannot touch [20:53:15] <sahafeez> http://undeadly.org/cgi?action=article&sid=20060929080943 [20:53:15] <sahafeez> http://undeadly.org/cgi?action=article&sid=20060927091645 [20:53:15] <sahafeez> http://undeadly.org/cgi?action=article&sid=20060928081238 [20:53:20] *** stevel has quit IRC [20:53:28] <ZeuSun> http://sourceforge.net/projects/zd1211sol/ and http://zd1211sol.sourceforge.net/ [20:53:43] <sahafeez> ah, start with the middle one. sorry [20:54:58] <sahafeez> so is anyone running sxcr50? if so are you running as a normal (not root user) and if so can you start up rhythmbox and tell if it works (on a sparc). i do not want to update unless it still works. (seems to break every other update) [20:56:53] *** flynux has quit IRC [20:56:58] *** sparc-kly has quit IRC [20:57:28] *** flynux has joined #opensolaris [20:57:43] <sahafeez> there has to be someone with 50 on a sparc? [20:57:58] <gdamore> i was in the process of trying to update to b50 on sparc. [20:58:15] <gdamore> but i'm still trying to download the DVD images. opera crashes when trying to do so. [21:02:18] *** yippi has joined #opensolaris [21:06:20] <gdamore> does anyone know if Acrobat 7 will ever be made available on x86? there have been rumors in the past that Sun was trying to get Adobe to fix this... [21:10:28] <gdamore> omg! there is no admin present! :-) [21:10:56] <AbeFroman> stupid adobe [21:11:15] *** B|nTaRa has quit IRC [21:11:31] *** esproul has quit IRC [21:12:15] <sahafeez> the channel is mine!!!! ahahahah. the power! [21:15:04] <sickness> oxygene: tnx :) [21:17:00] *** mega has joined #opensolaris [21:17:41] <myrkraverk> gdamore: dunno, but have you tried it on janus? [21:17:41] *** mega has quit IRC [21:17:54] <gdamore> nope. [21:18:04] <myrkraverk> gdamore: I'm curious mysely (I get my x86 laptop tomorrow) [21:18:37] *** ShadowHntr has joined #opensolaris [21:18:49] <myrkraverk> is janus part of excr b50 or will i have to download it somewhere? [21:19:09] <gdamore> dunno. [21:19:18] <gdamore> restarting X, back in a few. [21:19:18] <myrkraverk> hmm [21:19:21] *** gdamore has left #opensolaris [21:21:18] *** gdamore has joined #opensolaris [21:21:28] <myrkraverk> gdamore: wb ;) [21:22:01] <myrkraverk> would anyone here know where I can find info about built in modem support in solaris? (is there some?) [21:22:14] <gdamore> i don't think there is any [21:22:20] *** solarisjon has quit IRC [21:22:22] *** esproul has joined #opensolaris [21:22:24] <myrkraverk> hmmphfff [21:22:32] <Error_404> i would be suprised if there was [21:22:42] <Error_404> 'built in modem' is usually 'junk winmodem' [21:22:46] <gdamore> meaning, you can use a normal hardware modem that supports AT style commands. but if you need support for a Winmodem or ac97 modem, you're probably SOL [21:22:58] <jengelh> Error_404 : That was once upon a time. [21:23:18] <jengelh> Now routers with built-in modem exist, meaning, you don't need a PPPoE-PPPoA box from your ISP. [21:23:20] <gdamore> even ac97 modems need "special" support. [21:23:27] <myrkraverk> I do have an external real serial modem - in which case I may need to make/find a driver for my usb<--->serial stuff [21:24:03] <gdamore> solaris has some USB->serial drivers. but not too many. [21:24:04] <myrkraverk> but I'd like to be able to "use" everything in my 'puter [21:24:16] <myrkraverk> gdamore: maybe I'll add one then ;P [21:24:21] <gdamore> feel free to write a driver. ac97 is "non-trivial". :-) [21:24:36] <myrkraverk> ;) [21:24:36] <gdamore> (or at least so it seemed to me) [21:24:55] <myrkraverk> is it hard to find info about "win-modems" ? [21:25:13] <myrkraverk> last time I checked (ca 2000) it was some sort of voodoo ;P [21:25:47] <gdamore> winmodems are terribly hard. there is a large bit of code needed to support those. some companies (e.g. linuxant) have written closed source linux drivers. [21:27:56] <myrkraverk> hmm, well, at least I can try to get info, even if it means signing an NDA (at least drivers for solaris are not required to be open-source for some ideological reasons) [21:28:13] *** hell` has joined #opensolaris [21:28:48] <gdamore> good luck. even under NDA this info is hard. I think linuxant, for example, paid licensing fees to conexant [21:28:54] <hell`> when you set parameters in /etc/system on a non-global zone, will they only work for that non-global zone or they won't do anything at all and have to be set at the global zone level? [21:29:03] <myrkraverk> ;) [21:29:12] *** andrei has quit IRC [21:30:02] <gdamore> as the zones run the same kernel, i would expect that they only take effect in the global zone, but i'm mostly ignorant of zones, so i am not sure. [21:30:29] *** B|nTaRa has joined #opensolaris [21:34:37] *** solarisjon has joined #opensolaris [21:34:43] *** solarisjon has left #opensolaris [21:35:44] <alanc> myrkraverk: janus has been replaced by brandz, which went in a few builds ago I thought [21:37:50] *** stevel has joined #opensolaris [21:37:51] *** ChanServ sets mode: +o stevel [21:37:55] <myrkraverk> alanc: so, with build 50, I don't need to download anything extra? [21:38:20] <myrkraverk> (where do I find info an this brandz?) [21:38:21] <alanc> I don't think so - other than the libraries for whatever Linux distro you're using it with [21:38:26] <myrkraverk> ok [21:38:41] <alanc> http://opensolaris.org/os/community/brandz/ [21:38:44] <Error_404> you'll need the centos tarballs [21:38:44] <myrkraverk> alanc: can I have several distros in zones? [21:39:01] <dclarke> sounds radical .. the idea that I can stuff RHEL into Solaris and then VMWare into RHEL and then Windows XP into VMware and then ... [21:39:10] <myrkraverk> ;) [21:39:21] <dclarke> er .. CentOS .. thats what I meant [21:39:26] <andersmo> myrkraverk: right now the only distros that work are rhel (and centos, as it's basically rhel without rh =) [21:39:29] <Error_404> myrkraverk: you're limited to a certain version of the kernel & a certain version of glibc [21:39:53] <Error_404> someone got a version of gentoo working almost [21:40:00] <myrkraverk> oh [21:40:10] <Error_404> why you'd want to do that, who knows, but there have you [21:40:13] <alanc> in theory, since brandz is about giving zones a "brand" to run other binaries, I think you should be able to have different distros in different zones, but the implementation hasn't caught up yet [21:40:34] <myrkraverk> alanc: any chance whips will help? [21:40:39] <stevel> i had a debian zone, and a centos zone on my laptop [21:41:08] <alanc> myrkraverk: I don't want to know if they would [21:41:55] <myrkraverk> alanc: the issue is a bit thorny, I'd have to hire a dominatrix, and then get a clearance for her into sun internals for her to whip the engineers ;P [21:42:24] <myrkraverk> . o O (it might help though) [21:42:26] <alanc> see, I was sure I didn't want to know [21:42:33] <myrkraverk> ;) [21:43:23] <Error_404> personally, i don't really see the point of brandz [21:43:30] <Error_404> *shrug* but that's just me [21:44:04] <myrkraverk> can one "add" an OS support for brandz? [21:44:35] <myrkraverk> so that I might start to run X binaries [21:44:42] <myrkraverk> and Z binaries and so on? [21:45:16] *** glagasse has quit IRC [21:46:47] <andersmo> myrkraverk: yeah. See http://www.opensolaris.org/os/community/brandz/ - right now there's only a "lx" brand that emulates linux interfaces, but if you feel like emulating, say, a BSD, you could make a foobsd brand. =) [21:47:13] <alanc> I run X binaries all day... [21:47:27] <myrkraverk> ;) [21:48:06] <gdamore> i see brandZ with lx mostly intersting for ISPs or machines that have to run servers that are Linux based. It would be far better, to get the ISVs to properly support Solaris, IMO. [21:49:04] <gdamore> vmware, on the other hand, is truly interesting. the idea of a complete virtual machine, with separate kernels, is something i find really attractive. [21:49:16] <myrkraverk> gdamore: well, as an ISV (well, part of me is) - I see interesting posibilities with this ;) [21:49:35] <_william_> is there security issues known yet with zones ? [21:50:19] <gdamore> heh. i'd imagine zones get whatever security bugs may exist in the Solaris kernel, plus any security bugs in Linux binaries, plus any added by the zones infrastructure itself. :-) [21:50:59] <andersmo> But at least you don't have to worry about the security problems in the linux kernel. ;) [21:51:01] <_william_> gdamore, sure :) i was thinking of some issues like get root access to the global zone from another zone [21:51:09] <_william_> andersmo, ;) [21:51:13] *** gnu2it2 has joined #opensolaris [21:51:22] <gdamore> andersmo++ [21:51:37] *** Odin- has quit IRC [21:52:00] <gdamore> personally, i'd far, far rather have a Solaris userland + Solaris kernel, then some hybrid. i suppose if you want a hybrid you could also run nexenta. [21:52:48] <_william_> i was justing wondering myself about securities ussues :) i prefer 100% Solaris [21:53:35] <_william_> especially since two days ago one of my Linux boxes running raid 1 software screwed things up... one of the mirror was not synced since two monthes without a single error or warning [21:53:39] <gdamore> at that point, you're no longer running Linux at all, its GNU/Solaris. Linux is only the kernel. The userland is all GNU. [21:54:06] <myrkraverk> gdamore: no, only a small part of userland is gnu [21:54:07] <gdamore> (After saying GNU/Solaris, I feel dirty. So dirty....) [21:55:29] *** Andrew____ has quit IRC [21:55:31] <myrkraverk> (which is why a friend of mine stopped using the term gnu/linux - I never have, I find the notion stupid) [21:56:17] *** mega has joined #opensolaris [21:56:22] <gdamore> it was an RMS thing. but i am pretty sure most of the userland, apart from X, is from FSF. [21:56:50] <gdamore> e.g. bash, gawk, gsed, etc. :-) [21:57:19] <myrkraverk> oh, what about mozilla? openoffice? fvwm? zsh? xemacs? .... [21:57:52] <_william_> let say that a significant part of command line programs comes from FSF :) [21:57:57] *** yippi has quit IRC [21:58:13] <gdamore> xemacs has a lot of code from FSF in it, and is GPLd. Mozilla and OpenOffice are apps and don't count as part of the OS in my opinion [21:58:15] <myrkraverk> _william_: yes, but really, those are only small part of the system at large [21:58:18] *** yippi has joined #opensolaris [21:58:29] <_william_> myrkraverk, sure :) [21:58:33] <gdamore> fvwm is debatable. But I already declared "X" a major exception [21:58:52] <myrkraverk> even utilities like "ls" are not always gnu in linux [21:59:04] <alanc> didn't a lot of the Linux network stack & daemons originally come from BSD? [21:59:11] <_william_> actually on my i ahve 5 applications running under XFce and none of these applications, including XFce comes /directly/ from FSF [21:59:12] <gdamore> some, but not much. [21:59:14] <alanc> GNU/MIT/BSD/Linux [21:59:17] <_william_> but may be it has some code from FSF [21:59:44] <gdamore> almost all CLI utilties on Linux are from FSF, unless you run busybox. [22:00:13] <gdamore> alanc: most of the Linux network stack was written from scratch. [22:00:56] <gdamore> the linux goons used this to advance their case during the Net2 litigation fiasco back in the mid 90's. [22:01:01] <myrkraverk> strange, I don't use them so often (I use zsh, icpc, pdflatex, xemacs (gnu or not), even CC lately) [22:01:38] <gdamore> if you use the compiler, its CC. but we're talking about the OS, not the apps. [22:01:46] *** ZeuSun has quit IRC [22:01:50] <gdamore> i think zsh is GPL'd too, though I'm not 100% sure. [22:02:01] <gdamore> (it isn't from FSF tho) [22:02:14] <myrkraverk> we are? I thought we were talking about command line apps, not oses - and gpl != gnu [22:02:50] <gdamore> i said the OS. The OS includes the basic command line apps, infrastructure, kernel, and maybe the basic graphical interface. [22:03:13] <gdamore> for linux every application you use is linked against libc, which is GNU. [22:03:56] <sickness> oxygene: ping [22:05:53] <sickness> oxygene: http://www.rafb.net/paste/results/B2LLm441.html [22:12:50] *** koolniczka has joined #opensolaris [22:15:19] *** Tzoa has left #opensolaris [22:19:10] <sickness> oxygene: pleaze help =') [22:20:50] <gdamore> hmmm.... has anyone ever tried to network install S10u2? i have had crashes on both stock sun hardware (Ultra 10) as well as some tadpole hardware. the failure is in NFS mount root. [22:21:11] *** bougie has quit IRC [22:21:12] <oxygene> sickness: you updated the version in your tree? [22:21:26] <sickness> yeah [22:21:35] <sickness> with your .tar.bz2 [22:21:45] <delewis> gdamore: you mean a Jumpstart? [22:21:48] <gdamore> what i normally do is lofiadm the dvd on my install server, export it via NFS, and then run addclient. is there some problem with NFS export of lofs ? [22:21:49] <delewis> i do it all the time [22:21:57] <gdamore> yes. [22:22:08] <delewis> I've done Jumpstarts on my SB1000 since Solaris 10 FCS (tons of Solaris Express releases), and S10u2 [22:22:16] <oxygene> sickness: two or three patches were folded in in the meantime and thus disappeared.. best you delete everything in multimedia/mplayer, and then re-unpack that tarball [22:22:31] <oxygene> sickness: it's hard to add "delete" statements to a tarball :) [22:22:34] <gdamore> am i missing something obvious? i'm not using custom jumpstart. [22:22:40] <delewis> gdamore: I do an lofiadm on the ISO, mount it, and run the install_server script [22:22:43] <sickness> oxygene: tnx, I'll try this now :) [22:22:46] <sickness> oxygene: I know... :) [22:23:11] <gdamore> install_server installs a full copy of the DVD onto local disk, IIRC [22:23:31] <delewis> oh, you aren't doing a Jumpstart then [22:24:15] <gdamore> i just want to do a network install. not _custom_ jumpstart. i.e. install using an NFS lofs cdrom. this works with earlier releases, IIRC. [22:24:16] <sickness> oxygene: the dir is also 0777, is it ok if I chmod it back to 0644 by hand? [22:24:26] <oxygene> sickness: sure [22:24:33] <oxygene> 0777? *looks* [22:24:42] <oxygene> heh, indeed [22:24:43] <sickness> yeah, it's 777 :) [22:24:46] <sickness> but no prob :) [22:25:45] <sickness> stupid newbe users like me usually notice those things, because I'm too stupid to actually read code :P [22:27:32] *** nwf has quit IRC [22:27:46] <sickness> oxygene: ok, compiling now :))) [22:29:12] *** nwf has joined #opensolaris [22:29:55] <sickness> oxygene: ok it was too nice to be true... new prob: http://www.rafb.net/paste/results/tfa0CB22.html [22:29:58] <sickness> =) [22:31:00] <oxygene> sickness: what are you running on, nevada? [22:31:10] <sickness> yeah, snv50 [22:32:18] <oxygene> okay.. on 10u2 (what I'm using exclusively right now for various reasons), support for cddb doesn't get compiled, so I didn't run into that [22:33:05] <delewis> the libcdio that's shipped with Solaris is broken [22:33:12] <delewis> it's missing various symbols (or some symbols are hidden) [22:33:50] <sickness> oxygene: can I remove cddb? [22:34:01] <delewis> sickness: --disable-libcdio [22:34:23] <oxygene> sickness: add that to CONFIGUREOPTS in pkg.mk [22:34:32] <sickness> k [22:35:05] <sickness> trying now :) [22:36:21] *** Odin- has joined #opensolaris [22:36:59] <sickness> again :/ [22:37:39] <sickness> http://www.rafb.net/paste/results/hSWKyK78.html [22:37:48] <oxygene> hmm.. I really have to get a nevada box up and running again :) [22:38:46] <oxygene> sickness: there are 4 implementations of this function in stream_cddb.c - I have no idea which is the right one (or if any of these is actually usable :) [22:39:20] <sickness> heh [22:39:55] <oxygene> sickness: ugly workaround: add "int read_toc(const char*dev){return -1;}" at the end of the file (basically makes any attempt to read cddb data fail) [22:39:56] <sickness> what if I go back to the original mplayer dir of pmpkg? [22:40:07] <sickness> ok [22:41:21] <oxygene> the original mplayer dir should work.. look in files/ if there's a patch-200609* - that's the one that fixed some win32 codec lookup issues [22:41:57] <sickness> what's the file to add that line? stream.a ? [22:42:11] <sickness> uhm, I suppose it's not a .a file =) [22:42:39] <oxygene> stream/stream_cddb.c [22:42:44] <sickness> tnx! [22:43:35] *** sylvain has quit IRC [22:44:28] *** Cybernd has joined #opensolaris [22:44:49] <sickness> oxygene: iii built :) [22:45:06] <sickness> Installation of <PMmplayer> was successful. [22:45:08] <sickness> :P [22:45:10] <oxygene> :) [22:45:21] <oxygene> just don't expect that cddb stuff to work ;) [22:46:48] <sickness> ok, what's cddb after all? that online http database of cd titles? [22:47:24] <postwait> anyone know what this means? ldap_search: Can't connect to the LDAP server - Not owner [22:47:59] <postwait> Google shows me nothin' for "Not Owner" [22:48:07] <oxygene> sickness: yes [22:48:15] <sickness> ok, so I don't need that :))) [22:51:46] *** ShadowHntr has quit IRC [22:56:45] *** jmcp has joined #opensolaris [22:58:44] *** gnu2it2 has quit IRC [23:05:18] *** coffman has joined #opensolaris [23:07:18] *** mega has quit IRC [23:15:10] *** deather_ is now known as deather [23:16:27] *** GmanAFK is now known as Gman [23:16:29] <Gman> hey dudes! [23:16:40] <_william_> hi Gman [23:16:51] <Error_404> hmm... in theory BIND is already installed [23:19:36] <Gman> hi _william_ [23:21:41] * Gman giggles reading roy's reply to john on the arc stuff [23:22:20] <alanc> yeah - I don't know why plocher insists on using the e-mail system as a membership roster instead of a web page beyond "that's what ARC has always done" [23:22:56] <Gman> exactly :) [23:23:17] <Gman> the assumption that the arc process works so why change it perhaps [23:30:05] <alanc> *sigh* [23:30:12] <alanc> The reason it is being held: Message has implicit destination [23:30:24] <alanc> PSARC-EXT MUST DIE! [23:31:12] <Gman> heh [23:31:50] <stevel> "In addition, fix the "implicit destination" moderation hell resulting from the sac mail filter Bcc'ing to the opensolaris aliases." [23:31:53] <stevel> didn't i fix that? [23:32:01] <asyd> \_o< [23:32:18] <Gman> [sigh] [23:32:24] <Gman> i wish the opensolaris lists would archive like this - [23:32:26] <Gman> http://mail.gnome.org/archives/desktop-devel-list/2006-September/author.html [23:32:30] <Gman> so nice. so nice. [23:33:02] <stevel> i think we need a newer mailman version [23:33:10] <stevel> it's not awful though [23:33:13] <stevel> http://mail.opensolaris.org/pipermail/opensolaris-arc/2006-October/author.html#start [23:33:32] <Gman> no, not terrible [23:34:01] <Gman> i think gnome uses an external archiving utility [23:34:06] <Gman> doesn't rely on mailman directly [23:34:53] <Gman> stevel, would it help if the current mailman version has security vulnerabilities? :) [23:35:07] <stevel> sure. security vulnerabilities always help [23:35:29] <Gman> i'm pretty sure that i've seen a bunch of mailman releases to fix those recently [23:36:09] *** nwf has quit IRC [23:36:34] <Gman> http://lwn.net/Articles/203086/ [23:36:37] *** Tzoa has joined #opensolaris [23:37:20] *** Mazon is now known as mazon [23:39:17] *** laca is now known as lacaAFK [23:42:05] <_william_> bn [23:42:08] <_william_> gn in english :) [23:42:09] *** _william_ has left #opensolaris [23:42:24] *** Cybernd has quit IRC [23:43:24] *** sahafeez has quit IRC [23:47:18] <trs81> Gman: mhonarc vs pipermail [23:47:30] <Gman> nod [23:47:36] <trs81> gah [23:47:46] <trs81> I have libucb entering my build somehow [23:47:58] <trs81> and I don't know where it's coming from [23:48:21] <alanc> make sure you're not using /usr/ucb/cc [23:51:41] *** mage2 has quit IRC [23:54:19] <trs81> pretty sure I'm not [23:56:22] <trs81> hmm, I am using /usr/ucb/ld though [23:56:37] <jengelh> *cough* ON lacks a decent build system like in linux-2.6 [23:56:58] <Error_404> heh [23:57:08] <trs81> ON? I'm building gnome [23:57:38] <jengelh> i'm quite uncomfortable with the fact that `gcc -D_KERNEL` is all you need for kernel modules ;-/ [23:57:49] * trs81 fiddles his PATH