March 1, 2010 
< | 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 | >
March 1, 2010 [00:06:17] *** iceq_ has joined #opensolaris[00:07:05] *** iceq has quit IRC[00:08:15] *** ingenthr has joined #opensolaris[00:08:37] *** Tiger^ has quit IRC[00:08:44] *** Tiger^ has joined #opensolaris[00:11:20] *** throatwarbler has joined #opensolaris[00:11:25] *** throatwarbler has left #opensolaris[00:11:47] <longcat> LOL[00:14:20] *** Stellar has quit IRC[00:18:29] *** chilicuil has quit IRC[00:19:30] *** The-spiki has joined #opensolaris[00:21:28] *** twx has quit IRC[00:22:02] *** AxeZ has quit IRC[00:22:50] *** raichoo has left #opensolaris[00:23:01] *** alecmuffett has quit IRC[00:23:09] *** ingenthr has quit IRC[00:23:59] *** Stellar has joined #opensolaris[00:25:40] *** vraa has quit IRC[00:27:41] *** vraa has joined #opensolaris[00:27:52] *** jiraia has joined #opensolaris[00:29:22] *** alecmuffett has joined #opensolaris[00:33:26] *** ruomad has quit IRC[00:34:24] *** silas- has quit IRC[00:45:57] *** gosx has joined #opensolaris[00:48:50] *** wdp_ has quit IRC[00:49:18] *** gosx_ has quit IRC[00:50:18] *** gebi has quit IRC[00:50:20] *** bantu has quit IRC[00:50:46] *** gm152 has quit IRC[00:50:49] <Atomdrache> win 23[00:50:51] <Atomdrache> Oops.[00:54:56] *** Stellar has quit IRC[00:55:45] *** Stellar has joined #opensolaris[00:55:46] *** ma3x has joined #opensolaris[00:57:14] *** datadigger has quit IRC[00:57:29] *** winstonw has quit IRC[00:57:58] * nachox is running nightly and thinks it will actually take all night[00:58:47] *** wbkang has joined #opensolaris[00:59:39] *** fOB_2 has quit IRC[01:00:18] *** snuff-work has quit IRC[01:01:19] *** skeeziks has joined #opensolaris[01:05:06] *** coffeetime has quit IRC[01:07:02] *** vraa has quit IRC[01:08:29] <nachox> hmm, ok, it didnt take all night because it failed[01:08:43] *** snuff-work has joined #opensolaris[01:09:04] <oninoshiko> i hate it when that happens[01:09:16] <nachox> i think i missconfigured opensolaris.sh becuase it is looking for a compiler in /home/nacho/src/onnv/usr/src/tools/proto/opt/onbld/bin/i386/cw[01:09:28] <nachox> my workspace is in /home/nacho/src/onnv/[01:09:43] <sommerfeld> that's actually correct.[01:09:47] *** winstonw has joined #opensolaris[01:09:49] <sommerfeld> "cw" is the "compiler wrapper"[01:10:03] <nachox> cw is in /opt/onbld/bin/i386/cw[01:10:23] <sommerfeld> cw is built out of $CODEMGR_WR/usr/src/tools/[01:10:43] <sommerfeld> and nightly builds it by default as part of the tools build before moving on to the rest of the system[01:10:46] <nachox> cw is what is used to pass the correct options to each compiler as far as i know, its used for shadow compilation[01:10:49] *** snuff-work has quit IRC[01:11:01] <sommerfeld> right.[01:11:13] *** snuff-work has joined #opensolaris[01:11:26] <sommerfeld> takes Sun CC arguments, translates to gcc for gcc and passes through as-is for cc[01:11:49] *** snuff-work has quit IRC[01:12:18] <sommerfeld> nightly now builds tools by default because it doesn't take very long and it eliminates a major source of build errors (when your onbld is too old to build your source tree).[01:13:38] *** snuff-work has joined #opensolaris[01:14:25] <nachox> http://pastebin.com/fpUrK1Ea that is the relevant nightly part[01:14:36] <nachox> nacho@saturn:/opt/sstudio$ /opt/SUNWspro/bin/CC -V[01:14:37] <nachox> CC: Sun C++ 5.9 SunOS_i386 Patch 124864-12 2009/04/21[01:14:47] <nachox> that is where i have the compiler[01:15:02] <sommerfeld> what's the version of /opt/SUNWspro/bin/cc ?[01:15:18] *** bremby has joined #opensolaris[01:15:26] *** snuff-work has quit IRC[01:15:34] *** niq has quit IRC[01:16:05] <nachox> nacho@saturn:/opt$ /opt/SUNWspro/bin/cc -V[01:16:05] <nachox> cc: Sun C 5.9 SunOS_i386 Patch 124868-10 2009/04/30[01:17:02] <nachox> SPRO_ROOT=/opt/SUNWspro; export SPRO_ROOT[01:17:03] <nachox> is what is in opensolaris.sh[01:17:33] *** snuff-work has joined #opensolaris[01:18:12] *** tnoergaard has joined #opensolaris[01:18:13] <sommerfeld> argh, gotta run.[01:18:17] *** sommerfeld has quit IRC[01:20:10] *** iceq has joined #opensolaris[01:20:23] *** snuff-work has quit IRC[01:20:30] *** iceq_ has quit IRC[01:21:01] <nachox> crap :P[01:21:36] *** snuff-work has joined #opensolaris[01:22:05] *** ma3x has quit IRC[01:22:23] *** snuff-work has quit IRC[01:24:18] *** bremby has left #opensolaris[01:26:25] *** ma3x has joined #opensolaris[01:33:39] *** Stellar has quit IRC[01:34:25] *** Stellar has joined #opensolaris[01:35:23] *** gm152 has joined #opensolaris[01:35:42] *** Megaf_ has joined #opensolaris[01:36:58] *** hrist has quit IRC[01:37:32] *** hrist has joined #opensolaris[01:38:46] *** Megaf has quit IRC[01:41:57] *** JakeSays has quit IRC[01:42:47] *** rmorse has quit IRC[01:43:13] *** JakeSays has joined #opensolaris[01:44:20] *** atomicpunk has quit IRC[01:44:37] <nachox> ok, i set BUILD_TOOLS to /opt, lets see what it does[01:48:29] <nachox> damn, the rss of the package manager were a few hundread mbs of ram[01:49:11] *** dalibor has joined #opensolaris[01:49:28] *** piwi has quit IRC[01:50:46] *** snuff-work has joined #opensolaris[01:58:58] *** coffeetime has joined #opensolaris[02:02:04] *** timeless_mbp has quit IRC[02:02:44] *** timeless_mbp has joined #opensolaris[02:02:44] *** timeless_mbp has joined #opensolaris[02:09:50] <nachox> its working now[02:17:53] *** The-spiki has quit IRC[02:18:53] *** Canar has quit IRC[02:19:44] *** fleyta has joined #opensolaris[02:21:02] <fleyta> alanc: Are you awake?[02:21:51] <fleyta> Gman: Are you awake?[02:23:53] *** burungbesar has joined #opensolaris[02:23:59] *** derchris has quit IRC[02:24:05] *** derchris has joined #opensolaris[02:24:40] <hrist> fleyta: you are awake![02:24:59] <fleyta> yes[02:26:31] *** dalibor has quit IRC[02:29:22] <fleyta> Is anyone awake in this channel awake?[02:29:36] <burungbesar> :)[02:31:14] <hrist> fleyta: you are awake :P[02:31:17] <hrist> so are burungbesar and me[02:33:16] *** jfisc has quit IRC[02:33:52] <fleyta> Anyone worked with Roland Mainz or know him?[02:34:03] <g4lt-mordant> who?[02:34:31] <fleyta> Roland Sebastian Mainz?[02:34:54] <g4lt-mordant> never head of him, does he work for fraunhofer or something?[02:35:46] <fleyta> not employed currently[02:36:00] *** logic855 has quit IRC[02:37:32] *** kwinz3 has quit IRC[02:37:59] *** logic has joined #opensolaris[02:38:07] *** logic is now known as logic855[02:38:52] *** Mdx4 has quit IRC[02:45:22] *** Odin- has quit IRC[02:46:17] *** drewj has joined #opensolaris[02:47:48] *** piwi has joined #opensolaris[02:47:57] <nachox> g4lt-mordant, you have heard of him, the ksh lead[02:47:59] <fleyta> Is there a different channel where I can ask community questions?[02:48:27] *** atomicpunk has joined #opensolaris[02:49:54] <supahdupah> anyone have experience with link aggregates?[02:50:07] <piwi> fleyta: i think his irc name is gisburn[02:50:18] <fleyta> ah[02:50:25] <supahdupah> i have to set it up after a reboot[02:50:39] <fleyta> piwi: gisburn or nrubsig[02:50:39] <supahdupah> dladm still shows it[02:50:47] <piwi> both :)[02:50:55] <supahdupah> have to plumb and ifconfig again[02:51:04] <piwi> dont know what he uses in irc, but i think he uses both names[02:51:25] <piwi> fleyta: as you see here: http://www.nrubsig.org/[02:51:30] *** kwinz3 has joined #opensolaris[02:51:59] <piwi> <meta name="Author" content="Roland Mainz, nrubsig.org"> <meta name="Description" content="gisburn's home page">[02:56:21] <g4lt-mordant> nachox, yeah, but I don't tolerate metaquestions[02:57:07] *** coffeetime has quit IRC[02:57:55] <nachox> g4lt-mordant, sorry, i ruined you cover then :P[02:58:02] *** kwinz3 has quit IRC[02:58:18] <nachox> it seems nightly is running now[02:58:21] <nachox> i'm happy[02:59:05] *** brontide has joined #opensolaris[03:01:46] *** wbkang has quit IRC[03:03:11] *** burungbesar has quit IRC[03:03:20] *** vraa has joined #opensolaris[03:06:59] *** CodeWar has left #opensolaris[03:08:12] *** forq has joined #opensolaris[03:10:06] *** kwinz3 has joined #opensolaris[03:15:07] *** piwi has quit IRC[03:16:44] *** vraa has quit IRC[03:20:23] <lewellyn> sstallion_work: pong?[03:24:10] <brontide> does anyone know if there are problems with dns/multicast or avahi-bridge-dsd in dev_133? I think it's setup right, but just doesn't want to do what I expect.[03:25:05] *** forq has quit IRC[03:28:26] *** milek_ has quit IRC[03:29:32] *** cr37 has joined #opensolaris[03:35:49] *** forq has joined #opensolaris[03:36:09] *** cr37 has left #opensolaris[03:42:21] *** psychicist has quit IRC[03:42:49] *** brontide has quit IRC[03:44:29] <nachox> lewellyn, i saw you're running for an ogb place, congrats[03:44:38] <lewellyn> thanks :)[03:45:52] <nachox> i didnt know you were a core contributor[03:47:46] <nachox> i'll have to leave this nightly running it seems, i didnt start it on a screen[03:50:50] <lewellyn> i'm not... (note that i didn't nominate myself)[03:50:59] <lewellyn> (at least at this moment i am not)[03:51:07] *** TheBentZone has quit IRC[03:51:36] <nachox> oh, i see[03:51:37] *** babyju has quit IRC[03:53:18] *** vraa has joined #opensolaris[03:53:32] *** jiraia has quit IRC[03:54:39] *** sstallion has joined #opensolaris[03:54:47] <lewellyn> sstallion: pong[03:54:54] <sstallion> lewellyn: no worries, figured it out[03:55:21] <lewellyn> ok. :)[03:55:22] <sstallion> fm is giving me grief after an update to snv_132[03:56:27] <lewellyn> uh oh :([03:56:28] <sstallion> qlc is being a bit of a PITA[03:57:23] <CIA-37> Roger A. Faulkner <Roger.Faulkner at Sun dot COM>: PSARC 2009/657 delete obsolete system call traps, 6906485 delete obsolete system call traps[03:57:41] *** sstallion has quit IRC[04:00:26] *** vraa has quit IRC[04:02:58] <nachox> lewellyn, you dont happen to know how long on takes to build on a single 2.2ghz proc with 1.5 gb of memory, right? :P[04:04:41] *** vraa has joined #opensolaris[04:04:42] *** sstallion has joined #opensolaris[04:04:43] <sstallion> hrmm[04:04:43] <lewellyn> haha. no :([04:04:53] * sstallion regrets not spending more time with fmadm[04:05:05] *** forq has quit IRC[04:06:02] <sstallion> lewellyn: how can I get fmdump to stop reporting entries?[04:06:13] <sstallion> theres nothing wrong with the hardware - this was just an artifact of the upgrade[04:06:59] <snuff-work> sstallion i think u want the 'fmadm aquit'[04:07:04] <snuff-work> and the uuid[04:07:13] <sstallion> snuff-work: already did. Its still popping up in fmdump[04:07:34] <lewellyn> sstallion: i don't know. fm is still something that i love existing but i fear using[04:07:50] *** reflect has quit IRC[04:08:01] <sstallion> I remember fixing this a year or so ago - should have taken notes[04:09:18] *** babyju has joined #opensolaris[04:11:16] <nachox> fmadm acquit uuid[04:11:49] <sstallion> nachox: already acquited; entries are still showing up in fmdump[04:12:18] <nachox> you mean new entries[04:12:21] <nachox> right?[04:12:43] <sstallion> is it possible to purge the fmdump history?[04:12:59] <Triskelios> I've been wondering the same thing, fmdump shows several years of history even if I acquit the UUIDs[04:13:03] <sstallion> fmadm faulty looks fine, I'd rather clear out the fmdump history[04:13:39] * sstallion debates playing with /var/fm/[04:15:45] <nachox> hmm fmadm flush fmri?[04:17:02] <kohju> excuse me, someone knows where I can buy the OpenSolaris subscription ?[04:17:03] <sstallion> yeah[04:17:08] <sstallion> trying to figure out the fmri[04:17:55] <sstallion> hrmm.. that didn't seem to do it[04:18:18] <nachox> if you dont care about any of the things in fmdump you can always rotate the log and move out the old log backup i guess[04:18:31] <lewellyn> are the opensolaris.com maintainers still around? :P[04:19:01] <g4lt-mordant> kohju, opensolaris subs are in a state of flux, orrible's reworking them right now[04:19:08] <nachox> fmadm rotate fltlog[04:19:50] <sstallion> nachox: that worked well enough. thanks![04:20:23] <nachox> no problem[04:21:08] <jamesd2> sstallion did that sxce transfer complete?[04:21:33] <nachox> sstallion for the record, you always want fmdump faulty as that will tell you the current status and not historic date[04:21:41] <sstallion> jamesd2: it did, but I removed it earlier[04:22:03] <sstallion> nachox: right. I wanted to make sure the history was cleared in the event of a genuine failure later on[04:22:14] <nachox> err fmadm[04:22:15] <kohju> g4lt-mordant, hmm.... I thought so, too. ..[04:23:50] <sstallion> jamesd2: hang on I'll see about getting the file moved to a more permanent location[04:24:00] <jamesd2> ok[04:25:18] *** skeeziks has quit IRC[04:26:08] <sstallion> jamesd2: 30min ETA[04:26:26] <jamesd2> np[04:27:10] *** forq has joined #opensolaris[04:27:13] *** TommyTheKid has joined #opensolaris[04:27:25] <nachox> i'll leave this running and get some sleep, night guys[04:28:21] *** CodeWar has joined #opensolaris[04:29:32] *** Kedare has quit IRC[04:32:44] *** sstallion has quit IRC[04:37:07] *** forq has quit IRC[04:40:51] *** taemun has quit IRC[04:41:25] *** m3lling has joined #opensolaris[04:42:54] *** sommerfeld has joined #opensolaris[04:42:54] *** ChanServ sets mode: +o sommerfeld[04:43:31] *** fleyta has quit IRC[04:44:36] *** CodeWar has quit IRC[04:48:30] *** lblume has joined #opensolaris[04:52:44] *** darrenb` has quit IRC[04:56:14] *** taemun has joined #opensolaris[05:01:21] *** darrenb has joined #opensolaris[05:01:36] *** wrapster has quit IRC[05:07:04] *** ball has joined #opensolaris[05:07:48] *** sstallion has joined #opensolaris[05:07:52] <sstallion> jamesd2: around?[05:07:59] <jamesd2> yeah[05:08:24] *** m3lling has quit IRC[05:08:27] <ball> alanc: awake?[05:08:37] <alanc> sort of[05:09:46] <ball> alanc: I tried launching vncserver from the command line again and metacity didn't launch. Not a big deal of course because I was able to launch it manually, but I did say that I'd let you know.[05:11:28] <alanc> might have to force compiz off - http://defect.opensolaris.org/bz/show_bug.cgi?id=9891 mentions the issue, but only tells how to fix via the preferences dialog[05:16:27] <ball> Thanks, I'll try that.[05:16:48] *** GnuWulf has joined #opensolaris[05:18:42] *** djl1 has joined #opensolaris[05:21:51] *** gosx_ has joined #opensolaris[05:25:07] *** gosx has quit IRC[05:25:27] <ball> alanc: 'Visual Effects' is set to 'none'. Does that mean Compiz is turned off?[05:25:36] <alanc> yes[05:25:42] <alanc> guess that's not the issue then[05:26:12] <alanc> I don't remember where errors are logged by vncserver to see if there's any reason given for metacity not starting[05:26:58] <ball> Would you expect gnome-session to launch metacity, or is that something I should add to my xstartup file?[05:29:45] <alanc> I'd expect gnome-session to do it[05:30:19] <alanc> I'd have to ask gnome folks why it's not[05:30:39] *** Crothers has quit IRC[05:30:46] <ball> I was impressed that Gnome even worked on vncserver.[05:31:20] *** Crothers has joined #opensolaris[05:31:22] *** atomicpunk has quit IRC[05:31:33] <lewellyn> why? it's worked in vnc for years. it's just a bunch of X apps[05:32:23] <ball> lewellyn: I'm used to a platform where I've never got Gnome working (though in fairness I've not tried all /that/ hard ;-)[05:33:43] <alanc> compiz on the other hand... well, VNC isn't a good environment for heavy opengl use[05:33:53] <lewellyn> does any opengl work in vnc yet?[05:34:04] <lewellyn> i didn't think it implemented any of the GL extensions[05:34:18] <ball> lewellyn: ...but I also find the odd app that makes assumptions, including the assumption that they're running locally (e.g. Thunderbird taking a long time to pop up the animated new mail announcement in the bottom right corner of my screen)[05:34:36] <ball> lewellyn: VNC helps with that though.[05:34:44] <lewellyn> um. as far as anything is concerned, vnc is local[05:36:08] <ball> lewellyn: Right, I noticed the thunderbird thing when I was running it as a remote X client across the Internet[05:36:16] <ball> (tunnelled through ssh)[05:36:38] <lewellyn> even then, it wouldn't know you're non-local if you're using tunneling[05:37:09] <lewellyn> gecko apps tend to be slow with screen updates though. they're heavy. you just don't tend to notice it if you really are local[05:37:18] <lewellyn> ball: were you using cygwin/x or xming?[05:37:28] <ball> lewellyn: No, just X.[05:37:36] <lewellyn> on what platform?[05:37:46] <lewellyn> there are few windows x servers which don't suck, imo :P[05:37:47] <sommerfeld> sounds like there's a need for a "even if it looks like i'm local, pretend i'm not" option.[05:37:54] <ball> lewellyn: NetBSD usually.[05:37:56] <sommerfeld> i'd leave that turned on by default because i'm a ui minimalist[05:38:04] <ball> lewellyn: Sometimes Linux or OpenSolaris[05:38:13] <lewellyn> sommerfeld: that or make the ui less heavy :P[05:38:26] <lewellyn> ball: hm. and you used a lightweight cipher and compression?[05:38:31] <lewellyn> (X compresses pretty well)[05:39:43] <lewellyn> rdp over an ssh tunnel is just as affected by proper cipher selection and enabling compression[05:39:48] <lewellyn> it's not unique to X :)[05:39:59] <ball> lewellyn: I used compression and whatever the default cipher is there.[05:40:18] <ball> ...don't think there's much overhead with the cipher once the keys are exchanged.[05:41:31] <lewellyn> i spent many hours testing a while back. blowfish really improved interactive performance for tunneled X. i've gotta sit around and figure out the best choice of the current batch of acceptable ciphers[05:42:19] <ball> lewellyn: I tend to use VNC whenever I think of it (for remote connections)[05:42:21] *** ingenthr has joined #opensolaris[05:42:31] <lewellyn> ball: it benefits from ssh's compression, too[05:43:04] * ball nods[05:43:52] <ball> I think if Thunderbird just popped up a status message it wouldn't be so noticable. It's that they try to make it slide onto the screen and then slide back off. This takes a long time when it has to happen over the Internet.[05:44:22] <ball> ...fancy effects because they assume you're always running it locally on the machine with the X server.[05:44:24] <lewellyn> s/ when .*$//[05:44:34] <ball> That's my suspicion, anyway.[05:44:45] *** c0t0d0s0_ has joined #opensolaris[05:44:50] <lewellyn> firefox is just as bad about random redraws[05:45:12] <alanc> lewellyn: I think Xvnc reports that GLX is available, via the Mesa software renderer[05:45:13] <ball> PC mentality perhaps.[05:45:23] <lewellyn> alanc: ew :([05:45:27] <sommerfeld> i actually turn off notifications because they interfere with my work[05:45:41] <lewellyn> ball: optimization of useless shit, that's my suspicion[05:46:09] <ball> sommerfeld: I didn't even know that was possible.[05:46:49] <ball> That reminds me, I need to see whether that chap from IBM has written back to me yet.[05:47:40] <ball> oops. Take back all I said about Thunderbird. I'm using SeaMonkey on the box in question.[05:48:32] <sommerfeld> ball: Edit -> Preferences -> General: "When new messages arrive..."[05:49:06] <supahdupah> anyone have some tips on network trouble shooting?[05:49:09] <lewellyn> ball: seamonkey has lots of other issues. but it is lighter on ram than firefox + thunderbird :)[05:49:10] <sommerfeld> show an alert/play a sound. uncheck both boxes[05:49:18] <supahdupah> have an aggregate set up[05:49:23] * ball nods[05:49:32] <supahdupah> i get an ip via dhcp[05:49:33] *** ingenthr has quit IRC[05:49:42] <supahdupah> can't ping anything including lan[05:50:04] *** c0t0d0s0_ has quit IRC[05:51:14] <ball> lewellyn: That's pretty much our default Internet suite. I was confused because I recently configured Thunderbird for my personal email account.[05:51:28] <supahdupah> everything was working great until i shutdown the server and physically moved it[05:52:10] <ball> supahdupah: Does your server have a Link LED on the Ethernet jack?[05:52:19] <supahdupah> yeah they are both active[05:52:32] <supahdupah> on the switch as well[05:52:41] <supahdupah> the link has an ip from the router[05:52:45] <ball> Are both Ethernet ports linked to the same switch?[05:52:49] <supahdupah> yes[05:53:04] *** Norrin has joined #opensolaris[05:53:11] *** lattera has quit IRC[05:53:30] <ball> supahdupah: Do they have different MAC and IP addresses from each other?[05:53:33] <sommerfeld> does the switch see the aggregation as an aggregation?[05:54:15] *** c0t0d0s0_ has joined #opensolaris[05:54:18] <Triskelios> supahdupah: does the switch support LACP or do you have the aggregation configured on the switch (switch settings might call it bonding or trunking)?[05:55:13] <supahdupah> it does[05:55:22] <supahdupah> this was working before i moved the server[05:55:26] <supahdupah> same switch[05:55:51] <ball> Did you try cold-booting the server before you moved it?[05:55:59] <lewellyn> ball: seamonkey 2 is heavier in some ways and lighter in others. perhaps you should play with both 1.1.x and 2.0.x to see performance results over X :)[05:56:04] * lewellyn heads off to dinner[05:56:06] *** NorrinRadd has quit IRC[05:56:19] *** swa_work has quit IRC[05:56:30] <supahdupah> i've rebooted numerous times since[05:56:51] <ball> supahdupah: Did it work after you rebooted?[05:57:02] <supahdupah> well sorta[05:57:08] <supahdupah> it was super slow[05:57:30] <supahdupah> so tried setting LACPACTIVITY to active[05:58:07] <ball> define "super slow"[05:58:16] <ball> ...while I go and put the kettle on.[05:59:33] <supahdupah> well ssh was like i was typing on a serial terminal from the 60's[05:59:55] *** jkimball4 has quit IRC[05:59:56] <supahdupah> and one link on the switch kept blinking off[06:00:36] *** _setuid_H has quit IRC[06:00:58] *** jkimball4 has joined #opensolaris[06:01:01] *** jkimball4 has quit IRC[06:01:01] *** jkimball4 has joined #opensolaris[06:01:07] *** sstallion has quit IRC[06:01:59] <ball> supahdupah: Are the network interfaces on plug-in cards, or built into the planar?[06:03:03] <supahdupah> plugin[06:03:17] <supahdupah> i get an ip so i don't think its a physical issue[06:03:24] <supahdupah> but i am listening[06:03:37] <ball> supahdupah: Try it with one cable plugged in. See if it goes faster.[06:03:44] <supahdupah> its not going at all[06:03:49] <supahdupah> i can't ping anything[06:03:56] <supahdupah> besides localhost[06:04:18] <ball> Are you plugged into the same port(s) on the switch?[06:05:06] <supahdupah> probably not[06:05:56] <supahdupah> hmm i can't ping the server[06:06:16] <supahdupah> where is it getting an ip from?[06:06:21] *** torchie_ has quit IRC[06:06:25] *** ingenthr has joined #opensolaris[06:06:35] <ball> supahdupah: what kind of server?[06:06:45] <supahdupah> what do you mean?[06:06:54] <Triskelios> supahdupah: ping isn't a great diagnostic tool. check snoop, arp -n, etc...[06:06:57] *** Error404NotFound has joined #opensolaris[06:07:16] <supahdupah> gonna remove the mac assigned ip from the router[06:07:23] <supahdupah> and see if it still gets an ip[06:08:13] <Triskelios> supahdupah: make sure it's not from the local DHCP cache either[06:08:46] <Triskelios> see if it's actually receiving DHCP responses[06:09:50] <supahdupah> not sure how to do that[06:10:22] <Triskelios> truss/snoop/whatever in parallel with starting dhcp[06:10:27] <supahdupah> ok unplumbed the aggr[06:10:54] <supahdupah> how should i bring the aggr back up[06:10:58] <supahdupah> maybe i am doing ti wrong[06:11:58] <ball> Try making a single link work first.[06:16:29] <supahdupah> ok single works[06:16:40] <supahdupah> lemme try and use the other one by itself[06:18:00] <supahdupah> hmmm no go on the other link[06:18:13] <ball> There you go.[06:18:26] <supahdupah> weird was fine earlier[06:19:02] <ball> I have to go to bed.[06:19:05] *** ball has quit IRC[06:19:50] *** c0t0d0s0_ has quit IRC[06:20:24] <mui> http://img203.imageshack.us/img203/2134/meanwhileinfinland.jpg[06:21:40] *** c0t0d0s0_ has joined #opensolaris[06:22:07] <supahdupah> dam.. can't get an ip with the second device[06:22:58] *** jkimball4 has quit IRC[06:23:27] *** jkimball4 has joined #opensolaris[06:23:27] *** jkimball4 has joined #opensolaris[06:32:52] *** gm152 has quit IRC[06:37:30] *** tosh has quit IRC[06:46:47] *** radsy has quit IRC[07:01:43] *** c0t0d0s0_ has quit IRC[07:05:48] *** bitvector2 has joined #opensolaris[07:09:38] *** GnuWulf has quit IRC[07:09:43] *** c0t0d0s0_ has joined #opensolaris[07:13:15] *** Toxic__ has quit IRC[07:14:51] *** tnoergaard has quit IRC[07:20:42] *** merzo has quit IRC[07:29:15] <trochej> Lo[07:30:31] <mgssancho> ol[07:34:23] *** _madmax_ has quit IRC[07:35:13] *** encephalon has joined #opensolaris[07:40:09] *** ehf has joined #opensolaris[07:40:45] *** relling has quit IRC[07:42:09] *** TommyTK has joined #opensolaris[07:47:09] *** babyju has quit IRC[07:50:53] *** hsp has joined #opensolaris[07:51:46] *** vraa has quit IRC[07:51:52] *** hajma has quit IRC[07:52:42] * lewellyn hands trochej some carob coffee[07:53:07] * mgssancho intercepts[07:53:11] *** Gman has quit IRC[07:53:25] *** Gman_ has joined #opensolaris[07:57:13] <lewellyn> trochej: aren't you glad there was someone to save you from that hell? :)[07:57:39] * lewellyn finds carob "coffee" revolting[07:59:16] *** Zauberpo1y has joined #opensolaris[07:59:24] *** TommyTheKid has quit IRC[07:59:41] <bitvector2> carob coffe?[08:00:56] *** Zauberpony has quit IRC[08:01:11] <pitty> anyone have any sun related research on nahelem chips, the pros and cons of using[08:01:15] <mui> coffee ack[08:01:18] <pitty> sorry[08:01:21] *** pitty has quit IRC[08:01:33] <mui> o.O[08:01:36] <gdamore> code review request: removal of legacy esp/dma bits: http://cr.opensolaris.org/~gdamore/esp/[08:01:38] <mui> and that was all about[08:02:15] <trochej> lewellyn: Now situation is more to my liking :)[08:02:57] <trochej> lewellyn: Now I can run around and bitch, how fucking I'd revolutionise (?) OGB if only I'd be allowed. :)[08:03:05] <lewellyn> hahaha[08:06:41] *** ehf has quit IRC[08:08:45] *** kwinz3 has quit IRC[08:09:58] *** phimic has joined #opensolaris[08:13:21] <g4lt-mordant> trochej, you'd port opensolaris to coffeemakers? ;P[08:20:30] *** Chipdancer has quit IRC[08:23:12] <trochej> Yup :)[08:24:19] *** LouisJB has joined #opensolaris[08:25:44] <sickness> http://www.tomshardware.com/reviews/wd-4k-sector,2554-3.html <- what about opensolaris+zfs?[08:25:52] <lewellyn> which will involve bigger coffeemakers :)[08:26:39] <sickness> can you use 4Kb sector disks?[08:27:11] <lewellyn> you've asked this already[08:27:17] <lewellyn> or someone has very recently[08:31:35] <sickness> not me, sorry, what was the response anyway? =)[08:32:12] *** LouisJB has quit IRC[08:32:25] <lewellyn> probably that zfs will use whatever the drive reports[08:32:32] *** LouisJB has joined #opensolaris[08:34:37] <lewellyn> and if you're concerned about sector alignment, pre-partition however the hell you like :)[08:35:22] *** Chipdancer has joined #opensolaris[08:36:39] *** chilicuil has joined #opensolaris[08:36:43] *** easytiger has quit IRC[08:37:13] *** LouisJB has quit IRC[08:40:47] *** merzo has joined #opensolaris[08:45:42] *** Yu\2 has joined #opensolaris[08:46:25] *** |AbsyntH| has joined #opensolaris[08:47:44] *** sponix has joined #opensolaris[08:49:19] *** kwinz3 has joined #opensolaris[08:50:45] *** supahdupah has quit IRC[08:57:23] <CIA-37> Victor Li <Victor.Li at Sun dot COM>: 6927316 cfgadm -c unconfigure followed by mpathadm list lu panics system[09:03:11] *** c0t0d0s0_ has quit IRC[09:06:27] *** achester_ has joined #opensolaris[09:09:44] *** hajma has joined #opensolaris[09:12:04] *** _madmax_ has joined #opensolaris[09:12:27] *** __schily_ has joined #opensolaris[09:15:24] *** dunc has quit IRC[09:15:51] *** _schily__ has quit IRC[09:17:29] <lewellyn> gdamore: it's confusing that the webrev title is audioemu10k[09:18:00] *** mgssancho has quit IRC[09:22:02] *** houst0n has joined #opensolaris[09:22:49] *** hnhn has joined #opensolaris[09:23:25] *** chilicuil has quit IRC[09:24:19] *** comay has quit IRC[09:39:14] *** tsoome has quit IRC[09:39:45] *** raichoo has joined #opensolaris[09:51:17] *** c0t0d0s0_ has joined #opensolaris[09:52:33] *** gosx has joined #opensolaris[09:54:39] *** pogs has quit IRC[09:55:55] *** gosx_ has quit IRC[09:59:05] <trygvis> thunderbird 3.0.2 has a bug fix that I desperately need, how can I figure out if it will be included in b134 or any particular build?[09:59:28] <trygvis> and, in general, how are applications like that updated? When someone at sun has the time? a customer files a bug requesting an update?[10:00:59] *** Vinayak has joined #opensolaris[10:06:47] *** tosh has joined #opensolaris[10:11:45] *** LouisJB has joined #opensolaris[10:14:02] *** tnoergaard has joined #opensolaris[10:15:45] *** Mick27 has joined #opensolaris[10:22:23] *** _madmax_ has quit IRC[10:23:08] *** perlmongo has quit IRC[10:23:39] *** g4lt-mordant has quit IRC[10:24:59] *** c0t0d0s0_ has quit IRC[10:25:33] *** kwinz3 has quit IRC[10:25:47] *** kwinz3 has joined #opensolaris[10:27:30] *** wdp has joined #opensolaris[10:27:49] *** c0t0d0s0_ has joined #opensolaris[10:28:39] *** g4lt-mordant has joined #opensolaris[10:28:45] *** _madmax_ has joined #opensolaris[10:28:53] *** dunc has joined #opensolaris[10:30:40] *** pablomh has joined #opensolaris[10:30:46] *** KermitTheFragger has joined #opensolaris[10:31:02] *** Alasdairrr has quit IRC[10:32:22] *** Alasdairrr has joined #opensolaris[10:35:28] <lewellyn> trygvis: i try to file bugs and see if they get closed :)[10:35:58] <trygvis> would bugs.opensolaris.org be the right place to request an update of thunderbird?[10:36:38] <lewellyn> i think that's jds, which alanc said pays more attention to bugzilla[10:37:04] <trygvis> oh, right[10:41:49] *** c0t0d0s0_ has quit IRC[10:44:08] *** Erwann has joined #opensolaris[10:44:38] <Kapsel> i'm watching sysstat and nicstat - both give very different results on network traffic. which one should I consider the most correct reading?[10:45:03] <Stric> fabricate some load and see which one matches[10:45:49] <lewellyn> Kapsel: it's possible they are both correct. it's also possible that neither is correct.[10:45:49] <trygvis> yay, it's fixed already (#14836)[10:45:52] <lewellyn> trygvis: :D[10:47:06] *** timsf has joined #opensolaris[10:50:59] <PatrickS> is it expectable that the device names in the output of 'format', 'cfgadm -vl' and 'zpool status -v' did not match?[10:53:28] <lewellyn> PatrickS: "did not match"?[10:54:48] <PatrickS> lewellyn: hmm, there are devices names in 'zpool status' that do not occur in the 'format' output[10:55:20] <houst0n> Blimey. How did you manage that?[10:55:50] *** jamesd2 has quit IRC[10:55:59] <PatrickS> I assumed that the disk devices which make up a vdev of a zpool are listed under the same name in format and cfgadm[10:56:14] *** jamesd2 has joined #opensolaris[10:56:36] <houst0n> A safe assumption....[10:57:29] <CIA-37> Gangadhar Mylapuram <Gangadhar.M at Sun dot COM>: 6881648 running reboot in single user mode ( reboot -- -s ) after 141445-09 is applied gives errors[10:57:30] <CIA-37> Jan Kryl <Jan.Kryl at Sun dot COM>: 6927244 share_nfs charset option doesn't work when creating directories[11:00:04] <PatrickS> my motivation is: If zpool status gives a degraded pool where particular disks are faulty, I have to determine which one to 'cfgadm -c unconfigure' to remove it savely[11:02:08] *** Phenox_ has quit IRC[11:03:28] <lewellyn> PatrickS: do they show in rmformat?[11:05:15] <PatrickS> lewellyn: no, a simple 'rmformat' only show 2 USB devices[11:05:25] <lewellyn> hm.[11:06:11] <lewellyn> i dunno. did you ever move the disks around or something?[11:06:24] <PatrickS> is there a way to match the device names from 'zpool status -v' to the physical IDs?[11:06:34] *** MattMan has joined #opensolaris[11:06:51] <lewellyn> "physical IDs"?[11:07:07] <PatrickS> yes sure, I exported the pool, changed SATA cabeling and imported the pool[11:07:38] <lewellyn> PatrickS: perhaps that's the original disk IDs. i've never moved a pool[11:08:04] <lewellyn> but you could see what happens with an ls -l /dev/c4d0[11:08:11] <PatrickS> "physical ID" something like "/pci@0,0/pci8086,244e@1e/pci-ide@0/ide@1/cmdk@0,0"[11:08:29] * lewellyn repeats himself[11:09:47] <PatrickS> ls: /dev/c4d0: No such file or directory[11:09:57] <g4lt-mordant> note taht physical ID and /dev/dsk/blah are pretty strongly associated in solaris, you have to REALLY foul things up before a physical ID won't reassociate with the dame /dev/dsk[11:09:59] <lewellyn> so zpool is misreporting[11:10:11] <lewellyn> g4lt-mordant: that's what i was wondering about[11:10:38] <PatrickS> puhhh[11:10:48] <PatrickS> that pool is quite old[11:10:57] <lewellyn> i'd do a reconfiguration boot, for giggles[11:11:29] <PatrickS> touch /reconfigure + init 6 ?[11:12:04] *** kwinz3 has quit IRC[11:12:25] *** Newbie_ has joined #opensolaris[11:13:09] <Newbie_> currently need some help starting opensolaris[11:13:17] <Newbie_> problem booting[11:13:36] <Newbie_> can someone please help?[11:14:01] <g4lt-mordant> and we're supposed to read the problem off your forehead?[11:14:15] <Newbie_> ???[11:14:53] <raichoo> If you don't ask a question no one will be able or even willing to help.[11:15:00] <Newbie_> sorry[11:15:03] <lewellyn> Newbie_: give us enough information and we can try to help. otherwise, we can't do much except amuse ourselves at your expense[11:15:11] *** sponix has quit IRC[11:15:18] <Newbie_> i just received the cd[11:15:31] <Newbie_> tried running it on a pc without any os[11:15:52] <Newbie_> went into the screen which is similar to msdos command prompt[11:16:04] <Newbie_> and got stuck[11:16:10] <lewellyn> what did the screen say?[11:16:25] <Newbie_> it says subject to the license[11:16:33] <Newbie_> then stopped at that line[11:16:44] <Newbie_> waited for 3 hours[11:16:48] <Newbie_> no response[11:16:51] *** fredx_ng has joined #opensolaris[11:17:12] <lewellyn> is this 2009.06?[11:17:17] <Newbie_> yeah[11:17:24] <Newbie_> the version is 2009.06[11:17:24] <lewellyn> is your hardware on the HCL?[11:17:25] *** fredx_ng has left #opensolaris[11:17:45] <Newbie_> doesn't seems to be on the list[11:18:15] *** Phenox_ has joined #opensolaris[11:18:37] <lewellyn> it's possible that your system isn't compatible with 2009.06. a lot of hardware has been added since then. it may be worthwhile to try again on a different system to be sure it's not the disc.[11:19:10] *** fredxng has joined #opensolaris[11:19:18] <Newbie_> i just found out my motherboard is on the list[11:19:23] <Newbie_> on the hcl list[11:19:32] <raichoo> You could even try to boot in verbose mode. Maybe it's a single module that hangs and you just could disable that one.[11:19:51] <raichoo> I had a similar issue on my craptop.[11:20:09] <raichoo> Disabled the culprit and all went smooth.[11:20:26] <lewellyn> raichoo: it's also possible that it's a bum disc. the best replicators still can't guarantee 100% :([11:20:46] <raichoo> agreed[11:21:07] <Newbie_> how do i start the verbose mode?[11:21:15] <Newbie_> can kindly explain?[11:21:26] *** fredxng has left #opensolaris[11:21:27] *** fredxng has joined #opensolaris[11:21:56] <raichoo> add -v to the kernel line in grub[11:23:30] <raichoo> But to be sure that the disk isn't bum, do as lewellyn said. It's very possible that he is right with his assumption[11:24:09] *** swift has joined #opensolaris[11:24:17] *** dunc has quit IRC[11:24:38] <Newbie_> during the first bootup, I saw a list of choices[11:24:42] <Newbie_> and selected the first[11:24:46] <Newbie_> and waited[11:25:02] <raichoo> hit e to edit then edit the kernel line[11:25:10] <raichoo> just append -v to it end hit b to boot[11:25:21] *** kwinz3 has joined #opensolaris[11:25:25] <Newbie_> thanks[11:25:39] *** Newbie_ has quit IRC[11:25:53] <raichoo> Grub also offers online help[11:26:54] <nachox> damn nightly is still running after more than 8 hours[11:27:21] <raichoo> time for breakfast[11:28:20] *** himanshu_ has joined #opensolaris[11:28:36] <himanshu_> hello friends.[11:29:03] <himanshu_> does opensolaris support bluetooth technology?[11:29:12] *** milek_ has joined #opensolaris[11:29:45] <lewellyn> nachox: i tried it on my u5. the machine outright died after a week of nightly :D[11:29:46] *** Yu\2 has quit IRC[11:29:50] <lewellyn> himanshu_: a little[11:29:54] <lewellyn> smrt: explain bluetooth[11:29:54] <smrt> In build 107 and newer, you can install Bluetooth support. Only mice are supported at this time. See: http://www.opensolaris.org/os/project/bluetooth/[11:30:15] *** g4lt-mordant has quit IRC[11:30:15] <nachox> heh[11:30:46] <nachox> also the directory with the workspace has grown from 800mb to more than 3gb[11:30:53] <himanshu_> i have inbuilt bluetooth device in my laptop , i m unable to send images from my cell phone into my laptop[11:30:57] <lewellyn> nachox: yup.[11:31:15] <lewellyn> himanshu_: yes. as smrt said, you have to install the support, and it's currently limited to mice[11:32:08] <himanshu_> so , that means , not possible to connect to cell phone like device[11:32:36] <lewellyn> correct, for now.[11:32:41] <lewellyn> at least not via bluetooth[11:32:56] <himanshu_> fine , i will use the data cable then[11:33:48] <himanshu_> there's lots of development work to be done for this OS , opensolaris.[11:34:13] <lewellyn> of course. every operating system needs lots of development work :)[11:34:24] <himanshu_> :-)[11:35:45] *** g4lt-mordant has joined #opensolaris[11:36:50] <himanshu_> any idea , whats kind of knowledge will be required to create an application (as such mentioned above to detect bluetooth device on open solaris ), if its easier , may be i could develop it , as my major project[11:37:11] <himanshu_> it would be like writing down device drivers...[11:37:42] *** BSD_ has joined #opensolaris[11:37:58] <BSD_> got stuck in the use subjected to license screen[11:38:06] <BSD_> mine is 2009.06[11:38:36] <BSD_> can some help?[11:39:07] <himanshu_> BSD_ , start a fresh installation[11:39:43] <BSD_> can i use the live cd instead of installing?[11:40:50] <BSD_> tried editing the kernel and module lines[11:40:59] <BSD_> kept getting line 15 error[11:41:08] <BSD_> and press any key to continue[11:41:21] <BSD_> my hardware is supported on the hcl list[11:41:39] <BSD_> help would be appreciated[11:42:11] <lewellyn> BSD_: you're "Newbie_" from earlier?[11:42:11] <BSD_> mine is a 64bit system[11:42:15] <BSD_> ???[11:42:29] <lewellyn> 02:15 Newbie_ (cm249.omega13.maxonline.com.sg/218.186.13.249 - ht) [~daba0df9@gateway/web/freenode/x-wxaerlgdvwvnopdl] has joined #opensolaris[11:42:33] <lewellyn> 02:40 BSD_ (cm249.omega13.maxonline.com.sg/218.186.13.249 - ht) [~daba0df9@gateway/web/freenode/x-wulbvkposvnejxym] has joined #opensolaris[11:42:34] <BSD_> i think we shared the same isp[11:42:34] *** Beket has joined #opensolaris[11:42:47] <lewellyn> not the same isp, the same account ;)[11:43:06] <BSD_> all users who uses starhub's maxonline services in singapore uses the same ip[11:43:16] <BSD_> most likely will show up as same user[11:43:26] <lewellyn> ok. well... what, exactly, does this "line 15 error" say?[11:44:01] <BSD_> just error in line 15[11:44:07] <BSD_> nothing was mentioned[11:44:07] *** sponix has joined #opensolaris[11:44:13] <lewellyn> that's all the screen says?[11:44:24] <lewellyn> surely it said more than "Error in line 15"[11:44:26] <BSD_> that's all[11:44:38] <lewellyn> i've never seen that error[11:45:16] <lewellyn> i've seen similar errors, but none with that wording[11:45:25] <BSD_> maybe i try to screenshot it and post it in this channel[11:45:37] <lewellyn> that would be helpful :)[11:45:40] *** calumb has joined #opensolaris[11:45:46] <Berny> error 15 is just a "file not found"[11:45:46] <BSD_> but how do i screenshot it when i am not even in any os?[11:45:49] <BSD_> can help?[11:45:55] <lewellyn> Berny: line 15 is different[11:46:01] <lewellyn> BSD_: does your phone have a camera?[11:46:07] <Berny> ah my bad[11:46:12] <BSD_> sorry[11:46:14] <BSD_> no camera[11:46:21] *** dunc has joined #opensolaris[11:46:25] <BSD_> how to screenshot the error?[11:46:36] <lewellyn> you can't, without a separate camera[11:46:45] <BSD_> i try to see what i can do[11:46:53] <lewellyn> k. good luck[11:46:58] <BSD_> maybe i come back to this channel later[11:47:01] * lewellyn heads toward bed[11:47:03] <BSD_> thanks[11:47:17] <BSD_> thanks for the suggestions and help[11:47:38] <BSD_> logging out[11:47:42] *** BSD_ has quit IRC[11:47:45] *** Seemone has joined #opensolaris[11:49:11] * timeless_mbp tries to figure out liane's time zone[11:49:35] <Seemone> hello, I have a question about zfs-auto-snapshot… I've been told the maximum expert is here but I'm not sure he's awake right now :)[11:49:36] <lewellyn> liane?[11:49:45] <timeless_mbp> Praza[11:49:51] <lewellyn> oh. dunno[11:50:05] <timeless_mbp> seems to be up awfully late to be in Cupertino[11:50:20] <lewellyn> Mon Mar 1 02:53:14 PST 2010[11:50:25] <lewellyn> i'm still up *shrug*[11:51:16] <PatrickS> I have done a 'zpool export tank + touch /reconfigure + init 6....rebooted...zpool import tank' but all devicenames are as before[11:51:36] <Seemone> timsf: are you awake?[11:51:41] <lewellyn> PatrickS: i dunno. it appears to be a zpool reporting issue, to me[11:52:05] <timsf> Yeah Seemone[11:52:21] <Seemone> great. first of all thanks for the wonderful zfs-auto-snapshot :)[11:52:21] <PatrickS> ok, so should I try in #zfs channel?[11:52:47] <Seemone> then… what should I install on a sol10u8? 0.12?[11:53:24] <Berny> PatrickS, why do expect different devicenames?[11:53:34] <timsf> No problem. It's not backported to s10, but I think I have a post on zfs-auto-snapshot at opensolaris dot org somewhere that tells you which changeset you can run there[11:53:49] <timsf> (the one just before we introduced ksh93 specific shell scripting)[11:53:57] <timsf> I'll look...[11:54:06] <lewellyn> PatrickS: are you on 2009.06?[11:54:19] <lewellyn> Berny: because the ones reported by zpool status don't exist? ;)[11:54:29] *** RoyK has joined #opensolaris[11:55:03] <Berny> lewellyn, well just exporting/importing shouldn't make them disappear ;-)[11:55:15] *** torchie has joined #opensolaris[11:55:16] *** _madmax_ has quit IRC[11:55:50] <Berny> anyway... lunchtime[11:55:51] <PatrickS> cat /etc/issue -> NexentaCore 3.0 (Hardy 8.04/b124+)[11:56:20] <timsf> Seemone: you want changeset 752814468d17 from ssh://anon at hg dot opensolaris.org/hg/jds/zfs-snapshot[11:56:45] <timsf> You'll need to hg pull and build the package yourself, then work out how many of the subsequent changesets you want to backport[11:57:14] <lewellyn> Berny: zpool status "should" show the devices currently in the pool, not the devices which were originally in the pool on other hardware, imo[11:58:14] *** Odin- has joined #opensolaris[11:58:16] *** ma3x__ has joined #opensolaris[11:58:21] <nikolam> hmm, is it intentionally that package manager can not search package database when offline?[11:58:46] <lewellyn> nikolam: afaik, yes :P[11:58:51] <nikolam> shouldn`t it have cached package lists somewhere[11:59:09] * nikolam would like to kno short reason why[11:59:17] <timsf> Seemone: there was some stuff on the mailing list at http://mail.opensolaris.org/pipermail/zfs-auto-snapshot/2009-January/000082.html which may help[11:59:42] <lewellyn> nikolam: you'd have to ask the designers :P[12:00:22] <nikolam> Also I see that mostly one is unable to, say, stay with one dev release version and apply the other updates[12:00:30] <nikolam> except updating to newer dev release[12:00:45] <nikolam> I have updates in /extra per example[12:00:59] <nikolam> and I want them while staying on current dev I use[12:01:15] *** ma3x has quit IRC[12:01:18] <lewellyn> extra should be independent of dev[12:01:28] *** forq has joined #opensolaris[12:01:38] <nikolam> Would be nice to say: stick to snv_xyz for now[12:01:45] *** LenZ has joined #opensolaris[12:01:47] <lewellyn> so don't image-update[12:01:57] <lewellyn> install the individual updates[12:01:58] <nikolam> but to see in updates, only updates not related to release, etc[12:02:11] *** ma3x__ is now known as ma3x[12:03:21] <Seemone> timsf: thanks. do you know how the reimplementation is going and/or who I could ask about that?[12:03:35] <nikolam> maybe one wanting to use that particular dev release would need to make its own local copy of repository for that particular dev release[12:04:11] *** Cripps has quit IRC[12:04:35] <nikolam> Anyone using Zfs encryption yet?[12:05:02] <timsf> Well, it was going fine - integrated into snv_128 or 129 then got backed out - no idea what's happening with it at the moment.[12:05:05] <nikolam> There was a video on youtube about zfs encryption few days ago[12:05:52] <timsf> Seemone: A mail to zfs-auto-snapshot at opensolaris dot org might help: Niall Power is the guy who was working on it.[12:06:11] <nikolam> from that video it seems like its just.. around corner or integrated on 13x if I catched that right[12:06:54] <Seemone> timsf: great, thanks![12:07:09] *** forq has quit IRC[12:07:16] *** himanshu_ has quit IRC[12:08:33] *** Cripps has joined #opensolaris[12:08:36] *** torchie has quit IRC[12:09:55] *** nachox has quit IRC[12:12:09] *** himanshu_ has joined #opensolaris[12:20:15] *** dalibor has joined #opensolaris[12:20:16] *** p3rror has quit IRC[12:22:02] *** moepmoep has quit IRC[12:22:25] *** |AbsyntH| has quit IRC[12:26:47] *** moepmoep has joined #opensolaris[12:27:50] <RoyK> hi all. I have an ich9 controller and the drives are visible from bios, but not from osol. any idea how to fix that?[12:28:38] *** torchie has joined #opensolaris[12:36:57] <timsf> RoyK: I found a similar sounding issue, does http://opensolaris.org/jive/thread.jspa?messageID=190077 help ?[12:37:07] *** djinni has quit IRC[12:37:16] *** djinni has joined #opensolaris[12:40:23] *** timeless_mbp has quit IRC[12:40:36] *** torchie has quit IRC[12:53:21] <PatrickS> I guess I found something about the "format / cfgadm / zpool name not match" topic[12:54:11] <PatrickS> the same physical devicepath is mapped to differnt devicenames[12:54:47] *** MuffiXXL has joined #opensolaris[12:55:59] <lewellyn> except that the device names apparently no longer exist[12:57:12] <PatrickS> well that depends on the viewpoint[12:58:03] <PatrickS> for zpool the device "c4d1" exists[12:58:07] *** calumb has quit IRC[12:59:34] <PatrickS> for format this is "c3d1"[12:59:35] <lewellyn> i'm kind of curious how you managed to get new device names[13:00:28] <lewellyn> as g4lt-mordant mentioned, it's not an everyday occurrence[13:00:51] *** Trede has joined #opensolaris[13:01:16] *** himanshu_ has left #opensolaris[13:01:24] <Berny> shift the controllers around or reattach disks to different controllers?[13:01:44] <Berny> bios update that makes the controllers enumerate differently?[13:01:51] <PatrickS> oh ist was actually you "[11:10] <@lewellyn> but you could see what happens with an ls -l /dev/c4d0"[13:02:17] <PatrickS> if done a 'll /dev/dsk/ | grep ":wd"'[13:02:38] <PatrickS> and the a 'll /dev/dsk/ | grep ":wd" | cut -d- -f2- | sort | uniq -c'[13:03:11] <PatrickS> with ll as an alias for 'ls -l"[13:04:00] <PatrickS> yes I have done several changes to that system[13:04:23] <PatrickS> started with 6 disk pool[13:04:41] *** Error404NotFound has quit IRC[13:04:58] <PatrickS> then attached another SATA HBA and add 6 disks[13:05:19] <lewellyn> Berny: it's still not an everyday occurrance for the same physical device path to have a new logical name[13:05:40] <PatrickS> changed the BIOS settings for the onborad SATA from IDE to AHCI[13:06:48] *** nikolam_ has joined #opensolaris[13:08:07] <PatrickS> do you have an idea how i could get that in line again[13:08:20] *** ma3x has quit IRC[13:09:33] <Berny> changig ide to ahci changes the devname from cxdy to cxtydz[13:09:54] *** nikolam has quit IRC[13:09:56] <Berny> different drivers yield different device names[13:10:09] *** Trede has quit IRC[13:10:17] <Berny> i've seen this happen after the nvidia sata driver integrated too[13:13:00] <Berny> so besides zpool giving back a different device whats the problem?[13:14:26] <PatrickS> I'm planing a kind of "zfs nas web interface"[13:15:35] <PatrickS> so if zpool returns some errors I need to get the devicename for cfgadm to unconfigure + detach this disk properly[13:16:26] <PatrickS> OSol seems not to like it if I'm pulling out it's disks[13:16:46] <PatrickS> (which is understandable)[13:18:58] *** luisbg_ has joined #opensolaris[13:19:33] *** gm152 has joined #opensolaris[13:19:48] <Berny> how is your pool setup?[13:19:57] <Berny> mirrored or raidz?[13:21:17] <PatrickS> my pool consists of 2 x raidz2 (6x0.75 + 6x1.5)[13:21:36] *** kinkie has joined #opensolaris[13:21:40] *** g4lt-mordant has quit IRC[13:21:52] <Berny> are the devicenames wrong for all disks?[13:23:59] <Berny> if it's only a small number you could "forcefully" remove single disks an replace them with the new devname[13:24:35] *** luisbg__ has joined #opensolaris[13:27:21] *** g4lt-mordant has joined #opensolaris[13:27:40] <PatrickS> it is only a relavively small number of wrong disks[13:28:04] *** luisbg_ has quit IRC[13:28:21] *** jbit has quit IRC[13:28:21] *** jbit has joined #opensolaris[13:30:15] <Berny> you could also try to export again, remove the "wrong" devlinks (or let devfsadm clean up), run devfsadm again to see if the wrongs names reappear and if they don't importing should pick up the right devnames[13:31:19] <RoyK> timsf: thanks - works like a dream :)[13:32:09] <timsf> Great stuff.[13:36:28] *** gosx_ has joined #opensolaris[13:36:45] *** calumb has joined #opensolaris[13:39:06] *** Phenox_ has quit IRC[13:39:30] *** Phenox_ has joined #opensolaris[13:39:47] *** gosx has quit IRC[13:45:39] *** betchou has joined #opensolaris[13:47:50] *** MuffiXXL has quit IRC[13:47:53] *** timeless_mbp has joined #opensolaris[13:47:53] *** timeless_mbp has joined #opensolaris[13:48:20] <timeless_mbp> is http://pastebin.mozilla.org/705504 expected behavior, or could i file a bug? (yes i know, giving it a partition instead of a disk was mean, stupid, wrong, etc.)[13:48:30] *** corretico has joined #opensolaris[13:50:24] <PatrickS> Berny: I have to admit that I'm not very experienced with OSol...so after exporting the pool, I will do a 'devfsadm -Cv" ?[13:50:28] *** raichoo has quit IRC[13:52:43] <lewellyn> timeless_mbp: i'd ask the #vbox folks[13:53:13] <lewellyn> timeless_mbp: that paste seems to have nothing to do with opensolaris ;)[13:53:30] *** dalibor has quit IRC[13:53:41] <timeless_mbp> lewellyn: oops, wrong tab[13:53:55] <lewellyn> ;)[13:54:02] <timeless_mbp> no wonder tab completion didn't work :)[13:54:12] <lewellyn> (yes, i looked there to see if you asked there, first) :)[13:54:19] <timeless_mbp> thanks :)[13:54:55] <timeless_mbp> lewellyn: you know how 0x0b means fat[something] and 0xee means efi, do you know what means zfs in that scheme?[13:55:24] <RoyK> hi all. where is a good tutorial on how to partition disks on osol? I'm a linux alien, so I find it rather confusing. have a pair of x85m SSDs that I want to use for root, swap and Zil. no l2arc.[13:55:37] <lewellyn> timeless_mbp: you're asking a question other than what you think you're asking :)[13:55:44] *** gm152 has quit IRC[13:55:45] *** yippi has joined #opensolaris[13:56:06] <timeless_mbp> lewellyn: i have a disk where a partition claims to be marked as fat32 but is actually formatted as zfs[13:56:20] <timeless_mbp> i'd like to give osx enough of a hint so it stops trying to mount it as fat32 :)[13:56:24] <lewellyn> timeless_mbp: on solaris, zfs lives in a "Solaris" (0x82) or "Solaris2" (0xbf) partition[13:56:47] <lewellyn> ufs can also live in partitions of the same "number"[13:56:59] <lewellyn> note that linux coopted 0x82 for themselves, hence 0xbf[13:57:43] * timeless_mbp ponders[13:59:00] * lewellyn wanders off and wishes timeless_mbp luck[13:59:39] <Berny> PatrickS, devfsadm -Cv should clean up...[14:01:30] <PatrickS> I'm about to reboot (keeping my fingers crossed)[14:04:55] *** alech has quit IRC[14:07:10] <timeless_mbp> oh my[14:07:31] <timeless_mbp> pool: rpool; state: FAULTED; rpool FAULTED corrupted data | c1t1d0p0 UNAVAIL corrupted data[14:07:54] *** mroconnor has quit IRC[14:09:24] <timeless_mbp> http://www.sun.com/msg/ZFS-8000-5E is not particularly happy/helpful[14:10:13] *** eXeC001er has joined #opensolaris[14:10:34] *** darrenb has quit IRC[14:10:39] *** nanase has joined #opensolaris[14:11:43] *** raichoo has joined #opensolaris[14:12:13] *** darrenb has joined #opensolaris[14:12:14] *** encephalon has quit IRC[14:12:20] *** wizeman has quit IRC[14:15:32] *** g4lt-mordant has quit IRC[14:21:03] *** Animal-X has joined #opensolaris[14:24:04] *** derchris has quit IRC[14:24:09] *** derchris has joined #opensolaris[14:24:32] *** Animal-X has quit IRC[14:25:58] *** Animal-X has joined #opensolaris[14:26:01] *** ma3x has joined #opensolaris[14:26:48] <ma3x> is it safe resizing the partition where osol is installed with gparted? (it recognizes it as swap)[14:27:33] <PatrickS> timeless_mbp: could you put 'zpool status' to pastebin?[14:28:46] *** dalibor has joined #opensolaris[14:29:38] *** bpmcd has quit IRC[14:31:06] *** wdp has quit IRC[14:31:53] *** effremn has joined #opensolaris[14:42:43] <timeless_mbp> PatrickS: gimme a bit[14:42:57] <timeless_mbp> i'm currently copying data from a sparse bundle[14:43:10] *** g4lt-mordant has joined #opensolaris[14:43:54] *** MuffiXXL has joined #opensolaris[14:46:17] <timeless_mbp> http://pastebin.mozilla.org/705508[14:46:28] <ma3x> can anyone tell me if it's possible to resize active partition. i shrinked the linux partition so i have now some free space before the solaris partition (i wanna grow it to fill that space)[14:46:54] <ma3x> can i do it directly from solaris or should i boot and do it with g/parted or some similar tool?[14:47:28] <madwizard> Before?[14:47:36] <madwizard> May not be so easy[14:48:20] <ma3x> yeah it's like partition1(linux) 10GB, unallocated space 4 GB, second partition (solaris) 8GB[14:48:29] <ma3x> i wanna use those 4 gb[14:49:30] <ma3x> i'm afraid using a tool like gparted, it might mess things up[14:49:42] <madwizard> I'm not sure you can grow partition to the space before it[14:50:12] <timsf> Probably easier to either reinstall, or send those 8gb to an external disk, boot from that disk then recv the content to a newly created pool taking up the full 12g.[14:50:22] <madwizard> Yup[14:50:30] <madwizard> What this Mr said[14:51:10] <ma3x> what will happen if i try to resize it with gparted (i have gparted on usb flash drive). it treats it as swap. will it destroy the data[14:52:46] <madwizard> Probably[14:52:52] <timsf> Probably won't destroy the data, but may well make the data inaccessible: don't know how smart zfs is if it can't see labels at the beginning of a partition.[14:53:01] <timsf> (I know it stores labels towards the end too iirc)[14:53:09] <timsf> Try it and tell us how you get on.[14:53:28] * timsf is happy to risk your data experimenting :-)[14:53:28] <madwizard> Yup[14:53:34] <madwizard> A good blog post it would be[14:55:22] *** th0mz has joined #opensolaris[14:55:27] <th0mz> hi[14:55:27] <madwizard> ma3x: I'm gonna do this tonight (it's 3 PM here)[14:55:31] <madwizard> ma3x: In VM[14:55:36] <th0mz> how could i do a permanent ip alias please ?[14:55:37] <madwizard> I'll tell you then result[14:55:51] <madwizard> th0mz: I'm not sure if nwam can do this[14:55:53] <ma3x> madwizard: yup just msg me[14:56:04] *** ziyourenxiang has joined #opensolaris[14:56:08] <madwizard> Is so, klick on it or right click and see preferences[14:56:11] <madwizard> If it can't[14:56:17] <madwizard> svcadm disable nwam[14:56:43] <th0mz> i'm not in automatic mode[14:57:27] <madwizard> configure your net according to http://docs.sun.com/app/docs/doc/816-4554/ipconfig-1?l=en&a=view this guide[14:57:30] <madwizard> basically[14:57:31] <madwizard> You need[14:58:10] <madwizard> /etc/hostname.INTERFACE (for e1000g0 it would be /etc/hosntame.e1000g0) with either hostname or ip adress[14:58:18] <madwizard> /etc/ipnet/ipnodes[14:58:33] <madwizard> /etc/defaultrouter (or /etc/ipnet/defaultrouter?)[14:58:35] *** OclkdMan has joined #opensolaris[14:59:03] <th0mz> i'll read the guide, thank madwizard[14:59:10] <th0mz> read/follow[14:59:24] <Mick27> Hello all[14:59:24] <madwizard> I think you want specificcaly http://docs.sun.com/app/docs/doc/816-4554/ipconfig-64?l=en&a=view this[14:59:58] <Mick27> I'm looking for something similar to DRBD to use with an opensolaris based setup, any hint for me ?[15:00:15] <th0mz> Mick27: zfs mirroring maybe ?[15:02:16] <madwizard> Rather qfs[15:03:34] <madwizard> Or http://wiki.lustre.org/index.php/Main_Page[15:03:49] <Mick27> th0mz, is it live , remote & automatic as DRBD ? Google gave me AVS[15:04:10] <madwizard> Hm[15:04:14] <madwizard> no lustre for Solaris?[15:04:18] <madwizard> Did I dream some things?[15:04:42] *** mroconnor has joined #opensolaris[15:04:42] *** mroconnor has joined #opensolaris[15:06:15] <bda> Mick27: There's AVS or you can set up rolling snapshot replication.[15:07:47] *** perlmongo has joined #opensolaris[15:09:18] *** |AbsyntH| has joined #opensolaris[15:09:29] *** wdp has joined #opensolaris[15:16:46] *** OclkdMan has quit IRC[15:17:46] <jamesd2> sstallion_work, you around?[15:18:15] *** c0t0d0s0_ has joined #opensolaris[15:19:27] <taemun> should jack / jack still work on b133?[15:20:43] <timeless_mbp> PatrickS: http://www.webwizardry.net/~timeless/zpool-unhappy sorry for the delay[15:21:11] <jkimball4> entic offers ssl certs for *.auroraux.org for 113[15:21:34] <jamesd2> how did wget get included with no largefile support... 2010-03-01 00:26:02 (0.00 B/s) - Connection closed at byte 2147475456. Giving up. ... it stops after 2GB? that is just wrong...[15:21:53] <jkimball4> oops wrong channel[15:22:24] <taemun> nvm[15:22:24] *** c0t0d0s0_ has quit IRC[15:24:41] *** MuffiXXL has quit IRC[15:25:02] *** niq has joined #opensolaris[15:25:03] *** niq has joined #opensolaris[15:26:40] *** Vinayak has quit IRC[15:27:40] *** proti has quit IRC[15:28:18] <PatrickS> timeless_mbp: indeed not particular meaningfull[15:28:42] <timeless_mbp> PatrickS: should i file a bug somewhere?[15:28:49] <timeless_mbp> osx doesn't seem to mind the volume too much[15:28:56] <timeless_mbp> i mean, the volume is damaged, but importable[15:29:20] *** proti has joined #opensolaris[15:29:27] *** ma3x has quit IRC[15:29:40] <PatrickS> did you create that pool on that particular machine?[15:29:52] <timeless_mbp> yes[15:29:59] <timeless_mbp> running 10.5.8 or maybe 10.5.7[15:30:14] <timeless_mbp> the pool was created in feb 2009 or so iirc[15:30:29] <timeless_mbp> maybe a bit later… zpool history from osx would tell[15:30:36] <PatrickS> how is it connected?[15:30:47] <PatrickS> sata / usb?[15:30:51] <timeless_mbp> it was the internal primary sata volume[15:30:57] <timeless_mbp> it's now in a usb enclosure[15:30:58] *** MuffiXXL has joined #opensolaris[15:31:13] <timeless_mbp> because the disk was dying so it got replaced and i'm trying to repair it and copy over my data[15:32:13] <PatrickS> so you created another rpool for that system which is fine[15:32:35] <timeless_mbp> another rpool (named 'pool') on the new osx system on the new esata hdd[15:32:39] <timeless_mbp> and it seems fine, yeah[15:33:19] *** dalibor has quit IRC[15:33:39] <PatrickS> is there a indication in the logfiles what is failing on that disk?[15:33:42] <timeless_mbp> note that zpool import from osx works while the device is in this usb enclosure[15:34:00] <timeless_mbp> yeah, when i tried to use zfs send, i got a panic explaining a file was corrupt[15:34:21] <PatrickS> so the pool is FAULTED but you can import?[15:34:23] <timeless_mbp> the log is in the hfs+ (?) volume on the same disk, i can copy it once i ask jeos to release the disk[15:34:35] <timeless_mbp> the pool is only listed as faulted by jeos, not by osx[15:34:38] <timeless_mbp> which doesn't make sense[15:34:51] <timeless_mbp> note that jeos is running in virtualbox, it's possible i've misconfigured the wrapper for it[15:34:56] <PatrickS> thats right[15:35:20] <timeless_mbp> http://pastebin.mozilla.org/705516[15:35:23] <timeless_mbp> is how i created the wrapper[15:35:29] *** bobbyz_ has joined #opensolaris[15:35:33] <timeless_mbp> it doesn't seem particularly interesting[15:36:59] <PatrickS> how is ZFS supported by jeos? is it the FUSE implementation?[15:37:09] <timeless_mbp> jeos = osol stripped of gui[15:37:38] <swift> hi guys, what is the best tool which can be used for centralised logging and message notification if some warnings +errors are shown in the logs?[15:37:39] <timeless_mbp> sorry, not to be confused w/ someone else's jeos ![15:37:53] <timeless_mbp> http://hub.opensolaris.org/bin/view/Project+jeos/[15:38:26] <PatrickS> ah[15:38:53] <PatrickS> ok, I googled it and came up with an ubuntu flavor[15:39:25] <PatrickS> so to get the things straight....[15:40:31] <PatrickS> the rpool has been created on OSX then exported / disk phys removed / put into USB enclosure[15:40:46] <timeless_mbp> it was never officially exported[15:40:51] *** michuel has joined #opensolaris[15:40:53] <PatrickS> imported by OSX works ok?[15:40:55] <timeless_mbp> to import to osx i need to use -f[15:40:56] <swift> hi guys, can i use Spiceworks to monitor solaris logs?[15:40:58] <timeless_mbp> yeah[15:41:09] <houst0n> swift: Splunk[15:41:19] *** phimic has quit IRC[15:41:24] <timeless_mbp> PatrickS: but otherwise, yes[15:41:28] <swift> timeless_mbp, is that yes for me?[15:41:34] <swift> oh[15:41:36] <swift> my bad[15:41:37] <timeless_mbp> no :)[15:41:42] <swift> :D[15:41:56] <swift> houst0n, can't i use Spiceworks?[15:42:06] *** phimic has joined #opensolaris[15:42:28] <swift> Splunk is platform independant right?[15:42:38] <PatrickS> timeless_mbp: that pool with its data is now accessible from OSX[15:42:52] <timeless_mbp> PatrickS: when virtualbox isn't using it, yes[15:42:56] *** michuel has left #opensolaris[15:42:58] <timeless_mbp> (exclusive disk access..)[15:43:18] *** Phenox_ has quit IRC[15:43:23] <PatrickS> but as you setup a jeos in a vbox[15:43:47] <PatrickS> maybe you should tell vbox not using the raw disk[15:43:58] <timeless_mbp> and instead do what?[15:44:41] <PatrickS> maybe only c1t1d0p0 instead[15:44:56] <PatrickS> just guessing[15:46:45] *** nikolam_ has quit IRC[15:47:12] *** MuffiXXL has quit IRC[15:52:25] *** Phenox_ has joined #opensolaris[15:53:47] *** rcarcamo has joined #opensolaris[15:53:51] *** MuffiXXL has joined #opensolaris[15:54:21] *** ma3x has joined #opensolaris[15:54:40] <swift> guys, please give me the best way to manage and monitor my syslogs[15:54:43] <swift> event logs[15:55:15] <DrEeevil> hire some guys to look at them 24/7 ;)[15:56:42] *** jthunder has quit IRC[15:57:00] <lyarick> swift: (1) use a real loghost, with no services enabled (not even SSH). (2) look for failed login attempts and service failures.[15:57:07] *** sponix has quit IRC[15:57:26] *** nachox has joined #opensolaris[15:57:47] *** leito_ has quit IRC[16:04:28] *** leito_ has joined #opensolaris[16:04:40] <timeless_mbp> PatrickS: http://pastebin.mozilla.org/705520 is what i got this time :([16:05:07] <timeless_mbp> so… no, i can't seem to open it anymore :(~[16:06:02] <timeless_mbp> oh, that's just osx being silly[16:06:05] <timeless_mbp> the volume was in use :)[16:06:22] *** c0t0d0s0_ has joined #opensolaris[16:06:56] <timeless_mbp> http://pastebin.mozilla.org/705521[16:07:15] *** Tom-W has joined #opensolaris[16:07:55] *** nikolam has joined #opensolaris[16:11:40] <lyarick> does anyone know why the "ether" tag wouldn't show on an ifconfig -a for a network card that is plugged in and working?[16:12:16] <lyarick> it shows up in "arp -a", just not in the ifconfig command.[16:14:37] <lyarick> oop, nm wasn't running as admin.[16:16:19] *** ziyourenxiang has quit IRC[16:20:15] *** Tristam has quit IRC[16:20:42] <swift> guys, i see a lot of snmp related errors on my solaris box[16:21:00] <swift> http://pastebin.com/8496Hub1... here's a sample of the logs, please advise why this is happening[16:21:18] *** Tristam has joined #opensolaris[16:24:03] <swift> please advise[16:25:09] *** ravenslay3r has joined #opensolaris[16:25:10] *** pancake has joined #opensolaris[16:25:17] <pancake> how can I upgrade 2009.06 to snv133?[16:25:30] <Beket> smrt, explain /dev repository[16:25:31] <jamesd2> smrt: explain /dev[16:25:31] <smrt> Um. I seem to not know anything about /dev repository...[16:25:32] <smrt> see "dev repository"[16:25:35] <ravenslay3r> pancake - carefully ->[16:25:42] <pancake> ravenslay3r: :)[16:25:43] <Beket> crap :P[16:25:44] <lyarick> pancake: I also had problems with this, here's how i did it..[16:25:49] <Beket> smrt, explain dev repository[16:25:50] <smrt> To switch to the /dev repository (OpenSolaris development releases, see the IPS version in /topic): pfexec pkg set-publisher -O http://pkg.opensolaris.org/dev opensolaris.org && pfexec pkg refresh --full && pfexec pkg install SUNWipkg && pfexec pkg image-update -v # Then wait. You *must* read the release notes before starting out or you *will* be unhappy. See also: ips mirrors[16:26:11] <jamesd2> ravenslay3r, we have beadm its not as major change as it sounds.. just don't zpool upgrade and all is good for a fall back[16:26:22] <ravenslay3r> pancake: remvove "release" repo from your list and ADD http://pkg.opensolaris.org/dev/ named as "opensolaris.org"[16:26:22] <lyarick> mkdir /tmp/mnt-osol133[16:26:29] <swift> wow!!.. amazing ![16:26:31] <lyarick> beadm create osol133[16:26:45] <lyarick> beadm mount osol133 /tmp/osol133[16:27:16] <pancake> ravenslay3r: i can rollback just using any other boot entry right?[16:27:21] <lyarick> pkg -R /tmp/osol133 image-update[16:27:38] <lyarick> bootadm update-archive -R /tmp/osol133[16:27:50] <lyarick> beadmin activate osol133[16:28:00] <lyarick> beadm unmount osol133[16:28:06] <lyarick> seemed to work...[16:28:12] <lyarick> you might have to pkg refresh first[16:28:33] *** pizdec has joined #opensolaris[16:28:49] *** xapiens has joined #opensolaris[16:29:04] *** sponix has joined #opensolaris[16:29:12] <ravenslay3r> pancake: If image-update fails the first time try going from release to b128 then 128->133 : using these instructions (looking)[16:30:12] <ravenslay3r> pancake yes, as jamesd says - create new BE for each "step of the way" untill you get it right ..[16:30:36] <ravenslay3r> pancake: when you get it right keep your "old working BE" and the "new One" at least, but you can delete some of the incremental ones[16:31:39] *** kimc has quit IRC[16:31:42] <ravenslay3r> pancake: http://blogs.sun.com/observatory/entry/updating_to_a_specific_build[16:32:11] <ravenslay3r> if you have trouble going straingt to b133, try using that to go to b128 first, then to 133 ... did it last week :-P[16:32:35] <ravenslay3r> smrt: /msg[16:33:01] <lyarick> ravenslay3r: good link, I wish I had that link instead of having to slog through it myself.. guessing and testing..[16:33:57] * ravenslay3r Yeah I helped someone this week[16:33:59] <Tom-W> wonder when blogs.sun will become blogs.oracle ?[16:34:09] <houst0n> ... never[16:34:14] <Tom-W> hehe[16:34:26] <ravenslay3r> Tom-W: hopinng not too soon[16:35:31] <calumb> I'd imagine that will take a while, as there's an existing blogs.oracle.com, so they'll have to decide which platform is merging with which...[16:35:48] <jamesd2> Tom-W, not sure but current patterns show that sun bloggers are jumping blogs.sun.com faster than people left the titanic even after there feet got wet... i think blogs.sun.com may die all on its own before a name change is possible.[16:37:03] <ravenslay3r> Updated 2009.06-RELEASE => b133(dev) last week, then change my publisher back to RELEASE[16:37:36] <ravenslay3r> this morning update-manager is listing 30+ updates apparently related to b133.. i'm confused[16:38:17] *** skeeziks has joined #opensolaris[16:38:34] *** ma3x has quit IRC[16:38:39] <ravenslay3r> Now that I got the feature i needed "early" I want to get migrate back to 2010.03-RELASE when it's out.[16:39:06] *** bobbyz has quit IRC[16:39:52] *** bobbyz_ has quit IRC[16:40:06] *** bobbyz has joined #opensolaris[16:42:26] * timsf doesn't get the impression oracle blogs are actually blogs at all, as opposed to non-tech-writer produced documentation[16:42:43] *** LouisJB has quit IRC[16:42:44] *** Leal has quit IRC[16:43:05] *** sndcrb has joined #opensolaris[16:43:05] *** LouisJB has joined #opensolaris[16:43:14] <ravenslay3r> why am I still getting update notices for /dev/ if I removed it from my publsiher list?[16:43:15] *** raichoo has quit IRC[16:43:21] <timsf> s/documentation/marketing collateral/[16:43:24] <timsf> (whatever)[16:43:39] *** psychicist has joined #opensolaris[16:43:42] <ravenslay3r> Also is "trackerd" SUN or GNU utility?[16:44:41] <pancake> isnt GNOME?[16:44:44] <timsf> ravenslay3r: http://www.gnome.org/projects/tracker/index.html[16:44:50] <lblume> Refresh your catalog, maybe?[16:45:04] <pancake> reboot time, will try later to set snv133[16:45:07] <pancake> thanks for the tips[16:45:12] *** pancake has quit IRC[16:45:16] *** atomicpunk has joined #opensolaris[16:46:37] <timsf> the update manager cron job claims it refreshes the cache[16:47:04] <ravenslay3r> i updated in the GUI .. just ran from CLI to see if it makes a difference[16:47:31] <ravenslay3r> timfs i see, i think[16:48:03] *** g4lt-mordant has quit IRC[16:51:52] *** nanase has quit IRC[16:52:20] *** BSD_ has joined #opensolaris[16:52:49] <BSD_> help with opensolaris would be appreciated[16:53:01] <BSD_> got stuck in loading screen[16:53:09] <nachox> press the tab key[16:53:10] <houst0n> answer: wait for it to load[16:53:10] *** MuffiXXL has quit IRC[16:53:18] <joshua_> houst0n, :)[16:53:22] <houst0n> it takes around 70 hours to properly boot solaris[16:53:23] *** vraa has joined #opensolaris[16:53:25] <BSD_> already waited for 4 hours[16:53:32] <nachox> then its stuck[16:53:32] <houst0n> only 66 to go[16:53:39] *** kop has quit IRC[16:53:42] <joshua_> msg #cslounge [10:55:43] <BSD_> help with opensolaris would be appreciated[16:53:44] <joshua_> er, whoops[16:53:51] <BSD_> need 70 hours to boot?[16:53:55] * houst0n raises an eyebrow[16:53:55] <BSD_> that's long[16:54:00] <nachox> of course not![16:54:05] <BSD_> ???[16:54:14] <houst0n> BSD_: Yeah, vast aprts of it were re-written in java and python[16:54:20] <houst0n> 70 hours is pretty good.[16:54:27] <Tom-W> lol[16:54:41] <joshua_> houst0n, And just wait until you start pkg...[16:54:42] <BSD_> that's too long[16:54:45] <Tom-W> takes about 5 minutes in VB for me[16:54:50] <nachox> BSD_, boot again, add a -v to the kernel line in grub, press tab as soon as youre booting[16:54:56] <houst0n> joshua_: Lets not open *THAT* can of worms ;)[16:55:05] *** RoyK^ has joined #opensolaris[16:55:07] <BSD_> previously i happened to have error in booting[16:55:08] *** shbwvl has joined #opensolaris[16:55:11] <BSD_> managed to solve it[16:55:12] <houst0n> BSD_: See the console=graphics shit in grub? rm that too[16:55:17] <BSD_> but now stuck again[16:55:30] <BSD_> added -v to kernel line[16:55:37] <BSD_> and pressed b to boot[16:55:40] <BSD_> now stucked[16:55:51] <houst0n> rm console=graphics and also splash/fg/pretty-pics options[16:56:15] *** RoyK has quit IRC[16:57:23] <CIA-37> Jan Pechanec <Jan.Pechanec at Sun dot COM>: 6875954 fork error is reported with wrong errno in sshd.c, 6916082 sshd(1M) ignores nologin(4) if /etc/nologin is not world readable[16:57:47] <houst0n> Another showstopper fixed. >_<[16:58:10] <BSD_> does pressing tab help?[16:58:14] <BSD_> in booting?[16:58:28] <joshua_> it will give you a nongraphical display to see where you get stuck[16:58:52] <BSD_> there's no graphics during the booting[16:58:58] <BSD_> only command lines[16:59:13] <BSD_> i am now trying to upload my screenshot[16:59:22] <houst0n> Err, is this a vm?[16:59:26] <BSD_> imageshack just took forever to upload[17:00:08] <BSD_> maybe by looking at my situation someone can help me with my problem[17:00:12] *** ravenslay3r has left #opensolaris[17:00:22] <BSD_> now trying to get the screenshot uploaded[17:01:25] *** alfism has quit IRC[17:01:52] *** alfism has joined #opensolaris[17:02:16] <houst0n> Well, quickleh. I'm contemplating leaving the office soon[17:02:35] <Kapsel> Is there any way to show what files are actively being read/written? I've got a huge huge load on my fileserver, and i'm not sure what the heck is going on.[17:02:40] <Kapsel> shared via samba.[17:03:10] <houst0n> Kapsel: You're shit out of luck. I built lsof for this and it didn't show files shared over the "built in" cifs[17:03:13] <houst0n> smb might be different[17:03:20] <houst0n> I haaaave a binary somewhere, lemme look[17:03:30] <BSD_> http://i47.tinypic.com/21zg2q.jpg[17:03:36] <BSD_> this is what i encountered[17:03:45] <bda> kasDTrace.[17:03:45] <BSD_> stucked at this screen[17:03:50] <bda> er. Kapsel DTrace.[17:03:57] <Kapsel> houst0n: uhm - ok.[17:03:57] <joshua_> heh![17:04:00] <bda> Kapsel: Look at the DTraceToolkit. iosnoop, iotop.[17:04:16] <bda> You could also write a script that only fires on probes from the CIFS module.[17:04:17] <houst0n> ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/NEW/lsof_4.83K.sun.tar.bz2[17:04:19] *** swift has quit IRC[17:04:22] <Kapsel> bda: ok, thanks. I just figure that dtrace is a bit overkill for what I want?[17:04:31] <bda> Kapsel: DTrace is never overkill. That's the point. :)[17:05:00] <houst0n> Kapsel: Lost my bins, you probably want to compile it yourself anyway though[17:05:02] <bda> Kapsel: DTT has lots of really useful scripts, so it's no work on your part.[17:05:03] <houst0n> It's trivial to build[17:05:23] <Kapsel> houst0n: ok. I might want to look into dtrance instead, though. if it's that easy.[17:05:30] <BSD_> does the opensolaris really need 70 hours to boot?[17:05:32] <Kapsel> I've never used dtrace before.[17:05:35] <joshua_> running dtrace with any sort of complex script (or script that triggers often) is a huge performance hit, though, isn't it?[17:05:35] <bda> BSD_: No.[17:05:36] <houst0n> BSD_: Nah I lied[17:05:39] <houst0n> Like 73[17:05:41] <bda> joshua_: No.[17:05:49] <BSD_> ???[17:06:04] <BSD_> my dvd writer will break down before it can even boot[17:06:15] <houst0n> BSD_: It's only 70 hours if you have VT exts[17:06:16] <bda> BSD_: How much RAM does that system/VM have?[17:06:23] <bda> BSD_: houst0n is fucking with you.[17:06:26] <BSD_> only 4gb[17:06:32] <bda> BSD_: 4GB should be more than enough.[17:06:50] <BSD_> this is what i get http://i47.tinypic.com/21zg2q.jpg[17:06:58] <BSD_> during the boot up[17:07:03] <houst0n> Nothing more?[17:07:06] *** alitzell has joined #opensolaris[17:07:09] *** ma3x has joined #opensolaris[17:07:11] <BSD_> stuck at this for 4 hours[17:07:12] <houst0n> fuck. try booting with acpi disabled?[17:07:14] <joshua_> at some point I used to remember the incantations to turn on moddebug from kmdb, but I cannot remember how now[17:07:37] <houst0n> you should see -B $ZFS_somefoo on your kernel line in grib[17:07:38] <BSD_> is there anyway boot into opensolaris fast?[17:07:50] <houst0n> add ,acpi-user-options=0x8,disable-ehci=true[17:07:59] <ma3x> why running a program like transmission causes mouse screen freeze (2secs) every 20-30 secs and drops all connections[17:08:01] <houst0n> and it'll boot... probably[17:08:17] <houst0n> ma3x: which transmission are you using?[17:08:27] <ma3x> 1.80[17:08:47] <joshua_> ma3x: filesystem? how much RAM do you have? etc[17:08:49] <ma3x> rtorrent crashes immediately after starting the download[17:09:06] <ma3x> zfs 1 gigabyte ram + swap 1gb[17:09:14] <houst0n> BSD_: so your -B part becomes -B $ZFS_BLAHBLAH,acpi-user-options=0x8,disable-ehci=true blah blah /rest /of /options[17:09:17] <houst0n> yeah[17:09:23] <joshua_> every 20 to 30 seconds sounds like it happens with a TXG commit, although I am not an expert[17:09:26] <houst0n> ma3x: that's your problem[17:09:44] <joshua_> the claim is that ZFS really wants more than 1GB of RAM; I would believe it[17:09:45] <houst0n> ma3x: zfs doesn't play well with 2gb of mem, let alone one[17:09:54] *** BSD_ has quit IRC[17:10:20] <ma3x> so it's the ram then[17:10:25] <houst0n> *thanks houston, you're my hero*[17:10:28] <bda> 2GB is doable, but can suck if you're using X + Firefox, et al.[17:10:31] <houst0n> no probs BSD_[17:10:37] *** RoyK^ is now known as RoyK[17:10:52] <ma3x> also sometimes while im typing something after a 2 sec freeze the key gets pressed like thissssssssssssssssss[17:11:26] <houst0n> ma3x: You can buy 4gb of ddr2 for like $20[17:11:45] <houst0n> Otherwise, you're out of luck[17:11:51] <ma3x> i thought 1 gb is like $30[17:12:03] <lblume> OpenSolaris does work rather well on my !GB netbook though. Sure it's not optimal, but it's perfectly usable. Reducing all problems to lack of RAM is a bit lame.[17:12:05] <houst0n> Even if it is, :P[17:12:21] <houst0n> Not all problems....[17:12:34] <joshua_> lblume, Well, if you are doing *huge* txg commits (i.e., from a torrent), ...[17:12:40] <houst0n> I've managed to get it running in far less, not with zfs tho (at least, without some horrible hacky shit in /etc/system)[17:12:58] <houst0n> You need lots of memory for zfs, I thought we all understood this by not :P[17:13:16] *** gosx has joined #opensolaris[17:14:13] <ma3x> any idea why rtorrent crashes?[17:14:21] <x58> lblume: How big is !GB? As a mere mortal I have never seen a !GB stick of ram :P[17:14:21] <joshua_> argh, the evil zfs tuning guide does not have words about how to tune txg commit intervals[17:14:33] <lblume> There are some issues with freezes and ZFS that are certainly not RAM related. Or you'll tell me that my 4GB are just not enough?[17:14:33] <bda> You have plenty of tools to determine if performance problems are caused by memory shortfalls.[17:15:20] <houst0n> joshua_: to dtrace![17:15:24] <bda> joshua_: There's no "huge txg commits". If the commit buffer is full before 30s, it will commit more frequently. Look at fsstat, iostat, etc.[17:15:28] <lblume> bda: Exactly, so that's where it should start, analysis, not conclusions.[17:15:29] <bda> You don't even need DTrace for that.[17:15:30] <Tom-W> port blackbox to OS[17:15:35] <bda> vmstat -S ..[17:15:46] <bda> ma3x: No. Look at the core dump? Error logs?[17:15:59] <joshua_> bda, How large is the commit buffer usually?[17:16:11] <Tom-W> memory problems will reduce[17:16:25] <joshua_> (I guess the other option is to turn down commit buffer size? The goal is to get more granular commits, to avoid long periods of I/O.)[17:16:25] <bda> joshua_: Depends on avail memory, and iirc, the speed of your disks. I think ZFS tries to be smart there.[17:16:33] <lblume> x58: Sometimes, my fingers remember my French keyboard and use <shift> to get numbers :-)[17:16:35] <houst0n> Tom-W: Already done[17:16:38] <bda> joshua_: Do you even know that's the problem?[17:16:43] <joshua_> Nope! Throwing darts![17:16:51] *** gosx_ has quit IRC[17:16:51] <Tom-W> :)[17:16:54] <bda> Throwing darts is still when you have tools that will tell you what the problem is.[17:16:58] <bda> s/still/silly/[17:17:07] <ma3x> core is 27 mbytes binary[17:17:16] <houst0n> bda: It's more fun than reading output though, and usually widly innacurate[17:17:25] <ma3x> Caught Floating point exception, dumping stack:[17:17:27] <houst0n> bda: The once or twice you nail it you feel sort of like yoda though, so it's worth it[17:17:40] <joshua_> "Freezes for an I/O period of time under high I/O load every period of time that looks like a TXG interval" seems like a good set of information to work from, though.[17:17:43] <houst0n> I aeer teeeh servuhhhh whisperererrrrr[17:17:45] <bda> houst0n: I suppose if you enjoy going to palm readers, it's a good time. ;)[17:17:46] <houst0n> etc[17:17:47] <effremn> Hey all - I have a dladm aggr created out of e1000g0 and e1000g1. I think each might be plugged into wrong switches. How can I check based on say arp where each e1000g is connected?[17:18:02] * bda work &[17:19:17] <lblume> joshua_: It's the very symptoms I've had forever on my box, and which are not RAM-related.[17:20:15] <joshua_> lblume, Hmm, well, then, that would discount the amount of memory that he has as being the problem, then![17:21:37] *** Tom-W has quit IRC[17:22:40] <lblume> joshua_: My point is, as explained earlier, that rather than providing a ready-to-eat conclusion that'll improve the values of your Kingston stock, giving help on how to analyze the problem would help find the actual cause of the issue.[17:23:20] *** alka_ has joined #opensolaris[17:23:28] <ma3x> is there a pkg i can install for light weight display manager?[17:24:15] <ma3x> seems when i close firefox, the freezups disappear[17:24:26] <ma3x> that means only one thing right?[17:24:38] <ma3x> ah, no here we go it happened aggggggggggggggggggggggain[17:24:53] <ma3x> that wasn't intentional :p[17:25:28] <lblume> sapphire is in /contrib, I think. That's what I found when I needed a WM to run inside VNC.[17:25:56] *** TheBentZone has joined #opensolaris[17:26:01] *** aruntomar has joined #opensolaris[17:26:10] <timsf> we still ship /usr/bin/twm, right?[17:26:15] *** lattera has joined #opensolaris[17:27:00] <lblume> It's a bit too light for me, but yes :-)[17:28:01] <alanc> timsf: yes we do[17:28:05] <steven_> iv'e seen someone has got awseome running on opensolaris good lightweight wm tis a tiled window manager...[17:28:24] *** houst0n has quit IRC[17:28:33] <steven_> /awseome/awesome/[17:28:41] <alanc> metacity doesn't seem that heavy-weight, but I don't know how well it runs outside of the whole gnome desktop[17:29:31] *** baza11215 has joined #opensolaris[17:30:29] *** Erwann has quit IRC[17:30:54] <lblume> alanc: I rather assumed that starting metacity would automatically launch all kinds of gnome things too.[17:31:05] <alanc> dunno[17:31:16] *** stevel has joined #opensolaris[17:31:22] *** MuffiXXL has joined #opensolaris[17:32:11] *** ma3x has quit IRC[17:35:28] *** Alasdairrr has quit IRC[17:35:42] *** kop_ has quit IRC[17:35:48] *** Auralis has quit IRC[17:35:57] *** Auralis has joined #opensolaris[17:38:08] *** Kedare has joined #opensolaris[17:39:35] <aruntomar> how can i start osol only in text mode?[17:40:07] <atomicpunk> svcadm disable gdm[17:41:35] <aruntomar> atomicpunk: ok, i read there are milestones but haven't got a hang of them[17:42:07] <aruntomar> atomicpunk: in the live cd it gives the option of starting in text mode but not after the installation[17:42:50] *** LouisJB has quit IRC[17:43:14] *** ma3x has joined #opensolaris[17:43:14] *** LouisJB has joined #opensolaris[17:43:26] <ma3x> how do i start twm or metacity[17:43:58] *** suprsonic has joined #opensolaris[17:44:03] *** g4lt-mordant has joined #opensolaris[17:44:13] <ma3x> .xinitrc contains exec metacity, xorg launches but i only get mouse pointer and black background[17:44:24] <suprsonic> is there a recommended max number of disks for a raidz array in regards to disk failure percentages?[17:44:59] <timeless_mbp> PatrickS: http://pastebin.mozilla.org/705539[17:45:11] <lblume> suprsonic: More in regard to performance.[17:45:31] <suprsonic> how so lblume?[17:45:49] <asyd> more?[17:45:53] <asyd> I thought to none :)[17:46:06] <sommerfeld> suprsonic: as the number of disks per vdev increases you want to move to higher parity (raidz2 and raidz3)[17:46:19] *** MuffiXXL has quit IRC[17:46:34] <suprsonic> sommerfeld happen to have a chart for those numbers?[17:46:58] *** bpmcd has joined #opensolaris[17:47:08] <sommerfeld> http://blogs.sun.com/relling/entry/raid_recommendations_space_vs_mttdl[17:47:16] <aruntomar> is the new pkg manager in build 133 fast? in build 111b it takes a lot of time to create a plan[17:47:16] <suprsonic> sweet![17:47:18] <suprsonic> thanks![17:47:33] <sommerfeld> has a chart. doesn't include raidz3[17:47:39] *** sactodave has joined #opensolaris[17:47:50] *** Odin- has quit IRC[17:47:55] *** LouisJB has quit IRC[17:48:21] <sommerfeld> note that the time scale on the left is a log-scale chart[17:48:30] <aruntomar> i really loved the spkg pkg manager in belenix and its speed[17:48:32] *** timeless_mbp1 has joined #opensolaris[17:48:35] <timsf> aruntomar: it's faster[17:48:50] <timeless_mbp1> PatrickS: oops, zpool status -v causes panic :([17:49:39] *** timeless_mbp has quit IRC[17:49:46] <timsf> (133 is faster than 111b - haven't compared against spkg, which I suspect will be quicker)[17:49:55] *** phimic has quit IRC[17:50:24] <sommerfeld> note that the upgrade past the 132-133 bump seems to be expensive because of all the package renames[17:51:01] <sommerfeld> and your first pkg operation after upgrading to 133 will spend a while deleting the memory of the renamed packages[17:51:04] <aruntomar> timsf: thanx, i'm currently stuck with build 111, and it takes like ages to install anything new, the creating plan stage takes a lot of time[17:51:21] <timeless_mbp1> aruntomar: imo it's worth it to let it plan an update[17:51:54] <aruntomar> timeless_mbp1: upgrade is failing from build 111b to build 133, i've filed a bug report, let's c[17:52:16] <timeless_mbp1> :([17:52:21] <timeless_mbp1> at least your data is safe[17:52:36] *** merzo has quit IRC[17:52:38] * timeless_mbp1 sheds tears for 187 data errors and a 53% completed scrub that was interrupted[17:52:53] <timeless_mbp1> scrub: scrub in progress, 0.30% done, 16h46m to go[17:52:59] <aruntomar> timeless_mbp1: that's the best thing i love on osol, that even if ur upgrade fails, u can work with the existing BE[17:54:36] *** ma3x has quit IRC[17:56:46] <RoyK> timeless_mbp1: raidz?[17:56:53] <suprsonic> sommerfeld what does log scale represent?[17:57:29] <CIA-37> Ali Bahrami <Ali.Bahrami at Sun dot COM>: 6929607 ld generates incorrect VERDEF entries for ET_REL output objects[17:57:33] <sommerfeld> time. each major division is probably a factor of 10 increase[17:57:35] <CIA-37> Reza Sabdar <Reza.Sabdar at Sun dot COM>: 6789796 Restoring files does not restore data using NBU, 6925219 ndmp: ndmp log shows error when restoring hardlinked files to original location if file exists[17:57:35] *** aruntomar has quit IRC[17:58:00] <timeless_mbp1> RoyK: for an internal hdd of a mbp?[17:58:02] <suprsonic> very good, thanks![17:58:08] <timeless_mbp1> actually, i can do raidz, can't it[17:58:11] *** fredxng has left #opensolaris[17:58:22] * timeless_mbp1 tries to remember if you can upgrade to raidz later[17:58:30] <sommerfeld> but relling left the numbers off because with enough replication the math will claim trillions of years MTTDL[17:58:34] <timeless_mbp1> scrub: scrub in progress, 2.41% done, 5h56m to go[17:58:38] <sommerfeld> but any individual disk will only last a decade or two[17:59:10] <sommerfeld> and the likelihood of data loss due to an uncorrelated cause (asteroid hit!) overwhelms the loss due to individual disk failure.[17:59:23] <suprsonic> hehehe[17:59:29] <sommerfeld> gotta run.[17:59:36] <suprsonic> is there a recommended limited on performance?[18:00:44] *** timeless_mbp1 is now known as timeless_mbp[18:00:46] *** timeless_mbp has joined #opensolaris[18:01:34] <RoyK> timeless_mbp: heh - add another[18:01:38] <RoyK> zpool attach[18:01:43] <timeless_mbp> sommerfeld:[18:01:44] <timeless_mbp> 2009-03-25.14:52:39 zpool create rpool disk0s3[18:01:57] <timeless_mbp> RoyK: again, on a unibody macbook pro[18:02:03] <timeless_mbp> where exactly do you suggest i stick it?[18:02:12] <timeless_mbp> sommerfeld: that's roughly when i got the laptop[18:02:22] * RoyK hands timeless_mbp a drive, a usb cable and a roll of gaffer tape[18:02:32] <RoyK> why do I have to do all the thinking?[18:02:35] <timeless_mbp> and the disk started dying probably w/in the last 2 or 3 months[18:03:02] <RoyK> Real Men Never Take Backup[18:03:04] *** ma3x has joined #opensolaris[18:03:21] <ma3x> why when i disable ahci from the bios, opensolaris won't boot?[18:03:30] <ma3x> can i get past this[18:04:04] <joshua_> don't do that, then[18:04:21] *** milek_ has quit IRC[18:05:11] <RoyK> doctor, if I hit my thumb with this hammer, it hurts![18:06:01] *** crichardso has quit IRC[18:06:35] <tomww> disabling ahci means chaning the device paths (in most cases) ... so the device tree would need to be rebuilt..[18:09:39] *** Seemone has quit IRC[18:09:39] *** jamesd2 has quit IRC[18:10:31] <ma3x> so what should i do change grub lines or what?[18:11:23] *** aruntomar has joined #opensolaris[18:11:38] *** Jenkens has joined #opensolaris[18:11:42] <Jenkens> hey kiddos[18:12:15] <Jenkens> I'm finally getting around to testing out dedup.. though I seem to be rather unsuccessful at this.[18:12:24] <Jenkens> anyone have experience with dedup?[18:12:39] * RoyK lifts hand[18:12:43] *** IanLee has joined #opensolaris[18:12:44] * timeless_mbp wonders if the zfs volumes are available while the pool is being scrubbed[18:12:46] <Jenkens> i can copy two folders to different parts of a pool and see dedup working[18:13:06] <RoyK> worked well for me with similar testing[18:13:12] <Jenkens> but if i copy a backup of my sql db from thursday night and friday night, for almost all intensive purposes they should be relatively identical[18:13:14] *** JordanG has joined #opensolaris[18:13:18] <Jenkens> yet i get no dedup on it[18:13:28] *** suprsonic has left #opensolaris[18:13:48] <RoyK> keep in mind that default block size is 128k, so smaller files can be put in the same block and then the blocks will differ[18:13:54] <RoyK> try with smaller blocksize[18:14:10] <RoyK> performance is likely to go down if you do so, but you'll see if dedup works or not[18:14:23] <Jenkens> well the backups of my sql db's are right at the 3GB mark[18:14:32] <Jenkens> i'm just not seeing any dedup between them[18:15:00] <RoyK> have you copied the files into the dataset _after_ turning on dedup?[18:15:08] *** relling has joined #opensolaris[18:15:09] <sommerfeld> are the backups compressed?[18:15:10] <Jenkens> i know it works when i copy the same folder contents to two different directories and see dedup working[18:15:15] <Jenkens> RoyK: yes of course, after[18:15:21] <RoyK> just checking[18:15:22] <timeless_mbp> hrm[18:15:23] <Jenkens> sommerfeld: yes, i wondered if this was part of the problem[18:15:25] <sommerfeld> change one bit at the start of a file and the compressed images will probably differ completely after the first change[18:15:28] *** goksin has joined #opensolaris[18:15:40] <timeless_mbp> it looks like zpool import -R doesn't like me[18:15:42] <RoyK> sommerfeld: are they identical or just almost?[18:15:55] <timeless_mbp> i did: zpool import -R /Volumes/rpool/[18:15:57] <timeless_mbp> and as i result, i got;[18:15:59] <timeless_mbp> rpool/export/home/timeless/devel-old 22.9G 46.4G 75.1M /Volumes/rpool//Users/timeless/devel-old[18:16:07] <Jenkens> i think that was directed to me..? anyway they have a days worth of transactions difference[18:16:09] <timeless_mbp> note the extra / after rpool[18:17:34] <Jenkens> royk: http://opensolaris.pastebin.ca/1817466[18:17:37] *** lblume has quit IRC[18:18:27] *** Mick27 has quit IRC[18:19:09] <RoyK> Jenkens: zfs get dedup and zpool get compression on pooll/dedupe, and zpool list pooll[18:19:39] *** Kedare has quit IRC[18:19:53] <RoyK> Jenkens: but I would have decreased block size[18:20:11] <RoyK> 4k gives very good dedup, but might be overkill[18:20:13] *** vraa has quit IRC[18:20:29] <Jenkens> http://opensolaris.pastebin.ca/1817472[18:20:33] *** jamesd2 has joined #opensolaris[18:21:08] *** WireWulf has joined #opensolaris[18:21:25] <RoyK> try to recreate the pool with 4k block size[18:21:40] <RoyK> the big guys are using block sizes that low to get good dedup[18:22:42] <bda> Why do you need to recreate the pool? Can't you just set recordsize on the filesystem?[18:23:06] <Jenkens> royk: alright, lemme see what happens. moving the files[18:24:38] <Jenkens> whew, this has the CPU angry :p[18:24:53] <RoyK> :)[18:25:10] <relling> NB using smaller blocksize means you need more RAM for the DDT, too...[18:25:15] <RoyK> bda: erm - yes - I wasn't thinking[18:25:16] <Jenkens> iguess that's the 4K block size[18:25:31] <RoyK> 4k might be overkill[18:25:36] <RoyK> 16 or 32 might do fine[18:25:41] <Jenkens> yeah[18:25:47] <Jenkens> wait, can i just change the blocksize like bda said?[18:25:48] <jamesd2> RoyK, i saw your pool layout, it would be better if if you spread drives from each controller in each vdev, its always best to have one drive per controller in each pool so if a controller fails you don't loose the entire pool untill the controller is replaced, and you get better performance as well.[18:25:55] <Jenkens> or does that only apply to data after the cahnge is made?[18:26:32] <Auralis> Jenkens: you can change blocksize on the fly, but the changes only effects files written after the change[18:26:38] <bda> jamesd2: Ask relling about that. ;-)[18:26:39] <Jenkens> that's what i thought[18:26:41] <Jenkens> alright[18:26:54] <RoyK> jamesd2: I'll keep that in mind - what happens if I change this now? will zfs autoorder it back?[18:27:29] <jamesd2> RoyK, yes zfs has serial numbers on each drive so should be able to recover, after an import... but always best to have a backup.[18:27:42] <relling> I don't worry about controller failure (but I do worry about crappy controller firmware :-)[18:28:15] <Jenkens> i need to get my wife a bday gift[18:28:24] <Jenkens> no idea what to get her[18:28:27] <Jenkens> I suck at giving gifts[18:28:41] <jamesd2> you also get better performance[18:29:27] *** drewj has quit IRC[18:30:12] *** drewj has joined #opensolaris[18:32:00] *** Animal-X_ has joined #opensolaris[18:32:18] *** MuffiXXL has joined #opensolaris[18:34:19] *** Jiraia has joined #opensolaris[18:35:12] *** Animal-X has quit IRC[18:35:13] *** MuffiXXL has quit IRC[18:35:45] *** Mdx4 has joined #opensolaris[18:37:23] <Jenkens> hrmm[18:37:39] <Jenkens> ok i moved some smaller backups to it and i'm getting very little dedupe (1.04x) on them[18:38:09] <sstallion_work> smrt: explain mirror[18:38:10] <smrt> Um. I seem to not know anything about mirror...[18:38:11] <Jenkens> i guess i'll have to toy around[18:38:13] <sstallion_work> smrt: explain zfs mirror[18:38:15] <smrt> Um. I seem to not know anything about zfs mirror...[18:38:17] * sstallion_work sighs[18:39:14] <RoyK> Jenkens: how's the dedup now?[18:39:20] *** comay has joined #opensolaris[18:39:33] *** ChanServ sets mode: +o comay[18:39:43] <Jenkens> royk: well, i moved some smaller backups over and i'm not seeing a lot of dedupe (1.04x)[18:39:49] <Jenkens> these backups are near identical[18:40:01] <RoyK> compressed?[18:40:03] <Jenkens> within kilobytes of eachother[18:40:07] <Jenkens> but yes,compressed[18:40:17] <Jenkens> perhaps that's the problem[18:40:21] <RoyK> guess so[18:40:41] <RoyK> compression and dedup are quite alike[18:40:51] <RoyK> well, not really, but it's the same ballpark[18:41:11] <bda> One is playing stickball, and the other is breaking your knees with a bat.[18:41:14] *** moepmoep has quit IRC[18:41:21] <RoyK> :)[18:41:44] *** aphrek has joined #opensolaris[18:41:45] <RoyK> hrmf[18:41:46] <RoyK> Mar 2 00:53:41 mime unix: [ID 954099 kern.info] NOTICE: IRQ23 is being shared by drivers with different interrupt levels.[18:41:46] <RoyK> Mar 2 00:53:41 mime This may result in reduced system performance.[18:41:59] <Jenkens> Royk: with compressed files i'd think that the dedupe would be near 2.x..[18:42:19] <Jenkens> it's like taking a movie, then editing it and cutting off 30 seconds and re-saving it, the dedupe on that should be quite a bit[18:42:22] *** goksin has quit IRC[18:42:46] <bda> Jenkens: That isn't how it is at all. The compression changes the bits. If the bits are not the same, what duplication is there to remove?[18:43:11] *** bubbva has joined #opensolaris[18:43:15] <bda> Jenkens: Trivial test: Write something that grabs $recordsize from each file and hashes it. Are the hashes the same?[18:43:23] <mui> I got storage independent live migration with xen and zfs working[18:43:29] <bda> (seek over the entire file)[18:43:30] <mui> iz cool[18:43:42] <Jenkens> well, i just copied over a zip file and rename it and copied the same zipfile into the pool[18:43:48] <Jenkens> my dedup is now 2.x (for the identical file)[18:43:53] <Jenkens> which makes sense[18:43:59] <RoyK> yeah[18:44:02] <bda> Jenkens: But your backups are not identical.[18:44:21] <bda> Jenkens: Try decompressing the files, then copying them. See if you get a decent dedup.[18:44:22] <Jenkens> bda: right, but they should be damn near identical.. within a couple hundred meg at most[18:44:38] <bda> Jenkens: You are looking at the entire file when you need to bee thinking about $recordsize, no?[18:44:42] *** tsoome has joined #opensolaris[18:44:51] <Jenkens> which i thoguht the purpose of dedupe would have been to only 'use' the difference between them on the filesystem[18:44:52] <RoyK> if you compress with lower record size than dedup size, I guess it'll be possible[18:45:03] <RoyK> s/record/block/[18:45:31] <atomicpunk> isn't dedup block level?[18:45:37] <bda> Yes.[18:46:16] <atomicpunk> I'd be kinda surprised if moderately changed db dumps got significant deup even uncompressed. One small addition at the beginning of the dump could throw off alignment all future blocks, no?[18:46:30] <bda> Quite.[18:46:57] <bda> Though having never tried it, I don't know what results you would actually get.[18:47:03] <RoyK> atomicpunk: see above. he's not even getting good dedup with 4k bs[18:47:09] <atomicpunk> now, hot backups ... uncompressed ...[18:47:25] <atomicpunk> RoyK: I don't think smaller block sizes would do much towards that problem...[18:47:40] <RoyK> with compressed backups, no, but otherwise it does[18:47:46] <RoyK> the larger blocks, the worse dedu p[18:47:48] <RoyK> dedup[18:47:50] *** bewst has joined #opensolaris[18:48:06] <atomicpunk> very true, but I still don't think it would help much in the scenario I described. You could still end up with blocks completely misaligned for the entire file.[18:48:37] <atomicpunk> Jenkens: try a hot backup for giggles. :)[18:48:58] <Jenkens> a hot backup?[18:49:11] <atomicpunk> aren't these database dumps ?[18:49:12] <bewst> Need help understanding OSOL release numbers and how to get them. I have 2009.06 installed. This corresponds somehow to 111b, right? Could someone explain that? How do I upgrade to a version that supports ZFS deduplication? Thx.[18:49:25] *** crichardso has joined #opensolaris[18:49:26] <Jenkens> atomicpunk: well they are database backups from my ms sql server[18:49:33] <atomicpunk> oh. yuck. :)[18:49:38] <bda> bewst: 111b is the ONNV build number. 2009.06 is the release.[18:49:38] *** Hedonista has joined #opensolaris[18:49:48] <bda> smrt: dev repo[18:49:52] <atomicpunk> but these are exports/dumps, not copies of the data files?[18:49:55] <bda> >:|[18:50:00] <bewst> bda: now I need someone to explain "ONNV" :-/[18:50:01] <Hedonista> afternoon y'all[18:50:15] <Jenkens> meh, i gotta quit fiddling with this and figure out what to get the wife[18:50:20] *** sstallion_work has quit IRC[18:50:24] <bda> bewst: You need to upgrade to the /dev repository to dedup. Or wait another couple weeks (hopefully) for the next stable release (2010.03).[18:50:37] <bda> bewst: ONNV is OS/Net Nevada. The source tree.[18:50:59] <atomicpunk> wife won't appreciate dedup? :)[18:51:22] <Jenkens> noope[18:51:31] <bewst> bda: OK, thanks. Just how unstable is the /dev repo? I know, qualitative question.[18:51:36] <bda> Jenkens: Flowers, dinner, chocolate, a nice walk.[18:51:39] <bda> bewst: Depends on the build. :)[18:51:41] <Jenkens> hah[18:51:47] <Jenkens> we're doing dinner tonight[18:51:53] <Jenkens> need to get her a gift though[18:52:03] <bda> bewst: Always read the release notes before upgrading. Happily if a given build is totally hosed, you can roll back to the previous one.[18:52:04] <Jenkens> her valentines day flowers were a farking flop, so i'm not buying flowers again for a whiloe[18:52:15] <Jenkens> those flowers died so fast, it was insane[18:52:25] <bda> Jenkens: You want to get uncut flowers that have not opened yet.[18:52:35] * bda is getting married next month; has recently learned these things.[18:52:35] <Jenkens> bda: they were not opened yet[18:52:40] <bda> Jenkens: Lame-o, then.[18:52:42] *** aphrek has quit IRC[18:52:44] <Jenkens> yep[18:52:47] <Jenkens> it was lame-o[18:53:02] <bda> If you cut the stems and change the water every few days they will last longer.[18:53:04] <timsf> (or cut flowers and snip an inch off the bottom with a sharp scissors underwater)[18:53:06] <tsoome> bda: really? congrats:D[18:53:12] <bda> tsoome: ta :)[18:53:15] *** sommerfeld_work has joined #opensolaris[18:53:17] *** sommerfeld has quit IRC[18:53:23] *** sommerfeld_work has quit IRC[18:53:32] *** sommerfeld has joined #opensolaris[18:53:32] *** ChanServ sets mode: +o sommerfeld[18:54:11] *** PatrickS has quit IRC[18:54:45] *** KermitTheFragger has quit IRC[18:55:03] *** sstallion_work has joined #opensolaris[18:55:48] *** hajma has quit IRC[18:57:29] <CIA-37> William Gorrell <william.gorrell at sun dot com>: 6928826 panic: assertion failed: ab->b_buf == NULL, file: ../../common/fs/zfs/arc.c, line: 1738 - nightly[19:01:17] *** betchou has quit IRC[19:01:53] *** |AbsyntH| has quit IRC[19:01:55] *** sactodave has quit IRC[19:02:24] <RoyK> wtf is this? dedupditto YES <threshold (min 100)>[19:02:51] <trygvis> sounds like deduplication of ditto blocks[19:02:59] <trygvis> which in itself sounds quite .. backwards[19:03:00] <RoyK> ditto......[19:03:30] <trygvis> ditto[19:03:34] <sommerfeld> it's ditto blocks for dedup'ed blocks.[19:03:51] <sommerfeld> backwards[19:03:59] <RoyK> wtf are ditto blocks?[19:04:01] <trygvis> aha[19:04:18] <trygvis> RoyK: want to give you a link to a google search? :)[19:04:23] <sommerfeld> RoyK: what you get when you set copies=2 or copies=3 on a filesystem.[19:04:30] <RoyK> ok[19:05:00] *** hnhn has quit IRC[19:05:01] <sommerfeld> the interaction between copies and dedup is reasonable -- "copies=N" means "I want at least N copies of this block"[19:05:21] <sommerfeld> so if you have a filesystem with copies=3 sharing a block with 100 filesystems with copies=1, there are 3 copies stored[19:05:39] <sommerfeld> (without dedup there would be 103 copies stored)[19:05:43] *** jthunder has joined #opensolaris[19:06:12] <sommerfeld> but, well, if you have 100 filesystems with references to a block, maybe it makes sense to store two copies instead of one...[19:06:18] <tsoome> i did stomp on stupid issue today regarding to zfs:/[19:06:29] <sommerfeld> and that's what dedupditto lets you accomplish[19:08:33] <tsoome> zpool import should really have option to generate new ID for pool.[19:10:16] *** moepmoep has joined #opensolaris[19:10:30] *** wrapster has joined #opensolaris[19:10:53] *** shbwvl has quit IRC[19:11:04] *** bjw has quit IRC[19:11:20] <wrapster> in the onnv-gate while going through the pkgdefs dir i found that there was a etc directory which had a few driver definitions.. what are they and why are they not packaged?[19:11:30] <wrapster> or provided as any SUNW pkg?[19:11:50] <tsoome> there is no SUNW left.[19:12:09] *** MattMan has quit IRC[19:12:37] <wrapster> tsoome: huh?[19:12:52] <tsoome> oracle bought it all:P[19:12:59] <wrapster> :)[19:13:01] <wrapster> ok..[19:13:07] <tsoome> and opensolaris packages got renamed as well;)[19:13:13] <wrapster> to?[19:13:34] <sommerfeld> names that aren't compressed to the point of incomprehensiblity[19:13:38] *** jthunder has quit IRC[19:13:48] <tsoome> :)[19:14:09] *** phips|away is now known as phips[19:14:17] <wrapster> tsoome: you serious? coz in b134 onnv-gate I still see SUNW pkgs only[19:14:27] <wrapster> from 135 it will be changed is it?[19:15:15] <sstallion_work> wrapster: there are older pkgdef artifacts that remain to be cleaned up.[19:15:42] <sstallion_work> b134 doesnt exist yet - you are looking at onnv_134[19:15:44] *** Jiraia has quit IRC[19:15:52] <wrapster> sstallion_work: yes i am[19:15:55] <sstallion_work> actually nevermind, looks like b134 was pushed out.[19:16:08] <wrapster> sstallion_work: yes to that as well...[19:16:24] <wrapster> I pulled the gate from genunix.org[19:16:44] <wrapster> sstallion_work: thats secondary.. I would like to know however if the names of these pkgs have changed?[19:16:52] <sstallion_work> wrapster: yes.[19:16:54] <wrapster> or will be changed in b135...[19:17:03] <sstallion_work> package names were changed in snv_133[19:17:08] <RoyK> bbl[19:17:12] *** RoyK has quit IRC[19:17:19] <sstallion_work> generating IPS packages from the gate is still a little interesting.[19:17:45] <sstallion_work> Check the flag days - I think there are a couple of heads up in there for finishing up the conversion[19:18:07] <wrapster> sstallion_work: if its was changed in 133 then i still see usr/src/pkgdefs/ mentioning them as SUNW[19:18:24] <alanc> wrapster: SVR4 packages can't have names with '/' in, so the package names don't change in the onnv-gate until on-ips integrates to the gate tomorrow[19:18:28] <sstallion_work> wrapster: like I just told you, there are some older artifacts still sitting around.[19:18:36] * sstallion_work nods emphatically.[19:18:41] <wrapster> ok[19:18:50] <alanc> for builds 133-135, the package names are changed when the SVR4->IPS conversion is done[19:19:08] <wrapster> in all these I actually lost my original question...[19:19:17] <alanc> so the package names are changed in the IPS repositories as of 133, just not in the source gates[19:19:23] *** MuffiXXL has joined #opensolaris[19:20:00] <wrapster> i see in pkgdef/etc dir.. that there are a few drivers mentioned... arent they available as pkgs.. and if i have to install them..what do i do?[19:20:15] <eXeC001er> Hi[19:20:53] <wrapster> alanc: platform/i86pc/kernel/drv/cadp160[19:20:57] <wrapster> for example[19:21:11] *** dunc has quit IRC[19:21:12] <alanc> sounds like an encumbered driver[19:21:30] <alanc> I'd have to go look to find out, since I don't keep track of all the ON drivers[19:21:42] <wrapster> ok[19:21:50] <alanc> (you've built ON more times than I have)[19:22:18] <wrapster> alanc: yeah.. i like doing that :) and thats how i learnt quite a bit about libs and a whole lot more[19:22:19] <wrapster> :)[19:23:04] <alanc> yes, cadp160 is on the closed source, non-redistributable driver list - it's being considered for the extra repo, but for now, effectively doesn't exist for OpenSolaris[19:24:06] <wrapster> alanc: ok.. thanks.. but where did you find this info? weblink,pls?[19:24:32] <e^ipi> http://hub.opensolaris.org/bin/view/Main/no_source[19:25:09] <alanc> sorry, the link I checked is behind the Sun firewall, since it contains confidential information about Sun's licenses with the third party vendors[19:25:31] <wrapster> alanc: ok i understand...[19:25:41] <wrapster> but if i want to use these drivers.. how do i install them?[19:25:44] <wrapster> from that dir..[19:25:46] *** sommerfeld has quit IRC[19:25:49] <wrapster> could you tell me that at least?[19:26:04] <alanc> I have no idea if it's possible, or if so how[19:26:15] <wrapster> ok[19:26:22] <alanc> did I mention yet that I don't work on ON or the kernel in any way, shape or form?[19:26:48] <wrapster> alanc: ok :)[19:27:09] <sstallion_work> wrapster: if its not included in the closed bins, you are SOL.[19:27:14] <alanc> I suspect the first challenge would be getting a copy of the driver binary if it's not redistributable, since it wouldn't be in anything on opensolaris.org you can download[19:27:43] <wrapster> sstallion_work: SOL ?[19:27:52] <sstallion_work> wrapster: alternately, you can contribute an opensource equivalent like the rest of us poor saps ;)[19:28:09] <sstallion_work> wrapster: software term; out of luck :P[19:28:14] <wrapster> sstallion_work: seems interensting...[19:28:19] <wrapster> hee hee[19:28:20] <wrapster> ok[19:28:47] *** vraa has joined #opensolaris[19:29:33] *** calumb has quit IRC[19:29:37] *** sommerfeld has joined #opensolaris[19:29:37] *** ChanServ sets mode: +o sommerfeld[19:29:51] *** vraa has quit IRC[19:30:53] <alanc> "Simply Out of Luck" for the non-crude form, the more common substitutes a cruder S-word[19:31:52] *** vraa has joined #opensolaris[19:33:36] *** Yu\2 has joined #opensolaris[19:34:39] *** tg has quit IRC[19:35:47] *** tg has joined #opensolaris[19:35:56] *** xapiens has quit IRC[19:36:42] *** mega has joined #opensolaris[19:36:50] <Jenkens> mm jimmy johns[19:39:06] *** c0t0d0s0_ has quit IRC[19:40:14] *** g4lt-mordant has quit IRC[19:41:18] <bubbva> we're still about 40 members short of reaching quorum in #opensolaris-meeting. Reaching quorum in the meeting is a constitutional requirement in order to hold the election.[19:41:41] <bubbva> if you're an OpenSolaris core contributor, please come down the hall and announce yourself (format is in forum info). thanks[19:42:10] *** sactodave has joined #opensolaris[19:44:39] *** bewst has quit IRC[19:45:18] *** g4lt-mordant has joined #opensolaris[19:45:41] <eXeC001er> i try to create mirrored pool. but in 'zpool status' i have mirror-0. What is it?[19:46:38] <Auralis> its the number of the vdev mirror in the pool[19:47:43] *** xapiens has joined #opensolaris[19:48:41] *** perlmongo has quit IRC[19:53:02] *** fOB_2 has joined #opensolaris[19:54:00] *** bewst has joined #opensolaris[19:57:28] <CIA-37> George Wilson <George.Wilson at Sun dot COM>: 6923572 cannot dump to zvol when spa_config_lock is held as writer[19:57:29] <CIA-37> Sam Falkner <Sam.Falkner at Sun dot COM>: 6911541 'zfs mount' coredump[19:57:32] <CIA-37> Tim Haley <Tim.Haley at Sun dot COM>: 6912425 zpool import updates cache even with altroot[19:57:32] <CIA-37> Tim Haley <Tim.Haley at Sun dot COM>: 6499398 zdb has unreferenced file in ON[19:57:33] <CIA-37> Mark J Musante <Mark.Musante at Sun dot COM>: 6923102 spa_load_impl() leaks nvconfig if type is SPA_IMPORT_ASSEMBLE[19:57:35] <CIA-37> Jeff Bonwick <Jeff.Bonwick at Sun dot COM>: 6914746 sprintf_blkptr() causes NULL pointer dereference[19:57:36] <CIA-37> George Wilson <George.Wilson at Sun dot COM>: 6891437 DEADCODE in zfs_ioc_recv after dmu_recv_end.[19:57:37] <CIA-37> Neil Perrin <Neil.Perrin at Sun dot COM>: 6875045 zil log blocks can have too much stuff in them[19:57:39] <CIA-37> Chris Kirby <Chris.Kirby at sun dot com>: 6505339 dsl_pool_close() needs to call txg_list_destroy(&dp->dp_sync_tasks) to be nice, 6569720 zpool import -d <readonly filesystem> <pool> core dumps, 6566868 'zinject -c all' should not permit by regular user, 6561382 zpl version incompatability needs better error messaging, 6909405 concurrent replication actions fail to destroy each others' snapshots, 6794306 df.c calls libzfs_init with an argument when none is expec[19:57:40] <CIA-37> Eric Taylor <Eric.Taylor at Sun dot COM>: 6887372 DTLs not cleared after resilver if permanent errors present[19:57:41] <CIA-37> Matthew Ahrens <Matthew.Ahrens at Sun dot COM>: 6928246 spa_history_log should use dsl_sync_task_do_nowait[19:57:43] <CIA-37> George Wilson <George.Wilson at Sun dot COM>: 6929202 "zpool status" command hang when all the disks in a pool failed.[19:57:45] <CIA-37> Mark J Musante <Mark.Musante at Sun dot COM>: 6917573 ztest hangs on dsl_pool_sync()[19:57:46] <CIA-37> Tim Haley <Tim.Haley at Sun dot COM>: 6929453 panic in zfs-test hotspare_replace_003_pos[19:57:48] <CIA-37> Victor Latushkin <Victor.Latushkin at Sun dot COM>: 6915289 check for completed replacement in spa_vdev_resilver_done_hunt() in incomplete[19:57:50] <CIA-37> Sam Falkner <Sam.Falkner at Sun dot COM>: 6895223 zfs userspace/groupspace/allow/unallow/holds IOError in Japanese locale[19:57:52] <CIA-37> Matthew Ahrens <Matthew.Ahrens at Sun dot COM>: 6929652 dsl_sync_task_group_wait() can wait too long[19:57:55] <CIA-37> Matthew Ahrens <Matthew.Ahrens at Sun dot COM>: 6930022 zfs rename of non-shapshot is slower than necessary[19:57:55] <CIA-37> Mark Shellenbaum <Mark.Shellenbaum at Sun dot COM>: 6863610 ZFS mount can lose correct error return[19:57:56] <CIA-37> Mark J Musante <Mark.Musante at Sun dot COM>: 6918569 zpool_in_use() reports s2 (backup slice) as being in use[19:57:57] <CIA-37> George Wilson <George.Wilson at Sun dot COM>: 6808244 changing vdev state doesn't update config file, 6851389 zdb -l <block-device> should be discouraged, 6922161 zio_ddt_free is single threaded with performance impact[19:57:58] <CIA-37> Rod Evans <Rod.Evans at Sun dot COM>: 6918143 symbol capabilities, 6910387 .tdata and .tbss separation invalidates TLS program header information[19:57:59] <CIA-37> Ali Bahrami <Ali.Bahrami at Sun dot COM>: 6916788 ld version 2 mapfile syntax (fix relobj)[19:58:07] <mui> wat[19:58:16] <jbk> it's the zfs motherlode[19:58:46] <mui> what it says[20:00:19] *** hajma has joined #opensolaris[20:00:34] *** swa_work has joined #opensolaris[20:00:54] <jamesd2> yeah looks like 135 should be a good release... lots of zfs fixes.[20:02:29] <alanc> it's also the first build after the floodgates reopened for all the changes pent up waiting for 2010.03 to be finished[20:02:37] <alanc> like NWAM Phase I[20:03:06] <mui> why those fixed are not respun into release?[20:03:22] <jamesd2> alanc, then i'm not likely going to bea fan if they kept those things back... dedup performance sucks and needs a lot of work and that looks like the start of it...[20:04:09] <alanc> some fixes are being pulled back into 134a for 2010.03 - I have no idea if any of those ZFS ones are or are not[20:04:34] <tsoome> jamesd2: what hw you are using under dedup?[20:04:35] <mui> mokay[20:04:50] <alanc> I also don't know if the ZFS fixes were kept back or were just not done until now[20:05:10] <alanc> I was referring to other changes, like garrett's elxl replacement, that were waiting[20:05:27] *** aruntomar has quit IRC[20:05:34] <jamesd2> tsoome, amd dual core 2.4ghz, 6GB of ram, 4x 500GB raidz pool... without dedup, i get 40-50MB/s writes with dedup sub 1MB/s[20:05:42] *** Kedare has joined #opensolaris[20:05:44] <alanc> or some of the linker enhancements Rod & Ali have putback for 135[20:05:58] <tsoome> and how big is your arc?[20:06:44] <jamesd2> looking.. its zdb -S poolname?[20:06:54] <alanc> 136 will be entertaining as well, with the ON transition to native IPS packaging[20:08:48] <tsoome> arc_summary.pl or kstat -m zfs ...[20:09:15] <tsoome> or arcstat[20:09:15] <jamesd2> oops i mis read that.. currently my max arc is set at 3GB because of Sun VDI[20:09:28] * g4lt-mordant wonders when IPS will officially jump the IA32 shark, requiring more memory than possible on IA32[20:09:29] *** aruntomar has joined #opensolaris[20:09:31] <jamesd2> i thought you wanted to know the size of my dedup list.[20:09:33] <tsoome> 3G and you are complaining about speed?;D[20:10:03] <jamesd2> tsoome, well it was bad before... and lots of people will want to use dedup with only 2GB of ram...[20:10:10] <tsoome> ofc you should. poor thing is grabbing thise checksums from disk instead of using in memory cache[20:10:22] *** ptribble has joined #opensolaris[20:10:45] <tsoome> they may want whatever they wanna to. it does not mean its possible.[20:10:56] <g4lt-mordant> jamesd, uhm "only 2G of RAM"? that would be the entirety of IA32[20:11:24] <jamesd2> i'm currently trying to get the best mix settings for cache devices using cheap flash/sd memory cards and a vzol on a non deduped pool... poor man's ssd's[20:11:26] <tsoome> symantec puredisk is using postgres database for dedup engine[20:13:25] <tsoome> basically the point is, you are saving disk space with dedup, the *cost* is cpu power and memory.[20:14:07] * atomicpunk makes a mental note not to use puredisk.[20:14:19] <tsoome> there is no free meals.[20:14:33] <tsoome> lol[20:14:36] <atomicpunk> that's not what our politicians tell us! :)[20:14:43] <jamesd2> tsoome, and currently its using too much cpu if you dont have lots of ram and ssd l2arc/slogs[20:14:47] <timsf> tsoome: though you may be getting improved io as a happy side effect[20:15:10] <tsoome> sure, many thing can be improved.[20:15:26] <tsoome> but point stil remains, dedup is not for free[20:15:29] <timsf> [ but more deduped data means a better chance of a cache hit ][20:15:32] <timsf> indeed.[20:15:54] <jamesd2> yes reads with dedup are as good as without it... but writes totally suck[20:16:17] <tsoome> writes do the work there[20:16:45] <tsoome> read is just grab the block and feed it into app, with writes all the fun starts.[20:17:03] <mui> hum[20:17:12] <tsoome> actually, even with reads you will suffer perfomance hit.[20:17:34] *** yippi has quit IRC[20:17:42] <tsoome> as there is just 1 instance of block, and reads will load your spindles more.[20:17:45] <jamesd2> i know i understand that... but is it really necessary to get 1/100th of the performance of dedup on common home hardware...[20:18:06] <tsoome> give it more arc and see what will happen.[20:18:16] <mui> anything one could do improve reads in pool that is made of three raidz-sets and is gettin saturated with read ops?[20:18:38] <mui> resilver doesnt seem possible[20:18:54] <tsoome> mui: arc or l2arc will remove some burden[20:18:59] <jamesd2> tsoome, i didn't i saw this issues with 5GB of arc... i also saw lots of long standing locks that were also an issue, multithreading stuff in dedup will help a lot.[20:19:02] <mui> zil is disabled[20:19:19] <mui> but could adding l2arc help?[20:19:22] <jamesd2> mui, l2arc on a fast ssd's[20:19:22] <tsoome> zil has no influence on reads, lol[20:19:41] <mui> tsoome: yes i know :P[20:19:46] <mui> hm[20:19:59] <tsoome> i wouldnt disable my zil. ever.[20:20:00] *** bewst has left #opensolaris[20:20:40] <tsoome> i need my data;)[20:20:57] <mui> pool seems to saturate more if i turn prefetch on, is l2arc used even without prefetch?[20:21:17] <e^ipi> yes, it's a cache[20:21:24] <mui> i understand theres switch for using prefetching with l2arc[20:21:32] <tsoome> l2 will take time to be filled, once its filled, it will help a lot[20:22:06] <jamesd2> mui, yes.. l2arc is for stuff that is being kicked out of ARC memory that would of needed to be fetched from disk, but instead is fetched from your ssd instead.[20:25:28] *** Kedare has quit IRC[20:26:51] *** sommerfeld has quit IRC[20:26:54] <effremn> well i aggregated the wrong 2 nics together with dladm and had one on each switch - good times[20:27:29] <effremn> wish i knew some super cool ninja admin way to have figured that out without going to physically look at it[20:27:29] *** juwi has joined #opensolaris[20:28:47] <odyi> You could have looked at the Switches mac address table[20:29:39] *** MuffiXXL has quit IRC[20:29:39] *** juwi is now known as MuffiXXL[20:30:12] <effremn> @odyi - no switch access allowed for me - the other admins handle those but you're right - could have asked for a dump from them for my macs[20:30:32] * relling link aggregation bites another...[20:31:13] * odyi lives in the very weird world where I have access to everything from the physical through the application.[20:31:13] *** sommerfeld has joined #opensolaris[20:31:14] *** ChanServ sets mode: +o sommerfeld[20:31:28] <odyi> Networking and Computers alike.[20:31:49] <effremn> the larger the enterprise, the lower your control[20:32:00] <effremn> at least in my experience of it[20:41:33] <jamesd2> but even in the largest enterprise and you had the control it would be pointless anyway since you couldn't remember it all[20:42:15] *** milek has joined #opensolaris[20:44:04] <jamesd2> hi milek[20:44:14] <aruntomar> can i run osol on vitualbox with 512 mb ram without GUI[20:44:41] <jamesd2> not advisable... it takes at least 1GB of ram to install, and ZFS loves ram[20:45:20] <jamesd2> if you are a masocist after install you may get it to run in 512MB but don't complain about performance to us if you do.[20:46:21] <aruntomar> ok[20:50:09] *** hsp has quit IRC[20:50:13] *** bpmcd has quit IRC[20:50:36] *** hsp has joined #opensolaris[20:53:54] <g4lt-mordant> actually, you won't pkg will not work in the 132->133 transition with less than 1G of RAM[20:54:24] *** hsp has quit IRC[20:54:56] *** hsp has joined #opensolaris[20:57:30] <CIA-37> Petr Sumbera <Petr.Sumbera at Sun dot COM>: 6913894 logadm -c should be much faster if not atomic, 6578671 smf services keep logging to old smf logs after log rotation as well, until we restart them.[20:57:54] *** Kedare has joined #opensolaris[21:02:10] *** ryah is now known as ryah_away[21:02:19] *** Odin- has joined #opensolaris[21:02:57] *** Jenkens has quit IRC[21:03:40] *** gosx_ has joined #opensolaris[21:03:48] <milek> james: hello[21:05:13] *** hsp has quit IRC[21:05:46] *** hsp has joined #opensolaris[21:06:56] *** gosx has quit IRC[21:07:09] *** LenZ has quit IRC[21:07:12] *** merzo has joined #opensolaris[21:07:29] *** timsf has quit IRC[21:10:01] *** jthunder has joined #opensolaris[21:11:32] *** hsp has quit IRC[21:11:58] *** hsp has joined #opensolaris[21:12:08] *** g4lt-mordant has quit IRC[21:13:33] *** Bateau has joined #opensolaris[21:14:36] *** hsp has quit IRC[21:15:07] *** hsp has joined #opensolaris[21:15:11] <bobbyz> There's a tool I've seen referenced in here before for testing tcp performance between two servers on a local network...doesn't anyone remember off the top of their head the name of this tool? I believe you run it in a listening/server mode on one box and then fire up the client on the other box and test the throughut[21:15:23] <bobbyz> does anyone remember, rather?[21:16:34] <Auralis> netperf?[21:16:48] <bobbyz> ahh, yes I think that's it. Thanks...I was drawing a blank[21:16:51] <seanmcg> uperf, netperf, ttcp, nttcp... take your pick[21:17:15] *** Mdx4 has quit IRC[21:19:14] *** RoyK has joined #opensolaris[21:19:18] *** Mdx4 has joined #opensolaris[21:20:38] <RoyK> hi all. I got this new server today with 8 2TB drives for test[21:20:40] <RoyK> which is nice[21:20:43] *** LenZ has joined #opensolaris[21:20:44] <RoyK> but it's SLOOOW[21:21:08] *** vraa has quit IRC[21:21:20] <tsoome> get some fast drives?[21:21:33] *** hsp has quit IRC[21:21:40] <relling> yeah, some 2TB drives are 5,900 rpm :-([21:21:50] <RoyK> iostat -xc shows one drive responds very slow to commands (svc_t) and incidentially this is a drive I use temporary since they only sent me seven drives[21:22:02] <RoyK> it's up to several millisecs[21:22:03] *** hsp has joined #opensolaris[21:22:11] <tsoome> iostat -xn will show you both svc_t values[21:22:12] <RoyK> and it spins at 7200[21:22:30] <tsoome> from hba and from drive[21:22:31] <RoyK> tsoome: the other seven drives are ok, it's just this one[21:22:41] <tsoome> same firmware?[21:22:45] <tsoome> same vendor?[21:22:47] <RoyK> different make[21:22:54] <RoyK> different everything, really[21:22:58] <tsoome> :D[21:23:03] <RoyK> it was only meant to be used temporary[21:23:30] <RoyK> I guess I'll make a seven drive raidz2 for testing...[21:23:39] <tsoome> but if you create pool with 7?[21:23:50] <tsoome> is the load even then?[21:23:51] <RoyK> it's only for testing anyway[21:23:58] <RoyK> dunno - will try[21:24:13] <tsoome> maybe its disk connection causing that?[21:24:20] <RoyK> maybe[21:24:22] <RoyK> will see[21:24:32] <tsoome> or - rotate drives:)[21:24:42] <RoyK> I'm not at the office[21:24:49] <tsoome> then you can test with same 8 disk pool[21:24:51] <RoyK> but I'll try to exclude that one[21:26:13] *** IanLee has quit IRC[21:26:47] *** Animal-X_ has quit IRC[21:26:49] <RoyK> but I'll try to exclude that one[21:26:49] *** mroconnor has quit IRC[21:28:11] <nikolam> sooo, one can not delete files that are inside another BE, while it is mounted?[21:30:03] <nikolam> Hmm, I have some problems with sites that have a name something.domain.com etc[21:30:20] <nikolam> seems like I can not open them from Opensolaris machine[21:30:27] <seanmcg> nikolam, I just deleted a file in another BE.. mounted via beadm mount...[21:30:30] *** jthunder has quit IRC[21:30:34] <nikolam> neither if I give IP adress directly to FF[21:30:51] <nikolam> seanmcg, so it can be done. great :)[21:31:16] *** hsp has quit IRC[21:31:42] *** hsp has joined #opensolaris[21:33:02] <nikolam> Also i think i found a way, how to stay in One particular /dev release: pfexec pkg set-property pursue-latest false If that is it.[21:33:06] *** hsp has quit IRC[21:33:34] *** hsp has joined #opensolaris[21:33:36] <aruntomar> i installed mysql server with the following command, pfexec pkg install SUNWmysql5 but there is not mysql client with it, how can i install that[21:34:03] *** _madmax_ has joined #opensolaris[21:34:51] <RoyK> this needs further investigation[21:34:58] *** Mick27 has joined #opensolaris[21:35:32] *** luisbg__ has quit IRC[21:35:45] <nikolam> aruntomar, take a look inside contents of that package in package manager and see what files comes with it. i never installed it, but I guess that may help to see what is there[21:36:27] *** WireWulf has quit IRC[21:36:59] <aruntomar> nikolam: ok, i found it, it's not in the path, that's y it was not displaying it, also i'm i've stopped gdm, so anything i need to do, i've to rely on command line[21:37:49] <nikolam> you can ssh -X to it from another machine and run X apps, also[21:38:43] <ewdafa> what's a good way to monitor bandwidth usage via the console?[21:39:18] <tsoome> which bandwith?[21:39:32] <ewdafa> er tcp on one of the interfaces[21:39:47] *** mroconnor has joined #opensolaris[21:39:48] *** mroconnor has joined #opensolaris[21:39:57] <tsoome> netastat probably then...[21:40:05] *** TommyTK has quit IRC[21:41:19] * timeless_mbp ponders[21:41:27] <seanmcg> nicstat[21:41:34] <timeless_mbp> so, once my 'zpool scrub' finishes, should 'zpool status -v' _not_ cause my system to panic? :)[21:41:43] <seanmcg> google nicstat, very nice to bandwidth usage[21:42:15] <timeless_mbp> historically 'zpool status' caused my mac to panic, but at this point i need '-v' to get a panic, i'd like to see the errors hinted at by zpool status, but i don't really need more panics :)[21:42:48] *** gebi has joined #opensolaris[21:43:10] <aruntomar> how to list the contents of a pkg from command line? pkg info pkg_name shows the description, but no info on what and where the pkg's were installed[21:43:52] <aruntomar> ewdafa: u can also try iftop[21:45:27] <ewdafa> seanmcg: yeah i just found out about that, thanks[21:45:45] <tsoome> and loads of DTT scripts:P[21:46:55] *** hsp has quit IRC[21:47:20] <aruntomar> ewdafa: if u need historical usage and try ntop, u can get it from blastwave[21:47:23] *** hsp has joined #opensolaris[21:48:19] <crichardso> anyone here do aggr links with hp procurves?[21:52:53] *** aruntomar has quit IRC[21:53:17] *** PatrickS has joined #opensolaris[21:54:33] <gdamore> lewellyn: r.e. the webrev title...its because I reused a workspace....[21:57:46] <alanc> aruntomar: "pkg contents" shows the contents[21:57:59] <alanc> (or anyone who didn't know since he already left)[21:59:26] <nikolam> Do we have some useful usage of so called "Super key" or Win key?[21:59:51] *** RoyK has quit IRC[21:59:57] <sstallion_work> nikolam: whatever you want to bind to it[22:00:07] <sstallion_work> (its typically called the meta key)[22:00:20] <nikolam> but not default, yes[22:00:29] <nikolam> oh and meta, yes[22:00:37] <sstallion_work> not that I've noticed. I usually bind meta-R to vbox[22:01:08] <nikolam> there is yet to see Osol logo on that key on any computer[22:01:38] *** eXeC001er has quit IRC[22:02:02] <gdamore> i have meta bound to.... meta... for use on emacs and in shells. :-)[22:03:00] *** Gjermund has joined #opensolaris[22:06:25] *** wdp has quit IRC[22:06:36] *** rotondo_ has joined #opensolaris[22:06:38] <Gjermund> Hi, regarding Xscreensaver, I get ld.so.1: molecule: fatal: LibGL.so.1: open failed: No such file or dierectory when using 'Molecule' saver. Anyone know this. LD_LIBRARY_PATH probs? Or other things?[22:07:07] <Gjermund> For 2009.06 SPARC[22:07:16] <Gjermund> x86 is fine[22:07:43] <rotondo_> ogbot register rotondo as Scott Rotondo[22:07:47] *** wdp has joined #opensolaris[22:07:53] *** cwebber has joined #opensolaris[22:07:55] <e^ipi> rotondo_: #opensolaris-meeting[22:08:30] <rotondo_> sorry, I see I forgot to set the room[22:08:33] *** rotondo_ has left #opensolaris[22:10:02] *** olivier_54 has joined #opensolaris[22:11:08] <timeless_mbp> http://pastebin.mozilla.org/705578[22:11:45] *** engrxyz has quit IRC[22:12:17] *** nikolam has quit IRC[22:12:21] *** ryah_away is now known as ryah[22:12:41] <e^ipi> timeless_mbp: yep, that's why you run with redundancy[22:12:44] *** nachox has quit IRC[22:12:54] <e^ipi> zfs would just fix it in that case[22:12:56] <timeless_mbp> e^ipi: that's really hard to do w/ a laptop...[22:13:02] <e^ipi> set copies=2[22:13:03] <bda> oof.[22:13:05] <e^ipi> so, not really[22:13:14] <e^ipi> it doesn't help if a disk dies but it should help fix cksum erros[22:13:14] <bda> Presumably no ECC either.[22:13:14] <timeless_mbp> on the pool?[22:13:15] *** nikolam has joined #opensolaris[22:13:19] <e^ipi> timeless_mbp: yes[22:13:33] <Auralis> or get a laptop with dual drive bays :)[22:13:36] *** rulerrrr has quit IRC[22:13:41] <timeless_mbp> cannot set property for 'pool': invalid property 'copies'[22:13:54] <e^ipi> my bad, on the filesystems then[22:13:56] <timeless_mbp> e^ipi: what version of zfs / zpool do i need for that?[22:14:14] <timeless_mbp> ok, that's working[22:14:15] <tsoome> copies is fs property, not pool[22:14:17] *** olivier_54 has left #opensolaris[22:14:18] <jbk> yeah, it's per fs[22:14:28] <timeless_mbp> it's inherited[22:14:33] <timeless_mbp> so setting it w/ zfs on the pool works[22:14:46] <timeless_mbp> will zfs automatically dupify existing data?[22:14:50] <tsoome> no[22:15:12] <tsoome> nor compress, nor dedupe, nor decompress:P[22:15:25] <timeless_mbp> yeah yeah[22:16:28] *** hsp has quit IRC[22:16:58] *** hsp has joined #opensolaris[22:17:06] <jbk> you have to wait for bp rewrite to arrive[22:17:41] <timeless_mbp> is that a planned or wished feature?[22:17:49] <jbk> planned[22:17:52] <jbk> but it's kinda tricky[22:18:07] <tsoome> almost wished:D[22:18:34] <jbk> IIRC, it works for idle pools (what's in development)[22:18:44] *** logic855 has quit IRC[22:18:56] <tsoome> need guid regen option for zpool import :([22:19:15] *** hsp has quit IRC[22:21:04] *** palle has joined #opensolaris[22:21:23] *** hsp has joined #opensolaris[22:21:25] *** Termina_ has joined #opensolaris[22:21:26] *** Termina_ is now known as Termina[22:21:36] *** LouisJB has joined #opensolaris[22:22:08] <Termina> I'm trying to install a Solaris 10 container in osol, but am unable to find s10u8b8ax.flar[22:22:10] <Termina> http://hub.opensolaris.org/bin/view/Community+Group+zones/s10brand_dev_guide[22:22:14] *** tnoergaard has quit IRC[22:22:35] <Termina> The full path they use is: /net/kodiak.sfbay/gates/s10brand/public/images/s10u8b8ax.flar Google shows other people using this path but I have no idea how to mount it. :P Can someone help?[22:22:44] *** logic has joined #opensolaris[22:22:51] *** logic is now known as logic855[22:23:20] *** Gjermund has quit IRC[22:23:42] <nikolam> too bad eye of gnome can not play animated gifs..[22:24:14] <Termina> haha, yeah...[22:24:24] <Termina> isn't it fun just using firefox? ;)[22:24:42] <ampex> people actually use opensolaris as a desktop os?[22:25:03] <seanmcg> ampex, sure[22:25:05] <Termina> I tried that for a little bit xD[22:25:07] <tsoome> Termina: you cant unless you are in sun network;)[22:25:13] <ampex> why on earth?[22:25:16] <joshua_> Termina, that seems like a Sun internal path[22:25:17] <nikolam> Termina, yes, also because you can not deinstall Firefox, unless you want to get a rip of your system[22:25:24] <Termina> ah[22:25:25] *** swa_work has quit IRC[22:25:32] <timeless_mbp> Termina: /net/*.sfbay only works on the 'sun wan' (swan)[22:25:35] <seanmcg> ampex, heard of sunray ?[22:25:42] <comay> ampex because it has technology that others distros lack that are useful on a desktop os[22:25:45] <Termina> I see. Is there a public place to get things from othere?[22:25:48] <Termina> *there[22:27:20] <bda> No.[22:27:38] <bda> You might want to make a FLAR of an existing Solaris 10 system you manage.[22:28:06] <bda> Or, presumably, you could install Solaris 10 in the branded zone?[22:28:14] * bda guessing, hasn't looked into s10 containers.[22:28:28] *** palle has quit IRC[22:28:51] <Termina> Hm, is a cpio archive the same as an iso?[22:29:04] <alanc> no[22:29:04] <bda> No.[22:29:08] <Termina> boo xD[22:29:14] <Termina> so i'll have to install solaris 10[22:29:17] <Termina> to make a zone for solaris 10[22:29:18] <Termina> :/[22:29:18] *** mroconnor has quit IRC[22:29:19] <alanc> it's an archive format like tar[22:29:42] <bda> Termina: Maybe you should read up a bit more on how branded zones work.[22:29:43] <seanmcg> Termina, if you're on x86 just install to a vbox[22:29:59] <seanmcg> cheap[22:30:01] <Termina> bda, I'm familar with Linux branded zones.[22:30:31] <Termina> seanmcg, virtualbox?[22:30:31] <bda> Termina: If you're familiar with Solaris licensing, too... ;)[22:30:41] <Termina> oh, right... :/[22:30:49] <seanmcg> Termina, yup[22:31:10] * bda pets JET.[22:31:13] <seanmcg> can create the flar from the vbox install thereafter if need be[22:31:15] *** wrapster has left #opensolaris[22:31:24] <bda> 15 minutes of work, hands-off Solaris 10 installs forever.[22:32:13] <Termina> hm... can you pxe boot solaris?[22:32:13] <joshua_> Is there a way to pin pkg at a specific version?[22:32:23] <bda> Termina: Yes.[22:32:40] <joshua_> i.e., I haven't upgraded to 133 yet, and I want to always install @*-0.132.[22:32:45] <bda> Termina: Jumpstart. Or the Jumpstart Enterprise Toolkit.[22:32:51] <seanmcg> Termina, on x86 pxeboot is the net install method..[22:33:02] <seanmcg> so, yes..[22:33:07] <tsoome> pxeboot is boot method:P[22:34:39] *** ryah has left #opensolaris[22:34:59] <seanmcg> joshua_, read the faq http://hub.opensolaris.org/bin/view/Project+pkg/FAQ[22:35:14] *** CoolMa has joined #opensolaris[22:37:03] *** Guest32219 has joined #opensolaris[22:37:05] *** alitzell has quit IRC[22:37:33] <joshua_> seanmcg, Hmm... I appear to have 'entire' for this build already installed, but something like 'pkg install SUNWphp52' attempts to bring the whole system up to .0.133 alogn with it.[22:37:43] *** alitzell has joined #opensolaris[22:39:10] <seanmcg> joshua_, try asking on #pkg5 ? The pkg folks hang out there.[22:39:31] *** Cripps has quit IRC[22:39:48] <joshua_> okay -- thanks for the pointer[22:39:48] * seanmcg heads offline for the night[22:39:58] <joshua_> nighto[22:43:24] <Termina> how does one create a "flar"?[22:43:32] <Termina> assuming one had solaris 10 installed[22:44:07] *** Cripps has joined #opensolaris[22:44:30] <bda> flarcreate(1M)[22:46:12] *** c0t0d0s0_ has joined #opensolaris[22:47:01] <Termina> excellent! thanks[22:47:33] <Termina> solaris sure is weird when you're used to linux :P[22:47:44] <Termina> I DEMAND EVERYTHING TO BE THE SAME[22:47:50] <bda> Then go use Linux. :)[22:48:00] <Termina> Yeah... but then no ZFS or containers :P[22:48:05] <Termina> *good containers[22:48:39] <Termina> how many years has solaris had containers for?[22:48:46] *** Beket has quit IRC[22:49:04] *** Guest32219 has quit IRC[22:49:12] <atomicpunk> I find solaris / linux no more frustrating than say, gentoo / fedora. :)[22:49:15] <bda> I'm pretty sure U1 had them, so 2005?[22:49:29] <Termina> nice[22:49:39] *** drewj has quit IRC[22:50:03] *** drewj_ has joined #opensolaris[22:50:05] *** drewj_ is now known as drewj[22:50:22] * bda tries to find somewhere to stash 4TB of logs so he can rebuild this pool.[22:50:31] <jamesd2> solaris had ldoms 10+ years ago... at least on the sun fire 10k[22:50:38] *** g0ku has joined #opensolaris[22:50:41] *** g0ku has left #opensolaris[22:50:59] <Termina> 4TB of logs? god damn[22:51:12] <bda> jamesd2: Ah, good call.[22:51:26] *** jollyd has joined #opensolaris[22:51:30] *** jollyd has left #opensolaris[22:52:21] <jamesd2> 10 years ago solaris had technology that linux wont have 5 years from now... and it wont be as well designed and though out even after 10 more years.[22:53:06] *** Lacap has joined #opensolaris[22:53:36] <tsoome> ldoms are not exactly domains from midrange/high[22:53:49] <tsoome> those domains were electrically isolated[22:54:19] <Bateau> anyone know of a good location to find a good bind walkthrough? I need to act as a DNS server for my LAN, as well as for serveral domains to the internet at large (but I do not want to recurse for the whole world)[22:54:26] <jamesd2> and solaris had projects which are pseudo containers for resources in Solairs 9 maybe even 8.[22:54:33] *** piwi has joined #opensolaris[22:54:40] <e^ipi> Bateau: BIND is the same on solaris as anywhere else[22:54:50] <ptribble> Termina: I've been using zones in production since early 2004, Solaris 10 has always had them (the container term has been used for several related things)[22:55:05] <Bateau> e^ipi, I'm having trouble with the getting-it-running part, I've done config on a few other platforms before[22:55:18] <jbk> heh.. another genius marketing move =][22:55:29] <jbk> are they zones or containers or zone containers? :)[22:55:43] <e^ipi> /etc/named.conf and bob's your uncle[22:55:45] <joshua_> bda, The good news, I guess, is that logs compress well, right/[22:55:46] <tsoome> svcadm for dns/server, rest is normal /etc/named.conf[22:56:04] *** swa_work has joined #opensolaris[22:56:32] <tsoome> and named_checkconfig etc...[22:57:28] <Bateau> ok, ty, will work on it some more[22:57:45] <CIA-37> Rod Evans <Rod.Evans at Sun dot COM>: 6918143 symbol capabilities (fix lint), PSARC/2010/022 Linker-editors: Symbol Capabilities[22:57:57] <tsoome> check syslog messages for errors as well.[22:58:52] <bda> joshua_: Yes. compressratio 2.19x :)[22:59:14] <joshua_> bda: This is with lz? or gzip?[22:59:17] <bda> joshua_: But zfs send doesn't keep the compression, so... 1.67 in-use on the logserver will be ~4TB real wherever I zfs send it to.[22:59:25] <bda> lz.[23:00:17] <jamesd2> bda, you can just pipe the send stream or file to bzip2 or compression program of your choice[23:00:28] <bda> jamesd2: Sure, if I want to store it as a file.[23:00:38] <joshua_> Oh, loss loss.[23:00:45] <bda> jamesd2: That isn't very helpful for failing over the syslog zone that houses the data, though. :)[23:00:50] <jamesd2> bda, "|" is your friend...[23:01:32] <jbk> and if you set compression on the destination fs before you start the zfs recv ?[23:01:35] <bda> jamesd2: zfs send | bzip2 | zfs recv is going to do what for me?[23:01:51] <bda> jbk: shrug, doesn't work, unless my info is out of date.[23:02:12] <jamesd2> bda, you add bunzip2 in there.. and it uses less data over a network and zfs will compress on the other end for you...[23:02:24] <joshua_> hmm? zfs send does keep the compression AFAIK[23:02:49] <joshua_> at leas,t it did when I migrated my storage pool around[23:03:13] <bda> Maybe I'm crazy, then.[23:03:47] <joshua_> It's easy enough to determine on a small scale, anyway.[23:04:04] <bda> Sure, like:[23:04:27] <bda> Fileserver: used 759M (1.19x)[23:04:35] <bda> Backup: used 908M (1.00x)[23:04:50] <joshua_> Really?? Are you running s10 or OSol?[23:04:57] <bda> Solaris 10 U8 on both sides.[23:05:20] <joshua_> Oh, hm. I dd this on OSol.[23:06:29] *** jthunder has joined #opensolaris[23:07:21] *** vraa has joined #opensolaris[23:08:09] <joshua_> Oh, interesting/exciting: I am talking to my advisor on Thursday about spending a semester getting research credit for doing a rewrite of ZFS on Linux.[23:09:02] <joshua_> (Well, not "my" advisor, but the appropriate advisor; Garth Gibson.)[23:09:06] <jamesd2> joshua_, you mean fuse, or you going to do it in the kernel so will need a lawyers help if you ever want to distribute it?[23:09:07] <bda> You mean like btrfs?[23:09:13] <bda> ha.[23:09:16] *** sergiusens has joined #opensolaris[23:09:33] <gebi> you'd only need a lawyer if you want to distribute it in binary form *g*[23:09:34] <bda> What does "rewrite" mean in this context? A feature-complete copy or a port?[23:09:46] <bda> Because, uh, ZFS has a lot of manhours behind it.[23:09:50] <joshua_> jamesd2, (I plan to do this with another student) We don't plan to do a port; we plan to do a feature copy.[23:09:58] <joshua_> My understanding is that a lot of the man hours are design work.[23:10:09] <e^ipi> patented design work[23:10:47] <e^ipi> you have patent immunity if you use the CDDL implementation[23:11:04] <e^ipi> if you don't, well... lawyer up[23:11:12] <gebi> yes, your only chance afaik is to use the sun zfs implementation and plug exactly that into linux[23:11:25] <e^ipi> gebi: in which case, lawyer up[23:11:26] *** hsp has quit IRC[23:11:32] <e^ipi> because you're in violation of the GPL[23:11:36] <gebi> no[23:11:38] <jamesd2> joshua_, you picked the wrong time to do this.. oracle is too much of an unknown, if you had started this a year ago, the ZFS devs may of been more freindly than they are currently being they are now new employees of Oracle.. the DTrace developers came out and said they were willing to help port DTRace to linux...[23:11:47] <gebi> you just can't distribute it in binary form[23:11:52] <gebi> source is ok, no gpl there[23:12:18] <e^ipi> *shrug* it's your lawsuit[23:12:22] <joshua_> gebi, I don't want to start this debate here; but not quite true. (In particular, the argument is made that source is a derived work.)[23:12:43] <joshua_> I would be surprised if I got open patent-related hostilities for a ZFS reimpl.[23:13:04] <joshua_> jamesd2, Yeah; we mostly intend to do it on our own.[23:14:02] <gebi> some time ago on the zfs-fuse ml there was a note about a company writing a zfs kernel module for linux[23:14:10] <jamesd2> joshua_, something could be said with starting with the read-only gpl'd version of ZFS then you have really muddied the waters because you are taking a gpl'd version released by the code and patent owners and forked it into something else that is also gpl'd... which is both good and bad from the legal perspective...[23:14:12] <MindDrive> I suspect reimplementing ZFS with just two people will take far more time than you think it will...[23:15:42] <joshua_> jamesd2, You mean the GRUB implementation? I don't believe that the GRUB implementation has the level of layering needed to ever transform it into a version with read/write support without rewriting most of it...[23:16:06] <e^ipi> well, good luck with that[23:16:28] <joshua_> yes, it is very much of a "gg gl hf dd" proposition; were it to happen, it would happen about a year from now, so it is quite a ways out[23:16:29] *** MuffiXXL has quit IRC[23:16:44] <joshua_> it would be a senior thesis for sully, and research credit for me[23:17:03] *** timeless_mbp has quit IRC[23:17:04] <jamesd2> joshua_, well it understands single disk and mirrored... so its a good start, you can use it as a starting point... get minimal read/write support going, and then release see how much legal flack you are going to get.[23:17:14] <joshua_> hmm[23:17:24] *** omry_|work has quit IRC[23:18:04] <joshua_> There is precedent for other OSes with pure-other-licensed bootcode (namely, FreeBSD), too.[23:18:39] <jamesd2> once they release most likely you will only get a cease and decist order from legal... if you stop at that point.. it will probably just go away.. since you haven't hurt oracle/sun bottom line at that point[23:19:04] <bda> Could also talk to your .edu's lawyers.[23:19:05] *** torchie has joined #opensolaris[23:19:12] <joshua_> again, I would be surprised if I picked up patent-related flak[23:19:13] <e^ipi> joshua_: the other os's are using the CDDL version, which grants patent immunity[23:19:37] <joshua_> it is considered "very unfriendly" to hit open source projects in fields in which you are trying to build a community with patent threats[23:19:54] *** baza11215 has quit IRC[23:20:30] <jamesd2> joshua_, billion dollar companies often have legal departments that dont communicate with anyone else in the company they are a group unto them selves.[23:20:32] <alanc> I don;'t think Sun ever told an open source project to stop using methods that Sun owned patents on, and was usually liberal in licensing patents to open source projects[23:20:43] <alanc> I have no idea what Oracle's policies on the matter are though[23:20:47] <joshua_> whether this will be the case with Oracle, though[23:20:48] <joshua_> righ[23:20:49] <joshua_> t[23:21:09] <e^ipi> hey, until it's litigated nothing a lawyer tells you is fact either[23:21:12] <jamesd2> alanc, exactly i did mention to joshua_ that he picked the wrong time for this.. a year ago would of been much better.[23:21:20] <e^ipi> you are exposing yourself to legal risk, if you're okay with that go ahead[23:21:25] <joshua_> it is potentially also the case that CMU has patents that could be used in a "no, fuck off" response[23:21:27] *** ptribble has quit IRC[23:21:28] <Bateau> no joy trying to get named to run, log shows [ Mar 2 00:04:31 Method "start" exited with status 1. ][23:21:54] <joshua_> yes, I will suck it up w.r.t. legal risk, especially if I have CMU backing me[23:22:03] <jamesd2> Bateau, have you tried running named by hand not inside smf, perhaps enable more debugging info.[23:22:05] <joshua_> we are concerned about copyright, not patents[23:22:12] <e^ipi> suffice it to say that every couple months someone comes up with the idea of porting ZFS to linux, and we usually hear nothing of it a month or two later[23:22:15] <alanc> Oracle's also in the business of trying to sell Linux, so may be more friendly to Linux porting efforts than Sun was - who knows?[23:22:35] <Bateau> trying to call it just returned me to the prompt, no output, what is the argument for more debugging?[23:22:38] <e^ipi> i'd place my wager with "it's harder than people think"[23:22:40] <joshua_> e^ipi, Yes. You are unlikely to hear about this for another year if it happens.[23:23:29] <joshua_> Both of us who woudl be involved with this have worked on pretty large scale projects in C (on the order of tens of thousands of lines), pretty much done from scratch, and both of us have pretty good source-diving experience, though.[23:23:46] <snuff-work> i really think zfs will never goto linux.. they have their brtfs..[23:24:11] <jamesd2> joshua_, you should deffinetly wait a few months before contacting sun employees officially on the subject since they are all walking on egg shells.. they don't know what there new bosses and managers want so are more likely to say "No" when if you wait till the honeymoon is over between oracle and sun you will have better luck.[23:24:12] <e^ipi> 10klines is'nt a large project, zfs is over 100,000[23:24:30] <Bateau> named -d {1-4} -c /etc/named.conf give me nothing, back to command line silently[23:24:33] <bda> Is it really a honeymoon when you've bought the bride? :)[23:24:38] <jamesd2> there are some files in ZFS that are over 10k...[23:25:30] <joshua_> jamesd2, Yes, we don't intend to really move forward for 6 months still.[23:26:08] *** funde has joined #opensolaris[23:26:21] <joshua_> Also, apropos of nothing: how do pNFS and ZFS interact?[23:26:44] <joshua_> If pNFS feeds locations to blocks elsewhere, how do ZFS's checksums interact with that?[23:27:14] <e^ipi> doesn't NFS deal with files, not blocks?[23:27:23] <funde> Hi. I want to dive into the Solaris kernel source code (like reading it). Can anyone recommend some starting points (file-names)? Thanks.[23:27:40] <e^ipi> funde: what are you interested in?[23:27:43] <relling> pNFS can use ZFS datasets for stores[23:27:44] <jamesd2> joshua_, nfs doesn't care about ZFS checksums... ZFS keeps its knowledge of checksums hidden.[23:27:49] <e^ipi> filesystems? dispatcher ?[23:27:53] <joshua_> NFS does not; pNFS does deal with blocks[23:28:00] <jamesd2> funde, cvs.opensolaris.org rocks[23:28:15] <sommerfeld> funde: http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/uts/common/os/main.c[23:28:18] <sommerfeld> :-)[23:28:21] *** timeless_mbp has joined #opensolaris[23:28:24] *** timeless_mbp has joined #opensolaris[23:28:34] <sommerfeld> (actually, a fair bit of initialization happens before the kernel's main() is entered..)[23:29:00] * timeless_mbp cries[23:29:02] <jamesd2> sommerfeld, isn't that in the same area where the last profanity in the kernel was removed before release?[23:29:05] <joshua_> http://hub.opensolaris.org/bin/view/Project+nfsv41/WebHome[23:29:11] <snuff-work> pNFS is still very alpha.. i think the current package drop will destroy any data on restart..[23:29:26] <timeless_mbp> hey, can random people remove random whitespace from e.g . _6701797 in http://hub.opensolaris.org/bin/view/Community+Group+zfs/12 ?[23:29:28] <funde> sommerfeld: thanks. I should have known... :)[23:29:28] <joshua_> heh! :)[23:29:49] <joshua_> I have a very vague concept of how pNFS works in my mind, and I then noted that OSol supported it, and went like "??? how does that interact with ZFS?"[23:29:54] <joshua_> and the answer, then, is "not at all?"[23:29:57] <sommerfeld> as_gap is still in there.[23:30:38] <alanc> timeless_mbp: the ZFS community group would have to grant you access to edit their pages, or set them to be publicly editable (default is editable by group members only)[23:30:54] <timeless_mbp> alanc: ok, because http://hub.opensolaris.org/bin/view/Community+Group+zfs/7 has 'tthe' in the middle[23:31:28] <sommerfeld> (as in "tear someone a new as_gap"[23:33:05] *** chototsu has quit IRC[23:33:57] *** funde has quit IRC[23:34:17] <alanc> and panics often involve "assfail" messages, which we always tell managers is an abbreviation for "assertion failure", not "failed to find your own ass with both hands tied behind your back" or a particularly nasty gastro-intestinal outbreak[23:34:32] *** radsy has joined #opensolaris[23:34:40] <timeless_mbp> heh[23:37:52] <timeless_mbp> so...[23:38:02] <timeless_mbp> what does zpool scrub actually *do*?[23:38:14] <timeless_mbp> it listed a certain number of issues[23:38:28] <timeless_mbp> but after triggering another panic, zpool status no longer remembers this[23:38:29] <bda> Verifies block checksums.[23:38:53] <timeless_mbp> is it an in memory only operation?[23:38:58] *** CoolMa has quit IRC[23:39:13] <Stric> reads all blocks (not just "enough" as regular reads) and verifies the checksums[23:39:44] <timeless_mbp> roughly, once a scrub has found all of my failed checksums, what can i actually *do*?[23:39:49] <timeless_mbp> can i safely use zfs send?[23:40:04] <Stric> then you know that your hardware sucks[23:40:11] <e^ipi> if you were running with redundancy it fixes them[23:40:14] <timeless_mbp> well, i already knew that :)[23:40:19] <timeless_mbp> i wasn't...[23:40:26] <bda> If your data is indeed corrupt, and you are not running redundant, you're just going to copy hosed data.[23:40:36] <bda> Restore the hosed files from backup.[23:40:46] <Stric> zfs won't allow you to read bad blocks[23:40:47] <timeless_mbp> the last time i tried to use zfs send, zfs triggered a panic[23:40:52] <bda> Stric: Huh.[23:41:06] <Stric> bda: returning EIO or something[23:41:13] <bda> Well, that makes sense. :)[23:41:19] <timeless_mbp> i presume that if i try to use zfs again now, having panic'd since my last zfs scrub, it'll panic again?[23:41:38] <Stric> try? :P[23:41:39] <bda> Only the Small Gods of Screwing With Your Day know.[23:42:00] <timeless_mbp> Stric: oddly, i don't actually *enjoy* panic'ing[23:42:10] * timeless_mbp can't imagine why[23:42:18] <e^ipi> beats the alternative[23:42:41] <e^ipi> keep trying forever and have to reset without knowing wha's wrong[23:44:03] *** gm152 has joined #opensolaris[23:44:04] <Stric> timeless_mbp: what's your zpool failmode (or somesuch) set to?[23:44:43] <timeless_mbp> ?[23:45:25] <timeless_mbp> e^ipi: what i'd like right now is a list of corrupt files[23:45:25] <Stric> zpool get all yourpool[23:45:32] <timeless_mbp> so i can do something about it[23:45:35] <bda> zdb hackery.[23:45:46] <timeless_mbp> http://pastebin.mozilla.org/705601[23:45:56] <e^ipi> timeless_mbp: pull the tapes[23:46:25] <bda> Is /usr/sbin in your PATH?[23:46:36] <e^ipi> hacking around in zdb is decidedly unintuitive[23:46:38] <bda> But uh, zdb is going to just cause you pain. Run another scrub.[23:46:39] *** niq has quit IRC[23:46:44] <Stric> how old is this system?[23:47:02] <relling> FreeBSD?[23:48:05] <bda> Heh.[23:48:08] <relling> zpool status -v shows b0rken files, at least on Solaris it does[23:48:12] *** alecmuffett has left #opensolaris[23:49:34] *** mega has quit IRC[23:49:39] <timeless_mbp> osx[23:49:44] <bda> haha.[23:49:51] <timeless_mbp> relling: for me, zpool status -v shows the osx panic picture[23:49:51] <bda> Sorry. :)[23:50:01] <timeless_mbp> and then i get to turn it off...[23:50:11] <bda> From what I've been told, isn't panicing what ZFS does on OS X. ;)[23:50:16] <e^ipi> i didn't think i was in #osx[23:50:28] <e^ipi> how is OSX's zfs being broken at all relevant here?[23:50:34] <bda> Also, isn't the most recent build for that like.. b73?[23:52:07] <relling> timeless_mbp: you might have better luck on the zfs-macos google group[23:52:42] <Spencer_tt> does open source software mean linux to oracle or I'm just totally confused..[23:54:46] *** phips has quit IRC[23:55:14] *** phips has joined #opensolaris[23:55:50] <e^ipi> Spencer_tt: speculation... not here[23:55:52] <basic`> hmm[23:56:26] <Spencer_tt> fmm.[23:56:28] *** kwinz3 has quit IRC[23:56:38] <basic`> i have a rdr ipnat rule that isn't working on b133, im not sure if this is related to upgrading from 2009.06 to the dev builds or something else causing issues, but it doesnt work now (and it used to)[23:57:02] <basic`> did a lot change wrt ipfilter/ipnat?[23:57:51] *** sndcrb has quit IRC[23:57:53] *** kimc has joined #opensolaris[23:58:03] *** Chubbz has joined #opensolaris[23:58:25] *** ma3x_ has joined #opensolaris