March 11, 2009 
< | 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 11, 2009 [00:00:31] <jsoftw> Ahh of course. Cant have that looping in dnode_special_close() going wrong at this time of the year[00:02:08] <jmcp> jsoftw: you might want that fix if you're trying to import a pool with an error[00:02:24] *** jacobs has quit IRC[00:02:58] *** syamajala has joined #opensolaris[00:05:33] *** piwi has joined #opensolaris[00:07:34] *** coffeetime has quit IRC[00:09:16] *** rv-_ is now known as rv-[00:10:05] *** c00p_ is now known as c00p[00:10:09] *** Tpenta has joined #opensolaris[00:11:24] <c00p> nothing wrong with being a sysadin :\[00:11:34] <c00p> *admin[00:12:39] *** mikefut has quit IRC[00:14:00] <ry_laptop> the janitors of the computer industry.[00:14:11] <ry_laptop> unite![00:14:24] <eviljames> haha, hardly.[00:14:42] <ry_laptop> hmm, that's a window's admin huh?[00:14:45] <eviljames> I think it's more fair to call them "people programmers rely on to do anything useful"[00:14:54] <ry_laptop> heh[00:15:21] <ry_laptop> or "people that deduce WTF your programs are supposed to do"[00:15:21] *** ChanServ sets mode: +o Tpenta[00:15:34] <e^ipi> not entirely true, i can write code on a machine that has no sysadmin[00:15:49] <eviljames> e^ipi: A machine that you're the sysadmin on?[00:16:04] <Topdeck> Hi all, I'm after some more information about how idmap works. I want to have no users on the OpenSolaris box but still be able to do something like "chown user@domain /path/to/file". Is that possible?[00:16:21] *** radsy has joined #opensolaris[00:16:23] <e^ipi> eviljames: "we're all winners?"[00:16:33] *** NoFX_SBC has quit IRC[00:17:17] <e^ipi> not saying sysadmins aren't useful ( or rather, necessary )[00:17:27] <e^ipi> just that a developer doesn't technically depend on a sysadmin[00:18:15] <scoffin> one developer does not, but 40 do :-)[00:18:52] <e^ipi> fair enough[00:19:01] <eviljames> Indeed, look at my statement: "programmerS" (!)[00:19:02] <eviljames> heh[00:19:21] <e^ipi> plural means any number greater than or equal to 2[00:19:33] <e^ipi> 2 programmers can get away without a sysadmin just fine[00:19:44] <eviljames> Nope. One of them is doubling as a sysadmin.[00:19:52] <ry_laptop> heh[00:19:56] <e^ipi> :P[00:19:57] <ry_laptop> what have I started.[00:20:00] <eviljames> Either: managing the git repo or the web server or some other detail other than programming..[00:20:02] <oninoshik1> eviljames: atleast[00:20:08] *** rmorse has joined #opensolaris[00:20:22] <ry_laptop> well. a programmer that doesn't know how to sysadmin isn't worth much IMHO.[00:20:30] <eviljames> ry_laptop: I disagree with you again![00:20:33] <ry_laptop> :)[00:20:36] <ry_laptop> damnit![00:20:50] <eviljames> ry_laptop: There's a programmer in here who couldn't sysadmin for a shits worth. However, he has a master's in computing science and claims to write "bug-free" programs.[00:20:59] <ry_laptop> ...[00:21:01] <e^ipi> he's lying[00:21:06] * ry_laptop gets a bridge to sell eviljames.[00:21:06] <eviljames> ry_laptop: Considering his track record, I believe him so far. I have yet to find a bug in any program he's written.[00:21:09] <eviljames> e^ipi: Of course he's lying[00:21:24] <eviljames> but, the results do speak for themself.[00:21:31] <jbk> alanc: do you by any chance have time for an X question? :)[00:21:31] <e^ipi> or incompetant[00:21:32] <eviljames> e^ipi: Maybe lying is the wrong word. Embellishing :D[00:21:36] <ry_laptop> there are of course exceptions.[00:21:48] <oninoshik1> ry_laptop: he/she/it is worth a programmer's salery.[00:21:56] <ry_laptop> *nod* yeah, but that's an exception.[00:22:05] <ry_laptop> some people are born with 2 sets of genitals.[00:22:05] <alanc> jbk: sure[00:22:09] <ry_laptop> you can't make rules based on that.[00:22:41] <e^ipi> ry_laptop: gender and sex are two distinct concepts[00:22:45] <jbk> do you happen to remember how to configure xrandr on s10 (presenter is running s10 at our local ug meeting) -- trying to get external display port working :)[00:22:54] <ry_laptop> I'm just speaking of gender! :)[00:23:07] <e^ipi> gender isn't what hangs between your legs[00:23:09] <e^ipi> sex is[00:23:15] <e^ipi> gender is a self-definition[00:23:18] <ry_laptop> how you use it? :)[00:23:31] <e^ipi> heh, yeah... that's an apt description as well[00:23:33] <eviljames> Point of order: Sex isn't what hangs between your legs, genitals are.[00:23:34] <jbk> i'll buy you a beer next time i'm in norcal :)[00:23:45] <trichobezoar> jbk: this is the extent that I use xrandr - xrandr --output DVI-1 --right-of DVI-0[00:24:06] <eviljames> trichobezoar: That's one of my start up scripts for X :D[00:24:20] <ry_laptop> it would be awesome to get a macbook, a 30 inch cinema display, and an Iphone.[00:24:22] <oninoshik1> and you think you started something... i apperently get the converstaion really weird -_-[00:24:30] <ry_laptop> to nothing but programmin from now on.[00:24:33] *** awkfu has left #opensolaris[00:24:34] <ry_laptop> and never have to deal with hardware.[00:24:41] <eviljames> oninoshik1: hahaha, yeah this went off on a bizarre tangent, eh?[00:24:43] <e^ipi> ry_laptop: to extend your sex ?[00:24:48] <ry_laptop> It may[00:24:49] <e^ipi> like an expensive car[00:25:00] *** ericjray has joined #opensolaris[00:25:01] <ry_laptop> "hey baby, I've got a 30" montor at home, it does 2560x1600"[00:25:05] <e^ipi> but the nerd equivalent to it[00:25:15] <ry_laptop> 'oh baby'[00:25:23] <ry_laptop> "and I have an iphone biatch!"[00:25:30] *** ceri has left #opensolaris[00:25:31] <oninoshik1> hrm... ive seen mac people have to deal with hardware... it's not pretty...[00:25:51] <ry_laptop> Yeah, I was just fantasizin about it for a second, in the light apple gives it's hardware.[00:25:59] <ry_laptop> "it just works"[00:26:10] <ry_laptop> but in my fantasy, it plays all games too.[00:26:15] <alanc> jbk: not sure what you mean - if you are using Xorg on S10, you should have randr support and just need to run xrandr on the command line or the GNOME display resolution tool[00:26:19] <oninoshik1> (except when it doesn't)[00:26:25] <e^ipi> it just works... until it doesn't... then you buy a new one[00:26:26] <ry_laptop> Not even close[00:26:29] <ry_laptop> heh[00:26:32] *** boyd_ has joined #opensolaris[00:26:43] *** kate21de1 has joined #opensolaris[00:26:48] <ry_laptop> I just can't wait until virtualbox supports opengl in solaris guests.[00:26:54] <alanc> or if you're on nvidia chipset, /usr/bin/nvidia-settings[00:27:00] <ry_laptop> I'll just migrate 99% of my work into a virtual machine.[00:27:08] <ry_laptop> usin zarafa for my mail.[00:27:16] <ry_laptop> I won't need much else.[00:27:53] <Operator> hi , can some people test free opensolaris wine online game patch ? http://www.l2wolf.com/en/linux_download.jsp , i donno if its working or not let me know my mail is : admin at l2wolf dot com[00:28:11] <Operator> is it also working on OpenSolaris too ?[00:28:30] *** nachox has joined #opensolaris[00:29:23] <Dominic> didn't you just get kicked from #solaris for spamming your crappy game?[00:29:50] <Operator> no , i want to learn its working on OpenSolaris or not[00:29:56] <Dominic> then download it and try[00:29:57] <Operator> i didnt test it yet[00:30:04] <Operator> i dont have OpenSolaris yet[00:30:13] <trichobezoar> Heh... It's like trying to spam ADA users... Not enough of a market[00:30:23] <Dominic> heh[00:30:28] <Operator> market ? lol[00:30:30] <crichardso> does anyone use opensolaris with esx or vmware servcer in the sense of a san?[00:30:40] <Operator> private server is a market place ?[00:30:50] <LeftyBSD> crichardso: we did until recently[00:30:57] <LeftyBSD> we had too many performance problems[00:31:54] <crichardso> leftybsd what kind of performance issues? i am seeing it in both vmware server and esx[00:32:14] <crichardso> i am seeing high iowaits[00:32:21] <c00p> I don't thnk zfs like it ...[00:32:25] <LeftyBSD> we were using iscsi with zfs volumes to export to vmware, and we had unacceptably low performance[00:32:28] <crichardso> i have tried iscsi and nfs[00:32:37] <trichobezoar> zfs volumes == fail[00:32:54] <LeftyBSD> we discovered that it was mostly to do with the fact that shareiscsi is single threaded and broken[00:32:54] <trichobezoar> until the 'disable zil per dataset' is here[00:32:58] <c00p> lol - they work well on my x4500 here - they are not 100% production tho[00:33:20] *** capaz has quit IRC[00:33:21] <c00p> more for backups - used with time machines for some machines and some windows crap servers for our windows admn ...[00:33:54] <crichardso> i was kinda thinking that might be the issue to[00:33:56] <c00p> yeh I am only export 3 ISCSI targets ... so it doesen't scale it seems ...[00:34:13] <LeftyBSD> the solution from Sun was "use comstar"[00:34:19] <LeftyBSD> which wasn't ready yet[00:34:23] <LeftyBSD> it may not even be ready now[00:34:58] <trichobezoar> did you try with !zvol?[00:35:07] <trichobezoar> as a test...[00:35:13] <crichardso> but i never see high load and if i mount a nfs volume it seemed to copy and use full bw but when hosting vms or datasets it seems to act like 3m/s compared to 60[00:35:32] <LeftyBSD> I don't think so, trichobezoar[00:35:42] <crichardso> isnt that recomened against[00:35:44] <trichobezoar> yes. if you disable the zil, it will fix that. but until you can disable it only on your zvol sets, it's not recommended[00:35:57] <crichardso> ah[00:36:15] <crichardso> i will try it since it is just a dev box with nothing on it yet[00:36:26] *** Operator has left #opensolaris[00:36:45] <crichardso> do u really think comstar would fix that anyways[00:36:52] <LeftyBSD> the guy who was handling that particular machine was in a real rush to change to CentOS anyway[00:37:04] *** dunc has quit IRC[00:37:10] <LeftyBSD> why, I will never understand...[00:37:20] <trichobezoar> have you had problems with centos as it was set up?[00:37:39] <LeftyBSD> yes, but for different reasons[00:37:55] *** boyd has quit IRC[00:38:08] *** kim0 has quit IRC[00:39:10] <LeftyBSD> our storage solutions are completely broken right now, and we can't put any more primary storage in the ESX boxes[00:39:10] *** bahumbug has quit IRC[00:39:16] <LeftyBSD> it's infuriating[00:39:51] <trichobezoar> crichardso: no[00:40:07] <trichobezoar> But you asked what I think, not what I know.[00:41:08] <scoffin> is there some way to stop the updater from trying to create a new boot env each day, even when there is no "real" updates?[00:41:31] <trichobezoar> stop clicking upgrade?[00:42:19] <c00p> lol[00:42:44] <c00p> is the disabiling of the zil on zvol's promised ?[00:42:45] <oninoshik1> LeftyBSD: Im testing out COMSTAR's iscsi target, and having some issues... just sent an email to the list about it[00:42:55] <trichobezoar> c00p: I think there was a bug report submitted for it[00:43:04] <c00p> ok[00:43:13] <trichobezoar> Worth checking out ;P[00:43:21] <oninoshik1> OTOH: the COMSTAR's Fibre-Channel ran like to to[[00:43:23] <oninoshik1> top*[00:43:47] <trichobezoar> so it helped?[00:43:53] *** erast has joined #opensolaris[00:44:03] *** kate21de has quit IRC[00:45:41] <oninoshik1> trichobezoar: was that directed at me?[00:45:45] <trichobezoar> yes[00:46:18] <oninoshik1> oh.. it seemed to. I dont think it's a valid test though, because it also removes the overhead of the TCP/IP stack[00:48:08] <oninoshik1> I am really excited at the prospect of doingan apples-to-apples test for iSCSI. then see what i can really do with some of the caching that didn't seem to have much (read any) affect for iscsitgtd[00:49:43] <crichardso> how hard is comstar to manage[00:49:51] <crichardso> i briefly looked at it[00:50:06] <crichardso> one of my admin counter parts is a tad on the lazy side[00:50:12] <oninoshik1> based on my experence with the FC target. it's not hard, just different[00:51:50] <oninoshik1> I think i like it a bit better, really. because i can change the backing store for any LUN just by changing the view. that way it doesnt change the IQN... additionally i can actually set the IQN withing my own domain-name-space, resulting in much less human error in typeing iqns[00:52:02] <oninoshik1> (that was always our complaint)[00:52:13] *** bigjocker has quit IRC[00:52:21] <oninoshik1> well that an performance issues[00:52:42] <oninoshik1> anywho, it's late enough, im going home[00:52:43] <crichardso> ya i would like something more readable lol[00:52:54] <crichardso> alright thanks guys for your input[00:53:25] *** syamajala has quit IRC[00:53:30] <oninoshik1> crichardso: take a look at my post... it lists how im planning on doing them... it's on storage-discuss[00:55:45] *** stevel has quit IRC[00:58:12] <crichardso> ok thanks mate[00:58:51] <crichardso> just for the record zil on avg 1m/s with zil off 30-40m/s for iscsi[00:59:06] <crichardso> this is on 108[01:00:12] *** scoffin has quit IRC[01:08:10] *** micols has joined #opensolaris[01:08:16] *** micols has quit IRC[01:09:40] *** micols has joined #opensolaris[01:11:11] *** ericjray has quit IRC[01:11:41] *** Chipdancer has quit IRC[01:13:12] *** Chipdancer has joined #opensolaris[01:18:19] <Topdeck> Hi all, how do I create a user that I never want to be able to logon?[01:19:19] *** Chipdancer has quit IRC[01:19:25] <trichobezoar> lock the account[01:20:03] <Topdeck> can I use `useradd username -s /bin/false`?[01:20:30] *** Odin- has quit IRC[01:20:32] <Topdeck> yay for asking before trying :/[01:21:42] <Topdeck> and yes, that works (except the username goes at the end)[01:21:42] <nachox> Topdeck, you want your user to be able to run cron jobs?[01:21:59] <Topdeck> No, I'm trying to get idmap working against AD[01:22:08] *** victori_ has joined #opensolaris[01:22:19] <Topdeck> I'd really like there to be no users on the OS box at all, but I can't find any way to do that[01:22:32] *** tosh has quit IRC[01:23:05] * bubbva is back (gone 03:23:49)[01:23:34] *** e-jat has joined #opensolaris[01:24:52] <nachox> Topdeck, is this opensolaris or SXCE?[01:25:21] <Topdeck> nachox: OpenSolaris[01:26:30] *** void79 has joined #opensolaris[01:26:34] <nachox> Topdeck, if you use the default configuration you need at least 2 active users, one of them a real user, root which is a role and a regular user that has that role assigned[01:26:48] <void79> good day all[01:27:28] *** SQlvpapir has quit IRC[01:28:05] <Topdeck> nachox: Sorry, I'm not sure what you mean. Are you talking about mapping root to administrator@domain?[01:28:07] *** SQlvpapir has joined #opensolaris[01:28:51] <nachox> no, you said you wanted to block all the users in the solaris, i pointed out that was not possiblw[01:29:23] <_Lewellyn> nachox: well, technically it is... but erm... yeah... ;)[01:29:50] <Topdeck> nachox: Ah I see what you mean[01:30:22] <Topdeck> Is it possible to have (essentially) no users on the solaris box but still use idmap against AD?[01:30:37] <nachox> yes[01:30:56] <nachox> idmap will assign ephemerals uids to windows users[01:31:02] <Topdeck> nachox: any write-ups from ppl who've done it?[01:31:15] *** OpenBuntu has joined #opensolaris[01:31:18] <piwi> hi, little bit offtopic: i'm planning to buy a x2200 with 2x amd 2376. sun try&buy lists it for ~1500eur. can anybody tell me why a 2356 (barcelona) is 3 times the price of a 2376 (shanghai)? amdcompare lists identical specs for them?![01:31:19] <_Lewellyn> and the ephemeral uids are documented :)[01:31:50] <oninoshiko> _Lewellyn: i was thinking that there is no reason that you couldnt lock down all accounts... well except for the obvious lack of ability to do basic maintance[01:32:02] <nachox> Topdeck, http://blogs.sun.com/nico/entry/dealing_with_windows_sids_in[01:32:15] <nachox> check that blog entry to know what idmap is all about[01:32:26] <_Lewellyn> oninoshiko: you can delete all accounts on a machine. including the root role, iirc. but you end up with a noisy doorstop ;)[01:32:39] <oninoshiko> _Lewellyn: exactly[01:33:16] <nachox> _Lewellyn, once i removed shadow and password from a linux box, it did boot... mostly :P[01:33:41] <_Lewellyn> nachox: you can still go single-user then. or could under kernel 2.0.x[01:33:45] <Topdeck> nachox: hmmm, interesting[01:33:48] *** [JT] has joined #opensolaris[01:33:54] <_Lewellyn> not that i've ever done it... on accident, even... ;)[01:34:42] <oninoshiko> even when you can't do single user mode... init=/bin/bash always worked[01:35:01] <oninoshiko> well useually[01:35:08] <Topdeck> But I still don't think I'll be able to acheive my dream of having no users on the solaris box and being able to `chown user@domain /path/to/file`[01:35:25] *** victori has quit IRC[01:35:28] <nachox> _Lewellyn, i booted it normally and it complained a lot but worked. forget about logging in though[01:35:34] <Topdeck> Not with ephemeral mapping anyway[01:36:02] *** syamajala has joined #opensolaris[01:36:07] <nachox> i dont think that can be done, no[01:36:33] <nachox> _Lewellyn, that linux was almost as useless as when it had the password and shadow files :P[01:36:53] <_Lewellyn> hee[01:37:14] <_Lewellyn> Topdeck: that's what ACLs are for? ;)[01:37:36] *** wdp_ has quit IRC[01:38:45] <Topdeck> _Lewellyn: Heh, yeah. But what I'm after is for a windows user to own a share (not just a dir under another share) on the solaris box before they ever login, and do it completely from solaris[01:41:36] <_Lewellyn> you can't have it owned by a group?[01:44:11] <Topdeck> _Lewellyn: it's their home drive shared via SMB & NFS[01:44:37] <_Lewellyn> :P[01:44:48] <_Lewellyn> this is where i'd say to use nis... ;)[01:47:17] <nachox> i think solaris' cifs server has share ACLs already, but i'm not sure whether what your asking for is possible or not[01:48:45] *** Cass has joined #opensolaris[01:49:18] *** plavcik has quit IRC[01:49:19] *** Rocket2DMn has quit IRC[01:50:16] *** bnitz1 has quit IRC[01:50:26] <nachox> Topdeck, check for autohome http://dlc.sun.com/osol/docs/content/SSMBAG/smbshares.html[01:50:59] *** Gman has quit IRC[01:51:33] *** SQlvpapir_ has joined #opensolaris[01:52:36] <Topdeck> nachox: hmmm, interesting, but what's the point of a transient share, why not just have it shared all the time?[01:53:21] *** plavcik has joined #opensolaris[01:54:07] *** SQlvpapir__ has joined #opensolaris[01:59:10] *** axisys has joined #opensolaris[01:59:11] *** pitrucci has joined #opensolaris[01:59:28] <lesterc> jmcp, c00p: :) (referring your comments on being a sysadmin)[01:59:48] <lesterc> man this channel can get chatty.[02:00:25] <lesterc> I was off for an hour and boom my chat window is filled with rants.[02:01:03] *** alfism has quit IRC[02:01:08] <e^ipi> it's better when the rants are random and make little sense[02:02:08] *** kate21de1 has quit IRC[02:03:49] <nachox> Topdeck, ok, i'd need to actually do it myself to see if what you want to do can be done, but i'd start integrating my solaris with AD for which you might need SFU in windows[02:04:51] *** _Lewellyn has quit IRC[02:05:09] <Topdeck> nachox: already done, the solaris box is a member of the doman and can auth against AD and windows connects to the shares fine[02:05:11] *** e1kg has quit IRC[02:07:15] <nachox> you're using something like SFU?[02:07:33] *** SQlvpapir has quit IRC[02:08:08] *** rmorse has quit IRC[02:08:16] <Topdeck> not really, I didn't need to make any changes to the Windows server at all to get to this stage[02:08:31] *** ry_laptop has quit IRC[02:08:44] *** Andrew1 has quit IRC[02:09:36] *** qiyong has joined #opensolaris[02:09:44] *** m0zzzy has quit IRC[02:10:14] *** homeboy_ has joined #opensolaris[02:10:18] *** fallenhitokiri has quit IRC[02:10:20] *** micols_ has joined #opensolaris[02:10:22] *** fallenhitokiri has joined #opensolaris[02:10:32] *** SQlvpapir__ has quit IRC[02:11:41] *** SQlvpapir_ has quit IRC[02:14:07] *** oxygen-addict has joined #opensolaris[02:14:27] *** Andrew1 has joined #opensolaris[02:15:53] *** micols__ has joined #opensolaris[02:15:55] *** estibi has joined #opensolaris[02:16:10] *** Kitty_ has joined #opensolaris[02:16:11] *** fkr_ has joined #opensolaris[02:16:15] *** \amethys1 has joined #opensolaris[02:16:34] *** micols_ has quit IRC[02:16:34] *** [JT] has quit IRC[02:16:34] *** estibi_ has quit IRC[02:16:34] *** NeZetiC has quit IRC[02:16:34] *** nitrile has quit IRC[02:16:34] *** oxygen-a1dict has quit IRC[02:16:34] *** djgregor has quit IRC[02:16:34] *** fkr has quit IRC[02:16:34] *** SplasPood has quit IRC[02:16:34] *** \amethyst has quit IRC[02:16:34] *** Kitty has quit IRC[02:16:35] *** mitrox has quit IRC[02:16:35] *** dahlblom has quit IRC[02:16:35] *** Saltsa has quit IRC[02:16:38] *** djgregor has joined #opensolaris[02:16:44] *** NeZetiC has joined #opensolaris[02:16:45] *** [JT] has joined #opensolaris[02:16:45] *** estibi_ has joined #opensolaris[02:16:45] *** nitrile has joined #opensolaris[02:16:45] *** oxygen-a1dict has joined #opensolaris[02:16:45] *** fkr has joined #opensolaris[02:16:45] *** SplasPood has joined #opensolaris[02:16:45] *** \amethyst has joined #opensolaris[02:16:45] *** Kitty has joined #opensolaris[02:16:45] *** mitrox has joined #opensolaris[02:16:45] *** Saltsa has joined #opensolaris[02:16:49] *** Kitty has quit IRC[02:16:49] *** oxygen-a1dict has quit IRC[02:16:49] *** Saltsa_ has joined #opensolaris[02:16:49] *** dahlblom has joined #opensolaris[02:16:50] *** mitrox has quit IRC[02:16:51] *** victori_ has quit IRC[02:16:53] *** estibi_ has quit IRC[02:16:56] *** mitrox has joined #opensolaris[02:16:59] *** xy|ox has joined #opensolaris[02:17:16] *** \amethyst has quit IRC[02:17:21] *** Saltsa has quit IRC[02:17:25] *** fkr has quit IRC[02:17:26] *** Saltsa_ is now known as Saltsa[02:18:32] *** nitrile_ has joined #opensolaris[02:18:40] *** janil__ has joined #opensolaris[02:19:12] *** _Lewellyn has joined #opensolaris[02:21:04] *** nitrile has quit IRC[02:21:10] *** janil_ has quit IRC[02:21:17] *** homeboy has quit IRC[02:21:30] *** micols has quit IRC[02:21:43] *** homeboy_ is now known as homeboy[02:22:25] *** Fullmoon_ has quit IRC[02:27:15] *** Cass has quit IRC[02:28:44] *** kohju is now known as KOHJU[02:29:04] *** Cass has joined #opensolaris[02:30:39] *** comay has quit IRC[02:34:35] *** Gman has joined #opensolaris[02:34:35] *** ChanServ sets mode: +o Gman[02:34:44] *** bubbva has quit IRC[02:35:39] *** alain10 has quit IRC[02:42:00] *** nachox has quit IRC[02:42:13] *** ShadowHntr has joined #opensolaris[02:48:18] *** OpenBuntu has quit IRC[02:48:19] *** cypromis has quit IRC[02:52:25] *** segy has quit IRC[02:57:02] *** e-jat has quit IRC[02:57:10] *** ericjray has joined #opensolaris[03:00:02] *** segy has joined #opensolaris[03:00:24] *** gm152 has quit IRC[03:01:30] *** gm152 has joined #opensolaris[03:04:40] *** void79 has quit IRC[03:05:27] *** Gnu_Raiz has quit IRC[03:05:54] *** timsf has quit IRC[03:06:49] *** m0zzzy has joined #opensolaris[03:07:49] *** e1kg has joined #OpenSolaris[03:09:26] *** piwi has quit IRC[03:09:40] *** Fullmoon has joined #opensolaris[03:11:02] *** jub has joined #opensolaris[03:15:26] *** syamajala has quit IRC[03:17:47] *** JoergB has quit IRC[03:20:10] *** prav33n has quit IRC[03:24:13] *** Gman has quit IRC[03:24:41] *** e1mer has joined #opensolaris[03:28:42] *** jerry101 has joined #opensolaris[03:29:05] <jerry101> hello...[03:29:45] <Topdeck> it's vewy vewy quiet[03:29:57] <oninoshiko> (im hunting wabbits)[03:30:15] *** jsoftw has left #opensolaris[03:30:36] <jerry101> can someone tell me.. if there are any free blog sites that are (such as blooger or wordpress) that are run by opensolaris community. trying to start a blog which also promotes my fav. system[03:30:52] <eviljames> I would contribute.[03:31:23] <eviljames> But I don't know if there are any run by the community[03:31:28] *** esok has left #opensolaris[03:31:36] <eviljames> there's blogs.sun.com, but I think that's sun internal.[03:31:48] <jerry101> i registered @ opensolaris.org... but does not look like you are have a blog page or someting like that[03:32:05] <jerry101> yeah.. blogs.sun i think is for developerrs..[03:32:16] *** jub has quit IRC[03:32:20] <jerry101> they should start it..[03:32:38] *** javatexan has joined #opensolaris[03:32:42] <jerry101> hell, i can have my own blog profile page at chapters.ca[03:32:51] *** jub has joined #opensolaris[03:33:22] <jmcp> blogs.sun.com is for any Sun employee[03:33:40] <jmcp> if you're part of one of the OpenSolaris projects or communities, you can ask to get your blog aggregated[03:33:52] <jmcp> otoh, there's planet.opensolaris.org[03:34:54] *** rv- has quit IRC[03:35:31] <jerry101> ah. yes, planet.opensoalris.org[03:36:07] <jerry101> seem to be simialr the sun employees and developers one... but interesting idea.[03:37:42] <eviljames> How did I not know of this?[03:37:47] * eviljames hangs his head in shame[03:38:08] * oninoshiko props it up with a stick[03:38:28] *** \etc\bin has joined #opensolaris[03:38:36] <jmcp> jerry101: search for "Glynn Foster" on the p.o.o page[03:38:38] <bda> There is also http://planetsolaris.org/ which contains many of the same feeds.[03:38:41] *** ry_laptop has joined #opensolaris[03:39:35] *** e1mer has quit IRC[03:40:03] <eviljames> jmcp: Are you kidding me? This site even has a fun name like "the poo page" and I didn't know if it?!?[03:41:09] <jerry101> it seems.. if you want to start a free blog... you have to be affiliated with M$ @live.spaces, or google @blogger or yahoo @360. you should be able to start a blog and have it affiliated with someting you are inerested it and hence help promote it a little..[03:42:10] *** \etc\bin is now known as e1mer[03:42:34] *** pitrucci has quit IRC[03:42:58] *** a13x212 has joined #opensolaris[03:45:08] <_Lewellyn> um. are there foreseen issues with booting sxce when the machine has a date in the past?[03:45:20] <jmcp> eviljames: nope, no kidding[03:45:27] <jmcp> eviljames: goes well with boo and doo :-)[03:45:27] *** Topdeck_ has joined #opensolaris[03:45:40] *** javatexan has left #opensolaris[03:45:43] <eviljames> jmcp: I am saddened by this turn of events.[03:45:44] *** joanie has joined #opensolaris[03:46:14] <oninoshiko> what, that we are the only OS with scary fecial matter?[03:47:15] <Plazma> uhh[03:47:19] <Plazma> what[03:47:51] <tamr> oninoshiko: honestly, my favorite is "Unix Tip: Using SCAT (Solaris CAT) for analyzing crash dumps"[03:48:03] <jmcp> ah, SolarisCAT[03:48:10] <jmcp> I enjoyed working on that project[03:48:19] <jmcp> would continue doing so if i had cycles[03:48:36] <oninoshiko> oww.... my side hurts[03:48:49] <jmcp> oninoshiko: stop lying on the knife[03:50:37] <eviljames> hahaha[03:52:03] *** KOHJU is now known as kohju[03:52:15] *** Cass has quit IRC[03:52:19] <a13x212> i just added a repository using pkg authority and did a pkg refresh but when I do a search I only seem to be looking through the repository[03:52:32] <oninoshiko> tamr: i guess ill have to remember that tool... (somehow i dont think it will be hard)[03:53:19] <tamr> I don't know how that acronym got the greenlight, seriously :)[03:53:55] *** Topdeck has quit IRC[03:54:39] <oninoshiko> oh that's easy... it was probibly listed as "Solaris CAT" and everyone who got it, thought it was hilarious.[03:55:07] <oninoshiko> and everyone who was too prudish, didn't get it[03:55:17] *** ken has joined #opensolaris[03:57:05] <Plazma> ah skeet skeet skeet[03:58:00] <eviljames> Plazma: k, I actually asked someone on reddit once, and I've seen that before. wtf does it mean?![03:58:48] <Plazma> watch chapelles show[03:58:50] <Plazma> and learn my son[03:59:05] *** chrisr has quit IRC[04:00:22] <jmcp> oninoshiko: we used to call it "fm"[04:00:28] *** jerry101 has left #opensolaris[04:00:30] <jmcp> for .....ng magic[04:02:49] <jbk> hey jmcp[04:02:51] *** ry_laptop has quit IRC[04:03:08] <jmcp> hi jbk[04:03:53] *** xRaich[o]2x has joined #opensolaris[04:12:11] *** LouisJB has quit IRC[04:13:31] *** _Lewellyn has quit IRC[04:15:54] *** alanc has quit IRC[04:29:45] *** kohju is now known as KOHJU[04:34:01] *** alanc has joined #opensolaris[04:34:01] *** ChanServ sets mode: +o alanc[04:34:43] *** alanc has quit IRC[04:35:28] *** KOHJU is now known as kohju[04:35:54] * samc tries to learn Cisco PIX NAT :s[04:37:36] *** alanc has joined #opensolaris[04:37:36] *** ChanServ sets mode: +o alanc[04:39:18] *** tamr has quit IRC[04:40:30] <oninoshiko> ill trade you[04:41:02] *** ShadowHntr has quit IRC[04:41:18] *** a13x212 has quit IRC[04:42:22] *** ShadowHntr has joined #Opensolaris[04:46:06] * svcadm sympathizes with samc[04:48:22] *** ken has quit IRC[04:51:07] *** tamr has joined #opensolaris[04:57:16] *** m3t4l has joined #opensolaris[04:58:17] *** ddosia has joined #opensolaris[04:58:29] *** m3t4l has left #opensolaris[05:00:58] *** cwebber has joined #opensolaris[05:02:16] *** LordIllpalazo has joined #opensolaris[05:06:24] *** javatexan has joined #opensolaris[05:06:43] *** gm152 has quit IRC[05:06:45] *** javatexan has left #opensolaris[05:08:23] *** xy|ox has quit IRC[05:12:31] *** \amethys1 has quit IRC[05:12:39] *** \amethyst has joined #opensolaris[05:15:50] *** swa has quit IRC[05:29:10] *** ken has joined #opensolaris[05:29:20] *** ken is now known as ken-work[05:29:31] *** RULER_ has joined #opensolaris[05:30:14] *** javatexan has joined #opensolaris[05:31:41] *** javatexan has left #opensolaris[05:38:29] *** Topdeck_ has quit IRC[05:40:14] *** Catsceo has quit IRC[05:42:16] *** Topdeck has joined #opensolaris[05:58:16] *** ken has joined #opensolaris[05:58:52] <CIA-40> Yong-Feng Du <Yongfeng.Du at Sun dot COM>: 6814757 Using stmsboot -e to enable mpxio may destroy fp.conf file[06:05:06] *** victori has joined #opensolaris[06:08:07] *** ken-work has quit IRC[06:09:05] *** tamr has quit IRC[06:18:26] *** RULER_ is now known as Catsceo[06:21:52] *** ShadowHntr has quit IRC[06:21:52] *** ddosia has quit IRC[06:24:26] *** Roums has joined #Opensolaris[06:24:35] *** kohju is now known as KOHJU[06:26:53] *** Roms has quit IRC[06:26:55] *** Topdeck_ has joined #opensolaris[06:30:43] <jkimball4> this is a pretty good SNL[06:32:08] *** jub has quit IRC[06:32:21] *** jub has joined #opensolaris[06:33:36] *** ball has joined #opensolaris[06:33:58] *** Chipdancer has joined #opensolaris[06:34:50] *** Topdeck_ has quit IRC[06:35:28] *** Chipdancer has quit IRC[06:35:55] *** lesterc_ has joined #opensolaris[06:37:02] *** Topdeck has quit IRC[06:37:39] *** Topdeck has joined #opensolaris[06:39:21] *** gnut has joined #opensolaris[06:40:21] *** Chipdancer has joined #opensolaris[06:41:30] <gnut> hi[06:41:43] <gnut> has anyone tried installing openoffice 3?[06:42:49] <asyd> well, ooo 3 is available from ips[06:43:18] <gnut> oh[06:43:25] <gnut> how about if you're using sxce[06:43:55] <asyd> download packages?[06:44:24] <gnut> well.. the issue I have is I run setup[06:44:32] <gnut> and the next button is grayed out[06:44:37] <gnut> so I can't click next[06:45:08] <jmcp> gnut: just wait[06:45:13] <jmcp> it takes some time to trundle through[06:45:39] <gnut> jmcp: oh... how long usually?[06:45:47] <jmcp> I don't recall[06:45:52] *** cwebber has quit IRC[06:46:02] <jmcp> I went and did other stuff for several minutes, when I came back the button was ready to press[06:46:18] <gnut> okay. I'll try that. thanks. I thought I waited a while, but I'll wait longer :)[06:46:25] <gnut> ah...[06:46:48] *** colyte has quit IRC[06:48:45] <ball> I could build myself an OpenSolaris candidate machine for US$ 800\[06:48:52] <ball> ...with an upgrade path to boot.[06:51:05] <xRaich[o]2x> ball: so you are finally considering buying new hardware? ;)[06:51:23] <e^ipi> ball: easy.[06:52:17] <ball> xRaich[o]2x: I'm always /considering/ buying new hardware... just never buying new hardware for myself.[06:52:29] *** radsy has quit IRC[06:52:35] <xRaich[o]2x> ^^[06:52:41] <ball> e^ipi: Celeron 420, 4 Gbytes RAM, a pair of 120 Gbyte SATA drives.[06:52:50] <e^ipi> yep[06:52:54] <e^ipi> that works[06:52:56] <ball> upgradeable to quad cores, 8 Gbytes RAM[06:53:07] <ball> more disk drives.[06:53:10] *** lesterc has quit IRC[06:56:11] *** kate21de has joined #opensolaris[06:56:29] *** Topdeck has quit IRC[06:57:40] <ball> One Xeon box could probably replace four of my old Pentium III junkers.[06:58:11] <ball> That's an extra US$ 250 though.[06:58:44] <CIA-40> Stuart Maybee <Stuart.Maybee at Sun dot COM>: 6787135 Solaris kernel panic: pfn_to_mfn(): illegal PFN 0x8063ef0, 6730876 System crashing while running snv_93 on Celestica AMD64 (32bit)[06:58:56] <e^ipi> a cheap dual core core2 chip of some description could replace a dozen of your P3's[07:00:50] *** taltamir has joined #opensolaris[07:00:55] <ball> Core 2 Duo doesn't have VT though[07:01:56] <e^ipi> yes it does[07:02:28] <ball> The ones offered with this server seem not to.[07:02:33] <ball> E4600[07:02:58] *** gnut has quit IRC[07:03:03] <e^ipi> "server" ? you're putting together a PC[07:03:05] *** lesterc has joined #opensolaris[07:03:48] <jzu_> hmm, yeah gonna install Fire V490 with OpenSolaris[07:04:02] *** oxygen-a1dict has joined #opensolaris[07:04:03] *** freakazoid0223 has quit IRC[07:07:12] *** lesterc has quit IRC[07:07:28] *** lesterc__ has joined #opensolaris[07:08:59] <codestr0m> e^ipi: you mentioned you'd possibly be interested to mentor a gsoc student.. did you have a specific area in mind?[07:09:14] <e^ipi> nope[07:09:40] <e^ipi> usually students submit ideas, all the mentors go through them and pick ones they want to mentor[07:10:10] <codestr0m> e^ipi: we're doing it a bit backwards.. we're making ideas and trying to find students who are qualified and interested[07:10:28] <e^ipi> seems like a bad idea to me TBH[07:10:48] <codestr0m> it's working so far.. we have 6-10 mentors and 2-3 students interested[07:10:50] <e^ipi> if the students can't even come up with ideas, what's the likelihood they'll be able to complete them[07:10:53] <e^ipi> ?[07:11:01] <e^ipi> or won't lose interest halfway through, etr[07:11:03] <e^ipi> *etc[07:11:20] <codestr0m> and the likelihood is based on their qualifications of course.. a couple are really good c coders[07:11:37] *** [JT] has quit IRC[07:11:52] <codestr0m> well. the ideas are exciting so I think the risk of dropping out is the same as if it was their original idea[07:12:14] <e^ipi> *shrug* fair enough i guess[07:13:07] <codestr0m> e^ipi: just wanted to ping you.. if we're accepted maybe you can help out.. on the solaris specific side I have mostly zfs related ideas, but other things are welcomed[07:13:13] <Doc> http://www.youtube.com/watch?v=5RpDNTbSsI0[07:13:31] <e^ipi> yeah, let me know[07:13:38] <Doc> that's pretty cool - anination of the plane going into the hudson[07:16:41] *** oxygen-addict has quit IRC[07:17:16] <ball> e^ipi: it's a PC server, entry level from HP[07:17:43] <e^ipi> right, so whatever... buy what's cheap beacuse that's what you're getting anyways[07:17:50] <e^ipi> bucket of parts, ncix.com[07:18:07] <ball> three year support contract though, and they're whisper quiet[07:18:59] <ball> ...and I don't have to build it, though I don't mind that much.[07:19:11] *** LordIllpalazo has quit IRC[07:19:13] <ball> We bought a similar machine for a site about 130 miles S. of here.[07:19:34] <ball> ...though that one has a Xeon and hardware RAID[07:19:58] *** xRaich[o]2x has quit IRC[07:20:23] <e^ipi> well, you don't want hardware raid anyways[07:21:12] <ball> I would prefer not to have it for my own box.[07:21:21] *** _Lewellyn has joined #opensolaris[07:21:35] <ball> ...rather wish we hadn't bought it for the other machine too.[07:21:44] <ball> Oh well, live and learn I suppose.[07:22:11] *** lesterc_ has quit IRC[07:22:33] <_Lewellyn> ok. home. yay.[07:22:52] <ball> I wonder if that hardware RAID controller would work with OpenSolaris[07:23:01] * _Lewellyn has decided his two big itches to scratch, if he can figure out how to help with those projects.[07:23:29] <e^ipi> ball: maybe... why would you want to ?[07:23:31] <e^ipi> you have ZFS[07:24:11] <ball> e^ipi: I've not tried that, so I'm not sure how it works.[07:24:13] * _Lewellyn wonders when hardware-assisted zfs is coming out.[07:24:20] <_Lewellyn> e^ipi: well.[07:24:22] <_Lewellyn> erm.[07:24:24] <telexicon> a hardware raid controller can assist zfs[07:24:25] <_Lewellyn> ball: well.[07:24:32] <ball> ...but this ML110 would be a sensible box for me to start tinkering on OpenSolaris[07:24:36] <telexicon> some of the nice ones have like 256MB of battery backed cache[07:25:03] <_Lewellyn> telexicon: i am aware of this. i mean controllers specifically designed to assist zfs, however.[07:25:16] <telexicon> oh right[07:25:22] <telexicon> assist how[07:25:35] <telexicon> couldnt do checksums, then they wouldnt protect against hardware failure, driver mess ups[07:25:36] <e^ipi> yes, assist how?[07:26:05] <ball> It could periodically bring ZFS pizza, soda etc.[07:26:11] <ball> ...just to make sure it's on top form.[07:26:13] <_Lewellyn> well, only supporting JBOD, having a half gig or more of backed cache, a built in flash device...[07:26:15] <telexicon> good idea[07:27:24] <_Lewellyn> it probably wouldn't be terribly expensive to make such a controller. but people would probably pay almost as much as they would for raid[07:27:40] <jzu_> is there different ISO image for SPARC and X86 OpenSolaris?[07:28:14] <e^ipi> yes[07:28:16] <_Lewellyn> and you don't need to have usb thumb drive sticking out of your server, if you want maximum performance :)[07:28:17] *** lesterc__ has quit IRC[07:28:24] <e^ipi> _Lewellyn: thumb drive?[07:28:36] <_Lewellyn> next question: is sxce or osol recommended for sparc?[07:28:54] <jzu_> e^ipi: heh so I downloaded WROOOONG ISO-image :-)[07:28:59] <_Lewellyn> e^ipi: i've heard of putting stuff on a flash drive of some sort to speed it up.[07:29:03] <e^ipi> thumb drives are crazy slow, you'd be better off not spinning off slog[07:29:09] *** ken has quit IRC[07:29:15] <e^ipi> a flash drive, yes... that gives a perf. boost[07:29:17] *** nitrile_ has quit IRC[07:29:19] <e^ipi> thumb drive... no[07:29:33] <_Lewellyn> not all thumb drives are slow. the VAST majority, however...[07:29:35] <telexicon> _Lewellyn, like, a sata SSD disk[07:29:41] <telexicon> not a thumb drive[07:29:51] <_Lewellyn> telexicon: or a flash device embedded on a controller.[07:30:16] <e^ipi> or some DRAM chips on a PCI bus[07:30:22] <_Lewellyn> e^ipi: same thing :)[07:30:44] <_Lewellyn> if it's battery-backed, at least, it's the same thing, i suppose.[07:31:02] <e^ipi> http://www.ncix.com/products/index.php?sku=16611&vpn=gc-ramdisk&manufacture=gigabyte[07:31:06] <e^ipi> *shrug* one of those guys[07:31:10] *** codestr0m is now known as codestr0m_afk[07:31:44] <ball> brb.[07:32:50] *** gerard13 has joined #opensolaris[07:43:39] *** mib_knsghv has joined #opensolaris[07:46:06] *** _setuid_H has joined #opensolaris[07:46:22] *** estibi has quit IRC[07:46:26] *** jkimball5 has joined #opensolaris[07:46:31] *** jkimball4 has quit IRC[07:46:57] <_setuid_H> Morning all[07:47:06] *** devians has joined #opensolaris[07:47:23] <_Lewellyn> evening[07:47:43] *** jkimball5 is now known as jkimball4[07:47:43] <_setuid_H> _setuid_H: :-)[07:47:56] <_setuid_H> _Lewellyn: :-)[07:48:12] <devians> rtorrent seems to need libcurl v4, how would i install this so that rtorrent can see it?[07:48:31] *** sophokles has joined #opensolaris[07:48:44] <_Lewellyn> devians: if you install rtorrent from SFE, that gets dealt with automatically.[07:48:59] <_setuid_H> devians: try to add it to the LD_LIBRARY_PATH[07:49:18] <_Lewellyn> shouldn't need to touch LD_LIBRARY_PATH[07:49:42] <_Lewellyn> devians: would you like a package for it? :P[07:50:10] <_setuid_H> _Lewellyn: well than you need to build it with the same prefix as the sfe packages[07:50:15] <_setuid_H> or whatever[07:50:33] <_setuid_H> I don't know which authority is your rtorrent[07:50:43] <_setuid_H> or what repository[07:50:48] <_Lewellyn> _setuid_H: i have packages i put online for someone else a week ago[07:51:07] <devians> SFE? im not familiar with this sorry :) a package would be nice[07:51:18] <_setuid_H> _Lewellyn: ok than just build it with the same prefix and maybe --enable-shared[07:51:28] <devians> im using the rtorrent packages from whacked.net[07:51:44] <_Lewellyn> SFE is Spec Files Extra. it's like a "contrib" thing where you can build your own packages from source.[07:51:50] <_Lewellyn> google sfe-bootstrap[07:51:51] * _setuid_H builds most of the things by himself[07:52:12] <_Lewellyn> if you need a package, grab both of these: http://looking-glass/Solaris/PKGs/SFErtorrent.pkg.bz2[07:52:25] <_Lewellyn> oh bah. stupid browsers stripping domain on the lan[07:52:33] <_Lewellyn> http://looking-glass.greenviolet.net/Solaris/PKGs/SFErtorrent.pkg.bz2[07:52:57] <_Lewellyn> and http://looking-glass.greenviolet.net/Solaris/PKGs/SFExmlrpc-c.pkg.bz2[07:53:14] <_setuid_H> _Lewellyn: from the url this is a page of some enthusiast of looking glass project?[07:53:34] <_Lewellyn> _setuid_H: all my machines are names after things related to lewis carroll[07:53:50] <_setuid_H> _Lewellyn: ok[07:53:53] <_Lewellyn> another server is snark, i'm on cheshire-cat right now.[07:54:04] * ball grins, enigmatically[07:54:09] <_setuid_H> _Lewellyn: I used starwars naming[07:54:12] <_setuid_H> e.g. corusant taris ...[07:54:15] <_setuid_H> korribean[07:54:25] <_Lewellyn> ball: is that all we can see?[07:54:31] <jzu_> Heh, I'm unable to find OpenSolaris SPARC ISO :O[07:54:35] <jzu_> could someone point me to it?-)[07:54:42] <_setuid_H> jzu_: genunix.org[07:54:42] <devians> _Lewellyn, i already have the whacked package for rtorrent installed. so what you're saying is i need yet another package manager so i can get the libs it needs?[07:54:54] <_setuid_H> jzu_: or it is .com?[07:54:56] <_Lewellyn> i have no clue what whacked is[07:55:10] <devians> http://whacked.net/2008/05/08/updated-rtorrent-libtorrent-packages-for-solaris/[07:55:18] <jzu_> _setuid_H: it's org[07:55:27] <jzu_> _setuid_H: is Global ISO X86 & SPARC?[07:55:43] <_setuid_H> jzu_: no just automated installer[07:55:48] <_Lewellyn> jzu_: sxce is the only "official" opensolaris for sparc, right now[07:56:11] <_Lewellyn> devians: um. looking at that date, those are old packages[07:56:44] <jzu_> _Lewellyn: sxce?[07:56:56] <_Lewellyn> solaris express community edition[07:57:12] <_Lewellyn> there's a link to the dvd on opensolaris.org/os/downloads[07:57:24] <_setuid_H> jzu_: http://opensolaris.org/os/downloads/sol_ex_dvd_1/[07:57:26] *** ball has left #opensolaris[07:57:57] <jzu_> ah, thanks[07:58:11] *** colyte has joined #opensolaris[07:59:17] <_Lewellyn> devians: fwiw, my packages have been run on sxce and osol. and it's the current version of rtorrent.[08:00:15] <_Lewellyn> i guess it's the same version as what you found, but those look problematic, based on the comments[08:00:27] *** svcadm has quit IRC[08:00:50] <devians> fair enough. ill have a go at uninstalling them and ill look into your package :)[08:01:12] <devians> do i need to install anything sfe related to install the pkg you linked?[08:01:29] <_Lewellyn> nope. let me grab the requirements. one moment.[08:02:03] *** div110 has joined #opensolaris[08:02:30] <_Lewellyn> ok. SFErtorrent requires SUNWsigcpp, SFExmlrpc-c, SUNWcurl, SUNWncurses[08:02:53] <_Lewellyn> SFExmlrpc-c requires... nothing :)[08:03:08] <_Lewellyn> i would assume that you have those 3 SUNW packages[08:03:09] <devians> will pkgadd -d realise this and dl those? sorry im still new to how os works[08:03:30] <_Lewellyn> pkgadd will complain if you're missing dependencies, if that's the question.[08:03:49] <_Lewellyn> pkgadd is "dumb". that's one reason for pkg[08:04:13] <devians> mk, how do i use pkg install with a downloaded file?[08:04:46] <jzu_> could someone list pros and cons of Solaris 10 vs. OpenSolaris?=)[08:05:24] <_Lewellyn> i'm on sxce. i don't get pkg :)[08:05:41] <_Lewellyn> unbzip2 those files. then use pkgadd on them :)[08:05:52] <devians> kk[08:06:05] *** teknomega has quit IRC[08:06:24] <_Lewellyn> when sxce gets pkg, i'll know more about it ;)[08:06:50] <_Lewellyn> jzu_: pros? newer software, more features, better hardware support[08:08:07] <_Lewellyn> cons? under development and isn't as "stable" as the 4-years-since-release solaris 10, it can be slower for some people for some tasks (again, under development), if you follow the development versions you may encounter bugs that may be gone 2 weeks later :)[08:08:25] <_Lewellyn> oh, more zfs features in opensolaris :)[08:08:40] <_Lewellyn> and if you're using "opensolaris" vs sxce, you get a new package manager.[08:09:01] <_Lewellyn> another con... the dt stuff is going away. but you'll have to deal with that in solaris 11 anyhow.[08:09:19] <_Lewellyn> my /usr/openwin is shrinking constantly :([08:09:39] <devians> gah, package management in solaris makes my brain hurt. trying to figure out how to remove the 3 packages i installed earlier[08:09:41] <_setuid_H> jzu_: I can give you some steps how to create an ips image in sxce ...[08:09:56] <_setuid_H> jzu_: or even in solaris 10[08:13:36] <devians> ok got that sorted... now[08:13:42] *** phimic has joined #opensolaris[08:13:44] <_Lewellyn> devians: if they are svr4 packages, use pkgrm[08:14:04] <_Lewellyn> _setuid_H: you mean for me? i can do it. but i can't use it myself then ;)[08:16:22] *** stux|RC has joined #opensolaris[08:17:13] <devians> _Lewellyn, i figured that out its just a case of figuring out what they're named for pkgrm to remove them[08:17:44] <_Lewellyn> devians: pkginfo -l | grep blah[08:17:45] <_Lewellyn> :)[08:18:18] *** _coredump_ has quit IRC[08:18:58] *** ball has joined #opensolaris[08:19:02] <devians> yeah thats prettymuch what i ended up doing, just forgot that it was pkginfo not pkglist[08:19:12] *** Deesl has joined #opensolaris[08:19:58] <_Lewellyn> :)[08:20:24] <_Lewellyn> man pkgadd (or any other pkg* tool) and look at the list of related commands at the bottom :)[08:20:43] <_Lewellyn> also helpful: /usr/sbin/pkg*[08:20:51] * _Lewellyn hopes you're taking notes for the future ;)[08:21:19] <devians> yeahuh :)[08:21:25] *** _coredump_ has joined #opensolaris[08:21:43] *** ball has left #opensolaris[08:21:52] <_Lewellyn> we lost the ball[08:26:04] <Deesl> how do I remove the patch backout data stored? I am running short of space :-s[08:26:24] <devians> hrm, whats the command to extract a bz2[08:26:29] <_Lewellyn> devians: bunzip2[08:26:56] <devians> thanks[08:28:39] *** carl- has joined #opensolaris[08:31:27] <devians> xmrpc package went in fine, rtorrent had three dependencies. how to i get it to install these dependencies? can i install them with pkg or with blastwave and have it work?[08:31:51] <_Lewellyn> Deesl: usepkg[08:31:55] <_Lewellyn> use pkg[08:31:55] *** stux|RC-only has quit IRC[08:32:00] <_Lewellyn> i don't know the syntax, though[08:32:05] *** janil_ has joined #opensolaris[08:32:10] <_Lewellyn> SUNW packages are sun-provided[08:32:31] <Deesl> well I am on Solaris.. not opensolaris...[08:32:39] <Deesl> and there is nothing called pkg :-s[08:32:43] *** luc^ has joined #opensolaris[08:32:47] <_Lewellyn> i told you its dependencies in advance ;)[08:32:54] <_Lewellyn> Deesl: sorry.wrong completion[08:33:03] <Deesl> :)[08:33:06] <_Lewellyn> i've been d<tab>ing at devians[08:33:37] <_Lewellyn> Deesl: you may wish to wait for an answer in "the other channel" :)[08:33:55] <_setuid_H> Deesl: you should know that on solaris pkg is not included by default, but you can "somehow" get it to that, but it's not so safe[08:33:58] <Deesl> _Lewellyn: :)[08:34:01] <_Lewellyn> this one is specific to opensolaris, and i don't think anyone's needed patchadd yet:)[08:34:31] <_Lewellyn> _setuid_H: Deesl needs help with removing patch backout data[08:35:06] <_setuid_H> _Lewellyn: ok I just saw there is no pkg in solaris and .... reacted :-)[08:35:13] *** Auralis has joined #opensolaris[08:35:22] <_Lewellyn> _setuid_H: you overreact ;)[08:35:41] <_Lewellyn> are you willing to support everyone you try to get running with pkg outside of osol? ;)[08:36:01] <_Lewellyn> i'm willing to bet that not many IPS packages available will work right on S10[08:36:59] <_setuid_H> _Lewellyn: that is why I wrote unsafe[08:37:28] <_setuid_H> _Lewellyn: but anyway if your image won't be rooted in / :-) it won't take so much risk[08:37:43] *** __ms has quit IRC[08:37:44] <_Lewellyn> _setuid_H: for most people, the proper answer is really "you can't" :)[08:38:10] *** __ms has joined #opensolaris[08:38:31] *** mib_2itjfn has quit IRC[08:39:21] <_setuid_H> _Lewellyn: ok I just make some test on running it under sol 10 (due mirroring) ..., and also had to make a presentation with demos on my sxce[08:39:46] <_setuid_H> I found the package for solaris 10 and sxce unusable due not including mako[08:41:05] <_Lewellyn> there's good reasons for sxce to not get pkg yet ;)[08:43:29] *** ericjray has quit IRC[08:44:12] *** jamesd has quit IRC[08:44:15] *** janil__ has quit IRC[08:44:19] <devians> _Lewellyn, seems to be working :) now i just have to figure out its config voodoo :P[08:44:27] <devians> many thanks for your help[08:44:34] <_Lewellyn> yeah. rtorrent bugged me for that[08:44:50] *** Fullmoon has quit IRC[08:44:52] <devians> now all i need is to figure out whats broken in opensolaris that flexget wont write to dbm properly[08:44:54] <_Lewellyn> but transmission doesn't do dht and Qt's on a messed-up tracker (or it was)[08:45:03] <_Lewellyn> flexget?[08:45:15] <devians> very intelligent rss downloader, written in python[08:45:35] *** janil__ has joined #opensolaris[08:46:24] <devians> works great on everything but os, for some reason it has a hernia when trying to write to dbm to record what its already downloaded[08:46:38] <_Lewellyn> oh yeah. you linked to that bug the other day.[08:46:45] <_Lewellyn> i still suspect that the code's broken[08:48:24] <devians> quite possible[08:48:56] <_setuid_H> _Lewellyn: sxce will be dead soon so I'm not wondering why there is no pkg :-)[08:49:16] <devians> although im suspect on why it only fails on os[08:49:38] <_Lewellyn> _setuid_H: i'm kind of curious quite what they're going to do with sxce vs osol.[08:49:54] *** devians has quit IRC[08:50:12] <_Lewellyn> after all, SX is well-known as a "next version preview" of solaris... osol has a "linuxy toy os" sound to it[08:51:04] *** sophokles has quit IRC[08:52:53] *** sophokles has joined #opensolaris[08:53:04] *** lblume has joined #opensolaris[08:53:17] <_Lewellyn> oh no. run! :)[08:53:22] <_Lewellyn> hi, lblume[08:53:43] *** hsp has joined #opensolaris[08:53:48] *** joanie has left #opensolaris[08:53:55] *** kim0 has joined #opensolaris[08:54:15] *** niq has joined #opensolaris[08:55:31] <lblume> _Lewellyn: Hi to you too[08:55:35] <_Lewellyn> sh: usr/gnu/bin/date: not found[08:55:40] <_Lewellyn> oops[08:55:44] <_Lewellyn> 1236758238.216735215[08:55:54] <asyd> \_o<[08:56:00] <hrist> *quack*[08:56:09] <_Lewellyn> dude. who moved my shotgun?[08:57:00] <lblume> Here, use mine.[08:57:59] <lblume> As a reward for helping me count milliseconds.[08:58:22] <_Lewellyn> they let you keep guns, there? :D[08:58:40] *** janil_ has quit IRC[08:58:42] <lblume> SHHHHHHHHHH[08:58:46] <_Lewellyn> hee[09:00:18] <lblume> It's cunningly concealed inside a pack of Daliyuan Pies, nobody would think of trying to eat that.[09:02:45] <palowoda> Is anyone running coolstack, if so is it running the 64bit version of 5.1 mysql with x86?[09:03:00] *** ken has joined #opensolaris[09:05:03] <lblume> Isn't the coolstack version 32 bit only on x86? I seem to remember something like that from the readme?[09:05:44] <mib_knsghv> mmmh chinese milk products....[09:06:51] <palowoda> Maybe, I just wondered if I wasted my time building my own 64bit apache/php/mysql. All except I used the 5.1.8 client libs from the mysql site.[09:07:45] <palowoda> I was wondering what compiler they used. I ended up using gcc 4.3.[09:08:41] *** taltamir has quit IRC[09:10:30] *** cypromis has joined #opensolaris[09:10:45] *** Auralis_ has joined #opensolaris[09:10:49] <lblume> Coolstack uses GCC for MySQL on x86[09:10:59] <lblume> However, it builds fine using Studio 12[09:11:26] <lblume> I'm using it 32 bits, and I've successuflly tested it as 64 (but I was too lazy to rebuild the whole stack)[09:12:05] *** nitrile has joined #opensolaris[09:12:08] <lblume> I shun MySQL.org packages, they've not proven reliable in my experience[09:12:44] <_Lewellyn> palowoda: i only run coolstack on sparc, since sxce/osol include current versions of webstack stuff[09:13:14] <_Lewellyn> (my lone x86 box [this laptop] runs sxce)[09:13:26] <palowoda> Hmm, is there a test I can use to see this unrealiablity with the 5.1.8 version of mysql's x86 64bit port?[09:13:33] <_Lewellyn> for my needs, 5.0 is current enough :)[09:13:36] <flyingparchment> 5.1.8??[09:13:37] <lblume> mib_knsghv: Ever tried Inner Mongolian cheese snacks? And I'm told they live on it. *shudder*[09:13:48] <palowoda> Hang on.[09:13:51] <flyingparchment> is 5.1.8 even a release version, or some beta/alpha?[09:13:57] <flyingparchment> (current is 5.1.31...)[09:14:13] <palowoda> Your right I'm using 5.1.31.[09:14:24] <lblume> palowoda: It wasn't unreliable, just buggy scripts, and installs in/usr/local, of all places (their system must be lacking filesystem(4) )[09:14:41] <lblume> Also wasn't built with the options I needed.[09:14:50] <_Lewellyn> filesystem(5)?[09:15:30] <palowoda> I guess I didn't notice that bug because I install it in /usr/local.[09:15:33] *** perlmongo has joined #opensolaris[09:15:34] <lblume> Slippery keyboard - soy sauce :-P[09:16:27] <_Lewellyn> palowoda: you've not read filesystem(5) either then?[09:16:46] <palowoda> Nah I rarely read.[09:16:51] <lblume> It was an issue for me at that time, since it was conflicting with another older version I wanted to get rid of. Also, IIRC, the RUN_PATH wasn't great, and needed LD_LIBRARY_PATH or such hacks.[09:17:09] <lblume> Americans and reading....[09:17:09] <_Lewellyn> ugh. no -R?[09:17:15] <_Lewellyn> hey[09:17:21] *** Deesl has quit IRC[09:17:23] <_Lewellyn> i'm american. and i read lots[09:17:25] <flyingparchment> current versions definitely have -R[09:17:49] <flyingparchment> didn't notice old versions not having it either... but perhaps it was missing from mysql_config[09:18:24] <lblume> Hmmmm, I *think* it was messing up with my config - but again, a long time ago, I don't remember details. Just be cautious.[09:18:30] <palowoda> Heck I have it running and I'm satisfied with the speed of the compiled gcc4.3. But what a PITA to get it all compiled.[09:18:41] *** jollyd has joined #opensolaris[09:19:13] *** jollyd has left #opensolaris[09:19:18] *** DTEIT has joined #opensolaris[09:19:19] <palowoda> I just ran into some really strange problems with studio 12.[09:19:26] *** Auralis has quit IRC[09:19:26] <lblume> It's quite straightforward with Studio, though, at least, after reading the online docs for some flags.[09:19:29] <_Lewellyn> speaking of mysql... i still need to get php to work right against SUNWmysql5test[09:19:31] *** Deesl has joined #opensolaris[09:19:42] <palowoda> And I'm not willing to put up with Sun's gcc.[09:19:46] <DTEIT> morning[09:20:18] <lblume> palowoda: What kind of issue did you have? I've been running MySQL in small prod servers built with Studio for years, and it just works?[09:20:21] *** codestr0m_afk is now known as codestr0m[09:20:53] <palowoda> studio 64bit compiled apache with all the dressings.[09:21:03] *** Deesl has quit IRC[09:21:59] *** arcame has joined #opensolaris[09:22:24] *** ken1 has joined #opensolaris[09:22:45] <palowoda> I guess I was expecting too much, the latest apache, latest php, latest mysql and latest UofW imap and latest openssl 64bit.[09:23:35] <lblume> Ah, I've been using it 32, ot 64. I'd like to go 64 on my home server soon, at least for a try. I've built everything, but not activated yet.[09:23:53] <lblume> Only a short MySQL run, for a few days, and it seemed to work.[09:24:37] <palowoda> Well got it running, http://www.fiver.net/server-info just wondering what the coolstack issues are.[09:24:56] <_Lewellyn> palowoda: why do you need 64-bit mysql, btw?[09:25:15] <_Lewellyn> from what i understand, 64-bit actually takes a performance hit most of the time[09:25:27] <palowoda> Because I wanted everything 64bits.[09:25:28] <flyingparchment> _Lewellyn: mysql is one of the few things where 64-bit is often required - because it needs a lot of memory[09:25:38] <flyingparchment> (ours is using 28GB, which would be hard if it was a 32-bit build)[09:26:04] <_Lewellyn> flyingparchment: no. the coolstack docs explain that well. you only need 64-bit if you have tables over 4gb, iirc.[09:26:10] <flyingparchment> _Lewellyn: no, that isn't true[09:26:14] <palowoda> I want no modules 32bit and I don't want the compiler backend producing shitty code.[09:26:16] * _Lewellyn digs for that[09:26:22] <flyingparchment> _Lewellyn: well, it is, sort of. we _could_ run mysql in 32-bit using, say, 1GB, but it would be much slower[09:26:27] <_Lewellyn> maybe it's only true on sparc.[09:26:28] <flyingparchment> _Lewellyn: the 28GB is for the InnoDB buffer cache[09:26:39] <flyingparchment> _Lewellyn: it's a bit like the SGA on oracle[09:26:40] <palowoda> And I don't care about sparc[09:26:51] <flyingparchment> bigger cache = runs faster. it way offsets the lose of 64-bit slow down[09:26:58] <colyte> is there a place to request packages to be included on opensolaris.org?[09:27:10] <_Lewellyn> colyte: which packages?[09:27:22] <colyte> usually most software that does math benefits from 64bit.[09:27:30] *** KOHJU is now known as kohju[09:27:39] <colyte> _Lewellyn: whatever I desire.. ;)[09:27:48] <flyingparchment> dont think mysql does much maths ;)[09:28:02] <_Lewellyn> colyte: have you checked the software porting project?[09:28:03] <codestr0m> is the sol10 and SXCE installer almost the same? I'm making a comparison chart for some of the key differences.. other than sol10 is just rigidly tested and features/bugs are backported for longer support[09:28:09] <colyte> _Lewellyn: In my case these days, irssi, screen, nano, rtorrent[09:28:36] <_Lewellyn> irssi and screen are SUNW packages, at least[09:28:38] <_Lewellyn> dunno about nano[09:28:42] <_Lewellyn> and rtorrent is in SFE[09:28:54] <colyte> _Lewellyn: csw packages for irssi and screen, not sunw.[09:29:36] <colyte> _Lewellyn: nano csw as well. rtorrent doesnt compute from sfe on opensolaris. works for solaris but yeah. i was fortunate enough to get the old packages compiled for me.[09:31:01] <_Lewellyn> SUNWirssi[09:31:15] <_Lewellyn> SUNWscreen[09:31:18] <_Lewellyn> both are installed, here[09:31:28] <palowoda> codestr0m: How do you explain the testing of S10 vers SXCE?[09:31:29] *** ken has quit IRC[09:31:29] <flyingparchment> codestr0m: the SXCE installer is the same because no development is happening on it. (except a little bit to add basic zfs support)[09:31:36] <_Lewellyn> define "doesn't compute"...[09:31:54] <_Lewellyn> i have no problems with rtorrent from sfe[09:32:18] <codestr0m> palowoda: for sol10 I describe as "Rigidly tested and long term support contracts available" for SXCE "Best preview of what may become integrated in the next version of Solaris"[09:32:23] <_Lewellyn> flyingparchment: S10U6 supports zfs root, too ;)[09:32:38] <flyingparchment> _Lewellyn: yes, i know. afaik that's the only reason it was added to sxce - so it could go into S10[09:32:51] <_Lewellyn> codestr0m: the sxce installer is still the s10 installer, until there's something better available[09:32:52] <palowoda> Rigidly tested?[09:33:01] <palowoda> What the heck does that mean?[09:33:33] <_Lewellyn> palowoda: there's plenty of packages that haven't changed since before release. 4 years ago.[09:33:52] *** kohju is now known as KOHJU[09:33:54] *** KOHJU is now known as kohju[09:34:01] <lblume> palowoda: Not what a dictionary would tell you, for sure.[09:34:05] <palowoda> I guess it could mean wack it over the head with a board.[09:34:43] *** andy8 has joined #opensolaris[09:35:20] *** andy8 has quit IRC[09:35:22] <palowoda> Or maybe the tester had rigamortis.[09:35:44] *** andy0 has joined #opensolaris[09:35:54] *** andy0 is now known as Andrewski6667[09:36:01] <palowoda> Or maybe the tester had a hardon for the customer.[09:36:10] <Andrewski6667> i think i totally screwed myself[09:36:21] <Andrewski6667> by using the update facility in gnome[09:36:22] <lblume> I never thought of those explanations... They sound so plausible....[09:36:33] <_Lewellyn> oh neat. i just discovered that gnome itself can't access my cdrom drive for the first few discs after boot on 109. :([09:37:15] <Andrewski6667> does it error out?[09:37:25] <Andrewski6667> put anything on the console/[09:37:26] <Andrewski6667> ?[09:37:52] <_Lewellyn> nothing on console. dunno if the console's being redirected to a log somewhere.[09:38:02] <Andrewski6667> hm[09:38:19] <flyingparchment> _Lewellyn: cd-roms are an obsolete technology and support for them is being removed from solaris. the network is the computer![09:38:24] <_Lewellyn> but "Unable to scan CD-RW/DVD+-RW Drive for media changes" is the headline in the error box[09:38:34] <_Lewellyn> dbus access denied stuff[09:38:34] <Andrewski6667> well at least yours boots[09:38:35] *** ken1 is now known as ken-work[09:38:47] <_Lewellyn> takes forever, but yes, it boots[09:38:48] <Andrewski6667> i used the gnome updater (yeah, stupid i guess) and now it's broken[09:38:52] <palowoda> flyingparchment: I wish.[09:39:00] <Andrewski6667> anyway, i got tired after 2 hours[09:39:08] <Andrewski6667> rebooted it into text mode, worked fine[09:39:14] <_Lewellyn> something networky broke in 109, too. not sure what. but my startup errors for networking have changed[09:39:17] <Andrewski6667> except no gnome[09:39:25] <mib_knsghv> @lblume: sorry was afk. no, never tried inner mongolian cheese cake... or inner mongolian anything for that matter. They must be using very nuticious plastics, if they live on it :P[09:39:26] <_Lewellyn> Andrewski6667: file a bug :)[09:39:32] <palowoda> Too bad only DL DVD's where only allowed.[09:39:34] <_setuid_H> _Lewellyn: another problems with nwam?[09:39:44] <Andrewski6667> heh[09:39:47] <Andrewski6667> yeah, ok[09:39:50] <Andrewski6667> can't hurt i guess[09:39:52] <_setuid_H> _setuid_H: this is why I'm still using physical:default[09:40:05] <_Lewellyn> _setuid_H: i have no clue what's up. but the messages at boot have changed. i'll take pics when i file my bugs, don't worry.[09:40:08] <Andrewski6667> there are a few people out there who have potential workarounds[09:40:20] <_setuid_H> _Lewellyn: ok[09:40:24] <_Lewellyn> Andrewski6667: imo, it shouldn't break things.[09:40:31] <Andrewski6667> but my opensolaris box is at work and won't boot, otherwise i'd be all hooked up with x11[09:41:31] <Andrewski6667> i guess word on the street is, if you see the little upgrade notification, don't use it[09:41:54] <palowoda> Oh that, did it ever work.[09:42:07] <_setuid_H> well for those who are afraid of non-stability do not use dev authority[09:42:27] <Andrewski6667> instead do pkg img-update[09:42:37] <Andrewski6667> or pkg image-update rather[09:43:12] <_setuid_H> or install entire@version to which you want to udpate[09:43:16] <Andrewski6667> I think FreeBSD has it beat right now if only because it "works"[09:43:29] <Andrewski6667> what?[09:43:44] <_setuid_H> Andrew1: that was to the entire post?[09:44:12] <Andrewski6667> is there some document somewhere that explains the prefered way to update?[09:44:27] <colyte> _Lewellyn: you were right, they were there. I feel stupid for not using -r with pkg search, because that yielded me nothing.[09:44:28] <palowoda> Use the dev authority.[09:44:46] <_setuid_H> palowoda: i'm using it but time by time something does not work[09:44:47] <palowoda> Other than that buy the opensolaris support contrackt.[09:44:51] <palowoda> contract.[09:45:04] <_Lewellyn> colyte: i use sxce. if i have a SUNW package, it's packaged by sun ;)[09:45:05] <Andrewski6667> shoot[09:45:11] *** Fullmoon has joined #opensolaris[09:45:18] <Andrewski6667> i'll buy a contract right here if they can get my business booting again![09:45:20] <palowoda> high, aim low.[09:46:17] <palowoda> Don't worry nobody knows what you really get with a opensolaris support contract.[09:46:29] <Andrewski6667> i'd rather understand the problem and how to fix it, rather than paying a mentat at Sun to know all the answers, though[09:46:41] <Andrewski6667> heh[09:46:58] <palowoda> Well what problem do you have to begin with?[09:47:17] *** Madkiss has joined #opensolaris[09:47:18] <Madkiss> hi folks[09:47:28] <Andrewski6667> i installed opensolaris (have been running it in virtualbox for a while, wanted it on the metal)[09:47:33] <Andrewski6667> works fine[09:47:48] <Madkiss> just a quick question; when using svcadm to enable autostart of a service, where do I specify command-line parameters for that service? iow, where d I find the "bind9-init-script" in opensolaris?[09:47:52] <Andrewski6667> I installed some packages with the package manager, from Sun[09:48:01] <Andrewski6667> then I discovered blastwave and installed other stuff[09:48:05] <Andrewski6667> works fine[09:48:11] <palowoda> continue[09:48:14] <colyte> _Lewellyn: About requestion packages tho in general, where to go?[09:48:16] <Andrewski6667> heck, i even had windows up in a virtualbox, running really smoothly[09:48:25] <palowoda> and[09:48:34] <Andrewski6667> then I get this little notifier in my taskbar[09:48:42] <Andrewski6667> said "updates available"[09:48:46] <palowoda> there is your problem[09:48:47] <_Lewellyn> colyte: the software porting project[09:48:50] <Andrewski6667> so I clicked it and proceeded with the update[09:48:54] <Andrewski6667> yeah[09:48:59] <palowoda> big mistake.[09:49:04] <_Lewellyn> colyte: you also never elaborated on your rtorrent issue[09:49:13] <Andrewski6667> now it just sits on the blue OpenSolaris screen and the darn orange thing goes from left to right[09:49:22] <Andrewski6667> so, I can boot into the text mode environment[09:49:29] <palowoda> shot the button with a gun.[09:49:32] <Andrewski6667> but I can't even boot to the previous boot environment[09:49:32] *** pgr has joined #opensolaris[09:49:43] <Andrewski6667> the X11 one anyway[09:49:51] <Andrewski6667> text mode, as I said, works[09:49:57] <Andrewski6667> so I can get a terminal, not a problem[09:50:02] <_Lewellyn> Andrewski6667: reinstall SUNWgmone*?[09:50:09] <_Lewellyn> gnome even[09:50:15] <Andrewski6667> but, is it gnome?[09:50:15] <palowoda> But those are known bugs.[09:50:33] <Andrewski6667> it won't even launch gdm[09:50:43] <_Lewellyn> gdm == gnome display manager[09:50:47] <Andrewski6667> won't even change video modes and then fail, or anything[09:51:00] <Andrewski6667> so I think it's something not related to gnome[09:51:11] *** Fullmoon has quit IRC[09:51:27] *** div111 has joined #opensolaris[09:51:28] <codestr0m> can you get long term or only short term support contracts for SXCE?[09:51:35] <codestr0m> (or do they not offer support on it?)[09:51:49] <flyingparchment> what does long term and short term mean?[09:52:10] <palowoda> Platinum contracts are long term.[09:52:19] <palowoda> And they are cheap too.[09:52:47] <lblume> They still do contracts on SXCE? I thought it had been discnotinued in favor of OpenSolaris.com?[09:52:49] <Andrewski6667> I guess nobody else has upgraded with the button and recovered?[09:52:50] *** Deesl has joined #opensolaris[09:52:51] <_Lewellyn> hm. iirc, if you get a platinum contract on a solaris express release, it carries over onto final release. is that still the case?[09:52:53] <Andrewski6667> or had a problem[09:52:55] <Andrewski6667> that's somethin[09:53:10] <flyingparchment> _Lewellyn: afaik, one buys support for solaris in general, not a specific release[09:53:23] <Andrewski6667> I could spend a whole day getting everything set up again[09:53:23] <flyingparchment> it's attached to the machine, so whatever solaris you run on the machine is supported, as long as it's a supported release[09:53:30] <palowoda> If you can afford a platinum contract you would be on this irc.[09:53:36] <palowoda> wouldn't[09:53:37] <_Lewellyn> flyingparchment: hm. i seem to recall that solaris express contracts were cheaper, once upon a time.[09:53:52] <lblume> _Lewellyn: If you have a metal contracts, all flavors of Solaris are supported (if you're on the right hardware)[09:53:53] <flyingparchment> the old SX contracts were pretty useless[09:53:59] <flyingparchment> iirc, they allowed you to submit bugs, and nothing else[09:54:12] <_Lewellyn> flyingparchment: but i recall them "upgrading" upon release.[09:54:21] <colyte> _Lewellyn: Through SFE, it simply didnt start the compile because of dep issues.[09:54:33] <_Lewellyn> colyte: it requires 3 SUNW packages[09:54:38] * _Lewellyn scrolls back[09:55:03] <_Lewellyn> [00:04] <_Lewellyn> ok. SFErtorrent requires SUNWsigcpp, SFExmlrpc-c, SUNWcurl, SUNWncurses[09:55:03] <_Lewellyn> [00:04] <_Lewellyn> SFExmlrpc-c requires... nothing :)[09:55:22] <_Lewellyn> the build reqs are a bit harsher though, iirc.[09:55:33] *** Andrewski6667 has quit IRC[09:55:45] <colyte> _Lewellyn: Thanks for the headsup :)[09:55:49] <_Lewellyn> i extend my normal offer, though. i have prebuilt x86 binaries for that[09:56:13] *** jteo has joined #opensolaris[09:56:17] <_Lewellyn> and if you're using a recent pkgbuild, it should use pkg to pull in SUNW deps, from what i can tell[09:56:19] *** MattMan has joined #opensolaris[09:56:31] <codestr0m> flyingparchment: yes the support contracts are useless for SXCE and osol if they are the same.. and long terms means 5-10 years and short is 6 months - 2 years[09:56:35] <_Lewellyn> i have sxce and don't know for sure[09:56:53] <flyingparchment> codestr0m: contracts are sold for 1 year and 3 years. unless that's changed[09:56:59] <codestr0m> flyingparchment: thanks[09:57:01] <palowoda> So how much is a platinum contract these days?[09:57:05] <codestr0m> I think osol is 1 year max[09:57:46] <lblume> 1 year after the end of distribution (18 months from the start, then)[09:58:03] *** pgr has quit IRC[09:59:02] <colyte> _Lewellyn: the most recent beta when i tried. I havent been using opensolaris for that long ;)[09:59:14] <palowoda> Wait what is the best contract, gold silver or platinum?[09:59:22] *** mib_knsghv has quit IRC[09:59:44] <flyingparchment> i want a "solaris everywhere" contract.[09:59:47] <flyingparchment> but i suspect we're a little small.[10:00:16] <palowoda> Last I known a platinum was 56K.[10:00:24] <flyingparchment> it depends on the hardware[10:01:03] <palowoda> You mean how many mill's of hardware total?[10:01:18] <flyingparchment> no, the number of CPUs (for x86), or the system category (tier/whatever) for sparc[10:01:40] <palowoda> Ok explain.[10:01:42] <lblume> What about the non-metal support contracts? Like those for developers?[10:01:49] <flyingparchment> if your system has more cpus it costs more to buy support for it[10:02:08] <flyingparchment> that's why a contract for an SF15k costs more than a contract for a V210[10:02:12] <_Lewellyn> palowoda: 56k isn't much at all, really.[10:02:26] <palowoda> You mean platinum cost is based on cores?[10:02:34] <lblume> Sockets.[10:02:40] <flyingparchment> no, i said it's based on CPU (not cores) for x86 and system category for sparc[10:03:08] <palowoda> Example?[10:03:37] <flyingparchment> http://www.sun.com/service/subscriptions/prices.html[10:04:01] *** Deesl has quit IRC[10:04:28] *** Atomdrac1e has joined #opensolaris[10:04:35] *** neoxed has quit IRC[10:05:22] <palowoda> Oh it says sockets. Silly little people.[10:05:23] <flyingparchment> hmm.. what do people use for batch job scheduling in solaris?[10:05:47] <flyingparchment> say i have 10 compute servers and 1 head node, and i want something to schedule jobs over those systems[10:06:13] *** div110 has quit IRC[10:07:09] <palowoda> I thought culstering has obsoleted job scheduling.[10:07:25] *** csjp_ has quit IRC[10:07:42] <palowoda> Or cloud computing.[10:07:59] <flyingparchment> no, that's different[10:08:00] *** e-jat has joined #opensolaris[10:08:19] <flyingparchment> i have more work than available resources, and i need something to schedule that work, so the more important things are done first, and the less important things wait for resources[10:08:50] <palowoda> But cloud computing has more resorces than you right?[10:08:58] <flyingparchment> i don't know what that means[10:09:09] <flyingparchment> a "cloud" has as many resources as the number of machines in the cloud[10:09:12] <flyingparchment> (i hate that word, btw.)[10:09:25] <palowoda> It means job scheduling is out date.[10:09:39] <palowoda> The problem is lack of resources.[10:09:40] <flyingparchment> why?[10:09:46] *** mib_3y8xvg has joined #opensolaris[10:09:49] <palowoda> Hardware problem.[10:09:53] <flyingparchment> uh, no.[10:10:00] <palowoda> Yeah.[10:10:01] <flyingparchment> that would be true for some workloads, it's not true for this one[10:10:18] <flyingparchment> we need enough hardware for each job to finish in the time required for it[10:10:23] *** pgr has joined #opensolaris[10:10:28] <flyingparchment> that doesn't mean every job has to be running at once, competing for resources until they all finish[10:10:31] <palowoda> Is it a hardware problem?[10:10:38] <flyingparchment> there is no problem at all[10:10:46] <flyingparchment> i just need some software that can do job scheduling[10:11:42] <palowoda> Some third party companies do try to sell job scheduling on Solaris. Don't know if they are successful at it.[10:11:50] <flyingparchment> such as?[10:12:06] <palowoda> The ones that no longer exist?[10:12:29] <palowoda> Dead end road dude.[10:12:48] <monsted> while (read commandline); do ssh $node $commandline; done[10:13:06] <ofu> hooray, my opensolaris bible just arrived[10:13:15] *** KermitTheFragger has joined #opensolaris[10:15:09] *** fftb_ has joined #opensolaris[10:17:50] *** Atomdrache has quit IRC[10:19:50] *** Erwann has joined #opensolaris[10:20:29] <libkeise1> flyingparchment: most people i know are running openpbs or sge for that type of thing[10:20:42] *** tsoome1 has quit IRC[10:20:49] <flyingparchment> hmm, was just looking at grid engine. is it any good?[10:21:22] *** dunc has joined #opensolaris[10:21:28] <libkeise1> one of my friends runs a pretty big install for a bunch of bioinformatics compute clusters. they're quite happy with it[10:22:21] <flyingparchment> hmm, openpbs looks unsupported[10:22:27] <flyingparchment> unless you buy the commercial version[10:22:57] <fraggeln> how to i list the mac-address of a nic?[10:23:14] <asyd> ifconfig?[10:23:15] <flyingparchment> ifconfig -a, as root[10:23:48] <fraggeln> doh :/[10:23:51] <fraggeln> i did run as a normal user[10:24:15] *** Dar has joined #opensolaris[10:25:43] *** mikefut has joined #opensolaris[10:25:44] <lblume> arp[10:26:34] <lblume> It doesn't need root[10:30:04] *** wewek has joined #opensolaris[10:33:51] <Stric> fraggeln: arp -an as regular user can find out too[10:36:39] <fraggeln> Stric: thanks[10:37:51] *** Atomdrac1e has quit IRC[10:37:59] *** boyd_ is now known as boyd[10:38:54] *** wdp has joined #opensolaris[10:40:47] *** sophokles has left #opensolaris[10:40:59] *** Atomdrache has joined #opensolaris[10:41:12] *** pgr has quit IRC[10:48:56] *** stux|RC-only has joined #opensolaris[10:49:21] <mib_3y8xvg> Hi all[10:49:47] *** e-jat has quit IRC[10:49:48] <mib_3y8xvg> please excuse my overwhelming n00bness in IRC.[10:50:28] *** nollan has quit IRC[10:51:19] *** tsoome has joined #opensolaris[10:53:14] *** oenone has quit IRC[10:55:43] <mib_3y8xvg> I'd like to get some pointers where I could get started with OpenSolaris. Ive fiddled around with GNU linux in the past, but never got particularly attached to it. Its been a repeated, but always passing interest[10:56:38] <Stric> http://opensolaris.org[10:58:02] <mib_3y8xvg> thanks, ive been there. I feel quite overwhelmed by the information there. I need a more "for Dummies" approach, I think :P[10:58:24] <Stric> Maybe an "Asking question for Dummies" too? :P[10:58:32] <Stric> I have no idea what kind of information you're looking for[10:58:55] <Stric> Getting it, installing it, getting the source, compiling new kernel, using it, ...[10:59:12] <Stric> "get started" isn't all too specific ;)[11:00:10] <mib_3y8xvg> Hehe. Well, i got OpenSolaris running in a VBox virtual machine. Installing it was no problem. Im looking to use the powers of ZFS (which I read is the most advanced FS out there) to build a small file server[11:00:46] <Stric> http://opensolaris.org/os/community/zfs/demos/[11:00:53] <mib_3y8xvg> unfortunately, searching stuff like "samba opensolaris" on google, doesnt get me very far.[11:01:38] <Stric> google://"sharecifs"[11:01:43] <palowoda> Err what does zfs have to do with samba, which is cifs in solaris.[11:01:45] *** alex25 has joined #opensolaris[11:02:28] <palowoda> zfs best practices on http://www.solarisinternals.com/wiki/index.php/Solaris_Internals_and_Performance_FAQ is helpful too.[11:02:37] <_setuid_H> palowoda: well this is not about that solaris renamed it this is about that microsoft renamed it :-)[11:02:41] <Stric> samba is a userspace CIFS server, opensolaris also has a kernel integrated server with better performance (but less features in the "domain" part)[11:03:02] <mib_3y8xvg> palowoda: im just saying i want to make use of the reliability of ZFS and build a fileserver that'll play nice in an otherwise windows only network.[11:03:26] <palowoda> You have my permission.[11:03:36] *** lblume has quit IRC[11:03:45] <Stric> "play nice" is also a big vague.. depends if you just need it as a file server, or AD integration, NT4-domain integration, ....[11:04:22] <mib_3y8xvg> stric: thanks. thats helpful. Domain integration wont be an issue. Its just needed as a plain fileserver for a network share. Once thats up and running, I'd like to get things like ACL sorted out.[11:04:54] *** stux|RC has quit IRC[11:06:07] <Stric> then google://"sharecifs" will help[11:06:31] *** jteo has quit IRC[11:07:05] <mib_3y8xvg> stric: thanks for pointing me in the right direction. Ill give it a shot[11:07:42] <Stric> you can use samba as well, but the kernel cifs thingie will most probably give better performance[11:07:57] <stewart> Is it normal to only get 10-20 fsyncs()/second out of OpenSolaris ?[11:08:15] <Stric> how much can your storage do?[11:08:47] <stewart> Stric: it'll be a desktop drive (remote machine... haven't seen inside it)[11:09:09] <Stric> I was referring to the system that gives you 10-20 fsyncs/sec[11:09:18] <stewart> yeah, that's it.[11:09:27] <stewart> it's just OpenSolaris on a 16 core xeon[11:09:29] <Stric> fsync is a whole lot about how many IOPS your storage does[11:09:41] <stewart> yep[11:11:14] *** lesterc has joined #opensolaris[11:11:16] *** legolasw has joined #opensolaris[11:11:53] <legolasw> what is opensolaris command which works like fdisk -l[11:12:45] <stewart> but - for example. T1000 running linux completes this test in 41seconds. opensolaris on 16 core xeon times out after 5 or 10 minutes.[11:13:40] <seanmcg> legolasw, format ?[11:13:49] <legolasw> A list of all available disks[11:13:54] <seanmcg> format[11:14:08] <legolasw> Thank you.[11:14:17] <Stric> stewart: it does sound low though, but I guess a whole lot is about circumstances..[11:14:26] <Stric> legolasw: or iostat -En[11:14:39] <Stric> or cfgadm -la[11:14:46] <Stric> depends on what type of info you want[11:15:27] <Stric> seanmcg: just tested on a sol10 machine on zfs, writing 3 chars, then fsync().. *100 times.. 0.5 secs[11:15:39] <Stric> err. wrong s* ;)[11:15:45] <Stric> stewart: ^^^[11:15:56] <stewart> hrrm...[11:16:24] <stewart> Stric: http://paste2.org/p/162401[11:16:32] <stewart> Stric: using dtrace to find out how long each is taking[11:17:39] *** timsf has joined #opensolaris[11:18:00] <Stric> stewart: zpool iostat 1 at the same time gives what?[11:18:18] <Stric> write operations should be roughly the number of iops the storage does[11:19:01] <stewart> Stric: write ops staying constant at around 7 or 8 (maybe up to 10,11).[11:19:14] <Stric> that's not right[11:19:21] <stewart> write bandwidth anywhere from 32k to 1mb[11:19:32] <Stric> unless you're writing to a floppy or something[11:19:50] <Stric> assuming you're writing to ZFS.. otherwise zpool iostat is kinda worthless[11:20:45] *** oenone_ has joined #opensolaris[11:20:46] <stewart> Stric: yep. standard OpenSolaris install. 'zfs list' shows things mounted on /[11:20:53] *** mib_3y8xvg has quit IRC[11:20:56] <Stric> but if this is on ZFS, then no, that's plain wrong. you should see about 15x more or so[11:22:09] <oenone_> for servers (router, firewall, fileserver), is it safer to run 2008.11, or can i run sxce there without having to worry?[11:22:13] <Stric> for a single disk that is[11:22:20] *** otep has joined #opensolaris[11:22:28] <flyingparchment> oenone_: safer in what way? most people would suggest solaris 10 for production systems[11:22:46] <oenone_> safer in more reliable[11:22:59] <stewart> Stric: any ideas?[11:23:17] <Stric> stewart: driver issues or something similar?[11:23:28] <stewart> possibly... any way to check?[11:23:49] <Stric> stewart: just to confirm.. this is on a single or possibly mirrored regular disk, right?[11:23:54] <stewart> yep.[11:24:07] <stewart> AFAIK[11:24:23] <Stric> You don't know if it's on a HD or a floppy?[11:24:28] *** axisys has quit IRC[11:24:34] <stewart> well.. it's not a floppy :)[11:24:35] <Stric> or iscsi/...[11:25:04] <oenone_> flyingparchment: it's not quite a "production system" - it should serve as router/firewall for our student's house (about 11 students), so small downtimes (for upgrades, etc) are okay[11:25:06] <stewart> rpool is 136GB...[11:25:40] <stewart> how do i find out the devices that make up the pool?[11:25:45] <Stric> zpool status[11:25:49] <Stric> iostat -En[11:25:51] <Stric> cfgadm -la[11:26:15] <stewart> Stric: it says "pool is formated using an older on[11:26:18] <Stric> /usr/X11/bin/scanpci (or somesuch similar path) to list pci bus[11:26:20] <stewart> disk format" could that have anything?[11:26:28] *** legolasw has quit IRC[11:26:38] <Stric> nah, that won't affect this much[11:27:04] <stewart> c3t0d0s0 - which is: c3t0d0 Soft Errors: 0 Hard Errors: 0 Transport Errors: 0[11:27:04] <stewart> Vendor: Sun Product: STK RAID INT Revision: V1.0 Serial No:[11:27:04] <stewart> Size: 146.69GB <146685296640 bytes>[11:27:04] <stewart> Media Error: 0 Device Not Ready: 0 No Device: 0 Recoverable: 0[11:27:04] <stewart> Illegal Request: 0 Predictive Failure Analysis: 0[11:27:10] <Stric> you have a car that has all 4 wheels replaced with pingpong balls, not a scratch on the hood[11:27:56] <Stric> I would: 1) try regular solaris 10 and/or 2) ask Sun what's wrong since it seems to be a sun machine[11:28:38] <stewart> Stric: i'll go ask myself :) (i work for sun... but not on solaris or hardware. databases.).[11:28:49] <Stric> ;)[11:31:15] <stewart> Stric: thanks a lot though![11:33:48] *** oenone_ is now known as oenone[11:35:47] <Keso> anyone facing "pkg: image-update cannot be done on live image[11:35:51] <Keso> problem ?[11:36:02] <Keso> pkg: unable to create BE None[11:36:20] *** pt has joined #opensolaris[11:36:29] <DerSaidin> that sounds like your trying to update the livecd without installing it[11:37:50] <Keso> DerSaidin: I'm running 108 OSOL image and trying to do regular update[11:38:05] <DerSaidin> did you install it onto a harddisk?[11:41:34] <Keso> DerSaidin: yes, I'm running it from disk[11:41:51] <DerSaidin> you'll need to install it to a disk to do an image update[11:42:07] <Keso> DerSaidin: maybe it's problem I have removed original BE, because I had 5BE allready[11:42:23] *** Cass has joined #opensolaris[11:42:24] <Keso> DerSaidin: I'm booting it from my disk, no liveimage[11:42:24] <DerSaidin> oh, I didn't read that <_<[11:46:42] *** cemerick has joined #opensolaris[11:48:17] <eklof> Hi, I have some question on sharing a pool with smb. I have a /storage pool shared by smb. I then have a /storage/Documents filesystem also shared by smb. If I connect to the /storage share, I can see the /storage/documents folder but it's empty. But when I connect directly to /storage/document I see the content (browsing the filesystem remote) However listing the filesystem directly on the server things look fine.[11:48:30] <eklof> Is this some issue with smb or is it expected behaviour?[11:49:15] <eklof> Shouldn't one be able to have filesystem inside other filesystems and be able to see them when mounting the parent share? If you know what I mean.[11:50:41] <tsoome> its shared with kernel cifs?[11:51:12] <tsoome> probably is...[11:51:13] <eklof> Oh it could be a permission problem I just detected, hang on :)[11:52:36] *** penguini has joined #opensolaris[11:52:37] <eklof> Hrm noNo it wasn't :([11:52:48] <eklof> Yes I use the ZFS share property[11:53:12] <tsoome> the thing is, kernel cifs server scope is one file system; if your client is mounting share, its share corresponding to that zfs file system[11:53:39] <eklof> But if another filesystem reside inside, that one will not be shared unless mounted specifically?[11:53:48] <tsoome> so from client point of view, you need to mount all shares, then browse[11:53:56] <eklof> Hrm all right.[11:54:08] <eklof> Not so good idea to nest filesystems then if shared[11:54:21] <eklof> because they look empty and I get freaked out :)[11:54:22] <tsoome> depends[11:54:25] *** bizkut has joined #opensolaris[11:55:44] <eklof> I want to nest filesystems to easily see space allocation with df -h but only want to mount the top share with clients and still be able to browse all filesystems[11:55:52] <eklof> insted of mounting 50 filesystems :)[11:55:57] *** ericjray has joined #opensolaris[11:56:25] <eklof> tried nfs as well, but that gives even more strange things.[11:56:34] <tsoome> you can use samba ofc, as samba is not working on filesystem level, but its browsing directories[11:56:56] *** alex25 has left #opensolaris[11:56:58] <eklof> All right, so cifs,nfs and samba are not the same?[11:57:22] <_setuid_H> eklof: nfs is different[11:57:25] <eklof> oh sorry[11:57:27] <eklof> smb[11:57:42] <_setuid_H> it's like are those systems the same windows solaris windooze[11:57:54] <eklof> ok:)[11:58:03] <eklof> thought it was using the same protocol[11:58:05] <_setuid_H> wall smb stands for server message block[11:58:11] <_setuid_H> which is the protocol[11:58:16] <eklof> So CIFS is the stuff included in ZFS properties?[11:58:19] <_setuid_H> but Microsoft renamed it to cifs[11:58:23] <eklof> oh[11:58:28] <_setuid_H> so solaris is using the new name[11:58:32] <eklof> I see.[11:58:49] <_setuid_H> samba stands for the complete solution[11:58:50] <oenone> solaris has two methods to share, it's own implementation, or by installing samba[11:58:54] <_setuid_H> or how to say it[11:58:56] <tsoome> still there is sharesmb parameter in zfs and not sharecifs :D[11:59:13] <_setuid_H> tsoome: request for update rfu :-)[11:59:24] <tsoome> tbh, i cant care less:D[11:59:30] <_setuid_H> tsoome: :-)[12:00:21] <tsoome> http://en.wikipedia.org/wiki/Server_Message_Block[12:00:36] <eklof> Still think the behaviour is strange, but perhaps it's my Mac-klient that is the problem[12:00:48] <eklof> Only have macs to test with[12:01:02] <tsoome> why strange?[12:01:32] *** ruse39[home] has quit IRC[12:01:47] <eklof> If I mount the /storage filesystem I think I should be able to navigate to /storage/documents and see the contents even though it is a separate filesystem.[12:01:51] <tsoome> you have separate shares, what you expect is client to mount other shares automatically while browsing directory tree[12:02:02] *** penguini has left #opensolaris[12:02:08] <eklof> Ah, so it is a client problem?[12:02:12] <eklof> Just as I suspected.[12:02:16] <eklof> Damn Apple![12:02:20] <tsoome> ?[12:02:33] *** bizkut has left #opensolaris[12:02:37] <tsoome> how should the client know you have /storage/documents share?[12:02:57] * _setuid_H is going on lunch[12:03:11] <eklof> tsoome: It should know :)[12:03:29] <oenone> i think it's a understanding problem[12:03:58] *** timsf has quit IRC[12:04:42] <eklof> I nearly had a heartattack when navigating the document folder, knowing it sure had 1GB of content and it was totally empty... :) I had to check using ls on the server and my pulse went down :)[12:04:48] <tsoome> ig you have share names like "storage", "documents", there is no way the client can map them to /storage/documents by just using share names[12:04:53] *** incipit has joined #opensolaris[12:04:54] *** cypromis has quit IRC[12:05:03] <eklof> tsoome: yes I understand the problem.[12:05:16] *** Anne80 has quit IRC[12:05:55] <eklof> tsoome: I will change the folder and filsystem structore so i don't nest shared folders.[12:06:05] <eklof> And mount them individually.[12:06:49] *** ruse39[home] has joined #opensolaris[12:07:33] *** incipit has left #opensolaris[12:08:15] <oenone> raidz2 allows two disks to fail.. if i have raidz1 and one hot spare, two disks could fail, too.. any advice, which of these solutions would have better performance or reliability?[12:09:06] <Dominic> with raidz2, you don't have to wait for your hot spare to sync up[12:09:20] <Dominic> (which could in turn, induce an additional disk failure because of the load)[12:09:30] <eklof> Using spare, you have a vurnability-window when the first disc fail and the spare is synced. If another disk is failing during that time, you are toast! :)[12:09:30] <tsoome> microsoft is using DFS to solve this kind of issues, so if the client is browsing the dfs tree, it will mount needed resources while browsing. unfortunately the samba bits in macos are old and it cant really cope with dfs[12:10:05] <eklof> tsoome: ok, I will complain to Mr Jobs! :)[12:10:41] <tsoome> also, you need AD to set the dfs namespace up, therefore plain solaris server is no use for it....[12:10:46] <eklof> oenone: I'd say raidz2 is better.[12:10:47] *** oenone_ has joined #opensolaris[12:11:29] <eklof> tsoome: problem is that using NFS is really not better. It displays the document share as a strange icon and when i click it, it disappears :)[12:11:40] <eklof> Maybe the nfs-bit in mac osx is lame ass to! :([12:11:56] * eklof is soon installing opensolaris on his mac[12:12:12] <tsoome> you need to mount the document share first[12:12:43] <oenone_> did anybody answer my question? my server went offline again :([12:13:21] <eklof> Using spare, you have a vurnability-window when the first disc fail and the spare is synced. If another disk is failing during that time, you are toast! :)[12:13:33] <eklof> So I think raidz2 is better.[12:13:41] <Dominic> oenone_: with raidz2, you don't have to wait for your hot spare to sync up (which could in turn, induce an additional disk failure because of the load)[12:13:47] <eklof> Performance-wise, I have no idea :)[12:13:57] *** timsf has joined #opensolaris[12:14:06] <eklof> Should be faster since it have one more disk to read from[12:14:28] *** pt has quit IRC[12:14:32] <eklof> maybe a tad slower to write?[12:15:47] <oenone_> okay, thanks[12:17:55] *** kohju is now known as KOHJU[12:18:34] *** dunc__ has joined #opensolaris[12:19:04] *** e1mer has quit IRC[12:22:21] <oenone_> maybe even better would be raidz2 with hot spare(s)[12:23:03] *** phimic has quit IRC[12:25:09] *** phimic has joined #opensolaris[12:26:28] *** ericjray has quit IRC[12:27:22] *** phimic has joined #opensolaris[12:40:12] *** calumb has joined #opensolaris[12:40:49] *** calumb_ has joined #opensolaris[12:43:02] *** Odin- has joined #opensolaris[12:45:24] <qiyong> how do i dtrace a function?[12:48:13] *** dunc__ has quit IRC[12:48:56] <eklof> If I want to rename a pool I need to export it and import it?[12:50:38] *** calumb has quit IRC[12:53:36] <oenone> on zfs, is there this 5%-for-root thing, too?[12:55:57] *** gerryxiao has joined #opensolaris[12:56:11] <Stric> eklof: that's one way at least[12:56:22] <Stric> oenone: hm. not that I know..[12:56:23] <gerryxiao> IPS 109 not out?[12:56:31] <seanmcg> qiyong, userland function or in the kernel ?[12:56:36] <telexicon> oenone, what for[12:56:39] <eklof> I have had a bad experience with exporting and importing and that makes me nervous :)[12:56:43] <Stric> oenone: can be solved if you want to by creating something with reservation[12:56:43] <telexicon> oenone, what was it for on ext?[12:58:11] <oenone> telexicon: on other systems there is 5% of the disk space reserved for root, no user can use more space if it's 95% full (even df shows it as 100% full then)[12:58:29] <oenone> Stric: i just wanted to know if it was there by default.. if not it's okay[12:58:55] <Stric> telexicon: performance goes down the drain on most filesystems when you're getting short on free blocks[12:59:26] <telexicon> Stric, yeah, but zfs is copy on write so i dont know how that would interact[12:59:50] <Stric> telexicon: it still needs to find free blocks to put data.. and yes, zfs suffers from this too[13:00:19] <telexicon> Stric, doesnt zfs track a free area?[13:00:32] <Stric> telexicon: dunno. but i've had massive slowdowns when it's almost full[13:00:37] <telexicon> ok[13:00:41] <qiyong> seanmcg: kernel func[13:00:58] <jmcp> qiyong: use the fbt provider[13:01:11] <jmcp> dtrace -n 'fbt::*attach*:'[13:01:12] <jmcp> etc[13:01:28] <oenone> i just saw "zpool list" tells me 915GB free, but "zfs list" or "df" only shows 899GB free[13:01:39] <flyingparchment> oenone: raidz?[13:01:46] <qiyong> jmcp: fbt is function backtrace?[13:01:46] <oenone> no, single 1TB disk[13:02:01] <jmcp> qiyong: I believe it's "functional boundary tracing"[13:02:19] <jmcp> it doesn't really matter what fbt stands for though[13:02:32] <oenone> zpool tells me 13G used, zfs tells me 14.6GB used..[13:03:28] <eklof> oenone: can it be something with compression? Only a guess[13:03:36] <eklof> Do you have it on?[13:04:26] <qiyong> jmcp: dtrace: invalid probe specifier fbt:::iprb_m_start: probe description fbt:::iprb_m_start does not match any probes[13:04:43] <jmcp> qiyong: too many ":"[13:04:48] *** authentic has quit IRC[13:04:50] <jmcp> fbt::iprb_m_start:[13:05:13] <jmcp> it's "providername" : "modulename" : "functionname" : entrypoint[13:05:16] <oenone> eklof: no, default values for everything during install.. maybe swap space is not included in "zpool list"?[13:06:43] *** bigjocker has joined #opensolaris[13:08:01] *** jteo has joined #opensolaris[13:08:47] <seanmcg> as jmcp says is true. The docs refer to it being: 'providername : Probe Module : Probe Name : Entry point'[13:09:23] *** dguitar has joined #opensolaris[13:09:26] <seanmcg> where ProbeModule can be fbt, or syscall or in userspace a library name (libc.so.1 say), and Probe Name is generally the function name.[13:09:35] *** lewellyn_ has joined #opensolaris[13:09:35] *** lewellyn has quit IRC[13:10:04] <colyte> oenone: it's reported as so because a filesystem use storagespace simply because it stores it in a structure and such.[13:10:55] <qiyong> seanmcg: providername can be fbt ?[13:10:59] <jmcp> yes[13:11:12] <krisis> qiyong: i did a dtrace -l fbt:::| grep iprb and no such probe was found?[13:11:14] <qiyong> Probe Module is blank here[13:11:40] <jmcp> qiyong: for a userland function, I might use the pid provider instead, too -- dtrace -n'pid$target::open64:{print "%s", arg1};'[13:12:02] <jmcp> rather, dtrace -n'pid$target::open64:{print "%s", arg1};' -c "/path/to/command -w -i -t -h arguments"[13:13:58] <krisis> qiyong: fbt provider is only for kernel functions. are you trying to trace a userland function?[13:14:40] <qiyong> krisis: kernel[13:15:35] <jmcp> qiyong: the only mandatory part of your dtrace probe tuple is the provider and three ":"[13:15:36] <qiyong> what else i can do? show its args?[13:15:49] <jmcp> qiyong: you can see the value of the arguments, yes[13:16:00] <jmcp> it helps to know what sort of data to expect :)[13:16:03] *** szt_ has joined #opensolaris[13:16:17] <jmcp> qiyong: this is where src.opensolaris.org comes in handy[13:16:41] *** phimic has quit IRC[13:16:44] *** pandarilho has joined #opensolaris[13:16:45] *** phimic has joined #opensolaris[13:17:17] <krisis> jmcp: arent user guides easier places to start learning dtrace ?[13:17:30] <jmcp> krisis: easier than irc? sure[13:17:49] <jmcp> and you wouldn't learn how to *use* dtrace from src.opensolaris.org[13:18:09] <krisis> jmcp: i meant easier than src.opensolaris.org. and in my humble opinion[13:18:13] <qiyong> undefined function name: print[13:18:19] <tsoome> dtrace manual in docs.sun.com was like 800 pages;)[13:18:53] <jmcp> tsoome: I printed it out a few years back, got it bound and then managed to get bmc/ahl/mws to autograph it :-)[13:19:00] *** szt has quit IRC[13:19:02] <tsoome> :D[13:19:04] <tsoome> nice[13:19:06] *** phimic has quit IRC[13:19:22] <krisis> jmcp: i wish i get an autographed copy too ;)[13:19:26] <jteo> but no tshirt?[13:19:28] <jteo> ;)[13:19:38] <jmcp> jteo: no, no tshirt[13:19:54] *** lathiat has joined #opensolaris[13:19:55] <jmcp> qiyong: do yourself a favour and read through the dtrace doco - http://docs.sun.com/app/docs/doc/819-3620?l=en&q=dynamic+tracing+guide[13:20:08] <jmcp> it'll explain things a whole lot better (with examples) than we can here in the channel[13:20:08] *** jaek has quit IRC[13:20:12] <jmcp> more efficiently, too[13:20:17] *** phimic has joined #opensolaris[13:20:26] *** jaek has joined #opensolaris[13:20:28] <tsoome> its really favof for us as well;)[13:20:32] <tsoome> favor*[13:20:56] <jmcp> that's certainly true too[13:20:59] <krisis> i use this www.sun.com/bigadmin/content/dtrace/d10_latest.pdf, has helped me a lot. I am not sure if this is the best though.[13:21:20] <jteo> whatever happened to systemtap?[13:22:49] *** Caino has quit IRC[13:23:09] <krisis> there is a dtrace beta of sorts in linux. ftp://crisp.dynalias.com/pub/release/website/dtrace[13:23:43] <jmcp> krisis: wow[13:23:47] *** _Lewellyn has quit IRC[13:24:55] <krisis> jmcp: i cant vouch for its working, have not tried it :([13:25:04] <jteo> intriguing.[13:25:05] <jmcp> heh[13:27:15] <qiyong> jmcp: can it investigate it's local variables?[13:27:20] *** bnitz has joined #opensolaris[13:27:52] <jmcp> qiyong: I think so, but I don't know for sure[13:27:55] <jmcp> I haven't needed to[13:28:42] *** authentic has joined #opensolaris[13:30:22] *** ^authentic has joined #opensolaris[13:30:57] *** pgr has joined #opensolaris[13:31:07] <krisis> qiyong: where are these local variables, in kernel or userland?[13:31:37] <qiyong> kernel[13:34:08] <qiyong> prototype: char, short, int, long, or long long[13:34:08] <qiyong> argument: struct msgb *[13:34:21] <qiyong> jmcp: how do i print a struct * ?[13:34:27] <qiyong> "%x" doesn't work[13:34:35] <jmcp> qiyong: try %llx, or $p[13:34:38] <jmcp> %p, rather[13:34:53] <jmcp> if you want the structure's elements, you'll need to include the relevant header file in your D script[13:34:59] <jmcp> this is covered in the manual I pointed you to[13:35:17] <qiyong> jmcp: i was doing in command line[13:35:38] <jmcp> so now you'll have to start using a script instead[13:35:40] * jmcp shrugs[13:35:44] <jmcp> it's not like it's hard![13:36:13] <krisis> qiyong: about seeing local variables in kernel, you might have to add your own SDT probes. which is widely blogged about. ( i have no first hand experience )[13:36:44] <qiyong> jmcp: it's is func(arg0, arg1), right?[13:36:51] <jmcp> krisis: qiyong's example functions are from the iprb driver, which is Closed. SDT probes require source so you can compile them in[13:37:00] <jmcp> qiyong: yes, args number from 0 upwards[13:37:19] <krisis> jmcp: qiyong: yes, sorry i forgot to mention about source recompiling :([13:38:42] *** PicCard__ has quit IRC[13:38:55] *** Caino has joined #OpenSolaris[13:39:51] *** bnitz has quit IRC[13:40:05] *** Odin- has quit IRC[13:40:08] *** ^authentic has quit IRC[13:40:08] *** authentic has quit IRC[13:40:12] <Keso> ALL: is it know but that upgrade fail when you have zone?[13:40:24] *** authentic has joined #opensolaris[13:40:30] *** sipior has joined #opensolaris[13:40:33] <jmcp> Keso: non-global zones can get in the way, yes[13:40:45] <jmcp> Keso: you should generally detach the zone before running your upgrade[13:41:13] *** lesterc has quit IRC[13:41:32] <jmcp> then on reboot to the new build/version/... attach the zone with "zoneadm attach -u nameofzone"[13:41:36] <jmcp> it's what I do with LU[13:42:00] <jmcp> and I've been informed that it's something very similar for beadm and pkg(5)[13:42:51] *** hsp is now known as hsp_away[13:43:43] *** bnitz has joined #opensolaris[13:44:05] *** cypromis has joined #opensolaris[13:44:17] <krisis> jmcp: won't that mean, the detached zone be in the older BE?[13:44:29] <jmcp> krisis: that's why you use the "-u" flag on attach[13:44:43] <jmcp> so when it is attached, the zone management code knows to go and update all its bits[13:45:02] <krisis> jmcp: oh ok, thanks that solves an old problem of mine :)[13:45:19] <jmcp> it's something that affects me too - I run my home webserver in a non-global zone[13:45:38] *** lathiat_ has quit IRC[13:45:40] <jmcp> I used to have to backup all the data and config, blow away the zone and then recreate it from scratch after LUing to the next build[13:45:45] <jmcp> *most* aggravating[13:46:43] *** luc^ has quit IRC[13:46:56] *** MattMan is now known as MattAFC[13:47:02] <krisis> oooh, i dont depend heavily on my zone, i use it occasionally for testing. So never bothered to solve my 'old problem'[13:52:58] <Keso> jmcp: detach zone ?[13:53:32] <jmcp> Keso: zoneadm -z zonename detach[13:53:37] <jmcp> it's all there in the zoneadm(1m) manpage[13:54:06] <Keso> jmcp: thank you, maybe that was my problem[13:58:13] *** mlh has quit IRC[14:01:21] *** _coredump_ has quit IRC[14:04:01] *** legolasw has joined #opensolaris[14:04:34] *** _coredump_ has joined #opensolaris[14:04:41] *** mlh has joined #opensolaris[14:07:18] <legolasw> Where I can find information about disk naming in opensolaris? for example c5t0d0 and c4d1 means and things like that[14:07:28] <crichardso> anyone here fimilar with comstar i have some questions about if i can use it to do something[14:09:00] <timsf> legolasw: let me google that for you.[14:09:04] <timsf> http://www.sun.drydog.com/faq/9.html#s9.37[14:09:12] *** KOHJU is now known as kohju[14:10:26] <crichardso> like first does it support exporting volumes via iscsi? here is the thing i would like to do i would like to take our existing sans connect them via iscsi to comstar than use comstar for cutting up into volumes and dishing them out to our esx server. this way we can make all our sans in diffrent locations be able to work togher by doing snapshots/relication via the comstar. is this something that is possible with comstar?[14:11:14] <holcomb> yes.[14:11:23] <timsf> and http://docs.sun.com/app/docs/doc/806-4073/6jd67r9fj?l=en&a=view&q=disk+names[14:11:53] <crichardso> holcomb was that yes for me?[14:12:38] <holcomb> yes[14:13:13] <crichardso> ok cool and the stuff that comes into comstar you can manage with the same zfs style right?[14:13:25] <crichardso> in regards to snapshoting and stuff[14:13:33] <holcomb> for the snapshots, yes[14:13:41] <legolasw> I have two partitions in c5t0d0, how these partitions are named? are they c5t0d0p1 and c5t0d0p2?[14:13:43] <holcomb> setting up comstar to do iscsi is a little confusing, but there are examples out there[14:13:58] *** axisys has joined #opensolaris[14:13:58] <crichardso> ya i expected that much[14:14:07] <legolasw> I create the partitions using format and then fdisk sub command, but I can not make a sample raidz pool out of them.[14:14:19] <crichardso> but thats fine cause the normal shareiscsi doesnt work anyways with esx[14:16:37] <crichardso> do you know if the only package in ips that is needed is SUNWiscsit ?[14:17:28] <holcomb> one sec...[14:17:57] <holcomb> that and SUNWstmf[14:18:50] <crichardso> alright thanks a bunch mate[14:19:00] *** telexicon has quit IRC[14:19:16] *** telexicon has joined #opensolaris[14:22:18] *** gerryxiao has quit IRC[14:23:17] *** ken-work is now known as ken-away[14:24:27] <seanmcg> legolasw, you need to have solaris slices, not fdisk partitions. solaris slices live within a single fdisk partition[14:25:02] *** sparc-kly has joined #opensolaris[14:25:02] *** ChanServ sets mode: +o sparc-kly[14:25:52] <legolasw> does format command create those slices?[14:27:49] <asyd> s/create/define/ yes[14:29:59] *** ken-away has quit IRC[14:30:27] *** pgr has quit IRC[14:32:09] *** calumb_ is now known as calLNCH[14:32:47] *** _coredump_ has quit IRC[14:33:06] *** _coredump_ has joined #opensolaris[14:36:33] *** sparc-kly has quit IRC[14:40:29] *** pgr has joined #opensolaris[14:41:55] *** barfoed has joined #opensolaris[14:41:57] *** Odin- has joined #opensolaris[14:43:12] *** ericjray has joined #opensolaris[14:43:44] *** digifor has joined #opensolaris[14:44:21] *** spike- has joined #opensolaris[14:44:23] <spike-> heya[14:44:25] <spike-> whats up world[14:44:27] <spike-> got a major problem here[14:44:30] *** VenerZen has joined #opensolaris[14:45:28] <spike-> If anybody can take a min -> http://pastebin.com/m3d0466bf[14:45:41] <asyd> hmm spike- ?[14:46:05] <asyd> ah, not the same than spike, my bad[14:46:39] <asyd> sounds bad[14:47:50] <spike-> huh?[14:48:08] *** Deesl has joined #opensolaris[14:48:08] <spike-> Anybody have any idea how to rectify this situation?[14:48:22] *** christophsturm has joined #opensolaris[14:50:11] <spike-> can anybody help?[14:50:11] *** axisys has quit IRC[14:51:09] <asyd> tried zpool scrub? and check errors using fmadm[14:51:18] *** niq has quit IRC[14:51:23] <spike-> did you look at my pastebin?[14:51:28] <spike-> it's not an imported tank[14:51:34] <spike-> thus i don't think i can scrub it (maybe i'm wrong here)[14:52:10] <spike-> last time it was mounted was in b106 (now i'm running 109) and it was imported while booted into opensolaris fail safe command line mode[14:52:26] <spike-> and i scrubbed it the last time it was mounted (can't remember if i exported or not though, the machine was unstable and having major issues)[14:53:08] *** digifor has quit IRC[14:53:08] <spike-> but even if the second raidz1 array is missing one vdev that shouldn't prevent me from importing it![14:54:32] *** alain10 has joined #opensolaris[14:54:47] *** hsp_away is now known as hsp[14:57:23] *** Jiraia has joined #opensolaris[14:57:36] *** proberts has joined #opensolaris[15:01:32] *** ericjray has quit IRC[15:01:41] *** LordIllpalazo has joined #opensolaris[15:04:39] <spike-> Hello everybody, if anyone could please provide some input on my problem here -> http://pastebin.com/m3d0466bf[15:04:45] *** Deesl has quit IRC[15:04:46] <spike-> kinda freaking out here...[15:04:58] <phips> win 2[15:05:01] <phips> whoops[15:09:29] <spike-> Asyd - As far as using fmadm, what exactly am i looking for? The vdev that is reporting as unavail is a vdev that was -removed- and replaced.[15:09:43] <asyd> so you need to ack the error[15:10:34] <spike-> what fmadm command should i run?[15:10:38] <spike-> i'm not seeing any errors[15:10:58] <alain10> fmadm faulty :)[15:11:04] <asyd> you have nothing when you type fmadm faulty?[15:11:07] <spike-> yup that's what i was running[15:11:09] <spike-> and i'm getting nothing[15:11:13] <asyd> ok[15:11:26] <spike-> tried fmadm faulty -a too[15:11:27] <spike-> still nothing.[15:11:35] <spike-> all 15 drives are properly functioning[15:12:02] <spike-> (most are <1 month old)[15:12:21] <spike-> have any other ideas as to what i can do?[15:12:41] <spike-> I think it may have something to do with the fact that the last time i mounted it was under a different os version[15:13:12] *** NoFX_SBC has joined #opensolaris[15:14:42] *** X-Ray has joined #opensolaris[15:14:50] *** twisti has quit IRC[15:16:36] *** MattAFC is now known as MattMan[15:16:36] *** axisys has joined #opensolaris[15:17:31] <spike-> Anybody?[15:18:09] *** timsf has quit IRC[15:18:21] <tsoome> dont believe the os version does matter:P[15:18:40] <spike-> i wonder if the zfs version changed in 109[15:18:45] <alain10> searching spike ...[15:18:49] <spike-> i can't seem to find when 14 was released[15:18:57] <spike-> hehe thanks alain10, so am i :p[15:19:10] <spike-> this sucks, i'm soo worried that i'm not going to be able to recover this data[15:19:18] <spike-> its not the -only- copy but it sure would be nice if i could[15:19:20] <asyd> no backu?[15:19:35] *** twisti has joined #opensolaris[15:19:38] <tsoome> even if there is new zfs version - it wont upgrade the on disk version by itself[15:19:39] <spike-> there is, but it's on 100+ DVD's, a huge incremental[15:20:39] <spike-> hmmm, but don't I have to be able to import in order to upgrade?[15:20:47] <spike-> i'm thinking about installing b106 again to see if it can mount it[15:21:48] *** szt_ is now known as szt[15:21:57] *** wdp has quit IRC[15:22:09] <tsoome> no you cant upgrade zfs if its not imported[15:22:24] <spike-> hmmmmpf looks like i'm in trouble here[15:22:36] *** wdp has joined #opensolaris[15:22:37] *** timsf has joined #opensolaris[15:22:52] *** tombhadAC has joined #opensolaris[15:23:02] <tsoome> but you should be able to import unless zfs version is higher than currently supported one[15:23:27] <spike-> how do i get more of a log as to whats going on with zfs[15:23:30] <spike-> like why it can't import[15:23:46] <tsoome> if you get error about disk access, try to check if disks are actually here and accessible[15:23:50] <spike-> doesn't make sense, one vdev should be able to be unavail in a raidz1 array[15:24:03] <spike-> yes they are all present and accessible, did you view the pastebin?[15:24:25] <spike-> spike- Hello everybody, if anyone could please provide some input on my problem here -> http://pastebin.com/m3d0466bf[15:24:48] <tsoome> vdev can fail in working set, im not really sure if you are supposed to be able to import one with failed vdev[15:25:15] <oenone> well, there should be a possibility to resolve it[15:25:25] <tsoome> can you access all the disks with format for example?[15:25:42] <spike-> oh shit[15:25:45] <spike-> i thought i could[15:25:47] <spike-> but i missed something[15:25:49] <spike-> i'm one disk short[15:25:51] <tsoome> :D[15:25:54] * spike- goes to figure out whats going on[15:25:55] <spike-> thanks man[15:25:57] <tsoome> indeed[15:26:01] * spike- stops flipping out for a second[15:26:03] <seanmcg> spike-, its also not really recomended to have more that 5 or 6 disks in a raidz[15:26:06] *** ericjray has joined #opensolaris[15:26:19] <spike-> i thought 8 or 9 was kinda an upper limit[15:26:20] <oenone> seanmcg: why that?[15:26:41] <tsoome> oenone: pure statistics and probability[15:26:45] <seanmcg> performance, see solarisinternals.com for more[15:27:08] <tsoome> the more disks you have in raid the more chance you have there will be failures[15:27:53] <tsoome> performance is another issue ofc, overlooked that:D[15:27:57] <seanmcg> and what tsoome says too :)[15:28:39] <spike-> okay right on[15:28:47] <spike-> thanks for the advice, maybe i'll break it up into 3 sets of 5[15:28:51] <spike-> but first the data[15:28:51] <spike-> heh[15:28:59] <oenone> thanks for the link, seanmcg[15:29:12] <spike-> yeah i'm there now too[15:29:12] <spike-> heh[15:29:21] *** bodie has joined #opensolaris[15:29:22] <tsoome> spike you have pool with like 2x raidz1 groups[15:29:38] <tsoome> ?[15:29:43] <spike-> YUP[15:29:47] <spike-> err sorry for caps[15:30:09] <spike-> one of them has 10 500GB and the other is 5 750GB[15:30:28] <spike-> i've never had performance issues when its online though, its mainly for backup[15:30:28] *** stewart has quit IRC[15:30:29] <tsoome> ah you hot huge group in second one...[15:30:38] <spike-> yup[15:30:50] *** stewart has joined #opensolaris[15:30:58] <tsoome> well, the perfomance "issue" depends on your needs:D[15:31:01] <spike-> i may break that one into 2 different groups for purposes of reliability (once i do a bit more reading as to what was mentioned earlier)[15:31:09] <spike-> but as far as speed i've always had pleanty[15:31:17] <spike-> yeah exactly, my needs are -very- basic[15:31:29] <spike-> this machine mainly runs automated rsync backup tasks in the background[15:31:38] <tsoome> if you have like 99% of reads, large R5 is just ok[15:31:48] *** bodie has left #opensolaris[15:32:07] <tsoome> its the writes which will hit hard...[15:32:11] <spike-> yeah i have mainly reads[15:32:19] <spike-> i probably am around 99% reads[15:32:49] <spike-> because basically this machine pulls a backup from a bunch of other machines at least once daily, but most files don't change, and are thus only checksummed and then skipped over[15:32:54] <tsoome> as to write block, you need to read like full stripe, change block and write down new lock and new parity[15:32:58] <spike-> anyhow brb i gotta figure out whats going on with this disk[15:33:10] <tsoome> lock=block[15:33:52] *** codestr0m is now known as codestr0m|away[15:33:53] <spike-> and unfortauntely[15:33:55] <spike-> no disks are missing[15:34:12] <spike-> i just miscounted, there are 15 including 0, so 16 total[15:34:15] <spike-> so all disks do show up in format[15:34:49] <spike-> so all disks are in format, and thus should be accessible, yet the zpool import command reports that one vdev is screwed...[15:35:07] <spike-> replacing DEGRADED 11144084490198638692 UNAVAIL cannot open c5t0d0 ONLINE[15:35:18] <spike-> it shows the replacement drive as being online[15:35:22] *** bizkut_ has joined #opensolaris[15:35:27] <codestr0m|away> hahahaha.. greate pastebin guys.. can we get this changed in the topic please http://opensolaris.pastebin.ca/index.php > "You have no permission to post here (suspected spam). If this is not right, please send your post to the administration so the filters can be adjusted."[15:35:29] <spike-> the only one that isn't avail is the drive that i replaced, and then pulled from the system,[15:35:44] <asyd> codestr0m|away: lol[15:35:56] *** asyd changes topic to "OpenSolaris Annual Meeting is on #opensolaris-meeting | SXCE 109, ON 109, IPS 108 || Step one: see if SAG answers your question: http://docs.sun.com/app/docs/coll/47.16 || The answer to "I do $foo on *ux, how do I do it on Solaris" is http://bhami.com/rosetta.html || Clipboard: http://www.pastealacon.com/"[15:36:03] <codestr0m|away> asyd: thanks[15:36:16] *** calLNCH has quit IRC[15:36:18] <codestr0m|away> btw. when is the submission deadline.. it's like a couple days isn't it?[15:36:52] <oenone> submission deadline for what?[15:37:32] <tsoome> spike-: your pastebin suggests like its replacing 3 drives, not one?[15:38:12] <spike-> oh yes it was[15:38:17] <spike-> but i did them one at a time[15:38:36] <spike-> and the drives were all connected when i did the replacing[15:38:53] <tsoome> err, did you check if resilver was completed?[15:38:54] <spike-> i.e. when I did the replacing there were 17 drives total[15:38:59] <spike-> yes it was[15:39:01] <spike-> i scrubbed too[15:39:06] <spike-> then exported[15:39:07] <spike-> rebooted[15:39:09] <spike-> imported[15:39:11] <spike-> scrubbed again[15:39:15] <spike-> and then i don't know if it got exported or not[15:39:20] <spike-> scrubs came back clean[15:39:33] <crichardso> anyone here fimilar with compstar i need some assistence lol i enabled the services both the iscsi and stmf i created the logical unit based off of zfs i created a target with itadm but how do i set it to show my volumes to my esx server?[15:39:59] *** legolasw has quit IRC[15:40:12] <crichardso> sorry if it is a stupid question but i am really weak in iscsi and comstar commands are just adding to my lack of knowledge[15:40:16] <spike-> tsoome: but to answer your question more directly - Yes I did ensure the resilver completed 100%[15:40:17] *** Guest14948 has joined #opensolaris[15:40:23] *** Guest14948 has left #opensolaris[15:41:09] <spike-> I had 13 sata drives + 2 IDe drives, i decided to go all sata (for purposes of hot swap bays), so I added 2 more SATA drives (while leaving the 2 IDE drives connected) and issued the replace command (on them one at a time, letting them resilver).[15:41:22] <spike-> Also, one disk died, so I replaced this disk, and let it resilver[15:41:57] <noptrix> anyone using opensolaris under qemu?[15:42:01] <spike-> that's why there are 3 replacing commands going on, yet this pool has imported before, with all 3 running, and has scrubbbed fine, it just has something todo with the replacing not being cleared because i was performing maintenance on the pool in os failsafe mode[15:42:11] *** RavenSlay3r has joined #opensolaris[15:43:34] <spike-> So any ideas?[15:43:35] <spike-> heh[15:44:12] <spike-> tank UNAVAIL insufficient replicas[15:44:20] <spike-> maybe it doesn't have anything todo with the replacing's[15:44:26] <spike-> maybe it has to do with replicas[15:45:00] <bda> spike-: Email zfs-discuss@.[15:45:10] <bda> You're much more likely to get someone who knows how to help you there.[15:45:15] <spike-> heh i've had to do that before but didn't get much help[15:45:19] <spike-> well not for days[15:45:20] *** bigjocker has quit IRC[15:45:26] <spike-> but alright i guess that's what i'll do[15:45:53] <bda> Also, idle in #zfs. Someone may pick up there as well.[15:48:31] *** nsuperbus has joined #opensolaris[15:48:36] *** bizkut_ has left #opensolaris[15:49:04] *** X-Ray has quit IRC[15:51:52] *** oenone_ has quit IRC[15:56:13] *** LordIllpalazo has quit IRC[15:57:19] <spike-> i am, thanks :)[15:57:23] *** mega has joined #opensolaris[15:57:25] <spike-> also am awaiting my email to appear on the list.[15:58:46] <CIA-40> Lubomir Sedlacik <Lubomir.Sedlacik at Sun dot COM>: 6802008 webrev generates relative URI for workspace names containing colon which confuse browser[15:59:45] *** yippi has joined #opensolaris[16:00:06] *** luc^ has joined #opensolaris[16:01:52] <eklof> Can someone find why this is not accepted: http://pastebin.se/197666[16:02:00] *** qiyong has quit IRC[16:02:14] <eklof> syntax error on line 11[16:03:06] *** calumb has joined #opensolaris[16:03:37] *** jfisc has joined #opensolaris[16:05:03] *** spike-- has joined #opensolaris[16:05:13] <spike--> darn did i miss anything i got d/c'd for a minute[16:05:20] <spike--> i've been reading all over the place[16:05:24] <spike--> and my error just makes no sense[16:05:39] *** spike- has quit IRC[16:10:46] <trichobezoar> where did you get this using variables in ipf.conf?[16:11:46] <eklof> Dunnu, that doesn't work?[16:12:48] *** christophsturm has quit IRC[16:13:06] *** bnitz1 has joined #opensolaris[16:13:12] *** RaD|Tz has joined #opensolaris[16:13:17] *** christophsturm has joined #opensolaris[16:13:19] *** urbanmonk has joined #opensolaris[16:14:29] <spike--> rrrrrrrrg[16:14:36] <spike--> this makes no sense at all[16:14:40] <spike--> i really want my array back[16:14:51] <trochej> :/[16:15:00] <trochej> coffee[16:15:06] <spike--> i subscribed to the zfs-discuss list[16:15:10] <spike--> and sent an email to it with my problem[16:15:12] *** bnitz has quit IRC[16:15:36] <spike--> but i haven't got a subscription confirmation yet, nor have i received any sort of a reply to the email i sent to the list (the listbot must be really lagged...its been like 10+ minutes now)[16:17:52] *** RaD|Tz has quit IRC[16:23:30] <spike--> The only command one can issue to an array that isn't imported is import right?[16:23:50] *** _setuid_H has left #opensolaris[16:24:52] <spike--> Cool finally got a reply back from it[16:24:58] <spike--> anybody else got any advice on my problem?[16:25:16] *** mynhier has joined #opensolaris[16:25:18] <spike--> http://pastebin.com/m3d0466bf[16:25:26] *** mega has quit IRC[16:25:29] *** cri__ has joined #opensolaris[16:26:14] <trichobezoar> random idea Did you try removing the zpool cache?[16:26:22] *** luc^ has quit IRC[16:26:39] <spike--> i bet that happened when I reinstalled opensolaris, but no ididn't explicitly do that[16:26:49] <spike--> should i? And how?[16:27:54] *** pjlv has joined #opensolaris[16:28:19] <spike--> If anybody here moderates the zfs-discuss list, please approve my post[16:31:11] *** christophsturm has quit IRC[16:33:06] *** christophsturm has joined #opensolaris[16:33:44] <spike--> hmm[16:33:47] <spike--> okay so i looked around a bit[16:33:53] <spike--> are you talking about thie ZIL cache thing?[16:34:46] *** bubbva has joined #opensolaris[16:35:30] *** ken1 has joined #opensolaris[16:36:34] <spike--> Anybody around?[16:36:42] <pandarilho> no[16:36:50] <pandarilho> :-)[16:36:55] <spike--> hehehe[16:36:59] <spike--> Any ideas pandarilho?[16:37:12] <spike--> http://pastebin.com/m3d0466bf[16:37:18] *** tsoome has quit IRC[16:38:00] <pandarilho> may be hard disk problems?[16:38:07] <sactodave1> spike--: there is a file /etc/zfs/zpool.cache that you can remove and then reboot and see if ZFS find your pool and the disks it's suppose to be on. It seems that you moved the disks or had 3 die at once.[16:39:15] <pandarilho> whow, 3 disks failures at once... you are a lucky guy ;-)[16:39:24] <spike--> pandarilho: Not hdd problems, all disks are <1 month old, the array was scrubbed last night in os 106, but then i installed 109 and i can't get it to re-import[16:39:29] <spike--> naw they didn't all fail at once[16:39:36] <spike--> 1 disk failed and i replaced it with another and issued the replace command[16:39:39] <spike--> then i added 2 more disks[16:39:50] <spike--> and replace 2 existing (still working) disks with the 2 new disks[16:39:57] *** christophsturm has quit IRC[16:39:59] <spike--> then i removed the 2 old disks once the resilvering completed.[16:40:06] <spike--> then i rebooted and scrubbed and all and everything was just find[16:40:36] <asyd> you should ask on the zfs mailing list[16:40:41] <spike--> well except for a completely unrelated reason opensolaris wouldn't boot fully, so i could only use safe mode, which is why the replacing never went away[16:40:42] <spike--> i did[16:40:49] <spike--> still awaiting moderator approval[16:40:57] *** alfism has joined #opensolaris[16:40:59] <asyd> moderator approval?![16:41:02] <spike--> sactodave: i'm trying this right now, i'll be back in a few minutes (as i'm talking on the computer i'm working on)[16:41:14] <spike--> asyd: yup, evidently it's a moderated list (makes sense)[16:41:31] <spike--> hah[16:41:49] <spike--> sactodave1: Thanks for the idea, but unfortunately /etc/zfs/zpool.cache doesn't exist?[16:41:49] <asyd> hmm, I never saw a moderated list on opensolaris.org (but that may exists)[16:42:17] <spike--> oh[16:42:18] <spike--> shoot[16:42:29] <spike--> for some reason my subscription request hadn't finished before i posted to the list[16:42:32] <spike--> so it thinks i'm a non-member[16:42:35] <spike--> lets see if i can fix that[16:42:55] *** comay has joined #opensolaris[16:43:06] *** ChanServ sets mode: +o comay[16:43:57] <spike--> thanks for the heads up on the lack of moderation on opensolaris lists, i would've been waiting around forever before that got approved (if ever at all)l[16:44:05] *** stewart has quit IRC[16:44:13] *** stewart_ has joined #opensolaris[16:44:41] <spike--> heya comay[16:44:46] <spike--> you know much about zfs?[16:44:53] <asyd> spike--: you can use jive (the webforum) if you already subscribed to opensolaris.org[16:45:51] <spike--> asyd: once again thanks, but now i think my post to the email list has already worked :)[16:46:08] <asyd> well, you're welcome, and good luck :)[16:46:09] <spike--> in the future though ... :)[16:46:15] <spike--> thanks i'll need it[16:46:20] <spike--> hopefully the data is all still here and recoverable[16:46:31] *** Cass has quit IRC[16:47:39] *** perlmongo has quit IRC[16:49:24] *** cri__ has quit IRC[16:49:26] *** Cass has joined #opensolaris[16:50:13] *** spike-- has quit IRC[16:50:18] *** fftb_ has quit IRC[16:50:32] *** `Spike has joined #opensolaris[16:52:01] *** gerryxiao has joined #opensolaris[16:53:53] *** `Spike has quit IRC[16:53:57] *** phimic has quit IRC[16:54:56] *** bam123 has joined #opensolaris[16:56:07] *** Roms has joined #Opensolaris[16:58:10] *** duri has quit IRC[16:59:36] *** carl- has quit IRC[17:00:34] *** mega has joined #opensolaris[17:01:29] *** proberts_ has joined #opensolaris[17:02:45] <eklof> Hey, why can I ping the machine with pass in on rge0 proto icmp from any to rge0[17:03:01] <eklof> and not with pass in on rge0 proto icmp from any to rge0 icmp-type 0[17:03:18] <asyd> becaus icmp-type is only the request, not not the answer?[17:03:26] *** pgr has quit IRC[17:03:29] <asyd> +0[17:03:30] *** cwebber has joined #opensolaris[17:03:52] <eklof> asyd: I see, so what more type do I need to allow?[17:03:57] <asyd> 8[17:04:02] <eklof> Ah thanks.[17:04:06] <asyd> iirc[17:04:41] *** duri has joined #opensolaris[17:04:52] <asyd> yes, 0 and 8[17:05:01] *** `Spike has joined #opensolaris[17:05:04] <asyd> and 0 is the reply, not the request..[17:05:13] <`Spike> heya okay back[17:05:21] <`Spike> had to give a restart a try[17:05:23] <`Spike> didn't help though.[17:05:39] <eklof> asyd: yep works fine now, you are my friend! :)[17:06:27] *** Dar has quit IRC[17:07:46] <eklof> Another question, where is the log-file for ipf ?[17:08:07] *** Jiraia has quit IRC[17:10:18] *** freakazoid0223 has joined #opensolaris[17:10:26] *** jollyd has joined #opensolaris[17:10:34] *** jollyd has left #opensolaris[17:10:53] <`Spike> http://www.opensolaris.org/jive/thread.jspa?threadID=96801&stqc=true[17:13:06] *** Roums has quit IRC[17:13:55] *** __ms has quit IRC[17:14:16] <eklof> Will this rule allow outgoing on all my state-enabled rules for example ssh?[17:14:20] <eklof> pass out on rge0 proto tcp/udp from any to any keep state[17:14:45] <eklof> if I before that rule has a block out all[17:14:55] <eklof> have even[17:15:13] <eklof> I don't want to lock myself out :)[17:15:29] <asyd> well, add a crontab which flush firewall rules every 5 minutes..[17:17:21] *** urbanmonk has quit IRC[17:19:01] <eklof> asyd: good idea[17:19:37] <asyd> yes, trust me, I already fuck a router ~10 years ago by playing with firewall rules :)[17:19:38] *** appletizer has joined #opensolaris[17:19:41] *** appletizer has left #opensolaris[17:19:44] <asyd> "ooops"[17:20:11] *** proberts has quit IRC[17:20:23] *** wewek has quit IRC[17:20:49] *** MiraOsladil has joined #opensolaris[17:20:56] <trichobezoar> and "sorry"[17:20:59] *** VenerZen has left #opensolaris[17:21:18] *** tamr has joined #opensolaris[17:22:46] <eklof> asyd: :)[17:23:18] <eklof> what's the command for flushing all rules?[17:23:49] <eklof> simply svcadm disable network/ipfilter perhaps?[17:24:09] <asyd> ipf -F a[17:24:48] <eklof> thanks.[17:25:02] <asyd> you should try before, however :)[17:25:52] *** tsoome has joined #opensolaris[17:27:55] *** rv- has joined #opensolaris[17:29:14] *** Jiraia has joined #opensolaris[17:30:52] <eklof> asyd: :)[17:30:53] *** gerard13 has quit IRC[17:31:14] <eklof> Maybe I dont need that rule at all if I have keep-state on the incoming ssh-rule?[17:31:35] <eklof> So I *should* be able to have block log all[17:32:09] <eklof> and then pass in log on rge0 proto tcp from 192.168.1.0/16 to rge0 port = ssh keep state[17:32:21] <eklof> And then ssh should work both ways?[17:32:46] *** NoFX_SBC has quit IRC[17:32:57] <eklof> ehr, incoming ssh-sessions i mean, not ssh from inside the server[17:33:17] *** MiraOsladil has left #opensolaris[17:35:24] *** `Spike has quit IRC[17:36:05] *** gerryxiao has quit IRC[17:36:20] <trygvis> how can I unescape the output from svccfg -p tm_common_name <fmri>?[17:36:23] *** KermitTheFragger has quit IRC[17:36:51] *** jteo has quit IRC[17:39:26] *** KermitTheFragger has joined #opensolaris[17:41:10] *** Cass has quit IRC[17:41:20] *** bubbva has quit IRC[17:41:32] *** bubbva has joined #opensolaris[17:44:48] *** bam123 has left #opensolaris[17:46:21] *** alfism has quit IRC[17:46:43] *** sactodave1 has left #opensolaris[17:49:02] *** bubbva has quit IRC[17:51:24] *** xRaich[o]2x has joined #opensolaris[17:53:56] *** Cass has joined #opensolaris[17:55:47] *** codestr0m|away is now known as codestr0m[17:59:14] *** sommerfeld has joined #opensolaris[17:59:15] *** ChanServ sets mode: +o sommerfeld[18:05:09] *** `Spike has joined #opensolaris[18:09:44] *** esok has joined #opensolaris[18:13:36] *** NoFX_SBC has joined #opensolaris[18:14:48] *** pablomh has quit IRC[18:14:48] *** doug_out is now known as doug_outout[18:22:47] *** DTEIT has quit IRC[18:24:36] *** Auralis_ has quit IRC[18:24:58] *** doug_outout is now known as doug_out[18:27:05] *** rv- has quit IRC[18:27:24] <crichardso> man i am getting horrible performance on iscsi/nfs i was hoping comstar might be better but it didnt[18:27:43] *** Cass has quit IRC[18:27:49] <crichardso> goes from max of 3m/s to like 30-50m/s when i disable zil ;/[18:27:51] <tsoome> on top of zfs?[18:27:56] <crichardso> ya[18:28:07] <tsoome> get ssd/nvram card[18:28:22] <tsoome> and attach as a zil log device[18:28:48] *** estibi has joined #opensolaris[18:28:49] <tsoome> you can even test zil log on another fast hdd i guess[18:28:54] *** doug_out is now known as doug_outout[18:28:55] *** Auralis has joined #opensolaris[18:29:05] <holcomb> even a regular hd greatly helps[18:30:13] *** Erwann has quit IRC[18:30:13] <crichardso> how do you move it? how big does it typicaly get?[18:30:23] *** doug_outout is now known as doug_out[18:30:25] *** kohju is now known as KOHJU[18:30:30] <crichardso> would a flash drive work or is that a bad idea[18:30:53] *** yippi has quit IRC[18:30:54] <tsoome> altho i had like 50% of perfomance drop with zil when i did test iscsi from my mac to snv_108[18:31:58] <holcomb> depends on the flash drive[18:32:09] <holcomb> some of them have horrible sequential write[18:32:53] <tsoome> i had no zil log at all[18:32:54] *** _coredump_ has quit IRC[18:33:10] *** sndcrb has joined #opensolaris[18:33:11] <tsoome> just was playing with zil disable/enable[18:33:15] *** _coredump_ has joined #opensolaris[18:33:17] *** bubbva has joined #opensolaris[18:33:43] *** bahumbug has joined #opensolaris[18:34:12] <tsoome> but its was really simple test - did copy some huge avi, sequential io and not so random.[18:35:47] <tsoome> but from another hand, iscsi initiator and target in same host was rather horrible:D[18:40:42] *** alain10 has quit IRC[18:41:57] *** bnitz has joined #opensolaris[18:42:12] <crichardso> how would i go about moving zil to a diffrent device?[18:42:19] <crichardso> is it just the log command?[18:42:27] <holcomb> i think you remove the log then add the new one[18:42:28] <holcomb> never done it[18:43:38] <tsoome> i guess so yeah[18:43:50] *** div111 has quit IRC[18:43:56] <Stric> crichardso: There are some issues with separate log, like if it's missing when you boot - your pool is inaccessible etc..[18:44:03] <Stric> so do tests on a non-critical system first[18:44:16] <Stric> not sure if you can remove the separate log either.. as I said, test first[18:46:14] *** alfism has joined #opensolaris[18:46:44] *** bnitz1 has quit IRC[18:47:00] <trichobezoar> which is why its not allowed on rpool. :-([18:47:58] *** Tusk2 has joined #opensolaris[18:48:02] <Tusk2> hey[18:49:25] *** ravv has quit IRC[18:49:32] <Tusk2> what's the best way to check my system temperatures?[18:49:37] <crichardso> hmm disabling flushcache made a huge improvement[18:49:41] <holcomb> rectally[18:50:06] <holcomb> ... tough crowd[18:50:22] <trochej> Tusk2: ILOM/ALOM may be able to do this[18:50:25] <trochej> I think[18:50:37] <tsoome> they can[18:51:16] <Tusk2> trochej: on a non SUN hardware actually[18:51:28] <crichardso> if i do echo zfs_nocacheflush/W0t1 | mdb -kw is it perm accross reboots or do i need to add it to a cfg[18:51:53] <tsoome> crichardso: whAT VERSION OF OS?[18:51:58] <tsoome> oops damn caps[18:52:10] *** ry_laptop has joined #opensolaris[18:52:14] <ry_laptop> woot woot![18:52:18] <crichardso> 108[18:52:23] <crichardso> indiana[18:52:24] <ry_laptop> active/active ipmpd over vnics.[18:52:39] <ry_laptop> I love solaris.[18:52:43] <tsoome> no its not permanent[18:52:52] <ry_laptop> I want it to have my babies.[18:52:55] <tsoome> and i wouldnt really make it so.....[18:52:59] <Stric> crichardso:[18:53:15] <Stric> grmbl. sorry[18:53:21] <tsoome> if you wanna destroy your data, just install linux...[18:53:28] <ry_laptop> lol[18:53:33] *** prav33n has joined #opensolaris[18:53:43] <trochej> Tusk2: Hmm[18:53:44] <tsoome> :P[18:53:47] <trochej> Sorry, dont know then[18:53:56] <Tusk2> trochej: ok[18:54:44] <eviljames> tsoome: And make ext4 your root![18:55:06] *** magnon has joined #opensolaris[18:55:08] *** kim0 has quit IRC[18:55:12] <tsoome> i see, you have earned your name....[18:55:15] <tsoome> :D[18:55:28] <crichardso> lol i dont want to i just want my vms to stop bugging out due to the iowaits i am seeing[18:55:39] <trichobezoar> /etc/system for perm changes[18:56:13] <tsoome> crichardso: you have full disk in pool or slice?[18:56:13] <pandarilho> did anyone install opensolaris 2008.11 on virtualbox?[18:56:22] <crichardso> i had to rebuild a silly dc twice cause AD got out of sync[18:56:28] <crichardso> tsoome full disks[18:56:31] <magnon> pandarilho: sure, works fine[18:56:50] <pandarilho> magnon: how much ram did you need?[18:56:55] *** dunc__ has joined #opensolaris[18:56:55] <magnon> pandarilho: gave it 1g[18:57:07] <pandarilho> magnon: hmm, thanks[18:57:22] <magnon> I'm trying to run pkg image-update from 101b to 108 and got this: http://pastebin.com/d4f1f35d6[18:57:24] <pandarilho> magnon: i tried with 512 mb (the minimum required) but no way to install[18:57:34] <magnon> Anyone able to shed some light on my problems?[18:57:42] <tsoome> well, if you make it permanent, make sure your disks will *never* loose power.[18:58:12] <holcomb> magnon: one thing that helped me was to create the be before hand, then pkg image-update -R /path/to/mounted/be[18:58:48] <CIA-40> Reza Sabdar <Reza.Sabdar at Sun dot COM>: 6810894 OFAST NDMP: Remote restore failed via NBU, 6787737 Amber Road NDMP: Backup fails with "Out of memory" error on NDMP client, 6807614 EBS744:old file recover on NAS 7000 (amber road) over new file, only recovers ACLs, 6807326 ndmpd should use timeout on SCMD_TEST_UNIT_READY[18:59:45] <magnon> holcomb: -R before image-update, otherwise it works :) thanks, that makes introspection easier at least[19:00:27] *** jfisc1 has joined #opensolaris[19:04:54] *** ravv has joined #opensolaris[19:06:08] *** jfisc has quit IRC[19:06:34] *** cypromis has quit IRC[19:07:37] *** arcame has quit IRC[19:10:32] *** saohh has joined #opensolaris[19:10:33] *** pandarilho has quit IRC[19:13:51] *** KermitTheFragger has quit IRC[19:14:12] *** stevel has joined #opensolaris[19:14:12] *** ChanServ sets mode: +o stevel[19:21:46] <CosmicDJ> adobe reader for opensolaris very soon -> http://blogs.adobe.com/acroread/2009/03/sneak_preview_of_adobe_reader.html[19:24:20] *** saohh has quit IRC[19:27:11] *** bnitz has quit IRC[19:29:17] *** Odin- has quit IRC[19:29:46] *** coffeetime has joined #opensolaris[19:29:52] *** coffeetime has quit IRC[19:29:59] *** coffeetime has joined #opensolaris[19:30:09] <magnon> holcomb: no cigar :/[19:31:53] <holcomb> :([19:32:40] *** MattMan has quit IRC[19:32:59] *** cwebber has quit IRC[19:33:04] *** sipior has left #opensolaris[19:34:51] *** aadikr123 has joined #opensolaris[19:37:11] <e^ipi> CosmicDJ: *shrug* evince works fine[19:37:20] <e^ipi> i've never needed adobe reader for anything[19:37:29] *** morpholo has joined #opensolaris[19:37:33] <aadikr123> hi..i have connected to solaris through ssh from my laptop[19:37:41] <aadikr123> ii was trying to run a GUI application from ssh[19:37:52] <aadikr123> iit gave me a run time error[19:38:05] <aadikr123> can any one tell me how to run a graphical program through a ssh shell[19:38:05] <aadikr123> ??[19:38:27] <Auralis> it would help if you tell us the error[19:38:40] <e^ipi> aadikr123: the ssh man page can[19:38:51] *** morpholo has left #opensolaris[19:39:49] *** morpholo has joined #opensolaris[19:41:14] <morpholo> hi, i'm having some troubles with exclusive-ip zones on 2008.11, more specifically I'm experiencing a panic on zoneadm halt due to bug 6746721[19:41:28] <morpholo> the bug report says it was fixed in snv_102[19:41:47] *** jstrunk has joined #opensolaris[19:42:08] <morpholo> are there plans to release a patch for 2008.11 or do I have to track development repositories?[19:43:40] *** jstrunk has quit IRC[19:43:58] *** jstrunk has joined #opensolaris[19:45:23] *** cwebber has joined #opensolaris[19:45:25] <jstrunk> hello, how can I run a service such as pkg.depotd as a different user via SMF?[19:46:07] <jstrunk> I see the dlmgmtd is running as dladm, but the smfs properties do not show anything like that?[19:47:13] *** meshuggah has joined #opensolaris[19:47:28] <meshuggah> hello[19:47:57] <meshuggah> i got a free cd-live of opensolaris, but i cant get it boot[19:48:00] <meshuggah> someone can help me>[19:48:50] <tsoome> the computer shop can, most probably....[19:48:50] <meshuggah> it show me some line like it was booting, but it is very slow[19:49:02] <meshuggah> do you think it is because i only have 256mb ram?[19:49:12] <Auralis> its not enough ram[19:49:20] <meshuggah> they say 512 min[19:49:20] <eviljames> meshuggah: Either you're crazy or named after my favorite band. Either way 256 is not enough ram :D[19:49:35] <meshuggah> eviljames, i am crazy and i like this band[19:49:57] <timsf> jstrunk: look in the SMF docs for method_context and method_credential[19:50:11] <e^ipi> morpholo: you'll need to be on the dev repo... pkg(1) set-authority subcommand[19:50:41] <e^ipi> jstrunk: man rbac . the daemon runs with the least amount of privileges it needs[19:51:06] <eviljames> Soon to be pkg set-publisher :D[19:51:10] <e^ipi> ( assuming you see "root" and think "security hole" and that's why you're asking )[19:51:28] <e^ipi> eviljames: yes... nobody respects authority anymore[19:52:11] *** vitorio has joined #opensolaris[19:52:15] <timsf> The RIAA is a bunch of publishers though, and I'm not sure I respect them either :-)[19:52:20] <meshuggah> so anybody can give me old rambus ram :D ?[19:53:12] <meshuggah> i think of it, i even tried to load the livecd at the job, it has 512mb ram and still didnt boot there[19:53:13] <trichobezoar> Obama will soon be giving rambus handouts[19:53:19] <meshuggah> ok cool[19:53:22] <meshuggah> lol :)[19:53:42] *** Odin- has joined #opensolaris[19:53:56] <meshuggah> i think it is the asshole who us a sharp pencil to write the number 2 on the package, i can see the 2 on the cd[19:54:03] <meshuggah> who used[19:54:12] <e^ipi> what?[19:54:33] <meshuggah> someone wrote an encircled 2 on the package[19:54:50] <meshuggah> and this show on the cd[19:54:54] <meshuggah> he probably broke ir[19:55:00] <jstrunk> timsf: thank you for pointing me there.[19:55:04] <meshuggah> anyway[19:55:05] <meshuggah> thank[19:55:11] <meshuggah> have a nice day[19:55:21] <e^ipi> running on a tiny amount of ram is the problem, not a broken cd[19:55:38] <meshuggah> e^ipi, yeah but i tried with 512mb ram at the job[19:56:04] <meshuggah> and it never happened to me that siomething refused to run cause i didnt had enough ram, usually they only get really slow but run[19:56:27] <meshuggah> it is not like trying to run a 386 program on a little 286[19:56:31] <meshuggah> like dos4gw[19:56:33] *** airjump has joined #opensolaris[19:56:37] <meshuggah> i am off-topic sorry[19:56:43] <meshuggah> bye :)[19:56:47] <trichobezoar> have fun[19:56:51] <meshuggah> :)[19:56:55] <airjump> hello[19:57:00] *** meshuggah has left #opensolaris[19:57:10] <airjump> i hope the question is not off-topic[19:57:28] <airjump> run opensolaris on a pda with intel xscal cpu[19:57:28] <CosmicDJ> we'll see ;)[19:57:34] <morpholo> e^ipi: thanks, the only thing that scares me a bit is running a development version on a production machine, but since the alternative would be a crash every zoneadm halt... :)[19:58:03] <e^ipi> most people use S10 for prod[19:58:07] <e^ipi> *shrug*[19:58:20] <e^ipi> airjump: good luck with that. document it[19:58:31] <e^ipi> oh, xscale ? as in ARM? not gonna happen...[19:58:54] <CIA-40> <gerald.jelinek at sun dot com>: 6814626 p2v isn't setting suid bit on whole-root zone usr files[19:59:12] <airjump> ok[19:59:35] <e^ipi> you could port it to ARM, but that'd take a while[19:59:55] <airjump> ARM! Ok[20:00:09] *** joergb has joined #opensolaris[20:00:41] <e^ipi> in case it's not obvious, i'm being facetious ... solaris is decidedly un-portable[20:00:49] <airjump> google not help to finde some people who do opensolaris on a pda[20:01:17] *** jfisc1 has quit IRC[20:01:18] *** joergb has quit IRC[20:01:28] <e^ipi> atom pda maybe[20:01:30] <trichobezoar> how is it decidedly unportable?[20:01:30] <e^ipi> netbook, sure[20:01:40] *** jfisc has joined #opensolaris[20:02:00] <airjump> good news atom but i search now google for a atom cpu in a pda[20:02:02] <e^ipi> trichobezoar: the PPC team was working for 2 and a half years and managed to get as far as the cyclic backend in the kernel[20:02:12] <airjump> netbook i found many docs[20:02:13] <e^ipi> didn't even fork() a shell properly[20:02:33] <timsf> Not true - I ran 2.6 dev bits on an old IBM ppc[20:02:45] <timsf> but that was Solaris, and a long time ago[20:02:58] <timsf> there's the System Z port as well[20:03:19] <codestr0m> e^ipi: why is it so unportable.. do you have specific examples?[20:03:47] <morpholo> e^ipi: does that mean that it's still too early to run ON in production environments?[20:03:47] <e^ipi> codestr0m: the PPC community pages[20:03:48] <codestr0m> there's not *that* much asm in onnv-gate[20:03:59] <e^ipi> no, there are a lot of assumptions though[20:04:13] <e^ipi> if it were just asm, it'd be simple[20:04:20] <e^ipi> SPARC isn't that much different from PPC[20:04:37] <e^ipi> you could probably even write a SPARC -> PPC translation script/compiler[20:04:38] *** ruse39[home]_ has joined #opensolaris[20:04:51] <e^ipi> it's the rest of the architecture that's wierd[20:05:06] *** jstrunk has quit IRC[20:05:11] *** KOHJU is now known as kohju[20:05:36] *** cri__ has joined #opensolaris[20:05:49] <e^ipi> and the machine specific bits is all tied together in ways that make it difficult to decouple easily[20:06:23] *** dunc has quit IRC[20:06:35] *** capaz has joined #opensolaris[20:06:41] <codestr0m> e^ipi: do you mean ppc 32 bit or 64 bit as well?[20:06:53] <e^ipi> ppc32 was the target[20:07:03] <e^ipi> since that's the hardware that was kicking around[20:07:09] <codestr0m> because I'd be highly interested in a ppc to sparc comparison on this compiler project[20:07:21] <e^ipi> ask jbk[20:07:30] <e^ipi> he's an assembly wizard[20:07:45] <trochej> timsf: I vaguely recall your blogpost about some property of zpool that disables dialog Unable to mount filesystem, when dealing with mobile pools[20:07:55][20:08:17] <timsf> zpool import -R ?[20:09:52] *** mega has quit IRC[20:10:12] <airjump> thanks for help i found a pda / phone with ATOM CPU i will read the tech-specs and hope the wlan chip run with os[20:10:37] <e^ipi> it'll probably be nontrivial to fit it on there[20:10:52] *** dunc__ has quit IRC[20:10:54] <jbk> translating assembly wouldn't be too difficult, but that doesn't address the hardware aspects -- you don't just hook up a power supply to a cpu chip and have it do stuff[20:11:18] <jbk> there's your memory controller, pci bridges, etc. etc. that all need drivers[20:11:38] <CosmicDJ> well pci stuff is in sparc/x86 too, you could use them[20:11:42] *** morpholo has quit IRC[20:11:45] <airjump> yes but i don-t wont a pda with #wm[20:12:04] <trichobezoar> that's a shame[20:12:12] <jbk> CosmicDJ: no, i'm talking about manipulating the actual chips[20:12:15] <jbk> thing of a scsi card[20:12:32] <trichobezoar> so solaris doesnt have a layer to abstract this?[20:12:34] <jbk> the sd driver sends the same commands regardless of the scsi card[20:12:35] <CosmicDJ> yep, have one in my mind right now[20:12:44] <e^ipi> trichobezoar: like i said... nonportable[20:13:02] <jbk> but you still need a driver specific to that scsi card to take the scsi commands from the sd driver and get them onto the wire[20:13:58] <jbk> trichobezoar: you're assuming every pci bridge has the same programming interface[20:16:26] <trichobezoar> no, im assuming it's been abstracted at the right point[20:16:28] <CosmicDJ> well, I still think that some hardware components are shared (i.e. you find them in x86,sparc,ppc) because they're mass-produces and thus cheap...[20:16:30] <airjump> bye[20:16:33] *** airjump has quit IRC[20:20:32] *** pjfloyd has joined #opensolaris[20:20:46] *** ruse39[home] has quit IRC[20:21:05] *** derchris has quit IRC[20:21:26] *** calumb has quit IRC[20:21:40] *** derchris has joined #opensolaris[20:24:20] <jbk> possibly[20:24:42] *** Fullmoon has joined #opensolaris[20:24:48] <jbk> i can't recall what platform (that contained a PPC chip) was selected[20:24:58] *** Rocket2DMn has joined #opensolaris[20:25:01] *** Fullmoon has quit IRC[20:25:09] <jbk> but for example, a lot of Sun's servers use a sun-designed PCI bridge ASIC[20:25:45] <e^ipi> jbk: which, the ppc dev target?[20:25:49] <jbk> yeah[20:25:50] <CosmicDJ> mh sure, 10 years ago things were different... (very different)[20:26:00] <e^ipi> some weird specific thing "opendev workstation"[20:26:12] <e^ipi> which was PPC + BIOS for reasons of video cards[20:27:07] <jbk> but the whole point, it's not just a matter of having a compiler spit out PPC machine language instead of sparc or x86[20:27:09] <e^ipi> if i could convince my PS3 to boot anything other than games or linux i'd see about shoehorning it in[20:28:12] <jbk> if it or MIPS ever revivies, I can take care of the disassembler piece pretty easily :)[20:28:25] <e^ipi> seeing as how i don't understand linux's kexec that's unlikely[20:28:40] <e^ipi> ( ps3 bootloader is linux, it kexec's more linux to boot it )[20:28:45] <jbk> in fact, I could converge the sparc, mips, and ppc into a common framework due to simalarities[20:30:18] *** ruse39 has joined #opensolaris[20:30:51] <trichobezoar> https://lists.linux-foundation.org/pipermail/fastboot/2004-December/007849.html[20:31:10] *** twisti_ has joined #opensolaris[20:31:22] *** ruse39_ has quit IRC[20:32:07] *** Roms has quit IRC[20:32:34] *** Roms has joined #Opensolaris[20:36:31] *** cri__ has quit IRC[20:40:17] <ZOP> does anyone know how to disable mountd over tcp?[20:41:20] *** twisti has quit IRC[20:41:29] *** ravv has quit IRC[20:42:19] *** ravv has joined #opensolaris[20:44:03] *** tombhadAC has quit IRC[20:45:40] *** goo1 has joined #opensolaris[20:47:11] *** Mortas has quit IRC[20:48:19] *** vitorio has quit IRC[20:49:19] *** tamr has quit IRC[20:50:42] *** Jiraia has quit IRC[20:50:45] *** Jiraia has joined #opensolaris[20:53:16] *** goo1 has left #opensolaris[20:54:26] *** cypromis has joined #opensolaris[20:58:20] *** houst0n has quit IRC[21:00:00] *** `Spike has quit IRC[21:01:19] *** houst0n has joined #opensolaris[21:01:48] *** Gman has joined #opensolaris[21:01:48] *** ChanServ sets mode: +o Gman[21:02:39] <capaz> Citrix XenApp Client 8.5 won't install on OS2008.11. Any alternatives for an ICA client for OS?[21:03:52] *** kim0 has joined #opensolaris[21:04:17] *** sndcrb has quit IRC[21:04:30] *** oenone has quit IRC[21:05:05] <trichobezoar> why doesnt it install?[21:06:49] <trochej> capaz: Does it say why?[21:07:18] <fraggeln> anyone knows if SATA 300 TX4 is supported by sxce?[21:07:30] <fraggeln> (promise card that is)[21:08:16] <capaz> ld.so.1: echo_cmd: fatal: libucb.so.1: open failed: No such file or directory[21:08:16] <capaz> /export/home/tbaca/software/xenapp/./solx86/hinst: line 235: 2211: Killed[21:09:29] *** sndcrb has joined #opensolaris[21:09:49] *** sommerfeld has quit IRC[21:09:57] <capaz> in the tarball, there are some scripts and a cpio file. I haven't dug into the troubleshooting yet, hoping for a more seamless solution.[21:12:29] *** Tusk21 has joined #opensolaris[21:12:49] <capaz> ok, there's also a binary named "echo_cmd" that's actually throwing the linker error:[21:13:26] <capaz> ldd shows that libucb is the only missing lib[21:16:06] *** timsf has quit IRC[21:18:31] <pjfloyd> isn't libucb the old old old Berkeley compat lib?[21:18:44] <Auralis> yes[21:19:19] <Tusk21> how do i upgrade from snv_101b to 108?[21:19:34] *** coffeetime has quit IRC[21:19:39] <pjfloyd> liveupgrade?[21:19:43] <trichobezoar> I dont think 108 is really recommended right now[21:20:01] <trichobezoar> There are issues, although they can be worked around[21:20:05] <CosmicDJ> 101b, isn't that 2008.11?[21:20:07] *** dunc_ has joined #opensolaris[21:20:10] <pjfloyd> an 109?[21:20:28] <Tusk21> CosmicDJ: yep[21:20:46] <Tusk21> pjfloyd: yeah liveupgrade would be better[21:20:46] <CosmicDJ> then liveupgrade won't work[21:20:51] <Tusk21> oh[21:22:10] <fraggeln> Does anyone knows if a promise sata 300 TX4 is supported by opensolaris or sxce?[21:22:40] <Tusk21> fraggeln: if that's the same as mine forget about it[21:22:43] <Keso> trichobezoar: 108 is far better than 101b or 109 (at least for notebook)[21:22:44] <capaz> any hope then, of getting the libucb on OS2008.11?[21:23:04] <Tusk21> promise doesn't release any solaris driver[21:23:18] <capaz> short of that, is there any such thing as a 3rd party ICA client, or are they only done by Citrix?[21:23:19] <fraggeln> Tusk21: lets spam them and ask for a driver :)[21:23:21] <trichobezoar> Keso: oh really? I'm at 106 and I'm not moving until things settle down[21:23:43] <trichobezoar> I am pretty lucky that my desired feature, crossbow, was before all this beakage ;)[21:23:44] <Keso> trichobezoar: for me I saw performance improvement on 108[21:23:48] <fraggeln> 79b rocks! :D[21:23:51] <Tusk21> fraggeln: no thx i changed for a real motherboard and i don't need the card anymore :D[21:24:05] <Keso> trichobezoar: but 109 break suspend, it does not suspend at all[21:24:21] <Tusk21> Keso: where do i get that 108?[21:25:05] <Keso> Tusk21: pkg.opensolaris.org/dev[21:25:12] <Tusk21> and anyone using pkgsrc here??[21:25:20] <Keso> Tusk21: add this as repository and upgrade[21:25:56] <CosmicDJ> Tusk21: http://sites.google.com/site/solarium/how-to-update-opensolaris[21:26:04] <Tusk21> Keso: i did[21:26:17] <Tusk21> but it doesn't get me to 108 :([21:26:36] *** Roms has quit IRC[21:26:58] <Tusk21> i already did image-update[21:27:05] <CosmicDJ> and you got?[21:27:13] <Tusk21> 101b after reboot[21:27:15] <CosmicDJ> read the page again[21:27:28] *** syamajala has joined #opensolaris[21:27:31] <Keso> Tusk21: type beadm list[21:28:08] <Tusk21> oh[21:28:39] <Tusk21> i have to install entire first[21:28:43] <Tusk21> mmmh ok cool thx[21:29:49] *** Tusk2 has quit IRC[21:30:02] <IvanR_> capaz: libucb is in the SUNWscp package, "pkg search -r libucb.so.1"[21:32:07] *** aadikr123 has quit IRC[21:32:52] *** tombhadAC has joined #opensolaris[21:33:11] <Tusk21> CosmicDJ: pfexec pkg install entire at 0 dot 5.11-0.108 << it's saying that package didn't match...[21:34:05] <IvanR_> Tusk21: What does "pkg authority" say?[21:35:03] *** bnitz has joined #opensolaris[21:36:46] <IvanR_> df[21:37:01] <IvanR_> blah, wrong focus[21:37:13] *** master_of_master has quit IRC[21:37:55] <Tusk21> IvanR_: well displaying me the authorities with my Dev one[21:39:11] *** RavenSlay3r has quit IRC[21:39:58] <capaz> IvanR: thanks![21:40:21] *** master_of_master has joined #opensolaris[21:41:27] *** sactodave has left #opensolaris[21:42:08] <IvanR_> Tusk21: Is Dev the preferred repo?[21:42:26] <Tusk21> ah it must be?[21:42:32] <Tusk21> damn[21:42:56] <IvanR_> Yes[21:43:52] *** xRaich[o]2x has quit IRC[21:44:49] <Tusk21> while i'm checking this, anyone is using pkgsrc on OS ?[21:47:28] <Tusk21> IvanR_: no more luck with it prefered[21:48:25] <Tusk21> i did pfexec pkg refresh && pfexec pkg install entire at 0 dot 5.11-0.108[21:48:34] <Tusk21> and no package found[21:48:50] *** simNIX has joined #opensolaris[21:49:46] *** axisys has quit IRC[21:50:51] <seanmcg> Tusk21, if you want to update to b108 then try pkg image-update, that will create a new bootenvironment for you, then just reboot[21:51:26] *** cwebber has quit IRC[21:52:11] *** popcornPanic has joined #opensolaris[21:52:33] <Tusk21> seanmcg: as i said, i already did it and it just got me to snv_101b[21:52:56] *** bnitz has quit IRC[21:53:18] *** nsuperbus has quit IRC[21:53:27] <popcornPanic> Hi i am trying to create a branded zone and when i type 'create -t SUNWlx' it returns SUNWlx: No such zone configured[21:54:26] *** simNIX has left #opensolaris[21:54:53] <popcornPanic> nvm. i misread a line.[21:55:29] <bda> :)[21:55:47] <victori> does zfs do raid10 in software?[21:56:04] <Auralis> zfs does everything in software, so yes[21:56:19] <seanmcg> Tusk21, is the output of pkg authority '../release' or '../dev' ?[21:56:59] <victori> Auralis: is that "raidz" ?[21:57:01] *** bnitz has joined #opensolaris[21:57:07] <Auralis> no[21:58:01] <capaz> argh. Citrix ICAClient wants motif. :-([21:58:30] <Auralis> capaz: grab the motife pkgs from the standard solaris 10 media, or the SXCE[21:59:10] <capaz> Auralis: thanks for the tip.[21:59:12] <seanmcg> victori, raidz is akin to raid5 - but without the hole.[21:59:15] <Tusk21> seanmcg: nope[21:59:31] <victori> seanmcg: would that be ideal for a database setup? or stick to raid10?[21:59:35] <seanmcg> nope ? so its just 'http://pkg.opensolaris.org' ?[21:59:51] <Tusk21> yes[21:59:57] <Tusk21> and /dev[22:00:05] <seanmcg> victori, have a read on solarisinternals.com - the zfs section for some database + zfs tips.[22:00:10] <Tusk21> http://pkg.opensolaris.org/dev i meant[22:00:13] <victori> thanks[22:01:05] <seanmcg> Tusk21, odd, the /dev has the latest.. b108 at the moment.[22:01:07] *** tombhadAC has quit IRC[22:01:44] <Tusk21> i did the refresh[22:02:11] <Tusk21> but when doing a search it only displays the 101[22:02:18] *** popcornPanic has quit IRC[22:02:51] <seanmcg> Tusk21, can you pastebin the output from pkg authority ?[22:02:51] <Tusk21> sure[22:04:46] <Tusk21> http://pastebin.ca/1358508 here it is[22:08:07] <seanmcg> Ah, finally pastbin.ca loads. I see the issue (me thinks). Remove the Dev and change the opensolaris.org to be prefered and point it to http://pkg.opensolaris.org/dev[22:08:21] <Tusk21> oh[22:08:38] <Tusk21> lemme check[22:09:16] <Tusk21> how do you change preferred by command line?[22:09:53] <Tusk21> that graphic interface is way too slow :)[22:10:20] <victori> anyone know how how to setup raid-10 in zfs?[22:10:48] <Auralis> zpool create c0t0d0 c0t1d0 mirror c0t2d0 c0t3d0 for example[22:11:05] <victori> Auralis: thanks[22:11:19] <seanmcg> Tusk21, something like: pkg set-authority -P -O http://pkg.opensolaris.org/dev opensolaris.org[22:11:39] <seanmcg> man pkg for info[22:11:41] <Tusk21> ah ok you just set it back[22:11:58] <Tusk21> yeah yeah i just thought it would add a new one[22:12:38] <victori> Auralis: do you happen to know if you can convert a zfs mirror to raid10?[22:13:28] <Auralis> not 100% positiv, look through the man page, the attach subcommand is what you are looking for[22:13:40] <victori> Auralis: thanks for all the help[22:14:45] <Tusk21> seanmcg: still the same :( ???[22:14:55] *** twisti__ has joined #opensolaris[22:15:01] <seanmcg> did youu remove the Dev one ?[22:15:35] *** Kitty_ is now known as Kitty[22:16:38] *** pjfloyd has left #opensolaris[22:16:57] <Tusk21> sure[22:17:00] *** cemerick has quit IRC[22:17:04] <Tusk21> i let the contrib one btw[22:18:30] *** cwebber has joined #opensolaris[22:18:53] * Tusk21 lost[22:18:55] *** bubbva has quit IRC[22:21:28] *** sndcrb has quit IRC[22:22:10] <trygvis> anyone know how well this is supposed to work with opensolaris? http://www.supermicro.com/products/motherboard/Xeon1333/5400/X7DWA-N.cfm[22:22:36] <trygvis> I got one besides me, and I've installed opensolaris but I keep on getting io errors from the disc controller[22:24:58] *** bahumbug has quit IRC[22:26:04] *** twisti_ has quit IRC[22:26:19] *** BillyBop has joined #opensolaris[22:27:11] <BillyBop> Hi, I would like to install a version of solaris on my old sparc64 but I don't know which distribution I can choose[22:27:59] <BillyBop> any help in making my choice would be appreciated[22:28:19] <Auralis> define old[22:28:25] <jmcp> is that a sun4u system you're talking about?[22:28:32] <BillyBop> hum 500Mhz, no other information[22:28:45] <jmcp> you don't even know what model it is?[22:28:48] <BillyBop> it's a blade 100 workstation[22:28:51] <jmcp> ah, ok[22:28:54] <jmcp> that's sun4u[22:28:59] <jmcp> which is the architecture[22:29:04] <BillyBop> ok[22:29:10] <jmcp> whenever I see "sparc64" I think "linux has been here"[22:29:12] <jmcp> anyway[22:29:23] <jmcp> Solaris Express Community Edition is what you need at this point in time[22:29:24] <BillyBop> yes, and freebsd and openbsd ;)[22:29:25] *** hsp has quit IRC[22:29:26] <jmcp> aka SXCE[22:29:46] <jmcp> you will be hamstrung, somewhat, by the IDE bus in the blade100[22:29:54] <jmcp> and the motherboard limit on installable ram[22:29:59] <BillyBop> hamstrung?[22:30:01] <bda> Hamsterstrung.[22:30:09] <BillyBop> I'm french , sorry, what does that mean?[22:30:32] <jmcp> BillyBop: it's an English expression meaning "impeded"[22:30:34] <bda> Restricted or crippled.[22:30:37] <jmcp> that too[22:30:42] <BillyBop> ok[22:30:43] <jbk> limited[22:30:45] <jbk> :)[22:30:47] <jmcp> I'd run it headless and ssh into it as required[22:30:53] <bda> Hamsterstrung is similar, only furrier.[22:30:53] <jmcp> gnome will run *sloooooooowly* on it[22:30:58] * jmcp snorts[22:31:03] <bda> :)[22:31:06] <BillyBop> but will I have a lot of difficulties to install SXCE?[22:31:11] <jmcp> no[22:31:22] *** syd`_ has joined #opensolaris[22:31:24] <jmcp> I suggest using the text-mode installer rather than the gui installer[22:31:31] <BillyBop> ok[22:31:53] *** radsy has joined #opensolaris[22:32:06] <BillyBop> it's for a server, so no GUI. what's the problem with the amount of RAM?[22:32:08] <jmcp> BillyBop: the textmode installer also gives you the option of using ZFS for your root 'partition' --> choose that option[22:32:17] * jmcp checks the limits[22:32:41] <bda> BillyBop: ZFS loves RAM.[22:32:47] <jmcp> max of 2Gb[22:33:05] <jmcp> the cpu's cache (UltraSPARC-IIe) is also quite small[22:33:16] <jmcp> ZFS loves address space, and then it loves ram[22:33:25] <jmcp> so having 64bit CPU is an excellent start[22:33:51] *** proberts_ has quit IRC[22:33:51] <BillyBop> what's the best for me (1Gb of RAM), ZFS or not ZFS?[22:33:52] <jmcp> then you should add up to 4 PC133 168pin 512Mb dimms -- but they need ECC[22:33:56] <jmcp> ZFS[22:34:10] <jmcp> just make sure you tune the ARC _down_[22:34:17] <BillyBop> uh?[22:34:22] <jmcp> http://www.solarisinternals.com/wiki/index.php/ZFS_Evil_Tuning_Guide will tell you how to do it[22:34:31] <jmcp> ARC is part of the intelligence that ZFS has[22:34:46] <jmcp> http://www.solarisinternals.com/wiki/index.php/ZFS_Best_Practices_Guide << read this before the evil tuning guide[22:34:51] <bda> It shouldn't be that big of a deal.[22:35:01] <bda> I've run GUI on 1GB zfsroot and it was ok.[22:35:06] <bda> With no ARC pinning.[22:35:16] <jmcp> bda: I bet you had more MHz than just 500[22:35:19] <dguitar> BillyBop, I have a SunBlade 100, and OSOL would not install on it[22:35:19] <jmcp> and more cpu l2 cache[22:35:23] <jmcp> BillyBop: http://docs.sun.com/app/docs/doc/817-2271 << ZFS admin guide[22:35:26] <jmcp> very good to read[22:35:28] <bda> jmcp: haha, for sure. :)[22:35:40] * bda shuts up. :)[22:35:42] <jmcp> heh[22:35:50] <Tusk21> seanmcg: could it be long to "create plan" on pkg install entire@xxxxxx[22:35:50] *** rv- has joined #opensolaris[22:35:52] <trygvis> yay, zfs corruption[22:35:56] <jmcp> dguitar: OpenSolaris binary distro, or SXCE ?[22:36:08] <trygvis> box crashed while attaching a mirror, now it won't even boot[22:36:08] <dguitar> jmcp, Distro[22:36:21] <jmcp> dguitar: then that's correct, OSOL doesn't install on sun4u _at this time_[22:36:28] <jmcp> dguitar: I've already suggested SXCE instead[22:36:40] <BillyBop> dguitar: you said that OSOL does not install on a blade 100?[22:36:41] <jmcp> BillyBop: if you need to buy more ram for this box, look for part number 370-4151[22:36:51] <dguitar> SXCE didn't install either[22:36:52] <jmcp> BillyBop: there is a difference between OSOL binary distro and SXCE[22:36:54] <BillyBop> jmcp: no money[22:37:10] <jmcp> BillyBop: well, if you get some money, that's the part number you want to buy :)[22:37:11] <dguitar> Although I didn't 'really' try tbh..[22:37:28] *** scoffin has joined #opensolaris[22:37:35] <jmcp> dguitar: if you didn't 'really try' and didn't log bugs about it, why muddy the waters and spread FUD about it?[22:38:00] <BillyBop> jmcp: would perhaps keep my money for some x86 config :)[22:38:08] <jmcp> BillyBop: that might actually be more useful :)[22:38:12] *** syd` has quit IRC[22:38:14] <jmcp> anyways....[22:38:38] <dguitar> jmcp, No fud.[22:38:46] <seanmcg> Tusk21, I wouldn't do it that way, pkg image-update should be the way to go to upgrade.[22:39:06] <BillyBop> does SUN still produce good processors ? (not a troll, I really don't know)[22:39:09] <Tusk21> i'm following this way[22:39:10] <Tusk21> http://opensolaris.org/os/project/indiana/resources/relnotes/200805/image-update/[22:39:22] <jmcp> BillyBop: Sun is still designing and producing UltraSPARC processor[22:39:23] <jmcp> s[22:39:31] <bda> T2++[22:39:37] <jmcp> in fact, the verilog RTL for them is now free --> http://www.opensparc.net[22:39:45] <trichobezoar> verilog[22:39:47] <Tusk21> installing entire SUNWipkg and SUNWinstall-libs before image-update[22:39:55] <jmcp> BillyBop: and that's *GPLv2* free, as well[22:40:07] <seanmcg> Tusk21, ya, its a good idea that actually..[22:40:20] <BillyBop> ok, good philosophy[22:40:31] <trichobezoar> that means all the code it runs becomes gpl too right? ;P[22:40:40] * jmcp glares @ trichobezoar[22:40:40] <seanmcg> Tusk21, I see the dev repo off pkg.o.o has b109 bits now...[22:40:45] <jmcp> no, the FSF haven't tried that aspect[22:40:59] <Tusk21> yeah i saw too in my catalog file[22:41:10] <jmcp> BillyBop: any time somebody says "oh, why are you running a proprietary CPU architecture (sparc)?" I can say "it was never proprietary, it was always a publicly available standard. And now you can build your own if you want "[22:41:22] <BillyBop> are there any electronics guru trying to improve the GPLed verilog ? :)[22:41:28] <jmcp> dunno[22:41:38] <jmcp> BillyBop: it's not like CPU design is an easy thing to do[22:41:57] *** lenzi has joined #opensolaris[22:42:02] <jmcp> Sun, Fujitsu and lots of other organisations have plenty of veryvery smart people working cpu design already[22:42:18] <Tusk21> seanmcg: i just find the Create plan phase ultra long....[22:42:25] *** Cass has joined #opensolaris[22:42:46] <lenzi> hello... can someone please give a a clue about opensolaris version 108????[22:42:57] <jmcp> lenzi: define which clue you want[22:43:12] <BillyBop> well, thank you for your help, I'll certainly come here again for asking some silly questions :) see you all[22:43:12] <Tusk21> ok normal to be long :D[22:43:19] <seanmcg> Tusk21, it may be finding what needs and generating a list.. 'entire' as the name suggests can be rather large perhaps..[22:43:32] <tsoome> BillyBop: man google really:D[22:43:38] <Tusk21> yep yep i get it now[22:44:24] <BillyBop> tsoome: still I've no friends, I prefer come in a channel ask real people :D[22:44:51] <lenzi> I just boot from the cd osol-09-06-108-global-x86.iso, the cd boots, chooses the keyboard, (us) the language(us-english) and than stops in the login prompt, no graphical interface....[22:45:19] <lenzi> if I boot from the 0811 cd it starts ok and I can install it on disk using the same machine.[22:45:48] <jmcp> BillyBop: you haven't asked any silly questions[22:46:02] *** syd` has joined #opensolaris[22:46:05] *** koan has quit IRC[22:46:06] <jmcp> just remember to ask questions in here if you've got an issue with OpenSolaris in any flavour[22:46:13] *** cwebber has quit IRC[22:46:32] <lenzi> If I enter with user jack password jack, everything is there, the Desktop, the install icon... with the command gui-install.[22:46:54] <BillyBop> an issue==bug ? or an issue == something I can't achieve by myself ? thats a lot different[22:46:56] *** rv- has quit IRC[22:47:50] <jmcp> BillyBop: both[22:49:32] <lenzi> the question: what is the way to install opensolaris version 108 ????[22:49:47] <jmcp> lenzi: you've got osol2008.11 installed, correct?[22:50:06] <lenzi> Yes. this verwsion 2008.11 works ok...[22:50:10] <lenzi> in the same machine.[22:50:15] <jmcp> so use the pkg command[22:50:20] <jmcp> I think it's pkg image-update[22:50:26] <jmcp> you could try reading the doco about it[22:51:05] *** syd` has quit IRC[22:51:55] <BillyBop> damn ! I've no dvd rom player on my blade 100, May I use build 98 and upgrade without problems?[22:51:56] <lenzi> Hummm.... I came from the FreeBSD. I used Solaris on sparcs some years ago... now I have to move several server so opensolaris.. (about 100 servers, so I must understand the software).[22:52:16] <Tusk21> lenzi: http://opensolaris.org/os/project/indiana/resources/relnotes/200805/image-update/[22:52:23] *** syd` has joined #opensolaris[22:52:26] <Tusk21> what i'm using right now :D[22:52:26] <jmcp> BillyBop: yes - again, textmode install of 98, use ZFS for root, then to upgrade use LiveUpgrade[22:52:38] <BillyBop> ok[22:53:19] *** mikefut has quit IRC[22:54:53] *** LordIllpalazo has joined #opensolaris[22:54:54] <IvanR_> lenzi: Is this an older machine, or has little ram?[22:55:26] *** Cass1 has joined #opensolaris[22:55:52] <lenzi> Ok... excuse me if I am being "novice" but in the question of booting the 108, and get a console login, if I log in with jack, and than a su root, it works, when I try to start the gnome: gnome-session, it of course complains about dbus and hal. The question is: when I am in the login prompt, what is the command to start the whole thing (the dbus, hal, and then gnome-session) the same way opensolaris 2008.11 does????[22:56:23] <lenzi> the machine is new, an AMD X2 2gb , 160gb of disk nvidia 9600.[22:56:41] *** niq has joined #opensolaris[22:56:41] <lenzi> the network interfaces are nge0 and rtls0.[22:58:21] *** syd`_ has quit IRC[22:59:15] <lenzi> Here I am in the project of building opensolaris in servers and notebooks (core2 duo,2gb of ram...) I must customize the OS in order to enable the replication of the disk "by the dozen" about 1500 notebooks...and 30 servers...[23:00:23] <lenzi> so I must fully understand how thing is done... I have done with freebsd..[23:00:33] <Tusk21> is there a link sowhere of best practice concerning partitioning on OS ?[23:01:29] *** magnon has quit IRC[23:02:41] <jmcp> lenzi: you probably want to look at the Distro Constructor[23:02:45] <jmcp> Tusk21: "use ZFS"[23:02:59] <jmcp> lenzi: lemme find you a link[23:03:05] <seanmcg> lenzi, for replication or install of so many machines you may want to look at the caiman project (network install).. and what jmcp says[23:03:54] <jmcp> http://www.opensolaris.org/os/project/caiman, look at Current Projects[23:03:57] <lenzi> thank you... I have studying the "bulk" install of solaris that uses a "master" and than a small "slave" of about 120mb,[23:04:31] *** magnon has joined #opensolaris[23:04:47] <lenzi> but it is too slow... I am target in replicating the disk using a pen-drive boot, automate fdisk, zfs partigion, and than a tar via rsh.[23:04:53] *** syd`_ has joined #opensolaris[23:05:12] <Tusk21> jmcp: ok you just get all the disk and then logical parts using zfs[23:05:20] <Tusk21> sounds good to me[23:05:20] *** oenone has joined #opensolaris[23:05:21] <lenzi> may be a boot using pxe[23:05:35] <seanmcg> lenzi, one could possibly use the pen drive to boot off, and zfs send/zfs recieve to the target disk[23:06:00] *** syamajala has quit IRC[23:06:15] <jmcp> Tusk21: : http://docs.sun.com/app/docs/doc/817-2271 << ZFS admin guide[23:06:28] *** Cass has quit IRC[23:06:59] <seanmcg> or boot of the network and zfs send / recieve from there too[23:07:44] <lenzi> hummm... good idea... does zfs sends the whole zpool???? that will make things very easy... I think it is a question of fixing the initial grub only...[23:07:46] *** koobcam has joined #opensolaris[23:07:48] <koobcam> hola all...[23:08:05] *** Cass has joined #opensolaris[23:08:09] <seanmcg> lenzi, i think you need a zpool to send to thought..[23:08:39] <seanmcg> and zfs send, sends a snapshot.[23:08:43] <koobcam> can someone point towards an easy way to programmatically (even if it's shell, i guess) grab the Solaris Zone information (for example, what 'prstat -Z' gives u)[23:08:57] <koobcam> (id, nproc, swap, rss, memory, time, name)[23:09:47] <lenzi> Ok... I will try to install a opensolaris 2008.11 in a machine, than make a pen-drive, boot the pen-drive in another, the pendrive installs the grub, make the zpool (the entire disk) and than a rsh command will bring the zpool to the new machine... what do you think?????[23:10:06] <seanmcg> eg, zpool create rpool <disks>; (rsh $remote; zfs send $remote_zfs | zfs recieve rpool/ROOT/opensolaris) - or to that effect[23:12:15] <seanmcg> timf uses zfs send/recieve when upgrading his little laptop, there may be some info on his blog: blogs.sun.com/timf[23:12:17] <lenzi> good... what I have to do is build a master opensolaris with everything working.. I am having some problems in the Xorg that is for example some command is missing: xeyes, I have to build the entire libxine (with xine) and gstreamer with mplayer, mencoder, and ffmpeg...[23:12:38] <lenzi> I notice that there are 2 compilers (the SUN) and the gcc..[23:13:31] <seanmcg> <insert usual warning about legals issues around using mplayer ffmpeg etc>[23:13:40] <lenzi> What do you people recommend to build the software on top of: I need to build the wxgtk, mono 2.2, gnome-subtitles, and dvdstyle, and avidemux2 ....[23:14:33] *** dsch04 has quit IRC[23:15:03] <lenzi> no problem with legal issues here... the notebooks are for students, in universities, no commercial use,.[23:15:29] *** dsch04 has joined #opensolaris[23:16:05] <lenzi> I must build the ekiga, asterisk, and the g792/g723/ilbc codecs too.[23:16:08] *** dsch04 has quit IRC[23:16:51] *** syd` has quit IRC[23:17:04] *** nachox has joined #opensolaris[23:17:17] <seanmcg> lenzi, some of that software may be buildable via the jds common-build-environment, more at opensolaris.org/os/project/jds[23:17:31] *** jfisc has quit IRC[23:17:36] <nachox> evening[23:17:45] <lenzi> good...[23:18:20] <lenzi> In FreeBSD I used the ports..[23:18:36] *** pjd has quit IRC[23:18:37] *** Teltariat has joined #opensolaris[23:18:41] <Teltariat> Greets folks[23:18:54] <lenzi> Thank you people... it helped a lot... now one more quetion:[23:19:06] <Teltariat> What kind of utilities or tools does the Opensolaris use to help create packages?[23:20:15] *** jfisc has joined #opensolaris[23:20:24] <Teltariat> the Opensolaris *community[23:20:37] <seanmcg> lenzi, the jds build env, uses spec files, these hold the info where to pull the src, what patches, what compile options etc. works quite nicely.[23:21:15] *** Cass2 has joined #opensolaris[23:22:10] *** Cass1 has quit IRC[23:22:30] <lenzi> I must build a "thin client" enviroment I am thinking in build the thin client OS using FreeBSD 32bits. this small OS (about 200mb) with the Xorg hal and dbus will sit in the opensolaris file system, a setup of tftpserver in the opensolaris, wil transfer the pxeboot to the thin client (amd geode 64mb of memory, no hd). the pxe code mounts the nfs in the opensolaris machine, boots the FreeBSD kernel, starts dbus, hal and enter X -query opensolaris serve[23:22:31] <lenzi> r (running multi 64bits cpu)...[23:23:05] <Teltariat> seanmcg: the spec files are for IPS packages. For Solaris 10 proper without IPS, I would still have to use pkgmk, right?[23:23:36] <lenzi> Do you have an better aproach for the thin client solution??? will be about 200 thin clients for each opensolaris server.[23:23:39] <seanmcg> yes, for solaris 10 and older uses SRV4 packageing, so pkgmk there[23:23:41] *** piwi has joined #opensolaris[23:24:00] <seanmcg> lenzi, theres always sunray :)[23:24:32] <Teltariat> seanmcg: are there any convenience tools that build on top of the pkg* tools that help make it easier to make SVR4 packages?[23:24:55] <lenzi> I know... but here is too expensive... (it is marvelous... incredible ...) I can buy a thin client here for 200 dollars.... brand new....[23:25:02] <Teltariat> seanmcg: also, where does one find or get access to the IPS Build Environment that knows how to work with spec files?[23:25:05] <seanmcg> not that I know of, sorry. someone else here may though..[23:25:41] <Teltariat> I have OpenSolaris on my laptop and would not mind spending time to build packages if the need arose[23:26:32] <seanmcg> Teltariat, see opensolaris.org/os/project/jds for more on the jds cbd with spec files.[23:26:52] <seanmcg> s/cbd/cbe/ Common Build Env[23:27:20] <Teltariat> Thank you seanmcg[23:27:40] <seanmcg> Teltariat, well with opensolaris the old package system isn't used, see o.o/os/project/pkg for the new way[23:27:46] *** koan has joined #opensolaris[23:29:38] <seanmcg> wiki.sun.com also has some good info on createing pkgs with the new (IPS) package system[23:31:27] *** Cass has quit IRC[23:32:22] *** koobcam has left #opensolaris[23:32:23] *** coffeetime has joined #opensolaris[23:32:25] *** dsch04 has joined #opensolaris[23:32:53] *** Cass2 has quit IRC[23:34:12] *** dsch04 has quit IRC[23:34:18] <lenzi> Ok people thanks for the help... must go[23:34:19] *** dsch04 has joined #opensolaris[23:35:16] *** dsch04 has quit IRC[23:35:34] <seanmcg> same here, later folks.[23:39:00] *** lenzi has quit IRC[23:39:32] *** dsch04 has joined #opensolaris[23:40:01] *** dsch04 has quit IRC[23:40:14] *** Alasdairrr has quit IRC[23:40:26] *** dsch04 has joined #opensolaris[23:41:05] <ry_laptop> *sigh*[23:41:17] <ry_laptop> looks like seamless mode is broken for snv_101b guest ( nevada )[23:41:40] <dsch04> Hrm, my 2008.11 system won't boot[23:42:26] <ry_laptop> heh[23:42:35] <dsch04> I've turned off the graphical boot and can see it gets as far as "Hostname: hero" but there it seems to hang[23:43:36] *** codestr0m has left #opensolaris[23:43:52] <dsch04> Any ideas how to diagnose further?[23:43:56] <CosmicDJ> I wouldn't call that "won't boot" ;)[23:44:05] <dsch04> Heh[23:44:11] <dsch04> What would you call it?[23:44:19] <dsch04> won't complete startup?[23:44:29] <CosmicDJ> hangs...[23:44:36] <CosmicDJ> how long did you wait?[23:44:40] <dsch04> Hrm, my 2008.11 system hangs[23:44:54] *** lesterc has joined #opensolaris[23:44:57] <dsch04> 5 mins (ish) so far[23:45:07] *** Tusk2 has joined #opensolaris[23:45:16] *** webmink has joined #opensolaris[23:45:26] <dsch04> It did it once before (last time I rebooted)[23:45:27] <CosmicDJ> can you ping your box? can you ssh in?[23:45:28] *** codestr0m has joined #opensolaris[23:45:51] <dsch04> I seem to recall I just waited and eventually it came up, but I don't remember it taking this long[23:45:56] <dsch04> Let me try pinging...[23:46:06] *** Deiu has joined #opensolaris[23:46:10] <Deiu> Hello![23:46:24] <dsch04> I can ping it[23:46:29] <dsch04> But not ssh[23:46:43] <dsch04> (I'm not sure I have sshd enabled)[23:48:13] *** NoFX_SBC has quit IRC[23:48:29] *** Topdeck has joined #opensolaris[23:48:50] <CosmicDJ> hm...[23:48:52] <Deiu> I'm trying to build a network tool on opensolaris using libnet. During compilation I get the following error: Undefined symbol inet_ntop first referenced in file /usr/lib/libnet.so. Any ideas on how to solve this issue (there are more than one reference)[23:49:27] <CosmicDJ> Deiu: man inet_ntop[23:49:32] *** cwebber has joined #opensolaris[23:49:44] <Deiu> If I try to check the man pages for the functions, it works. So I guess they are correctly installed. Also, libnet was installed using the package manager.[23:49:59] <jmcp> add -lsocket -lnsl[23:50:25] <CosmicDJ> thats right under "SYNOPSIS" in the manpage..[23:50:47] <Deiu> Yeah, sorry.[23:50:54] <Deiu> I feel like an idiot now.[23:50:57] <Deiu> Thanks guys![23:51:05] <CosmicDJ> you're welcome[23:51:40] <dsch04> Still no progress...[23:51:47] *** Auralis_ has joined #opensolaris[23:53:19] <dsch04> Woot![23:53:26] <dsch04> Progress...[23:54:00] *** dsch04 has quit IRC[23:56:32] *** comay changes topic to "OpenSolaris Annual Meeting is on #opensolaris-meeting | SXCE 109, ON 109, IPS 109 || Step one: see if SAG answers your question: http://docs.sun.com/app/docs/coll/47.16 || The answer to "I do $foo on *ux, how do I do it on Solaris" is http://bhami.com/rosetta.html || Clipboard: http://www.pastealacon.com/"[23:56:47] *** sstallion_work has quit IRC[23:57:09] <nachox> comay, what's the opensolaris annual meeting? :)[23:57:54] *** tosh has joined #opensolaris[23:57:55] <seanmcg> ogb elections[23:58:10] *** dsch04 has joined #opensolaris[23:58:40] <dsch04> Aha[23:58:48] <CIA-40> Jim Dunham <James.Dunham at Sun dot COM>: 6794624 SCSI-3 PGR data for LUs whose backing store is a ZVOL, need to persist data in a ZVOL attributes, PSARC/2009/168 iSCSI Target PGR directory for ZVOLs[23:58:52] * dsch04 back on my OpenSolariis box