[00:01:31] *** pablo_costa has quit IRC
[00:03:14] *** pablo_costa has joined #opensolaris
[00:05:34] <pablo_costa> hi all
[00:05:35] <pablo_costa> !
[00:07:39] *** solarisjon has joined #opensolaris
[00:14:24] *** jwtodd has quit IRC
[00:16:16] *** koolniczka has quit IRC
[00:19:57] *** jmcp has joined #opensolaris
[00:27:40] *** pablo_costa is now known as solaris_user
[00:30:13] *** jcsmith has joined #opensolaris
[00:34:33] *** mrdeviant has joined #opensolaris
[00:35:57] <boyd> Morning, all
[00:36:56] *** Error_404 has quit IRC
[00:38:02] *** Gman has joined #opensolaris
[00:44:48] <jmcp> morning
[00:45:28] <Gman> hey jmcp
[00:45:55] *** mrdeviant has quit IRC
[00:47:30] <jmcp> how's the world?
[00:47:37] <boyd> Hey Gman, jmcp
[00:47:38] <hali> cold and cruel
[00:47:59] <jmcp> aww
[00:48:56] <Gman> howdy boyd
[00:49:11] * Gman just back from a short weekend in sydney, so body clock is a bit fucked up again
[00:49:26] <jmcp> Gman: you should have emailed - would have been great to catch up with you
[00:49:29] * jmcp sulks
[00:49:30] * boyd nods. I get that from Sydney too :)
[00:49:44] <jmcp> boyd: you're just soft :)
[00:49:49] <boyd> Hehe
[00:50:56] <Gman> jmcp, yeah, unfortunately i was fully involved all weekend - on the paper review committee for linux.conf.au in jan
[00:51:09] <Gman> otherwise i'd totally have caught up
[00:51:48] * jmcp nods
[00:52:00] <jmcp> ok, fair enough .... but only just
[00:52:43] <jmcp> Gman: you're on the committee?
[00:52:56] <Gman> yep
[00:53:17] <jmcp> cool
[00:53:37] <jmcp> gonna have another quick trip across before the conf?
[00:53:40] <Gman> 350 papers to get down to 70 :(
[00:53:45] * jmcp whistles
[00:53:48] <Gman> jmcp, might well do - if i do, i'll let you know
[00:54:53] <jmcp> cool
[00:56:12] *** solarisjon has left #opensolaris
[00:59:59] *** deedaw has quit IRC
[01:00:06] <jmcp> wow ... a bloke who used to be a CPRE intern just walked in as a new contractor here
[01:01:44] *** stabet has joined #opensolaris
[01:12:21] *** delewis has quit IRC
[01:15:06] *** delewis has joined #opensolaris
[01:15:08] *** coffman has quit IRC
[01:15:28] *** vigos has joined #opensolaris
[01:17:03] <solaris_user> anyone has problem with gnome fonts ( dt freeze when starting)?!?!?!?!?!?
[01:22:02] *** Error_404 has joined #opensolaris
[01:34:28] *** delewis has quit IRC
[01:42:02] *** deather_ has joined #opensolaris
[01:42:15] *** vigos has quit IRC
[01:47:02] *** Odin- has quit IRC
[01:55:32] *** Cybernd has quit IRC
[01:55:52] *** Odin- has joined #opensolaris
[01:57:30] *** kb7sqi has quit IRC
[01:59:21] *** aTypical has joined #opensolaris
[02:00:11] *** deather has quit IRC
[02:01:26] *** swa_mobil is now known as bar
[02:01:48] *** bar is now known as swa_mobil
[02:16:43] *** lisppaste3 has quit IRC
[02:25:59] *** solaris_user has quit IRC
[02:26:53] <aTypical> Anyone awake in here?
[02:27:17] <jamesd> no he went on a keg run a while ago
[02:27:37] <aTypical> :-)
[02:29:36] <jmcp> aTypical: somebody or nobody might be awake though
[02:29:44] <jmcp> depends on what your perspective is
[02:29:54] <jmcp> ?
[02:30:02] <aTypical> It's usually upside-down.
[02:30:59] *** Odin- has quit IRC
[02:32:12] <aTypical> jmcp, I'm whining on #solaris how I'm bored with Linux and Windows.  I want to try something else.  I was just wondering if other non-technical people are running Solaris on their laptops and what their success was.
[02:33:31] <jamesd> do you always walk into to sears and  ace hardware and whine that your hardware and screwdrivers are boring?  computers are tools...
[02:34:06] <aTypical> james_, I don't always, but often.
[02:34:20] <jamesd> how aTypical ...
[02:34:27] <aTypical> :-P
[02:34:34] <Kitty> aTypical: have you considered taking up golf ?
[02:34:51] <aTypical> I play golf, tennis, guitar, and piano.
[02:35:23] <Kitty> urgh, I hate it how sarcasm just doesn't project properly in text based medii
[02:35:52] <aTypical> I really didn't mean to piss off the group.  I'm sorry if I did.  I just thought that solaris might be a fun OS to work with and was wondering how others were using it.
[02:36:27] <Kitty> I run solaris on my coffee table and my foot rest
[02:36:34] <aTypical> :-)
[02:36:46] <Kitty> and I am deadly serious
[02:37:29] <jamesd> http://video.google.com/videoplay?docid=-8563337315465387385&q=label%3Aweird+al
[02:40:35] *** aTypical has left #opensolaris
[02:43:11] *** Odin- has joined #opensolaris
[03:01:51] <jbalint> is there a dbx option that can be set to print large structs?
[03:23:03] *** stabet has left #opensolaris
[03:23:18] <Gman> OpenSolaris Weekly News #32 - http://mail.opensolaris.org/pipermail/opensolaris-announce/2006-October/000286.html
[03:25:43] *** lloy0076 has joined #opensolaris
[03:26:18] <hile_> evening gman
[03:26:29] <Gman> hey hile_
[03:27:30] <hile_> How goes?
[03:28:00] <Gman> knackered ;)
[03:28:40] <Gman> also trying to build bits and pieces of gnome on a sunblade 100
[03:28:49] <hile_> oh fuck dude...
[03:28:52] <Gman> more exciting watching paint dry
[03:28:55] <hile_> that hurts just talking about it.
[03:32:29] *** yongsun has joined #opensolaris
[03:33:33] * jmcp sends Gman a sympathy packet
[03:33:42] <jmcp> Gman: minspec sb100 no doubt
[03:33:53] <Gman> yeah, suspect so
[03:34:04] <Gman> i want a niagara :)
[03:34:18] <jmcp> so do I
[03:34:27] <jmcp> I just have to convince J that it's the right thing
[03:34:30] <richlowe> You want to throw that sb100 out a window.
[03:34:38] <jmcp> more important that I get the dslr that I want first :)
[03:34:48] <jmcp> richlowe: yes, maybe even open the window first, too
[03:35:27] <richlowe> Hmmmm though I guess it could be considered karma to make the JDS folks tolerate their own stuff on one.
[03:35:30] *** elflord_ has joined #opensolaris
[03:36:28] * jmcp grins broadly
[03:36:29] <Gman> richlowe, brrr!
[03:43:48] *** elflord has quit IRC
[03:45:44] *** test_ has joined #opensolaris
[03:47:44] * jmcp shops
[03:52:54] <jteo> moin.
[03:53:15] <hile_> howdy jteo
[03:53:26] <hile_> Is it 2000 US/Eastern Monday yet?
[03:53:29] <hile_> so that it can be my weekend?
[03:55:44] <jteo> yes it is.
[04:02:07] *** silly_girl22 has quit IRC
[04:05:12] <Error_404> hile_: say you're canadian & you need the day off
[04:05:20] <Error_404> it's thanksgiving
[04:05:43] <hile_> oh fuck dude i forgot about that
[04:05:51] <hile_> how the hell would I of all people for get that
[04:05:57] <hile_> Happy Thanksgiving John :)
[04:06:04] <Error_404> thanks man
[04:06:29] <hile_> If I didn't have work ethic I'd take a sick day
[04:08:00] <Error_404> heh
[04:19:56] *** yongsun has quit IRC
[04:22:40] *** yongsun has joined #opensolaris
[04:32:15] *** Dr_Jekyll has joined #opensolaris
[04:36:46] *** Elendal has joined #opensolaris
[04:37:32] <Elendal> Hi, want to connect Ultra 1 to wireless network, any ideas how to do this?
[04:38:38] <hile_> plug it into a bridge of some sort?
[04:39:52] <Elendal> Hmm, yes.
[04:42:34] <Error_404> it's about the only way to do it
[04:42:38] <libkeiser> if you own an sbus to pcmcia bridge, then openbsd might be an option
[04:45:13] *** jsubl2 has joined #opensolaris
[04:46:15] *** Dr_Jekyl1 has quit IRC
[04:46:33] <hile_> hey libkeiser
[04:48:04] <libkeiser> evening
[04:59:52] *** jsubl2 has quit IRC
[05:03:29] <Gman> http://positivesharing.com/2006/10/10-seeeeeriously-cool-workplaces/
[05:03:31] <Gman> freaking cool
[05:10:19] *** jsubl2 has joined #opensolaris
[05:19:25] <jteo> the versioning FS discussion has taken on a life of it's own. :|
[05:28:15] <steleman> 'evening
[05:28:55] * jmcp wanders back in
[05:29:03] *** lloy0076 has quit IRC
[05:29:17] <jmcp> jteo: I think it's the most long-lived of any thread I've seen since opensolaris.org when live
[05:30:04] *** g4lt has joined #opensolaris
[05:33:03] <Sir-Al> versioning fs is a cool idea
[05:35:52] <jteo> jmcp: unforunately everyone has a different idea of what a versioning FS should do.
[05:36:12] *** bank has joined #opensolaris
[05:36:40] <Sir-Al> they should make zfs more robust before they go on to implement features like versioning
[05:36:58] <bank> versioning?
[05:37:02] <bank> subversion control?
[05:37:45] <bank> morning all.
[05:38:09] <Sir-Al> bank: nite
[05:38:52] *** jwtodd has joined #opensolaris
[05:41:05] <boyd> Depends if you count the gpl/debian meta-thread
[05:41:26] <boyd> Oops... I have no idea how I did that
[05:42:29] <jmcp> boyd: you channeled Joerg you bad bad man
[05:42:44] * steleman waits for the inevitable
[05:42:45] <boyd> Haha!
[05:43:20] <steleman> PO ... PO ...
[05:43:23] <steleman> POSIX ?
[05:43:26] <boyd> What's the inevitable? "That thread is the only one to be Posix complian since 6 years" ?
[05:43:46] <steleman> something like that
[05:43:53] *** delewis has joined #opensolaris
[05:44:28] <boyd> :)
[05:44:41] * steleman bats eyelashes innocently
[05:44:46] <boyd> heh
[05:44:53] * jmcp snorts
[05:45:02] <jmcp> steleman: hi! long time no see
[05:45:09] <steleman> hiya jmcp :-)
[05:45:12] <jteo> "star solved all these problems in 1988"
[05:45:25] <jmcp> nono, "solved... since at 15 years"
[05:45:32] * jmcp chuckles
[05:45:42] <jmcp> hmm needmorecaffeine
[05:46:16] <Sir-Al> star?  not star again
[05:46:21] <boyd> jteo: You know, I was wondering (but didn't want to add to the thread) How much of what's wanted would be solved with a sorta half-way solution: an easy too to say "What snapshotted versions of this file are there (ignoring identical snapshots)" and "copy back the version from snapshot x"
[05:46:41] <Sir-Al> i think that an api would be the best bet
[05:46:48] <Sir-Al> shell extensions are easily integrated
[05:46:49] <boyd> umm  s/easy too to/easy to use/
[05:46:58] <Sir-Al> gui tools into gnome could be done in almost a day
[05:47:12] <bank> hi boyd,jmcp
[05:47:13] <Sir-Al> the only tricky part is nfs
[05:47:24] <boyd> I agree... I suspect that an API could get that answer faster than using stat
[05:47:30] <Sir-Al> the rest, samba, afp, cli, is all good
[05:48:00] <boyd> Sir-Al: Yeah... Seems to me that NFS being deficient is not a reason to not implement it. NFSv4 allows extensions to the protocol
[05:48:45] <Sir-Al> and by the time we have versioning in production, i would assume nfs v4 would be more prevalent
[05:49:17] <Sir-Al> linux is mostly nfs v4, the later bsds are as well.. mac isn't
[05:49:36] <jamesd> except linux's v4 is mostly broken
[05:50:34] <boyd> That's so people used to their v3 are comfortable
[05:52:36] <jteo> boyd: a shell script would do for that.
[05:53:32] <boyd> jteo: I agree... I just wonder if it could be more optimized with help from the fs later... otherwise it's at least 1 stat/snapshot plus a stack of metadata lookups just to list them
[05:54:23] <jteo> boyd: imho, implementation wouldn't be tough. the question is how people want it. i'm leaning towards a specialize api via openat right now.
[05:54:26] <Sir-Al> yeah, i don't have much experience with linux nfs, but from what i used, i had a few issues with using v4, so i reverted to v3 and it was all good...
[05:54:56] <Sir-Al> linux nfs is better than mac nfs by a large margin though, i can only get like 500k to 12000kb/sec between my mac and solaris server
[05:55:05] <Sir-Al> while using CIFS i get around 50 - 100mb/sec
[05:58:40] *** g4lt has quit IRC
[06:01:51] <delewis> Sir-Al: copying files through Finder or using 'cp' and friends?
[06:02:00] <Sir-Al> delewis: mr Finder
[06:02:08] <delewis> that's the problem :-)
[06:02:17] <delewis> Try using 'cp' -- you'll see a difference
[06:02:18] <Sir-Al> ya well i don't expect my users to use cp
[06:02:24] <Sir-Al> especially when most things can't be copied with cp
[06:02:37] <delewis> Finder does something silly like reading back every block written or something
[06:02:39] <Sir-Al> delewis: actually i did have performance problems with cp as well
[06:02:56] <delewis> I get normal speeds when not using Finder
[06:02:57] <Sir-Al> delewis: maybe ditto works too
[06:03:00] <Sir-Al> hmm
[06:03:13] <Sir-Al> delewis: too bad i can't test it out because my zpool won't mount
[06:04:32] <boyd> jteo: Any thoughts on the interface?
[06:05:00] <boyd> delewis: Are you aware that cp -Rp preserves more attributes than ditto these days?
[06:05:09] <jteo> boyd: nope. ;(
[06:05:23] <jteo> do we really want special tools to access versions?
[06:05:59] <boyd> jteo: Not really... I was just trying to think of a way to make the existing facilities provide much of the target.
[06:06:49] <jteo> boyd: the thread has proponents stating that versioning doesn't give us anything we don't have already.
[06:08:40] <boyd> Which I think is untrue, but it's hard to think of an automatic way to make it work. Maybe on close if there has been any write() s
[06:10:07] *** NilsAN has quit IRC
[06:10:22] <bank> :)
[06:10:55] *** NilsAN has joined #opensolaris
[06:11:19] <jteo> it's a vicious cycle: apps only expect a simple POSIX bucket of bits, and no one wants to use new apps.
[06:12:20] <boyd> Yeah... I was thinking the same thing about some of the reiser semantics.
[06:13:03] <boyd> It's a shame, since I think there are some facilities that could be added to what the FS provides beyond Posix..
[06:13:45] <jteo> a way out of this would be to extend JDS to be version aware first, and then start extending apps.
[06:14:04] <jteo> and it just ocurred to me OpenOffice is probably more complex than zfs.
[06:14:15] <boyd> I'm sure it is.
[06:14:37] <jteo> it shouldn't be. (OpenOffice)
[06:14:37] *** bank has quit IRC
[06:14:50] <boyd> Yeah, JDS integration would be a good kick. In fact, if the open dialog supported it it could be largely automatic in many apps
[06:15:22] <jteo> boyd: excellent idea! ;)
[06:15:26] *** jwtodd has quit IRC
[06:15:34] 
[06:16:11] <boyd> s/uses/users
[06:18:08] <jmcp> I don't think it's bad for naive users, I think it's a pretty good idea actually - helps with the "oh crap what did I do with my file when I had X in it?" questions
[06:18:48] <boyd> Look at what apple are doing with Time Machine. It's a nice front end on versioning (in effect) with some application integration
[06:18:51] <jteo> do people really lose that much work in between snapshots?
[06:18:52] <jmcp> what's needed is a consistent interface and semantics
[06:18:59] <jmcp> jteo: dunno
[06:19:34] <boyd> Depends how often they're taken... also, taking snapshot while a file is open is probably bad
[06:19:41] <boyd> ... or at least not optimal
[06:19:59] <jteo> boyd: it's not bad. matt says it should be really quick now.
[06:20:26] <boyd> I actually mean bad for the file contents integrity, not performance
[06:21:32] <jteo> boyd: it doesn't affect integrity because all pending transactions have to complete before a snapshot is comitted.
[06:21:43] <jteo> integrity from the app perspective is another issue. ;)
[06:21:44] <richlowe> jteo: integrity from the view of the application, not the FS
[06:22:00] <boyd> For me, the biggest problems with 5-minute snapshots are: 1) filesystem granularity (someone mentioned the 4GB video file that's deleted) and 2) it's hard to simply see the list of *different* versions of a single file. 2) is easy to solve with user space code. 1), less so.
[06:22:05] <richlowe> oh, sure, think ahead and beat me to it. ;)
[06:22:06] <jteo> fflush would be a good candidate.
[06:22:07] <boyd> Yeah, I mean integrity from the ap
[06:22:08] <jteo> ;)
[06:22:09] <boyd> app
[06:22:55] <boyd> There is also potentially 3) several changes in one 5-minute period
[06:23:24] <jteo> which begs the question: doesn't your app have undo?
[06:24:01] <boyd> Not if it's a script or I quit it :)
[06:24:48] <jteo> true.
[06:25:50] <boyd> I like the extended attributes thing. JDS integration would help. Only other q I can thing of at the moment is "how do I roll a file back 1 rev?" I can't copy an extended attr on top of the file it's attached to... oh, no.. maybe I can, in fact.
[06:26:11] <boyd> Are extended attributes (solaris style) visible across nfs?
[06:28:04] <boyd> Of course, extended attrs won't help in the "oops, I rm'ed the wrong file" scenario, since all the attrs go too. but there's still snapshots
[06:31:50] *** jsubl2 has quit IRC
[06:36:50] *** Kernel86 has quit IRC
[06:42:42] <jteo> in other words, the reference to the tree that represents the file versions should be stored outside the latest rev.
[06:42:45] <jteo> hmm.
[06:43:20] <boyd> Well, only for that case.... on the other hand from an interface POV *inside* the rev is much nicer
[06:44:33] <jteo> i'll spend a while reading before summarizing everything in a blog entry.
[06:45:49] * boyd nods
[06:47:01] *** klocze1 has quit IRC
[06:50:14] <jteo> boyd: you give me the impression it's an interface/packaging issue. like "how do we make this usable?"
[06:50:31] <boyd> Nicolas Williams is saying some sensible stuff
[06:50:34] <boyd> Umm..
[06:51:22] <boyd> Well, I'm more thinking along the lines of "what behaviour do we want, how will that affect all current uses of the FS, *then* think about implementation."
[06:52:08] <jteo> boyd: ah.
[06:57:17] *** tsoome has quit IRC
[06:58:54] *** Gman has quit IRC
[07:13:11] *** Kernel86 has joined #OpenSolaris
[07:29:18] <jmcp> back in a bit
[07:29:19] *** jmcp has quit IRC
[07:29:26] *** Griff has joined #opensolaris
[07:38:12] *** s3tup has quit IRC
[07:39:03] *** Gr|ffous has joined #opensolaris
[07:39:12] *** s3tup has joined #opensolaris
[07:47:43] *** Griffous has quit IRC
[07:51:55] *** Griff has quit IRC
[07:59:17] *** bank has joined #opensolaris
[08:00:43] *** NilsAN has quit IRC
[08:01:25] *** NilsAN has joined #opensolaris
[08:02:28] <twincest> hm.  studio express for linux doesn't ship any manpages
[08:02:53] <richlowe> I must admit, I still don't entirely see the point in general.
[08:03:01] <twincest> of what, the linux studio?
[08:03:07] <richlowe> beyond muttered curses at gdb every now and then, obviously.
[08:03:10] <richlowe> twincest: Yeah.
[08:03:50] <twincest> lint and dbx are nice, and the C compiler isn't bad
[08:04:00] <twincest> although i already have idb and gdb and icc.. :)
[08:04:34] <twincest> maybe it helps porting software using 4.x's ancient c++ dialect or something
[08:04:36] <richlowe> It's not the utility of it.
[08:04:55] <richlowe> it's the fact that no matter how good (or not good) it is, I doubt it's going to have enough users to cause it to have been worth the (assumed) effort.
[08:08:42] <twincest> eh well, i don't mind about that as long as they do it :)
[08:22:50] *** kloczek has joined #opensolaris
[08:25:50] *** tsoome has joined #opensolaris
[08:28:41] *** bank has quit IRC
[08:35:31] *** mazon is now known as Mazon
[08:35:32] <boyd> I presume it was there thanks to a thought that linux on SPARC could be made to be bigger than it is.
[08:35:55] <jengelh> Indeed.
[08:36:09] <jengelh> (And thank God debian is not the only one available for sparc)
[08:39:32] <twincest> the compilers don't even run on linux sparc, do they?
[08:41:26] <jengelh> asking whom?
[08:41:33] <twincest> boyd
[08:41:47] <jengelh> sunstudio?
[08:41:52] <twincest> yes
[08:41:59] <jengelh> should test :)
[08:42:16] <twincest> i only saw a linux x86 download on sdlc yesterday
[08:42:36] <jengelh> Generally it seems possible to run solaris binaries on linux-sparc, using emulation
[08:42:56] <twincest> well, okay, but that's not a good way to run your standard compiler :)
[08:43:06] <twincest> although i did manage to make Intel icc produce Solaris x86 binaries, it was quite a pain
[08:43:26] * jengelh mumbles: just use gcc
[08:45:11] * twincest thought gcc was quite suboptimal on sparc..
[08:49:11] <tsoome> gcc used to be quite crap:(
[08:49:26] <tsoome> not sure about current state
[08:49:29] <twincest> i've been quite unimpressed with it on x86 as well recently
[08:49:50] <twincest> i wrote something yesterday to benchmark a tst/hash table implementation, and icc ran it at least 3-4 times faster than gcc
[08:50:14] <twincest> not that that's a typical workload, but still
[08:50:56] *** Cyrille_ has joined #opensolaris
[08:51:05] <twincest> 2 and 3
[08:51:39] <twincest> tested icc using -xipo -O3 -xD and also with just -O2
[08:51:42] <jengelh> icc is quite tied to intel platforms
[09:22:59] <asyd> \_o<
[09:25:44] *** kloczek has quit IRC
[09:27:23] <noyb> >o_/    quack back
[09:27:28] *** swa_mobil has quit IRC
[09:28:14] *** kloczek has joined #opensolaris
[09:37:41] *** darkie33 has joined #opensolaris
[09:37:45] <darkie33> hi
[09:37:56] <darkie33> does solaris have good driver support?
[09:38:37] <quasi> darkie33: yes and no
[09:39:12] *** Andrew____ has joined #opensolaris
[09:39:54] <quasi> darkie33: it supports some very good drivers... but doesn't quite have as many as one could wish for
[09:40:15] <darkie33> quasi:what abt the display drivers?
[09:40:45] <darkie33> quasi:does it have this sis chipset support?
[09:40:50] <quasi> darkie33: probably fairly decent
[09:40:59] <quasi> darkie33: good question
[09:41:09] <twincest> for graphics you have Xorg
[09:41:56] <darkie33> i am not having any luck in linux for sis.. want to migrate to solaris now..
[09:43:49] * Stric used sis in linux on one computer up until the other day without much trouble
[09:45:03] <darkie33> Stric:well glgears gives really poor results..
[09:45:25] *** Dar_HOME is now known as Dar
[09:46:13] <Stric> darkie33: sis isn't really known for their top 3d performance..
[09:46:26] <Stric> if you want that, go ati or nvidia.
[09:46:40] <Stric> and if you want driver support in linux or solaris, go nvidia.
[09:47:11] <darkie33> Stric:oh well i just want a decent performance..
[09:47:44] <darkie33> in the sense that if in windows i can get a decent performance i want it same in linux or solaris..
[09:50:07] <Stric> in linux, enable dri/drm thingie in the kernel and you should get better speed than software rendering..
[09:50:45] *** ShadowHntr has joined #opensolaris
[09:51:10] <darkie33> Stric:oh i am not able to do that because of no driver support i guess..
[09:52:58] <Stric> I replaced my sis machine a few days ago so I can't verify the performance (and i didn't use it before because it was on my video/projector/tv machine).. but I believe I tested glxgears and got some performance.. never tried windows on it..
[09:55:31] * Stric -> meeting
[09:58:49] *** darkie33 has quit IRC
[10:05:01] *** |ReIkO| has joined #opensolaris
[10:05:04] <|ReIkO|> morning all
[10:07:12] *** swa_mobil has joined #opensolaris
[10:08:18] *** Error_404 has quit IRC
[10:08:42] *** swa_mobil has joined #opensolaris
[10:10:08] *** alanc_ has quit IRC
[10:10:51] *** alanc_ has joined #opensolaris
[10:17:39] *** elflord_ has quit IRC
[10:17:58] *** jmcp has joined #opensolaris
[10:22:07] <twincest> hey jm
[10:22:36] <PerterB> morning, chaps
[10:23:56] <jmcp> hiya
[10:25:20] <jmcp> I think technorati is having some issues
[10:25:28] <jmcp> I can't seem to get my login to stick
[10:31:01] *** jmcp has quit IRC
[10:31:52] <|ReIkO|> Do you know how can verify a x509 certificate signature ?
[10:33:47] <andersmo> $ openssl verify -h
[10:33:48] <andersmo> usage: verify [-verbose] [-CApath path] [-CAfile file] [-purpose purpose] [-crl_check] [-engine e] cert1 cert2 ...
[10:36:15] * quasi looks around in confusion - did I make a wrong turn and step into #openssl ;)
[10:36:46] <lasseoe> no no, this is #anything_goes
[10:37:53] <quasi> ah, that explains it ;)
[10:38:27] <andersmo> This is #reiko_asks_lots_of_ssl_related_questions, and I intended to give a not-so-subtle hint about where to look for an answer. =)
[10:38:50] <lasseoe> andersmo, yeah it was a real head scratcher :)
[10:40:42] <|ReIkO|> andersmo I need not command line
[10:40:45] <|ReIkO|> but C code
[10:40:48] <|ReIkO|> call
[10:41:02] <lasseoe> Reiko, try in #ssl #openssl #c or something
[10:41:09] <PerterB> use the source, |ReIkO|
[10:41:28] <quasi> or pay someone to do your work
[10:42:13] <andersmo> |ReIkO|: This channel isn't concerned with PKI and related problems, you're probably better off asking people who are actually interested in such problems. =)
[10:42:37] <|ReIkO|> asyd is my relator :)
[10:43:48] <andersmo> |ReIkO|: Anyway, if I had to figure out how to verify x509 certificates programmatically, I'd start with the openssl source and figure out how they do it. =)
[10:44:07] <|ReIkO|> andersmo I'm doing that
[10:44:18] <andersmo> But openssl is completely unrelated to OpenSolaris. =)
[10:44:31] <|ReIkO|> this is the problem
[10:44:44] <|ReIkO|> openssl is bad documented
[10:45:11] <quasi> |ReIkO|: don't blame that on opensolaris
[10:45:38] <andersmo> There are probably commercial well-documented libraries around. (They need to be well documented because you probably won't see the source? =)
[10:45:53] <andersmo> Still, not OS related. =)
[10:46:02] <|ReIkO|> of course
[10:46:08] *** Tpenta has joined #opensolaris
[10:49:16] *** Dar is now known as Dar_AFK
[10:53:14] *** MattMan has joined #opensolaris
[11:04:25] *** [tfb] has quit IRC
[11:09:48] *** Dar_AFK is now known as Dar
[11:13:58] *** denied has joined #opensolaris
[11:14:19] *** denied is now known as s3tup-
[11:16:27] <s3tup-> hi
[11:32:46] *** lloy0076 has joined #opensolaris
[11:33:02] <lloy0076> I have the most stupid question.
[11:33:15] <lloy0076> I've just installed OpenSolaris ON from DVD on my laptop...
[11:33:18] <lloy0076> I'm running Gnome.
[11:33:23] <lloy0076> It's running perfectly...
[11:33:32] <lloy0076> BUT there's no shutdown the computer button or option!
[11:33:33] <lloy0076> :(
[11:33:56] <lloy0076> I can't see one from the XDM/GDM or in any of my menus.
[11:34:22] <jengelh> manage solaris or it will manage you
[11:35:06] <quasi> lloy0076: init 5 in a console will do the trick
[11:35:12] <jengelh> init 0
[11:35:23] <lloy0076> Yes, that would however it would be nice to not have to do something like that.
[11:35:34] * quasi kicks jengelh
[11:35:38] <jengelh> quasi: what
[11:36:13] <quasi> lloy0076: I think you have to exit and then choose it from the general screen
[11:36:23] * lloy0076 hmmm
[11:36:34] <lloy0076> I'll try that again.
[11:36:41] <quasi> jengelh: 0 may be a problem in some cases - 5 is fine
[11:36:56] <jengelh> quasi: he said shutdown, not startx
[11:37:01] <lloy0076> :)
[11:37:22] <lloy0076> init 0 should initiate a shutdown.
[11:37:23] <quasi> jengelh: sure, and init 5 is poweroff
[11:37:28] <jengelh> huh
[11:37:32] <jengelh> weird solaris thing...
[11:37:37] <lloy0076> And I'll try init 5 if I can't find the whatsy.
[11:37:40] <lloy0076> Anyway, off I go.
[11:37:43] *** lloy0076 has left #opensolaris
[11:37:50] <jengelh> linux is more orthogonal - 0 being halt/poweroff, 5 x, 6 reboot
[11:37:51] *** Hunger- has quit IRC
[11:38:35] <quasi> linux is inconsistent as hell, don't suggest something from there because you assume it may be the same in solaris
[11:39:00] <quasi> then killall will get you in serious trouble on older solaris
[11:39:28] <PerterB> linux run levels are gratuitously different from historical System V run levels
[11:39:33] <jengelh> i don't think killall (sol) supports names so it just errors out
[11:39:40] <s3tup-> solaris cant detect my audio card
[11:40:06] <quasi> jengelh: take killall on sol7 - it does just what the command says it does and takes no options
[11:45:48] *** s3tup- has quit IRC
[11:48:50] <GreenHat> ipv4cksum() question: I am using this method to set checksum of a modified ip header in a module ; strangely in a sniffer I see "incorrect checksum" for these packets ; what is odd is that if we add 4 to each checksum we get the correct answer !!
[11:48:51] <asyd>    
[11:49:19] <GreenHat> for example : 0xb824 instead of the correct which is 0xb828
[11:49:27] <GreenHat> and this is consistent !
[11:49:44] <GreenHat> could it be something related to that I am using it on SPARC ?
[11:50:04] *** jmcp has joined #opensolaris
[11:50:21] <GreenHat> I knwo I can simply add 4 but I don't want to without understanding
[11:53:39] <PerterB> "Use the 4's, Luke"
[11:53:40] <PerterB> (sorry)
[11:55:13] <GreenHat> why?
[11:55:22] * jmcp notes that there have been 99 posts in jteo's versioning fs thread on zfs-discuss
[11:55:40] <GreenHat> not why sorry, but why does the original method not work?
[11:55:55] <GreenHat> because I see it is used in the kernel
[11:56:23] <GreenHat> dlpi_io.c,icmp.c ,ipv4.c  are example of such usage
[11:58:25] *** lloy0076 has joined #opensolaris
[11:58:27] <lloy0076> Heh
[11:58:49] <lloy0076> Well, I now have Gentoo Linux (which is currently my main distro on my laptop) and Solaris Express RATHER than Gentoo + Windows XP
[11:59:39] <lloy0076> Anyway, I'm off home for the night :)
[11:59:40] *** lloy0076 has quit IRC
[12:04:52] *** qbit has left #opensolaris
[12:06:51] *** apierre has quit IRC
[12:08:16] *** calumb has joined #opensolaris
[12:08:34] *** kimc has joined #opensolaris
[12:09:59] <GreenHat> anybody heard of "checksum adjust" in solaris kernel ?
[12:13:54] *** Hunger- has joined #opensolaris
[12:20:20] *** calum_ has joined #opensolaris
[12:20:39] *** [tfb] has joined #opensolaris
[12:26:25] *** [tfb] is now known as tfb
[12:26:28] *** calumb has quit IRC
[12:30:57] *** Vanuatoo has quit IRC
[12:31:39] *** Vanuatoo has joined #opensolaris
[12:35:41] *** calum_ is now known as calumb
[12:38:23] *** ShadowHntr has quit IRC
[12:38:30] *** yongsun has quit IRC
[12:39:15] *** Vanuatoo has quit IRC
[12:39:32] *** Vanuatoo has joined #opensolaris
[12:40:14] *** Snake007uk has joined #opensolaris
[12:40:57] <|ReIkO|> asyd my love :)
[12:41:22] <|ReIkO|> miss you in my cripto development
[12:41:23] <|ReIkO|> de eh!
[12:48:33] <GreenHat> any idea if I can know if my NIC performs chksum calc in hw ?
[12:50:05] <|ReIkO|> should be a default operation
[12:51:21] <GreenHat> so why for example there are checksum calcs in nat of solaris ? if the card does that job anyhow ?
[12:51:30] <Stric> I think ethereal/snoop/etc will say that all packets have the wrong checksum if it's done in hw
[12:53:05] <ofu> 120849-04 fixes bug 6380495 checksum is being calculated in software with ipge driver
[12:53:48] <twincest> green: not all cards can do hw checksums
[12:54:19] <GreenHat> I have e1000
[12:54:23] <GreenHat> (intel)
[12:54:30] <twincest> that's ipge
[12:54:35] <twincest> (the driver name changed)
[12:54:54] <GreenHat> is it true what Stric said about error when sniffing if it calc by hw ?
[12:55:38] <GreenHat> twincest, what do you mean ? I believe I work with /dev/e1000g0
[12:55:54] <twincest> yes, whether you have e1000g or ipge depends on which solaris release you use
[12:56:16] <twincest> but they're the same thing, intel 1GE
[12:56:18] <GreenHat> I am working on onnv-20060619 on sparc
[12:56:26] <GreenHat> (MIAGRA)
[12:56:56] <GreenHat> SO does it have hw checksum?
[12:57:12] <twincest> the card does, but whether solaris supports it presumably depends on what ofu was talking about
[12:57:27] *** janus_man has joined #opensolaris
[12:57:43] <GreenHat> ofu ?
[12:57:48] <GreenHat> you mean bfu?
[12:57:49] <twincest> < ofu> 120849-04 fixes bug 6380495 checksum is being calculated in software with ipge driver
[12:59:22] <ofu> this patch was added to 10_Recommended on April, 2nd 06
[12:59:23] <GreenHat> http://sunsolve.sun.com/search/advsearch.do?collection=PATCH&type=collections&max=50&language=en&queryKey5=120849&toDocument=yes
[13:00:15] <GreenHat> So I assume it is with the opensoalris I have
[13:01:34] <GreenHat> twincest, so I understand that even that it talks about /kernel/drv/sparcv9/ipge , which I am not using, it is also relevant to /kernel/drv/sparcv9/e100g ?
[13:01:53] *** kimc has left #opensolaris
[13:15:21] *** bunker_work has joined #opensolaris
[13:16:46] <bunker_work> hello, anyone known where i can find a working nasm package for my ultra10 64bit?
[13:17:22] <bunker_work> i'm using latest opensolaris distribution on it
[13:34:19] *** calumb is now known as calLUNCH
[13:35:47] *** Griffous has joined #opensolaris
[13:35:48] *** Gr|ffous has quit IRC
[13:38:19] *** qbit has joined #opensolaris
[13:49:36] *** eugene has joined #opensolaris
[13:51:43] <jmcp> bunker_work: there is no "nasm" for Solaris. at least, not in the MS sense. what functionality are you looking for?
[13:52:06] <twincest> i thought nasm was the free masm that ran everywhere
[14:00:18] *** mkhl has joined #opensolaris
[14:06:55] *** |ReIkO| has quit IRC
[14:20:18] *** kloczek has quit IRC
[14:21:37] *** Snake007uk has quit IRC
[14:21:58] *** kloczek has joined #opensolaris
[14:23:14] *** bronze_|| has joined #opensolaris
[14:29:06] *** calLUNCH is now known as calumb
[14:33:24] <GreenHat> I still have a question about checksum:
[14:33:38] <GreenHat> suppose I want to activate HW checksum
[14:33:50] <GreenHat> on e1000g event that there is that patch
[14:33:55] <GreenHat> HOw can I do it ?
[14:37:33] *** ChanServ sets mode: +o Tpenta
[14:38:10] <GreenHat> I found that : "The driver can now employ the card's onboard H/W checksum offload facility if it is available, resulting in improved performance (in some tests as much as 30% better throughput)."
[14:38:15] <GreenHat> in http://blogs.sun.com/dp/entry/what_s_new_in_solaris10
[14:40:40] *** triplah has joined #opensolaris
[14:42:19] * jmcp returns
[14:44:33] *** bank has joined #opensolaris
[14:45:03] <bunker_work> jmcp: i have to compile asm code and decode it from binary (ndisasm like)
[14:45:10] <bank> hi
[14:45:24] <bank> I got ultra 80 a uni lab.
[14:45:30] <bank> look old.
[14:45:32] <bunker_work> jmcp: at now i've installed "gas"
[14:46:12] <bank> I press the turn on button at front. it seem to not work.
[14:46:21] <bank> ancient server.
[14:47:02] <jmcp> bunker_work: you want to have a look at /usr/ccs/bin/dis
[14:47:09] <jmcp> bank: is there a Sun keyboard attached?
[14:47:17] <jmcp> bank: if there is, press the POWER key on the top right
[14:47:27] <bank> not sure sun
[14:47:29] <bank> wait a moment
[14:47:44] <s3tup> what should i do if Solaris said /dev/audio doesnt exist ?
[14:48:54] <bank> have a sound
[14:48:56] <bank> when press
[14:48:58] <jmcp> s3tup: are you sure you've got a sound device in your system?
[14:49:26] <bank> scary button
[14:49:41] <s3tup> jmcp : im using ASUS A3HFseries laptop
[14:50:52] *** denied has joined #opensolaris
[14:51:09] <jmcp> s3tup: that doesn't mean much to me, but since you say laptop I assume there's a sound chip on it
[14:51:11] *** denied is now known as s3tup-
[14:51:45] <s3tup-> jmcp: any guides on how i can configure my audio device ?
[14:54:19] <jmcp> what driver does Windows use?
[14:54:51] <s3tup> jmcp : this laptop came with No OS
[14:54:58] <bank> someone infront of monitor. I run away :P
[14:56:14] <jmcp> s3tup: so follow the instructions at http://blogs.sun.com/jmcp/entry/how_do_i_find_out
[14:56:23] <jmcp> granted, it's for a nic, but the same principle applies
[14:57:48] <GreenHat> for all NIC pundits:
[14:58:03] <GreenHat> How can I know 1) if my NIC supports HW checksum offload
[14:58:14] <GreenHat> 2) In case it does how to activate it ?
[14:58:26] <GreenHat> I am despereate from googling for it
[14:58:31] <eugene> GreenHat: in chinese, greenhat means that your wife has cheated on you :)
[14:58:42] <GreenHat> eugene, I am sorry to know
[14:58:59] <GreenHat> eugene, and I will think of starting learing
[14:59:00] <GreenHat> chinees
[14:59:05] * elektronkind makes a procmail recipe for Subject.*A Versioning FS
[14:59:07] <eugene> GreenHat: good luck
[14:59:16] *** axisys has quit IRC
[14:59:23] <GreenHat> I know about 4 words already
[14:59:26] <jengelh> chinese is boring, try japanese
[14:59:35] <GreenHat> eugene, do you know chineese?
[15:00:08] *** qbit has left #opensolaris
[15:00:11] <eugene> GreenHat: yes
[15:00:33] <GreenHat> jengelh, thinking HiTech , they say that china
[15:00:42] <eugene> jengelh: japanese uses a very simplified set of chinese characters plus some similar traditional chinese characters.
[15:00:43] <GreenHat> has a very good potential
[15:00:56] <jengelh> Yeah and starting with jp makes zh easier later :)
[15:01:00] <bunker_work> jmcp: thank's!
[15:01:00] <GreenHat> regarding my NIC question...
[15:01:05] <elektronkind> plus japan has Gundam units
[15:01:07] <GreenHat> any idea ?
[15:01:18] <eugene> jengelh: in terms of learning difficulty/curve, japanese is simplier and more boring compared to chinese :)
[15:01:43] <jengelh> i don't think it is boring exactly _because_ it is easier
[15:02:13] <eugene> jengelh: i know both language enough to judge if which is boring :)
[15:02:18] <jengelh> heh
[15:02:29] *** nachox has joined #opensolaris
[15:03:13] <bank> jmcp: how're you today?
[15:04:43] <GreenHat> people ! I am desperate. My wife betrayed me with somebody probably from  china
[15:04:47] <GreenHat> and I don't know
[15:04:48] <GreenHat> how to
[15:04:53] <GreenHat> switch the hardware checksum on
[15:04:54] <GreenHat> or off
[15:05:16] <GreenHat> I am talking abut Network Card (e1000)
[15:05:46] <GreenHat> any ideas ? maybe japanese people can help ? in english ...
[15:06:06] <jmcp> bank: doing ok, thankyou
[15:06:07] <GreenHat> or in chinees as someonce knows here this language
[15:06:13] <jmcp> bunker_work: you're welcome
[15:06:15] <GreenHat> I am going nuts...
[15:06:20] <GreenHat> HW checksum please ...
[15:06:51] <jmcp> GreenHat: has it occurred to you that since you've asked that same question here about 4 times and got no response, that maybe you should use eg google.com to try to find the answer?
[15:07:24] <Cyrille_> would he keep asking the question if it had occurred to him? ;-)
[15:07:44] <GreenHat> Cyrille_,
[15:07:47] <GreenHat> jmcp,
[15:08:02] *** triplah has quit IRC
[15:08:03] <GreenHat> I had mentioned it herer
[15:08:06] <GreenHat> here
[15:08:11] <GreenHat> I googled for it
[15:08:20] <GreenHat> googling is the default
[15:08:21] <jengelh> google harder
[15:08:26] <GreenHat> before IRC
[15:09:39] <GreenHat> "e1000g" "solaris" "hardware checksum offload"
[15:09:42] <GreenHat> gave 7 results...
[15:10:08] <GreenHat> non helped
[15:10:39] <asyd> check the driver source if available
[15:11:26] <GreenHat> it is NOT
[15:11:30] <GreenHat> available
[15:11:46] <GreenHat> I wish it would ; however
[15:11:51] <jengelh> maybe solaris can't do it at all
[15:11:57] <GreenHat> this NIC is on the T2000
[15:12:13] <GreenHat> so I assume people would want to use this feature
[15:12:18] <GreenHat> if possible
[15:12:29] <jengelh> there is more to sparc than solaris
[15:12:34] <GreenHat> they talk about 30% better performance
[15:12:56] <GreenHat> jengelh, did not understand your last sentence
[15:13:01] <jengelh> hehe
[15:13:02] <GreenHat> what do you mean by that ?
[15:13:13] <GreenHat> "there is more to sparc than solaris"
[15:13:16] <GreenHat> ?
[15:13:28] <GreenHat> more what?
[15:13:31] <jengelh> solaris is not the only thing to run on sparc, hence, there are other OS to run with e1000
[15:13:51] *** loke has joined #opensolaris
[15:14:03] <GreenHat> ok. so you suggest to see if it is supported on
[15:14:07] <GreenHat> the other OS
[15:14:08] <GreenHat> linux/
[15:14:12] *** calumb has quit IRC
[15:14:13] <GreenHat> bsd
[15:14:14] <jengelh> bsd, pick some
[15:14:19] <jengelh> whatever
[15:14:22] <bank> seem like you are busy during these 2-3 days
[15:14:24] <GreenHat> ok
[15:14:26] <jengelh> see if OBP can do it
[15:14:36] <jengelh> (though would be overkill)
[15:14:40] <trygvis> why would obp support hardware checksumming?
[15:14:40] <GreenHat> what is OBP
[15:14:40] <bank> I am converting data mining algo from C++ to java.
[15:14:48] <jengelh> what, you got a sparc and don't know what obp is?
[15:14:54] <jengelh> i need a drink...
[15:14:55] <GreenHat> I knew
[15:14:58] <GreenHat> and forget
[15:15:00] <GreenHat> really
[15:15:34] <jmcp> GreenHat: OBP = "OpenBoot Prom" - an implementation of IEEE-1275 aka Open Firmware
[15:16:00] <GreenHat> jmcp, ok , I remember now
[15:16:07] *** bunker_work has quit IRC
[15:16:25] <bank> oh bunker.
[15:17:45] <trygvis> the grub e1000 driver seems to support it
[15:17:47] *** bank is now known as bank__
[15:18:07] <GreenHat> trygvis, "grub e1000"
[15:18:15] <GreenHat> what is grub in this context?
[15:19:06] <GreenHat> and don't say I don't googel
[15:19:09] <GreenHat> since
[15:19:14] <GreenHat> this is what google gives
[15:19:15] <GreenHat> www-uxsup.csx.cam.ac.uk/~fanf2/hermes/src/grub-e1000
[15:19:16] <asyd> "you don't google"
[15:19:37] <GreenHat> did you mean this?
[15:19:52] <GreenHat> google will sue me
[15:19:53] <GreenHat> ...
[15:19:58] <loke> hmm... I tried to translate bunkers exit message, but babelfish came up as: "And rare "????" ??????"
[15:20:05] <loke> anyone speak japanese? :-)
[15:20:31] <GreenHat> trygvis, I assume you mean
[15:20:35] <jmcp> loke: plenty of people do
[15:20:38] <GreenHat> the e1000
[15:20:41] <GreenHat> for solaris boot
[15:20:44] <jmcp> loke: that's not the question you wanted to ask :)
[15:20:46] <loke> jmcp: hah... anyone here?
[15:20:58] <bank__> banker
[15:21:01] <Cyrille_> 154 persons at last count
[15:21:04] <loke> jmcp: Well, I deserved it. I usually answer just like that too :-)
[15:22:30] <GreenHat> trygvis, How do you know that e1000 of grub supports it?
[15:22:31] *** s3tup has quit IRC
[15:27:42] *** bank__ has quit IRC
[15:27:51] <jmcp> loke: not here, sorry :)
[15:35:46] *** logic has quit IRC
[15:35:48] *** logic_ has joined #opensolaris
[16:01:35] <xinkeT> anyone know if the T2000 will be upgradable to niagra-2?
[16:04:31] *** lisppaste3 has joined #opensolaris
[16:04:35] *** MattMan is now known as MattAFC
[16:05:56] *** aska_ has joined #opensolaris
[16:06:09] <jmcp> xinkeT: highly unlikely
[16:07:18] <xinkeT> jmcp: that was my inclination as well, which is a shame
[16:08:56] <elektronkind> it's a shame that I can't plug a opteron into this pentium2 motherboard I have :/
[16:09:20] <xinkeT> heh yeah, but that isn't quite an apples to apples
[16:09:40] <xinkeT> having an upgrade path for an expensive box doesn't seem that unreasonable
[16:10:41] *** cd32fan has joined #opensolaris
[16:11:00] <cd32fan> hello
[16:11:31] <cd32fan> i need some help installing Solaris on my pc
[16:12:03] <cd32fan> can someone help,please
[16:12:46] <cd32fan> oh
[16:13:27] <loke> cd32fan: do you have a question?
[16:17:58] *** aska has quit IRC
[16:18:02] *** Mazon is now known as mazon
[16:21:53] <cd32fan> hey loke
[16:21:57] <cd32fan> my question is
[16:22:04] <cd32fan> on the partition menu
[16:22:46] <cd32fan> is it possible that when im installing Solaris,that my xp install will be erased?
[16:23:20] <loke> cd32fan: very possible :-)  I have no idea. I have never had anything except solaris on my systems
[16:23:31] <cd32fan> hmmm
[16:23:43] <cd32fan> i was wise no to proceed with the install
[16:23:48] <tfb> cd32fan: do you have a spare partition for Solaris?
[16:23:58] <tfb> I mean, do you already have one?
[16:24:03] * jmcp sleeps
[16:24:04] <jmcp> gnite all
[16:24:06] <cd32fan> no
[16:24:07] <loke> cd32fan: why would you want something else on the system? It's just annoying
[16:24:20] <cd32fan> i only have C: and D: 2 hard disks
[16:24:23] <loke> cd32fan: if you just want to check out what Solaris is like, you can install it in WMware
[16:24:31] <tfb> yeah, do that
[16:24:46] <cd32fan> but if i like it
[16:24:55] <cd32fan> how do i partition my hard disk?
[16:25:01] <cd32fan> and keep xp
[16:25:04] <loke> cd32fan: then you flip it around. Install Solaris and then run windows in WMware :-)
[16:25:13] <tfb> I don't know how to shrink the XP partition(s), sorry
[16:25:20] <cd32fan> hmmm
[16:25:24] *** sparc-kly has joined #opensolaris
[16:25:24] *** ChanServ sets mode: +o sparc-kly
[16:25:32] <loke> tfb: he has two hard drives. No need to shrink.
[16:25:42] <cd32fan> right
[16:25:47] <tfb> Well, so long as one is not in use by XP?
[16:26:02] <cd32fan> but on install there is only partition available
[16:26:10] <tfb> Then it ought to be fine to install on the spare disk
[16:26:30] <cd32fan> there doesnt appear to be 2 harddisks recognized
[16:26:42] <cd32fan> only 1
[16:27:42] <cd32fan> weird
[16:27:43] <rydis> Not to be insulting, but is it really two disks, and not just one partitioned disk?
[16:27:50] <cd32fan> no
[16:27:56] <cd32fan> i have 2 hard disks ;)
[16:28:04] <tfb> are they on the same controller?
[16:28:08] <cd32fan> yes
[16:28:34] <hali> controller as in chipset as well, not just hooked up to the same motherboard.. my motherboard has three chips for SATA junk
[16:28:35] <tfb> Then I don't know, sorry - seems to me the installer should see both...
[16:29:02] <hali> whereof only the sillicon image one work in opensolaris...
[16:31:40] <janus_man> and only on intel/amd
[16:32:09] <cd32fan> so,in normal conditions the 2 hard disk should be recognized by Fdisk?
[16:32:36] <hali> cd32fan: yes, might be worth moving them around the sata ports on the motherboard if you want to play around a bit :)
[16:32:37] <cd32fan> even if they are full?
[16:33:00] <hali> or boot a knoppix CD and check what chipset they are detected under
[16:36:48] <cd32fan> thanks all for the help
[16:37:54] <cd32fan> i think i will try WMware first :)
[16:38:40] <loke> cd32fan: that's a good choice
[16:46:47] *** elflord has joined #opensolaris
[16:49:34] *** mrdeviant has joined #opensolaris
[16:52:08] <s3tup-> how i wanna install OSS on solaris ?
[16:52:16] *** janus_man has quit IRC
[16:53:14] *** stevel has joined #opensolaris
[16:53:14] *** ChanServ sets mode: +o stevel
[16:53:15] *** janus_man has joined #opensolaris
[16:54:08] *** stevel has quit IRC
[16:54:41] *** MattAFC is now known as MattMan
[16:54:44] *** stevel has joined #opensolaris
[16:54:44] *** ChanServ sets mode: +o stevel
[16:55:52] *** nachox has quit IRC
[16:55:57] *** stevel has quit IRC
[16:56:30] *** stevel has joined #opensolaris
[16:56:30] *** ChanServ sets mode: +o stevel
[17:00:31] *** bronze_|| has quit IRC
[17:01:13] *** Kush- has joined #opensolaris
[17:01:26] *** jsubl2 has joined #opensolaris
[17:01:44] *** glagasse has joined #opensolaris
[17:03:17] *** nwf has joined #opensolaris
[17:06:03] *** sommerfeld has joined #opensolaris
[17:06:03] *** ChanServ sets mode: +o sommerfeld
[17:09:03] *** s3tup- has quit IRC
[17:12:37] *** cd32fan has quit IRC
[17:14:06] *** nwf has quit IRC
[17:15:18] *** __schily_ has joined #opensolaris
[17:22:00] *** Jiko_ has joined #opensolaris
[17:22:29] *** s3tup- has joined #opensolaris
[17:26:49] *** _schily__ has quit IRC
[17:27:52] *** pseudo_ has joined #opensolaris
[17:30:33] *** tsoome has quit IRC
[17:32:33] *** estefg has joined #opensolaris
[17:32:39] *** MattMan has quit IRC
[17:35:11] *** stevel has quit IRC
[17:36:50] *** Jiko has quit IRC
[17:36:51] *** cd32fan has joined #opensolaris
[17:37:00] *** cd32fan has left #opensolaris
[17:37:07] *** cd32fan has joined #opensolaris
[17:37:24] *** Xh4 has quit IRC
[17:38:09] *** stevel has joined #opensolaris
[17:38:10] *** ChanServ sets mode: +o stevel
[17:38:40] <jteo> heya stevel.
[17:38:49] <stevel> 'morning jteo
[17:43:21] <jteo> isn't the hg beta today?
[17:43:45] <estefg> hg?
[17:43:51] *** tsoome has joined #opensolaris
[17:44:59] <cd32fan> hey loke
[17:45:33] <cd32fan> is this the right program?
[17:45:44] <cd32fan> VMware player
[17:47:37] <stevel> jteo: supposed to be - but i'm running into some issues, it may be delayed a couple of days
[17:47:47] <stevel> i'll have an update in the next few hours
[17:47:58] <s3tup-> what should i do if solaris didnt detect my wifi card ? ath0
[17:48:00] *** mazon is now known as Mazon
[17:48:01] <jteo> stevel: thanks!
[17:48:02] <stevel> (if you care, i'm running into issues deploying the SCM software onto the webapp machines)
[17:49:48] <jteo> stevel: :)
[17:51:50] *** cd32fan has quit IRC
[17:58:53] <asyd> anyone knows good kernel debugger and have few minutes to help me to understand why solaris 10 doesn't boot on a C7
[17:59:11] <tsoome> mdb
[17:59:14] <tsoome> :)
[17:59:37] <asyd> [0]> ::stack
[17:59:41] <asyd> kobj_init+0x1c3(100ff18, 0, 100fe98, 1025c00)
[17:59:41] <asyd> _kobj_boot+0x441()
[18:04:26] <eugene> stevel: hi there
[18:04:31] <stevel> hey eugene
[18:04:48] <eugene> stevel: how are you doing?
[18:05:50] <stevel> hurting. :-P i ripped a lot of skin off my hands this weekend climbing, and it's hard typing with blisters and skin missing from your fingertips
[18:06:01] <eugene> whoops :)
[18:06:29] <eugene> do take care.
[18:09:20] *** loke has quit IRC
[18:09:22] *** Error_404 has joined #opensolaris
[18:09:47] *** Dar is now known as Dar_HOME
[18:23:35] <sickness> i'm back
[18:24:04] *** Elendal has quit IRC
[18:29:28] *** Error_404_ has joined #opensolaris
[18:34:39] *** Error_404 has quit IRC
[18:46:46] *** estefg has quit IRC
[18:47:31] *** Cybernd has joined #opensolaris
[18:48:52] *** ktabet_ has joined #opensolaris
[18:58:49] *** Error_404_ is now known as Error_404
[18:59:10] *** ktabet_ has quit IRC
[19:05:09] *** Cyrille_ has quit IRC
[19:06:44] *** GreenHat has quit IRC
[19:07:26] *** bank has joined #opensolaris
[19:07:53] <bank> hii
[19:08:09] <bank> I am going to upgrade ultra sparc 80 from sol 8 to opensol
[19:09:39] *** bank has quit IRC
[19:14:11] *** phips has joined #opensolaris
[19:17:45] <Error_404> ultrasparc 80?
[19:18:10] <stevel> thanks for sharing
[19:18:11] <stevel> :-P
[19:18:40] *** aska_ has quit IRC
[19:20:04] *** bank has joined #opensolaris
[19:20:56] <bank> hi
[19:25:07] *** bondolo has joined #opensolaris
[19:27:31] *** bunker has joined #opensolaris
[19:29:35] *** ShadowHntr has joined #opensolaris
[19:34:16] <richlowe> afternoon all.
[19:34:42] *** bank has quit IRC
[19:45:48] *** cd32fan has joined #opensolaris
[19:46:17] <cd32fan> hello
[19:46:26] <trygvis> hmm .. has anyone else had a problem with a nfs share hanging for a few minutes once in a while?
[19:46:41] <trygvis> I'm using latest osol build and debian unstable
[19:46:43] *** alanc_ is now known as alanc_away
[19:47:14] <cd32fan> does anyone know why my solaris installation keeps aborting
[19:47:28] <cd32fan> it says abort core dumped.
[19:47:31] <tsoome> trygvis: and snoop/tcpdump is telling what?
[19:47:41] <cd32fan> im using WMware btw
[19:48:12] 
[19:49:32] <trygvis> tsoome: what should I be looking for? seems like I only have tcpdump here
[19:50:07] <trygvis> the debian server is under heavy/moderate load, but for some reason the osol client freezes
[19:50:15] <tsoome> what is the last nfs request/response before freeze
[19:51:07] <tsoome> snoop is included in solaris
[19:52:37] <tsoome> also what version of nfs, what transport (tcp or udp), the block size....
[19:53:31] <trygvis> I've limited the client to 3 on the solaris side, other than that I haven't set anything
[19:54:09] <cd32fan> i wish i could stay longer but i have to go to work now
[19:54:29] <cd32fan> cya
[19:54:48] *** yjsong has joined #opensolaris
[19:57:29] *** svoboda has quit IRC
[19:57:49] <trygvis> now I got NFS server deservio not responding still trying
[19:58:20] <trygvis> and now a whole bunch of nfs traffic
[19:59:26] <Kush-> anyone in the bay area and got a sunray DTU? (at home w a relatively good network conneciton)
[20:01:29] *** pseudo_ is now known as Xh4
[20:02:43] <dwc-> if I had to guess,  I'd bet it's linux nfs suckage ;)
[20:03:03] <tsoome> true:)
[20:03:35] *** svoboda has joined #opensolaris
[20:03:40] <trygvis> is that an educated guess or plain trolling?
[20:04:25] <tsoome> linux in not exactly known about reliable implementation of nfs....
[20:04:48] <trygvis> it has worked flawlessly for me for my servers ..
[20:04:57] <dwc-> that's an educated guess
[20:05:08] <tsoome> and this will prove what?
[20:05:10] <dwc-> the nfs implementation on linux isn't known for being compatible with anything but itself
[20:05:31] <sickness> well, every nfs implementation has it's own caveats/quirks :/
[20:05:36] <trygvis> tsoome: and what will your trolling prove? ..
[20:05:56] <trygvis> so, suckyness aside .. anyone got a clue on how I can improve the situation?
[20:06:03] *** Elendal has joined #opensolaris
[20:06:06] <tsoome> I had an client asking for help. they did migrate from linux to solaris server, and they did discover the code working "okay" on linux, was giving core dumps on solaris....
[20:06:23] <dwc-> try nfsv3... try switching from udp to tcp or vice versa
[20:06:35] <tsoome> and smaller block size
[20:06:38] <sickness> try the various nfs options that you have, like udp/tcp proto, modify the buffers, the retry, and the wait time
[20:06:39] <dwc-> tsoome: sparc or x86?
[20:06:40] <trygvis> I'm using v3 only
[20:06:46] <tsoome> sparc
[20:06:50] <trygvis> how/where do I set these options?
[20:06:55] <dwc-> linux on sparc or x86?
[20:06:56] *** [tfb] has joined #opensolaris
[20:07:39] <sickness> usually in the /etc/fstab mount line
[20:07:47] <sickness> or mount -t nfs -o options
[20:07:51] <tsoome> and when I used debugger, the code was just horryble - writing past the end of array etc...
[20:08:22] *** tfb has quit IRC
[20:08:33] <dwc-> core dumps when switching endianness sounds more like doing some non-byte-endian-independent operations more than any inherent brokenness or superiority of one OS over another
[20:09:00] <tsoome> treue, but in that case, the code was plain broken...
[20:09:36] * dwc- shrugs
[20:09:43] <dwc-> every OS has code problems
[20:09:51] <tsoome> but it dis run fine in linux was the argument....
[20:09:56] <trygvis> is the nfsv4 fs in linux useful or too unstable?
[20:10:20] <tsoome> you can try to find answer via google
[20:11:08] <tsoome> of course, your problem may be just plain dns/name resolver issue as well....
[20:11:21] <trygvis> oh sorry, I forgot I wasn't supposed to ask any questions here ..
[20:11:32] <tsoome> ?
[20:11:41] <trygvis> "use google"
[20:11:43] *** jcea has quit IRC
[20:11:48] <tsoome> :)
[20:11:56] <tsoome> well, I'm not the linux expert
[20:12:11] <tsoome> so I can't really tell how good it is or is not
[20:12:42] *** Elendal has quit IRC
[20:12:49] <tsoome> therefore if I do the google or you;) I'd prefer the last option:)
[20:13:17] <trygvis> oki
[20:13:51] <tsoome> anyhow what was the request  before server not responding?
[20:13:58] <trygvis> what's the deal with the dns/name resolver issue?
[20:14:24] <tsoome> well, it depends how you are defining an access to your nfs share
[20:14:31] <trygvis> I'm capturing now but it's currently frozen
[20:14:37] <trygvis> world has access to the share
[20:14:39] <tsoome> if you are using names, thei must resolv
[20:15:03] <trygvis> I do, but the are already connected
[20:15:26] <tsoome> frozen? but the client is repeating the request?
[20:15:34] <trygvis> nope
[20:16:03] <tsoome> ah, then you are probably using tcp transport, try udp instead
[20:16:14] <trygvis> now I suddenly got this: http://rafb.net/paste/results/RzxC6c67.html
[20:16:35] <trygvis> I don't see how I could be using tcp, I haven't set that option anywhere
[20:17:04] <tsoome> it's the default for nfsv3
[20:17:06] <tsoome> and 4
[20:17:15] <trygvis> oh
[20:17:53] <trygvis> are you sure about that? from debians's nfs(5) man: udp            Mount the NFS filesystem using the UDP protocol.  This is the default.
[20:18:02] <tsoome> also, solaris will try to default 32k blocks, you may try to make this smaller at first
[20:18:08] <trygvis> and for the tcp option: Many NFS servers only support UDP.
[20:18:22] <tsoome> udp only is for nfsv2
[20:18:24] <delewis> NFSv3 uses UDP, IIRC, and falls back to TCP
[20:18:31] <delewis> NFSv4 does TCP by default
[20:18:38] <trygvis> http://rafb.net/paste/results/FpdNPl21.html
[20:18:43] <trygvis> next set of traffic
[20:18:54] <trygvis> I guess the three minutes is about a tcp timeout
[20:19:16] <trygvis> now it's back to full speed again
[20:19:27] *** nwf has joined #opensolaris
[20:19:47] *** solarisjon has joined #opensolaris
[20:20:19] <tsoome> I'm pretty sure solaris NFSv3 is tcp by default
[20:20:32] <tsoome> of course, I may be wrong....
[20:20:37] <delewis> maybe it's TCP and falls back to UDP
[20:20:43] <delewis> I always get it mixed
[20:21:02] <delewis> why it would try UDP in the first place when TCP works doesn't make much sense, anyway.
[20:21:08] <tsoome> but afterall I have worked quite a while with nfs servers:)
[20:21:15] <trygvis> now it locked up again, last tcpdump lines: http://rafb.net/paste/results/MQyzRj87.html
[20:22:43] <tsoome> at first - try to use smaller blocks - rsize option for mount_nfs
[20:22:48] <trygvis> how do I set protocol to be udp in solaris (no man pages in osol yet). mount -o proto=udp?
[20:22:54] <trygvis> oki
[20:23:04] *** jcea has joined #opensolaris
[20:23:09] <tsoome> man mount_nfs will help
[20:23:22] <trygvis> not if I don't have any man pages
[20:23:27] <trygvis> :)
[20:23:43] <tsoome> man google ;)
[20:23:51] <tsoome> the net is full of them
[20:24:02] <tsoome> including reference man pages in docs
[20:24:03] <trygvis> yeah, but you also know the line
[20:24:05] <tsoome> docs.sun.com
[20:24:15] <trygvis> ..
[20:24:20] <trygvis> c'mon
[20:24:43] <richlowe> trygvis: exactly as you said.
[20:24:58] <trygvis> thanks
[20:25:24] <richlowe> but udp is not usable for NFSv4
[20:25:37] <trygvis> I'm using v3 againt the linux machine
[20:25:41] <trygvis> against
[20:26:10] *** jacotton_ has quit IRC
[20:26:10] <tsoome> if you read the manual, thare *may* be some hints or clues besides just the option syntax, thats the point
[20:27:36] *** bank has joined #opensolaris
[20:27:39] *** jsubl2 has quit IRC
[20:27:50] <bank> hellooooooo
[20:28:04] <bank> I try to setup opensolaris on ULTRA 80
[20:28:09] <bank> THere are sol8!
[20:28:22] <bank> I finding where is cdrom ... it can't boot from cdrom!
[20:28:28] <bank> so ... I have to mount it ????
[20:28:39] <bank> I try to find ... but I don't know where it is ....
[20:28:44] <tsoome> you may have too old firmware
[20:28:54] <bank> I have c0t0d0s0 at /
[20:29:12] <tsoome> or just wrong cd;)
[20:29:21] <bank> and c0t0d0s7 at /mnt
[20:30:16] <bank> errr ....
[20:30:25] <bank> DO you know where is cdrom?
[20:30:36] <bank> I have ultra 80 beside me.
[20:30:48] <bank> :'(\
[20:31:19] <bank> how can i upgrade sol8 in ultra 80?
[20:31:24] <mrdeviant> um, what does "boot cdrom" at the OF prompt do?
[20:31:56] <bank> command not found
[20:31:57] <bank> at the OF?
[20:32:00] <bank> what is OF?
[20:32:14] <bank> I am searching ...
[20:32:36] <tsoome> do init 0 and then once you will get ok> prompt, type boot cdrom
[20:32:55] <mrdeviant> open firmware
[20:33:12] <bank> hey init 0 is restart. ..
[20:33:32] <dwc-> no
[20:33:37] <dwc-> init 0 is shutdown to firmware
[20:33:52] <dwc-> reboot is 5
[20:33:56] <dwc-> errr 6
[20:34:06] * dwc- can't type today
[20:34:22] *** dolske has joined #opensolaris
[20:34:31] *** slowhog has joined #opensolaris
[20:34:33] <bank> I ever upgrade sol . I use init 6 ...
[20:35:01] <mrdeviant> you need to drop to firmware and boot from the cdrom
[20:35:05] <bank> | / _ | \
[20:35:10] <mrdeviant> unless you're trying to do live upgrade
[20:35:25] <bank> there are / moving now.
[20:35:54] <bank> slow motion :(
[20:36:07] <jbalint> what is dvma?
[20:36:10] <bank> so excite found the old ultra sparc at university lab.
[20:36:47] <bank> DVMA - Direct Virtual Memory Access
[20:36:47] <bank> or Delaware Valley Medieval Association :P
[20:37:08] <jbalint> hehe, thanks
[20:37:15] <tsoome> .oO so, someone still knows about google:D
[20:37:27] <trygvis> o.O
[20:37:34] <bank> :P
[20:38:26] <tsoome> but I liked the U2 question the best;)
[20:39:00] <bank> the " / " still turn around ...... so long......
[20:39:05] <bank> ultra 80
[20:39:06] *** mrdeviant has quit IRC
[20:39:46] <bank> setting up java please wait. ....
[20:40:03] <tsoome> it's old machine and your cdrom dev is probably not the fastest one....
[20:40:46] <bank> Could you compare ultra 80 400 mhz to intel x mhz.
[20:41:00] <bank> 10 years :O
[20:41:18] <tsoome> it depends.
[20:42:13] <tsoome> on workload type etc...
[20:43:22] <bank> I got the gui now :D
[20:45:48] <bank> solaris 8 is look so classic
[20:46:02] *** bubbva__ has quit IRC
[20:46:39] *** _william_ has joined #opensolaris
[20:46:55] <quasi> you can make sol10 look just like it
[20:47:27] <jbalint> hi _william_
[20:48:47] *** cd32fan has quit IRC
[20:48:48] <_william_> hi all
[20:48:51] <_william_> hi jbalint
[20:49:31] <bank> hi quasi
[20:49:38] <bank> batman time.
[20:52:34] <bank> sun monitor is very very big
[20:53:45] <tsoome> :D
[21:02:52] *** andre1 has joined #opensolaris
[21:22:27] <bank> hey
[21:22:34] <bank> I finish CD1 then i say restart
[21:22:41] <bank> so ... how about CD 2 3 4 5 6
[21:23:02] <tsoome> after restart you will be asked to insert...
[21:24:37] *** axisys has joined #opensolaris
[21:25:03] <axisys> can someone explain why it says 30 shaes instead of 60 shares here http://www.rafb.net/paste/results/NF2dz299.html
[21:25:14] <axisys> this is an exerpt from brendan's page
[21:25:22] <axisys> http://users.tpg.com.au/adsln4yb/zones.html this page
[21:35:44] <quasi> axisys: because shares != %
[21:35:52] <gdamore> hi *
[21:36:50] <axisys> quasi: is it exactly half? he is setting up four zones and he has 8 cpus
[21:36:51] <quasi> axisys: if you hand out 50 shares in total, 30 of those make 60% of the total shares
[21:37:18] <axisys> quasi: i c
[21:38:06] <axisys> so shares would be multiple of number of zones i have.. sorta?!
[21:38:11] <quasi> no
[21:38:45] <quasi> shares is a function of how many you hand out
[21:38:46] <axisys> well would be a multiple of how many zones i put under FSS?
[21:38:58] <quasi> no
[21:39:05] <axisys> hmm..
[21:40:19] <axisys> on this page http://users.tpg.com.au/adsln4yb/zones.html brendan decided to give 60% to zone1 .. so how did he pick 30 shares?
[21:40:53] <quasi> because the total number of shares handed out is 50 - 60% of that is 30
[21:41:27] <quasi> get it?
[21:41:55] <axisys> how did u decide how many shares to hand out?
[21:42:06] <axisys> or how do u rather
[21:42:07] <richlowe> axisys: shares aren't anything but a value relative to other shares.
[21:42:40] <richlowe> The total number would seem to do little other than affect the granularity you have.
[21:42:43] <bank> I think it not gonna ask CD 2 anymore
[21:42:48] <bank> it goes in to CDE
[21:43:40] <axisys> richlowe: so say i have 5 zones and I have 8 cpus.. i decided to have zone1 with 50% of shares
[21:43:55] <axisys> richlowe: so that would be 20 share?
[21:44:04] <_william_> gn
[21:44:05] *** _william_ has left #opensolaris
[21:44:14] <bank> can I run setup again..
[21:44:28] <quasi> axisys: could be 20 or could be 200 - it all depends on what scale you choose
[21:44:44] <axisys> quasi: gotcha
[21:44:55] <bank> oh i got step 2 now
[21:44:56] <bank> sorry
[21:45:53] <quasi> axisys: so you want 50% for one zone and the other 4 to share the rest equally?
[21:45:56] <axisys> can i have zone1 with 4 cpus fixed and other 4 zones to share 4 cpus?
[21:46:07] <quasi> yes
[21:46:13] <bank> and it disappear...
[21:46:21] <axisys> quasi: (yes)
[21:46:32] <bank> see the blue background with sendmail notify.
[21:46:47] <quasi> axisys: so give 4 shares to the big zone and 1 to each of the other
[21:47:11] <quasi> (except you _really_ don't want to forget the global zone ;)
[21:47:50] <bank> I see..because the machine is too slow
[21:47:55] <axisys> quasi: how do i make sure global zone get enough shares beside putting in SMF script?
[21:48:15] <axisys> with prctl command in it
[21:48:30] <quasi> axisys: read the rest of the guide ;)
[21:49:04] <axisys> quasi: according to that guide use SMF for global to get enough shares.. i was wondering if there is any other ways
[21:51:39] <quasi> keep it out of the FSS
[21:54:07] *** golqm-noob has quit IRC
[21:56:16] *** dolske_ has joined #opensolaris
[21:57:42] <trygvis> just FYI, setting proto=udp,rsize=4096 made the nfs stuff work
[21:57:57] <bank> there are white screen said about media error over the installation progress..
[21:58:00] <bank> paint over
[21:58:10] <axisys> quasi: global zone is out of FSS now ..
[21:58:34] <axisys> quasi: so can i have one zone with a fixed number of cpus while other zones share the rest of cpus?
[21:59:07] <axisys> quasi: i guess i can .. i guess that what u unanswered when u said keep them out of FSS
[21:59:42] <quasi> yes, I'm pretty sure Brendan had an example of that - binding processor sets to different pools
[22:00:43] <axisys> yeap he does .. not sure if i could mix and match
[22:01:13] <quasi>         pool httpd-pool
[22:01:13] <quasi>                 int     pool.sys_id 8
[22:01:31] <quasi> string  pool.scheduler FSS
[22:01:43] <quasi> pset    pset_default
[22:02:02] <quasi> could have been another pset
[22:02:29] <asyd> the kernel flag for a reconfigure boot is -r right ?
[22:02:38] <richlowe> Yes.
[22:02:42] <asyd> quasi: you remember my story about the dedicatedserver ?
[22:02:58] <asyd> I imported via dd a ufs from another server to a new one
[22:03:11] <asyd> when I try to boot it, I have a kernel error 'can't find root fs'
[22:03:14] <axisys> quasi: i c
[22:03:19] <asyd> but when I boot in safe mode
[22:03:36] <asyd> it ask me if I want to mount c0t0d0 as /a
[22:03:41] <axisys> so pool.sys_id 8 means processor number 8 correct?
[22:03:44] <asyd> so the problem is not a driver one
[22:04:06] <quasi> axisys: wrong
[22:04:23] <axisys> quasi: never mind :P
[22:04:24] <quasi> axisys: just a number on the pool that the httpd zone is in
[22:04:43] <axisys> quasi: i realized after i asked u the question..
[22:04:44] <quasi> asyd: let me get you a howto for updating the boot archive
[22:04:46] <asyd> I think I'm pretty close to success to boot this server with solaris ~{~#
[22:04:53] <asyd> quasi: thanks
[22:05:56] <bank> there are media fail white paint over the setup progress windows
[22:06:10] <bank> white and black font print over the setup progress windwos
[22:06:13] <asyd> quasi: I now have a serial access to the server, it's more easy :)
[22:06:48] <quasi> asyd: boot in failsafe from the grub menu and say y to mount rw under /a
[22:07:08] <quasi> then  rm -f /a/platform/i86pc/boot_archive
[22:07:18] <quasi> bootadm update-archive -R /a
[22:07:37] <quasi> when that is done, init 6 and you should be fine
[22:07:52] <asyd> hmm ok
[22:08:00] <asyd> let me try that
[22:13:02] *** dolske has quit IRC
[22:13:18] * asyd crosses his finger
[22:13:42] <bank> media error a lot
[22:15:45] <asyd> quasi: same error:/
[22:15:50] <asyd> I don't understand
[22:16:25] <asyd> how solaris x86 read the partition table and or mapping beetween /dev devices and real device ?
[22:16:29] <quasi> root must be set wrong - different ordering of disks
[22:16:32] <asyd> panic[cpu0]/thread=fec1e1a0: cannot mount root path
[22:17:03] <asyd> quasi: do you think it's possible than my disk will have a differrent device name in fail safe rather in the normal 'boot' ?
[22:17:21] <quasi> shouldn't have
[22:17:42] <asyd> and now I'm pretty sure it's not a SATA drier problem since the failsafe kernel is able to mount my /
[22:18:38] <quasi> check it out in failsafe and see where it is and then look to adding mount root stuff to the regular boot menu
[22:18:47] <asyd> Properties can be set in /boot/solaris/bootenv.rc
[22:18:52] <asyd> hmm i should take a look in this file
[22:19:35] * quasi nods and also points to the grub menu
[22:20:52] <bank> quasi: What is your current time?
[22:21:08] <bank> do you live in australia.
[22:24:08] <asyd> hmm
[22:24:18] <asyd> setprop bootpath /pci@0,0/pci-ide@7/ide@0/cmdk@0,0:a
[22:24:23] <asyd> I think this line is not good
[22:25:17] <quasi> so boot in failsafe and see what that gets you
[22:25:28] *** glagasse has quit IRC
[22:25:31] <asyd> I'm in failsave, but not sure how to have the correct oone :)
[22:25:56] <bank> no one talk to me :'(
[22:27:05] <asyd> hmm I think I guess
[22:27:18] <asyd> /dev/dsk/c0d0s0 -> ../../devices/pci@0,0/pci-ide@f/ide@0/cmdk@0,0:a
[22:27:26] <quasi> uhm, mount -> dev path -> link to something liek /pci@0,0/pci-ide@7/ide@0/cmdk@0,0:a
[22:27:31] <quasi> yeah
[22:28:25] <bank> The CD 2 is failed.
[22:28:37] <bank> and It ask for CD3 and I continue
[22:28:42] <jbalint> whats the cc option to show which headers in include which?
[22:29:25] <jbalint> THanks.
[22:29:39] <asyd> pfff dunno why but editing a file in safe mode even with vi TERM=vt10 is very difficult
[22:30:07] <quasi> vt10 ?
[22:30:31] <quasi> asyd: usually because it is set to a fixed size
[22:30:49] <quasi> asyd: ctrl-l usually helps
[22:30:50] <asyd> vt100 sorry
[22:30:54] <asyd> not here :/
[22:31:08] <bank> Will that a problem if CD2 is failed...
[22:31:17] <rydis> Just use ed, like ghod intended.
[22:31:19] <asyd> though I access the serial port via an old cisco 2511
[22:31:34] *** elflord has quit IRC
[22:31:41] <asyd> rydis: well, I thinking about ed, but no, no way for me to use ed in 2006 :)
[22:32:56] <quasi> asyd: if you have grep, you can find the line and the line no
[22:33:10] <bank> gd night
[22:33:13] *** bank has left #opensolaris
[22:33:14] <asyd> yeah that's exactly the way I'll do :p
[22:33:21] <quasi> asyd: then it is a simple thing to replace in vi
[22:33:23] <asyd> hmhm err, echo >> is my friend
[22:33:57] <asyd> both, in fact
[22:34:40] <asyd> ok, time to test
[22:36:13] <dwc-> what's wrong with vt100?
[22:36:40] <dwc-> I think every terminal emu in existence supports vt100
[22:36:51] <asyd> yeah, me too
[22:37:40] <asyd> it'svery strange than backspace, arrows work. But editing a file with vi is horrible, more or less impossible
[22:37:54] <rydis> The problem most likely is that you have cols and rows wrong, I think. Does stty give you any help there?
[22:38:08] <dwc-> why would arrows work?
[22:38:08] <richlowe> that sounds like a reasonable assumption.
[22:38:27] <dwc-> backspace sounds like an stty problem as well
[22:38:32] <quasi> rydis: yes, that is the problem because the console goes by a fixed number
[22:38:40] <asyd> ahh
[22:38:46] <dwc-> or you can alwyas try the delete key, the ^H combination, shift-backspace, etc.
[22:38:49] <asyd> I maybe doesn't have 80 cols right
[22:39:00] <asyd> hmm i have
[22:39:11] <quasi> delete usually works as backspace
[22:40:18] <asyd> pff, still have the same error :/
[22:41:16] *** gisburn has joined #opensolaris
[22:41:30] <quasi> check your grub setup for oddities
[22:41:49] <asyd> can I set any booten.rc value from grub with the -B argument ?
[22:42:05] <gisburn> Hi!
[22:42:30] <gisburn> Why does "add_install_client"'s "-i" option not support IPv6 addresses ?
[22:42:41] <asyd> ah, interesting, booting in fail save update my bootarchive
[22:44:30] <quasi> isn't that what we did earlier?
[22:44:36] <asyd>  hmm, the size are far to be the same
[22:45:37] <quasi> is the clock that far off?
[22:46:41] <quasi> or is you tz another than I guess?
[22:46:48] <asyd> yeah, though I never set it to the correct date, it's just a test box
[22:47:07] <asyd> oh oh oh
[22:47:15] <asyd> yeahhhhhhhhhhhhhhhhhh
[22:47:27] <tsoome> gisburn: because bootparams and tftp is IPv4 only
[22:47:31] <tsoome> ?
[22:47:44] <asyd> Root password for system maintenance (control-d to bypass):
[22:47:45] <asyd> \o/
[22:47:53] <gisburn> tsoome: I tought tftpd can use UDP6, too...
[22:47:56] <alanc> or it could just be a bug - ask on install-discuss or file a bug report
[22:48:07] <tsoome> true:)
[22:48:33] <gisburn> alanc: Yeah, I still have to file a zillion new bug repprts
[22:48:56] <gisburn> alanc: someone turned off rstatd by default in B48 and then sdtperfmeter runs amok
[22:49:01] <quasi> gisburn: no time like the present to get started ;)
[22:49:08] <Xh4> ...
[22:49:16] <alanc> did you ever get your old web page list of bugs recovered and start turning those into bug reports?
[22:49:18] * gisburn kills quasi to have some fun
[22:49:31] <asyd> it's very stupid, I don't have the root password \o/
[22:49:32] <alanc> gisburn: probably part of the secure-by-default changes in b43
[22:49:34] <gisburn> alanc: I have a new bug list.
[22:49:37] <elektronkind> GooTube is official.
[22:50:00] <quasi> asyd: that's easy to fix
[22:50:16] <gisburn> alanc: but I have more trouble right now. Seems they want to close our institute... finally... ;-(
[22:50:17] <asyd> yeah i know
[22:50:50] <richlowe> alanc: though if it's depended on in such a way, it should have been one of the "add a local-only flag" services.
[22:51:17] <richlowe> gisburn: how exactly does it run amok?
[22:51:23] <richlowe> gisburn: all it does to me is tell me to enable rstatd. :)
[22:51:36] <gisburn> 100% CPU usage
[22:52:24] <alanc> yeah - but the people who did it may not have realized sdtperfmeter needed it
[22:52:26] <richlowe> alanc: fixed in snv_49 perhaps?
[22:52:39] <richlowe> since gisburn is seeing it thrashing, and I'm getting a one-line message then exit
[22:53:55] <alanc> 	4168119 "disabling rstatd causes an empty sdtperfmeter window which consumes a lot of cpu" was supposed to be fixed in S7 to keep it from running amok
[22:54:26] <alanc> I don't see any recent changes to sdtperfmeter
[22:55:19] <tsoome> perhaps there is a rstatd client library:)
[22:55:45] <gisburn> is there any bug that B48/SPARC network traffic can make a S8/SPARC box hang hard ?
[22:56:49] <alanc> if there was a client library, it'd be in ON so you'ld see the bug listed in the ON changelogs
[23:00:04] *** jcea has joined #opensolaris
[23:02:20] <gisburn> is there any bug that B48/SPARC network traffic can make a S8/SPARC box hang hard ?
[23:03:06] <gisburn> umpf
[23:03:14] <gisburn> where are the links in the /topic ?
[23:03:15] <gisburn> grrrr
[23:03:57] <alanc> they seem to have been moved to the welcome /msg
[23:03:59] *** deedaw has joined #opensolaris
[23:04:08] *** Andrew____ has quit IRC
[23:04:12] <alanc> chanserv said "[#opensolaris] Welcome to #opensolaris. Please see the following URLs for more information. 'The OpenSolaris Wiki' @ http://www.genunix.org/wiki, 'What is OpenSolaris?' @ http://whacked.net/what_is_opensolaris, and the #opensolaris IRC channel log @ http://www.uwyn.com/drone/log/bevinbot/opensolaris. Of course, please visit the main OpenSolaris website at http://www.opensolaris.org"
[23:04:40] <gisburn> umpf
[23:04:56] <quasi> gisburn: blame stevel
[23:05:07] *** yjsong has quit IRC
[23:05:32] <gisburn> alanc: the chanserv message is already lost in /dev/null - the history in this client doesn't go back that far
[23:05:34] *** Gman has joined #opensolaris
[23:05:39] * gisburn looks at stevel
[23:05:59] <alanc> it showed up in my server messages tab, which gets very little traffic, so I still had it
[23:06:02] <richlowe> what is it you're actually looking for?
[23:06:16] <richlowe> ... beyond something to complain about, obviously ;)
[23:07:29] * gisburn consideres the options: Boiling stevel in hot oil, boil him in hot water, boil him in chicken soul, rip stevel into pieces, use stevel as test dummy for the next north korean atom bomb test, throw stevel into a deep pit filled with komodo dragons... does anyone have better ideas ?
[23:07:37] <Gman> now i know what it's like to run gnome on a sunblade 100 as my 'work' machine :(
[23:07:39] <libkeiser> thanks for making the topic sane, stevel :)
[23:07:43] <gisburn> s/soul/soup/
[23:07:51] * gisburn kicks libkeiser
[23:08:04] <gisburn> libkeiser: yeah, the /topic is now sane - and useless
[23:08:12] <Gman> no it's not
[23:08:15] <alanc> any of those options would seem to keep stevel from working on the external SCM, which would be bad
[23:08:20] * gisburn kicks Gman
[23:08:23] * Gman agrees with both libkeiser and stevel
[23:08:29] * gisburn kicks Gman harder
[23:08:40] *** stevel sets mode: +o Gman
[23:08:42] <Gman> you're in the minority ;)
[23:08:43] <stevel> feel free to kick back
[23:08:44] <axisys> i had to run this to give global zone 10 shares of CPU `prctl -n zone.cpu-shares -v 10 -r -i zone global'
[23:08:45] <stevel> ;-)
[23:08:50] <gisburn> Gman: want to be booiled in chicken soup ?
[23:08:57] <alanc> Gman: that's what I've got as my home desktop - upping the RAM to 1GB helped a bit
[23:09:03] <axisys> how do i make sure it always get 10 shares .. SMF is the only way?
[23:09:14] <Gman> alanc, this is a bog standard machine i think
[23:09:20] <alanc> which since it was something sane, like PC100 or PC133, wasn't too expensive
[23:09:21] <Gman> i'm not going to waste any of sun's money on it ;)
[23:09:44] <asyd> ah, very strange
[23:09:50] <asyd> on /pci@0,0/pci-ide@f/ide@0/cmdk@0,0:a
[23:09:54] <asyd>  / on /pci@0,0/pci-ide@f/ide@0/cmdk@0,0:a
[23:09:55] <asyd> even
[23:09:58] <alanc> definitely not a speed demon, even when it was new
[23:09:59] <asyd> I'm not able to write on the fs
[23:10:52] *** karrotx_ has joined #opensolaris
[23:12:08] <richlowe> asyd: single user?
[23:12:15] <asyd> yeah
[23:12:20] <richlowe> remout it read-write
[23:12:26] <richlowe> remount, too.
[23:12:57] <asyd> how ? mount -o remount,rw / ?
[23:13:01] <gisburn> wheeee, ksh93 gets a "sync" builtin
[23:13:38] *** dolske_ has quit IRC
[23:14:27] <asyd> # ls -l /dev/rdsk/c0d0s0
[23:14:27] <asyd> /dev/rdsk/c0d0s0: No such file or directory
[23:14:30] <asyd> interesting
[23:14:51] <asyd> 0
[23:15:26] <asyd> format doesn't found any disk \o/
[23:15:37] <quasi> ls -l /dev/rdsk
[23:15:52] <asyd> I have some entries
[23:16:49] *** dolske has joined #opensolaris
[23:17:07] <asyd> but hmm, since format doesn't found entry, I don't know which one I should use
[23:17:26] <asyd> I even wonder how I can have a prompt in the solaris :) sounds a bit magical
[23:17:41] <quasi> mount
[23:18:40] <asyd> on /pci@0,0/pci-ide@f/ide@0/cmdk@0,0:a
[23:18:57] <asyd> that is the one I specified in bootpath in bootenv.rc
[23:19:30] <quasi> so what else is in /pci@0,0/
[23:19:55] <asyd> hmm
[23:20:01] <asyd> c1d0s5 -> ../../devices/pci@0,0/pci-ide@7/ide@0/cmdk@0,0:f,raw
[23:20:13] <asyd> ah sorry
[23:20:29] <asyd> well, no device with ide@7 in /dev/rdsk
[23:20:39] <axisys> let me rephrase my question.. how do i make sure global zone always get 10 shares? run prtctl everytime it reboots is the only way?
[23:21:01] <axisys> 10 share of CPU that is .. or 1 CPU.. either way
[23:23:07] <asyd> I tried fstyp on every deviced I had on /dev/rdsk , found nothing
[23:23:51] *** axisys has quit IRC
[23:24:08] *** solarisjon has quit IRC
[23:24:08] <asyd> the /dev is dynamically and automagically populated, right ?
[23:24:16] <quasi> asyd: devfsadm -C and devfsadm -v
[23:24:36] <asyd> well I tried to run devfsadm but it complain that / is in ro
[23:24:49] <asyd> let me retry
[23:26:20] <asyd> # devfsadm -C
[23:26:20] <asyd> devfsadm: mkdir failed for /dev 0x1ed: Read-only file system
[23:26:55] *** karrotx has quit IRC
[23:27:47] <quasi> of course
[23:27:58] <quasi> as long as it is read only
[23:28:00] *** karrotx_ has quit IRC
[23:28:10] <asyd> yeah.. but I don't know how to remount it in rw :/
[23:30:37] <asyd> since -o remount complain about the /dev file doesn't exist
[23:33:50] <gisburn> Does anyone know when "The Return, Part 2" will be aired in the US ?
[23:34:27] <gisburn> Or "The Quest, Part 2" ?
[23:37:17] <asyd> I don't success to make a reconfigure boot i don't understand why (at least I don't see the "Reconfiguring devices" message)
[23:39:55] *** Mazon is now known as mazon
[23:40:22] <asyd> hmm, the -a kernel option seem very interesting
[23:40:25] *** LeftWing has quit IRC
[23:42:16] <asyd> err, useless
[23:42:28] *** LeftWing has joined #OpenSolaris
[23:42:30] *** Jtara has joined #opensolaris
[23:46:48] * dolske grumbles...
[23:47:42] <dolske> Anyone familiar with running Nevada under Parallels? I've got it installed, but using DTrace locks it up hard. *sigh*
[23:49:19] *** bunker has quit IRC
[23:49:46] <asyd> http://kaoru.asyd.net/~asyd/dedibox-solaris.txt
[23:49:54] *** nwf has quit IRC
[23:50:11] <Gman> the default setup in thunderbird is really not very nice
[23:50:16] <Gman> we should try fixing that..
[23:50:25] <ShadowHntr> what skin of evolution do they use with nexenta?
[23:52:02] <Gman> theme? likely to be something liike clearlooks
[23:52:02] <Jtara> Gman.. when would it be fixed?
[23:53:21] <asyd> it's is possible to populate /a/dev from a failsafe kernel ?
[23:54:01] <Gman> Jtara: i dunno, i guess i have to get people agreeing with me first...
[23:54:17] <Gman> plus there's the issue of modifications to thunderbird not preserving their branding thing
[23:54:26] <Jtara> ah okay
[23:58:53] <asyd> hmm running devfsadm in a chroot was a stupid idea





top