[00:07:38] *** rmorse has quit IRC
[00:08:22] *** rmorse has joined #opensolaris
[00:10:39] *** rmorse has quit IRC
[00:12:25] *** eokyere has quit IRC
[00:12:41] *** smferris has quit IRC
[00:13:58] *** rmorse has joined #opensolaris
[00:17:33] *** Tpenta has joined #opensolaris
[00:17:38] *** trs81 has quit IRC
[00:19:35] *** coffman has quit IRC
[00:19:39] *** laca has joined #opensolaris
[00:30:35] *** nwf has quit IRC
[00:30:51] *** rmorse has quit IRC
[00:31:35] *** rmorse has joined #opensolaris
[00:39:34] *** piwi has joined #opensolaris
[00:43:05] *** ChanServ sets mode: +o Tpenta
[00:51:00] *** aliquis has joined #opensolaris
[01:08:36] *** _dreams_ has quit IRC
[01:13:14] *** McBofh has joined #opensolaris
[01:13:27] * McBofh re-caffeinates
[01:13:57] <boyd> Morning, McBofh
[01:20:31] <McBofh> hi boyd, one-n-all
[01:23:44] *** Kernel86 has quit IRC
[01:25:32] *** B|nTaRa has quit IRC
[01:27:20] <Gman> http://mail.opensolaris.org/pipermail/opensolaris-announce/2006-October/000291.html - OpenSolaris Weekly News #33
[01:27:21] *** nwf has joined #opensolaris
[01:29:46] <McBofh> boyd: mirrored my root with svm for the moment, but I've just created a zpool called "sink".
[01:29:59] <McBofh> guess what I called its filesystems :)
[01:30:08] <Auralis>  kitchen
[01:30:32] <McBofh> sink/hole and sink/swim
[01:30:40] <McBofh> I should have a kitchen though, that's a good idea
[01:30:49] * McBofh fires up another putty session....
[01:35:33] *** Kernel86 has joined #OpenSolaris
[01:39:21] *** drio has quit IRC
[01:41:15] *** bengtf has quit IRC
[01:42:41] *** deather_ has joined #opensolaris
[01:44:34] *** sparc-kly has quit IRC
[01:47:18] *** MadMethod has joined #opensolaris
[01:47:26] <MadMethod> so who here runs tx?
[01:47:35] *** sparc-kly has joined #opensolaris
[01:47:36] *** ChanServ sets mode: +o sparc-kly
[01:47:41] <twincest> what is tx?
[01:49:27] <MadMethod> that means you don't run it :)
[01:49:30] <MadMethod> trusted extension
[01:49:39] <MadMethod> *s
[01:49:58] <twincest> i'm fairly sure i don't, i was just curious what it was
[01:50:47] <McBofh> MadMethod: I'm not trusted to run tx
[01:52:43] *** Kernel86 has quit IRC
[01:52:54] <MadMethod> ok, maybe you can answer this anyway
[01:53:04] <MadMethod> say there is a .class file in file browser
[01:53:15] <MadMethod> what do i need to do to make file browser correctly execute that class?
[01:53:30] *** rmorse has quit IRC
[01:54:06] <MadMethod> nm, guess i can jsut do it from a terminal
[01:59:26] *** deather has quit IRC
[02:01:23] *** mikaeld has quit IRC
[02:01:28] *** ShadowHntr has joined #opensolaris
[02:01:39] *** mikaeld has joined #opensolaris
[02:02:11] *** trs81 has joined #opensolaris
[02:02:26] <McBofh> MadMethod: .class files are java bytecode
[02:03:48] <McBofh> MadMethod: so if you run        java -classpath filename.class Main      (or whatever the main-equivalent method is) then you should be able to see something useful
[02:04:28] *** Kernel86 has joined #OpenSolaris
[02:07:24] *** lloy0076 has joined #opensolaris
[02:07:55] *** ManicMethod has joined #opensolaris
[02:09:30] *** ManicMethod is now known as Method
[02:17:06] *** MadMethod has quit IRC
[02:17:39] *** Elendal has joined #opensolaris
[02:17:40] *** OnkelSchorsch has joined #opensolaris
[02:26:44] <klocze1> can I use SFM on Sol 10 for change sendmail binding to ip/interface or it must be performed by chage sendmail configuration ?
[02:32:51] *** McBofh has quit IRC
[02:38:43] <lloy0076> Is the only way to update Open Solaris ON 48 to ON 49 to download the DVD again?
[02:40:11] <g4lt-mordant> look for BFU
[02:40:14] <Gman> you can download the bfu archives and update that way
[02:40:17] <Gman> only works for ON though
[02:40:29] <Gman> alternatively you can download the source and build it ;)
[02:41:36] <Method> Gman: do you know if any tx people ever come in here?
[02:41:53] <Gman> i dont' think so
[02:42:08] <lloy0076> Downloading the source and compiling would probably take too long on my Athlon XP 2400+ - the BFU's sound the way to go.
[02:42:10] <Method> :\ so much for it having a "vibrant" community
[02:42:22] <lloy0076> I don't suppose the BFUs are in any obvious place?
[02:42:27] <Gman> yeah, watch this space
[02:42:36] <Gman> i plan to call jim gris on that one ...
[02:43:40] <Gman> lloy0076: lemme find
[02:45:30] <Gman> lloy0076: http://dlc.sun.com/osol/on/downloads/current
[02:45:45] <Gman> contains bfu archives from the latest source drop
[02:46:06] <lloy0076> Thanks :)
[02:46:13] <Gman> read the README to start off
[02:46:46] <lloy0076> Why can't I be like everyone else and not read the README? :(
[02:46:52] * lloy0076 rofl
[02:47:16] <Gman> well, it'll probably help if you bfu an archive and get weird messages
[02:59:20] *** lloy0076 has left #opensolaris
[03:03:50] *** neoxed has quit IRC
[03:04:03] *** neoxed has joined #OpenSolaris
[03:04:17] * sparc-kly is away: @ university design project
[03:08:42] *** telpochyaotl has joined #opensolaris
[03:08:52] *** McBofh has joined #opensolaris
[03:10:50] <telpochyaotl> i've been looking unsuccesfully for the instructions on how to allow regular users to run inetmenu, can someone share with me what the file in /etc is to add the username?
[03:10:51] <telpochyaotl> ty!
[03:11:51] *** lisppaste3 has quit IRC
[03:14:27] *** lisppaste3 has joined #opensolaris
[03:17:10] <McBofh> telpochyaotl: I think you'll need to give those users the correct profile entries
[03:17:14] <McBofh> man profile(4) iirc
[03:27:44] *** simford has joined #opensolaris
[03:28:52] *** piwi has quit IRC
[03:32:14] <telpochyaotl> ty
[03:32:31] <telpochyaotl> thats what i needed... i'll try it now
[03:32:32] *** telpochyaotl has left #opensolaris
[03:32:51] *** B|nTaRa has joined #opensolaris
[03:36:06] *** nwf has quit IRC
[03:36:24] <Doc> evening
[03:40:11] *** udos has joined #opensolaris
[03:44:15] *** Method has quit IRC
[03:47:38] *** sparc-kly_ has joined #opensolaris
[03:48:23] *** sparc-kly__ has joined #opensolaris
[03:48:47] *** sparc-kly__ has quit IRC
[03:53:23] *** nyati has joined #opensolaris
[03:58:23] <Gman> richlowe: there?
[04:00:16] *** solarisjon has joined #opensolaris
[04:00:28] *** solarisjon has left #opensolaris
[04:04:29] *** sparc-kly has quit IRC
[04:07:35] *** B|nTaRa has quit IRC
[04:07:39] *** sparc-kly_ has quit IRC
[04:10:37] *** halton has joined #opensolaris
[04:10:38] <McBofh> Doc: !! welcomeback
[04:11:30] *** nyati_ has quit IRC
[04:12:25] <Doc> back?
[04:12:34] <McBofh> or are you still on holidays?
[04:13:17] <Doc> unless they have changed Sydney a lot whilst Ive been gone, i think im still in peru
[04:13:33] <McBofh> right
[04:13:42] <McBofh> enjoying it?
[04:13:55] <Doc> http://www.docbert.org/flickr  if you care
[04:14:06] * McBofh does
[04:14:10] <Doc> indeed
[04:15:59] <McBofh> are the trees in http://www.flickr.com/photos/docbert/270816925 what I suspect they are?
[04:16:39] <Doc> nfi what they are
[04:18:04] <Doc> anyway, outa here....
[04:18:13] *** Method has joined #opensolaris
[04:18:30] <McBofh> Doc: hope you're having fun
[04:21:20] *** ShadowHntr has quit IRC
[04:23:55] *** besonen_mobile has quit IRC
[04:24:42] *** yongsun has joined #opensolaris
[04:28:08] *** besonen_mobile has joined #opensolaris
[04:31:53] *** jteo_ is now known as jteo
[04:38:42] <jteo> http://www.red-bean.com/dav/presentations/Poisonous-people.pdf is intriguing. ;)
[04:39:10] <eugene> jteo: how's things
[04:40:04] <jteo> eugene: so-so. you?
[04:40:51] <eugene> jteo: in bangalore now
[04:41:06] <jteo> eugene: work or play?
[04:41:23] <eugene> jteo: heh, i have been traveling across asia pacific for the past 2 months. didn't i tell you? :P
[04:42:12] <jteo> eugene: ah.
[04:44:00] <eugene> 1jteinteresting slides
[04:49:14] *** laca has quit IRC
[05:02:13] <boyd> Gman: You still around?
[05:08:24] *** Kernel86 has quit IRC
[05:11:31] *** karrotx has quit IRC
[05:13:19] <Gman> boyd: yep
[05:13:23] <Gman> boyd: what's up?
[05:13:26] <boyd> Hey,
[05:13:35] <boyd> I thought you may have a definitive answer on this...
[05:13:40] <Tpenta> hi boyd & glynn
[05:13:56] <Gman> hey Tpenta
[05:14:05] <boyd> is it possible for a user to customise their launch->Applications menu at all?
[05:14:08] <boyd> Hey Tpenta
[05:14:15] <Gman> boyd: yes
[05:14:38] <Gman> the degree of difficulty depends on what version of gnome you're using
[05:14:47] <Gman> the absolute top level launch menu, you can't
[05:14:54] <Gman> just the applications/preferences
[05:15:28] <boyd> Ok... I'm interested in what's in b49 (2.14?) and 2.6
[05:15:40] <boyd> (it is 2.6 in s10GA?)
[05:15:45] <Tpenta> on the subject of gnome versions, ...
[05:15:50] <Gman> 2. in GA
[05:15:55] <Gman> erk. 2.6
[05:16:03] <Tpenta> i note that tamarack put back the other day; any news on time frame for 2.16?
[05:16:04] <Gman> b49 is 3.14
[05:16:09] <Gman> erk 2.14
[05:16:12] <boyd> hehe
[05:16:31] <Gman> 2 differnt menu mechanisms - 2.14 should be easy, 2.6 is buggy
[05:16:33] <boyd> I think your keyboard must be getting better at dodging
[05:16:41] <Gman> Tpenta: probably within about 2 or so builds
[05:16:48] <Tpenta> :)
[05:16:59] <Gman> boyd: it's a sunblade 100, it's slow, my fingers are fast
[05:16:59] <boyd> Ok, so just 2.14 then for the moment.... unless you can just point me at a URL or a search term
[05:17:48] * boyd loaded a sunblade 100 so badly a couple of weeks ago that it was actually dropping chars from the USB kb
[05:17:57] <boyd> .. and duping some of them
[05:18:40] *** Kernel86 has joined #OpenSolaris
[05:19:10] <Gman> sure, http://www.freedesktop.org/menu-spec/latest/
[05:19:24] <boyd> Sweet.. that's for both revs?
[05:19:30] <Gman> gives you an idea on how the .menu files work in /etc/xdg/menus
[05:19:35] <Gman> nope
[05:20:13] <Gman> you'll need the base directory spec too at http://www.freedesktop.org/wiki/Standards/basedir-spec
[05:20:18] <boyd> I think it's http://standards.freedesktop.org/menu-spec/latest/
[05:20:27] <boyd> Ok...
[05:20:34] <Gman> to find the local stuff which is effectively ~/.local/....
[05:20:34] <boyd> and these are for 2.14, right?
[05:20:39] <Gman> this only works for 2.14
[05:20:43] <boyd> ok...
[05:20:48] <Gman> 2.6 is similar
[05:20:52] * boyd wonders about a GUI
[05:20:55] <Gman> but doesn't use .menus, but rather .vfolders
[05:21:06] <Gman> and the local location is ~/.gnome2/vfolders iirc
[05:21:07] <richlowe> alacarte?
[05:21:15] <Gman> the sys admin igude should be able to help in 2.6
[05:21:28] <Gman> richlowe: it's lame, only allows you to show/hide stuff i think
[05:21:31] <boyd> Ok.. thanks... I didn't think to RTFM
[05:21:41] * boyd blushes
[05:22:34] <boyd> Hmm... the alacarte screenshot looks nice... it's part of 2.16 apparently
[05:23:41] <boyd> Thanks for the pointers Gman
[05:41:28] *** nwf has joined #opensolaris
[05:54:37] * McBofh heads off to a team meeting
[05:59:50] *** hile_ has joined #opensolaris
[06:09:09] *** nwf has quit IRC
[06:09:52] *** nwf has joined #opensolaris
[06:15:14] <Method> svc:/system/auditd:default (Solaris audit daemon)
[06:15:14] <Method> State: maintenance since Sun Oct 15 21:42:28 2006
[06:15:15] <Method> Reason: Start method failed repeatedly, last exited with status 98.
[06:15:20] <Method> does that make sense to anyone?
[06:15:36] <Error_404> yeah, it means something's preventing it from starting
[06:16:25] <Error_404> check /var/svc/log/system-auditd:default.log
[06:16:30] <Error_404> it'll tell you why
[06:17:48] <Method> [ Oct 15 21:42:28 Executing start method ("/lib/svc/method/svc-auditd") ]
[06:17:48] <Method> [ Oct 15 21:42:28 Method "start" exited with status 98 ]
[06:17:51] <Method> thats all it says
[06:21:26] *** nwf has quit IRC
[06:21:36] *** nwf has joined #opensolaris
[06:27:33] *** hile_ has quit IRC
[06:34:48] <Sieghard> how do i make passwords that are larger than 8 characters?
[06:43:25] *** besonen_mobile has quit IRC
[06:44:23] <Sieghard> btw, does anyone know which zfs revision date was used for solaris 10u2
[06:44:46] <Sieghard> i'm trying to see why i get a "internal error: unexpected error 5 at line 773 of ../common/libzfs_pool.c"
[06:49:39] <libkeiser> Sieghard: if you're using unix auth, change CRYPT_DEFAULT in /etc/security/policy.conf and change all the local account passwords
[06:49:53] <richlowe> is that error from lint, or cc, or?
[06:49:55] <Sieghard> libkeiser: cool, thanks
[06:50:06] <Sieghard> richlowe: i get that error while doing a zfs import
[06:50:15] <richlowe> oh, ah.
[06:52:49] *** besonen_mobile has joined #opensolaris
[06:52:52] <Sieghard> libkeiser: which one do you recommend, 1, 2a, or md5?
[06:52:54] <Triskelios> Sieghard: the string " at (#)libzfs_pool dot c ..." is in the binary
[06:53:01] <Triskelios> which has the version
[06:53:10] <Sieghard> Trident: excellent, thanks
[06:53:25] <richlowe> the ident string would normally be removed by mcs as part of the build, and replaced with the OS version.
[06:54:22] <Triskelios> richlowe: oh
[06:54:31] <Sieghard> i guess it doesn't show when i do a strings
[06:54:47] <richlowe> Sieghard: /usr/ccs/bin/what
[06:55:22] <Sieghard> January 2005
[06:55:30] <Sieghard> can't be that old
[06:59:34] <libkeiser> Sieghard: use 2a (blowfish) or md5.
[07:04:35] * McBofh wanders off to do some work
[07:04:36] *** McBofh has quit IRC
[07:23:19] *** |tsoome| has quit IRC
[07:31:21] *** neoxed_ has joined #OpenSolaris
[07:32:21] <Error_404> i wonder if cats get colds or flu
[07:32:32] <Error_404> because my cat's been lethargic all day
[07:32:47] <steleman> Error_404: yes they can get colds
[07:32:54] <steleman> bronchitis and pneumonia too
[07:33:20] <Error_404> she hasn't been coughing... she just hasn't been her normal crazy attention-starved self
[07:33:35] <steleman> does she have a fever ?
[07:33:57] * steleman is obviously a cat person :-)
[07:34:08] <Error_404> i'll ask her, but i don't think she'll tell me
[07:34:27] <steleman> lol you can tell if she feels "warm"
[07:35:52] <Error_404> not any warmer than usual
[07:36:31] <steleman> ok thats good then maybe she's just lazy
[07:36:50] <Error_404> lazy & on a diet?
[07:36:58] <steleman> ohh she hasn't been eating either ?
[07:37:14] <Error_404> she has, but she's still got half her dish from this morning
[07:37:22] <steleman> ohh
[07:37:28] <Error_404> heh... cat on a diet
[07:37:37] <steleman> well if this lasts tomorrow i would go to the vet
[07:38:08] <Error_404> yeah, i was thinking if she's not better tomorrow i'd dig up a vet
[07:41:32] <Error_404> so far i've been lucky enough to have cats that are mostly healthy through their entire life
[07:41:58] <Error_404> (the cat in question bit down on a fish hook a year ago, but that's still not illness)
[07:42:31] <Error_404> so i don't really know
[07:44:41] * boyd just stumbled across the "Advanced User's guide" I don't think it's named well. http://docs.sun.com/app/docs/doc/806-7612
[07:47:38] <Error_404> 'differences between command-line & gui interface'
[07:47:41] <Error_404> uh huh...
[07:47:45] * boyd nods
[07:48:14] <richlowe> "The Network is The Computer", but "Using the Network" is advanced.
[07:48:17] *** neoxed has quit IRC
[07:48:31] <richlowe> though then again, so is logging in, apparently.
[07:48:33] <boyd> hehe
[07:48:52] <boyd> I find that the non-advanced users that don't log on have far fewer problems
[07:49:12] <boyd> Well, just one problem, over and over
[07:52:03] <asyd> \_o<
[07:55:12] <Error_404> boyd: not being able to log in?
[07:55:21] <boyd> yep :)
[07:56:13] *** delewis has quit IRC
[08:02:07] <simford> I'd like to convert a teamware(sccs) workspace to mercurial(hg) one. does anybody here have any scripts to simply it? thanks
[08:09:29] *** gtc has joined #opensolaris
[08:10:02] <Gman> simford: stevel does, and i've forwarded him your mail
[08:10:17] <Gman> simford: you'll have to wait until west coast us time
[08:10:46] <timeless> richlowe: maybe for people used to notepad, vi is advanced :)
[08:10:51] <simford> thanks a lot, Gman!
[08:11:00] <timeless> although personally i'd use some other word for it
[08:12:27] * timeless ponders
[08:13:04] <timeless> so does the lx brandz emulate pthreads or npt?
[08:13:34] <Error_404> i would imagine not
[08:13:58] <Error_404> the kernel it emulates is too old for that
[08:14:44] <timeless> makes sense, i wonder if someone will work on maintaining such things
[08:15:43] <Error_404> most commercial software i've seen uses linuxthreads
[08:15:50] <richlowe> maybe I'm a pessimist, but I imagine it would suffer the a similar fate to other similar things.
[08:15:55] <richlowe> s/the//
[08:15:58] <Error_404> and if it's only open source, then why do you need lx zone?
[08:16:10] <richlowe> you can never chase a moving target and come close to drawing level.
[08:16:21] <timeless> i'm trying to deal with apt
[08:16:45] <timeless> not to actually install software, but to retrieve sources from a repository for a cross reference
[08:17:16] <timeless> and i'm trying to use ZFS to host the cross reference because i'm hoping ZFSs copy behaviors will work better for me than a normal file system
[08:18:23] <Sieghard> $ whatis love
[08:18:23] <Sieghard> love: nothing appropriate
[08:18:30] <timeless> but atm, i don't actually have any real use for npt support other than to work w/ dtrace in profiling gecko-linux instead of dtrace profiling gecko-solaris
[08:19:07] <timeless> (and yes, gecko can use solaris threads or pthreads, or npt, but it's just a what if)
[08:23:56] <Error_404> you could use kprobes
[08:23:57] <Error_404> heh
[08:24:15] <Error_404> or systemtap
[08:25:34] *** delewis has joined #opensolaris
[08:25:51] <Error_404> do it man.
[08:26:11] <delewis> ugh, it seems SSGD doesn't like users using more than one JDS session.
[08:26:12] *** gtc_ has quit IRC
[08:26:26] <delewis> it seems to think there's a window manager already running on the display it's trying to use.
[08:26:46] <delewis> CDE seems to work fine, though.
[08:29:04] *** gtc has quit IRC
[08:30:58] *** klocze1 has quit IRC
[08:33:41] <Error_404> hmm... the russians want to build a floating nuclear power plant
[08:33:50] <Error_404> seems like a bad idea if you ask me...
[08:34:04] <delewis> Error_404: they've already built plenty of them :-)
[08:34:11] <delewis> most of them are rotting away in the shipyeards.
[08:34:16] <delewis> shipyards, rather.
[08:34:24] <Error_404> i'm all about nuclear power, but it seems a little too risky to throw it on a barge
[08:34:39] <delewis> they could save cash and convert their nuclear subs to powerplants.
[08:34:52] <delewis> that's the epitome of the "russian way" :-)
[08:35:34] <Error_404> repurposing war machines?
[08:36:00] <delewis> no, making as much use of existing equipment as possible. :-)
[08:36:12] <Error_404> lousy commies
[08:36:51] * delewis needs more coffee
[08:40:18] *** libkeiser has quit IRC
[08:44:10] *** Tpenta has quit IRC
[08:55:03] *** Cyrille has joined #opensolaris
[08:58:00] <asyd> hello Cyrille
[08:58:11] <Cyrille> Hi asyd, morning everyone
[09:01:48] *** Sparcioni has joined #opensolaris
[09:01:50] <delewis> "Jeez, it's lovely of you guys to pleasantly make corrections or point out errors. Stop being such fanboy bitches, admittedly every article written by somebody who has a few minutes to spare writing *isn't* going to be a technical masterpiece, but still informitive enough to be worth the read. So maybe drop your Unix Bible and your bitchcunt attitude, and go out and get laid, if at all possible." -- one of the replies to
[09:02:32] <delewis> that probably got cut-off
[09:08:35] <oxygene> "one of the replies to"..
[09:09:33] <delewis> ah
[09:10:03] <delewis> one of the replies to my reply, regarding the latest OpenSolaris-related linux.com article. The author of the article conveyed a few inaccuracies. :-)
[09:10:20] <delewis> http://distrocenter.linux.com/article.pl?sid=06/10/06/150252&tid=127
[09:10:22] <delewis> this little gem.
[09:11:06] <oxygene> yay, "I'm talking shit, but at least I'm talking".. nice attitude ;)
[09:11:52] <delewis> http://distrocenter.linux.com/comments.pl?sid=37432&cid=91933
[09:12:01] <delewis> another series of my replies (related to the same article)
[09:12:15] <delewis> it's pretty obvious which replies are mine :-)
[09:14:41] <delewis> amazing. my Firefox session hasn't crashed in days, yet.
[09:14:50] <timeless> dbx: internal error: signal SIGSEGV (no mapping at the fault address)
[09:14:55] <timeless> "oops"
[09:15:15] <delewis> timeless: probably requires a patch
[09:15:16] <Triskelios> delewis: the real question is, how large is it now?
[09:15:31] <delewis> Triskelios: too large :-)
[09:15:35] <timeless> delewis: eh?
[09:16:45] * timeless kinda ran out of disk space again
[09:16:51] <delewis> timeless: awhile back, a particular Solaris Express release broke dbx (IIRC, it was a SIGSEGV) -- a patch remedied the issue.
[09:17:14] <timeless> ok, cute
[09:17:17] * timeless sighsd
[09:17:27] <andersmo> sighsegv?
[09:17:30] <trygvis> http://www.miguelcarrasco.net/miguelcarrasco/2006/10/linux_crash_top.html
[09:17:33] <timeless> ok, so, i have a snv_41 snapshot
[09:17:50] <timeless> andersmo: i sigh about twice as often as my apps sigsegv
[09:18:50] <timeless> so, if i have a 2gb core dump for an app and i'm out of disk space, should i kill the core dump or try to get a new dbx and see if it says anything useful? :(
[09:20:08] <Sparcioni> Could please someone who has access to SWAN look and see if in b51
[09:20:09] <Sparcioni> which is to be released tomorrow, there is now the packet filtering hooks?
[09:20:09] <delewis> apply the latest dbx patches
[09:20:10] <richlowe> delewis: it's worth noting that the dbx in the ss11 tarball still suffers from that.
[09:20:14] <Sparcioni> The check is simple and immediate: just look if there is
[09:20:14] <delewis> then see if it SIGSEGVs
[09:20:14] <Sparcioni> file named neti.h under /usr/src/uts/common/sys ?
[09:20:17] <Sparcioni> I am just waiting for it and igor to know...
[09:20:25] <delewis> richlowe: yes, without the patch(es)
[09:20:28] <richlowe> for what it's worth, this is the second time something has broken it too, first one was a hang in libproc.
[09:20:37] <delewis> yes, I remember that nasty one.
[09:20:43] <delewis> irritated the hell out of me.
[09:22:38] <timeless> richlowe: wait, so the only available dbx won't work out of the box w/ snv_49? i have to patch it?
[09:22:53] <delewis> timeless: correct
[09:23:08] <delewis> any version greater than what snv release originally caused the dbx bugs needs to be patched
[09:23:37] <timeless> great
[09:24:04] * timeless sighs
[09:24:10] <timeless> gnome system monitor is a bit silly
[09:24:19] <timeless> or maybe it's the df output it uses
[09:24:55] <richlowe> delewis: I guess 5033325 broke it?
[09:24:57] <timeless> do people here use zfs w/ lots of nested volumes?
[09:24:58] <delewis> dtperfmeter was at least somewhat useful :-)
[09:25:17] <delewis> timeless: I sort of do
[09:25:29] * timeless doesn't have dtperfmeter
[09:25:35] <delewis> I usually create a separate volume for each zone (with my export/zones volume), and each of my NFS exports
[09:25:48] <timeless> delewis: try running gnome system monitor and then select the devices tab and sort by 'used'
[09:25:58] <delewis> sdtperfmeter
[09:26:08] <delewis> I forget it's a Sun authored tool
[09:26:13] <richlowe> hm, though that seems pretty unlikely.
[09:26:31] <timeless> don't seem to have that either according to some script that greps installed pacakges
[09:26:52] *** simford has quit IRC
[09:26:59] <delewis> richlowe: no idea
[09:27:11] <delewis> timeless: /usr/dt/bin/sdtperfmeter
[09:27:40] <delewis> but of course it requires rstatd (which is disabled per SBD)
[09:28:02] * delewis imagines lots of unhappy CDE users until they found out about the magical script to disable SBD
[09:28:16] <richlowe> delewis, timeless: http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6477975
[09:28:17] <timeless> delewis: for me, i have: svn [26.5 GiB (74%)] root [9.3 GiB (51%)] timeless [7.1 GiB (44%)] c1d0s0 [2.9 GiB (74%)] libc.so.1 [2.9 GiB (74%)]
[09:28:19] <richlowe> that probably.
[09:28:22] <timeless> zsh: no such file or directory: /usr/dt/bin/sdtperfmeter
[09:28:31] <delewis> weird.
[09:28:44] <delewis> you must not have the CDE consolidation installed
[09:28:47] <timeless> i have a 4gib root partition
[09:28:51] <timeless> it doesn't really have space for much
[09:29:15] <delewis> ugh.
[09:29:15] <timeless> richlowe: yes, that's the right crash
[09:29:21] * delewis never knew about bugs.sun.com
[09:29:35] <delewis> I was looking on bugs.opensolaris.org (where the bug information was abit sparse)
[09:29:38] <timeless> heh
[09:30:30] <delewis> hmm, I haven't experienced that crash, yet
[09:30:38] <delewis> but I installed the latest dbx patch
[09:30:58] <richlowe> delewis: I die in libcpc instead.
[09:31:07] <timeless> richlowe: nice bug
[09:31:36] <richlowe> well, dbx does, not me ;)
[09:31:42] <delewis> hehe
[09:32:04] <timeless> programmers really are fatalists
[09:32:08] * delewis goes to get a coffee re-fill.
[09:32:14] <timeless> they talk about death all the time :(
[09:34:28] *** simford has joined #opensolaris
[09:34:34] <Sparcioni> I am repating my question:
[09:34:35] <Sparcioni> Please, can somebody answer: I am just waiting for it and igor to know...
[09:34:35] <Sparcioni> Could please someone who has access to SWAN look and see if in b51
[09:34:35] <Sparcioni> which is to be released tomorrow, there is now the packet filtering hooks?
[09:34:39] <Sparcioni> he check is simple and immediate: just look if there is
[09:34:39] <Sparcioni> file named neti.h under /usr/src/uts/common/sys ?
[09:35:00] <Sparcioni> can't wait anymore...
[09:35:15] <delewis> Sparcioni: I doubt anyone that has access to SWAN can say without being attacked lawyers :-)
[09:35:26] <timeless> heh
[09:35:29] <richlowe> they could.
[09:35:36] <richlowe> but the answer will be almost certainly "No"
[09:35:41] <richlowe> it would show up on the heads-up page if it had been putback.
[09:35:55] <Sparcioni> richlowe, thnkls
[09:36:05] <Sparcioni> I heard about that page
[09:36:16] <Sparcioni> but forgot the URL
[09:36:27] <richlowe> opensolaris.org/os/community/on/flag-days/all
[09:36:52] <Sparcioni> richlowe, thnks again
[09:37:05] <timeless> i read that page on friday, i was really impressed by the organization/process that keeps such a page so well maintained
[09:38:07] <delewis> interesting
[09:38:16] <delewis> not all of those putbacks seem to be ARC cases
[09:38:24] * delewis wonders what kind of putbacks require ARC cases
[09:38:33] <delewis> other than ksh93 :-)
[09:38:39] <Sparcioni> timeless, ant idea what is the difference between lines starting with head up: and other lines
[09:39:00] <Sparcioni> for example the PSARC underlined lines ?
[09:39:01] <timeless> sparcioni: i think some are just informational, the page actually explains
[09:39:19] <timeless> i'm not an expert, i just read it, and i'm just a user who breaks things, not an opensolaris dev
[09:39:21] <delewis> oh, finally
[09:39:26] <delewis> ISO9660v2 support
[09:39:45] <delewis> hopefully, that implies > 64 character filename support
[09:40:02] * delewis was going to get around to fixing that at some point -- afterall, it's just a single #define, I think
[09:40:05] <richlowe> delewis: anything that changes, adds, or removes an interface, pretty much.
[09:40:15] <delewis> richlowe: ah
[09:40:40] *** halton has left #opensolaris
[09:40:44] <Sparcioni> well, the packet filtering hooks in indeed not there ...
[09:40:59] <delewis> any particular reason why all the PSARC case links aren't valid?
[09:41:13] <delewis> well, some of them are
[09:41:16] <delewis> the majority aren't
[09:41:49] *** besonen_mobile_ has joined #opensolaris
[09:41:54] *** Tpenta has joined #opensolaris
[09:42:41] <timeless> wasn't that explained in the page?
[09:43:06] * delewis feels silly
[09:43:12] *** ChanServ sets mode: +o Tpenta
[09:43:19] <timeless> very top
[09:43:21] <delewis> a 1024x768 Firefox window just requires too much scrolling
[09:43:36] <delewis> thanks :-)
[09:43:41] <timeless> be happy, i'm working on a device w/ 800x480 screen
[09:43:45] <delewis> ouch.
[09:43:51] <delewis> and I thought I had a low-resolution fetish.
[09:43:54] <timeless> i also have somewhat of a parrot like memory
[09:43:59] *** besonen_mobile has quit IRC
[09:44:09] <timeless> so thankfully useless details lik a note about ARC links being bad stick with me :)
[09:44:29] <timeless> actually, it works really well for news.google.com/mail.google.com
[09:44:42] * timeless isn't sure how it'd do on this page
[09:44:49] <timeless> probably ok, but i wouldn't have any reason to read it
[09:45:11] <timeless> note: that's not my *only* device, just the one i'm developing
[09:45:39] <timeless> but i can ssh to the computer which hosts this irc session, so i could be irc'ing from it :)
[09:57:32] *** Dar_HOME is now known as Dar
[09:57:33] *** jcea has quit IRC
[10:00:52] <Vanuatoo> By accident I created directory called "~" in my home dir
[10:00:55] <Vanuatoo> How do I delete it
[10:01:12] <Tpenta> rmdir ./~
[10:01:14] <asyd> rm -- \~ ?
[10:02:19] <Vanuatoo> rmdir \~ made the trick :)
[10:02:37] <Vanuatoo> I was afraid not to empty my home dir :)
[10:03:09] <timeless> zfs snapshot it first ;-)
[10:04:00] <Vanuatoo> I don't have zfs :(
[10:05:43] <delewis> single-quotes would've worked, too, I think.
[10:06:20] <Vanuatoo> delewis: I knew about escape symbol, just wanted to make sure ;)
[10:06:28] <nightswim> and also
[10:06:35] <nightswim> without -r, rm ~ shouldnt do anything
[10:06:46] <nightswim> or rmdir
[10:07:09] <Vanuatoo> I first did ls \~ to make sure this dir is empty
[10:07:16] <Vanuatoo> and after that rm \~
[10:07:23] <Vanuatoo> no, rmdir
[10:08:02] <Vanuatoo> I think it should be denied to create a dir with the name like this
[10:08:16] <Vanuatoo> to avoid such situations
[10:08:45] <Vanuatoo> It's the strength and weakness at the same time
[10:10:05] *** Gman has quit IRC
[10:10:06] *** alanc_away has quit IRC
[10:11:02] *** alanc has joined #opensolaris
[10:27:25] *** peteh has joined #opensolaris
[10:31:08] *** alanc is now known as alanc_away
[10:44:12] *** Cyrille has quit IRC
[10:46:03] *** peteh has quit IRC
[10:46:12] *** Dar has quit IRC
[10:56:41] *** Dar has joined #opensolaris
[10:57:24] *** Cyrille has joined #opensolaris
[11:01:35] *** deather_ is now known as deather
[11:01:46] *** halton has joined #opensolaris
[11:02:08] <timeless> hrm, i'm reading http://blogs.sun.com/nilsn/entry/installing_a_debian_zone_with
[11:02:16] *** peteh has joined #opensolaris
[11:02:17] <timeless> zonecfg:debian> create -B lx
[11:02:26] * timeless sighs, lunch first
[11:02:37] <asyd> you don't have the good SE build ? ;p
[11:04:23] *** sylvain has joined #opensolaris
[11:05:33] *** yongsun has quit IRC
[11:06:35] *** Gman has joined #opensolaris
[11:06:57] <Sparcioni> I want to make sure that what I assume is true:
[11:06:57] <Sparcioni> in order tha rmformat util will detect USB (and other) devices
[11:06:57] <Sparcioni> which are pluuged to the machine, the volfs service should be disabled.
[11:06:57] <Sparcioni> Is it so?
[11:07:05] <Gman> so I'm wondering if level of 'Core Contributor' is too high for voting rights..
[11:12:34] <eugene> timeless: crunchyfrog. haha.
[11:14:10] *** jcea has joined #opensolaris
[11:15:22] <quasi> Gman: sounds rather restrictive
[11:15:57] *** halton has left #opensolaris
[11:18:17] *** dojtoll has joined #OpenSolaris
[11:19:29] *** LeftWing has quit IRC
[11:19:34] *** LeftWing has joined #OpenSolaris
[11:31:46] <dojtoll> Dose anyone know what "Screened for processor flatness" means? It's from the Full component list for the SB1000/2000.
[11:32:30] *** bengtf has joined #opensolaris
[11:32:52] *** logic__ has quit IRC
[11:34:33] *** r3boot has quit IRC
[11:35:22] *** r3boot has joined #opensolaris
[11:43:11] *** dojtoll has left #opensolaris
[11:47:26] <timeless> eugene?
[11:47:44] <timeless> not my computer
[11:47:50] <timeless> mine is swift
[11:48:51] <Vanuatoo> I'm trying to compile xmms on solaris express 9/06 x86. Configure step goes fine, but after a
[11:48:58] <Vanuatoo> I issue command gmake
[11:49:03] <Vanuatoo> it gives me an error
[11:49:58] <andersmo> Vanuatoo: Then you need to read the error message, understand what it says, understand what causes it, and remedy the problem. =)
[11:50:13] <quasi> andersmo++ ;)
[11:50:20] <andersmo> Step 2 and 3 are probably the tricky ones, but with the highest payoff later. =)
[11:50:31] *** MattMan has joined #opensolaris
[11:50:36] *** logic has joined #opensolaris
[11:50:48] <Vanuatoo> andersmo: I'm now posting error
[11:51:27] <Vanuatoo> pastebin.com sucks, any other site?
[11:52:04] <trygvis> rafb.net/paste
[11:52:18] <quasi> paste.getlinuxhelp.org
[11:53:03] <Vanuatoo> http://pastebin.ca/204868
[11:54:18] <timeless> paste.lisp.org
[11:55:12] <Vanuatoo> any ideas?
[11:55:17] <quasi> fullscreen.c:73:38: X11/extensions/xf86vmode.h: No such file or directory
[11:55:53] <Cyrille> and the rest comes from that missing header
[11:56:16] <Cyrille> presumably coming from XFree86 X extensions (just a guess)
[11:57:09] <Vanuatoo> strange
[11:57:22] <Vanuatoo> I have that file
[11:57:25] <Vanuatoo> in here /usr/X11/include/X11/extensions/xf86vmode.h
[11:57:56] <Vanuatoo> I can add this include path to configure, can't  I?
[11:58:36] <Cyrille> what would be important would be to have /usr/X11/include in the include path of the compilation step which is failing.
[11:58:38] <timeless> ok, i'm guessing i have to add lx support to brandz before i can make an lx brandz?
[11:58:59] <Vanuatoo> Cyrille: How do I achieve that?
[11:59:32] <Cyrille> not sure exactly, might depend on the configure.
[11:59:51] <Vanuatoo> it has a parameter --x-includes
[11:59:52] <Cyrille> configure --help might give you options you could use.
[12:00:44] <Vanuatoo> I did ./configure --x-includes=/usr/X11/include
[12:00:49] <Vanuatoo> but still the same error
[12:01:46] <Cyrille> that could be it
[12:01:46] <quasi> try pointing it at /usr/X11/include
[12:01:47] <timeless> hrm
[12:01:48] <timeless> is /zones a real file system thing?
[12:02:28] <Cyrille> Vanuatoo: do you see the include in the gcc line?
[12:03:03] *** calumb has joined #opensolaris
[12:04:46] <Vanuatoo> Cyrille: I don't understand what you mean by that
[12:06:12] <Cyrille> Vanuatoo: presumably you ran configure with the new flag and then launched make again, then got to the part which you copy/pasted earlier and got the same error. The question is whether in the first line "if gcc ..." you see something like "-I/usr/X11/include" which is what's missing for your file to be found.
[12:07:06] <Vanuatoo> No I don't
[12:07:33] <Vanuatoo> if gcc -DHAVE_CONFIG_H -I. -I. -I.. -I. -I/usr/sfw/include/gtk-1.2 -I/usr/sfw/include/glib-1.2 -I/usr/sfw/lib/glib/include -D_REENTRANT -g -O2 -Wall -Wpointer-arith -finline-functions -ffast-math -funroll-all-loops -DDATA_DIR=\"/usr/local/share/xmms\" -DPLUGIN_DIR=\"/usr/local/lib/xmms\" -DPLUGINSUBS=\"Output\",\"Input\",\"Effect\",\"General\",\"Visualization\" -DLOCALEDIR=\"/usr/local/share/locale\" -DI386_ASSEM -I../intl -I..  -
[12:07:34] <Vanuatoo>   -c -o fullscreen.o `test -f 'fullscreen.c' || echo './'`fullscreen.c; \
[12:07:53] *** kimc has joined #opensolaris
[12:08:39] <Cyrille> then this --x-includes doesn't seem to have the expected effect.
[12:10:37] <quasi> you could try: export CFLAGS='-I/usr/X11/include'
[12:10:38] <Vanuatoo> And where it takes other paths from?
[12:10:41] <Vanuatoo> ok
[12:12:24] <quasi> also, it may be worth to do a make distclean before running configure again
[12:12:32] <Vanuatoo> This is the one of the lines of ./configure
[12:12:33] <Vanuatoo> checking for XF86VidModeQueryExtension in -lXxf86vm... yes
[12:13:02] <Vanuatoo> OK It passed that step
[12:13:19] <Vanuatoo> bad script I guess
[12:13:29] <Vanuatoo> None is assuming it will be run on solaris :(
[12:13:52] <Vanuatoo> linux-oriented programming :)
[12:14:47] *** coffman has joined #opensolaris
[12:17:23] <timeless> http://pastebin.ca/204883 - why didn't something create /sbin ?
[12:17:46] <Vanuatoo> I've got xmms running fine, Thank you everyone :)
[12:19:44] <coffman> hm
[12:20:30] <coffman> if i disable the onboard sata controller (vt8237) the system (10u2) does not boot anymore
[12:20:37] <quasi> Vanuatoo: that did it?
[12:21:27] <Vanuatoo> yes
[12:21:36] *** ShadowHntr has joined #opensolaris
[12:21:37] *** digit has joined #opensolaris
[12:21:48] <Vanuatoo> it seems that --x-includes flag is ignored during compile
[12:22:16] <Vanuatoo> and I guess it's solaris specific, otherwise it would be fixed on linux
[12:23:58] *** digit has quit IRC
[12:24:24] <Cyrille> well normally the configure step should use the various flags to probe for and find the files it expects the compilation to need, and accordingly generate the makefiles.
[12:25:54] *** lloy0076 has joined #opensolaris
[12:26:03] *** lloy0076 has left #opensolaris
[12:26:27] <quasi> autoconf is "magic" enough that it doesn't surprise me if it doesn't work
[12:27:07] <asyd> :)
[12:30:24] <lplatypus_> anyone had any good experiences with alternatives to autoconf?
[12:30:52] <oxygene> autoconf can be good - it's the autoconf users (those writing m4 macros) that tend to mess things up
[12:30:59] <oxygene> (of course, autoconf could be more robust against that)
[12:32:04] <quasi> oxygene: I don't quite agree - autoconf gets to be a bit of a nightmare when you decide to add another platform later (and let's not talk about speed, or rather lack thereof ;)
[12:32:09] *** kimc has left #opensolaris
[12:32:53] <oxygene> quasi: "add a platform"? that's the point of it: no need to add "platforms"
[12:33:25] <quasi> oxygene: except it doesn't always work like that
[12:34:03] <oxygene> it often does, given a properly written set of tests
[12:35:25] <lplatypus_> does the survival of tools like autoconf indicate the failure of standardization efforts such as c99, the single unix specification?
[12:36:42] <quasi> or just the laziness of certain os implementers (and a certain compiler ;)
[12:37:06] <oxygene> no, even with strict standard implementation availability most apps would still have to go beyond those standards with other libraries
[12:37:37] <oxygene> (let's start with X and its extensions, paths and library setups)
[12:41:16] *** lplatypus_ has quit IRC
[12:42:30] *** ShadowHntr has quit IRC
[12:42:35] *** lplatypus has joined #opensolaris
[12:44:46] * timeless tries to figure out how to get this zone stuff to work :(
[12:44:58] *** Sparcioni2 has joined #opensolaris
[12:45:06] <lplatypus> yeah fair enough... configure scripts seem to do many more checks than should be necessary though
[12:45:44] <quasi> timeless: zones in general or brandz zones?
[12:45:47] *** Sparcioni2 has quit IRC
[12:45:49] <oxygene> yes, lots of that could be killed (my favorite: how about it just tries a 200kchars long command line before iteratively working up to the maximum?)
[12:45:52] <timeless> grandz lx
[12:45:57] <timeless> err brandz lx
[12:46:14] <timeless> specifically trying to get a debian lx brandz zone happy
[12:46:44] <quasi> timeless: I've seen plenty of howtos on the brandz list
[12:46:59] <timeless> i tried following http://blogs.sun.com/nilsn/entry/installing_a_debian_zone_with but it didn't work
[12:55:19] <timeless> quasi: i don't get it
[12:55:30] <timeless> every single thread says zonecfg -z blah ... creat e-B lx
[12:55:38] <timeless> but my zonecfg says:
[12:55:41] <timeless> zonecfg:debian> create -B lx
[12:55:42] <timeless> usage:
[12:55:42] <timeless> create [-F] [ -a <path> | -b | -t <template> ]
[12:56:54] <oxygene> seems like your zones version doesn't support brands
[12:56:58] <oxygene> (for whatever reason)
[12:57:30] <timeless> SunOS swift 5.11 snv_49 i86pc i386 i86pc
[12:58:09] *** Dar has quit IRC
[12:59:43] <timeless> according to my ignorant reading of brandz and sce docs, snv_49 is supposed to include it
[13:01:19] *** sparc-kly has joined #opensolaris
[13:01:19] *** ChanServ sets mode: +o sparc-kly
[13:05:36] *** logic has quit IRC
[13:05:47] *** mega has joined #opensolaris
[13:12:33] *** Sparcioni has quit IRC
[13:12:56] <timeless> oxygene: do i need somespecial package installed? the only one listed in the brandz page is SUNWpkgcmdsu and i have that
[13:13:32] <oxygene> no idea, sorry
[13:15:35] *** medar has joined #opensolaris
[13:15:35] *** coffman has quit IRC
[13:16:08] *** coffman has joined #opensolaris
[13:16:37] <ofu>     1.7  369.1   13.3 6033.5 67.5 14.5  182.0   39.0  80 100 c1t2d0, hmmm
[13:16:48] <timeless> # pkginfo|grep SUNWlxu
[13:16:49] <timeless> system      SUNWlxu                         lx Brand (Usr)
[13:16:59] <ofu> i suppose, i know why the system is so slow
[13:17:32] <quasi> ofu: slightly busy disk ;)
[13:19:42] <quasi>             MBytes Installed:  2078.24
[13:19:42] <quasi>             MBytes Remaining:  1640.63
[13:19:54] *** Tpent1 has joined #opensolaris
[13:19:55] <quasi> this t2000 install is taking forever
[13:20:41] <DataStream>  Wierd, my X4200 takes less len 45 mins, full install on sol q0
[13:20:44] <DataStream> 10
[13:21:09] <ofu> installing is single-threaded
[13:21:46] <Berny> hmm, why doesn't sxcr 49's installer like the touchpad of my ferrari?
[13:23:06] *** logic has joined #opensolaris
[13:23:45] * quasi concludes that installs haven't gotten faster for 11/06
[13:26:57] <Berny> .oO(i'll tell you once i've completed the update...)
[13:27:53] <timeless> can someone tell me what package owns or provides /usr/lib/brand/lx/install/
[13:28:23] <quasi> timeless: grep /usr/lib/brand/lx/install /var/sadm/install/contents
[13:28:36] <timeless> quasi: i *don't* have that
[13:28:41] <Berny> 8-)
[13:28:42] <timeless> there's no way i can find it on my system
[13:29:21] <Berny> timeless: i don't have it either on my on 49 box
[13:29:32] <sickness> morning all
[13:30:08] <quasi> afternoon sickness
[13:33:59] *** vmhobbes has joined #OpenSolaris
[13:34:30] <timeless> berny: i don't suppose you've used zones (esp brandz / lx)
[13:36:22] <timeless> it looks like the /install/ directory was collapsed into its parent
[13:37:43] <Vanuatoo> Is there any chance that we'll have some voice enabled messenger?
[13:37:55] <Vanuatoo> skype, google talk?
[13:39:33] <Berny> timeless: i use zones though not branded (yet)
[13:40:29] <IvanR_> timeless: create -t SUNWlx
[13:41:37] <timeless> ivanr: that's what i've been using
[13:42:53] *** vmhobbes has quit IRC
[13:43:03] <cormac__> here, if i want a virtual interface raised on boot, do i just need to make the /etc/hosts entry and the /etc/hostname.ce1:1 to list the ip
[13:43:06] *** cormac__ is now known as cormac
[13:44:20] <timeless> ivanr: http://pastebin.ca/204936
[13:44:57] <timeless> i'm trying to follow the steps at http://blogs.sun.com/nilsn/entry/installing_a_debian_zone_with
[13:46:38] <quasi> timeless: careful with that howto - it is a bit old
[13:46:54] <timeless> quasi: yeah, but um,...
[13:47:15] <cormac> if /etc/hostname.ce1:1 lists an ip, does the rc network script not read the hosts file ata ll?
[13:48:20] <IvanR_> cormac: IP should work, too. Are you seeing a problem?
[13:51:11] <IvanR_> timeless: What are you trying to do with the dummy.tar file?
[13:51:15] *** mega has quit IRC
[13:51:30] *** mega has joined #opensolaris
[13:51:41] <timeless> ivanr: follow the instructions from the blog which quasi says is old
[13:52:42] *** Kernel86|Laptop has quit IRC
[13:52:52] *** Kernel86|Laptop has joined #OpenSolaris
[13:55:21] <quasi> hmm, more than an hour just to copy the files
[13:56:08] *** apa has joined #opensolaris
[13:56:33] <IvanR_> I think the scripts behind the scenes of the zones/brandz creation take care of some of those steps now, but need disto-version information to work right.
[13:57:08] <IvanR_> And I don't have any around right now for trying an lx brand, either.
[13:58:03] *** apa has quit IRC
[13:59:11] *** calumb is now known as calLNCH
[14:01:01] *** svoboda has quit IRC
[14:14:27] *** aliquis has quit IRC
[14:19:41] *** laca has joined #opensolaris
[14:32:18] <ofu> does anybody know the release-date of honeycomb?
[14:32:55] <ofu> the docs tell me that i think i want to use it pretty soon
[14:37:20] <quasi> ofu: I don't think there's an official date - at least not one that's public
[14:38:32] <quasi> ofu: I'd probably recommend getting in touch with your sun rep and lining up as a "test customer" ;)
[14:41:16] <ofu> my account manager will visit me on friday
[14:41:33] <ofu> and today bluearc made a visit
[14:42:08] <ofu> we are still searching for something similar to netapp, but cheaper ;-)
[14:42:35] <ofu> quasi: so honeycomb is zfs-backed? And can cope with _many_ files?
[14:48:57] *** Dar has joined #opensolaris
[14:49:11] *** calLNCH has quit IRC
[14:52:58] *** OnkelSchorsch_ has joined #opensolaris
[14:55:28] <quasi> ofu: from what I've read, it is an "object storage" - wether it is zfs or not, I'm not 110% sure
[14:55:58] <quasi> ofu: actually, sun has a product similar to netapp, but supposedly at half the price
[14:56:30] <quasi> ofu: doing the usual, nfs and cifs
[14:57:13] <quasi> ofu: http://www.sun.com/storagetek/nas/5320/
[14:57:20] <quasi> (I think that's the one)
[15:00:03] <ofu> non-clustered and just one cpu
[15:00:24] *** laca is now known as lacaAFK
[15:01:10] <ofu> if our netapp-clusters fail, about 1/4 of germanys internet is offline... this should never happen
[15:01:43] <quasi> but unlike netapp they put a raid controller in each drive shelf - making potentially better performance
[15:02:10] <quasi> or at least making it less necessary to have more cpus
[15:02:23] <ofu> i thought, dedicated asics are dead, software-raid (as in zfs) is the future
[15:03:29] <quasi> not everyone thinks like that
[15:03:55] <ofu> but jonathan does
[15:04:09] <ofu> isnt that enough?
[15:04:10] *** OnkelSchorsch has quit IRC
[15:04:39] <quasi> well, I would imagine future versions of the 5320 being much smarter
[15:04:43] * andersmo would've loved to see a "NAS appliance" install for thumpers - pop in a dvd, install, get a ready-made Thumper-tuned Solaris 10 install with Samba, and integrated authentication/configuration for NFSv3/4 and SMB.
[15:05:23] <quasi> andersmo: that's pretty much what jumpstart is for ;)
[15:05:30] <ofu> why didnt they put additional bootdisks or compact flash into the thumper?
[15:06:02] <ofu> can you pxe-boot a live solaris system?
[15:06:22] <quasi> booting and live makes no sense
[15:06:42] <quasi> ofu: http://www.sun.com/storagetek/nas/5320cluster/ if you want clustering ;)
[15:07:05] <andersmo> quasi: I know I could make it myself with jumpstart - the value would be in having a ready-to-go massive but relatively cheap NAS without spending a lot of work (work costs money. =)
[15:07:29] <ofu> quasi: aah, fine.
[15:07:49] <quasi> andersmo: I wonder if they don't deliver it preinstalled?
[15:08:01] <ofu> but its just a galaxy, thats way too slow
[15:08:25] <quasi> andersmo: prior to 11/06 there's a few bits missing in zfs for it to be real fun with that many disks
[15:08:31] <quasi> ofu: not quite
[15:09:30] <coffman> yeah like raidz2
[15:09:42] <ofu> and auto-selecting disks
[15:10:23] <ofu> i like to say: take 5 disks and make a pool, not c*t*d0s*
[15:10:36] <coffman> hehe
[15:11:08] *** mrdeviant has joined #opensolaris
[15:12:03] *** mega has quit IRC
[15:12:14] <quasi> ofu: I'm hoping to see what difference it makes with the different setups - filebench and I have a plan with a thumper that should be arriving shortly
[15:19:15] *** calumb has joined #opensolaris
[15:25:11] <timeless> ofu: a bit of perl would work, or cut, right?
[15:25:20] <timeless> echo c*t*d0s*|cut ...
[15:26:19] <ofu> but thats too complicated, why cant zpool pick a disk?
[15:26:48] <timeless> oh, i'm not saying it shouldn't, but relatively speaking, it could be worse
[15:27:11] <ofu> if zpool picks disks based on performance and redundancy, it should do it
[15:33:01] <andersmo> What if zpool picks a disk you happen to use for something else, say, in a different os? Making a safe and sane zpool disk-grabber sounds like an intractable problem. =)
[15:33:16] <andersmo> (that is, safe, sane and still convenient. =)
[15:33:37] <timeless> andersmo: well, one thing would be to offer a gui (html is fine)
[15:33:38] *** svoboda has joined #opensolaris
[15:33:48] <andersmo> uuh. gooey. :=)
[15:33:51] <quasi> timeless: there already is one
[15:34:08] <timeless> but doesn't zpool usually only consider available (i.e. unused/unpartitioned) volumes?
[15:34:29] <timeless> quasi: hrm, i havne't met it, i probably ran out of disk space:)
[15:35:01] *** coffman has quit IRC
[15:35:05] <quasi> timeless: part of the admin web thingy
[15:35:39] *** coffman has joined #opensolaris
[15:36:52] * timeless wonders how to properly setup a zone to talk to the global zone using a loopback device
[15:36:58] * timeless will have to google for that later
[15:38:06] *** glagasse has joined #opensolaris
[15:38:40] <quasi> timeless: one way could be simply adding a range of RFC1918 ip to both the global zone and the local zone
[15:40:15] *** nwf has quit IRC
[15:42:04] <timeless> # zonecfg -z global
[15:42:06] <timeless> global: Bogus zone name
[15:42:09] <timeless> mighty friendly :)
[15:43:02] *** logic has quit IRC
[15:43:03] *** logic_ has joined #opensolaris
[15:43:36] <quasi> true though
[15:43:50] <timeless> but i want to configure the global zone :)
[15:45:47] * timeless sighs
[15:46:15] *** karrotx has joined #opensolaris
[15:46:36] <quasi> you do that with the usual tools, not zoneadm (since it isn't really a zone at all)
[15:47:44] *** hile_ has joined #opensolaris
[15:51:10] *** unixconsole has joined #opensolaris
[15:52:34] *** calumb has quit IRC
[15:52:35] *** apa has joined #opensolaris
[15:54:10] <timeless> yeah, but zoneadm's interface is easier :(
[15:54:34] <AbeFroman> ha
[15:57:15] * timeless tries to figure out how to change or remove a net portion of a zone
[15:57:57] <asyd> zonecfg ; .. remove net address <address> something like that
[15:59:47] <timeless> wow, i'm root on centos (zoned)
[16:03:53] *** sparc-kly_ has joined #opensolaris
[16:10:44] <timeless> i'm used to using zfs create foopy/bar/whack to create a mounted zfs item in foopy/bar named whack
[16:10:53] <timeless> this works fine if i already have a zfs partition for bar
[16:11:07] *** nyati has quit IRC
[16:11:26] <timeless> can i not do this if foopy is a mounted zfs partition and bar is a regular directory?
[16:12:03] <oxygene> no, must be a zfs partition.. you can, of course, create zfs partitions somewhere else in the namespace and shuffle them around with zfs set mountpoint=
[16:12:23] <quasi> timeless: the web gui thingy is on https://hostname:6789 if you haven't disabled it
[16:12:54] <timeless> i'm pretty sure it wasn't installed
[16:13:03] <timeless> host doesn't respond :)
[16:13:20] <quasi> if you do a full install, it is there
[16:14:06] <Cyrille> you may need to start it though (smcwebserver start)
[16:15:32] <timeless> # smcwebserver start
[16:15:33] <timeless> Starting Sun Java(TM) Web Console Version 3.0.1 ...
[16:16:11] <Cyrille> sounds about right.
[16:16:32] <timeless> wow, pretty
[16:17:56] <timeless> can i browse the actual files in the zfs system using this ui? :(
[16:19:18] <Cyrille> I have absolutely no idea, I just know how to start the web console :-)
[16:19:50] *** cypromis has quit IRC
[16:20:25] *** sparc-kly has quit IRC
[16:20:50] *** cypromis has joined #opensolaris
[16:21:07] *** hile_ has quit IRC
[16:23:24] <timeless> http://pastebin.ca/205072 - ok, now i've reached the interesting error state :)
[16:27:55] <gdamore> hi *
[16:28:34] *** lacaAFK is now known as laca
[16:34:58] *** hile_ has joined #opensolaris
[16:39:28] *** Sparcioni has joined #opensolaris
[16:39:54] <Sparcioni> is there a method for getting timeofday from a kernel module?
[16:42:26] <timeless> lx_brand.so.1: not found.ERROR: No such file or directory
[16:43:51] *** nightswim has quit IRC
[16:43:54] *** nightswim has joined #opensolaris
[16:53:41] <postwait> I'm playing zfs send and seeing awful performance.
[16:54:02] <postwait> I get only about 350Kb/second
[16:54:29] <asyd> you use a full disk ?
[16:54:35] <postwait> No
[16:55:06] <postwait> It's about 8% full or so
[16:55:53] *** Sparcioni has quit IRC
[16:58:06] *** sylvain has quit IRC
[17:02:04] <timeless> is sunsolve unrelated to the other main sun account logins?
[17:02:13] <timeless> i have an sdc account :(
[17:04:29] <quasi> postwait: what kind of disks under it?
[17:04:44] <mrdeviant> timeless, they should all use the same userid/password
[17:08:05] <delewis> greetings
[17:08:10] <timeless> thanks
[17:08:13] <hile_> morning delewis
[17:08:17] <delewis> morning, hile_
[17:08:22] <timeless> now i just need to find the latest version of sunstudio
[17:09:01] <delewis> timeless: http://developers.sun.com :-)
[17:09:47] <jengelh> http://developers.developers.developers.developers.microsoft.com/
[17:10:24] <timeless> delewis: when i do it wrong it asks me for anotehr login :)
[17:10:34] <delewis> jengelh: :-)
[17:10:39] <quasi> timeless: sun.com/studio
[17:11:27] <timeless> opensolaris.org has a reeasonable link that works
[17:15:38] *** sommerfeld has joined #opensolaris
[17:15:38] *** ChanServ sets mode: +o sommerfeld
[17:15:57] *** y___schil has joined #opensolaris
[17:26:07] *** ___schily has quit IRC
[17:33:02] <delewis> the link on the OpenSolaris page is to a tarball of Sun Studio
[17:33:05] <delewis> IIRC
[17:33:18] <delewis> not SVR4 packages, which you may want, instead.
[17:36:09] <postwait> quasi: Xserve RAID
[17:37:59] <quasi> postwait: that should be giving you a bit more
[17:38:13] *** stevel has joined #opensolaris
[17:38:14] *** ChanServ sets mode: +o stevel
[17:38:21] <postwait> the rest of the performance on those disks peaks (and sustains) much higher than that
[17:39:55] <timeless> zoneadm: zone 'centos': lo0:2: could not bring interface up: address in use by zone 'global': Cannot assign requested address
[17:39:58] <timeless> :(
[17:40:24] <timeless> delewis: err, yeah, a package is preffered, but i believe it's a package
[17:42:29] <timeless> darn
[17:42:48] *** nwf has joined #opensolaris
[17:42:59] <quasi> postwait: u2, u3 or nevada something?
[17:43:06] * timeless frowns
[17:44:09] <timeless> so, i'm at http://developers.sun.com/sunstudio/index.jsp and i clicked "Sun Studio 11" <http://developers.sun.com/sunstudio/documentation/ss11/index.htm> in "Sun Studio Components and Topics Area"
[17:44:12] <quasi> (just out of curiosity)
[17:44:15] <timeless> and got a pretty 404
[17:44:34] <quasi> timeless: still trying to download?
[17:44:37] <timeless> because the site was missing an l
[17:44:40] <timeless> quasi: yeah
[17:44:43] * timeless sighs
[17:44:54] <timeless> you'd think people would know *not* to include 'index.htm' in links
[17:45:26] <timeless> i keep getting enter username and password for sdn premium content
[17:45:45] <postwait> quasi: u2
[17:45:50] <postwait> u3 isn't out yet.
[17:45:59] <postwait> is it? I thought it was coming in Jan.
[17:46:46] <quasi> postwait: 11/06 is the current naming
[17:47:06] <quasi> postwait: it has quite a few zfs updates
[17:47:49] <quasi> timeless: you have to register for first - http://developers.sun.com/sunstudio/downloads/index.jsp - # If you are not already an SDN member, register here. SDN membership is free and entitles you to additional benefits.
[17:47:53] <ProfMikey> hi timeless...did you find the answer to the ultimate question or anything like that?
[17:48:16] <quasi> postwait: so far it is looking good in my tests, but I'll see what it does to the thumper
[17:48:17] <timeless> profmikey: my script turned up enough definitely stoeln files
[17:48:27] <timeless> and the rest seemed to not match, so i settled on my report
[17:48:34] <ProfMikey> :))
[17:48:37] <timeless> and reported, and now i'm back to bigger and more painful tasks
[17:48:40] <ProfMikey> glad to hear
[17:48:43] <timeless> like trying to get brandz working
[17:48:50] <timeless> or sunstudio or dbx
[17:48:55] <timeless> or scratchbox
[17:49:04] <ProfMikey> anyway I am off to home...shift handover so...later dudes and dudessses :)
[17:49:05] <postwait> is 11/06 out?
[17:49:22] <quasi> postwait: in beta
[17:49:50] *** nwf has quit IRC
[17:49:51] <quasi> postwait: I have it running on a couple of boxes so far
[17:49:58] <postwait> schduled for release in the first week of Nov?
[17:50:33] <quasi> any week of - i don't know yet
[17:52:17] <quasi> http://www.cuddletech.com/blog/ - 4th entry lists a few of the features
[17:53:10] * timeless is so very confused by zfs mount/friends
[17:54:00] *** nyati has joined #opensolaris
[17:55:05] <twincest> "network layer 7 cache" < hasn't that been in since S9 or so?
[17:55:41] <timeless> http://pastebin.ca/205159
[17:55:48] <timeless> ^ my current zfs problems :(
[17:56:19] <sommerfeld> twincest: no, the implementation was replaced in nevada
[17:56:23] <twincest> ah
[17:57:24] <elektronkind> twincest: re: your pastebin... "zfs get mountpoint root_pool/home/zones/linux/centos/install/scratchbox "
[17:57:31] <twincest> that wasn't mine
[17:57:59] <timeless> bad nick completion, it's mine
[17:59:39] * timeless waits for elektronkind to continue
[18:06:49] *** hell` has quit IRC
[18:08:14] *** MattMan has quit IRC
[18:15:44] <twincest> what is libtsol?
[18:15:59] <timeless>      libtsol - Solaris Trusted Extensions library
[18:16:06] <timeless> according to man libtsol :)
[18:16:34] <twincest> hm
[18:17:18] <quasi> yeah
[18:17:21] <timeless> http://docs.sun.com/app/docs/doc/819-2242/6n4hvao9s?a=view
[18:17:39] <quasi> twincest: looking at patches today? ;)
[18:17:39] <timeless> does anyone have a good google query to "always" get the right docs.sun.com man page for  a topic? :(
[18:17:44] <twincest> quasi: yes :)
[18:18:09] * quasi likes the new feature of auto installing on reboot
[18:18:17] <timeless> quasi: eh?
[18:18:26] <timeless> elektronkind: you were saying? :(
[18:18:27] <twincest> smpatch has done that for a while
[18:18:36] <quasi> yeah
[18:18:57] <quasi> it has, but still a lot nicer than the original having to go in single user
[18:19:15] <timeless> quasi: http://pastebin.ca/205159 is don't suppose you can explain what obvious thing i did wrong? :(
[18:19:54] <twincest> "To obtain the complete Solaris Trusted Extensions functionality support, please install the following patches: 118855-25 (or greater)  kernel patch"
[18:20:08] <twincest> this kernel patch is not so much with the existing at the moment
[18:20:24] *** Cyrille has quit IRC
[18:20:24] <timeless> heh\
[18:21:38] <quasi> timeless: missing zfs mount perhaps?
[18:21:59] <timeless> cannot mount 'root_pool/home/zones/linux/centos/install/scratchbox': mountpoint or dataset is busy
[18:24:11] <timeless> ok, halting the zone got that command to work
[18:26:21] <timeless> how do i get the standard automount back?
[18:27:18] *** anthony79 has quit IRC
[18:31:01] <timeless> oh
[18:31:08] <timeless> do i only get one mountpoint free?
[18:32:27] *** ZeuSun has joined #opensolaris
[18:33:49] *** vmhobbes has joined #OpenSolaris
[18:36:43] <ZeuSun> Is there something parallel to linux jiffies in solaris ?  I want to measure the time elapsed between entering a heavy function (in a kernel module) and the point of exit frpm that function; what is the recommended way ?
[18:37:02] <timeless> can't you use dtrace? :)
[18:37:28] <quasi> ZeuSun: do you need more resolution than microseconds?
[18:37:49] <ZeuSun> for the beginning, no
[18:38:12] <quasi> dtrace will do that for you very nicely then
[18:38:40] <ZeuSun> can i use also gettimeofday()?
[18:39:13] <timeless> what are you really trying to do? :)
[18:39:53] <ZeuSun> I want to know how heavy are some methods
[18:40:02] <ZeuSun> especially copymsg
[18:40:18] * timeless thinks that's one of the main reasons dtrace exists
[18:40:23] *** Dar is now known as Dar_HOME
[18:41:13] <ZeuSun> I am not familiar very much with dtrace ; the only thing I did with it was to use it against a java machine
[18:41:22] <ZeuSun> a java application, sorry
[18:41:23] <quasi> ZeuSun: at those units I wouldn't be surprised if you spent more time counting microseconds than doing the actual task
[18:41:25] <oxygene> and the best thing is that you don't need to have a "tuning-enabled" version and an optimized one
[18:41:49] <ZeuSun> quasi, I mean to call 10000 times copymsg in a method
[18:41:52] <ZeuSun> and then calc
[18:42:26] <quasi> ZeuSun: right
[18:42:42] <ZeuSun> suppose my module is called mymodule and my fuction is called "myfunction" ; how can I dtrace entry and exit times of this method ?
[18:42:59] <ZeuSun> Or can I simply use "gettimeofday"
[18:43:02] <ZeuSun> in such a case?
[18:43:48] <quasi> gettimeofday could work
[18:44:00] <ZeuSun> quasi, thnks
[18:44:16] <ZeuSun> BTW i saw that it the gettimeofday
[18:44:27] <ZeuSun> is almost not in use in the kernel
[18:44:57] <quasi> ZeuSun: otherwise, places to look for dtrace examples are /usr/demo and the dtrace toolkit
[18:45:47] <ZeuSun> quasi, ok, thnks again
[18:47:50] *** Sieghard has quit IRC
[18:49:52] *** alanc_away has quit IRC
[18:50:20] *** ZeuSun has quit IRC
[18:50:39] <timeless> oh
[18:50:43] <timeless> interesting
[18:51:08] <timeless> zoneadm doesn't appreciate it when you have a mounted filesystem under its management point
[19:08:29] *** alanc-away is now known as alanc
[19:14:09] <timeless> hrm, i wonder what magic is required to get /proc/sys/ working in brandz lx
[19:14:57] *** mlh has quit IRC
[19:18:25] <gdamore> has anyone ever considered using Solaris zones to run older releases of Solaris?  I would like, for example, to use a Solaris 8 userland in a zone.
[19:18:36] <timeless> it's been discussed
[19:18:55] <timeless> google will explain that it's "hard"
[19:18:56] <timeless> half the problem is implementing bugs
[19:18:57] <cypromis> how should that work ?
[19:19:02] <timeless> then you have to explain to a caller what version of solaris you have when someone asks
[19:19:03] <cypromis> lol
[19:19:37] <gdamore> heh.   yeah, i figured it would be "interesting" since we don't normally provide any compatibility layer for private userland->kernel APIs
[19:19:40] <quasi> gdamore: the answer is no - not doable at this time
[19:19:52] <jengelh> hey gdamore
[19:20:07] <gdamore> hey jengelh
[19:20:29] <jengelh> Back to that clone open...
[19:20:34] <gdamore> still, having some kind of VM to run Solaris 8 would be really, really nice.
[19:20:45] <gdamore> (i mean solaris 8/sparc, of course)
[19:20:51] <gdamore> yes, about the clone open....
[19:21:00] <quasi> gdamore: I was just about to say vmware ;)
[19:21:03] <jengelh> gdamore: I already have a 'working' streams driver.
[19:21:16] <gdamore> is it a true driver?
[19:21:21] <jengelh> gdamore: How true?
[19:21:23] <gdamore> (i.e. not a module or mux)
[19:21:35] <jengelh> gdamore: A module that is autopushed onto pts
[19:21:45] <twincest> the interesting part of the answer to the old solaris versions thing is it's actually easier to support linux because their public api is kernel<->libc
[19:22:24] <gdamore> that's true for NetBSD as well.  and they have versioning for doing backwards compatibility.  kinda cool.
[19:22:43] <gdamore> tho' for most cases i prefer the solaris API up in libc.
[19:23:07] <gdamore> jengelh: so if you are a module, then each stream _should_ get separate open/close pairs, IIRC
[19:24:20] *** lopa has joined #opensolaris
[19:24:50] <gdamore> however, if the user does non-clone duplicate opens on the pts device, i suppose you will only see the last close.
[19:25:13] <jengelh> gdamore: Yes, and the issue is that I cannot influence what the user does. (Basically, it's ">/dev/pts/1")
[19:25:28] <jengelh> gdamore: /etc/iu.ap contains "pts 0 16 rpldhk", rpldhk being the test module to show the "problem"
[19:25:51] <jengelh> as far as I can retell, I only observe "opens", but no closes.
[19:26:21] <gdamore> that's probably because pts does non-clone opens.  gimme a sec to look
[19:27:13] <jengelh> need source to rpldhk?
[19:27:18] <gdamore> no.
[19:28:46] <gdamore> yeah, pts doesn't do clone opens -- it always uses the same minor number.
[19:30:44] <gdamore> as a _module_, i think you're stuck.  you don't get any control over stream plumbing in this case.
[19:30:50] <jengelh> ah I see
[19:31:08] <jengelh> i'm still guessing about clone opening
[19:31:29] <jengelh> when userspace open("/dev/pts/1",O_RDWR), it always uses the same minor, no?
[19:31:34] <gdamore> clone opens are done by the _driver_, and basically you get one file descriptor to one minor number
[19:31:42] <gdamore> yes,.
[19:32:05] <gdamore> but for other (clone devices) this minor could be changed.
[19:32:28] <jengelh> So, to put it short: The Linux way is: two opens(/dev/pts/1) in userspace generate two open calls and two close calls in the driver, without any clone magic AFAICT. I would like to see this behavior in other OS too.
[19:33:02] *** Elendal has quit IRC
[19:33:28] <gdamore> maybe the right thing to do is to change pts into a cloning driver.  but the terminal subsystem in solaris is largely black magic to me. :-)
[19:33:54] <jengelh>  /dev/ptmx is a cloner, is not it?
[19:33:59] <gdamore> changing _streams_ is not likely going to happen.  far too much code depends on the existing behavior.
[19:34:02] <gdamore> let me look.
[19:34:25] <gdamore> looks that way
[19:34:51] <jengelh> then I am a little bit puzzled how pts could be made a clone device too when all /dev/pts/ nodes need a distinct minor number
[19:35:25] <gdamore> you can have multiple minor numbers that are fixed, and then use teh rest  for cloning
[19:35:34] <gdamore> i.e. its how you carve up your minor number space.
[19:35:58] <gdamore> this is, in fact, how old GLD style 1 device nodes work for most x86 NICs.
[19:36:16] <gdamore> some bits indicate the device instance, and some indicate the clone number.
[19:37:33] <jengelh> oh dear
[19:38:02] <gdamore> that's a bit too simple.  usually you set aside numbers e.g. values less than 256 are "master" opens (open of the cloen device), and the driver allocates clone minor numbers with values greater than 256 (or some other value)
[19:38:39] <jengelh> does it suffice to say that all opens under linux are cloned?
[19:39:25] <gdamore> linux doesn't have streams.  i don't think it uses the same framework for changing minor numbers.  but i have to look.
[19:39:27] <gdamore> just a sec.
[19:40:21] <jengelh> Basically it all comes down to: Do I need to do a hack[1] for Solaris too?
[19:40:22] <jengelh> [1] http://svn.sourceforge.net/viewvc/ttyrpld/trunk/kpatch/freebsd-6.1.diff?revision=2&view=markup
[19:41:01] *** benr has joined #opensolaris
[19:41:02] *** ChanServ sets mode: +o benr
[19:41:14] <gdamore> i dunno, its kinda hard to figure out out first blush what you are trying to do
[19:42:07] <jengelh> There is a "bug" in various Unix-style kernels (BSD, Solaris), that any device driver (streams or not) that is opened twice in userspace gets opened twice in the kernel (i.e. the function is called twice), but only closed once.
[19:42:12] *** yippi has joined #opensolaris
[19:42:25] *** mlh has joined #opensolaris
[19:42:31] <gdamore> right.  and this is not true for linux.
[19:42:42] <jengelh> Exactly.
[19:42:43] <gdamore> (or at least that appears to be the case)
[19:43:01] <jengelh> (let's not consider dup() - it's fine as it is)
[19:43:27] <gdamore> if you need to match open & close, then yes, you need a "hack".
[19:44:06] <gdamore> but, its not too hard to match first open vs. clone open against close.
[19:44:19] <gdamore> s/clone open/duplicate open/
[19:44:45] <gdamore> basically a duplicate open is going to have a different device number returned back to the caller.
[19:45:20] <gdamore> i'm not sure if you can see the dev_t value on the downstream module from a module, however.   that part might be tricky
[19:45:31] <jengelh> But the close() routine is never called, which... is the event I am actually depending on.
[19:46:15] <gdamore> ah, right, this is something probably with pts, which is that once plumbed, the configuration stays around forever.
[19:46:34] *** koolniczka has joined #opensolaris
[19:46:53] <gdamore> why do you need to know about close(), btw?
[19:47:10] <jengelh> that's part of the program :)
[19:47:39] <gdamore> what i mean is, are you just release allocated resources (alloc'd in open), or is there some other special action you need to take?
[19:48:02] <OnkelSchorsch_> hm. which board did sun base the ultra 20 m2 on?
[19:48:18] *** nwf has joined #opensolaris
[19:48:57] <quasi> OnkelSchorsch_: I'd guess something very similar to what goes into the x2100 M2 ;)
[19:49:16] <gdamore> wow, this cobalt (250MHz MIPS) is taking ~1 hour to compile bind 9.  yikes.
[19:50:15] <OnkelSchorsch_> thanks, quasi. the kingston site tells me, they use the same memory too. :(
[19:52:20] <jengelh> gdamore : I'll try to sort it out.
[19:52:59] <quasi> OnkelSchorsch_: I remember reading what chipsets it was, but that's probably not going to do you much good
[19:53:21] <jengelh> hm where is the 'd_open' member declared? [what .h file] - can't seem to find it using grep
[19:53:37] <jengelh> nm, called cb_open :/
[19:53:56] <gdamore> right. but for streams you want q_open
[19:54:21] <jengelh> yeah :)
[19:55:13] <gdamore> actually, its qi_qopen.  argh.
[19:55:39] <gdamore> and sys/stream.h.  (member of struct qinit)
[19:56:39] *** yippi has quit IRC
[19:56:48] <gdamore> btw, that's crucial.  if you declare a streamops, but use cb_open/close instead of qi_qopen/qi_qclose, they won't get called!
[20:01:14] <gdamore> does anyone know if QEMU can host guest Solaris OS?
[20:01:17] <gdamore> (on SPARC)
[20:03:13] <delewis> gdamore: I seriously doubt it but I haven't tried -- I think the emulation is good enough to run Linux, but that isn't say much :-)
[20:03:53] <jengelh> would a b45 kernel work with a b39 userland?
[20:03:58] <OnkelSchorsch_> quasi, does it have an nVIDIA or broadcom chipset?
[20:04:24] <gdamore> probably, but there are likely some gotchas.  :-)
[20:04:47] <jengelh> i am not in the mood to reinstall sol now :d
[20:05:03] <jengelh> aww too many gotchas already
[20:05:20] <delewis> jengelh: define userland
[20:05:29] <jengelh> delewis: Anything but the kernel.
[20:05:43] <delewis> if you mean a b45 kernel with the rest of b39 ON consolidation -- you're likely to run into some trouble, i would imagine.
[20:06:02] <jengelh> basically I have b39 installed, and the b39 usr/src/uts, but that wants ../Makefile.master whcih I don't have, and unless someone knows where to get the ON from ... b39,
[20:06:04] <delewis> if you're just referring to stuff that isn't in the ON consolidiation, in general, definitely not.
[20:06:09] <delewis> jengelh: ah
[20:06:19] <delewis> yeah, I can see having a few problems :-)
[20:06:26] <jengelh> so I would need the ON archive for b39 again
[20:06:35] <jengelh> if that's available somewhere - then the better
[20:07:35] <sickness> i'm back
[20:08:26] <delewis> jengelh: http://svn.genunix.org/repos/on/tags/
[20:08:31] <delewis> though, I don't see b39 listed
[20:14:26] *** solarisjon has joined #opensolaris
[20:17:11] *** gtc_ has joined #opensolaris
[20:17:12] * jengelh tickles gdamore
[20:17:28] <jengelh> gdamore: sunstudio11-ii-20060829-sol-x86.tar.Z -- why the outdated compression format?
[20:18:33] <timeless> jengelh: the version from dsc or whatever is .bz2
[20:18:37] <alanc> at least it's finally a patent-free compression format
[20:18:53] <timeless> alanc: which format has patent issues?
[20:18:58] <alanc> .Z
[20:19:09] <jengelh> timeless : I am on SDC, are not I?
[20:19:21] <jengelh> just followed the link on http://opensolaris.org/os/community/tools/sun_studio_tools/sun_studio_11_tools/
[20:19:21] <timeless> jengelh: one sec
[20:20:07] <delewis> I generally download the one from SDC
[20:20:10] <timeless> you want some other link
[20:20:12] <delewis> which uses SVR4 packages
[20:20:23] <timeless> http://developers.sun.com/sunstudio/downloads/index.jsp
[20:20:25] <delewis> (and thus patches can be applied to it)
[20:22:19] <jengelh> browser icon spins indefinitely
[20:22:26] <timeless> ok, i'm confused
[20:22:37] <timeless> i grabbed the studio thing from sdn
[20:22:43] <timeless> and the result is a .tar.bz2
[20:22:50] <timeless> but i don't see pkgs in it
[20:23:37] * timeless runs installer and ignores everything else (including the pdf readme)
[20:23:42] <jengelh> hm seems like i already downloaded them a while ago
[20:23:48] <jengelh> heh
[20:23:53] <jengelh> things pile up when you got a big harddisk :p
[20:24:09] <timeless> hrm
[20:24:16] <timeless> perhaps i want /opt to be in zfs land
[20:24:31] *** peteh has quit IRC
[20:24:57] *** peteh has joined #opensolaris
[20:30:07] <timeless> hrm, i guess it's kinda bad to have /opt on a file system that's different from the one w/ the package database
[20:30:43] * timeless wonders how hard it is to split a package database
[20:30:49] <quasi> OnkelSchorsch_: my x2200s have both nvida and broadcom nics, but the mobo chipset is nvidia
[20:32:29] * timeless grumbles
[20:34:19] <OnkelSchorsch_> quasi, thanks. I think the u20 m2 is using the tyan tomcat n3400b with a different on-board graphics. http://www.tyan.com/products/html/tomcatn3400b.html
[20:34:35] <OnkelSchorsch_> but this one also takes non-ECC memory. hmm. weird
[20:35:16] <quasi> OnkelSchorsch_: I thought it was opteron rather than amd64
[20:36:29] <OnkelSchorsch_> yep. strange. but it is also listed in the am2, opteron, sempron section of http://www.tyan.com/support/html/cpu_athlon_duron_opteron.html
[20:37:28] * quasi is seriously contemplating buying an u20
[20:38:06] <Error_404> do it
[20:38:07] <OnkelSchorsch_> I am buying one. just the base system without HDs and memory.
[20:38:16] <quasi> strange case though
[20:38:26] * OnkelSchorsch_ loves the case.
[20:38:36] <OnkelSchorsch_> ;)
[20:38:50] <Error_404> looks like a g5 powermac
[20:39:19] <quasi> OnkelSchorsch_: have you seen the inside pics from the system handbook? looks a bit crowded
[20:39:54] <OnkelSchorsch_> looks fine to me http://www.sun.com/desktop/workstation/ultra20/gallery/stills.xml?t=3
[20:40:17] <OnkelSchorsch_> or, what is bothering you? :)
[20:40:37] *** Symmetria has joined #opensolaris
[20:40:38] <Symmetria> hey all
[20:40:56] <Symmetria> hrm, for some reason mozilla dies when I try and run it through ssh -X
[20:41:01] <Symmetria> I get some BadWindow error back
[20:41:22] <timeless> symmetria: ./run-mozilla.sh -g -d `which dbx` ./mozilla-bin --sync
[20:41:28] <quasi> OnkelSchorsch_: looking at that pic - nvidia nforce pro 3400
[20:41:32] <timeless> at least, that's the normal way to debug things
[20:41:45] <timeless> if you're using an snv_ build, you should have enough symbols to give me something useful
[20:42:13] <timeless> i used mozilla via ssh forwarding w/ solaris 7/8/9 in 99/00
[20:43:35] <timeless> of course, lots of things changed since then, also please do indicate what version of mozilla you're using (and snv for that matter)
[20:43:48] <OnkelSchorsch_> quasi, hehe. so it's based on the tyan tomcat n3400b. nice.
[20:44:31] <Symmetria> timeless heh, I went and did a package add straight off the solaris 10 cd
[20:45:41] <quasi> OnkelSchorsch_: looks like it could be
[20:46:27] <quasi> OnkelSchorsch_: except that still isn't opteron
[20:46:41] <quasi> ehrm, yes it is
[20:47:16] <timeless> Symmetria: i'd expect that's equivalent
[20:47:54] <OnkelSchorsch_> hm. quasi, do you think, one needs to take ECC RAM for the opteron?
[20:48:40] <OnkelSchorsch_> do you know of anybody, besides peter tribble, who owns an ultra 20 m2 or x2100 m2?
[20:49:27] <quasi> I know of someone with an x2100m2
[20:52:49] <quasi> OnkelSchorsch_: the odd thing is that the small model (at least on the .nl site) seems to be delivered with 1*512M ddr - not ddr2
[20:53:32] <quasi> OnkelSchorsch_: but other than that, the system handbook does call for ECC - 512MB, 1GB, 2GB Unbuffered DIMMs only - DDR2-667
[20:54:03] <OnkelSchorsch_> yep. and that one is really expensive :(
[20:54:48] <quasi> well - have you checked kingston or crucial or someone like that?
[20:57:33] <OnkelSchorsch_> yep. the one kingston specified for the ultra 20 m2 costs 341 euros + tax per 2 GB kit
[20:58:00] <quasi> that's somewhat less than sun
[20:58:38] *** lopa has quit IRC
[20:58:58] *** tarzeau has quit IRC
[21:13:03] *** gisburn has joined #opensolaris
[21:14:21] <ProfMikey> err folks did you hear bout this libnspr4.so design error (quiet not sure if thats a new one)
[21:14:53] <ProfMikey> src: http://www.securiteam.com/exploits/6U00D1FH5C.html
[21:16:07] <timeless> it's fixed in cvs
[21:16:21] <ProfMikey> oh, good
[21:16:26] <ProfMikey> just wanted to ensure
[21:16:33] <quasi> and there was a patch to nspr also from sun iirc
[21:17:13] <twincest> 6467033 Security vulnerability in the way NSPR library creates log files
[21:17:20] <twincest> fixed in the patches listed in the page you just linked to
[21:17:35] <gisburn> ugh
[21:17:40] <gisburn> is this #mozilla ?
[21:17:43] <timeless> https://bugzilla.mozilla.org/show_bug.cgi?id=351470
[21:17:52] * gisburn checks the channel name...
[21:17:53] <quasi> 119214-10 NSS_NSPR_JSS 3.11.3_x86: NSPR 4.6.3 / NSS 3.11.3 / JSS 4.2.4
[21:17:55] <ProfMikey> aww updated
[21:17:58] <gisburn> timeless: Hi! :-)
[21:18:03] <ProfMikey> ta folks
[21:18:04] <timeless> hi gisburn
[21:18:15] <twincest> gis: yes, god forbid we discuss solaris security vulnerabilities in #opensolaris
[21:18:16] * gisburn looka for sickness
[21:18:43] *** andrei has joined #opensolaris
[21:18:44] <gisburn> twincest: who did forbid that ?
[21:18:46] <gisburn> who ?
[21:18:51] *** andrei has left #opensolaris
[21:18:57] *** andrei has joined #opensolaris
[21:19:04] <timeless> twincest: gisburn's parsing of sarcasm isn't that great, give him a break
[21:19:08] <timeless> he isn't a native speaker
[21:19:17] <timeless> and he's particularly tired of dealing w/ mozilla
[21:19:22] <gisburn> timeless: oh
[21:19:30] <gisburn> timeless: this is not a problem of the language
[21:19:41] <gisburn> timeless: I am very bad in sarcasm
[21:20:00] <gisburn> timeless: it usually backfires when people try it
[21:20:52] <gisburn> timeless: like that someone suggested years ago that I could install a large accerlator segment (around 2600kg) after I did lots of other work that day.
[21:21:05] <gisburn> timeless: after that weekend the stuff was installed, too.
[21:21:18] <gisburn> timeless: (which noone thought that this would be possible)
[21:21:48] <gisburn> timeless: since then most people avoided trying sacrasm in my direction
[21:23:12] <gisburn> !summon sickness
[21:23:24] <gisburn> !summon kupfer
[21:24:40] <gisburn> BOOORING
[21:24:58] <gisburn> We need a mozbot here... really...
[21:25:51] <quasi> gisburn: or you could just jump over in #mozilla instead of here... ;)
[21:27:17] <gisburn> Gman: ping!
[21:29:15] <gisburn> grumpf
[21:29:33] <timeless> quasi: can dbx debug things in brandz?
[21:29:42] <timeless> specifically i need to know why execve is failing
[21:29:53] * gisburn adds Gman to the "list of people who deserve 10mins in a deep pit filled with komodo dragons"
[21:30:20] <gisburn> timeless: in theory yes, but AFAIK there is a small problem with brandz and dbx and a missing symbol
[21:30:36] <quasi> timeless: just to get this absolutely clear - I do not use brandz, never have and probably never will
[21:30:49] <timeless> quasi: oops, sorry
[21:30:59] <quasi> maybe grab the linux version of sun studio?
[21:31:06] <timeless> yeah, i have it
[21:31:16] <gisburn> quasi: what's your problem with brandz ?
[21:31:19] <timeless> but i'm not sure how syscalls work on linux/solaris
[21:31:23] * gisburn curses http://bugs.grommit.com/show_bug.cgi?id=29
[21:31:52] <quasi> gisburn: I get enough linux elsewhere - no need to pollute my solaris boxes if I can help it ;)
[21:32:13] * timeless sighs
[21:32:40] <timeless> all i want is to be able to run enough of *any* apt that will let me retrieve apt source balls as directories so that i can run solaris apps on the result
[21:32:48] <timeless> i don't really need brandz, or shouldn't
[21:32:49] <Error_404> yeah, i'm not sure i understand the justification for brandz, apart from marketing "it runs linux binaries!"
[21:33:06] <gisburn> Error_404: I don't understand it either.
[21:33:14] * timeless shrugs
[21:33:16] <gisburn> Error_404: maybe related to the "zones" and SMF madless.
[21:33:19] <quasi> timeless: nextensa? you could probably use part of that
[21:33:25] <gisburn> s/madless/madness/
[21:33:32] <timeless> quasi: i can't figure out how to get the apt from nextenta
[21:33:40] <timeless> it seems to be in a .deb
[21:33:40] <gisburn> Does brandz support Solaris/SPARC ?
[21:33:45] <timeless> which is pretty unhelpful
[21:33:50] <Error_404> gisburn: no
[21:33:57] <gisburn> Error_404: ugh
[21:33:58] <timeless> gisburn: thats listed as a project someone could work on:)
[21:34:01] <gisburn> Error_404: why ?
[21:34:12] <gisburn> timeless: I have my burden.
[21:34:19] <timeless> how many commercial apps are there for sparc-linux? :)
[21:34:32] <gisburn> timeless: and I won't start a new project until ksh93-integration gets it's first putback.
[21:34:33] <Auralis> how about, like zero
[21:34:52] <timeless> that's a pretty good financial reason for sun not to worry about it
[21:34:56] <gisburn> Auralis: 1 / Inf
[21:35:23] <gisburn> timeless: yes, but for consistency it would be nice to have both.
[21:35:32] <gisburn> timeless: but Sun is weired in many ways.
[21:35:47] <alanc> "nice to have" doesn't get you resources in a company that just had to let 5,000 people go
[21:35:55] <alanc> sad facts of life
[21:36:29] <Auralis> brandz on sparc i think is pointless and a waste of resources
[21:36:31] <gisburn> timeless: the ksh93 thing was hidden under the carpet too long... with the result that I was hit by the full extend of "wrath against sun" collected in the last six (or more) years
[21:36:38] <gisburn> and this is NOT funny
[21:37:25] * delewis snickers
[21:37:26] <delewis> :-)
[21:37:27] * timeless shrugs
[21:37:41] <timeless> delewis: how about helping me fight execve instead of snckering
[21:38:40] <gisburn> timeless: what about trying % truss -u :: ... # ?
[21:38:45] <gisburn> timeless: or dtrace ?
[21:38:57] <gisburn> timeless: both should work with brandz ...
[21:41:20] <timeless> # truss -s -e -l -f -p 18166
[21:41:21] <timeless> truss: unrecognized signal name/number: -E
[21:42:58] <gisburn> ?!
[21:43:11] <gisburn> timeless: is that in the zone or from the global zone ?
[21:43:25] <timeless> that's from global
[21:43:35] <gisburn> weired.
[21:43:45] <gisburn> !summon Gman
[21:43:53] <gisburn> !summon tentaclemonster
[21:43:55] <timeless> i don't quite understand where -E comes from given that i didn't pass -E
[21:43:59] <gdamore> anyone here ever used simics with solaris before?
[21:43:59] <gisburn> !summon endofworld
[21:44:00] <Symmetria> stupid question, but is there some kinda interface for disk management and partition management in solaris in console mode (non-x in other words)
[21:44:05] <Symmetria> something interactive
[21:44:13] <gdamore> fdisk?
[21:44:19] <gisburn> Symmetria: fdisk ?
[21:44:24] <timeless> symmetria: elinks doesn't count? :)
[21:44:38] <hile_> format(1M) doesn't count?
[21:44:40] <Symmetria> gisburn I was looking for something interactive, and fdisk as such isnt it :p
[21:44:54] <Symmetria> heh since I dont know what the hell Im doing under solaris yet I'd rather try the easy route first :)
[21:45:15] <gisburn> Symmetria: try $ rm -Rf /* #
[21:45:30] <gisburn> Symmetria: that formats the whole system and downloads illegal warzs for free, too.
[21:45:36] <gdamore> fdisk is interactive, but not friendly
[21:45:42] <Symmetria> errr gisburn Im used to freebsd and linux, and Im a CTO, Im not an idiot
[21:46:13] * gisburn is unhappy... no victium for him today...
[21:46:21] <gisburn> s/victium/victim/
[21:46:34] <gdamore> if you are doing this on x86, you have to be aware that there is the MS-DOS partition table, and then there is the embedded solaris partition table. :_0
[21:46:49] <gisburn> Does anyone remebers rich's fork() bomb code ?
[21:46:56] <hile_> which rich?
[21:47:08] <Symmetria> gda, I just wanna partition and then put a zfs file system onto the san, I've managed to get the san drivers configured and I've got solaris seeing the san
[21:47:13] <gdamore> was it for (;;) fork(); ?
[21:47:22] <gdamore> i've nver used zfs yet.
[21:47:25] <gisburn> hile_: doesn't matter, I just want to get Symmetria to test it... =:-)
[21:47:39] <timeless> symmetria: mostly you just use zpool create
[21:47:48] <timeless> it generally just works
[21:47:51] <gisburn> gdamore: yes, but it used pipes to keep the children alive.
[21:48:19] <hile_> I don't use zfs
[21:48:27] <gdamore> heh.  the children won't die, they just keep trying to fork().
[21:48:41] <gdamore> now if you want to have a pipe to _kill_ the children, that's different.
[21:48:42] <gisburn> gdamore: AFAIK $ x() { x | x } ; ( x & ) or something like that.
[21:49:00] <gdamore> oh, you want _shell_ code to do it.
[21:49:32] <gisburn> gdamore: AFAIK $ x() { (x | x) & } ; ( x ) # may be better...
[21:50:03] * gdamore logs in to gisburn's system to try out a new bit of shell code someone just gave him....
[21:50:23] <Error_404> hile_: you should, it's fantastic...
[21:50:38] <timeless> gdamore: if you end up in some restricted zone, it shouldn't matter, right? :)
[21:50:39] <Error_404> unless it confuses your bios and causes the machine to not boot
[21:51:03] <gdamore> does zfs live on top of FAT partitions, Solaris partitions, or does it just use the raw disk?
[21:51:16] <Error_404> gdamore: yes, yes, and yes
[21:51:32] *** coffman has quit IRC
[21:51:38] <gdamore> heh.  should've guessed, i suppose.
[21:53:51] <elektronkind> ASEFJASDFK!@#$
[21:53:59] <elektronkind> I hate our sun VAR
[21:54:41] <timeless> # truss -f -p 18204
[21:54:41] <timeless> 18204:      Received signal #5, SIGTRAP [default]
[21:54:41] <timeless> 18204:        siginfo: SIGTRAP TRAP_BRKPT addr=0x08054811
[21:54:49] * timeless tries to figure out what in the world that means
[21:55:29] <gdamore> are you running this in a debugger?
[21:55:43] <gdamore> (e.g. gdb/dbx?)
[21:55:43] <timeless> 18204 is being debugged by gdb, yes
[21:55:57] <gdamore> then you just hit the gdb break point.
[21:55:58] <gisburn> timeless: http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6175434
[21:57:09] <gdamore> gisburn: heh.  I'm not entirely sure that in this case it should be viewed as a bug.  if the process takes the SIGTRAP signal, then I think truss should see it.
[21:57:27] <timeless> gisburn: that was marked as fixed
[21:57:36] <timeless> and the example works for me
[21:58:04] <gdamore> It only refereces to dtrace.  I think gdb break points should still see the SIGTRAP.
[21:58:20] <Gman> gisburn: what's up?
[21:58:27] <Symmetria> hrm
[21:58:42] <Symmetria> I cant see to see this SAN in solaris itself, the HBA management software shows teh session as active
[21:58:53] <Symmetria> and its definately talking to the HBA card correctly
[21:59:17] * timeless is confused
[21:59:42] * timeless tries to decide if gisburn and gdamore are the same
[21:59:48] <gisburn> Gman: !summon kupfer # is broken
[22:00:06] <gisburn> timeless: gdamore is my twin.
[22:00:25] *** glagasse has quit IRC
[22:00:40] <gisburn> timeless: (no, I am just making jokes)
[22:00:48] <Symmetria> woooooooooooot
[22:00:49] <Symmetria>        1. c2t3d0 <LEFTHAND-iSCSIDisk-63-3.88TB>
[22:00:49] <Symmetria>           /pci@0,0/pci8086,244e@1e/pci1077,124@0,1/sd@3,0
[22:00:50] <Symmetria> !
[22:01:44] <lasseoe> clear
[22:01:46] <lasseoe> oops
[22:03:12] *** mazon is now known as Mazon
[22:03:28] *** pjlv has joined #opensolaris
[22:04:17] <Symmetria> apool                  3.8T    24K   3.8T     1%    /mirror/iscsi-san-1
[22:04:18] <Symmetria> !
[22:04:19] <Symmetria> bingo
[22:04:25] * Symmetria celebrates
[22:04:45] <Symmetria> that was a lot easier than when I tried to do it under linux the first time ;p
[22:06:45] *** solarisjon has quit IRC
[22:06:57] <andersmo> 3.8T... yummy. =)
[22:07:23] <Symmetria> heh andersmo thats one of 3 of them
[22:07:46] <Symmetria> and internal array on machine Im aiming to install this setup on is another 6 x 400gig disks as well
[22:08:10] *** andrei has left #opensolaris
[22:08:19] <Gman> gisburn: ok, i've pinged him
[22:10:32] *** kupfer has joined #opensolaris
[22:10:34] *** ChanServ sets mode: +o kupfer
[22:11:02] <kupfer> hi gisburn, what's up?
[22:11:10] <gisburn> kupfer: Hi!
[22:11:29] *** y___schil is now known as schily
[22:12:14] <timeless> ok, i replaced gdb w/ printf/scanf
[22:12:23] <timeless> now i can watch from truss
[22:12:24] <benr> Anyone a fan of Solaris IPQoS Network Flow Accounting?
[22:13:57] <timeless> 18241:  execve("/bin/sh", 0x08047B4C, 0xFE8892E8)       Err#2 ENOENT
[22:15:55] <alanc> bah, security bug in nvidia closed-source driver...  http://www.rapid7.com/advisories/R7-0025.jsp
[22:16:29] <timeless> nice
[22:16:58] <Error_404> Symmetria: that's a lot of porn you got there
[22:18:32] <Symmetria> heh error eventually solaris will be the operating system I wanna run on mirror.ac.za once Im comfortable with it
[22:18:35] * delewis waits for the arguments against closed-source drivers
[22:18:35] <Symmetria> I wanna get linux off it
[22:19:00] <alanc> heh - sample exploit includes this comment:     *
[22:19:00] <alanc>      * "It's so hard to write a graphics driver that open-sourcing it would
[22:19:00] <alanc>      *  not help."
[22:19:00] <alanc>      *    - Andrew Fear, Software Product Manager (NVIDIA Corporation).
[22:19:04] <Symmetria> *shrug* linux hasnt proved to be stable, and considering thats the largest mirror server on the continent it needs to be stable
[22:19:22] <jengelh> so it speaks for itself
[22:20:16] <delewis> Symmetria: what stability problems are you encountering?
[22:21:39] <stevel> jengelh: did you find the b39 sources you were looking for?
[22:22:06] <trygvis> I wonder, why would the nfs directories have uid of nobody?
[22:22:10] <Symmetria> delewis linux has major memory leakage problems in certain aspects of its networking
[22:22:36] <trygvis> I thought if I did zfs set sharenfs=on pool0/home it would become a r/w share?
[22:22:37] *** LoBoGoL has joined #opensolaris
[22:22:43] *** LoBoGoL has left #opensolaris
[22:23:02] <Symmetria> take a linux box with an e1000 network card, adjust the tcp window sizing to counter a 300ms latency on 200mbit of bandwidth, then actually push between 100 and 200mbit for 1 to 3 hours
[22:23:07] <Symmetria> the machine will fall over and die, outta memory
[22:23:21] <Symmetria> even if there is nothing running on it except an ftp daemon of your choice which is taking no memory
[22:23:22] <gdamore> alanc: you have got to be kidding, right?  i've _done_ a graphics driver, how hard can it be?
[22:23:25] <Symmetria> and it will also refuse to use swap
[22:23:46] *** eokyere has joined #opensolaris
[22:23:51] <gdamore> (granted, getting OpenGL done is "hard".)
[22:24:21] <alanc> gdamore: that was a response from last year I think to the "Why won't nvidia open source?" question
[22:24:53] <alanc> of course it doesn't explain their forcing XFree86 to obfuscate the nvidia driver by replacing all constant names with hex values either
[22:25:09] <gdamore> they should keep their 3D OpenGL libraries closed but open up the 2D stuff.  that would be a huge sep forward, and probably avoid giving away the company jewels
[22:25:42] <kupfer> trygvis: if you're using NFSv4, the client and server probably disagree about the NFS domain name.
[22:26:14] <trygvis> hmm .. how can I check that? I'm using automount to mount /home dirs from a zone
[22:26:21] <trygvis> the global zone is sharing it
[22:26:45] <alanc> http://cvsweb.xfree86.org/cvsweb/xc/programs/Xserver/hw/xfree86/vga256/drivers/nv/Attic/README.RIVATNT.diff?r1=1.1.2.2&r2=1.1.2.3&hideattic=0&only_with_tag=xf-3_3_3
[22:26:57] <stevel> trygvis: cat /var/run/nfs4_domain
[22:27:16] <trygvis> they're the same on both sides
[22:28:01] <kupfer> stevel: are you sure the client zone has its own /var/run/nfs4_domain?
[22:28:23] <stevel> kupfer: i thought it did
[22:28:44] <stevel> though i have no proof :-P
[22:28:56] <kupfer> they added /var/run/nfs4_domain after I left the group, so I don't know how it's implemented.
[22:29:12] <trygvis> /var is not lofs mounted so yes, they both have one
[22:29:19] <stevel> the zone doesn't inherit /var or /var/run, so it has its own fs there
[22:30:33] <kupfer> well, if they're both using the same v4 domain, I don't know what the problem is.
[22:31:32] *** slowhog_home has joined #opensolaris
[22:31:33] <Symmetria> question, how do I register my solaris system so I can use smpatch
[22:33:23] <quasi> Symmetria: http://sunsolve.sun.com/search/document.do?assetkey=1-9-82688-1
[22:34:17] <trygvis> hmm .. I wonder what's happening
[22:38:59] <Symmetria> ok, next question *grin* how do I convince sun to let me mirror whatever Im allowed to mirror to add more content to mirror.ac.za, since its the central download point for every university on the african continent *grin* gotta find someone to ask about that
[22:39:38] <Error_404> you can mirror opensolaris
[22:39:49] *** dj2 has joined #opensolaris
[22:39:52] <Error_404> genunix does it
[22:40:38] <Symmetria> error sweet, definately something to add to the list
[22:41:03] <stevel> just to clarify.... you can mirrour the freely redistributable bits - but not SXCR
[22:41:25] <Error_404> yes, sorry
[22:41:34] <Symmetria> heh, I have *WAY* 2 much free disk space in that system, I need to find stuff to fill it up :p (got 7.8 terabytes free, and thats after I mirrored * linux distributions, * bsd distributions and the entire global bioinformatics databases!)
[22:42:13] <Symmetria> (less than 50% full)
[22:43:25] <Error_404> now don't you feel silly for paying for all that storage?
[22:44:36] *** Kernel86_ has joined #OpenSolaris
[22:44:44] *** Kernel86|Laptop has quit IRC
[22:45:25] <Symmetria> error nah *shrug* I had budget I needed to burn
[22:46:27] *** sommerfeld has quit IRC
[22:48:15] <quasi> Symmetria: http://www.apache.org/info/how-to-mirror.html
[22:48:56] <Error_404> speaking of apache, is it just me, or is apache actually supporting 2 or 3 competing products in a number of arenas?
[22:49:14] <mrdeviant> Symmetria, you may also want to mirror openoffice, blastwave, sunfreeware, fink, darwin ports, gnu
[22:49:53] *** benr has quit IRC
[22:50:29] <Error_404> geronimo / tomcat, jackrabbit/maven and so on
[22:50:36] <quasi> Error_404: it is not just you
[22:50:51] *** sommerfeld has joined #opensolaris
[22:50:51] *** ChanServ sets mode: +o sommerfeld
[22:50:52] <OnkelSchorsch_> Symmetria, you could also mirror docs.sun.com. it's only 5.5 to 6 GB though
[22:50:53] <trygvis> Error_404: ehm, are you comparing jackrabbit and maven?
[22:51:09] <quasi> trygvis: I was just about to ask that as well
[22:51:10] <trygvis> Error_404: and geronimo and tomcat are definitely not the same thing. geronimo are embedding tomcat
[22:51:42] <trygvis> though, apache is all about supporing all kinds of similar products .. they have a lot of competing projects
[22:51:52] <Symmetria> OnkelSchorsch_ cool, will do that
[22:51:59] <gisburn> stevel: ping!
[22:51:59] <Error_404> okie, so it's mostly just my confusion then
[22:52:00] <quasi> trygvis: it is a tad worse in the "web services" department - there's much morecompetition
[22:52:05] <stevel> gisburn: pong
[22:52:15] <trygvis> indeed, that's where all the fighting is
[22:52:18] <gisburn> stevel: when will the B51 sources be released ?
[22:52:27] <trygvis> OnkelSchorsch_: is it possible to mirror docs.sun.com? I'd love to mirror it internally
[22:52:41] <stevel> gisburn: when they're ready?
[22:52:57] <stevel> the schedule says 10/23
[22:53:09] <gisburn> stevel: ahhrrgglll
[22:53:26] * gisburn runs around and screams
[22:53:45] <gisburn> seven horrible days of waiting ?
[22:54:09] <OnkelSchorsch_> trygvis, you'd lose the search function though. but if you got the document ID you won't get bothered by "internal server error"s anymore
[22:54:25] *** leal has joined #opensolaris
[22:54:35] <richlowe> what b51 will contain isn't completely final currently.
[22:54:46] <trygvis> I can make those links myself as long as I don't have to go to the way to slo sun server
[22:55:16] <OnkelSchorsch_> downloading it is gonna take a while ;)
[22:55:24] <leal> hello, has solaris express 01/06 brandZ?
[22:55:34] <gisburn> richlowe: are there any specific problems ?
[22:55:37] *** lisppaste3 has quit IRC
[22:55:38] <gisburn> leal: unlikely
[22:56:20] <leal> "Now that BrandZ has integrated into Solaris, this whole section has been obsoleted"
[22:56:23] <gdamore> leal, i don't think so.
[22:56:25] <trygvis> OnkelSchorsch_: that's ok .. do you know where/how I can mirror it?
[22:56:27] <leal> http://www.opensolaris.org/os/community/brandz/install/
[22:56:30] *** delewis has quit IRC
[22:56:40] <Error_404> leal: when you upgrade to build 49, it has it
[22:56:43] <gdamore> its integrated into Solaris Nevada, not Solaris 10
[22:56:55] <richlowe> I don't know about specific problems, but the gate doesn't even close for several hours yet.
[22:57:07] <OnkelSchorsch_> trygvis, I just used wget -r and that convert internal links option
[22:57:13] <leal> solaris nevada is express, right?
[22:57:33] <sickness> nope
[22:57:41] <sickness> nevada is community release, a step ahead of express ;P
[22:57:48] <trygvis> aha, oki
[22:57:53] <richlowe> nope, both are Nevada-based.
[22:58:00] <timeless> that makes no sense
[22:58:03] <richlowe> but SX:CR is generally a more recent build thereof.
[22:58:12] <sickness> yeah but express undergoes a first "round" of testing, for what I know... correct me if I'm wrong
[22:58:20] <alanc> Nevada is the release after 10, that marketing won't let us call 11 yet
[22:58:23] <leal> ok, where can i download it?
[22:58:35] <sickness> http://www.opensolaris.org/os/downloads/
[22:58:45] <timeless> i think not calling it 11 until it's done is actually a good thing
[22:58:48] <alanc> SX & SX:CR are both builds of Nevada, SX is just time-delayed a bit to allow additional testing before release
[22:59:30] <alanc> SX ISO images should exactly match one of the sets of SX:CR ISO images from the previous month
[22:59:33] <leal> i'm looking in solaris express download center... and the iso image is: sol-nv-b48.
[22:59:46] *** ZeuSun has joined #opensolaris
[22:59:47] <leal> for me, nevada was the codename for solaris 11 (Express).
[22:59:52] *** nwf has quit IRC
[23:00:16] *** nwf has joined #opensolaris
[23:00:29] <alanc> right - b48's the latest one to go through the extended testing to be released as SX - b49 is only available as SX:CR so far
[23:00:39] <elektronkind> leal: don't worry. sun's weird culture of terminology confuses a lot of people.
[23:01:17] <alanc> Solaris Express is the program to release the current development builds - it started out with S10 builds before the S10 release, now does Nevada builds - someday we'll have released 11 and Solaris Express will be the builds on the next train leading to 112
[23:01:20] <alanc> er, 12
[23:01:22] <elektronkind> SX, Nevada, SX:CR, Sun ONE Applicataion Java Server, and so on.
[23:01:29] <ProfMikey> what does SX an SXCR stand for?
[23:01:32] <alanc> 112 will be after we're all retired 8-)
[23:01:44] <jengelh> guess: solaris express certified release
[23:01:49] <alanc> SX: Solaris Express, SXCR: Solaris Express Community Release
[23:01:52] <elektronkind> ProfMikey: Solaris Express, and Solaris Express: Community Release respectively
[23:01:53] <jengelh> whoops :)
[23:02:00] <ProfMikey> aww...cheers
[23:02:01] <sickness> well, maybe next "bump" will go straight for 100 :P
[23:02:08] <alanc> SXCR is the less tested one, so not certified
[23:02:09] <gisburn> leal: Maybe Solaris NGCECV/DVD (Next Generation Customer Evaluation Canidate Version DVD) will include it.
[23:02:14] <Gman> alanc: no retirement for you, get back to work you slacker! ;)
[23:02:24] <alanc> I've almost got Xorg 7.2 building!
[23:02:25] <sickness> like sol 2.x -> 5.x and java 1.5.x -> 5.x :P
[23:02:32] <sickness> sol 11 -> sol 100 :PPP
[23:02:43] <sickness> great marketing overlap over the competition >:)))
[23:02:46] <alanc> though I've gotten off track today doing code review for the Solaris DRI putback
[23:02:56] *** kupfer has quit IRC
[23:03:04] <elektronkind> a worthy thing to do code review time for
[23:03:12] * gisburn wonders whether leal will belive the NGCECV/DVD thing...
[23:03:15] <alanc> actually it was SunOS 5.x -> Solaris 2.x -> Solaris x
[23:03:31] <alanc> when we skipped from Solaris 2.6 straight to Solaris 7
[23:03:34] <gdamore> SunOS 5.x numbers still stand.
[23:04:03] <gisburn> alanc: When will be rename to SolNexTJavaOS happen ?
[23:04:16] <gdamore> in fact, i got rid of the use of "Solaris branding" version numbers in some code while at Sun, because of the heartache that the conversion from 2.6 to 7 caused.  i figured SunOS uname -r would never be subject to marketing insanity
[23:04:18] <sickness> well yeah, but then solaris 2.7 (7) and 2.8 (8) and 9 and 10 that way was not so clear =)
[23:04:18] <mrdeviant> gisburn, don't you mean SolNeXTJavaOS ? :P
[23:04:23] <alanc> that's a marketing secret - engineers like us aren't allowed to know in advance
[23:04:44] <sickness> alanc: lol :)
[23:05:30] <leal> alanc: Ok, so... :))
[23:05:45] <leal> alanc: SX:CR has brandZ?
[23:05:57] <leal> express does not?
[23:06:26] <alanc> you'ld have to compare the build number brandZ integrated in against the builds of each - I don't remember when it went in
[23:06:29] <gisburn> leal: sure, Solaris 10 has brandz - if you use it wth a Microsoft Vista2 license...
[23:06:42] <stevel> brandz went into 49, so yes - SXCR has brandz, and at the moment, SX does not
[23:06:46] <asyd> nice, I give a talk this monday about Solaris 10 in a rails event
[23:07:06] <elektronkind> can I have a brandZ zoned container?
[23:07:25] <alanc> what do you think the "Z" stands for?
[23:07:44] <Error_404> zettabyte?
[23:07:46] <gisburn> alanc: Zatan the ruler of the hell ?
[23:07:54] <alanc> you can't have BrandZ without zones...
[23:08:03] <elektronkind> I'm just committing a bit of socratic irony regarding terminology here...
[23:08:05] <gdamore> what marketing "genius" came up with the name BrandZ, anyway?
[23:08:08] <oxygene> "we had to move away from x, and well.. zones, zfs.. nice match"
[23:08:21] <leal> who is the gisburn... F...
[23:08:26] <gisburn> gdamore: don't ask
[23:08:28] <oxygene> it was called brandX for a while
[23:08:33] <alanc> the one who had to scramble when they found out "Brand X" was trademarked by someone else
[23:08:41] <gisburn> leal: sorry...
[23:08:50] <elektronkind> brand X... by Acme, inc
[23:08:55] <gdamore> heh.  and not Brand Y. :-)
[23:08:56] * gisburn is just quite happy and makes bad jokes...
[23:09:03] <alanc> so it's an abbreviation for Branded Zones or some such bit of retcon to make the name fit
[23:09:08] <elektronkind> makers of "Beer"-brand beer.
[23:09:24] <elektronkind> comes in a white can with black block lettering. just says "Beer".
[23:09:41] <alanc> if you want to annoy gisburn, just suggest "rm -rf usr/src/ksh93 ; ln -s /usr/bin/bash /sbin/sh"
[23:09:47] <alanc> 8-)
[23:09:57] <gisburn> alanc: grrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrxxx@@@@!!!
[23:09:58] <Error_404> elektronkind: if i actually saw that in the liquor store, i'd buy it
[23:09:59] <gisburn> damn you!!
[23:10:08] <asyd> S 12
[23:10:09] <asyd> oups
[23:10:56] * gisburn throws stones in alanc's direction...
[23:10:59] <alanc> fortunately, I don't have write access to the ON gate, so I can say that knowing I could never do that
[23:11:14] * gisburn throws more stones
[23:11:19] <elektronkind> Error_404: I thought that's how Kokanee beer was sold up there in BC ;)
[23:11:24] * gisburn throws a stunk
[23:11:33] <Error_404> elektronkind: no, kokanee comes in blue cans
[23:11:38] <Error_404> also, it's swill
[23:11:44] <alanc> a stunk?
[23:11:56] <gisburn> alanc: s/stunk/skunk/
[23:12:03] <schily> good evening!
[23:12:04] <elektronkind> Error_404: yeah, I likened it to Coors here in the US.
[23:12:11] <Error_404> that's a good analogy
[23:12:16] * gisburn growls
[23:12:38] <schily> to whom?
[23:12:58] <gdamore> Coors != Beer.
[23:13:05] <Error_404> mass market beer is gross no matter which country it comes from
[23:13:15] <AbeFroman> amen
[23:13:25] <gdamore> Coors == Piss In A Bottle
[23:13:32] <gisburn> schily: <alanc> if you want to annoy gisburn, just suggest "rm -rf usr/src/ksh93 ; ln -s /usr/bin/bash /sbin/sh"
[23:13:52] <gisburn> gdamore: erm
[23:14:16] <schily> I am not working on shells, I am just testing hsfs/mkisofs with working hardlink support
[23:14:17] <trygvis> how can I check which version a nfs mount is using?
[23:14:36] <gdamore> schily: any chance that cdrecord will ever be part of open solaris?
[23:14:50] *** leal has quit IRC
[23:14:52] <gisburn> gdamore: ah, Ok... I tought coors == person ... coors == beer == pi**s in a bottle is ok
[23:14:53] <schily> this only depends on Sun.....
[23:14:58] <gdamore> (we ship our own build of it to support xcdroast for our customers)
[23:15:08] <richlowe> cdrecord is in SFW
[23:15:10] <richlowe> SFW is open.
[23:15:20] <richlowe> I'd assume that thus the cdrecord integration with SFW also was.
[23:15:33] <schily> but mkisofs is installed in /usr/bin
[23:15:47] <stevel> cdrecord is installed in /usr/bin too
[23:16:03] <schily> I am interested to learn whether we will be able to ship the new mkisofs
[23:16:11] <richlowe> http://cvs.opensolaris.org/source/xref/sfw/usr/src/cmd/cdrtools/
[23:16:16] <gisburn> alanc: is there a reason why Xsun in B48 has no GLX extension ?
[23:16:16] <schily> in sync with the upcoming hsfs
[23:16:21] <richlowe> where something is installed, and what consolidation it lives in are unrelated.
[23:16:26] <schily> You need both to support working hard links
[23:16:30] <richlowe> currently bits of openssl are in /usr/sfw, openssl is in ON
[23:16:54] <alanc> gisburn: on sparc?   because stuart broke the code to list extensions loaded from separate modules 8-)
[23:17:20] <gdamore> oh, yes.  i now see that cdrecord is installed in /usr/bin on my b49 system.
[23:17:22] <gdamore> very cool.
[23:17:23] *** ZeuSun has quit IRC
[23:17:42] <gisburn> alanc: when is this going to be fixed ?
[23:17:53] <alanc> b50
[23:18:08] <gisburn> alanc: can I fix this somehow in a B48 installation ?
[23:18:39] *** ZeuSun has joined #opensolaris
[23:18:41] * gdamore wonders whether Xorg will replace Xsun on sparc before Solaris 11 ships
[23:18:41] <alanc> http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6465198
[23:18:50] * alanc does too
[23:19:06] <mrdeviant> i'm trying to do a Install with b50. when i reboot with the new kernel, i get an error that it can't open misc/forthdebug. then it dies, saying that it can't load tod module. Is there an easy fix for this?
[23:20:18] <alanc> gisburn: I don't know of any way short of installing the Nevada build 50 Xsun or reverting to a pre-build 45 Xsun
[23:20:20] * gdamore wonders if the solaris version of cdrecord includes DVD support.
[23:20:20] <gisburn> alanc: is there no hack/fix/whatever ?
[23:20:31] <gisburn> alanc: oh oh
[23:20:35] <schily> of course
[23:20:38] <gisburn> alanc I am dead meat.
[23:20:48] <gdamore> so no more ProDVD key required?
[23:20:49] * gisburn installed his GF's Blade1500 with B48.
[23:20:51] <alanc> why?
[23:20:54] <schily> the OSS cdrecord includes DVD support since May 15th
[23:20:55] *** kb7sqi has joined #opensolaris
[23:21:03] <gisburn> alanc: she needs OpenGL for her work.
[23:21:08] <ZeuSun> I tried to copy a large file (the belenix iso) from a disk on key to a local folder ; the iso is about 650 MB ; I got a message "io error". I did it twice; any idea what can it be ? I can view the contents of this iso  in the Disk On Key in linux
[23:21:09] <alanc> b50 ISO's should be up this week I think
[23:21:12] <gisburn> alanc: and I broke it.
[23:21:45] <gisburn> alanc: I am sure that I have to revert the installation to S10 or I am going to be roasted alive.
[23:21:48] <alanc> I know the feeling - my wife's upset I broke PHP on her Linux machine when I upgraded it from FC4 to FC5
[23:22:14] <gisburn> alanc: heh
[23:22:19] <gisburn> alanc: welcome to the clup.
[23:22:20] <gisburn> er
[23:22:22] <gisburn> club
[23:22:25] <gisburn> :-)
[23:22:50] <alanc> unfortunately, I only have done the legal paperwork to release binaries on opensolaris.org that come from the sources we've released, which doesn't include Xsun
[23:23:17] <alanc> (and I haven't even been bothering to do that since SX:CR covers them soon enough)
[23:23:39] <gisburn> alanc: It's Ok... I install Solaris 10 instead.
[23:23:48] <gisburn> alanc: I prefer peace.
[23:24:31] <alanc> that's what Live Upgrade is all about - never having to say I'm sorry for 2 hours while you re-install the older version - just "luactivate old_be ; init 6"
[23:25:13] <gisburn> alanc: I don't do any upgrades at all.
[23:25:29] <gisburn> alanc: I learned that "upgrades" are a way to generate more work.
[23:25:58] <gisburn> alanc: and I really do not need any artificial work right now - the normal way to get "more work" works quite efficient.
[23:27:10] <sickness> I really like lu* too :)))
[23:27:39] * gdamore wants to shoot the tadpole engineer(s) who wrote our tools in Tcl.
[23:27:53] <icon> haha
[23:28:09] <gisburn> gdamore: did you read your /msg stuff ?
[23:28:10] <icon> gdamore: ive been considering using it for the ports system
[23:28:14] <gdamore> of course, my wifitool in java is terribly slow
[23:28:30] <gdamore> gisburn: yes.
[23:28:46] <gdamore> icon: its Yet Another Dependency which makes it a PITA on build servers
[23:29:03] <icon> gdamore: agreed. python gives me an even more foul taste
[23:29:04] <gisburn> gdamore: but I never got a reply... you need to identify yourself against "nickserv" before you can send private messages via /msg
[23:29:27] <icon> gdamore: build systems get tough when you have to deal with multiple archs and have decent scripting support
[23:29:28] <Symmetria> hrm
[23:29:35] <icon> perl is well... perl
[23:29:40] <icon> which leaves python and tcl
[23:29:52] <gdamore> perl bites.  but at least its foulness is already present on anything S8 or newer.
[23:29:54] <gisburn> icon: you could use tksh
[23:29:57] <icon> java wouldnt perform well enough, so thats pretty well out too
[23:30:09] <icon> tksh?
[23:30:28] <slowhog_home> hi alanc
[23:30:28] <gisburn> icon: just google it.
[23:30:37] <gdamore> dtksh at one point.  ewww.  i'm probably gonna convert wifitool to gtk at some point
[23:30:50] <gdamore> it would be nice if someone would change dtksh into gtksh. :-)
[23:30:55] <icon> gisburn: already have. seems somewhere interesting
[23:31:00] <gisburn> icon: http://www.cs.princeton.edu/~jlk/tksh/
[23:31:11] <icon> already there ;)
[23:31:14] <icon> why tksh?
[23:31:21] <icon> and is it standard on sol8 and up?
[23:31:22] <gisburn> gdamore: feel free to employ me for that work...
[23:31:33] *** mrdeviant has quit IRC
[23:31:34] <alanc> good afternoon slowhog
[23:31:48] <gdamore> wtf?  Tcl implemented as ksh. weird
[23:31:54] <alanc> gdamore: that's what zenity is for
[23:32:04] <icon> meh, im a csh guy
[23:32:09] <icon> ksh gives me hives :D
[23:32:27] <gdamore> ksh is cool if you have to write real scripts.  but i try not to write real scripts. :-)
[23:32:44] *** jamesd has joined #opensolaris
[23:32:45] *** ChanServ sets mode: +o jamesd
[23:33:02] <icon> meh
[23:33:03] <gdamore> i wrote a very cool menu system using dtksh once.  it worked using tputs to emulate curses.  it was a jumpstart front end.
[23:33:27] <gdamore> scary the things you can do with dd and stty to change tty modes. :-)
[23:33:40] <icon> id honestly almost consider perl if it wasnt such a breeding ground for kruft
[23:34:11] <gdamore> it takes waay to much discipline to write maintainable perl code.  waay to much syntactic sugar.
[23:34:16] <icon> yup
[23:34:24] <Symmetria> hrm, anyone know what the solaris equivelant of the ar command is, or where I can get that command, since a bunch of stuff Im trying to build requires it
[23:34:37] <icon> i cant tell you how many times liberal use of $_ has made things extremely tough to deal with
[23:34:40] *** saltmiser has joined #opensolaris
[23:34:43] <gdamore>  /usr/ccs/bin/ar
[23:34:46] <icon> Symmetria: ar :)
[23:34:55] <gdamore>  or if you prefer, /usr/xpg4/bin/ar
[23:35:02] <Symmetria> thanks gdamore, so its just pathing then
[23:35:30] <gdamore> np
[23:35:33] <slowhog_home> alanc: do you have any suggestion for me to figure out why dtlogin stuck at the spalsh screen with 3 rectangles? :-)
[23:35:38] <icon> gods im actually considering tcl... i need professional help
[23:35:49] <slowhog_home> after typing username/password
[23:36:08] <alanc> no ideas coming to mind
[23:36:31] <icon> slowhog_home: have your X logs in /var/log?
[23:36:32] <slowhog_home> where is metacity coredump?
[23:36:35] <icon> or syslog output?
[23:36:39] <alanc> is it the GNOME splash screen after dtlogin has cleared the screen?
[23:36:47] <alanc> if so, try clicking the mouse on it
[23:36:54] <ZeuSun> I repeat my ( unanswered) question  with an update: I had copied successfully a large file , 650 MB, the belenix iso, from USB DiskOnKey to linux. Alas when trying to do the same on solaris, with the same DiskOnKey, I get an "I/O error "; any idea why ?
[23:36:59] <alanc> mine gets stuck there sometimes and a mouse click wakes it up
[23:37:16] <slowhog_home> I can only login with failsafe session, and start metacity causes segment fault
[23:37:31] <alanc> oh, metacity crashes, I'd ask Gman about those
[23:37:41] <Symmetria> libtool: link: `/usr/sfw/lib/gcc/i386-pc-solaris2.10/3.4.3/../../..//libsupc++.la' is not a valid libtool archive
[23:37:42] <Symmetria> hrm
[23:38:08] <slowhog_home> icon: there is a Xorg.log, syslog is empty
[23:38:13] *** hile_ has quit IRC
[23:38:41] <timeless> hrm, could someone here try mousing across the big "picture" on http://en.wikipedia.org/wiki/Image:KB_Dvorak_Right.svg
[23:38:59] <timeless> when i try mousing across it, my mouse seems to jump instead
[23:39:02] <slowhog_home> is there a font issue or linker issue ?
[23:39:15] <slowhog_home> I had been held the upgrade for those issues
[23:39:51] <slowhog_home> but seems like b35 to b50 still having problem
[23:40:00] <stevel> slowhog_home: there is a fontconfig issue
[23:40:21] <stevel> alanc pointed me at it the other day
[23:40:27] <stevel> i dunno if that would cause metacity to crash though
[23:41:31] <alanc> oh yeah, fontconfig crashes if you have any fonts with a font name (not file name) starting with any character other than A-Z or a-z
[23:41:41] <alanc> and all of GNOME uses fontconfig
[23:42:04] <slowhog_home> hmm, I only have default fonts
[23:42:19] <alanc> then that shouldn't be a problem
[23:43:14] <slowhog_home> time to ping Gman ;-)
[23:43:17] <trygvis> hm .. forcing both server and client to v3 seemed to do the trick
[23:43:23] <Symmetria> hrm I need libstdc++
[23:44:26] <alanc> you might try #jds on the Sun internal IRC too
[23:44:36] <alanc> haven't gotten all those guys out here yet
[23:47:30] *** dj2 has quit IRC
[23:47:53] <timeless> gisburn: do you have a file you use to setup a compiler env (for CC)?
[23:49:25] *** coffman has joined #opensolaris
[23:49:35] <gisburn> timeless: for OS/Net or which project ?
[23:49:48] * gisburn realises his tea is empty
[23:49:48] <timeless> i'm trying to build hello_static_world.c
[23:50:00] <gisburn> timeless: erm
[23:50:12] *** XVampireX has joined #opensolaris
[23:50:15] <gisburn> timeless: where is that source file from ?
[23:50:22] <timeless> i wrote it
[23:50:37] <gisburn> timeless: and the problem is... ?
[23:50:41] <timeless> it does something like printf("hello static world\n")
[23:50:51] <timeless> i want to link it statically which means i need to tell it where libc is
[23:51:05] <timeless> LD_LIBRARY_PATH=/opt/SUNWspro/lib /opt/SUNWspro/prod/bin/CC -Bstatic -g myfile.c
[23:51:07] <gisburn> timeless: libc or libC ?
[23:51:10] <timeless> ld: fatal: library -lm: not found
[23:51:10] <timeless> ld: fatal: library -lc: not found
[23:51:56] <Auralis> timeless: there is no static libc in solaris 10
[23:52:01] *** ZeuSun has quit IRC
[23:52:19] <timeless> Auralis: that could be a wee bit of a problem
[23:52:23] <gisburn> timeless: solaris >= 10 no longer has a static libc or libm
[23:52:24] <timeless> i'm trying to test chroot
[23:52:48] <timeless> and that kinda requires an app that isn't bound to an env i don't have
[23:53:22] <Symmetria> heh Im having problems with libstdc++
[23:53:32] <Symmetria> gonna install libg++ which may fix it
[23:54:19] *** XVampireX has quit IRC
[23:54:30] <timeless> Auralis: and i can't get someone else's? :(
[23:55:29] * icon &
[23:55:59] <timeless> kill -SIGUSR icon
[23:56:30] <gdamore> hmmm.... what happened to the /usr/j2se link in OpenSolaris?  Solaris 10 used to have it....
[23:56:46] <jamesd> timeless, screw  chroot... use a zone.
[23:56:56] <timeless> jamesd: i'm in a zone
[23:56:58] <alanc> Java 2 became obsolete - it's all about Java 6 now
[23:57:00] <timeless> and in there, i'm in a chroot
[23:57:10] <timeless> and that chroot isn't working
[23:57:15] <timeless> and i need to know whose fault it is
[23:57:22] <timeless> the zone, the chroot, the broken root
[23:57:34] <gdamore> heh.  well, on Solaris 8 /usr/java didn't work and i had to use /usr/j2se.  Now on Solaris 10 I have to go back to /usr/java.  arg.
[23:57:47] <alanc> though my build 50 machine does still have a /usr/j2se as well as /usr/java
[23:58:02] <jamesd> timeless, seems like over kill
[23:58:31] <jamesd> you can make  /etc and tother directories... read/only  if you are paranoid
[23:58:34] <timeless> jamesd: given that all i need is a working apt? yes
[23:58:35] <alanc> /usr/j2se is 1.4.2, /usr/java is 5.0
[23:58:52] <timeless> but i can't figure out how to get a working apt
[23:59:01] <timeless> and i can't figure out how to get a working debian brandz
[23:59:34] <timeless> but i can get centos, and i've installed scratchbox w/ debian devkit on normal linuxes
[23:59:50] <gdamore> alanc: what a PITA.  java versioning changes.... perl version changes... openssl changes... what's next?





top