[00:00:04] *** derchris has quit IRC
[00:00:06] *** Marv|Home has quit IRC
[00:00:11] *** derchris has joined #opensolaris
[00:00:21] <CIA-17> mlf: PSARC/2006/501 Nvidia ck804/mcp55 SATA HBA driver, PSARC/2007/100 Device Id Extensions, PSARC/2007/172 Reserved Devnames, 6296435 native sata driver needed for nVIDIA mcp04 and mcp55 controllers, 6418034 DMA resources need to be released for scsi commands emulated in sata framework, 6487018 driver binding for sata storage devices should be more flexible, 6487838 common definitions can be defined in sata_defs.h to be shared with other sata HBA's, 65
[00:00:33] <gnut> seanmcg: how do you pkgchk all of them?
[00:00:40] <jmcp> pkgchk -nv
[00:00:52] <jmcp> well, without specifying a pkgname it'll check every package on the system
[00:01:10] <jmcp> you probably want to run it like this       pkgchk -nv >> /tmp/pkgchk.out 2>&1
[00:01:11] <gnut> yeah. i just did that
[00:01:14] <jmcp> and go for a coffee  :)
[00:01:16] <gnut> i get lots of errors on even my working machine
[00:01:32] <boyd> And then you have to deal with the driver.conf files that aren't marked editable :(
[00:01:37] <gnut> does it have to come out error-free? or does it mark also conf files?
[00:01:38] <boyd> (Morning, all)
[00:01:56] *** cmihai has quit IRC
[00:02:10] <boyd> With the -n it won't complain about files that the pkg creator said were editable
[00:02:39] <gnut> i know this is a stupid question, but sometimes, when i pkgadd, say ... pkgadd -d ./SUNWhea (i just copied over the directory)... it says no packages found
[00:03:07] <boyd> I suspect you mean pkgadd -d . SUNhea
[00:03:35] <boyd> After -d you name the directory that the package is in. The package is the dir called SUNWhea
[00:03:56] <boyd> (in other words you want to point at the *parent* of the SUNWhea directory
[00:03:58] <gnut> boyd: no... i didn't mean that... thanks for pointing that out :)
[00:04:07] <boyd> Oh well.
[00:04:22] <gnut> but it works now. thanks
[00:04:41] <gnut> sometimes, the package name is given?
[00:04:56] <gnut> because there are packages where i just specify pkgadd -d ./dir and it works
[00:04:58] <jmcp> yeah, eg   pkgadd -d thisisapkg.pkg
[00:04:59] <kjetilho> I suggest you read what boyd wrote once more
[00:05:19] <gnut> got it
[00:05:22] <boyd> after -d you put what I call the "container" (parent dir) then last you can specify the pkg name or you get a menu.
[00:05:46] *** Murmuria has left #opensolaris
[00:06:20] <boyd> Naturally (in order to maximise confusion) there are two pkg formats. the one where the pkg is a dir ("Filesystem format") and the one where the pkg comes as a single file ("Datastream Format")
[00:06:32] <boyd> in both cases you point -d at the "container".
[00:06:45] <boyd> The container is the parent dir for the Filesystem format
[00:06:48] <gnut> and sometimes, the package name doesn't need to be specified... , right?
[00:06:54] <boyd> and the file itself for the datastream format
[00:07:01] *** JWheeler has quit IRC
[00:07:20] <boyd> If you dont' spec the pkg name it will give you a menu to choose packages from that "container"
[00:08:25] <boyd> (really the Datastream format is a cpio archive of the Filesystem format with an extra header on the front)
[00:08:51] *** mega has joined #opensolaris
[00:09:02] <boyd> gnut: or you can use "all" as the pkg name and it will do what you might expect
[00:09:34] <kjetilho> boyd: you're a veritable cornucopia of information :-)
[00:09:44] <boyd> I'm certainly full of something :)
[00:09:50] <gnut> heheh
[00:09:55] <gnut> thanks for this information
[00:10:06] <gnut> ndis compiles!!!
[00:10:11] <gnut> that was a good find
[00:11:26] * boyd waits for SXDE 70. Not really sure why :)
[00:13:31] *** jfndi has quit IRC
[00:14:01] *** bengtf__ has joined #opensolaris
[00:14:03] *** bengtf__ is now known as bengtf
[00:14:23] <jbk> evening
[00:20:06] <elektronkind> OMG
[00:20:18] <elektronkind> Is it true? AM I dreaming?
[00:20:21] <elektronkind> PSARC/2006/501 Nvidia ck804/mcp55 SATA HBA driver
[00:20:25] <elektronkind> was just put back
[00:20:26] <wesolows> yep
[00:21:13] <elektronkind> and open source, too
[00:21:18] <elektronkind> yippee
[00:21:27] <alanc> there's even a headsup for it at http://www.opensolaris.org/os/community/on/flag-days/pages/2007081601/
[00:22:57] *** BatonT has quit IRC
[00:26:47] *** nachox has joined #opensolaris
[00:27:10] *** hohum has quit IRC
[00:27:19] <nachox> evening
[00:27:51] <jbk> hello
[00:30:32] *** Teltariat has quit IRC
[00:35:27] *** gnut has quit IRC
[00:38:22] *** mega has quit IRC
[00:39:50] *** alphakiller|away is now known as alphakiller
[00:40:41] *** Pietro_S has left #opensolaris
[00:41:03] *** baijiutong has quit IRC
[00:42:51] *** freakazoid0223 has quit IRC
[00:45:46] *** freakazoid0223 has joined #opensolaris
[00:48:59] *** jafari_ has joined #opensolaris
[00:50:09] *** nrubsig has joined #opensolaris
[00:50:09] *** ChanServ sets mode: +o nrubsig
[00:50:18] * nrubsig looks at Tpenta
[00:50:24] <nrubsig> Tpenta: good news for you!
[00:50:43] <nachox> shh! do not disturb, he is sleeping :P
[00:50:58] <nrubsig> nachox: only loosers need sleep.
[00:51:12] <nachox> no, living people need sleep
[00:51:39] *** yippi has joined #opensolaris
[00:52:23] <nrubsig> Is there a channel for SMF ?
[00:53:40] <richlowe> Not out here, that I know of.
[00:53:43] <richlowe> there maybe one internally, I guess.
[00:53:57] <richlowe> nrubsig: what's the problem?
[00:54:58] *** cypromis has joined #opensolaris
[00:55:19] *** yippi has left #opensolaris
[00:56:37] <nrubsig> richlowe: Tpenta gets a promotion.
[00:56:57] <boyd> ?
[00:57:00] * wesolows is working in SMF right now actually.  Got a question?  I may be able to help
[00:57:12] * nrubsig giggles madly.
[00:57:22] *** yippi has joined #opensolaris
[00:57:31] <jmcp> nrubsig: wtf are you talking about?
[00:57:34] <nrubsig> wesolows: well, SMF events framework...
[00:57:39] <nrubsig> jmcp: secret
[00:57:41] <networkdump> I have one
[00:57:50] *** jafari_ has quit IRC
[00:57:53] <nrubsig> networkdump: ?!
[00:58:00] <networkdump> question
[00:58:03] <wesolows> You mean contracts?  That's not an area I'm very familiar with
[00:58:22] <networkdump> I have compiled a program under /usr/local
[00:58:31] <wesolows> You probably want to ask on smf-discuss
[00:58:38] <richlowe> Asking for help, and then not telling us what you want to know is unlikely to be particularly sucessful.
[00:58:41] <boyd> jmcp: I think he's in a good mod again. It's hard to distinguish from being on a controlled substance
[00:58:46] <richlowe> (to nrubsig not networkdump)
[00:58:49] <Tpenta> nrubsig: ??
[00:58:54] *** hile_ has quit IRC
[00:58:55] *** jafari has joined #opensolaris
[00:58:59] <jmcp> boyd: yeah, that's what I'm thinking too
[00:59:04] <nrubsig> Tpenta: you're promoted.
[00:59:06] <networkdump> and I have to export LD_LIBRARY_PATH in order to correctly work
[00:59:07] <richlowe> boyd: I think what Roland is on is too obscure to even be controlled.
[00:59:11] * nrubsig giggles
[00:59:19] <jmcp> clearly
[00:59:20] <richlowe> boyd: either that, or it's a combination of powders found in every kitchen cabinet.
[00:59:21] <Tpenta> to what? orange carrier?
[00:59:27] <jmcp> coolie?
[00:59:28] <jmcp> :)
[00:59:29] <nrubsig> nah
[00:59:32] <nrubsig> Tpenta: ---> /msg
[00:59:35] <jbk> networkdump: then you probably forgot to add -R to the link cmdline
[00:59:38] <networkdump> is there a way quick way to do it without writing a shell script?
[00:59:57] <networkdump> jbk: autoconf missed probably?
[01:00:02] <networkdump> missed it*
[01:00:03] <jmcp> networkdump: re-run your build, and add -R/usr/local/lib to your CFLAGS
[01:00:04] <wesolows> networkdump: You can do it globally with crle, but that's not really the right answer - as jbk noted.
[01:00:11] <jmcp> assuming that you need /usr/local/lib, that is
[01:00:20] *** yippi has quit IRC
[01:00:20] <networkdump> no, /usr/sfw
[01:00:36] <jmcp> same concept
[01:01:02] <wesolows> {cc,ld} -o some_binary ... -R /usr/sfw/lib -lsome_library ...
[01:01:06] <jbk> just do LDFLAGS=-R/usr/sfw/lib ./configure
[01:01:27] <nachox> just use a shell script, what's wrong with that?
[01:01:36] *** sarah has quit IRC
[01:01:41] <wesolows> nachox: shell scripts that set LD_LIBRARY_PATH are broken.
[01:01:42] <pitty> which language do you guys think is better to learn c++ or c ?
[01:01:54] <jmcp> c
[01:02:01] <wesolows> nachox: We hate that.  LD_LIBRARY_PATH is for the user to control, not for scripts to work around broken binaries.
[01:02:01] <jmcp> c++ is evil and should be destroyed
[01:02:08] <pitty> really
[01:02:11] <jmcp> pitty: yes
[01:02:12] <pitty> why so ?
[01:02:22] <jmcp> pitty: I think wesolows can explain it better than I can
[01:02:25] <jbk> well C allows you to shoot your foo off, c++ allows you to shoot off both legs, all while you think it's unarmed :)
[01:02:25] <bubbva> C++ is not so bad.  It's tricky going between the two.
[01:02:33] <nachox> wesolows, how so? it's much better than setting in your your profile if you cant or wont recompile the program, i am too against using it unless i am developing
[01:03:21] *** Jondice has joined #opensolaris
[01:03:32] <nachox> but if i have no other option i much rather have it in a script than the other much worse options
[01:03:33] <richlowe> if you're developing, you'd never need it.
[01:03:36] *** yippi has joined #opensolaris
[01:03:37] <richlowe> because you'd use -R.
[01:03:59] <jbk> hmm fry's has what supposedly is a wireless card that uses the atheros chipset for $18
[01:04:08] <kjetilho> LD_LIBRARY_PATH is evil, but LD_RUN_PATH is a convenient alternative to -R
[01:04:28] <networkdump> good it worked, thank you guys
[01:04:30] <kjetilho> (since syntax for -R varies between OS :-/)
[01:04:31] <nachox> actually you might when you need to test your program using a different set of libraries
[01:05:01] <jbk> and LD_* isn't universal either :)
[01:06:18] <kjetilho> right, but it's *more* universal :)
[01:06:24] <kjetilho> IME at least
[01:13:34] *** pitty has left #opensolaris
[01:15:40] *** alphakiller has left #opensolaris
[01:16:29] *** sstallion has quit IRC
[01:17:28] *** sstallion has joined #opensolaris
[01:18:27] * brendang is back
[01:18:47] <jmcp> hi brendang
[01:18:59] <brendang> jmcp: G'Day
[01:20:10] *** yippi has quit IRC
[01:27:51] *** rmorse has joined #opensolaris
[01:27:56] *** sparc-kly has joined #opensolaris
[01:27:57] *** ChanServ sets mode: +o sparc-kly
[01:29:22] *** slowhog has left #opensolaris
[01:34:49] *** jpdrawnee1 has quit IRC
[01:35:15] *** comay has quit IRC
[01:37:04] <Gman> richlowe, must...have...more...bug states!
[01:37:40] * Gman wants a 'this bug is being ignored' state :)
[01:37:45] <richlowe> Gman: 4-Defer
[01:37:50] * nrubsig injects more "bug states" into Gman's bloodstream ?
[01:38:00] <nrubsig> Gman: feel better ?
[01:38:00] <richlowe> or Dispatched/Accepted, and untouched for several years.
[01:38:04] <richlowe> which is effectively an implicit Defer.
[01:38:09] <Gman> :)
[01:38:23] <nrubsig> richlowe: "lost interest" would be nice, too
[01:38:39] <kjetilho> Gman: how about Priority: none-whatsoever?
[01:39:04] <Gman> heh
[01:39:12] 
[01:39:31] <richlowe> 12-Use-Star
[01:39:32] <richlowe> ;)
[01:39:43] <Gman> hawhaw
[01:39:44] <nrubsig> richlowe: 13-Bad-Day
[01:39:53] <nrubsig> richlowe: 14-Reporter-Sucks
[01:40:12] <nrubsig> richlowe: 666-!*>Satan-Rocks<*!
[01:40:38] <nrubsig> richlowe: Item 13 should only be accessible Friday afternoon
[01:40:54] <alanc> CLOSED - NOT_NEAR_INDIANA
[01:41:22] <nrubsig> alanc: CLOSED - customer dead, noone cares anymore.
[01:41:29] <kjetilho> "Indiana - the Bug State!"
[01:41:35] <richlowe> hah.
[01:42:00] *** hohum has joined #OpenSolaris
[01:42:01] <nachox> haha
[01:42:13] <alanc> "Nevada - the state where the bugs are really scary after all that nuclear bomb testing"
[01:42:22] <nrubsig> heh
[01:42:25] <nrubsig> alanc: +1
[01:43:01] <nrubsig> Nevada - the state where the bugs will hunt you at night.
[01:43:47] <nachox> "nevada - the state with the alien bugs"
[01:44:14] * alanc is a second-generation Nevadan - my mom used to see the glow of the nuclear tests over the hills before they went underground
[01:45:06] <nrubsig> alanc: is that the reason for your giant six-headed ... ?
[01:45:12] *** Gman_ has joined #opensolaris
[01:45:26] <kjetilho> alanc: that must be pretty scary
[01:45:53] *** Gman has quit IRC
[01:45:56] <nrubsig> kjetilho: alanc is scary. sometimes. If he wears pink shirts and a mikey-mouse mask.
[01:46:10] <Gman_> sodding battery
[01:46:17] <nrubsig> (and screams for goofy)
[01:46:23] * nrubsig hides from alanc's wrath...
[01:46:36] * nrubsig crawls away...
[01:47:14] <nrubsig> "sodding" or "sobbing" (the poor battery is lonely... and discharged... overworked... and no hope for a replacement...) ?
[01:47:27] <alanc> sorry, fresh out of wrath today - I can provide spite instead if you like?
[01:47:40] <nrubsig> sure
[01:47:41] *** rmorse has quit IRC
[01:47:47] <jmcp> how about bile?
[01:47:48] <Gman_> hrm, what good is youtube now if you can't upload anything longer than 10mins :(
[01:47:58] <nrubsig> what is "spite" ?
[01:48:03] <richlowe> jmcp: I have a decent amount of all 3, if that'd work.
[01:48:31] <alanc> the tar thread on PSARC has overused everyone's bile quota for the week, we have to wait until next week to get more
[01:48:36] <jmcp> richlowe: topped off with some vitriol?
[01:48:53] *** oninoshiko has joined #opensolaris
[01:49:11] <dlg> how do i uninstall sun studio 12?
[01:49:12] 
[01:49:23] <jmcp> dlg: there should be an uninstaller class somewhere under /var/sarm
[01:49:34] <nrubsig> jmcp: /var/sadm/
[01:49:36] <dlg> sarm?
[01:49:43] <alanc> the stomach is the fastest way to get to bile - just reach a little left to the gall bladder
[01:49:44] <wesolows> Actually it was blunt but polite, the sort of message one expects to receive after abusing a lot of people for a long time.
[01:49:45] <kjetilho> Gman_: use Google Video instead
[01:49:46] <nrubsig> dlg: typo
[01:49:47] <jmcp> dlg: when you're slurring your words ...
[01:49:48] <dlg> oh wait, its sun software
[01:49:50] <dlg> they have doco
[01:49:53] <Gman_> kjetilho, isn't it the same?
[01:50:19] * Gman_ checks
[01:50:23] <kjetilho> no, Google Video has a bit more "serious" profile, it has lots of lectures and stuff like that which obviously are longer than 10 minutes
[01:50:24] * dlg get a coffee before he hurts himself
[01:50:24] <alanc> I thought google was shutting down google video
[01:50:32] <Triskelios> is it bad to remove Studio with pkgrm instead of whatever other app management thing it's registered in?
[01:50:40] <alanc> or was it just video sales they killed there?
[01:50:43] <richlowe> alanc: makes you wonder what they're going to do with their talks stuff on there.
[01:50:44] <kjetilho> alanc: only the selling of videos
[01:50:46] <Gman_> kjetilho, cool, thanks for that
[01:51:13] <nrubsig> wesolows: yes, it's an emai which I can't write... I would explode long before that point.
[01:51:23] <nrubsig> s/can't/couldn't/
[01:51:33] *** jlc has joined #opensolaris
[01:51:46] <nrubsig> maybe in 30 years I learned enougth about people to handle such things better
[01:52:09] <nachox> dlg, use prodreg
[01:52:15] <nrubsig> (but the first place for a polite kick still goes to Stuart Kreitman IMO)
[01:52:32] <kjetilho> Triskelios: is should be ok
[01:53:28] <oninoshiko> the video's from the scalability conference were pretty good up there
[01:53:32] <boyd> The talks on Google video are really interesting. Not enough hours in the day :(
[01:55:27] *** NikolaVeber has quit IRC
[01:55:30] <nrubsig> yippiiiieeeee!!!
[01:55:48] <nrubsig> alanc: remeber the <esc><ctrl-l> bug in ksh93 ?
[01:57:17] * nrubsig fear alanc may be sulking and hides...
[01:57:22] <nrubsig> s/fear/fears/
[01:57:45] * alanc is answering e-mail and getting ready to leave
[01:57:52] <alanc> but yes, I remember the bug
[01:58:17] <nrubsig> alanc: fixed
[01:58:21] <alanc> cool
[01:58:32] <nrubsig> alanc: it was the memory corruption issue in the signal handler
[01:58:41] <nrubsig> alanc: and set -o multiline suddenly works, too.
[01:58:51] <alanc> that sounds like a nasty bug
[01:58:54] <nrubsig> it was all the xx@@@!!!-SIGCHLD handler
[01:58:58] <nrubsig> yeah
[01:59:07] <nrubsig> and April found it :-)
[01:59:53] <nachox> tar's psarc case is open?
[02:00:20] <CIA-17> mh27603: 6587576 cpu_info kstat is returning garbage in supported_frequencies_Hz
[02:00:21] <nachox> Triskelios, ops, it was for you, use prodreg
[02:00:29] <sommerfeld> nachox: no.  but people are still chattering
[02:00:33] <PerterB> nachox: don't mention the war
[02:00:58] <jmcp> nachox: psarc-ext, look for the most recent thread
[02:01:07] * nachox buries his head in the ground
[02:01:24] <jmcp> and it's not tar, but CIFS system attributes support for cpio(1), pax(1), tar(1) [ PSARC/2007/459]
[02:01:37] <richlowe> and you almost certainly don't want to read it.
[02:02:25] <PerterB> having read it, I can vouch for that
[02:02:38] <nachox> hehe, hmm, that reminds me of something i needed to do... i think frkit was updated
[02:02:39] <nachox> brb
[02:03:55] *** Darth_Wa1er has left #opensolaris
[02:05:12] *** nachox has quit IRC
[02:10:16] *** cypromis has quit IRC
[02:10:17] <boyd> The word "unhinged" comes to mind...
[02:11:03] <jmcp> boyd: more and more as time goes on
[02:11:20] *** alanc is now known as alanc-away
[02:11:31] <boyd> yes
[02:12:51] <Triskelios> frkit was updated?
[02:15:00] *** nachox has joined #opensolaris
[02:15:48] <nachox> no luck, i guess frkit now requires nevada and does not work in solaris 10 :)
[02:17:14] *** bubbva has quit IRC
[02:18:41] <jmcp> nachox: yeah
[02:19:01] <nachox> what's next? -% ?
[02:19:34] <jmcp> probablyu
[02:19:39] <jmcp> or an aneurism for Joerg
[02:22:16] <nachox> hehe
[02:22:33] * Triskelios wonders if it makes sense to install acpidrv when the battery driver is also installed
[02:22:37] <nrubsig> there are still 2^24 unicode symbols which could be used
[02:24:20] <nachox> you probably want to use symbols easily available in any keyboard
[02:24:37] *** vmlemon has quit IRC
[02:24:57] <nrubsig> nachox: why ? It's more fun to torture people.
[02:25:12] <libkeiser> they really need to reserve part of the unicode namespace for representing profane gestures
[02:25:31] <nrubsig> libkeiser: already done. porn wingdings
[02:25:38] <oninoshiko> OpenSolaris: the OS for the masochist in all of us
[02:26:41] <wesolows> heh, while I'm sure it can cater to masochists, I actually think the real masochists are the people who are developing on platforms that don't offer the kind of developer tools and documentation that's available for/in OpenSolaris distributions
[02:26:59] <libkeiser> wesolows: hear hear
[02:27:11] <wesolows> I can't imagine developing on GNU/Linux, or Windows, or AIX.  Pain!
[02:27:22] <nrubsig> wesolows: is that sarcasm ?
[02:27:35] <wesolows> no, not at all.  I'm 100% sincere.
[02:27:36] <libkeiser> in comparison to the other two, i find developing for aix relatively easy
[02:27:37] <nachox> er, aix has goodish docs... or did last time i used it
[02:27:46] <wesolows> I want my, I want my, I want my mdb
[02:28:08] <jmcp> hehe
[02:28:09] <wesolows> sure, it's all relative.
[02:28:10] <nrubsig> wesolows: I don't have problems with Linux development. Exept lots of tiny cratches and itches.
[02:28:18] <libkeiser> the aix docs are good, until you encounter the kernel debugger, then run in fear...
[02:28:24] <nrubsig> heh
[02:28:39] <nachox> hehe
[02:28:40] <nrubsig> libkeiser: isn't that mdb, too ?
[02:28:59] <richlowe> mdb isn't painful, just different.
[02:29:09] <wesolows> no DTrace.  no mdb.  primitive introspection facilities.  An inferior linker.
[02:29:18] <wesolows> libumem?  come on
[02:29:48] <libkeiser> aix kdb is, um, poorly documented to say the least
[02:30:00] <nachox> damn -% is actually used somewhere!
[02:30:35] <libkeiser> libumem? ha. they don't even have a general purpose object allocator in the aix _kernel_
[02:30:44] <nachox> i'd say you cant use -*
[02:30:45] <wesolows> libkeiser: yeah, hence my "come on" comment
[02:30:56] <nrubsig> nachox: yes, you can
[02:31:10] <nrubsig> nachox: $ foo '-*' ... #
[02:31:32] <nachox> nrubsig, ohh, you need to escape it or use ''
[02:31:39] <nachox> painfull
[02:31:40] <nrubsig> yes
[02:31:44] <nrubsig> yes
[02:31:51] <nrubsig> same may appliy to -%
[02:32:13] <nrubsig> whoover had that idea shall be roasted on a small fire.
[02:32:14] <nachox> i dont think % is used for anything in the shell, is it?
[02:32:22] <kjetilho> I suggest -^C :)
[02:32:33] <kjetilho> much more conveient
[02:32:39] <nrubsig> nachox: depends on the shell
[02:33:05] <nachox> C shell might, i'm not sure i'm not a cshell user, i use mainly ksh and bash
[02:33:16] <nachox> well, and sh
[02:33:39] * jmcp gets to Don Cragun's email
[02:33:40] <jmcp> ouch!
[02:34:18] <wesolows> job identifiers?  prompt expansion?  pattern-matching word expansion?  bash sure uses %
[02:34:30] <wesolows> jmcp: I thought it was polite.
[02:34:45] <jmcp> his sig lines ... leaves me, at least, in no doubt
[02:34:48] <jmcp> it was polite
[02:35:55] *** stevel has quit IRC
[02:40:43] <nachox> hehe, so much mail and all for a simple option :P
[02:41:27] 
[02:41:44] *** _Megaf is now known as Megaf|Busy
[02:41:45] <nachox> i did propose -$
[02:42:06] <nachox> things might get nasty when someone tries something like -$la :)
[02:42:34] <jmcp> that's getting a tad sillu
[02:42:36] <jmcp> silly
[02:43:15] <jbk> i hate wireless cards
[02:43:28] <nachox> hehe, give us a break, we're having a little fun
[02:43:46] <nachox> that does not even show here
[02:44:30] *** tsp has joined #opensolaris
[02:45:22] <boyd> What about morse options: -...---.--..---.---....
[02:45:31] <axisys> does NFS works accross two separate class A network ? is it possible to have a netapp server on one class A and the host on a different class A ?
[02:45:32] <boyd> This is infinitely expandable
[02:45:35] * jmcp sods off in a huff
[02:45:47] <nrubsig> axisys: yes
[02:45:48] <boyd> Aw, sorry jmcp
[02:45:51] <tsp> I understand how I can use usermod to add privileges to a user, but how do I list what privs the user has?
[02:46:07] <jmcp> su - $user -c ' ppriv -l'    I think
[02:46:08] <boyd> grep username /etc/user_attr
[02:46:11] <nrubsig> tsp: $ cat </dev/kmem
[02:46:28] * nrubsig hides from tsp
[02:46:29] <boyd> actually, jmcp's answer is better
[02:46:33] <jmcp> but wrong
[02:46:38] <nrubsig> boyd: my answer is generic
[02:46:57] <jmcp> tsp:   su - $user -c ' ppriv $$'
[02:47:02] <jmcp> nrubsig: your answer is crap
[02:47:05] <wesolows> your answer is actually not correct, not is it useful
[02:47:18] <jmcp> nrubsig: I might as well email you genunix, too
[02:47:21] <wesolows> at best it will lock up your terminal
[02:47:43] <tsp> all I'm trying to do is grant me and my friends access to my audio device
[02:47:58] <nrubsig> tsp: what about using ACLs ?
[02:48:22] <nrubsig> (not joking)
[02:48:35] <tsp> nrubsig: I was going to use chmod on it and chmod it 0666, but before I do that, I want a way to reset it to defaults without having to chmod it back - since something sets it in the first place
[02:49:05] <nrubsig> tsp: try to use $ setfacl ... # and add users to the ACL list... it's independent (more or less) from what chmod mode does
[02:49:19] <tsp> ah, so I won't have to tweak logindevperms
[02:49:28] * boyd still hasn't gotten to Don's famous email
[02:49:51] <nachox> i've only seen one of his mails but i was using jive which sucks
[02:50:40] *** cypromis has joined #opensolaris
[02:51:26] <wesolows> poor jive
[02:51:45] <wesolows> it's trying to do a very hard job, failing pretty miserably, and attracting so much hatred
[02:51:48] <axisys> nrubsig: so mount -F nfs host-on-classA:/home /mnt where /mnt is on host-on-classB  should work?
[02:51:51] <nrubsig> wesolows: seriusly: Jive sucks.
[02:51:54] <jbk> anyone have any experience with porting network drivers from open bsd?
[02:52:16] <nrubsig> axisys: yes, assuming no firewall blocks the RPC traffic.
[02:52:27] <boyd> wesolows: Which hard job? Being a web forum that is as usable as 20 year old mail clients? :)
[02:52:28] *** slowhog has joined #opensolaris
[02:52:34] <nrubsig> axisys: maybe it helps to use NFSv4 only
[02:53:14] <nrubsig> axisys: otherwise try "snoop" and check for problems
[02:53:17] <wesolows> boyd: The web forum <=> mail bridge
[02:53:18] <axisys> nrubsig: is there way to check if RPC is blocking?
[02:53:25] <axisys> nrubsig: u just answered
[02:53:27] <nrubsig> axisys: erm
[02:53:32] <boyd> axisys: There is nothing at all about NFS that cares what the IP addresses are. There *may* be routing problems or firewall issues.
[02:53:41] <nrubsig> axisys: I suggest to ask in networking-discuss@
[02:53:45] <wesolows> boyd: and that, yes.  The web is not really a good platform for mail.  They should really make an AJAX mutt and have done with it. :-)
[02:53:58] <boyd> wesolows: I vote +1 for that.
[02:53:59] <nachox> hehe, seems that cannonical was hacked...
[02:54:04] <axisys> boyd, nrubsig let me do the snoop
[02:54:18] <boyd> wesolows: Actually, gmail isn't that far off.
[02:54:19] <boyd> nachox: ?
[02:54:54] *** simford has joined #opensolaris
[02:55:02] <wesolows> I'm not really a fan.  It's comparable to thunderbird, which I hate too.
[02:55:09] <wesolows> graphical mail clients just feel clunky to me.
[02:55:10] <boyd> Heh...
[02:56:22] <nachox> hmm,  i was happy enough with kmail when i used it
[02:56:34] <boyd> What makes Joerg think that just ignoring all format changes to tar in the last 3 years is going to happen?
[02:56:52] <jmcp> boyd: megalomania
[02:57:07] <boyd> Maybe.. there's some paranoia in there too
[02:57:14] <jmcp> of course
[02:57:21] <nachox> star love?
[02:57:24] <wesolows> Joerg has some interesting pathologies.  His heart's in the right place, I think, but he's too obsessed with not just getting the right solution but having it be his solution.
[02:57:53] <wesolows> He also seems to work on some fairly hard stuff, making it more likely that there will be conflict.
[02:58:12] <boyd> I agree.. I said yesterday that I often see his point (and even agree sometimes) but cringe when I see him present it.
[02:58:28] <nachox> i think he lacks pr skills
[02:59:09] <wesolows> That too.  It's really a confluence of several problems.  I can definitely see why he's frustrated, but I also think he's sometimes not diligent or open-minded enough to grasp the big picture.
[02:59:37] <boyd> Yes, the big picture is an issue.
[03:00:17] <CIA-17> blakej: 4875742 PAGE_SE_MUTEX() macro needs maintenance, 6517224 pse_mutex show scaling issues with tpc-h
[03:00:19] <wesolows> I've been frustrated (and undoubtedly frustrated others as well) with PSARC in the past, but you have to work through the issues and actually try to reach some generally acceptable solution.
[03:00:36] <nrubsig> wesolows: well, calling Don a "liar" was iMO the worst imaginable thing which he could've done so far... ;-(
[03:00:42] <wesolows> Joerg sometimes seems too unwilling to be flexible.
[03:00:59] <boyd> I concur.. I've been (mostly) happy with the way PSARC has dealy with some issues... Sparc New Boot seems like one.
[03:01:11] <nachox> i think another part of the problem is that he is not a native english speaker i sometimes think he has problems translating his ideas
[03:01:16] <wesolows> nrubsig: Sure, bringing out ad hominem attacks is really the worst aspect of any of this.  It's hardly without precedent but it's not that common either.
[03:01:21] <richlowe> nachox: I suggest not saying that to him.
[03:01:29] <richlowe> that was a bigger argument than the argument, last time it happened.
[03:02:19] <boyd> IIRC nrubsig once told me that he's not much less abrasive in German
[03:02:33] <wesolows> If there is something we can do to help him feel less frustrated or avoid pissing off others, let us know.  But I don't really see what can be done here.  People are who they are and they're going to conduct themselves in whatever ways they prefer.
[03:02:35] <kjetilho> Germans are more abrasive
[03:02:40] <richlowe> well, German is a great language to be abrasive in, from what I can tell.
[03:02:43] <richlowe> all those wonderful compounds.
[03:02:59] <kjetilho> Norwegians are, too (we prefer to say "direct", though)
[03:03:00] <nachox> richlowe, why? i'm being honest, he is a great programmer, with great ideas, and there is nothing wrong with not being able to translate your ideas as accurately
[03:03:02] *** tsp has quit IRC
[03:03:59] <kjetilho> nachox: it will never be a fruitful discussion, it's not going to change how he writes or acts
[03:04:16] <wesolows> nachox: Actually I have little trouble understanding him technically (at least, no more than most native English speakers) but his language tends toward abrasiveness.  If you've been around long enough, you learn to overlook and ignore that sort of thing to a large extent.  But bringing out actual personal attacks is something else.
[03:04:23] * delewis without looking at the scrollback is able to determine the subject of the conversation is either schilly or nrubsig.
[03:04:28] <nrubsig> boyd: yes, he is. Up to a point where he damages other people's mind very quickly.
[03:04:37] <delewis> :-)
[03:04:50] <axisys> nrubsig: i check the nfs server is replying back to sunrpc but the host is not allowing the pkt in.. does the host needs an rpc client running as well.. i will check the firewall team once I check on possibilities on server side
[03:04:53] <boyd> delewis: :) I think you know which
[03:05:01] <axisys> host side and nfs server side that is
[03:05:09] <delewis> I'm guessing we've all seen the latest episode of "star(1) Wars!" on psarc-ext? :-)
[03:05:19] <brendang> perhaps opensolaris lacks role models from which people can learn acceptable behaviour
[03:05:22] * boyd is still catching up
[03:05:50] <wesolows> brendang: I don't think so.  But I do think it lacks them in the places where many people seem to hang out - especially the sewer that is -discuss.
[03:05:52] <sommerfeld> delewis: hah!  great name
[03:05:59] <boyd> brendang: you think? I think most people's behavior is acceptable
[03:05:59] * delewis bows
[03:06:08] <nachox> wesolows, i think that is the way he acts when he feels frustrated actually
[03:06:13] <wesolows> If you live on smf-discuss or dtrace-discuss or even ogb-discuss, or you routinely read ARC mail, there are plenty of good examples.
[03:06:41] <boyd> Yes, and zsf-discuss is downright pleasant.
[03:06:50] <brendang> boyd: most people, sure, but some (especially newbies) can act in a non-constructive way
[03:06:51] <boyd> err zfs-discuss
[03:07:05] <delewis> dtrace-discuss and zfs-discuss have always been extremely insightful, I think.
[03:07:05] <boyd> brendang: I think that's humanity right there.
[03:07:27] <delewis> psarc-ext is interesting to watch, too, when there's nothing silly going on.
[03:07:38] <axisys> nrubsig, boyd : nfs client seems to hanging at lockd http://rafb.net/p/tIlvNd50.html
[03:07:41] <delewis> (like there is now)
[03:08:08] * delewis goes back to making ksh93 packages
[03:08:22] <nrubsig> erm
[03:08:22] <axisys> but lockd is running
[03:08:25] <axisys>  root   208     1  0 15:43:55 ?        0:00 /usr/lib/nfs/lockd
[03:08:31] *** tsp has joined #opensolaris
[03:08:34] <nrubsig> delewis: packages for what ?
[03:08:34] <tsp> back
[03:08:42] <brendang> wesolows: there are plenty of examples, sure. but which people are easily identified as role models? I can think of a possible few.
[03:08:45] <nachox> delewis, nrubsig got you infected with ksh93 too?!
[03:08:48] * nachox hides
[03:08:48] <delewis> nrubsig: I got tired of not having ksh93 on my production Solaris 8, 9, and 10 systems.
[03:08:59] <wesolows> axisys: you can use rpcinfo to help diagnose this stuff
[03:09:12] <nrubsig> mom
[03:09:22] <delewis> I have a few scripts that make use of ksh93 extensions (namely, associate arrays) and therefore ksh88 doesn't cut it.
[03:09:27] <wesolows> brendang: In most places, it's the people who post the most, or who post a relatively small number of very prominent messages.
[03:09:59] <delewis> the ast build system is surprisingly nice to work with (vs. autoconf/automake/configure garbage)
[03:10:01] <nrubsig> delewis: it may be nice to use http://svn.genunix.org/repos/on/branches/ksh93/gisburn/scripts/buildksh93.ksh to build the packages (but I guess you need some adjustments fo Solaris 8 and 9)
[03:10:27] <delewis> nrubsig: I'm just using the standard ast system... bin/package setup.. bin/package make, etc.
[03:10:34] <axisys> wesolows: rpcinfo looks fine http://rafb.net/p/b8Sg3l86.html
[03:10:45] <delewis> copying the binaries to a PKGROOT and doing the appropriate package make magic (pkgproto, pkgmk, etc.)
[03:10:53] <axisys> wesolows: anything I am missing here for nfs.client to start
[03:10:56] <nrubsig> delewis: yes, but that doesn't include the C99 extensions by default... ;-(
[03:11:01] <delewis> its only two files, anyway (ksh93, man/man1/ksh93.1)
[03:11:12] <delewis> I'm not doing libast, nmake, sfio, etc.
[03:11:24] <nrubsig> delewis: and it may be nice to run the test suite before shipping the packages...
[03:11:37] <axisys> nrubsig: ok after commenting out the nfs mountpoints i could start nfs.client
[03:11:44] <delewis> nrubsig: yeah, I'm doing bin/package test.. if that's what you're referring to.
[03:11:52] <axisys> but then I am still on the issue
[03:12:07] * delewis is competent enough to read the README
[03:12:44] <nrubsig> delewis: you're better than RH&&Fedora then who needed two years to figure that out...
[03:12:51] <delewis> nrubsig: :-)
[03:13:06] <delewis> they were shipping broken packages, I take it?
[03:13:28] <nrubsig> half-broken
[03:13:37] <nrubsig> delewis: BTW: feel free to ship http://svn.genunix.org/repos/on/branches/ksh93/gisburn/scripts/gnaw.ksh &co. too
[03:13:51] * boyd laughs at James Carlson's comment about living to be insulted.
[03:14:04] <nrubsig> boyd: where ?
[03:14:04] <delewis> nrubsig: good idea. I've told a few friends about that and they're interested in seeing it.
[03:14:23] <boyd> nrubsig: Same thread
[03:14:44] <nrubsig> delewis: or better use the versions from http://svn.genunix.org/repos/on/branches/ksh93/gisburn/prototype005/usr/src/lib/libshell/common/fun/ for now.
[03:14:59] <nrubsig> delewis: */scripts/ is experimental stuff
[03:15:12] <delewis> ok
[03:15:21] <axisys> wesolows: rpcinfo is the way to troubleshoot? what looks missing in my rpcinfo output http://rafb.net/p/b8Sg3l86.html /
[03:15:24] <nrubsig> delewis: e.g. http://svn.genunix.org/repos/on/branches/ksh93/gisburn/prototype005/usr/src/lib/libshell/common/fun/termclock may be usefull, too.
[03:15:46] <nachox> ok, i need to review my scripting material, ksh93 seems to have typed variables....
[03:15:59] <delewis> nachox: ksh88 had those, too...
[03:16:05] <delewis> that's through typeset(1)
[03:16:21] <nrubsig> delewis: yes, but no float and no nameref
[03:16:26] <nrubsig> etc.
[03:16:28] <delewis> nrubsig: right
[03:16:31] <boyd> Gah... does he *willfully* not understand?
[03:16:45] 
[03:16:52] <nrubsig> boyd: short: *yes*
[03:16:53] * boyd nods
[03:16:58] <nachox> delewis, i think it had arrays through typeset
[03:17:14] <axisys> nfs mount: nfs-server: : RPC: Rpcbind failure - RPC: Timed out
[03:17:19] <delewis> nachox: upper case/lower case strings, too.
[03:17:24] <nrubsig> nachox: typeset -A for associative arrays and typeset -a for normal arrays
[03:17:29] <wesolows> axisys: I was just suggesting a tool you might not be aware of; I did not volunteer to troubleshoot the problem for you.  One thing to consider: NFS4 pretty much just uses a single port.  If you're having firewall problems, it might be easier to work with than NFS3.
[03:17:31] <nachox> but i dont know if it had integers
[03:18:00] *** laca has quit IRC
[03:18:03] <nrubsig> nachox: ksh88 had "Integer", too.
[03:18:05] <axisys> wesolows: my appology.. i will check w/ fw team .. thx
[03:18:20] <nrubsig> axisys: and maybe ask in networking-discuss@
[03:18:41] <axisys> nrubsig: i am assuming they dont have a irc chnl?!
[03:18:55] <axisys> nrubsig: i will go ahead email there
[03:18:56] <nrubsig> erm
[03:18:59] <nrubsig> unlikely
[03:19:17] *** sarahj has joined #opensolaris
[03:19:52] <nachox> nrubsig, really? i have searched tons of tutorials and never once seen integers, i guess i learn something new every day :)
[03:20:34] <nrubsig> nachox: "integer" is an alias for "typeset -i" or "typeset -li"
[03:23:45] <nachox> unalias integer :)
[03:29:23] *** tombhadAC has quit IRC
[03:43:00] <Jondice> anyone know how to set the default pager in solaris?
[03:43:07] * Jondice thinks less is more =/
[03:44:04] <nachox> exporting PAGER?
[03:44:07] <jbk> PAGER=/bin/less
[03:44:16] <Triskelios> you'll have to do it in /etc/profile
[03:44:31] <Jondice> thanks
[03:46:49] <Jondice> ah, so much better
[03:49:52] <jbk> you could also mapquest it :)
[03:49:54] *** karrotx_ has quit IRC
[03:49:57] <jbk> err
[03:49:58] *** Shiv__ has joined #opensolaris
[03:49:59] <jbk> wrong channel :)
[03:53:46] * nachox stabs jbk
[03:54:08] *** Teltariat has joined #opensolaris
[03:54:27] <jbk> i finally had to switch to a wired connection
[03:54:40] <jbk> the crappy 802.11b card i was using was just too flaky :(
[03:55:09] <jbk> fry's had a deal on a card that was supposed to use the atheros chipset, turns out it was marvell
[03:55:33] *** jamesd_ has joined #opensolaris
[03:56:26] *** jamesd2 has joined #opensolaris
[03:56:53] *** Megaf|Busy has quit IRC
[03:57:22] <tsp> the pcre in blastwave should be shot
[03:57:40] <tsp> they're up to version 7 and csw's still at 4 :)
[03:57:41] *** Gman_ is now known as Gman
[03:58:04] *** zacharym has joined #opensolaris
[04:00:15] <CIA-17> yt160523: 6550832 sata AHCI driver has to support SATA ATAPI CD/DVD devices, 6558736 ahci sata hba driver doesn't fetch the device signature correctly, 6559336 ahci sata hba driver doesn't handle correctly for POLLING/SYNC mode commands, 6583174 ahci driver: re-organize the interrupt handlers
[04:01:38] *** cypromis has quit IRC
[04:02:03] <nachox> nice!
[04:02:45] <wesolows> it was a good day for SATA
[04:02:47] <Triskelios> yay ahci. boo sata
[04:03:06] <wesolows> yes, SATA sucks, but if you have it, it was a good day for you
[04:04:23] *** Me[-_-]gaf has joined #OpenSolaris
[04:05:03] <nachox> SAS is ok
[04:05:21] <nachox> and sata is better than plain ide anyway
[04:06:40] <wesolows> SAS is more than ok
[04:06:50] <kjetilho> the only bad thing about SATA is that there is no market for fast disks.  i.e. no 15k RPM models
[04:07:25] <kjetilho> actually, the market is there, but the manufacturers prefer to charge a premium for SCSI/SAS versions
[04:07:50] * dlg heart sas
[04:09:37] <jamesd2> i'm happy with sata at home, and scsi at work...  its nice to have a team to handle all the storage... and all i have to do is ask for storage and execute a couple commands...    vxassist  take dg's and make volumes,  then newfs them and mount and add to  vfstab :-)
[04:10:21] <boyd> or zpool add ....
[04:10:22] <wesolows> wow, vxassist.  Do you use a flint when you are ready to cook dinner too?
[04:10:29] <boyd> heh
[04:10:32] * jmcp guffaws
[04:10:47] <jamesd2> wesolows: yes...   120 boxes are still running solaris 8
[04:11:02] *** cypromis has joined #opensolaris
[04:11:28] <wesolows> no wonder people think Solaris is "old" and "crusty" - they're all using a 10 year old version of it
[04:11:32] <nachox> ok these people are flying too high http://www.physorg.com/news106499717.html
[04:11:43] <jamesd2> i'm just glad my team doesn't manage any  2.5 or 2.6 boxes ... like some divisions at my company.
[04:12:02] <wesolows> ugh indeed
[04:13:06] <tsp> anyone have a box running.. lets see, really old... sunos 4.1.4?
[04:13:11] <nachox> well, solaris 8 was too successsfull for sun's own good :P
[04:13:26] <jamesd2> and they boxes are rock solid untill hardware fails or some bone headed app developer decides lets write our log files to /tmp
[04:13:41] <boyd> nachox: Funny, the copyright at the bottom says 2006!
[04:14:03] <nachox> hehe
[04:14:40] <tsp> jamesd2: tmp is tmpfs on sol8?
[04:15:58] <jamesd2> tsp: yes... and machines don't like it when you put 5GB large files in it.. when the machine only has a  4GB swap file and  7 copies of oracle running on it
[04:16:29] <nachox> can you write "crawl to death"? :)
[04:16:46] <boyd> There is something to be said for user quotas on /tmp. Personally I think a shared /tmp was a misdesign in the first place
[04:16:58] <jamesd2> nachox: yes... and of course... its being killed one line to the log file at a time.
[04:17:31] *** alfism has quit IRC
[04:17:58] <jamesd2> boyd: yes i agree, but corprate bean counters say that isn't friendly to our lines of businesses that are different divisions but still part of the same company.
[04:18:05] <nachox> /export/home/user/tmp :)
[04:18:16] <jbk> heh the mentality where i used to work was 'well there's 3 apps (out of over 1000) that are badly behaved and write to /tmp, therefore we _must_ put /tmp/ on ufs on every box
[04:18:27] <jamesd2> nachox: i would be happy with  /var/tmp  i can grow that filesystem  if needed.
[04:18:56] <nachox> symlink anyone?
[04:19:44] *** jafari has quit IRC
[04:19:46] <boyd> jamesd2: it isn't friendly when one line of business causes the others to break either.
[04:20:06] <jamesd2> boyd each business  has there own machine.
[04:20:28] <tsp> you think that's bad... I've got a U5 running zfs with 256mb of ram
[04:20:35] <boyd> jamesd2: Ah
[04:20:50] <Shiv__> Is there a way to prevent a user to fill up /tmp resulting in a denial of service ?
[04:20:53] <tsp> someone ran perl -e'fork while 1' and killed it :)
[04:20:54] <jamesd2> tsp that is painful, its bad enough on my  blade 1500  with 512MB
[04:21:15] <Shiv__> Quotas ?
[04:21:18] <tsp> jamesd2: I was going to upgrade it to 512, but sun memory is expensive
[04:21:19] <kjetilho> Shiv__: you can limit the size
[04:21:31] <kjetilho> or use quotas if you use UFS, yes
[04:21:34] <jamesd2> Shiv__: man quota
[04:21:57] <boyd> Limiting the size helps with the /tmp vs swap thing, but not with user vs use
[04:22:23] <jamesd2> tsp: you think  u5 memory is bad...  blade 1500 ram is  3-4x as much.. and i'm ungrading it to 1.5GB  ram is on the way.. found a steal...  2x512MB dimms for $75
[04:22:27] <flyingparchment> do USB card readers (as in SD/CF/etc cards) have any kind of standard interface that might work with non-Windows systems?
[04:22:48] <kjetilho> flyingparchment: yes, it's generally mass storage
[04:22:53] <jamesd2> flyingparchment: you can try it... i think most will show up as a dos filesystem.
[04:22:58] <wesolows> I know the one in the Ferraria works fine, shows up as a storage device
[04:23:05] <wesolows> *Ferraris
[04:23:59] <FrostCS> usb, sata, or ide card readers will all show up as mass storage
[04:24:22] <flyingparchment> that's surprisingly sane.  thanks
[04:24:36] <FrostCS> it's laptops that have problem, and not all, just the ones without a seperate interface to run them on..
[04:24:37] <Shiv__> boyd: Can you elaborate on your comment
[04:24:56] <flyingparchment> yeah, i'm looking at one of the desktop ones that fits in a floppy bay
[04:25:25] <FrostCS> your floppy memory card reader will work fine then.
[04:26:08] *** yongsun has joined #opensolaris
[04:29:55] <boyd> Shiv__: you can make /tmp have a max size, then it won't compete as much with other uses of swap (which from the kernel perspective normally refers to RAM + on-disk swap space)
[04:30:26] <boyd> Shiv__: On the other hand, it won't stop one user from using up all the available /tmp/ space and stopping others from putting stuff ther.
[04:30:32] <boyd> e.
[04:31:00] <boyd> Does that help?
[04:31:49] <nachox> isnt intels 945gm already supported in sxde?
[04:32:20] <Triskelios> nachox: snv_66 and later
[04:33:14] <nachox> ohh, it will be supported in the next sxde
[04:33:21] *** apersson has quit IRC
[04:35:28] <Shiv__> boyd: If a quota is set for a user, there will be a limit on how much space that user can takeup in /tmp
[04:35:48] <Shiv__> boyd: So, per user restriction (upper limit) is possible, right ?
[04:36:14] <boyd> Yes, per-user quotas help there, but they are not available on tmpfs AFAIK. You need to re-configure to make /tmp a UFS filesystem
[04:36:33] <Shiv__> Ok. Will check it out. thanks.
[04:36:41] <myrkraverk> why not zfs?
[04:36:51] <boyd> there are no per-user quotas in ZFS
[04:37:05] <myrkraverk> oh
[04:37:08] *** rbrown_ has joined #opensolaris
[04:37:20] <wesolows> which is actually good; you just need to follow a different administrative model there
[04:37:32] <myrkraverk> oh
[04:37:36] <boyd> Yes, but that doesn't help for a shared /tmp
[04:37:37] <wesolows> the filesystem is the unit of administration; if you want per-user quotas, use per-user filesystems.
[04:37:59] <myrkraverk> yeah -- am doing that, for myself and a friend ;-P
[04:38:04] <wesolows> indeed not, but anything that writes to /tmp and doesn't honour TMPDIR is broken so the right thing to do there is fix your shit
[04:38:24] <boyd> (which is part of why I said that shared /tmp was a misdesign from the start)
[04:38:43] <nachox> haha, now there is someone asking if sun's hardware is going to dissapear in osol-discuss :P
[04:39:08] <myrkraverk> well, one can (probably) patch the kernel to make /tmp point to a per-user fs ;-P
[04:39:22] <boyd> That must be one of the threads I catchup-ed
[04:39:34] <jmcp> nachox: in the "Unix belongs to Novell" thread?
[04:39:42] <nachox> no
[04:39:56] <nachox> [osol-discuss] OT: Sun hardware to disappear?!?!
[04:39:57] <jmcp> which one?
[04:39:57] * boyd wonders about making a pseudo FS that reports "Fix your shit - wesolows" on every attempt to access it.
[04:40:16] <jmcp> nachox: not seeing it
[04:40:28] <boyd> I just got it
[04:40:33] <myrkraverk> boyd: and wesolows being? you?
[04:40:38] <boyd> Well, at least it says "OT"
[04:40:41] <boyd> myrkraverk: ^^
[04:41:17] <jbk> boyd: just use dtrace :)
[04:41:23] <boyd> Heh
[04:41:36] <myrkraverk> ah, wesolows was outside my visible area ;-P
[04:41:52] <nachox> boyd, if it logs that to /var/log/messages all you will do is getting "fix your shit" related messages to fill your filesystem very quickly :P
[04:42:04] <boyd> something like if(TMPDIR is set && TMPDIR != "/tmp") { error("Fix your shit") }
[04:42:22] <boyd> Yes, maybe it should return that message in all reads :)
[04:42:34] <flyingparchment> ln -s '${home}/tmp' /tmp
[04:42:52] <boyd> It's one of those areas where Plan9-style bindings would be nice
[04:43:08] <kjetilho> why not give every user his own zone?
[04:43:11] <jmcp> nachox: I like kaiwai's response
[04:43:25] <boyd> flyingparchment: and when the second user logs on?
[04:43:36] <nachox> jmcp, i like mine better, i am just going to ignore it :P
[04:43:39] <boyd> jmcp: I don't see that yes :)
[04:43:44] <boyd> yet
[04:43:51] *** sarahj has quit IRC
[04:44:04] <FrostCS> should I open up my osol folder? I can troll it more I am sure.
[04:44:11] <flyingparchment> boyd: notice the quotes
[04:44:20] <flyingparchment> boyd: that was a feature suggestion, not an existing solution
[04:44:43] <rpm> what does opensolaris have for an ipsec? does have a built-in isakmp or ike daemon?
[04:44:45] <boyd> Oh, I see what you're getting at... like CDSLs
[04:44:55] <nrubsig> The logic behind T2 GPL'ed ---> SUn leaving hardware buisiness is... weird.
[04:45:03] <jmcp> rpm: there's a builtin ike daemon
[04:45:06] <nrubsig> Sun is dyyyyyyyiiiing.
[04:45:13] <nrubsig> would be the next claim
[04:45:20] <jmcp> nrubsig: and OpenBSD is dead!
[04:45:21] <jmcp> :-)
[04:45:22] <boyd> Why didn't T1 -> GPL mean that?
[04:45:32] <jmcp> rpm: ik.iked
[04:45:37] <boyd> rpm: man in.iked
[04:45:39] <boyd> <snap>
[04:45:46] <jmcp> (1m), too
[04:46:00] <nrubsig> boyd: dunno, the email reads like someone needed attention
[04:46:03] <jmcp> boyd: because terminator #1 "died"
[04:46:14] <boyd> Hehe
[04:46:22] <nachox> boyd, come on, you only need a who knows how many billions factory to make those cool t2 procs cheaper than sun! :P
[04:46:50] <nrubsig> nachox: still the patents are a problem
[04:47:01] <jmcp> nrubsig: which patents?
[04:47:09] <boyd> There are FPGA implementations of the T1 now
[04:47:49] <nrubsig> jmcp: uhm... isn't most of the T1/t2 design somehow protected by patents (at least I would do that... and I hear from time to time that I am not evil... therefore...) ?
[04:48:09] <nachox> only becuase sun didnt yet decide to sue them :P
[04:48:13] <nrubsig> jmcp: GPL license doesn't mean something doesn't violate patents
[04:48:36] <nachox> if only they released them using cddl...
[04:48:54] <boyd> But then everyone would complain...
[04:49:02] <richlowe> everyone complains anyway.
[04:49:03] <jmcp> I'm pretty darned sure that whenever Sun releases something especially in source, then the lawyers have gone over it with multiple fine-toothed combs to avoid *any* possibility of being sued for patent infringement
[04:49:04] <Doc> but CDDL sucks!  i read it on slashdot!
[04:49:05] <richlowe> I say we burn them.
[04:49:06] <boyd> Oh, wait.. they complain anyway...
[04:49:11] <boyd> <snap again>
[04:49:22] <jmcp> Doc: and OpenBSD is dying!
[04:49:25] <boyd> richlowe: All of them?
[04:49:28] <nachox> boyd, of course not, the gpl community is full of idiots ;)
[04:49:33] <boyd> Doc: Must be true then
[04:50:15] <nrubsig> jmcp: I mean Sun holding patents, not SUn violating patents
[04:50:24] <nachox> jmcp, the problem goes the other way round actually, sun holds the patents
[04:50:57] <boyd> All the hollerin' about philosophical positions that the GPL brings look suspicously flimsy when they're all using the cheaper proprietary CPUs rather than the GPL ones that are available.
[04:50:58] <jmcp> nrubsig: yeah, and so what's the problem? Sun's licensed the verilog etc under GPLv2, so that brings with it certain requirements regarding re-implementation
[04:51:23] <jmcp> boyd: yup. I long to ask HP fudsters when they're going to GPL their cpus, or OS
[04:51:29] <jmcp> ditto for IBM and Power
[04:51:47] <nachox> i will laugh to death when some company releases something they hold the patent to and then sues for patent violation everyone who uses the code :P
[04:51:59] <boyd> It shows it for what it really is: I want it to be cheap. Ideally free.
[04:52:15] <nrubsig> sue me!
[04:52:20] *** nrubsig was kicked by nrubsig (You have been sued by Sun. Pay $10000192949838 and go directly to jail. Do not collect any income in the meantime.)
[04:52:29] <nachox> hahaha
[04:52:35] <jmcp> nrubsig: you're being a dick. stop it
[04:52:36] <boyd> I knew there was a reason we went 64bit.
[04:52:37] *** Jondice has quit IRC
[04:52:44] * jmcp waits for nrubsig to return
[04:52:59] <boyd> So you can call himn a dick again?
[04:53:02] <jmcp> yes :)
[04:53:08] <flyingparchment> i guess #opensolaris needs to borrow the Wikipedia policy, "Don't be a dick"
[04:53:27] <nachox> jmcp, he drew the go to jail card from the monopoly game :)
[04:53:28] <richlowe> boyd: it's as practical as anything else :)
[04:54:00] <boyd> richlowe: Sorry, I'm not sure what thread that's in response to.
[04:56:04] <kjetilho> what is the correct command to tell SMF that I've fixed the service?
[04:56:09] <flyingparchment> svcadm clear
[04:56:22] <movement> why it's not repair ... grumble
[04:56:28] <kjetilho> flyingparchment: ah, thanks!
[04:56:41] <nachox> because you are not asking smf to fix anything
[04:56:43] <wesolows> because repair implies that it does the repairing for oyu?
[04:56:54] <nachox> youre just clearning the broken flag
[04:56:57] <boyd> yes, but it's fmadm repair :)
[04:57:05] <wesolows> (of course, debuggers are misnamed too)
[04:57:14] <wesolows> boyd: yeah, I don't get that either
[04:57:17] <boyd> ... which apparently sprouts an arm and fixes the HW
[04:57:29] <jmcp> debuggers should stop screwing around and actually help me
[04:57:30] <kjetilho> I think "retry" would be a good command verb
[04:57:32] * boyd lunches
[04:57:43] <boyd> It's clear on zpool too
[04:57:46] <movement> wesolows: I'd prefer clear for both, but fmadm was first, no?
[04:57:54] *** Shiv_ has joined #opensolaris
[04:58:02] <wesolows> movement: It was.
[04:58:27] *** Chipdancer has quit IRC
[04:58:44] *** Shiv_ has left #opensolaris
[05:00:17] <CIA-17> pl196000: 6579740 simplify aac_cmd_fib() for efficiency and scalability, 6579744 simplify aac_intr_norm() for efficiency and scalability, 6579757 aac_tran_init_pkt() should use the arguments passed down for DMA functions, 6579778 driver should allocate aac_cmd FIB dynamically, 6579786 SCSI_HBA_TRAN_CLONE is not needed for aac, 6579800 aac driver should do data synchronization for consistent packets, 6580283 New Comm. need not to send AIF response back t
[05:01:11] <kjetilho> hey, how nice of Sun to pre-image the disks in Thumper as a zpool!
[05:01:14] *** Shiv has joined #opensolaris
[05:02:03] <Triskelios> what's the pool named, out of curiosity?
[05:02:09] <kjetilho> zpool1
[05:02:12] <jamesd2> kjetilho: the truth is they want you to run solaris on it, because the phone support people dont want to think about setting up lvm in linux on it or   svm
[05:02:27] <jamesd2> over the phone.
[05:02:56] <kjetilho> hehe.  I was dreading entering all the device names :)
[05:03:17] <boyd> That's what shell wildcards are for.
[05:03:47] <kjetilho> hmm.  well, I was going to parse the output of format </dev/null
[05:04:06] *** Shiv__ has quit IRC
[05:04:19] <sommerfeld> I normally use c{1,2,3,4}t0d0 type expansion
[05:05:16] *** Marv|LG has joined #opensolaris
[05:05:25] <kjetilho> weird, the wildcards don't add up
[05:05:54] <kjetilho> only 46 entries in /dev/dsk, but 48 disks
[05:06:05] <Marv|LG> no power, 115 degress not a good day
[05:06:07] <boyd> someone stole 2?
[05:06:33] <boyd> Marv|LG: It's a good day to go swimming
[05:06:36] <sommerfeld> i'd think there would be 48 * 8 entries or something like that
[05:06:43] <sommerfeld> (8 slices per disk)
[05:07:08] <boyd> good point
[05:07:10] <sommerfeld> IIRC the stock config has an svm mirror of two disks for boot and 46 disks in a big pool
[05:07:17] <kjetilho> oh, more than that.  but I only counted c*t*d0
[05:07:43] <sommerfeld> what does metastat -c say ?
[05:07:46] <boyd> In that case you're missing the boot pair, which are SMI vtoc, not EFI
[05:08:08] <boyd> and hence don't have a c*t*d* entry... only ones with slices
[05:08:23] <kjetilho> ahhh, yes, it's the ones in SVM which are missing
[05:10:49] *** halton has joined #opensolaris
[05:11:03] <kjetilho> if I check c*t*d*s2, I get two extra -- probably DVD-ROM, and what else?
[05:11:22] <boyd> usb?
[05:11:39] <boyd> does the LON in them have a fake device?
[05:11:41] <boyd> LOM
[05:11:48] <kjetilho> ah, yes
[05:11:59] <Marv|LG> i swam 18 laps
[05:12:20] <boyd> Across, or along? :)
[05:13:05] <sommerfeld> yup, ilom has 3 usb connections to the host: keyboard, mouse, and hallucinated storage.
[05:13:34] <Triskelios> sommerfeld: that is the most awesome storage ever
[05:14:37] <sommerfeld> actually, no.
[05:14:55] <sommerfeld> my understanding is that you boot much faster via pxe
[05:15:11] <Triskelios> okay. it's all in the name =P
[05:15:25] <kjetilho> I guess it can be convenient when you need a Driver Assistant floppy
[05:17:28] *** Jondice has joined #opensolaris
[05:19:18] <tsp> how do I see how much ram is being swapped/used?
[05:20:52] <Triskelios> vmstat (-s)
[05:22:39] *** Chip has joined #opensolaris
[05:22:44] <tsp> hmm
[05:24:37] *** Chip is now known as Chipdancer
[05:26:02] <jbk> hmm.. hg is pretty neat
[05:26:09] <jbk> i hope i'm using it correctly :)
[05:33:38] *** rbrown_ is now known as phus
[05:33:48] *** stevel has joined #opensolaris
[05:33:49] *** ChanServ sets mode: +o stevel
[05:47:47] <Shiv> Flash archive restoration uses profile file that has partition information (along with device ids)
[05:47:59] <Shiv> If a DVDROM is removed the id numbering changes.
[05:48:26] <Shiv> Is there a way to make it independent of whether DVDROM is present or not
[05:49:16] <Shiv> The controller id that was say c3 for a hard disk would change to c2 after removal of the DVDROM.
[05:49:50] <boyd> you can use the term "rootdisk" I'm not sure how that is mapped to a device
[05:50:53] <Shiv> What about other hard disks (non-root)
[05:51:00] <Shiv> post setup scripts ?
[05:51:22] <Shiv> I will check out the "rootdisk" hint
[05:51:30] <boyd> Dunno about the others...
[05:52:38] *** Shiv has left #opensolaris
[05:53:38] *** nachox has quit IRC
[05:55:00] <richlowe> jbk: driver-discuss@ may have been a better choice (and a good bet if you don't get responses on networking-discuss)
[05:55:46] <jbk> oh
[05:55:52] <jbk> didn't see that one :)
[05:56:43] <axisys> what is a good audit tool for sol 10.. if a file permission get changed i need to find out who changed it
[05:56:59] <jbk> well
[05:57:06] <jbk> there's the audit subsystem :)
[05:57:09] <axisys> or if a file ownership got changed i want to know who changed it
[05:57:10] <jbk> you can log all chmod's
[05:57:23] <axisys> jbk: BART is it?!
[05:57:28] <jbk> no
[05:57:33] <axisys> BSM?
[05:57:38] <jbk> yes
[05:57:41] <jbk> bsm
[05:57:55] <axisys> jbk: i thought that one is a bitch.. :-)
[05:58:31] <axisys> jbk: let me revisit it
[05:58:56] <axisys> jbk: is there a man page? man bsm gets me nothing
[05:58:59] <jbk> just enable the fm class for any successes
[05:59:13] <jbk> audit_control
[05:59:17] <jbk> bsmconvert
[05:59:29] <axisys> jbk: what is fm class?
[05:59:39] <jbk> i think it stands for 'file modification'
[05:59:56] <jbk> basically, you have a bunch of different events (defined in /etc/security/audit_event)
[06:00:10] <axisys> jbk: ok
[06:00:13] <jbk> related events are grouped into classes (/etc/security/audit_class)
[06:00:15] <CIA-17> rs135747: 6490400 vhci_mpapi_update_tpg_acc_state_for_lu cann access freed memory, 6541881 mpapi panic when mdi_pi_online fails
[06:00:16] <CIA-17> jw149990: 6531069 SCSI2 (tc_mhioctkown test cases) testing are showing UNRESOLVED results for ST2530, 6588278 mpt driver doesn't check GUID of LUN when the path online, 6591973 panic in mdi_pi_free() when remapping devices
[06:00:59] <axisys> jbk: did not find a single sol blueprint on it
[06:01:16] <boyd> axisys: docs.sun.com
[06:01:23] <jbk> so you select which classes you want to audit
[06:01:32] <jbk> the syntax is a bit outdated imo
[06:01:39] <jbk> it feels like it's form the 80's :)
[06:02:00] <boyd> At least it doesn't feel like it's from the 2000's and is therefore in XML
[06:02:04] <jbk> though still has nothing on sendmail.cf :)
[06:02:05] <jbk> haha true
[06:02:44] <axisys> boyd: cool
[06:03:22] <axisys> boyd: found a cool page by Hal Stern
[06:03:27] <jbk> i know there was talk of being able to enable auditing without a reboot, don't know if it's been putback (or even if work is still going on with it)
[06:03:32] <boyd> axisys: URL?
[06:03:41] <axisys> oops I meant Hal Pomeranz ..url yes
[06:03:53] <boyd> jdk: still going on AFAIK
[06:04:08] <boyd> axisys: That wasn't meant to be a boolean question
[06:05:05] <jbk> if they get the remote auditing storage piece putback, i'll be happy :)
[06:05:28] *** Megaf has joined #OpenSolaris
[06:05:40] <axisys> boyd: http://www.deer-run.com/~hal/sysadmin/SolarisBSMAuditing.html
[06:05:49] *** Megaf has quit IRC
[06:06:03] <axisys> boyd: kind a old.. i think amy rich has one.. looking for that
[06:07:33] <boyd> jbk: http://opensolaris.org/os/project/audit/projects/remote_storage/
[06:08:38] <jbk> boyd: yeah..
[06:08:47] <jbk> doesn't indicate if it's been integrated yet
[06:12:20] *** cypromis has quit IRC
[06:12:51] *** Me[-_-]gaf has quit IRC
[06:16:40] *** Megaf has joined #OpenSolaris
[06:20:48] <axisys> sounds like bsm is pretty resource intensive.. would be nice if i could use some dtrace tool (like the way appcrash.d works) to watch for change in a file and generate an event
[06:23:25] *** edwardoc has joined #opensolaris
[06:23:34] *** stevel has left #opensolaris
[06:23:47] <edwardoc> hey all
[06:23:57] <edwardoc> ok my SGD
[06:24:02] <jbk> axisys: depends on what you're doing
[06:24:08] <jbk> if you audit _everything_, then yeah
[06:24:14] <edwardoc> and my bandwidth is getting rinced as we speak
[06:24:25] <edwardoc> I got a quick question
[06:24:32] <jbk> if you're auditing stuff that happens inrfrequently, then not going to be noticable
[06:25:32] <edwardoc> why after a SXDE install would it hang on bootup for the first two times just before the Homename : part ?
[06:25:59] <edwardoc> after a few 'hard' reboots it works fine now? did two installs and same thing ???
[06:26:35] <tsp> edwardoc: did you try boot -v?
[06:26:46] <Megaf> Bye All, Good Night
[06:26:47] *** Megaf is now known as _Megaf|Sleep
[06:27:12] <edwardoc> tsp: no I will do that, i don't know solaris boot parm
[06:27:18] <boyd> axisys: You *could* use dtrace, but it discards events under load...
[06:27:36] <edwardoc> boyd:Hey man
[06:27:39] <boyd> axisys: audit is only bad if you enable too many events
[06:27:42] <boyd> hey edwardoc
[06:27:49] <boyd> What happened to the long nick?
[06:28:14] <edwardoc> I am on Tempt SGD and my band width will run out any min now
[06:28:19] <edwardoc> hehe
[06:28:33] <edwardoc> i could not be bother sign in :p
[06:28:38] <boyd> Heh
[06:28:45] <edwardoc> hows things mate?
[06:28:57] <boyd> Tempt was telling me about that stoopid limit.
[06:29:03] <boyd> Not bad... nice to be home
[06:29:26] <edwardoc> yea always feels good ah
[06:29:43] <edwardoc> yes the limit is really getting to me
[06:29:54] <edwardoc> the worst part is its a Linux proxy
[06:30:06] <edwardoc> how poo poo is that
[06:30:33] <edwardoc> it always goes down at 11pm when it backs up its logs or something ?
[06:30:42] <edwardoc> every night for about 5min
[06:31:12] <edwardoc> I can get to the microsoft site without limit though ?
[06:31:24] <edwardoc> well most of it
[06:31:33] <edwardoc> must be fore updates
[06:32:11] <edwardoc> ~Sorry for my flood of IRC entropy
[06:32:45] <boyd> (sorry.. .phone)
[06:33:23] <edwardoc> I was readingg about the IBM deal today
[06:33:34] <edwardoc> maybe more drivers one hopes
[06:33:42] <boyd> could be...
[06:34:04] <boyd> Drivers for things you/we want is another question..
[06:34:25] <edwardoc> This weekend me and my class mate are going to do loads on JPack with NetBeans :D
[06:34:46] <edwardoc> true very true
[06:34:55] <boyd> Cool
[06:35:04] <edwardoc> I want SATA and NIC, they are at the top of my list
[06:35:19] <boyd> some SATA stuff went back today
[06:35:32] *** pg_user has joined #opensolaris
[06:36:31] <edwardoc> Next week I want to have a hack at making ext2 filesystem automount and be supported in a SXDE build
[06:36:57] <edwardoc> really, whats new on the driver front ?
[06:37:12] <boyd> NVidia SATA IIRC
[06:37:56] <pg_user> Hi guys
[06:37:57] <Gman> Tpenta, jmcp: you listening to marc?
[06:38:15] <pg_user> I know what Jon's strategy is regarding linux
[06:38:27] <pg_user> it's quite brilliant
[06:38:42] <Triskelios> edwardoc: the new ahci stuff probably should work with many intel, sis, via, and nvidia chips...
[06:39:02] <pg_user> sun is redefining UNIX to mean DTrace, ZFS and zones
[06:39:18] <pg_user> BSD will have them, OS X will have them
[06:39:23] <pg_user> and linux will not
[06:39:48] <richlowe> Gman: eh?
[06:39:58] <Gman> richlowe, internal townhall over in australia
[06:40:13] <edwardoc> great ! Man I got this crap Asus MB from when I was down in Melbroune
[06:40:18] <richlowe> Gman: Oh, thought maybe I was missing a usergroup with a dial-in.
[06:40:23] <Gman> richlowe, nope
[06:40:25] <edwardoc> Its has a nvid chipset
[06:40:29] <kjetilho> pg_user: and Solaris will be more GPL than Linux to boot!
[06:40:36] <Gman> though a pity there wasn't a dial in for san diego
[06:40:41] <boyd> You need linux to boot?
[06:40:46] <kjetilho> hehe
[06:40:50] <edwardoc> lol
[06:40:53] <kjetilho> only on the PS3
[06:41:02] <edwardoc> Ah yes I got a PS3
[06:41:13] <pg_user> kjetilho: I hope solaris doesn't go GPL
[06:41:20] <edwardoc> Mate would you be intersted to port with me ?
[06:41:21] <kjetilho> really?  why?
[06:41:33] <pg_user> GPL puts a undue burden on developers
[06:41:41] <richlowe> kjetilho: because the arguments are so much fun.
[06:41:43] <delewis>  /me watches #opensolaris descend into yet another licensing argument
[06:41:50] <pg_user> your are constantly hassled for the source
[06:41:53] <edwardoc> I _Really_ ! would like to see my SunOS kernel on Cell
[06:41:53] <richlowe> delewis: I think this is where we set people on fire.
[06:41:58] <boyd> Hehe
[06:42:00] * purserj does a change of pace and talks about removing X
[06:42:04] <delewis> richlowe: pass the kerosene.
[06:42:09] <boyd> purserj: now now
[06:42:12] <pg_user> CDDL is like the BSD license
[06:42:18] * Gman throws in an emacs vs vi debate
[06:42:21] <pg_user> Free as can be.
[06:42:26] <edwardoc> delewis:Hey mate hows things?
[06:42:33] * boyd thinks about founding #opensolaris_license
[06:42:38] <pg_user> Gman: emacs violates the philosophy of UNIX
[06:42:46] <kjetilho> pg_user: well, it would be dual licence, I guess
[06:42:48] <Gman> pg_user, so does colorls
[06:42:53] <richlowe> pg_user: that's a feature.
[06:42:56] <richlowe> :)
[06:43:10] <kjetilho> that is, patches which aren't dual licenced are basically ignored.
[06:43:16] <pg_user> emacs has the design idealogy of the boys in redmond
[06:43:39] <pg_user> i used to think the gpl was the greatest thing too
[06:43:41] * purserj hopes they find RMS just so he can watch pg_user say that to him in person
[06:43:48] <boyd> What, you mean the way you can customise it with your own code on the fly?
[06:43:51] <delewis> boyd: found #crackpottery while you're at it, too.
[06:43:54] <edwardoc> i see some linux distros only with nano :O these days!
[06:43:58] <boyd> delewis: I'm on it.
[06:43:59] *** oninoshiko has quit IRC
[06:44:48] *** karrotx_ has joined #opensolaris
[06:44:57] <Triskelios> pg_user: dude, I forgot to complain about emacs the last time I talked to RMS =\
[06:45:10] <edwardoc> edward sits back and enjoys his IRC ride
[06:45:55] <pg_user> emacs is OK.  I use it
[06:46:01] <boyd> dtrace -wn 'syscall::exec*:return /execname=="nano"/ {raise(9)}'
[06:46:18] <pg_user> but that doesn't mean it is consistent with the UNIX philosophy
[06:46:29] <edwardoc> I neaver really used it, only vim
[06:46:45] <richlowe> boyd: no, copyout over args.
[06:46:52] <boyd> Heh
[06:46:56] <tsp> heh, I use ed and sometimes vim for email :)
[06:46:57] <Triskelios> emacs confuses the hell out of me when I'm pair-programming
[06:46:57] <edwardoc> although the Solaris version of vi is diffrent to the BSD/Linux one ?
[06:46:58] <dlg> anyone here built ON on solaris?
[06:47:09] <Triskelios> dlg: many times...
[06:47:17] <boyd> of course, Plan9 is more consistent with the UNIX philosphy than Unix is...
[06:47:20] <dlg> Triskelios: any gotchas i should look out for?
[06:47:41] <richlowe> use the right compilers, don't try building on 10, or something silly like that.
[06:47:43] <Triskelios> dlg: not really, I think the documentation is basically correct
[06:47:50] <pg_user> Emacs is the equivalent of Microsoft bundling IE into the OS.
[06:47:53] <dlg> k
[06:47:59] <dlg> i just put ss11 on instead of 12
[06:48:01] <dlg> and patched
[06:48:02] <richlowe> and don't download the closed-bins to usr/closed, it hates that ;)
[06:48:09] <richlowe> (yes, I am a dumbass, sometimes)
[06:48:09] <dlg> got the right onbld
[06:48:12] <dlg> it should all be ok now?
[06:48:15] <edwardoc> how do I upgrade my SXDE 5/07 with a new ON, can I do that or do I need to reinstall ?
[06:48:54] <Triskelios> dlg: it does need closed-bins (preferably nd)
[06:49:01] <boyd> edwardoc: You can upgrade (Live or... dead) or you can BFU, but that will only update ON, not other consolidations and it breaks the normal upgrade path for the future
[06:49:09] <richlowe> Triskelios: "preferably" is a matter of opinion.
[06:49:24] <richlowe> dlg: Yeah, it's not rocket science.
[06:49:31] <pg_user> has anyone talked with Bill Joy lately?
[06:49:36] <edwardoc> boyd:I just want to get the extra drivers ?
[06:49:49] <dlg> richlowe: im dumb, humour me
[06:50:11] <richlowe> dlg: nah, I was just suggesting you don't seem to have missed anything other than actually doing it. :)
[06:50:16] <dlg> k
[06:50:24] <dlg> building on nfs is probably not a good idea either
[06:50:50] <boyd> edwardoc: Well, Either will do that. BFU is faster, but then you're stuck on the BFU track until you re-install
[06:50:53] <edwardoc> boyd:Is it a good idea to upgrade with BFU as i have very limited bandwidth ?
[06:50:58] <Doc> ohh yay to a dieing disk :(
[06:50:59] <richlowe> dlg: should work
[06:51:09] <boyd> There's less to download, so that's a help
[06:51:14] <pg_user> i have cdrw and cdrecord which is better?
[06:51:28] <boyd> cdrecord is better since 18 years.
[06:51:31] <edwardoc> boyd:I think thats the go for me at the moment
[06:51:32] <richlowe> boyd: hah.
[06:51:56] <Triskelios> Bill Joy wrote vi and co-founded Sun, BSD one might be the same --> <edwardoc> although the Solaris version of vi is diffrent to the BSD/Linux one ?
[06:52:04] <edwardoc> boyd:Any good howtos around ?
[06:52:04] <tsp> speaking of dying disks... can anyone ls /usr/bin/psset without getting an I/O error?
[06:52:18] <tsp> I just get /usr/bin/psset: I/O error
[06:52:21] <Doc> /usr/bin/psset: No such file or directory
[06:52:24] <boyd> edwardoc: For BFU?
[06:52:35] <edwardoc> boyd:yes
[06:52:49] <richlowe> the devref
[06:52:50] <boyd> Umm... benr had a post...
[06:52:53] <tsp> same with /usr/bin/texi2dvi4a2ps
[06:52:54] <richlowe> opensolaris.org/os/community/on
[06:52:55] <boyd> yes, that too
[06:53:12] <edwardoc> i did have a read of that yesterday
[06:53:20] <edwardoc> but not really clear on it
[06:53:22] <boyd> http://www.cuddletech.com/blog/pivot/entry.php?id=702
[06:53:23] <Triskelios> pg_user: cdrw is simpler to use if you just want to burn something quickly
[06:53:42] <pg_user> i want to do backups to my cdr
[06:53:51] <pg_user> and i want to have multiple sessions.
[06:54:37] <boyd> cdrecord had more functions. I don't know if you can do multi-session with cdrw.
[06:55:06] <pg_user> joerg is my hero!
[06:56:07] <Triskelios> you can with cdrw, but you'd have to write down the end of the last session, so if cdrecord can avoid that it would be better
[06:58:16] <pg_user> could someone write a patch for cdrw that would look at the last session and then burn past that?
[06:58:42] <pg_user> like this 'cdrw --next-session myimage.img
[06:59:04] <pg_user> writing down the next session is a chore
[06:59:26] *** phus has quit IRC
[06:59:43] <Triskelios> oh wait, cdrw -M does list the sessions
[06:59:44] <edwardoc> boyd:Thanks man
[06:59:52] <boyd> You can get it with -msinfo from cdrecord.
[07:00:06] <edwardoc> I will have a go and I can see me doing a reinstall :p and wasting lots of time
[07:00:07] <pg_user> i would like to automate this too.
[07:00:38] <pg_user> i would like to do a zfs backup and then be able to restore it on another machine if need be.
[07:00:39] <boyd> edwardoc: Well, worst case you leart something
[07:00:51] <pg_user> watch out for ubuntu and solaris
[07:01:10] <pg_user> ubuntu overwrites the MBR EVERY time you install software...
[07:01:15] <boyd> pg_user: I was looking at something like that... There are more docs on the net for multi-session with cdrecord than with cdrw
[07:01:38] <edwardoc> by the way, I tried that compiz for solaris and it installed ok but ended with a Segment Fault when config X for the extensions or something
[07:01:39] <pg_user> yes, i was looking at zfs send
[07:01:48] <edwardoc> is this a common problem
[07:02:00] <pg_user> but I'm a bit nervous at trusting it...
[07:02:03] <edwardoc> boyd:Yes true, I do love to learn
[07:02:12] <Triskelios> there is a good example in the man page for cdrw, actually. but obviously cdrecord is vastly more popular
[07:02:58] <pg_user> zfs send mysnapshot@bla > cdrw --nextsession /dev/rdsk/c0d0t01
[07:03:03] <boyd> pg_user: that's what I had half-built..
[07:03:10] <pg_user> that is what i want to do
[07:03:36] *** Doc has quit IRC
[07:03:39] <boyd> I was putting them into files and then burning a multisession iso9660
[07:04:17] <pg_user> boyd: yes, i want to do this too.
[07:04:45] <edwardoc> Not that I care to great for compiz but I would like to compare it to how stable it is compared to Linux
[07:05:12] <pg_user> i would like to be able to do this with zfs though.  My /home directory as well as my /opt  and /usr/local/pgsql/data directories are on ZFS
[07:05:24] <edwardoc> I do feel like X is getting to unstable these days, anyone else feel the same?
[07:05:44] <jmcp> Gman: nope
[07:05:56] <edwardoc> I hear this is a big push for a new framework that brakes somethings from Xlib to something I can;t remember the name of
[07:06:05] <pg_user> I lost a lot of data when I was doing an installgrub after ubuntu killed my boot partition and I made a typo.
[07:06:52] <pg_user> I had a bunch of DTrace scripts that I lost.
[07:07:03] <pg_user> which took me forever to create
[07:07:06] <edwardoc> ubunut is getting very bad these days, way unstable with some intel hardware
[07:07:15] <pg_user> I agree edward
[07:07:42] <pg_user> I only dual boot windows now.  Linux is going down hill. It peaked with SUSE 10
[07:07:51] <edwardoc> too many people working on the same thing, not enother leaders
[07:08:04] <pg_user> did you guys hear how ubuntu's servers got hacked?
[07:08:11] <boyd> pg_user: I was using my method with zfs send and it was zfs send files that I put on the DVD
[07:08:18] <edwardoc> yea Linux seems real bad last few Kernels
[07:08:19] <timeless> over statement
[07:08:26] <timeless> it wasn't a core server
[07:08:35] <pg_user> boyd: what did you type to do this?
[07:08:36] <edwardoc> They took out /dev/hda and everything is /dev/sda now
[07:08:45] <timeless> it'd be like saying that the helsinki open solaris user group server got hacked
[07:09:10] <pg_user> if ubuntu had RBAC and ZONES it wouldn't have happend
[07:09:13] <boyd> pg_user: zfs send blah > /some/path/to/file then mkisofs from the dir and add it to the session
[07:09:32] <Triskelios> pg_user: stupid sysadmins will always prevail
[07:09:33] <edwardoc> seems that messed up alot, also not sure how good this tickerless kernel is now?
[07:10:04] <pg_user> boyd: I get the general idea of it, I'm still nervous of writing it myself.  It would have to be a ksh script that would run daily.
[07:10:19] <pg_user> would you have a link to a tutorial on this?
[07:10:36] <edwardoc> edward wonders what would happen if Linux had DTrace and they used it to find all there bugs
[07:10:45] <boyd> yes. I never finished it since I have to dual boot the machine - it's running Debian ATM. I'll get back to it some time :(
[07:10:50] <pg_user> I actually liked it when you had options when installing linux
[07:11:07] <pg_user> now it just does whatever to your system, no way to specify on the desktop install.
[07:11:08] *** emealy has joined #opensolaris
[07:11:15] <Triskelios> pg_user: it's very straightforward... you can even just burn the file directly to disc if you want
[07:11:37] <pg_user> I'm going to try it tonight or tomorrow Triskelios
[07:11:45] <edwardoc> Slackware is the only Linux worth looking at now man
[07:12:05] *** zacharym has quit IRC
[07:12:11] <edwardoc> My view anyways
[07:12:13] <pg_user> I am a bit scared though, if I do this with zfs, and there is trouble I will only be able to get it back on solaris.
[07:12:16] <emealy> dlg: did you kill the nightly when you left? it died saying "hangup" in the build noise section of the mail
[07:12:23] <pg_user> and i'm a solaris newbie
[07:12:34] <dlg> emealy: i logged out, that probably did it
[07:12:38] <dlg> just start it again
[07:12:39] <emealy> dlg: its re-running now.. will update you on how it goes
[07:12:42] <dlg> or move it to disk and start again
[07:12:46] <emealy> already had
[07:12:49] <boyd> pg_user: you could use star. It's been POSIX complian since 324 years.
[07:12:50] <pg_user> what if sun dual licensed solaris. CDDL and GPL
[07:12:58] <WickedWicky> good morning world
[07:13:11] <quasi> morning
[07:13:20] <edwardoc> GPL is too flexible i think
[07:13:22] <Triskelios> pg_user: GPLv3 is tentative right now
[07:13:23] <pg_user> boyd: I want it to be really easy.  creating a zfs snapshot and sending it away is so friggen easy
[07:13:28] <WickedWicky> wow, still here or again quasi?
[07:13:45] <edwardoc> don't know aboutv3
[07:14:03] <quasi> WickedWicky: just got up and about to leave for work
[07:14:08] <pg_user> SUN should license solaris GPL and CDDL but KEEP all the copyrights and have the copyrights assigned to them from submitters.
[07:14:14] <WickedWicky> ahh, ok.
[07:14:17] <WickedWicky> yea same here
[07:14:29] <pg_user> that way we can have all the GPL goodies relicensed as CDDL as well!
[07:14:38] <WickedWicky> first a cup of coffee though.
[07:14:41] <edwardoc> I really don't see whats wrong with BSD ?
[07:14:50] *** richlowe has left #opensolaris
[07:14:55] <edwardoc> license works well
[07:15:08] <pg_user> i agree, the BSD license is the best out there.
[07:15:17] <pg_user> CDDL is basically the BSD license
[07:15:18] <dlg> yes
[07:15:20] <edwardoc> It would be great if the nvid driver was BSD
[07:15:23] <WickedWicky> see ya all later
[07:15:24] <e^ipi> no
[07:15:30] <Triskelios> pg_user: CDDL is much closer to the GPL
[07:15:31] <e^ipi> CDDL is the mozilla license
[07:15:44] <pg_user> how is the CDDL closer to the GPL?
[07:15:49] <Triskelios> right, it's a modified MPL
[07:15:52] <e^ipi> it's a per-file GPL, or IOW the GPL without the viral clause
[07:16:01] *** WickedWicky has quit IRC
[07:16:08] <edwardoc> I try to keep my nose out of law :P, I just use what i like the sound of
[07:16:12] <e^ipi> you /must/ share source, but it's on a file basis, not on a meaningless "project" basis
[07:16:15] <pg_user> I could start a Solaris distro that would be completely closed
[07:16:20] <edwardoc> and what histroy has proven
[07:16:21] <pg_user> and distribute it.
[07:16:32] <Triskelios> pg_user: no, you can't
[07:16:36] <pg_user> I just wouldn' t be able to call it solaris
[07:16:38] <e^ipi> pg_user: no, any CDDL files have to remain CDDL
[07:16:45] <Triskelios> it is still viral
[07:16:48] <e^ipi> and you have to distribute the source on request
[07:16:56] <pg_user> that's not what I've read.
[07:17:01] <e^ipi> you've read wrong then
[07:17:03] <edwardoc> I hate forks
[07:17:04] <Triskelios> pg_user: read the actual licence someday
[07:17:05] <g4lt-mordant> !seen useful license discussion
[07:17:06] <Drone> I've never seen useful license discussion talk in #opensolaris.
[07:17:13] <pg_user> I
[07:17:18] * g4lt-mordant pats drone
[07:17:32] <edwardoc> look at linux distros, look at what happen when compiz got forked and how unstable it is
[07:17:32] <pg_user> read jonathan say that anyone could create a binary only solaris distro
[07:17:34] *** e^ipi is now known as useful_license_d
[07:17:39] <useful_license_d> drats
[07:17:43] *** useful_license_d is now known as e^ipi
[07:17:47] <pg_user> they just couldn't call it solaris .
[07:18:36] <edwardoc> just keep on the same project and fix whats broken/don't like
[07:18:55] <Triskelios> pg_user: link?
[07:19:04] <pg_user> I'm looking for it.
[07:19:07] <pg_user> hold on
[07:19:08] <edwardoc> if something need to go diffrent then have a poll/vote on the opensolaris site, don't fork
[07:19:21] <Triskelios> pg_user: technically, nobody can call their distro Solaris, since that's Sun's trademark...
[07:19:33] <pg_user> I'm a newbie, there's no way I could fork anything
[07:19:57] <edwardoc> any way i am going to leave here and go to somewhere i can download the BFU thing
[07:20:34] <edwardoc> I would like to get Solaris to boot on my PS3 though !
[07:20:45] <pg_user> http://blogs.sun.com/jonathan/date/20050410
[07:20:50] <pg_user> here is the proof
[07:20:51] <edwardoc> Yellow Dog Linux is a Dog
[07:21:02] <edwardoc> Where they kidding or what!?
[07:21:30] <jmcp> edwardoc: I know one of the devs for ydl
[07:22:03] <pg_user> The risk of accidentally using GPL code is very severe
[07:22:18] <pg_user> the CDDL and BSD licenses are a relief
[07:22:19] <Gman> jmcp, having a good giggle at all the chirpy australian accents ;)
[07:22:27] <edwardoc> jmcp:Whats that like?
[07:22:48] <Triskelios> pg_user: I don't see anything about "binary-only" or non-release of source...
[07:22:52] <e^ipi> a blog; even jon's blog; is not "proof". read the license. read any of the numerous analysis of the license
[07:23:08] <edwardoc> What do they even do apart from a recomplie/bug fix for PPC Linux ?
[07:23:17] <e^ipi> the license is quite clear that it is viral. you /MUST/ redistribute source upon request
[07:23:18] <Triskelios> pg_user: please, you've never even read the licence. it's not far removed from the GPL
[07:23:47] <e^ipi> it doesn't spread to files in the same project though, which is the primary difference between CDDL and GPL
[07:23:52] <pg_user> First, we assume all developers (whether nations, universities, manufacturers or software developers) want choice: a developer using OpenSolaris under the CDDL can choose to weave their own IP into an OpenSolaris or derivative distribution without any obligation to deliver their privately developed intellectual property back to Sun or their competitors. Licensees can stand on our shoulders,...
[07:23:54] <pg_user> ...build derivative products based on OpenSolaris, without our hands in their pockets or business plans.
[07:24:01] <jmcp> edwardoc: no idea, just saying that I know one of the developers of it
[07:24:18] <Triskelios> pg_user: that is exactly the same for GPL
[07:24:25] <pg_user> no its not
[07:24:38] <Triskelios> pg_user: you can fork a project and sell it, but you must provide source to the licencees
[07:24:51] <Triskelios> pg_user: sorry, you don't know what you're talking about. how is the GPL different?
[07:25:06] <pg_user> Triskelios: did you read the above?
[07:25:07] <edwardoc> jmcp:Someone and me did a patch that DMA's GRAM over as SWAP so you get 512mb RAN
[07:25:24] <Triskelios> pg_user: yes, I've read the whole entry before
[07:25:25] <pg_user> that was jonathan talking about the cddl
[07:25:26] <edwardoc> s/RAN/RAM/
[07:25:44] <jmcp> edwardoc: using your gpu's attached ram?
[07:26:07] <pg_user> Under the CDDL, you are free to choose what to reveal, what to withhold, and how to price your products. The CDDL encourages self-determination by giving developers the basic building blocks of the entire OpenSolaris operating system without any obligation to disgorge their private property or predetermine its price. You can withhold from Sun, from the OpenSolaris community, from the world,...
[07:26:09] <pg_user> ...anything you build.
[07:26:12] <edwardoc> Its slow so SWAP is the only use for it :P
[07:26:32] <e^ipi> anything /YOU/ build
[07:26:43] <edwardoc> I was trying to get it to work faster but atlass i could not
[07:26:48] <e^ipi> what Sun builds ( the CDDL stuff ) is bound by the license the file was released under
[07:27:01] <e^ipi> and the license it was released under says you have to redistribute the source upon request
[07:27:02] <edwardoc> +The hypervisor is gay
[07:27:03] <Triskelios> pg_user: everything described in that is consistent with both the GPL and the CDDL
[07:27:22] <Triskelios> pg_user: the last two paragraphs you pasted, anyway
[07:27:38] <pg_user> so if i were to release a program that included a header file to solaris code, would I have to release the source?
[07:27:56] <Triskelios> pg_user: only to users who receive the software
[07:28:03] <pg_user> NO!
[07:28:11] <pg_user> read what jonathan wrote!
[07:28:13] <Gman> e^ipi, only the files you modify, files you create you don't have to
[07:28:23] <edwardoc> jmcp:YDL may want to look into a software GPU under Cell if they are not going to open up RSX
[07:28:47] <pg_user> I'm going to read the cddl again.
[07:28:48] <Triskelios> well, CDDL is per-file, so headers don't count (and are usually exempted anyway)
[07:29:03] <edwardoc> jmcp:What are they doing, they seem to be wasting so much time on I don't know what
[07:29:06] <pg_user> this is a lot worse than I thought if you guys are right
[07:29:29] <pg_user> i would puke if solaris was gpl
[07:29:35] <edwardoc> Anyways I better go.
[07:29:35] <Triskelios> but if you use source from say, ON. you would have to licence the files (but only those files) under the CDDL
[07:29:44] <Gman> it's similar to the lgpl, except that you have to release files that are already under the cddl that you modify
[07:29:45] <edwardoc> Thanks again lads talk to you laters
[07:29:52] *** edwardoc has quit IRC
[07:30:16] <pg_user> i'm really going to read the cddl very carefully now.
[07:30:18] <Chipdancer> has anybody here configured samba under opensolaris and is using it with a mac client?
[07:31:02] <palowoda> Here you go guys.  Couple of cheap systems that run opensolaris (if your around Fry's) http://www.fiver.net/misc/solarisamdbox.html
[07:31:10] <pg_user> i was under the impression the cddl was a version of the bsd license...with the only real difference being that any lawsuits would be settled in california
[07:31:13] *** Doc has joined #opensolaris
[07:31:23] <Doc> ok, solaris did not handle that disk dying very well
[07:31:31] <jmcp> Chipdancer: why would you want to use a Mac when you've got OpenSolaris?
[07:31:32] * jmcp trolls
[07:31:43] * Chipdancer smacks jmcp :)
[07:31:44] <jmcp> pg_user: did you read the faq associated with the CDDL?
[07:31:46] <jmcp> ow!
[07:31:54] <pg_user> i just read jon's blog
[07:32:30] <pg_user> and it says "Unlike the GPL, this choice permits participants to take control of their future - to build upon the assets created by Sun and others without any future obligation. The rigidity of the GPL is what renders it incompatible with the CDDL and the Mozilla Public License - both the CDDL and MPL gladly accept the intermingling of any other source code (including GPL). And why didn't we...
[07:32:31] <pg_user> ...use the MPL? For the most part, we did - we simply removed those portions that conveyed to the Mozilla Foundation the ability to revoke the license, and force any disputes to be heard in Santa Clara, California (see Section 11., which obviously doesn't work for a global market)."
[07:32:32] <jmcp> pg_user: do yourself a favour and read http://www.opensolaris.org/os/about/faq/licensing_faq asd well as all the other links under http://www.opensolaris.org/os/about/faq
[07:32:46] <jmcp> pg_user: fffs! PLEASE stop just copy/pasting entries from JS's blog here
[07:32:55] <pg_user> sorry
[07:32:56] *** Doc has quit IRC
[07:33:10] <Chipdancer> jmcp: interesting problem -- it's causing smbd to core dump whereas smbclient from the same client platform works better
[07:33:27] <jmcp> Chipdancer: you've got a core from smbd/nmbd?
[07:33:32] <pg_user> The ability to distribute executables under a different license;
[07:33:33] <Triskelios> wesolows: you fixed anonymous bind a while back, right?
[07:33:41] <Triskelios> er, context being ldap...
[07:34:00] <Chipdancer> jmcp: no, it seems to not actually write it to disk, just claims it does in the logfile
[07:34:14] <jmcp> Chipdancer: what's the output from "coreadm" run as root?
[07:34:57] <pg_user> oh shit.  you guys are right.
[07:34:58] <pg_user> Yes, for any source files that are licensed under the CDDL and any modifications you make. However, you don't need to share the source for your proprietary source files.
[07:35:17] <e^ipi> yes, finally...
[07:36:18] <pg_user> what's to stop someone from creating an interface to a cddl file and then having their propriatery files interact with that interface?
[07:36:32] <e^ipi> nothing
[07:36:34] <e^ipi> it's permitted
[07:36:42] <pg_user> ok, so that is not really an issue.
[07:36:58] <e^ipi> for that matter, you can take a CDDL file, and add it to your project, no need for adding an interface at all
[07:37:05] <e^ipi> you don't need to share your code, only the CDDL code
[07:37:12] <pg_user> so its a non issue
[07:37:57] <pg_user> why not use the BSD license?
[07:38:33] <jmcp> pg_user: (a) patent protection (b) didn't fit with the requirements Sun and Sun Legal had
[07:38:52] *** theRealballchalk has joined #opensolaris
[07:39:04] <pg_user> solaris kernel would make an excellent game console
[07:39:12] <pg_user> os
[07:39:13] <e^ipi> umm...
[07:39:23] <theRealballchalk> hello
[07:39:25] <e^ipi> so, i take it you've never taken an OS design principles class
[07:39:30] <e^ipi> or worked on a games console
[07:39:33] <jmcp> pg_user: and how do you come to that conclusion?
[07:39:59] <pg_user> game consoles are becoming multi-cpu systems
[07:40:13] <jmcp> and?
[07:40:15] <e^ipi> and this has nothing to do with anything
[07:40:46] <pg_user> so are you saying solaris sucks in that area?  What about physics modeling etc?
[07:41:17] <jmcp> pg_user: to me it seems that you're displaying a woeful lack of understanding
[07:41:29] <e^ipi> a games console is like a supercomputer
[07:41:39] <e^ipi> the kernel needs to do one thing
[07:41:46] <e^ipi> dispatch a user-mode process and get the hell out of the way
[07:42:04] <pg_user> not really
[07:42:17] <jmcp> justify your answer
[07:42:20] <pg_user> i developed games in DirectX
[07:42:25] <e^ipi> i give up
[07:42:46] <pg_user> and we used a ton of things in the NT kernel
[07:42:51] <pg_user> networking etc.
[07:43:10] <jmcp> pg_user: and every kernel is the same under the hood, eh?
[07:43:25] <dlg> jmcp: be nice
[07:43:34] <jmcp> dlg: he's pushing it
[07:43:36] <pg_user> I do agree that the NT kernel is more advanced than most
[07:43:47] <pg_user> but solaris is quite good for a unix kernel
[07:43:51] <timeless> jmcp: they're all kernels, aren't they?
[07:44:19] <pg_user> the point is, that for game developers, we are not going to reinvent the wheel.
[07:44:29] <timeless> use SDL
[07:44:35] <pg_user> that is why the playstation 3 is tanking.
[07:44:40] <theRealballchalk> pg_user: what's the differnece between the NT kernel and the solaris kernel?
[07:44:45] <theRealballchalk> i'm just curious
[07:44:49] <pg_user> night and day
[07:44:52] <timeless> bugs
[07:44:58] <pg_user> nt is basically VMS
[07:45:13] <boyd> Well, that's clear then.
[07:45:20] *** oninoshiko has joined #opensolaris
[07:45:35] <timeless> most of us haven't played w/ openvms and don't really know much about it
[07:45:43] <timeless> so your answer isn't helpful
[07:46:12] <jmcp> quite good for a unix kernel?
[07:46:13] <purserj> look at it this way: See a plate of spaghetti? Thats the nt kernel
[07:46:17] * jmcp boggles
[07:46:31] <boyd> purserj: Haha!
[07:46:32] <purserj> Its got twenty year old meatballs from windows 3.1
[07:46:41] <pg_user> jmcp: did you read what ken said about linux?
[07:46:48] <purserj> 15 year old parmasan cheese from win9.x
[07:46:59] <jmcp> pg_user: "ken" ? which "ken" ?
[07:47:07] <timeless> boyd: sorry, i can't /msg, but "heh, nah"
[07:47:07] <pg_user> you know the ken.
[07:47:08] <Triskelios> purserj: ...
[07:47:19] <pg_user> the guy who invented Unix
[07:47:20] <boyd> jmcp: Ken, like "BArbie and Ken"
[07:47:27] <purserj> and a fresh batch of sun dried tomato's from XP
[07:47:44] <purserj> then they've left it out in the sun for a few weeks and called it Vista
[07:47:55] <jmcp> pg_user: oh, Thompson
[07:48:01] <jmcp> no, I haven't read what he wrote about linux
[07:48:34] <pg_user> its pretty bad
[07:48:45] <pg_user> he says its worse than M$
[07:48:53] <jmcp> so what?
[07:49:05] <Jondice> any word on how long it'll be before zfs is a bootable fs?
[07:49:06] <theRealballchalk> pg_user: wait....so how does ps3's kernel contribute to lack of games that's killing it? is it really the kernel?
[07:49:22] <e^ipi> y'all are retarded.
[07:49:38] <e^ipi> Jondice: about a month ago
[07:49:39] <theRealballchalk> e^ipi: hahaha
[07:49:40] <purserj> pg_user: he says whats worse than MS? (Please note the lack of leet $ replacing S there) Linux?
[07:49:40] <boyd> Jondice: negative several weeks
[07:49:48] <Jondice> wow
[07:49:53] <Jondice> that's great =)
[07:49:58] <g4lt-mordant> kn thompson also trojanned more tahn 80% of all computers at one time
[07:50:01] <oninoshiko> jondice, its booting just fine for me
[07:50:12] <boyd> Jondice: Unfortunately nobody told the installer yet
[07:50:35] <theRealballchalk> e^ipi: about a month ago - do you mean nexenta?
[07:50:37] <pg_user> "I view Linux as something that's not Microsoft -- a backlash against Microsoft, no more and no less. I don't think it will be very successful in the long run. I've looked at the source, and there are pieces that are good and pieces that are not. A whole bunch of random people have contributed to this source, and the quality varies drastically.
[07:50:39] <pg_user> My experience and some of my friends' experience is that Linux is quite unreliable. Microsoft is really unreliable but Linux is worse." -Ken
[07:50:42] <Jondice> boyd, hmm, fun
[07:51:04] <e^ipi> theRealballchalk: no, I mean it integrated in to ON about a month ago
[07:51:11] <theRealballchalk> ohhh
[07:51:26] <boyd> pg_user: a) that's clearly old. b) who ever said quality is a determining factor in sales?
[07:51:42] <pg_user> boyd: it still holds true today
[07:51:47] <pg_user> the unstable ABI
[07:52:10] <pg_user> dropping of the odd numbered development branch
[07:52:13] <theRealballchalk> how often does SXDE gets updated?
[07:52:26] <Jondice> if only nvidia open sourced their hardware and drivers, we could make an open solaris sparc nvidia game console to rule the world
[07:52:28] <e^ipi> theRealballchalk: does it matter?
[07:52:33] <e^ipi> SXDE is just an old version of SXCE
[07:52:34] <Jondice> or maybe not
[07:52:38] <Gman> theRealballchalk, supposedly quarterly
[07:52:43] <theRealballchalk> i'm still runing 2/07
[07:52:51] <theRealballchalk> oh ok
[07:53:16] <e^ipi> hey Gman, long time no see
[07:53:52] <pg_user> are there any tutorials on how to write drivers in solaris?
[07:53:55] <Gman> g'day
[07:54:20] <jmcp> pg_user: yes - docs.sun.com, look for "Writing Device Drivers"
[07:54:21] <Gman> http://www.opensolaris.org/os/article/2005-03-31_inside_opensolaris__solaris_driver_programming/
[07:54:22] <Gman> next question :)
[07:54:29] <pg_user> that was fast
[07:54:42] <steleman> how tall is a giraffe ?
[07:54:43] <Gman> i use google.
[07:54:48] <purserj> Gman: whats the air speed of an unladen swallow/
[07:54:49] <purserj> ?
[07:54:59] <timeless> african or europeann?
[07:55:00] <e^ipi> step one might be to familiarize yourself with what precisely an operating system kernel /does/
[07:55:03] <jmcp> should probably include that in the topic for the channel
[07:55:12] <purserj> I don't know... arrrggggh
[07:55:24] <purserj> thank you this has been a Monty Python IRC recreation production
[07:55:24] <boyd> pg_user: There's also a book at docs.sun.com cryptically called "Device Driver Tutorial"
[07:55:38] <purserj> jmcp: which the swallow question?
[07:55:39] <pg_user> the nvidia drivers are really flaky on my box
[07:55:52] <jmcp> purserj: "which year did the swallow question appear in a NSW HSC maths paper?"
[07:55:53] <timeless> Suddenly the Dungeon collapses!! - You die...
[07:55:58] <pg_user> is anyone else here unable to play flash videos in solaris?
[07:56:05] <boyd> That is a feature.
[07:56:08] <pg_user> my pc locks up when I try
[07:56:32] <Gman> works for me, build 66 solaris express
[07:56:35] <palowoda> Which version of flash?
[07:56:40] <pg_user> i have 11/06
[07:56:45] <pg_user> with all the updates
[07:57:04] <pg_user> but it is on x86
[07:57:07] <palowoda> Works ok with opensolaris and the release version of flash9 from Adobe.
[07:57:37] <pg_user> I have flash installed now, just when I try playing a video it locks up everytime
[07:57:51] <palowoda> you already said that. say something new.
[07:58:16] <Triskelios> maybe it's your sound drivers?
[07:58:20] <pg_user> ZFS is really good.  Never lost a file yet with me pulling on the power cord.
[07:58:31] *** Doc has joined #opensolaris
[07:58:33] <pg_user> Triskelios: I have a soundblaster LIVE!
[07:58:37] <Doc> yay to a completely corrupt bootdisk
[07:58:40] <jmcp> I have a soundblaster DEAD!
[07:58:51] * timeless prefers pro dead
[07:58:54] <pg_user> I gave up playing and sound or video on solaris for now
[07:59:02] <e^ipi> I have a soundblaster SCHRODINGER'SCAT!
[07:59:10] <steleman> pg_user: have you tried libcaca ?
[07:59:23] <pg_user> what about libpoop?
[07:59:34] <pg_user> steleman: no
[07:59:46] * jmcp gives up
[08:00:10] <palowoda> pg_user: Your a game developer and have a soundblaster?
[08:00:19] <pg_user> yes
[08:00:30] *** Gman is now known as GmanAFK
[08:00:31] <pg_user> use what the average user is using
[08:00:32] <palowoda> Which games do you work on?
[08:00:36] <GmanAFK> jmcp, good idea...later all
[08:01:39] <pg_user> palowoda: I have a soundblaster on my solaris box.  I dedicated an older machine to solaris
[08:02:06] <palowoda> Don't waste your time with old boxes and opensolaris.
[08:02:18] <pg_user> it would be a waste to use a new machine on solaris.  Solaris runs fine on old hardware.
[08:02:27] <Doc> hi ho.. hi ho..  it's off to backup we go!
[08:03:00] <theRealballchalk> laters Gman
[08:03:01] <pg_user> more often older hardware is supported better
[08:03:17] <pg_user> in GNU/Linux / Solaris/Freebasd
[08:03:31] <pg_user> good night everyone
[08:03:49] <theRealballchalk> g'nite pg_user
[08:04:16] *** pg_user has quit IRC
[08:04:40] <palowoda> I really wish opensolaris supported old hardware but we all know it isn't ture.
[08:04:45] <palowoda> true.
[08:05:15] <Jondice> boyd, I suppose I could install the system to a USB hard disk (if the installer supports that)
[08:05:38] <palowoda> Too slow on usb disks.
[08:05:39] <Jondice> boyd, then mirror the installation once the zpool has been created
[08:05:59] *** Chip has joined #opensolaris
[08:06:21] <Jondice> palowoda, even just as a temporary install?
[08:06:46] <boyd> Maybe... I haven't done that.
[08:07:08] <palowoda> What your out of partitions? Or internal hard disks?
[08:07:09] <Jondice> i'm wondering what the easiest way to do a zfs root install would be
[08:07:36] *** Chipdancer has quit IRC
[08:07:49] <Jondice> i was thinking about getting a sun fire 4200n, and use zraid on all 4 disks
[08:07:51] *** Chip is now known as Chipdancer
[08:08:06] <Jondice> *x4200
[08:08:15] *** bobbyz has joined #opensolaris
[08:08:23] <palowoda> err I didn't think raidz was supported with zfs boot.
[08:08:46] <Jondice> ah, I see
[08:09:01] <Jondice> well, in that case, I guess it doesn't matter if zfs is bootable for me or not
[08:09:14] <palowoda> Now you have lots of bootable disks.
[08:10:50] <Jondice> is it possible to stick more than 4 hard drives in a 4200? not necessarily under a SAS controller
[08:11:16] <palowoda> Sure get a SATA controller up to 24 disks.
[08:11:37] <Jondice> I'm just wondering about the chassis itself
[08:11:52] <g4lt-mordant> palowoda, for 24 disks, you can get a sunswift FFS
[08:12:25] <Jondice> i'm trying to be as cheap as possible ... and buy from sun.
[08:12:26] <palowoda> g4lt-mordant: How much was that?
[08:12:42] <palowoda> How cheap is cheap?
[08:12:47] <g4lt-mordant> palowoda, I just spent like $0 on one a few months ago
[08:12:54] <g4lt-mordant> $20 even
[08:13:06] <palowoda> g4lt-mordant: Good for you now give it to me.
[08:13:36] <Jondice> I'm trying to build a file server for $3-4K
[08:13:55] <palowoda> How many terabytes?
[08:14:05] <g4lt-mordant> palowoda, why, so you cna put it in a craptacular X86?  I'd ratehr melt it for the metal
[08:14:48] <palowoda> g4lt-mordant: You got it for free, I hope you don't expect me to pay the shipping. :)
[08:15:20] <g4lt-mordant> no, I expect you to die, mr bond
[08:15:30] <palowoda> :)
[08:15:50] <bobbyz> I keep getting
[08:16:03] <Jondice> palowoda, 1TB of usable space would be great for our needs
[08:16:16] <Jondice> polowoda, looking like i'll want to get the disks from someone other than sun though
[08:17:50] <bobbyz> I keep getting "nfs mount: odin:/mnt/migrate: access denied" when trying to mount an NFS vers 3 share from a linux server.  I'm exporting the share as "/mnt/migrate 10.12.63.*(ro,insecure_locks,async,all_squash,nohide,anonuid=10000,anongid=100)" on the linux box and the solaris box I'm trying to mount with has the ip 10.12.53.3.  Does anyone have any idea what might be causing this?
[08:17:57] <palowoda> Jondice: So you priced out a basic Ultra20?
[08:18:27] <bobbyz> It mounts fine with other linux client boxes on the same 10.12.63.* network
[08:18:28] <rpm> bobbyz, probably incompatible nfs versions
[08:18:58] <boyd> there's an option on linux to allow client ports < 1024 I think it's called "insecure"
[08:19:09] <boyd> sorry *>* 1024
[08:19:41] <rpm> set NFS_CLIENT_MAXVERS to 3 in /etc/default/nfs.
[08:20:02] <boyd> I would be surprised if you get an access denied error from the wrong version
[08:21:17] <bobbyz> rpm: still the same problem.  Does the "-o vers=3" command line option do the same thing?  I was trying that too with no luck
[08:21:38] <rpm> bobbyz, yes.
[08:21:38] <bobbyz> boyd: I'll try enabling the insecure option then
[08:21:39] <Jondice> palowoda, that might be good, only real problem is it only has 2 disk bays
[08:21:47] *** nostoi has joined #opensolaris
[08:22:31] <bobbyz> no luck.  This is weird
[08:23:02] <palowoda> Jondice: Move up to an Ultra 40.
[08:23:20] <boyd> I think I found something in the logs on the server. But it's been about 2 years..
[08:24:51] <Jondice> palowoda, well said
[08:27:08] <bobbyz> well that was weird.  I think it was choking on the 'async' option from the linux server
[08:27:10] <palowoda> Hey I'm a cheap bastard. http://www.fiver.net/misc/solarisamdbox.html but I'm not going to complain anyone purchasing Sun equipment.
[08:27:13] <bobbyz> oh well,  it's working now
[08:27:42] * boyd wonders what "async" does on linux
[08:28:08] <moazamraja> holy shit that url is slow
[08:28:22] <moazamraja> that image is like being raytraced in real time on a 486dx2
[08:28:25] <moazamraja> <-- mean.
[08:28:52] <palowoda> That is because someone is beating me up.  Hang on.
[08:29:04] <boyd> Ah! it's an "I don't care about my data, please throw it away at your leisure" option
[08:29:04] <bobbyz> "async -- This option allows the NFS server to violate the NFS protocol and reply to  requests  before any changes made by that request have been committed to stable storage (e.g. disc drive)."
[08:29:29] <boyd> Maybe solaris was turning away in disgust.
[08:29:35] <bobbyz> well it's a /tmp type of mount, so that's why it was enabled  :)
[08:29:50] <bobbyz> not actually /tmp, but not important data
[08:31:20] <bobbyz> I wish I would have remembered the bit about it violating the protocol.  It has been a while since I'd touched that exports file.  I bet the solaris implementation adheres strictly to the protocol
[08:31:32] *** slowhog has left #opensolaris
[08:32:12] *** richlowe has joined #opensolaris
[08:32:17] *** estibi has quit IRC
[08:32:29] *** bengtf has quit IRC
[08:33:19] <networkdump> palowoda: slow on my end too
[08:33:59] <moazamraja> i want a new workstation in a SparcStation 20/Pizzabox style case
[08:34:02] <moazamraja> that'd be badass
[08:34:10] <moazamraja> i'm sick of desktop towers
[08:35:38] <palowoda> Hmm I didn't know you could kill a sftp session.
[08:35:45] <palowoda> couldn't
[08:36:14] <g4lt-mordant> root can kill ANYTHING
[08:36:54] <palowoda> I would have thought so but in anycase try it again.
[08:37:07] *** CIA-17 has quit IRC
[08:37:14] *** CIA-21 has joined #opensolaris
[08:40:12] *** Cyrille has joined #opensolaris
[08:41:38] *** WickedWicky has joined #opensolaris
[08:41:41] <WickedWicky> morning all
[08:41:49] <palowoda> moazamraja:  There is a high demand for T2 Workstations in the consumer market right now.  You might get your wish.
[08:42:26] * g4lt-mordant waits for the coolthreads Tadpole with bated breath
[08:42:41] <moazamraja> palowoda: i doubt it'll be pizzabox tho, even if they were to make it
[08:42:50] <moazamraja> and...I doubt they'll make a T2 based workstation
[08:42:50] <palowoda> New ones with Nvidia cards and all for 1999.95.
[08:43:01] <moazamraja> palod: my ass.
[08:43:08] <moazamraja> (politely said, of course)
[08:43:18] <WickedWicky> that's quite an expensive ass
[08:43:21] <palowoda> Yeah I know it will never happen.
[08:43:26] <moazamraja> T2 with nvidia, and some RAM, for $2k? not gonna happen.
[08:43:46] <g4lt-mordant> first, theere  is no nvidiot support for sparc
[08:43:49] <e^ipi> demand is an economics term with a well-defined meaning
[08:44:16] <palowoda> First there are no massive amounts of desktop users on utlrasparcs.
[08:44:22] <e^ipi> people may /want/ the machines
[08:44:36] <e^ipi> but unless they have the cash to back it up, there's no demand
[08:44:56] <palowoda> With Sun distribution channels?
[08:44:59] *** tsoome has quit IRC
[08:47:02] <palowoda> I don't know if you could get enough T2 desktops out in the market via one website to purchase them from to increase the numbers.
[08:49:24] <palowoda> Although if they wanted to put a T2 into a settop box, I'd be ripping it apart and altering it to make a desktop in two seconds.
[08:50:54] <WickedWicky> imagine Kreatel using T2 processors for their iDTV settopboxes
[08:50:58] <WickedWicky> *sigh*
[08:52:03] <palowoda> Hey they can put a cell in a PS3.
[08:53:23] <WickedWicky> I know. I just dont see kreatel (which is owned by motorolla) switch from PPC any time soon
[08:53:24] <moazamraja> a T2 in a settop...yeah, ok
[08:53:31] <moazamraja> that'd be one mega expensive 'set top' box
[08:53:56] <WickedWicky> depends, when you have a mass buyer it might be doable
[08:54:11] <palowoda> moazamraja:  Suggest a way to get the volume up.
[08:54:18] *** bengtf__ has joined #opensolaris
[08:54:18] <WickedWicky> taking in account you want every customer to have the same STB and you have 300.000 customers
[08:54:19] *** bengtf__ is now known as bengtf
[08:54:30] <moazamraja> not. gonna. happen.
[08:54:41] <WickedWicky> you're not much of an optimist, are you?
[08:54:50] <libkeiser> yeah, but then you'd have to rewrite your set top code to leverage the parallelism of the T2.
[08:54:55] <libkeiser> not gonna happen
[08:55:05] <WickedWicky> so
[08:55:13] <WickedWicky> didnt they have to do that for Cell either?
[08:55:25] <moazamraja> wtf would u need a settop box to be so parallel for?
[08:56:01] <WickedWicky> picture in picture support and recording a iDTV program at the same time
[08:56:08] <e^ipi> moazamraja: e-penis
[08:56:28] <WickedWicky> multiple tvs connected to one STB without needing a seperate STB for every TV in your house
[08:56:43] <libkeiser> sigh. T2 is not the right architecture for that type of thing, but whatever
[08:56:58] <palowoda> Just need a way to get the volume up on the T2.
[08:57:34] *** GmanAFK has quit IRC
[08:57:43] <WickedWicky> I dont think someone would say Solaris would be used as user desktop a couple of years ago either
[08:58:05] <WickedWicky> and honestly,I dont think Sun Microsystems will really care what their T2 is used for, aslong as it selsls
[08:58:08] <WickedWicky> sells too
[08:58:18] *** Doc has quit IRC
[08:58:19] <libkeiser> WickedWicky: this has nothing to do with imagining possibilities, and everything to do with understanding what hardware microarchitectures excel at, and what types of workloads they suck at executing
[08:58:29] <moazamraja> WickedWicky: sorry, but u can do that with tons of much cheaper processors which are out on the market right now
[08:58:50] <moazamraja> i love the T1/T2 archs, but....this stuff ain't gonna happen, it'd be silly
[08:59:00] *** Doc has joined #opensolaris
[08:59:09] <WickedWicky> yea? then why is there no STB available in the entire of europe that can decode more than two mpeg streams?
[08:59:36] <libkeiser> what the hell does that have to do with the fact that the T2 would be a bad fit?
[08:59:42] <palowoda> Ok T2=8999.95 in single quanity.  Discounts available.
[09:00:12] <WickedWicky> nothing libkeiser, but as palowoda pointed out, it doesnt hurt to look for posibilites to enlarge volume
[09:00:26] <CIA-21> sp92102: 6588379 brand interpose handlers stick around if a branded process dumps core
[09:00:28] <moazamraja> then look towards network switches
[09:00:35] <moazamraja> nas appliances (large ones)
[09:00:47] <libkeiser> now, those are places i'd love to see T2
[09:00:48] <moazamraja> and guess what? other vendors are not going to use a SUN chip to power their appliance
[09:00:57] <moazamraja> simple fact is, Sun will have to do it themselves
[09:01:16] <palowoda> moazamraja: bingo
[09:01:55] <moazamraja> getting some other company to showoff your product, not gonna happen
[09:02:15] <moazamraja> Apple tried that with hardware, and they tried it with stores (compusa/best buy, etc.) and it failed
[09:02:32] <moazamraja> apple finally had to take everything in house and show everyone how it's supposed to be done
[09:02:39] <moazamraja> and guess what? apple reaped the profits
[09:02:54] <palowoda> But apple is a desktop.
[09:03:11] <moazamraja> palowoda: it's a concept, don't take it 100% literally as the same as apple
[09:03:31] <moazamraja> apple took control of their sales channel cuz they saw retailers were doing a horrible job
[09:04:09] <palowoda> apple toke control of their server products and didn't get anywhere either.
[09:04:20] *** tsoome has joined #opensolaris
[09:04:25] <moazamraja> no, they didnt take control of their server products.
[09:04:31] <oninoshiko> actually everyone undercut apply and they brought in back inhouse, and contenued selling poorly, untill they made them up to look like fisher price
[09:04:45] <moazamraja> apple is exactly where it wants to be with it's server products, they are NOT concentrating on them, on purpose
[09:04:48] <palowoda> with lead paint
[09:04:53] <oninoshiko> apple*
[09:05:02] <moazamraja> oninoshiko: not true, but whatever
[09:05:13] <moazamraja> this whole thing with "ppl buy it cuz it looks pretty" is b.s.
[09:05:17] <moazamraja> that only lasts so long
[09:05:23] <moazamraja> but, again, whatever
[09:05:49] <moazamraja> getting back to the point
[09:06:02] <moazamraja> sun needs to control it's own destiny with it's new processors
[09:06:08] <moazamraja> oems/partners won't do squat
[09:06:33] <moazamraja> its, not it's :P
[09:06:49] * tsp didn't think malloc() would crash, but it did...
[09:06:52] <palowoda> moazamraja: You think Sun will follow the traditional pricing of the T2 compared to T1?
[09:07:07] <moazamraja> i have no idea about T2 pricing
[09:07:32] <richlowe> malloc would (rightfully) get angry if the allocator's metadata got clobbered.
[09:07:35] <richlowe> or similar.
[09:08:22] <tsp> hmm
[09:08:48] *** mikefut has joined #opensolaris
[09:08:53] <tsp> I don't even know where to start learning how to get my system not to die the moment someone types perl -e'fork while 1' :)
[09:09:27] <oninoshiko> tsp, dont type that
[09:10:03] <oninoshiko> oh... i misread that
[09:10:04] <tsp> oninoshiko: that's a start, but I don't like the fact that a normal user can bring down a box that easily
[09:12:59] *** phimic has joined #opensolaris
[09:14:50] <e^ipi> tsp: man ulimit
[09:18:41] <oninoshiko> you can also use containers, if you so desire
[09:18:52] *** pablomh has joined #OpenSolaris
[09:20:36] *** palowoda has quit IRC
[09:21:21] *** oninoshiko has quit IRC
[09:21:26] *** jfndi has joined #opensolaris
[09:23:06] *** Snake007uk has joined #opensolaris
[09:25:08] *** palowoda has joined #opensolaris
[09:25:40] <tsp> interesting
[09:29:26] *** theRealballchalk has quit IRC
[09:33:39] *** estibi has joined #opensolaris
[09:37:58] *** tsoome has quit IRC
[09:40:03] *** MattMan has joined #opensolaris
[09:40:27] <palowoda> Wow looks like ksh93 might make it to build 72. http://bugs.opensolaris.org/view_bug.do?bug_id=6437624   Love the Introduced field: Solaris_2.3.
[09:41:19] *** phips|macbook has joined #opensolaris
[09:41:31] <moazamraja> Mr. Phipps
[09:41:46] *** phips|macbook is now known as phips|mb
[09:42:02] *** tsoome has joined #opensolaris
[09:42:14] <moazamraja> phips: Flip is great, give him a raise! ;)
[09:42:20] <moazamraja> fine fine, he paid me to say that.
[09:42:21] <moazamraja> or not.
[09:42:51] *** JonathanW has quit IRC
[09:44:51] <phips|mb> moazamraja: I am probably not the person you think I am, since I have no idea what you're talking about :) Mornin anyway!
[09:45:34] <moazamraja> yeah, i just figured that out :)
[09:46:05] <sbahra> richlowe, unless you pet that metadata to your liking ;-p
[09:48:31] *** JonathanW has joined #opensolaris
[09:53:34] *** sneaker is now known as nightswim
[09:55:26] *** nikinana is now known as niki|installON7`
[09:55:41] *** niki|installON7` is now known as niki|install_ON7
[09:56:05] *** niki|install_ON7 is now known as niki|installON71
[10:00:11] *** dmarker has quit IRC
[10:00:22] *** dmarker has joined #opensolaris
[10:16:36] *** Gman has joined #opensolaris
[10:17:42] <Gman> anyone happen to know the bugtraq bash cat?
[10:19:58] <seanmcg> utility/bash
[10:21:09] *** Pietro_S has joined #opensolaris
[10:21:24] <Gman> seanmcg, ta
[10:21:27] <richlowe> seanmcg: yay consistency
[10:21:29] * Gman managed to make bash crap itself
[10:21:34] <Gman> first time in a long, long, long time
[10:21:40] <richlowe> 'cos you know, solaris/shell/bash would be way too... sensible.
[10:21:59] <WickedWicky> where's the fun in sense?
[10:22:27] <Gman> richlowe, seriously.
[10:22:28] <Gman> gar.
[10:22:30] <richlowe> so the shells SFW deliver are in solaris/utility, and the shells ON delivers (or soon will) are in solaris/shell
[10:22:47] <Gman> richlowe, most of solaris is bollox
[10:22:54] <Gman> solaris/
[10:22:58] <richlowe> I figured that was just JDS
[10:23:01] <richlowe> Oh, you meant the bug cat. ;)
[10:23:05] * Gman prides himself on re-inventing jds cats
[10:23:17] <Gman> i actually sat down to figure out what did and didn't make sense
[10:23:22] <Gman> and re-categorized :)
[10:23:58] <seanmcg> did you have one servicedesk call per re-cat ?-)
[10:24:44] * Gman actually sat down and thought about it some
[10:24:48] <Gman> but yeah
[10:24:54] <Gman> i ran a shit load of scripts
[10:24:56] <Gman> caused a lot of pain
[10:25:07] <Gman> and learned to use bt
[10:25:32] <richlowe> So, it's one of the things I'd like to fix with the new DTS
[10:25:41] <richlowe> because some of this is just completely bloody stupid
[10:30:39] *** sioraiocht has quit IRC
[10:31:52] <Gman> 6594123: bash had a fit and dumped core.
[10:32:08] <richlowe> Yeah, paste a bug none of us will be able to see yet.
[10:32:11] <richlowe> that'll help :)
[10:32:21] <richlowe> (seanmcg doesn't count) :)
[10:32:24] <Gman> huh, seemingly bugtraq crapped itself too, and created multiple bugs (6594122 and 6594123)
[10:32:27] <Gman> heh
[10:32:40] <richlowe> I'd blame you.
[10:32:41] <richlowe> :)
[10:32:42] <seanmcg> clicky fingers
[10:32:54] <richlowe> close the latter duplicate of the former, and look suitably ashamed. :)
[10:33:14] <richlowe> it's this bloody nscd crash that's bugging me.
[10:33:19] <Gman> http://pastebin.ca/660768
[10:33:50] * Gman goes for dinner instead
[10:33:53] <richlowe> Gman: I assume you attached the core to the bug?
[10:33:53] *** Gman is now known as GmanAFK
[10:33:56] <GmanAFK> richlowe, yep
[10:33:59] <Pietro_S> hmm, why man gcc doesn't work? Does I need to fidle with MAN_PATH?
[10:34:01] <richlowe> GmanAFK: k
[10:34:23] <Triskelios> if anyone cares, we finally got anonymous bind + self-cred auth working for ldap on our machines seems CR 6526896 was fixed in snv_63
[10:34:57] <Triskelios> (it was really aggravating earlier because 6526896 is not public)
[10:35:35] <richlowe> I suspect that's in a cat that's not shown (in error)
[10:35:39] <richlowe> bug website-discuss about it.
[10:36:32] *** Murmuria has joined #opensolaris
[10:36:38] *** migi_ has joined #opensolaris
[10:36:54] *** niki|installON71 has quit IRC
[10:38:39] <GmanAFK> richlowe: http://www.gnome.org/~gman/bash-core
[10:38:42] * GmanAFK really goes
[10:39:25] *** Cyrille has quit IRC
[10:41:05] *** tsoome has quit IRC
[10:48:15] *** BatonT has joined #opensolaris
[10:56:16] *** alanbu1 has joined #opensolaris
[10:59:00] <Pietro_S> sigc++ use SIGC_HAVE_SUN_REVERSE_ITERATOR, which 'should' allow compilation by studio, but when studio use stlport4, it fails because of this compatibility :-(
[10:59:45] <Pietro_S> the problem is that I can't find where this preprocessor variable is defined ...
[11:00:35] <moazamraja> re
[11:00:48] <moazamraja> anyone using Solaris Container Manager?
[11:00:53] *** Cyrille has joined #opensolaris
[11:01:11] * WickedWicky is off to run Explorer on 24 machines
[11:02:08] *** bnitz has joined #opensolaris
[11:02:17] *** tsoome has joined #opensolaris
[11:02:57] *** coraline has quit IRC
[11:03:53] *** coraline has joined #OpenSolaris
[11:06:11] *** Murmuria has left #opensolaris
[11:07:00] <Triskelios> Pietro_S: they use a standard auto* tools build system so it's in  sigc++config.h generated by configure
[11:08:58] <quasi> WickedWicky: you don't do that every week from cron?
[11:09:57] <Triskelios> Pietro_S: SFE (pkgbuild.sf.net) has a package for sigc++ (SFEsigcpp), so it definitely builds with studio
[11:10:26] <Triskelios> oh, they're not using stlport though, nevermind
[11:10:55] <Pietro_S> Triskelios: I installed the SFE package
[11:11:24] <Pietro_S> but the sf which use sigc++ doesn't build, because of this option
[11:11:48] <Pietro_S> from SFEsigcpp.spec stlport4 is used to compile it
[11:12:01] <WickedWicky> quasi: Siemens didnt install SUNWexplo on these machines
[11:13:26] *** coraline has quit IRC
[11:14:44] *** coraline has joined #OpenSolaris
[11:17:57] <quasi> WickedWicky: incompetents ;)
[11:18:21] <WickedWicky> let's say I find room for improvement
[11:18:32] <Triskelios> Pietro_S: when I add -library=stlport4 -staticlib=stlport4 to CXXFLAGS, it does not define SIGC_HAVE_SUN_REVERSE_ITERATOR
[11:18:40] <Triskelios> (Studio11)
[11:18:44] *** Vanuatoo has quit IRC
[11:20:56] <Pietro_S> Studio12: with this flags (taken from SFE)  CXXFLAGS="%cxx_optflags -library=stlport4 -features=tmplife -features=tmplrefstatic
[11:21:14] <Pietro_S> defined SIGC_HAVE_SUN_REVERSE_ITERATOR
[11:21:23] *** KermitTheFragger has joined #opensolaris
[11:22:07] *** tsoome has quit IRC
[11:23:05] *** halton has quit IRC
[11:24:09] <Triskelios> I'll check with Studio12 here
[11:26:15] <Pietro_S> I added -staticlib=stlport4 to see what happend
[11:26:51] <Triskelios> that just changes how it links (so it won't depend Studio internal libs at rruntime)
[11:27:28] <Pietro_S> it doesn't compile with -staticlib :-(
[11:27:48] <Triskelios> really?
[11:27:59] <Triskelios> it just finished here
[11:28:09] <Triskelios> had to patch one of the tests
[11:28:24] <Pietro_S> pkgbuild: "test_copy_invalid_slot.cc", line 22: Error: The function "malloc" mus
[11:28:27] <Pietro_S> t have a prototype.
[11:28:39] <Triskelios> hold on, I'll commit the patch
[11:29:09] <Pietro_S> oki
[11:30:23] <Triskelios> you really need stlport4 support?
[11:31:06] <Pietro_S> yes, without it lot's of error jump over me
[11:36:19] <Triskelios> svn up, please
[11:37:18] <Pietro_S> svn uped, compiling
[11:37:51] *** tsoome has joined #opensolaris
[11:42:22] <Pietro_S> Triskelios: passed and that option isn't defined so it works, thanks
[11:42:58] *** MattMan is now known as MattAFC
[11:43:32] <Triskelios> np
[11:45:51] *** Tpent1 has joined #opensolaris
[11:46:46] <Triskelios> Pietro_S: hold on, the library is linked incorrectly =\ - need to set LDFLAGS also...
[11:47:14] *** aruiz has joined #opensolaris
[11:47:34] *** aruiz has left #opensolaris
[11:54:26] *** Tpenta has quit IRC
[11:57:17] <boyd> I see that Sun Marketing are demonstrating that they can't do maths again
[11:57:50] <boyd> "Triple your Capacity" "3 Sunfire X4600 servers for the price of 2"
[11:57:52] <Cyrille> I thought marketing was explicitly forbidden from doing that (math).
[11:58:26] <boyd> Well, they're not doing it :)
[12:00:09] *** derchris has quit IRC
[12:00:14] *** derchris has joined #opensolaris
[12:00:26] <Cyrille> I remember (in another company) the salespeople written document which bid on a customer contract was somehow stating (albeit in a cleverly disguised way) that 7 * 3 = 10. We wondered once we had to do the project why we ended up being twice as slow as expected until I found out that paragraph...
[12:01:39] <boyd> Heh
[12:02:11] *** NikolaVeber has joined #opensolaris
[12:04:17] <Triskelios> Pietro_S: I'm going to have to drop stlport4 from the build, adding it to LDFLAGS screws up the configure tests (which are done with cc not CC)
[12:05:45] *** yongsun has quit IRC
[12:06:22] <Pietro_S> Triskelios: oops, what about to exclude -staticlib=stlport4 and patch sigc++config.h?
[12:06:55] *** bobbyz is now known as bobbyz_zzz
[12:07:03] <Triskelios> well, it builds correctly for me without stlport without changing anything...
[12:07:19] <Triskelios> in both Studio 11 and 12...
[12:08:58] <Pietro_S> I think that laca had reason why to add stlport there ...
[12:09:19] *** simford has quit IRC
[12:10:32] <Triskelios> it wasn't using stlport in the last rev
[12:13:47] <Pietro_S> it was - there was -library=stlport4 in CXXFLAGS
[12:15:17] <Triskelios> I checked the history, don't see it
[12:18:15] <Pietro_S> well then I will try to compile it without stlport ... :-(
[12:19:34] <Triskelios> just set CXXFLAGS="%optflags" and nothing else, I don't think you need to change the temp scoping rules either
[12:20:33] *** karrotx_ has quit IRC
[12:20:52] *** wisnios has joined #opensolaris
[12:23:03] *** MattAFC is now known as MattMan
[12:23:30] <Pietro_S> Triskelios: that surely didn't come from my head, because I don't knwo what it means! Did you checked right history?
[12:24:06] <Triskelios> svn log SFEsigcpp.spec and diff with the other 2 revisions (179 and 5)
[12:24:21] *** calumb has joined #opensolaris
[12:29:12] <Pietro_S> Triskelios: it's strange because changelog doesn't match
[12:31:27] *** bengtf has quit IRC
[12:32:51] *** purserj_ has joined #opensolaris
[12:33:12] <Pietro_S> Triskelios: svn up - PREV, proved that you are ight, stlpot wasn't used there at all
[12:33:15] <Pietro_S> shame on me
[12:33:20] *** purserj_ is now known as purserj[sol]
[12:33:38] *** Tpenta has joined #opensolaris
[12:34:12] <Pietro_S> the question is can I link library without stlport with program with stlprot?
[12:35:49] <seanmcg> Seems dylan could do with a point and shoot camera..
[12:35:58] <seanmcg> oops wrong channel.
[12:36:07] <Triskelios> my guess is they use different namespaces for the symbol names at least, so it might be okay
[12:40:16] *** Hystrix has joined #opensolaris
[12:41:23] *** Tpent1 has quit IRC
[12:42:03] *** alanbu1 has left #opensolaris
[12:42:27] *** dunc has quit IRC
[12:43:20] *** MattMan has quit IRC
[12:45:08] *** vmlemon has joined #opensolaris
[12:51:35] <trochej> http://www.flickr.com/photos/mala_mi/475686018/
[12:53:03] *** MattMan has joined #opensolaris
[12:54:50] *** bengtf__ has joined #opensolaris
[12:54:52] *** bengtf__ is now known as bengtf
[12:57:00] *** reinhold has joined #opensolaris
[13:02:08] <Pietro_S> is there any way how to say sunCC to undefine preprocessor variable?
[13:05:13] *** deather_ has joined #opensolaris
[13:05:21] <Cyrille> that's standard preprocessor command line instruction, I'd imagine Sun CC will accept it
[13:07:05] *** Triskelios has quit IRC
[13:12:47] *** vmlemon has quit IRC
[13:13:17] *** nostoi has quit IRC
[13:19:55] *** vmlemon has joined #opensolaris
[13:22:33] *** deather has quit IRC
[13:24:40] <WickedWicky> !kpn@1
[13:24:53] <WickedWicky> eep
[13:25:26] *** bunker has joined #opensolaris
[13:26:58] <boyd> Has anyone seen any announcements from Sun about Xen since this Citrix thing?
[13:28:30] <WickedWicky> naw, the only thing i got was an email from XenSource
[13:33:28] *** Hystri1 has joined #opensolaris
[13:35:13] *** Hystrix has quit IRC
[13:35:20] *** estibi_ has joined #opensolaris
[13:39:07] *** estibi has quit IRC
[13:45:43] *** wisnios has quit IRC
[13:46:35] <Berny> blogs.sun.com/mrj/entry/the_latest_solaris_on_xen was the last xen related thing i've seen
[13:50:10] * boyd sleeps
[13:56:15] *** reinhold is now known as reinhold_away
[13:57:15] *** Kush- has quit IRC
[13:57:18] *** DNSOUTH has joined #opensolaris
[13:58:34] *** DNSOUTH has quit IRC
[14:00:56] *** vmlemon has quit IRC
[14:01:27] *** vmlemon has joined #opensolaris
[14:01:30] *** laca has joined #opensolaris
[14:01:48] <Pietro_S> fof the hell: what this should mean "&::random" - advice part of C++ of code
[14:02:35] <trygvis> a reference to the global variable random
[14:03:17] <Cyrille> getting a pointer to the "random" variable which is in the global namespace.
[14:03:22] <Pietro_S> in current namespace or in global namespace?
[14:04:26] <trygvis> global, that's the "::" part
[14:05:05] *** Darth_Wader has joined #opensolaris
[14:06:38] *** yarihm has joined #OpenSolaris
[14:08:27] *** Vanuatoo has joined #opensolaris
[14:09:04] <sle> this is very strange
[14:09:25] <sle> i formatted a disk connected via a PATA to USB adapter with ZFS
[14:09:33] <Pietro_S> thanks, I think that soon I'll have nigthmares about C++
[14:09:41] <sle> and now it blocks boot on one of our supermicro machines
[14:09:57] <sle> just hangs at the POST screen
[14:10:18] <oxygene> sle: is there a usable mbr+partition table on it?
[14:10:50] *** master_o1_master has joined #opensolaris
[14:10:58] *** trede has quit IRC
[14:12:19] *** JonathanW is now known as JWheeler
[14:12:27] <sle> no, i don't think so
[14:12:50] <sle> i only noticed this problem after making it into a single disk ZFS pool
[14:13:12] <sle> i've disabled all other boot items, and it still hangs
[14:14:02] <trygvis> I think that is a know problem with some BIOSes not not being able to read the label that ZFS puts on it
[14:14:06] <trygvis> try google
[14:15:57] <sle> oh, so the EFI label is the problem
[14:16:13] * sle prays for a bios update
[14:22:53] *** master_of_master has quit IRC
[14:28:35] *** calumb is now known as calLNCH
[14:30:33] <g4lt-mordant> Pietro_S, nightmares about C++ is redundant
[14:31:50] *** SYS64738 has joined #opensolaris
[14:31:54] <SYS64738> hi
[14:33:50] *** nachox has joined #opensolaris
[14:34:55] <nachox> morning
[14:37:12] *** Gekkko[PDA] has joined #opensolaris
[14:37:41] <Gekkko[PDA]> hey
[14:38:11] *** reinhold_away has quit IRC
[14:38:23] *** trede has joined #opensolaris
[14:39:47] *** hohum has quit IRC
[14:52:47] *** nachox has quit IRC
[14:53:10] *** Gekkko[PDA] has quit IRC
[15:01:11] <splunk> morning
[15:02:34] *** SYS64738 has quit IRC
[15:03:39] *** mritun_ has joined #opensolaris
[15:03:59] <mritun_> Hello folks
[15:04:04] <mritun_> when is b70 arriving ?
[15:04:53] <PerterB> "real soon now"
[15:05:16] <mritun_> :(
[15:05:31] <mritun_> there used to be a page somewhere with schedules listed in
[15:05:51] <mritun_> I guess it got canned after 2 week delay around new year
[15:06:17] *** sioraiocht has joined #opensolaris
[15:06:23] <mritun_> today is friday, I think the "build day" right ?
[15:06:24] <PerterB> you mean like http://www.opensolaris.org/os/community/on/onnv_schedule.txt
[15:06:50] <PerterB> obviously thngs slipped since that was written ;)
[15:08:35] <mritun_> did gate close happen yet ?
[15:08:41] *** pablomh has quit IRC
[15:08:49] <mritun_> if yes, then we should be able to see it on monday, right ?
[15:09:30] <PerterB> no idea... but 70 was being targetted as a "more stable" release so I wouldn't be surprised if they do extra testing
[15:09:44] <mritun_> hmm
[15:10:33] <mritun_> also since b70 installer would be based on indiana, and will NOT upgrade b69, I'm just trying to hang around before installing natively on my laptop
[15:10:51] <mritun_> right have have to do with vmware :-/
[15:12:08] <PerterB> based on Caiman, not Indiana but still... even if it doesn't do upgrades (and I thought I saw in the requirements that it did), Liveupgrade should still work, surely?
[15:13:21] <mritun_> lu ! on a developer machine... nope
[15:13:31] <mritun_> just too much waste of space
[15:14:10] *** phimic has quit IRC
[15:14:29] <mritun_> the reason why I refuses to do upgrade is because previous installers install SUNWCall, and this one installs SUNWCXall
[15:16:49] <PerterB> well, guess you'll just have to wait then.
[15:17:45] <mritun_> hmm yes
[15:17:46] <Pietro_S> mritun_: so just because some packages that you have for nothing, you make new install which configuring your system?
[15:18:19] <mritun_> Pietro_S: problem is not reinstalling
[15:18:29] <mritun_> problem is reconfiguring the whole bunch
[15:18:59] <mritun_> esp since I'm on a low bandwidth connection... it takes whole weekend to download & install the thing
[15:19:37] * mritun_ has lowly 256k conn
[15:19:59] *** karrotx_ has joined #opensolaris
[15:20:10] <mritun_> but it's *totally* unlimited
[15:20:17] *** Kbuck70 has joined #opensolaris
[15:21:05] *** karrotx_ has quit IRC
[15:21:12] *** karrotx_ has joined #opensolaris
[15:22:18] <mritun_> is there any specific reason why Studio 11 is preferred over 12 for building ON and stuff ?
[15:23:16] <oxygene> mritun_: because ON relies on the presence of studio 11 bugs and the absense of the studio 12 bugs
[15:23:23] <jfndi> Qualification?
[15:23:27] *** Kbuck70 has quit IRC
[15:23:32] <movement> the presence of SS11 bugs? which ones?
[15:23:44] <mritun_> uh
[15:23:55] *** Kbuck70 has joined #opensolaris
[15:23:58] * mritun_ scracheshead
[15:24:09] <mritun_> ^^ bug in spelling
[15:24:38] <oxygene> movement: nothing specific - it's just the tested platform
[15:24:41] <mritun_> I was thinking of adding -xautopar and whatnot to the build, and see how far it goes before sinking :-D
[15:25:04] <mritun_> s/build/compiler options/
[15:25:15] <mritun_> Gentoo/Opensolaris :P
[15:25:34] <Kbuck70> I wanted to get involved in OpenSolaris development but not know where to begin
[15:25:57] <mritun_> Kbuck70: you need to begin by sending Sun the contributor's agreement
[15:26:15] <mritun_> you'll need to fax/snail-mail it after signing it
[15:26:34] <movement> Kbuck70: well actually you should begin by downloading it, installing it, then building it
[15:26:37] <movement> then looking around...
[15:26:50] <movement> then you can search for "oss-bite-size" bugs for some easy fixes to get used to the process
[15:27:19] <mritun_> yup, bite-sized bugs are nice way to start
[15:28:02] <Kbuck70> do I need a sparc computer to do this or can I do this on X86
[15:28:28] <jfndi> X86/amd64 will do
[15:29:06] <jfndi> But you can also try sparc :-)
[15:29:51] <mritun_> opensolaris is mostly platform agnostic
[15:30:05] <mritun_> so doesn't matter where you start...
[15:30:15] *** hohum has joined #OpenSolaris
[15:30:26] <mritun_> you won't need sparc unless you're working on sparc specific bugs and need to test them
[15:31:19] <kjetilho> it's much cheaper to get an x86 box which can compile quickly than a SPARC box
[15:32:52] *** calLNCH is now known as calumb
[15:34:31] <Kbuck70> that does help me, although I wonder how to understand the OS internals better
[15:35:08] <Kbuck70> so I know what I am looking at in a build
[15:35:48] <jfndi> Solaris Internals is certainly a must...
[15:36:16] *** Kbuck70 has quit IRC
[15:36:33] <mritun_> buy the solaris internals book
[15:36:39] <mritun_> and troll the projects pages
[15:36:59] <mritun_> though, the source code is remarkably beautiful... much better than e.g. Linux
[15:37:31] *** Kbuck70 has joined #opensolaris
[15:37:41] <mritun_> but I've found that to understand some functionalities we'd have to understand solaris use cases first
[15:38:00] <mritun_> e.g. deadman timer
[15:39:06] *** Kbuck70 has quit IRC
[15:40:04] *** jpdrawneek has joined #opensolaris
[15:40:23] <jfndi> troll the projects pages ?
[15:40:53] <mritun_> visit the specific project pages
[15:41:10] <mritun_> a lot of them have architecture guides, roadmaps, design documents etc
[15:41:16] <mritun_> which help understanding the code
[15:41:36] *** Netwolf_ has joined #opensolaris
[15:42:22] *** cypromis has joined #opensolaris
[15:42:43] <kjetilho> jfndi: he probably meant "trawl", not "troll" :)
[15:43:05] <jfndi> Ahh. OK.
[15:43:44] *** jamesd_ has quit IRC
[15:43:47] *** jamesd2 has quit IRC
[15:44:30] *** _Megaf|Sleep is now known as _Megaf
[15:45:04] <_Megaf> Hi
[15:49:51] *** calumb has quit IRC
[15:50:02] *** jfndi has quit IRC
[15:54:35] *** Netwolf_ has quit IRC
[15:54:43] *** nachox has joined #opensolaris
[15:58:22] *** Netwolf has quit IRC
[15:58:29] *** jfndi has joined #opensolaris
[15:58:56] <splunk> anyone here know of a script to automount external devices?
[15:58:57] *** jamesd has joined #opensolaris
[15:59:02] *** jamesd_ has joined #opensolaris
[15:59:24] <PerterB> sort of like rmvolmgr?
[15:59:55] <nachox> hal should do that in nevada
[16:00:26] <splunk> rmvolmgr is garbage
[16:00:30] <splunk> never works for me
[16:00:37] <splunk> unless I have to edit the vfast file
[16:00:44] <splunk> vfstab*
[16:00:52] <splunk> i don't wanna fuck that file up though
[16:00:53] <splunk> lol
[16:00:59] <splunk> figured there was another way
[16:03:16] <nachox> hmm, CONSOLE doesnt come commented out by default in solaris?
[16:05:07] *** freakazoid0223 has quit IRC
[16:05:47] *** freakazoid0223 has joined #opensolaris
[16:06:13] <PerterB> nachox: why would you want it commented out by default?
[16:07:24] <nachox> because i wouldnt want root to be able to login remotely via either telnet or ssh by default?
[16:07:42] <PerterB> right, so then you want it *uncommented*
[16:08:08] <mritun_> nachox: root login via telnet and ssh s disabled by default
[16:08:20] <mritun_> s -> is
[16:08:42] *** Netwolf has joined #opensolaris
[16:10:11] <nachox> ops, missread the comments :) which makes me wonder why i was able to login as root using ssh (i will reconfigure ssh now to dissallow it)
[16:11:41] *** cypromis has quit IRC
[16:12:31] <ofu> telnet? /me tries to remember what that is
[16:13:12] <quasi> ofu: telnet is a handy http and smtp client
[16:13:35] <nachox> hehe
[16:13:42] <WickedWicky> quasi: agreed
[16:13:59] <quasi> at least that's what I use it for
[16:14:02] <ofu> ah, the client
[16:14:05] *** pfa3rh has joined #opensolaris
[16:14:24] <ofu> i heard, there even was a telnet-server, somewhere in the past
[16:14:44] <WickedWicky> it's still there.. and unfortunatly used more than it should be
[16:15:52] <quasi> ofu: there still is - it is handy for paranoid network admins who want to be able to grab all your traffic to the servers (or those who are scared of port forwarding in ssh ;)
[16:17:03] <ofu> there are still people using telnet in production environments?
[16:17:15] <ofu> i refuse to believe that
[16:17:58] <quasi> ofu: yes, there are
[16:18:05] <PerterB> I know one environment for sure where telnet is still enabled
[16:18:49] <mritun_> telnet is very handy for limited environments
[16:18:56] <quasi> ofu: we didn't get the permission to implement ssh till two years ago - before that the notwork people wanted to be able to track our sessions
[16:19:08] <mritun_> and when used within an intranet, there is little security risk
[16:19:13] <ofu> hrhr, notwork people
[16:19:18] <kjetilho> I turn off telnet, but keep rlogin in case sshd falls down
[16:19:33] <quasi> ofu: not to mention a whole pile of annexes and other console thingies that still are plain telnet
[16:19:34] * ofu uses serial and ssh, nothing else
[16:19:39] *** jpdrawneek has quit IRC
[16:19:47] <ofu> our cyclades speak ssh2
[16:20:00] <WickedWicky> ours do to
[16:20:13] <PerterB> kjetilho: you're kidding, right?
[16:20:18] <quasi> good old annex doesn't
[16:20:41] <quasi> kjetilho: eeeeeeeeeeeeeek, that'd be insane
[16:20:55] <ofu> only our netapps are rlogin-enabled, very handy for scripting
[16:21:53] <kjetilho> PerterB: no, why?
[16:22:51] <PerterB> because historically rlogin is a much bigger security risk than telnet (modulo the recent telnet bug)
[16:23:04] <kjetilho> o rly
[16:23:13] <PerterB> ya rly
[16:23:22] <WickedWicky> w0rd
[16:23:41] *** Teknix has joined #opensolaris
[16:24:08] <nachox> crap ... /sbin/sh: /lib/svc/method/svc-scheduler: not found
[16:24:20] *** splunk has quit IRC
[16:24:50] <nachox> does anyone remember what package has smpatch?
[16:24:55] <kjetilho> PerterB: "past performance is no guarantee of future failure" :)
[16:25:13] *** _Megaf has quit IRC
[16:25:28] <kjetilho> anyway, you have to understand the security impacts and how to secure your network.  different protocols have different weaknesses
[16:25:31] *** Megaf has joined #OpenSolaris
[16:25:42] <kjetilho> e.g. ssh can be used to DOS a server's CPU generating keys
[16:26:19] <kjetilho> rlogin is vulnerable to forged IP packets
[16:26:26] <kjetilho> telnet is vulnerable to sniffing
[16:26:30] <PerterB> I'm aware of those, but thanks
[16:26:46] <kjetilho> yes, I didn't mean to imply otherwise
[16:27:28] <tsoome> use klogin;)
[16:27:48] *** Fullmoon has joined #opensolaris
[16:28:32] <kjetilho> hehe
[16:28:51] <kjetilho> it's sad, but I've never set up Kerberos.
[16:28:58] <Hystri1> did anybody manage to install the nv66 xen build on an amd64 machine as dom0 without the system clock running 4 times as fast as it's supposed to?
[16:29:00] <kjetilho> ten years ago I could blame US exports laws
[16:29:10] <kjetilho> now I have no excuse, really.
[16:29:17] *** Fullmoon has quit IRC
[16:29:20] *** Hystri1 is now known as Hystrix
[16:31:20] *** migi_ is now known as migi
[16:37:20] *** Megaf is now known as _Megaf|Away
[16:41:28] *** stevel has joined #opensolaris
[16:41:29] *** ChanServ sets mode: +o stevel
[16:44:17] *** Plazma_Work has joined #opensolaris
[16:44:51] <renihs> hmm anyone happens to have a netapp around? i somehow have issues exporting nfs (only) to solaris/linux whatever
[16:46:31] <kjetilho> try to mount it using version 3 explicitly
[16:46:34] <Plazma_Work> trying to make an opensolaris package with pkgmk.. i run it and it errors out saying ERROR in prototype: garbeled entry , -pathname <path> - problem - mode is not numeric
[16:47:34] *** deather_ is now known as deather
[16:48:19] <renihs> kjetilho, i can mount it but if i enter the dir -> error
[16:48:28] <ofu> renihs: what exactly is your problem?
[16:48:30] <ofu> ipfilter?
[16:48:33] <renihs> the netapps own check routine says everything is fine
[16:48:36] <renihs> ofu, no
[16:49:40] <ofu> what does mount tell you?
[16:51:23] <ofu> ontap7? is the exports of the filer correct? NFS licensed?
[16:52:11] <renihs> should be, 7.22
[16:52:16] <renihs> it sais mounted nfsv3, tcp bla
[16:52:30] <renihs> seems normal
[16:52:37] <renihs> however if i enter the dir -> permission denied
[16:52:53] <renihs> exportfs -c on netapp says has access/fine
[16:52:57] *** SYS64738 has joined #opensolaris
[16:52:58] <elektronkind> are you doing that as root?
[16:53:02] <ofu> maproot -> nobody?
[16:53:03] <renihs> yes
[16:53:10] <renihs> on netapp?
[16:53:22] <renihs> hmm dunno howto maproot on the netapp yet
[16:53:26] *** MattMan has quit IRC
[16:53:56] <renihs> i flushed everything and redid but not really a change
[16:54:30] *** Kush- has joined #opensolaris
[16:54:33] <renihs> exportfs -io rw=some_ip/mask /vol/nfs ...on the netapp, nothing fancy
[16:54:34] <ofu>  /vol/web1       -sec=sys,rw=192.168.40.0/21,root=host1:host2,anon=uid
[16:54:37] <renihs> but maybe i am missing something
[16:55:02] <ofu> use rdfile /etc/exports on the netapp to verify it is really in the exports file
[16:55:06] <renihs> . /vol/nfs        -sec=sys,rw=some_ip/mask
[16:55:41] <renihs> hmm i see the issue now
[16:55:48] <renihs> there are 2 uncommented nfs ...the ones i did
[16:55:53] <renihs> i mean commented
[16:56:03] <renihs> and one uncommented which is set for 1 ip only
[16:56:04] *** edwardocallaghan has joined #opensolaris
[16:56:14] <edwardocallaghan> hey
[16:56:17] <renihs> hmm i need to wipe the exports
[16:56:20] <elektronkind> renihs: try adding root=some_ip to your export
[16:56:36] <ofu> if you could please file a RFE to netapp: vi
[16:56:44] <edwardocallaghan> boyd:You around ?
[16:56:48] <ofu> my requests are always ignored
[16:56:56] <renihs> . /vol/nfs        -sec=none,ro,rw=ONE_IP,root=ONE_IP
[16:57:03] <renihs> thats the one i get from rdfile
[16:57:31] <renihs> elektronkind, hmm how i can i remove the export content?
[16:58:08] <renihs> ah zap
[16:58:14] <ofu> exportfs -a should be enough
[16:58:23] *** jfndi has quit IRC
[16:58:27] <edwardocallaghan> ping
[16:58:38] <renihs> if i do that i have the 1 ip access again
[16:59:05] <renihs> zapped it
[17:00:23] *** edwardocallaghan is now known as edwardocallagha
[17:00:28] <edwardocallagha> ping
[17:00:49] <edwardocallagha> My bandwidth is very low here :P
[17:00:52] <renihs>  /vol/nfs        -sec=sys,rw=192.168.0.0/24,root=192.168.0.0/24
[17:00:59] <renihs> shouldnt that work?
[17:01:12] <renihs> for the 192 subnet?
[17:01:19] <ofu> yes, i think so
[17:01:33] *** splunk has joined #opensolaris
[17:01:36] <ofu> did you remount it on the client?
[17:01:37] <splunk> ok back
[17:01:42] <edwardocallagha> Can anyone help me with ID 954099 kern.info IRQ20 prob ?
[17:01:54] <renihs> ofu, yes
[17:02:09] <renihs> still input output error on linux, and permission denied on solaris
[17:02:10] <edwardocallagha> I did not know x86 had a IRQ20 ?
[17:02:15] <renihs> it has???
[17:02:18] <renihs> reserved maybe :p
[17:02:41] <edwardocallagha> whats on IRQ20 then ?
[17:03:13] <renihs> dunno, i still dont believe there is :p
[17:03:53] <renihs> anyhow hmmm this netapp doesnt like me
[17:03:55] <renihs> it seems
[17:04:24] <edwardocallagha> I am trying to google it with nothing useful atm
[17:05:18] <renihs> software list
[17:05:18] <renihs> 722_setup_m.exe
[17:05:22] <edwardocallagha> how do i find what interrupts are what on solaris ?
[17:05:26] <renihs> ...why is there a *EXE* on my netapp?
[17:05:29] <ofu> m? mips
[17:05:36] <renihs> hmm
[17:05:38] <renihs> ah ok
[17:05:39] <ofu> because software install uses exe files
[17:05:42] *** Hystrix has quit IRC
[17:05:45] <renihs> they do?
[17:05:48] <ofu> yes
[17:05:52] <ofu> so this is a 250 or 270?
[17:05:59] <renihs> 270
[17:06:12] <renihs> netapp fas 270
[17:06:13] <renihs> :p
[17:06:37] <edwardocallagha> IRQ storms are bad :(
[17:06:46] <edwardocallagha> edward wonders
[17:06:58] <SYS64738> how can I make xserver to accept remote connections ?
[17:07:11] <ofu> renihs: sounds strange... does rpcinfo -p work correct from client to server?
[17:07:30] <renihs> good question but since mounting works, lemme try
[17:07:37] *** tsoome has quit IRC
[17:07:58] <renihs> yap looks sane
[17:08:24] <renihs> looks like nfsv3
[17:09:15] <edwardocallagha> SYS64738:Just make sure you have allowed it in the ssh config and X conf and do a export DISPLAY
[17:09:23] <ofu> i have no idea... license tells you, nfs is enabled and nfs is enabled?
[17:09:34] <SYS64738> edwardocallagha, thanks
[17:09:56] <elektronkind> SYS64738: see http://archive.netbsd.se/?ml=sunhelp-sunhelp&a=2007-06&m=4390559
[17:09:59] <ofu> nfs status tells you something?
[17:10:15] <renihs> lemme find the correct tab
[17:10:47] <renihs> says nfs is running
[17:10:47] *** cypromis has joined #opensolaris
[17:11:09] <renihs>   nfs site JSPKACN expires 16 Nov 2007
[17:11:30] <renihs> hmm everything seems fine
[17:11:31] *** MattMan has joined #opensolaris
[17:11:32] <edwardocallagha> SYS64738, no worrys mate
[17:11:35] <renihs> this is mighty strange
[17:11:51] *** pfa3rh has quit IRC
[17:12:16] * kjetilho resists the urge to /nick POKE53281,0
[17:12:24] <renihs> lol nfs is the *only* thing not expired
[17:12:29] <renihs> and the only thing not working?
[17:12:30] <ofu> renihs: is the volume online?
[17:12:31] <edwardocallagha> SYS64738:If you come up with a good HOWTO and you would like to share it, then join up to my new site over at http://moonshine.opn4.org/ thanks
[17:12:53] <ofu> or is the volume part of an aggregate?
[17:13:06] *** Plazma_Work has left #opensolaris
[17:13:07] <renihs> one aggregate, root vol and this one
[17:13:31] <renihs> at least when i watched my collegue creating it, lemme find the commands :P
[17:13:44] <edwardocallagha> Its more of a FAQ but I am looking to fill my site ASAP for the first issue
[17:13:54] <ofu> aggr status, vol status
[17:14:01] <renihs>  nfs online     raid_dp, flex
[17:14:12] <renihs> aggr0 online     raid_dp, aggr     root
[17:14:19] <renihs> hmm
[17:14:45] <ofu> looks fine
[17:15:44] <edwardocallagha> elektrokind:You had a topic and your sound OSS issue of some such I found with google and I am getting the same error msg in my dmesg
[17:16:05] <edwardocallagha> IRQ20 and reduced sys proformance ?
[17:16:47] <ofu> none      No  security.   Data ONTAP treats all of the NFS client's users as anonymous users.
[17:16:56] <edwardocallagha> I think its a nvidia problem
[17:17:23] <ofu> this is from man exports
[17:18:29] <renihs> ofu hmm? none?
[17:18:50] <ofu> you pasted security=none...
[17:18:55] <renihs> ah yes
[17:19:11] <renihs> but that i dont care atm, just want to be able to write a bit there :p
[17:19:52] <renihs> hmm confused me is, getting coffee
[17:20:12] <edwardocallagha> so can anyone tell my how to find out what intrrupts are what on a solaris system?
[17:20:29] <edwardocallagha> like cat /proc/intrrupts
[17:21:05] <seanmcg> intradm
[17:21:43] <seanmcg> sorry, intrstat
[17:21:44] <jamesd> or intrstat
[17:23:22] <edwardocallagha> Thanks
[17:23:49] <SYS64738> edwardocallagha, hwo can I restart X11 ?
[17:24:17] <ofu> svcadm restart something-with-X
[17:24:24] <edwardocallagha> svcadm restart gdm
[17:24:32] <edwardocallagha> if your using gdm
[17:24:56] <SYS64738> mmm I continue to have
[17:24:57] <SYS64738> XDMCP windows client
[17:25:00] <seanmcg> or just logout ?
[17:25:04] <SYS64738> DISPLAY=localhost:10.0
[17:25:17] <edwardocallagha> if your using default one then its something i can't remember svcs -a | grep login
[17:25:18] <SYS64738> it should mean that remote displays are disabled
[17:26:14] *** pfa3rh has joined #opensolaris
[17:26:27] <edwardocallagha> i am kind of out of bandwidth :P
[17:26:37] <edwardocallagha> stupid school computers
[17:26:48] <edwardocallagha> I need to get around this stupid linux proxy
[17:29:35] *** tsoome has joined #opensolaris
[17:30:25] *** Snake007uk has quit IRC
[17:30:28] <jamesd> edwardocallagha: find a  rsync or  torrent server with the same files....   perhaps your school doesn't enforce proxy use on those services
[17:31:02] <edwardocallagha> jamesd:No most ports are blocked
[17:31:11] *** sarah has joined #opensolaris
[17:31:19] <edwardocallagha> I think the microsoft update site is not
[17:31:29] <edwardocallagha> anyway to spoof it ?
[17:31:50] <edwardocallagha> its running squid 2.5(someting) stable
[17:33:00] <jamesd> edwardocallagha: one way to get around it is to use port 443,  and setup your home box to proxy on that port...  most sites wont block port 443, because its encrypted https  so is allowed out...
[17:33:32] <SYS64738> I reached to open a remote xterm in my local machine, now what can I do to make start something like twm or gnome ?
[17:34:12] <edwardocallagha> jamesd:Well Tempt has given me a SGD account that i use to get on here and seems to keep going but puts my account into neg figures
[17:35:10] <jamesd> SYS64738: install a  vnc solution, realvnc or something like it..   the X desktop as a whole uses udp and can't get through a firewall or  tcp pipe...   either dispute this or shut up...
[17:35:11] <edwardocallagha> SYS64738:Once exported you can do firefox&
[17:35:36] <SYS64738> jamesd, I am in ipsec vpn with the remote server
[17:36:22] <SYS64738> edwardocallagha, I make confusion with DISPLAY variable, it tells me cannot open display
[17:36:38] <edwardocallagha> ?
[17:36:48] <edwardocallagha> I can't see what your doing
[17:36:52] *** stevel has left #opensolaris
[17:37:46] <edwardocallagha> join my site, make a HOWTO and i'll fix it up and we can have it there for others who want o do the same thing
[17:37:47] <SYS64738> I launched ssh -X -f 172.16.0.241 /usr/openwin/bin/xterm from my local machine
[17:38:21] <edwardocallagha> is the server end lissening ?
[17:38:23] <jamesd> SYS64738: from there you can run any X app, but not a window manager...
[17:38:45] <edwardocallagha> you should not need to export a display if your using ssh, i think it does everything for you?
[17:39:11] <jamesd> it usually does...  and changing display when running over ssh, usually breaks things anyway.
[17:40:26] <edwardocallagha> all i know is setprop options/tcp_listen=true hope it helps
[17:40:29] <edwardocallagha> :P
[17:40:53] <edwardocallagha> VNC is always good
[17:41:16] <edwardocallagha> I think there is less overhead
[17:41:45] <edwardocallagha> I think you can tunnel it under ssh as well but i have not tried
[17:42:38] <WickedWicky> Ellows Tempt
[17:42:40] <edwardocallagha>  /bin/zsh its all you need :)
[17:42:43] <WickedWicky> jamesd, all
[17:43:03] <edwardocallagha> me thinks Tempt is asleep
[17:43:13] <WickedWicky> I thought you were tempt
[17:43:16] <WickedWicky> :P
[17:43:19] *** yarihm has quit IRC
[17:43:37] <edwardocallagha> jamesd _was_/and maybe still is alive ?
[17:43:37] <WickedWicky> but hello edwardocallagha then
[17:43:50] <edwardocallagha> ello ello
[17:44:20] *** SYS64738 has quit IRC
[17:45:05] <edwardocallagha> OK well i will ahve another hack at this IRQ problem on this stupid x86 system
[17:45:18] <edwardocallagha> this is were i like SPARC
[17:45:37] <edwardocallagha> and i hate cheap MB's
[17:46:39] <edwardocallagha> as i'll be offline can some people here chuck some useful debug commands at me please? - i'll man page them later
[17:47:33] <edwardocallagha> i know prtconf -v and now intrstat but any others before i go i should know ?
[17:48:00] <jamesd> man prtconf
[17:48:24] <edwardocallagha> will do, hopful it will give me some leads to others
[17:48:29] *** _Megaf|Away is now known as _Megaf
[17:48:50] <_Megaf> i'm back
[17:48:55] <edwardocallagha> anyother good commands for hardware issue solving on solaris ?
[17:49:27] <edwardocallagha> i am going to be offline for days now and this is making my bandwidth neg here
[17:49:46] <edwardocallagha> s/bandwidth alwance
[17:51:29] <_Megaf>  mount -o loop /media/ISOs/windowsserver2003.iso /media/megafisos/
[17:51:30] <_Megaf> mount: /media/ISOs/windowsserver2003.iso not a block device
[17:51:59] <_Megaf> something is wrong
[17:52:54] <seanmcg> _Megaf: man lofiadm; lofiadm -a /path/to/iso; mount -F hsfs /dev/lofi/X /mnt/point where X is returned from the lofiadm cmd
[17:53:42] <_Megaf> hm
[17:54:14] <edwardocallagha> Well I am going to bed to dream some useless entropy :P night all
[17:54:25] *** freakazoid0223 has quit IRC
[17:54:41] *** andresgd has joined #opensolaris
[17:55:07] *** edwardocallagha has quit IRC
[17:55:17] <andresgd> hi all
[17:55:31] <_Megaf> hello
[17:57:52] <_Megaf> seanmcg, than you very much,
[17:58:35] <andresgd> I am experienced user of FreeBSD, various flavours of linux and of course windows
[17:58:58] *** migi has quit IRC
[17:59:05] <andresgd> I want to give a try to OpenSolaris
[18:00:53] *** bunker has quit IRC
[18:01:23] <andresgd> Could anyone give me some URL I can read so I can see the differences between the Free Solaris 10 and OpenSolaris Dev Edition ?
[18:02:14] *** Jondice has quit IRC
[18:02:39] <Cyrille> http://whacked.net/2005/06/21/confused-so-was-i/
[18:03:06] <Cyrille> is a good start for understanding the difference between Solaris, OpenSolaris and its distributions.
[18:03:16] <seanmcg> _Megaf: np . this question should be in a faq though :)
[18:04:27] <Cyrille> or it should be in the first three hits of "solaris iso mount" on goog... oh wait, it is. ;-)
[18:04:43] <andresgd> Thanks Guys
[18:04:46] <andresgd> ;)
[18:06:19] *** Jondice has joined #opensolaris
[18:09:28] *** soro has joined #Opensolaris
[18:09:37] *** andyshack has joined #opensolaris
[18:10:21] <ofu> can you boot solaris/x86 from iscsi?
[18:13:11] <andyshack> i like this channel, i get at least 200 things to google a day. i wonder what a iscsi is..
[18:16:37] <andyshack> oo iscsi is cool! what do you use it for ofu ?
[18:16:45] *** soro has left #Opensolaris
[18:17:27] <seanmcg> ..  http://www.opensolaris.org/jive/thread.jspa?threadID=36780&tstart=15
[18:17:39] <ofu> i am not using iscsi right now, but i think it might be a good idea perhaps
[18:18:52] <ofu> so i am not the only one thinking it might be a nice idea
[18:21:35] <seanmcg> iscsi software boot psarc: http://www.opensolaris.org/jive/thread.jspa?threadID=36553&tstart=30
[18:22:14] <andyshack> nice
[18:23:33] *** MattMan is now known as MattGYM
[18:23:35] <elektronkind> we're going to use iscsi because it's cheaper than hooking up some of our smaller servers to our fibre channel SAN. It costs me roughly $2000 to connect a box to our SAN (cost of HBA, cables, SFPs, port on the switch) whereas every box we get as 2 or 4 gig-e interfaces.
[18:24:09] *** bondolo has quit IRC
[18:25:12] *** freakazoid0223 has joined #opensolaris
[18:26:22] *** bondolo has joined #opensolaris
[18:28:20] <ofu> i have never used iscsi, but what about cpu usage and offloading it to the nic?
[18:29:57] *** bondolo has quit IRC
[18:30:01] <elektronkind> ofu: you get the iscsi HBAs for that
[18:30:54] <ofu> what hbas are you using?
[18:31:13] <elektronkind> we're not using any yet, but the Qlogic ones are supposedly well-supported
[18:31:26] <Pietro_S> this is interesting problem- C++ source include C openssl header and you can guess what happend ;-)
[18:31:33] <Pietro_S> "/usr/sfw/include/openssl/sha.h", line 109: Error: size_t is not defined.
[18:31:38] *** dunc has joined #opensolaris
[18:31:57] <Pietro_S> I guess that std::size_t is wanted by compiler ;-)
[18:33:31] <andresgd> is anybody using SAS5IR from dell ?
[18:33:31] <Cyrille> could that be because the C++ code uses "uses std ;" which is a recipe for disaster?
[18:35:16] <Pietro_S> Cyrille: yes that's it
[18:35:59] <Cyrille> using blanket "uses whatever ;" is usually the short route to head on desk banging...
[18:36:30] <Pietro_S> Cyrille: nope, I don't understand to your statement well by the first time
[18:36:57] <Cyrille> I'm sorry but I didn't understand what you just meant there.
[18:38:14] <Pietro_S> Cyrille: C++ source code doesn't use "useing namespace std;"
[18:38:57] <Pietro_S> but sunCC wants std::size_t, suncc is happy with size_t
[18:39:09] *** FunkyELF has quit IRC
[18:39:09] *** sarahj has joined #opensolaris
[18:39:38] <andyshack> am i dreaming to use a usb kbd and pci card to speak with the open boot prompt on a e450 ?
[18:39:55] <sommerfeld> andy: yes
[18:41:17] <andyshack> thanks
[18:41:25] <oxygene> unless that pci card has the proper OF packages on it
[18:41:42] *** FunkyELF has joined #opensolaris
[18:42:38] <andyshack> oxygene : that i believe may be beyond me although thanks for the tip
[18:42:48] <jamesd> e450's were not made to be desktops ....   and hi :-)
[18:42:54] <andyshack> oh jimbo
[18:42:57] <andyshack> shouldnt you be working ?
[18:43:12] <Pietro_S> hmm, strange this lines in openssl include file should take care about C++:
[18:43:13] <Pietro_S> #ifdef  __cplusplus
[18:43:14] <Pietro_S> extern "C" {
[18:43:21] <jamesd> took the day off for my b-day... and  i have to work saturday night... patching 60 production servers
[18:43:42] <andyshack> happy birthday!
[18:43:49] <jamesd> thanks
[18:43:54] <tomww> congrats :-)
[18:44:02] *** bondolo has joined #opensolaris
[18:44:39] *** Cyrille has quit IRC
[18:46:21] *** sarah has quit IRC
[18:46:48] <andresgd> Anybody knows someone using Dell Poweredge SC440 with Solaris ?
[18:47:04] *** comay has joined #opensolaris
[18:47:17] *** sstallion has quit IRC
[18:48:26] *** sstallion has joined #opensolaris
[18:50:06] <g4lt-mordant> wait, jamesd is EVEN OLDER today? ;P
[18:52:14] <quasi> g4lt-mordant: roughly one day older than yesterday ;)
[18:53:15] <jamesd> yeap and a tiny bit more wiser, if you dont forget the 5 mb of data that the 4MB of information that acquired yesterday displaced.
[18:53:16] <Pietro_S> jamesd: at first sight, I though that you are celebrating 60, but then I read it carefully and it was `only` 60 servers ;-), congratulation
[18:53:49] <quasi> heh
[18:54:17] *** dunc has quit IRC
[18:54:40] <jamesd> i may feel 60 by the time i leave work on sunday morning...  these are prod servers and are not built to corprate baseline so it may get ugly...
[19:01:32] *** Shiv__ has joined #opensolaris
[19:03:21] *** Jondice has quit IRC
[19:03:24] <Shiv__> Hi Laca
[19:06:04] *** mikefut has quit IRC
[19:08:46] <laca> hi Shiv__
[19:09:44] <Shiv__> The specs are still pending cleanup. Difficult to attend during weekdays. Probably this weekend.
[19:10:20] <Shinden> when ON 71 will be on dvd ? :}
[19:12:32] *** alanc-away is now known as alanc
[19:13:00] <alanc> when you download it and burn it - though why you'ld want a DVD with just ON and not a full distro I don't know
[19:13:12] *** KermitTheFragger has quit IRC
[19:13:48] <laca> Shiv__: take your time (:  i was going to send you some comments too
[19:13:56] <laca> some specs are already in jds or sfe
[19:14:21] <e^ipi> alanc: there you are
[19:14:38] <alanc> yep, I am here
[19:14:52] <Shiv__> Yeah. I think there are 5 of them. Will verify it. If the ones in SFE are old, then these can make their way.
[19:15:09] <Shiv__> You may send comments after the cleanup.
[19:15:24] <e^ipi> you'd know better than anyone.... what'd be the best way to get this blade1000 to drive my 16:10 cheap PC monitor
[19:15:38] <Shiv__> Since not all ppl who worked on it were familiar with SFE structure.
[19:16:07] <e^ipi> the FFB2+ i've got in there only does 4:3 resolutions except on monitors that do Sync-on-green; and it looks like all the sun cards are that way
[19:16:23] <alanc> really?  weird
[19:16:35] <alanc> no idea
[19:16:46] <e^ipi> will DVI make a difference ( if I get, for example, an xvr-500 ) ? at the moment it's 13w3 -> vga converted
[19:17:02] *** laca is now known as lacaAFK
[19:17:05] <lacaAFK> bbiab
[19:17:25] <alanc> my Sun 24" LCD works fine with DVI on my XVR-600
[19:17:26] <Pietro_S> hello Shiv__
[19:17:31] <e^ipi> or can I use an apple PPC card in the sparc like a radeon7000 and expect it to work with solaris & Xorg?
[19:17:32] <Pietro_S> Shiv__: what specs?
[19:17:39] <Shiv__> Pietro_S: Hi
[19:18:03] <alanc> I don't think you'll get non-X console output working without the Sun/SPARC firmware version
[19:18:08] <Shiv__> Pietro_S: Some 35 specs a few of us put together last saturday as part of a local event.
[19:18:27] <e^ipi> alanc: but X should work?
[19:18:28] <alanc> martin bochnig's Xorg sparc driver stuff may work - never tried
[19:18:29] <Shiv__> Pietro_S: listed at the bosug news page
[19:18:44] <Shiv__> alanc: X server crashes if it is not able load keyboard driver, shouldn't this be a bug.
[19:18:56] <_Megaf> http://www.flickr.com/groups/solarisgroup/
[19:19:43] <Pietro_S> Shiv__: can you give me link to bogus page?
[19:20:06] <Shiv__> http://www.opensolaris.org/os/project/bosug/news
[19:20:06] <alanc> Shiv__: no, I think that's still considered a feature so that X doesn't autostart on headless servers
[19:21:52] <e^ipi> alanc: happen to have martin bochnig's email ?
[19:22:02] <alanc> it's all over xwin-discuss
[19:22:57] *** postwait has joined #opensolaris
[19:23:08] <Shiv__> alanc: But the system goes into maintenance state.
[19:23:20] <Shiv__> alanc: It is not just X-server not starting.
[19:23:35] <Shiv__> alanc: I connect a keyboard_1 & boot, system comes up. I connect keyboard_2 & boot system, system goes into maintenance state.
[19:24:13] <alanc> that sounds like a bug
[19:24:17] <delewis> http://209.85.165.104/search?q=cache:_V7Au3R059kJ:sunsolve.sun.com/masterlist/WHITEPAPER.html+sunsolve+whitepaper+collection&hl=en&ct=clnk&cd=1&gl=us&client=firefox-a
[19:24:18] <Shiv__> alanc: I connect keyboard_1 boot, X comes up, Now, I swap it with keyboard_2, and it works fine.
[19:24:33] <delewis> very interesting link -- apparently, SunSolve has all of Sun's old STBs and white papers.
[19:24:59] <delewis> (the link no longer seems to be valid, but the assetkeys are still there)
[19:25:16] *** phips|mb has quit IRC
[19:25:26] <Shiv__> alanc: Interestingly keyboard_1 was a dell keyboard with Sun V440 and keyboard_2 was a sun keyboard.
[19:25:35] <Shiv__> alanc: I had posted it at http://www.opensolaris.org/jive/thread.jspa?threadID=33388&tstart=30
[19:25:44] <Shiv__> alanc: for which you had responded.
[19:26:32] <Shiv__> alanc: (It was in Solaris_9). But unfortunately I no longer had access to the system to experiment.
[19:27:17] <Shiv__> alanc: Do not know if it continues to exist.
[19:27:24] <alanc> keyboard handling has changed a whole lot since Solaris 9
[19:27:32] <alanc> both in X and the kernel
[19:27:43] <Shiv__> alanc: Ok. Then I'll just let it rest. Thanks.
[19:27:59] <alanc> it's only very recently that SPARC even officially supported non-Sun keyboards in OBP even
[19:36:40] <Shiv__> I just came across the HA community site which talks about a 1 node cluster !
[19:36:50] <Shiv__> Why would anyone have a 1 node cluster.
[19:37:05] <Pietro_S> Shiv__: from libnjb.spec (first line of comment nicely show what was source of copy&paste) - I think that SUNW_Pkg doesn't need to be defined unless you want something special, include files will create it
[19:37:07] <Shiv__> What benefits does it offer over just the OS
[19:37:56] <Shiv__> Pietro_S: I suggest you wait till I do some cleanup. Some of the book keeping errors will be corrected.
[19:38:19] <Shiv__> Pietro_S: I will then ask for a review on the desktop discuss and you may volunteer :)
[19:38:43] <Shiv__> Pietro_S: Maybe by sunday I should be able to do it.
[19:39:19] <Shiv__> Pietro_S: The specs are contributed by 9 different people of whom 7 were completely new to spec files.
[19:39:45] <Shiv__> Back to my question to the forum members at large:
[19:40:09] <WickedWicky> Eya all
[19:40:09] <Shiv__> How is a 1-node cluster better than a system with just the OS ?
[19:40:38] <Shiv__> Does it influence any aspect of availability?
[19:40:48] <quasi> Shiv__: as in it provides ways to monitor and manage applications
[19:40:53] <WickedWicky> yeps!
[19:40:58] <WickedWicky> I was about to write that
[19:41:11] <WickedWicky> you dont have to waste time on writing oracle monitor scripts for example
[19:41:11] <quasi> Shiv__: but something similar could probably be done with smf
[19:41:24] <WickedWicky> you just use the "off the shelf" methods that come with the cluster agents
[19:42:04] <WickedWicky> and instead of running 495489054 scripts with ps -ef | grep something you just run scstat to check your application's health
[19:42:26] <WickedWicky> granted it will do jack, availability wise
[19:43:12] <quasi> I have been thinking about porting some of the cluster agents to be usable in smf instead
[19:43:25] <WickedWicky> not sure if the term 'better' fits.. I think it's 'easier' to keep track of what goes on with mission critical applications
[19:43:34] <Shiv__> Am new to sun cluster, so will ask some simple questions.
[19:43:37] <WickedWicky> quasi: that should be very possible
[19:43:51] <WickedWicky> I'd be interested in how you do it and what you do
[19:43:54] <Shiv__> With smf in place is there still a case for 1-node clusters
[19:45:04] *** andresgd has left #opensolaris
[19:45:11] <quasi> WickedWicky: my first aim would have been the weblogic cluster agent, but I've moved on from the customer where it would be relevant, so I'm not sure what will happen
[19:45:20] <WickedWicky> when you define your dependancies within smf well enough and include reporting when things go screw-all, IMO: nopt
[19:46:28] *** Kush- has quit IRC
[19:46:39] <quasi> I agree
[19:47:18] <Shiv__> Was having a look at scstat man page on the net.
[19:47:24] <WickedWicky> quasi: I think we have some old Sun box running weblogic... I will have a look at that on monday when I am in the office
[19:47:33] *** Jondice has joined #opensolaris
[19:47:48] <quasi> Shiv__: sounds like cluster 3.1 or older
[19:48:08] <WickedWicky> scstat is Sun Cluster 3.1
[19:48:23] <WickedWicky> we run 3.1u4 at work with Oracle 9iAS cluster agents
[19:48:35] <andyshack> anyone touched smoothwall ?
[19:48:52] <WickedWicky> used to work with it before our company got taken over by KPN Telecom
[19:48:57] <andyshack> ie : do you still touch it, is it touchable, should i not touch it ?
[19:48:58] <Shiv__> So, what is it with this Open HA cluster from HA community?
[19:49:04] *** Kush- has joined #opensolaris
[19:49:06] <WickedWicky> andyshack: i was happy with it
[19:49:13] <quasi> WickedWicky: I've just had a brief look at the wls cluster agent when the code was first released and it  didn't look like it did all that much
[19:49:15] <andyshack> WickedWicky nice thanks
[19:49:16] <Shiv__> How do I make a process part of this monitory framework or whatever it is called.
[19:49:41] <WickedWicky> andyshack: you can do IPSec with it as well
[19:50:00] <quasi> WickedWicky: I can think of better ways to integrate with smf - but those bits vary very much based on how fscked up your wls setup is
[19:50:01] <Shiv__> I intend to setup a cluster with 2 VMware based systems sometime this weekend.
[19:50:17] <WickedWicky> quasi: it doesnt do much indeed, hence my comment regarding SMF and my opinion when you set it up properly ;-)
[19:50:34] <quasi> Shiv__: aim for 3.2 rather than 3.1 if you can
[19:50:37] <WickedWicky> Shiv__: Sun Cluster is really a nice thing to play with
[19:50:40] <WickedWicky> try 3.2 though
[19:50:42] <WickedWicky> yes
[19:50:43] <Shiv__> I do not have any other infrastrcutre to experiment.
[19:50:46] <WickedWicky> 3.2 is freely downloadable
[19:50:53] <Shiv__> On x86 ?
[19:50:57] <WickedWicky> yes
[19:51:05] <cypromis> yes
[19:51:07] <cypromis> and works fine
[19:51:13] <cypromis> and will take you far less time to implement
[19:51:36] <Shiv__> Ok, will check it out.
[19:51:50] <WickedWicky> http://www.sun.com/software/solaris/cluster/index.xml
[19:51:53] <quasi> there's some nice and simple docs on osol.org in the ha-cluster community
[19:52:07] <quasi> http://www.opensolaris.org/os/community/ha-clusters/
[19:52:21] <WickedWicky> there, that should give you something to read this weekend
[19:52:24] <quasi> ehrm, http://www.opensolaris.org/os/community/ha-clusters/ohac/Documentation/
[19:52:50] <Shiv__> What should I look at, to make a process part of the monitoring framework for availability?
[19:52:59] <Shiv__> resource groups?
[19:53:09] *** estibi_ has quit IRC
[19:53:11] <Shiv__> Sounds like actual devices (like hard disks)
[19:53:18] <WickedWicky> no
[19:53:23] <WickedWicky> a resource group is a container
[19:53:38] <WickedWicky> in which you define your resources, like shared storage, your actuall application
[19:53:50] <WickedWicky> shared IP address
[19:53:52] <WickedWicky> etc etc
[19:54:28] <WickedWicky> every individual resource can/should be monitored by either the cluster software or additional cluster agents
[19:55:47] <Shiv__> Ok, I think I need lot more intro info and will visit docs.sun.com. An IRC session will not help. Thanks.
[19:55:52] <WickedWicky> example:
[19:56:02] <WickedWicky> snippet of an scstat on a server at work is:
[19:56:03] <WickedWicky>      Group: oracle-rg           ORA-1               Online
[19:56:04] <WickedWicky>      Group: oracle-rg           ORA-2               Offline
[19:56:22] <WickedWicky> here you see a resource group called oracle-rg, it is defined on two nodes, and active on ORA-1
[19:56:36] <WickedWicky>   Resource: oracle-hastorage    ORA-1               Online    Online
[19:56:36] <WickedWicky>   Resource: oracle-hastorage    ORA-2               Offline   Offline
[19:56:41] *** bnitz has left #opensolaris
[19:56:54] <WickedWicky> and this is a resource within the resource group, in this case shared storage on our SE3510
[19:57:13] * Shiv__ Hmm.....watching the outputs...
[19:57:49] <seanmcg> Theres a link from blogs.sun.com/timf/13th_ie_osug_meeting_report to a presentation on HA cluster.  Its a broad overview - mainly about the opensourcing of it, but its good intro.
[19:58:06] *** Kernel86 has joined #OpenSolaris
[20:00:23] <CIA-21> jp151216: 6552729 db handle should be reused instead of opened per-request, 6588930 idmapd crashes in processing AD LDAP response
[20:00:25] <CIA-21> okie: 6591834 nwamd core dump in get_user_preference()
[20:05:49] *** polk_ has joined #opensolaris
[20:06:34] *** Shiv__ has left #opensolaris
[20:08:39] *** polk__ has quit IRC
[20:12:07] *** bengtf has quit IRC
[20:15:44] *** lacaAFK is now known as laca
[20:17:09] *** iceq has quit IRC
[20:17:45] *** alfism has joined #opensolaris
[20:25:19] *** agliodbs has joined #opensolaris
[20:25:31] *** ket has joined #opensolaris
[20:30:30] <ket> query; am i correct in the reading that /usr/sbin/format will -always- traverse the entirety of /dev/dsk/* and/or /dev/rdsk/* even when passed -d?
[20:34:10] *** bengtf__ has joined #opensolaris
[20:34:12] *** bengtf__ is now known as bengtf
[20:38:41] <tomww> ket: seems to me thet format wants to make shure, the symlink points to a configured disk which is alive (=configured as in cfgadm -avl)
[20:39:01] *** estibi has joined #opensolaris
[20:39:04] *** sartek has joined #opensolaris
[20:39:36] <ket> tomww: so i'm not hallucinating as i look at the code here then?
[20:39:56] <ket> tomww: and it is poking the entirety of /dev/dsk/* and dev/rdsk/*?
[20:41:13] *** mikefut has joined #opensolaris
[20:46:41] *** ChanServ sets mode: +o jamesd
[20:47:19] *** slowhog has joined #opensolaris
[20:48:01] * bda sets kawaii on fire.
[20:48:22] <bda> Or I guess I could set up osol-discuss@ rules instead.
[20:49:04] <wesolows> the rule should be that if mail arrives from osol-discuss, automatically send an unsubscribe request for osol-discuss
[20:49:24] <bda> Seems sound.
[20:52:34] *** Mdx4 has joined #opensolaris
[20:56:32] *** sommerfeld has quit IRC
[20:56:48] *** Red_Cloud has joined #opensolaris
[20:57:37] *** gursikh has joined #opensolaris
[20:58:14] <rpm> neat, my free solaris 11/06 cd's just came
[20:59:15] *** sommerfeld has joined #opensolaris
[20:59:15] *** ChanServ sets mode: +o sommerfeld
[20:59:26] <WickedWicky> welcome back
[21:00:05] <alanc> holy flaming putbacks!
[21:00:09] <alanc> it went in!
[21:00:14] <elektronkind> ??
[21:00:17] <CIA-21> dwoods: 6589158 ddi_dma_addr_bind_handle with DDI_DMA_FORCE_PHYSICAL fails on OPL
[21:00:20] <WickedWicky> sta?
[21:00:29] <Tpenta> alanc?
[21:00:35] <alanc> hmm, must have just barely missed the CIA cutoff
[21:00:44] <nachox> ?
[21:00:45] <alanc> 6437624 RFE: Add ksh93 (as /usr/bin/ksh93) and libshell.so to OS/Net
[21:00:53] <alanc> Contributed by Roland Mainz <roland.mainz at nrubsig dot org>
[21:00:53] <nachox> holly sh*t!
[21:01:00] <Tpenta> woooooo
[21:01:09] * elektronkind raises a beer in Roland's direction
[21:01:17] <nachox> roland must be real happy today
[21:01:21] <wesolows> Too bad Roland didn't listen to our advice about the heads-up.  No one will read all that.
[21:01:34] <wesolows> But still, it's good that this is done.
[21:01:42] <movement> wow.
[21:02:12] *** gursikh has quit IRC
[21:02:17] <elektronkind> I would say it's the largest RFE putback by a OS.org community member to date (non-Sun) ?
[21:02:23] *** bunker has joined #opensolaris
[21:02:31] <wesolows> of work done by people outside Sun, yes
[21:02:35] *** ChanServ sets mode: +o Tpenta
[21:02:36] <jamesd> but will they ever put back the change that tells  useradd to use ksh93 as the default shell when creating non root users.
[21:02:55] <alanc> yeah, that is a long one
[21:03:18] <nachox> ksh's integration is a major milestone imho
[21:03:46] <jamesd> next up  star/sfind ;-)
[21:03:57] <alanc> it's been added to http://www.opensolaris.org/os/community/on/flag-days/71-75/ but the linked page for the headsup isn't there yet
[21:03:58] <tsp> hmm, I tried making a proejct and adding a user to it with a limit on the number of max-lwps they could create. That works, but a 2nd or 3rd login will still work and get a new limit. Is there a way I can make that global per-user, and not per user login?
[21:04:13] *** nachox has left #opensolaris
[21:04:36] *** nachox has joined #opensolaris
[21:05:14] <WickedWicky> will I be able to get ksh93 when I do an hg update on onnv-gate repository?
[21:05:49] <_Megaf> how to load a kernel module? like modprobe on linux
[21:05:56] <nachox> what was the link to the ksh thing? centerclick on the link actually closed the channel in gaim
[21:06:05] *** baijiutong has joined #opensolaris
[21:06:08] <nachox> _Megaf: you shouldnt need to use it
[21:06:13] <estibi> _Megaf: modload
[21:06:30] <elektronkind> _Megaf: modload, modinfo, modunload
[21:06:42] <nachox> nod
[21:06:43] <_Megaf> estibi, modload module?
[21:06:53] <alanc> WickedWicky: you should - if not, then stevel's bridge broke again
[21:06:56] *** Yamazaki-kun has joined #opensolaris
[21:06:57] <wesolows> you should never need to do that
[21:07:00] <Pietro_S> silly question with no meaning to start flame, but is ksh93 so much better than bash (considered about so big applause, never tried it)?
[21:07:02] <elektronkind> _Megaf: modload has a nice man page.
[21:07:05] <estibi> but like said nachox you don't neet to do it
[21:07:09] <wesolows> modload is unlikely to do what you want
[21:07:32] <WickedWicky> alanc: it's in the 'tip' branch then, I presume?
[21:07:34] <nachox> Pietro_S: ksh93 is great, much better than bash imho
[21:07:43] <_Megaf> elektronkind, not here
[21:07:56] <_Megaf> No manual entry for modload
[21:08:09] <_Megaf> ok
[21:08:11] <_Megaf> google...
[21:08:14] <wesolows> _Megaf: You either didn't install man pages or your MANPATH is broken.
[21:08:18] <alanc> I would assume so, haven't checked out onnv from mercurial myself
[21:08:29] <WickedWicky> alanc: trying it now, thank you :)
[21:08:30] *** Fish- has joined #opensolaris
[21:08:35] <tsp> does everything in the flag day pages mean its in/going in?
[21:09:07] <Fish-> hello
[21:09:11] <e^ipi> unless it slips for some reason
[21:09:33] *** Fish- is now known as Fish
[21:09:36] <tsp> neat
[21:09:43] <Tpenta> hello Fish
[21:09:48] <alanc> the flag days are stuff that's been putback, so is in, unless it's found to be so broken after going in that it has to be backed out
[21:10:33] <wesolows> (found to be so broken AND the gate staff have the balls to back it out)
[21:10:41] <tsp> heh
[21:10:45] *** JonathanW has joined #opensolaris
[21:10:54] <Pietro_S> nachox: I know that the next statement is non ok, but I still have to ask: "if you can say only *one* advantage ksh93 over bash which will it be?"
[21:11:36] <alanc> for instance, you may notice File Events Notification API appears under both nv_71 and nv_72 since it went in, got backed out, fixed its bugs and went back in
[21:11:37] <WickedWicky> I hope I can do regexp, like in the current ksh
[21:11:47] <e^ipi> Pietro_S: complex number support
[21:11:57] <WickedWicky> yea I see it can do float calculations!
[21:12:01] <WickedWicky> is this true?
[21:12:08] <nachox> Pietro_S: usefull tab completion, a history that actually works, threads, posix compliant built-ins, floating point number support
[21:12:17] <e^ipi> the reals are a subset of the complex numbers...
[21:12:25] <wesolows> e^ipi: If that's the ONE advantage, what a colossal waste of time.  How many people actually use complex numbers?  And how many of them don't use fortran?
[21:12:54] <nachox> Pietro_S: also associative arrays
[21:13:02] <e^ipi> it's the one i find most entertaining to point out
[21:13:02] <alanc> http://www.opensolaris.org/os/community/on/flag-days/pages/2007081701/ is available now with the ksh93 heads up message in all it's wordy glory
[21:13:08] <e^ipi> precisely because of it's pointlessness
[21:13:13] <Pietro_S> wesolows: you saved me responce to e^ipi ;-)
[21:13:50] <_Megaf> # modload -p /kernel/drv/kqemu
[21:13:50] <_Megaf> can't load module: No such file or directory
[21:13:50] <_Megaf> root@spacesol:/export/home/megaf/Virtual_Machines_QEMU# ls /kernel/drv/ | grep kqemu
[21:13:50] <_Megaf> kqemu
[21:14:28] <nachox> i'm off
[21:14:33] *** nachox has quit IRC
[21:14:34] <WickedWicky> I wasnt aware that qemu had an acceleration module for solaris
[21:14:48] <wesolows> The right question is not "why is ksh93 better than bash" but "why is ksh93 better than ksh88"
[21:14:55] <wesolows> One obvious answer is that we have the source.
[21:15:18] <Tpenta> correct
[21:15:24] <Tpenta> much better question
[21:16:34] <g4lt-mordant> the best question is "why ksh?"
[21:16:45] <wesolows> g4lt-mordant: The answer to that question is "POSIX"
[21:16:49] <ket> g4lt-mordant: because ksh93 makes me happy
[21:16:50] *** JWheeler has quit IRC
[21:16:52] <seanmcg> wesolows: the accelerator got gpl'd couple of months ago. (me thinks)
[21:16:54] * g4lt-mordant believes in using as few words as possible
[21:17:05] <Tpenta> no. choice is good. WE provide shells to provide choice.
[21:17:13] <ket> so, who wants to confirm that format -always- traverses all of /dev/dsk/ and /dev/rdsk even when passed a -d argument?
[21:17:15] <g4lt-mordant> wesolows, fine, make it /usr/xpg4/bin/sh
[21:17:57] <wesolows> g4lt-mordant: Hopefully there will be a time when that's possible.  But having /bin/ksh93 and /bin/ksh does no harm.  If you prefer bash (as I do) then use bash.
[21:18:17] <wesolows> It's stupid to say that something shouldn't exist just because you don't like it.
[21:18:26] <wesolows> It's not hurting you any.
[21:19:14] * g4lt-mordant forgot the ;P
[21:19:47] <wesolows> with ksh and libdisasm, 2 large pillars of closed sourceness are falling
[21:19:57] <wesolows> I wonder how the libc_i18n people are doing
[21:21:50] <alanc> http://i18n-freedom.blogspot.com/
[21:22:37] <g4lt-mordant> personally, I never change root's shell, but I [ab]use tcsh pretty heavily
[21:22:57] <tsp> zsh over here
[21:23:53] <g4lt-mordant> my thing is "don't mess with /bin/sh until you're 1,000% sure that it won't break anything, then don't anyway"
[21:24:10] <wesolows> I've made changes to /bin/sh
[21:24:17] <wesolows> it's very buggy
[21:24:20] <wesolows> a pile of shit, really
[21:24:38] *** bunker has quit IRC
[21:24:54] <wesolows> If we ever do SunOS 6, that fucker'll be the first thing I delete.
[21:25:13] *** mikefut has quit IRC
[21:25:54] <delewis> it does need to go.
[21:26:01] <g4lt-mordant> well, I've seen linsux and the Magical Moving Bash target, and I'm not suree that solaris needs to follow that road
[21:26:07] <tsp> it doesn't even have command history, so it has to go
[21:26:16] <delewis> its every week that I hear someone griping about /bin/sh brokeness and why Sun hasn't gotten rid of it, yet.
[21:26:18] <Tpenta> keith, at least the sh code looks infinitely better than it did in v7
[21:26:31] <wesolows> Tpenta: You mean the pseudo-ALGOL?  Indeed. :-)
[21:26:45] <delewis> do you guys still have the Algol macros?
[21:26:47] <Tpenta> well it was more srb wanting the source to sh to look like sh
[21:26:50] <WickedWicky> ksh93 didnt make it to the mercurial repository yet
[21:26:51] <delewis> or did you get rid of those?
[21:27:13] *** sartek has quit IRC
[21:28:01] <g4lt-mordant> tsp you hav to ralize that /bin/sh is not supposed to be comfortable, it's supposed to work under ALL circumstances and you can always launch $SHELL_OF_CHOICE from it if you can get your system up that far
[21:28:49] <delewis> there's a distinction between an interactive and programming shell, you know. tcsh is an excellent interactive shell, but its virtually useless for scripting.
[21:29:25] * delewis waits for some tcsh fanatic to argue otherwise
[21:29:30] <e^ipi> wesolows: people?
[21:29:33] <e^ipi> there's me...
[21:29:40] <g4lt-mordant> bunt now it's payday lunchtime.  don't wait up, it's five o'clock somewhere ;)
[21:29:51] * Tpenta uses csh as interactive most of the time ;)
[21:30:16] <g4lt-mordant> delewis, I'd disagree, my best shellscripting is done in tcsh, but it's literally a large case/switch script that's impossible on most other shells
[21:30:33] <delewis> I'd probably be satisifed with csh/tcsh as an interactive shell, but I do find myself doing quite a few for loops during my interactive sessions.
[21:31:12] <delewis> I could exec sh or whatever, but that's not the point.
[21:31:52] <g4lt-mordant> yeah, loops in tcsh throw you for onee, but it's the only one with a sane case/switch
[21:32:29] <ket> so, nobody can answer the format question? :P
[21:32:45] *** MattGYM is now known as MattMan
[21:32:55] <delewis> g4lt-mordant: I assume you've read "csh programming considered harmful"?
[21:33:36] <e^ipi> libc_i18n-emancipated.a builds, but it won't build in to libc yet
[21:34:11] <g4lt-mordant> yes, and agreed with as much as I did Dijkstra, in fact, it made me make the joke "consdiered harmful screeds considered improper use of TFM"
[21:34:13] <e^ipi> i'm fighting with the linker to recognize that I do, in fact, have those functions that it says it can't find in there
[21:34:45] <g4lt-mordant> asa in, ignoring it is improper use.  I'm out now
[21:36:28] <delewis> if anyone wants to experience a bit of nostalgia -- here's the original Solaris x86 OE white paper from '94 -- http://24.176.104.6/~dlewis/solaris_x86_oe_wp.pdf
[21:37:43] *** MattMan has quit IRC
[21:41:03] <ket> delewis: please no
[21:41:20] <ket> delewis: i'll have to dredge up the ultra AXi OE white papers and such
[21:42:56] <ket> oeming rather :P
[21:43:09] <andyshack> slash opt/csw/bin/pkg-get upgrade <-- should i be scared of that ?
[21:44:08] <delewis> Sun also published this gem -- http://24.176.104.6/~dlewis/computer_viruses_and_what_to_do_about_them.pdf
[21:44:13] <delewis> (the title says it all :-)
[21:44:27] *** pfa3rh has left #opensolaris
[21:44:57] <ket> delewis: you've never seen the ultra AXi/AXmp OEMing stuff from about '95 or '96, have you? :)
[21:46:46] *** RicardoSSP has joined #opensolaris
[21:49:21] <splunk> DOes Opensolaris come with Basic Security Module (BSM)
[21:51:34] *** Red_Cloud has quit IRC
[21:57:39] <Pietro_S> laca: when 2 patches modify same file, should I use %patch13 -R -p1 or should I check if my patch is weel done?
[21:58:38] <laca> Pietro_S: the patch should be written such that it applies to the previously modified file
[21:59:01] <Fetch> Just curious, I saw something about a svn70a respin, is svn70's release going to hold for that?
[21:59:34] <Fetch> snv*
[22:01:07] *** Murmuria has joined #opensolaris
[22:01:13] *** tsp has quit IRC
[22:01:20] *** theRealballchalk has joined #opensolaris
[22:03:08] <alanc> snv70 was already a week late due to the integration of the new installer
[22:03:40] <Fetch> *nod* actually, I was reading an IRC log quoting you on Friday :)
[22:03:46] <alanc> but I have no idea what they're doing with the SXCE releases with the respins of 70 going on in parallel with 71
[22:06:45] <splunk> BSM anyone?
[22:08:27] *** andyshak has joined #opensolaris
[22:08:38] <andyshak> IT WORKS!
[22:08:53] *** mritun_ has quit IRC
[22:08:58] *** andyshak has quit IRC
[22:09:27] <andyshack> i installed an irc client!
[22:09:56] <andyshack> im fkn stoked
[22:09:57] <andyshack> nite folks
[22:10:03] *** andyshack has quit IRC
[22:11:27] *** tsp has joined #opensolaris
[22:11:51] *** Plazma_Work has joined #opensolaris
[22:12:20] <Plazma_Work> I have an intel gma950 integrated video card.. how do i get DRI and openGL enabled on it?
[22:13:13] *** kaiwai has joined #opensolaris
[22:14:46] <Pietro_S> Plazma_Work: do you still have problem with pkgmk? I was afk when you asked and when I came you wasn't here ...
[22:15:28] <Plazma_Work> no i got it, thx
[22:17:41] <Pietro_S> Plazma_Work: was the problem ' '  in the file which was packaged by pkgmk? - that was my guess
[22:17:55] <seanmcg> splunk: read http://www.cuddletech.com/blog/pivot/entry.php?id=647 for a nice one on bsm
[22:18:58] <Plazma_Work> Pietro_S:  no, it was a problem with my prototype file, i debugged it and fixed it, thx for chekin
[22:19:39] <splunk> read it
[22:19:40] <splunk> nice article
[22:19:52] <splunk> thanks
[22:20:43] <seanmcg> splunk: and of course the security community page on o.s.o. has links to security audit goods
[22:21:57] <seanmcg> o.s.o. search is good for most thinks splunk.
[22:22:08] *** esaxe has joined #opensolaris
[22:22:26] <Pietro_S> hmm, I think that after creating 10 different patches I should say that this sf is gcc only, right now I have 15 patches and the latest error I already solved in other file, but I can't remember which diff was the cure :-(
[22:24:26] <Fetch> writing compiler-agnostic C code is good practice
[22:24:48] <Fetch> otherwise, the next thing you know you'll be writing Linux networking code, and that's just sad
[22:27:06] *** tsp has quit IRC
[22:27:24] <kaiwai> hmm, eta on when SXCE will be delivered?
[22:27:31] *** Plazma_Work has left #opensolaris
[22:27:34] *** sstallion has quit IRC
[22:28:42] *** sstallion has joined #opensolaris
[22:29:09] <e^ipi> if there's a specific feature you need, BFU
[22:29:19] <e^ipi> otherwise, it doesn't much matter, you can LU later
[22:29:56] <Pietro_S> e^ipi: I want to try new installer in action, I doubt bfu will help ;-)
[22:30:00] <splunk> thanks
[22:30:59] *** Pietro_S has left #opensolaris
[22:35:27] *** Chipdanc1r has joined #opensolaris
[22:38:31] *** tsp has joined #opensolaris
[22:38:38] *** iceq has joined #opensolaris
[22:38:56] <tsp> hmm, fork bomb caused the box to panic
[22:39:27] * quasi rolls eyes - ksh93 integration - I never thought that would happen
[22:40:45] <tsp> how do I see waht crashed my kernel? I wasn't able to get a capture of the serial console, but I do have the /var/crash/ files
[22:41:14] <wesolows> cd /var/crash/<hostname>; mdb 0
[22:41:16] <quasi> tsp: mdb is a good start for debugging crashdumps
[22:41:27] <wesolows> (or whatever crash dump number it is)
[22:41:55] <wesolows> sometimes the panic message ends up in the system log, but the dump is much more reliable
[22:43:47] <tsp> hmm, I'm way out of my depth here. I might be able to get it to panic again though - managed to do it twice so far
[22:44:11] <seanmcg> tsp: post the fork bomb ?
[22:44:28] <Tpenta> if you have a crash dump, why would you want to force it to panic again?
[22:44:31] <e^ipi> tsp: man ulimit
[22:44:36] <quasi> tsp: do the mdb command as wesolows suggested and type ::status for starters
[22:44:58] <Tpenta> $C (or ::stack) will also be useful
[22:45:12] <tsp> BAD TRAP: type=31 rp=2a100cd5710 addr=0 mmu_fsr=0 occurred in module "genunix" due to a NULL pointer dereference
[22:45:23] <wesolows> that's a start, ok - now the stack
[22:45:44] <tsp> I was screwing with projects and resource controls along with perl -e'fork while 1' - one of those caused it to die
[22:46:53] * tsp will pastebin it
[22:46:55] <seanmcg> odd, thought perl had a limit on how many forks it can do.
[22:47:51] *** Yamazaki-kun has quit IRC
[22:47:59] *** Chipdancer has quit IRC
[22:48:00] <tsp> http://pastebin.ca/661500
[22:48:12] <tsp> not sure if that'll help anyone, but its worth a shot
[22:50:39] <seanmcg> looks like: http://bugs.opensolaris.org/view_bug.do?bug_id=6583594
[22:51:51] *** rpm has quit IRC
[22:52:16] *** dunc has joined #opensolaris
[22:52:19] <kaiwai> hmm, the korn officiando's must be happy with the merging into opensolaris
[22:52:39] <tsp> hmm, probably related to that. Hopefully it won't keep happening, as long as I stay away from ed /etc/project
[22:53:38] <e^ipi> aficionado kaiwai.... it's spelled aficionado...
[22:54:08] <kaiwai> e^ipi: sorry oh masterful e^ipi! *worships the ground which e^ipi walks on*
[22:54:33] <WickedWicky> just for my understanding, how can a normal user let a box panic by forking lots of stuff? when I type ulimit -a I see the max. processes per user is 8277
[22:54:49] <WickedWicky> surely the default should be a sane value for not crapping a box out of bussiness
[22:54:57] <wesolows> any panic is a bug
[22:55:03] *** Yamazaki-kun has joined #opensolaris
[22:55:08] <tsp> WickedWicky: perl -e'fork while 1' - brings the box down in ~2 seconds
[22:55:25] <wesolows> except the kind caused by uadmin/reboot -d, or intentional horkage via mdb.
[22:55:35] <Tpenta> sounds reprodi=ucable to me. you might want to log a bug
[22:56:01] <WickedWicky> wesolows: well yes, my box paniced every 6 minutes as well but not cause of a kernel issue but bad RAM.
[22:56:02] <tsp> Tpenta: if I can figure out how to get it to happen again, I will :)
[22:56:26] <Tpenta> thought you said that perl would do it in 2 seconds
[22:56:28] <WickedWicky> I just wonder if the default ulimit size for max. user processes could be too high, if the forking really is an issue
[22:56:35] <wesolows> fork bombs may render the system unusable if reasonable resource controls aren't in place, but it should not panic
[22:56:48] <WickedWicky> hm
[22:56:49] <tsp> Tpenta: I'm not sure the fork is the issue; something iwth rctl is
[22:57:01] <tsp> as soon as I edited /etc/project and su'd to the user, it paniced
[22:57:20] <wesolows> oh yeah, it panicked in the rctl syscall
[22:57:28] <wesolows> 100% probability it's a bug
[22:57:33] <Tpenta> oh yes
[22:57:35] <tsp> but before that I was su'd in and forking away without problems; kill the fork, ed /etc/project, su, it went bang :)
[22:58:05] <tsp> but http://bugs.opensolaris.org/view_bug.do?bug_id=6583594, as was pointed out, seems to be it anyways
[22:58:34] <quasi> kaiwai: I'm a fan of ksh, but quite happy with the ksh88 variety
[22:58:54] <wesolows> yes, that's clearly the bug
[22:58:55] * WickedWicky uses ksh for god knows how long now
[22:59:34] <tsp> is making a project for every user on my system with the project.max-lwps variable set the proper way to limit per-user processes?
[23:00:08] <kaiwai> it would be nice if they updated their caselog
[23:00:36] * quasi has prolly used ksh since before ksh93 was made (if the name means what I think it does ;)
[23:00:41] <kaiwai> case in point: http://opensolaris.org/os/community/arc/caselog/2007/027/
[23:00:59] <WickedWicky> quasi: that old eh... ;-)
[23:01:01] <wesolows> there are still cases waiting for Andy Tucker to write an opinion
[23:01:24] <quasi> WickedWicky: aye, feeling very, very old tonight ;)
[23:01:37] * WickedWicky tosses you a beer
[23:01:44] <kaiwai> I want AmigaSHELL
[23:02:00] <kaiwai> REXX would be good
[23:02:02] * quasi sniffs the cap and falls over drunk
[23:02:03] <wesolows> I want windows command.com
[23:02:21] <quasi> and a pony
[23:02:30] <WickedWicky> I want a Lamborghini countach
[23:02:31] <wesolows> a shitless pony
[23:02:38] * kaiwai gives quasi a pony
[23:02:45] <wesolows> cleaning up after the other kinds is annoying
[23:03:10] * WickedWicky goes to Toys 'r us and buys you a "my little pony"
[23:03:56] *** apersson has joined #opensolaris
[23:04:15] * kaiwai squeals with delight
[23:05:33] *** RicardoSSP has quit IRC
[23:05:36] <quasi> kaiwai: I didn't want a pony, but I just assumed wesolows wanted one to go with his command.com
[23:06:02] * kaiwai gives weslows a pony
[23:06:07] * kaiwai gives quasi a hug
[23:07:09] * quasi is beginning to think the pony would have been better ... at least if I could have it cut into nice steaks and get some coal for the barbecue ;)
[23:07:25] <WickedWicky> when is ksh93 expected to be in the tip branch of the mercurial respository?
[23:07:56] <quasi> WickedWicky: the announce said integrated in 72
[23:08:11] <WickedWicky> ah ok
[23:08:30] <WickedWicky> I was hoping it'd be merged into the current tip :s
[23:08:38] <kaiwai> now, waiting for OpenSound to be merged
[23:09:15] * quasi can live without opensound - noone will hear the servers anyway ;)
[23:09:47] <WickedWicky> and when you hear them you're not a happy camper anyway
[23:10:05] *** theRealballchalk has left #opensolaris
[23:11:10] * tsp wants a z command in ed
[23:11:21] <tsp> oh well, gnu ed solves that, or edbrowse
[23:17:55] <kaiwai> hmm, or having the server go, "I'm lonely! talk to me!"
[23:18:40] <tsp> kaiwai: nothing wrong with audio on servers if your right next to them, start blasting the mp3s through them
[23:19:14] <tsp> unfortunately opensound won't work with the cs4231 on my U5
[23:20:01] *** hohum has quit IRC
[23:21:56] *** jcsmith has quit IRC
[23:23:31] <kaiwai> ah
[23:24:54] *** jcsmith has joined #opensolaris
[23:26:57] <sommerfeld> wickedwicky: code is in the mercurial tip by the time you see a CIA-N announcement here.
[23:30:22] <kaiwai> hmm, the team up with IBM is interesting
[23:31:19] *** estibi has quit IRC
[23:34:49] *** Yamazaki-kun has quit IRC
[23:34:54] <alanc> but the CIA bot & onnv-notify has been quiet for a couple of hours, so the Teamware->Mercurial bridge may have collapsed again, and stevel isn't here to poke about it
[23:38:05] *** Red_Cloud has joined #opensolaris
[23:38:47] <Mdx4> tsp: apropos opensound there are plannings of port of soundcards ?
[23:39:48] <kaiwai> http://dlc.sun.com/osol/on/downloads/20070813/ <-- that turned up for some reason
[23:48:28] *** Murmuria has quit IRC
[23:48:49] <Red_Cloud> I want to learn the difference between "export PATH=/usr/sfw/bin:$PATH" in my shell file as opposed to "PATH=/usr/sfw/bin".  What is the advantage of one over the other?
[23:49:08] *** laluciol has joined #opensolaris
[23:49:35] <holcomb> export VAR is like a global variable, just VAR= is like a local variable
[23:49:44] <Bartman007> Red_Cloud: the first one add /usr/sfw/bin to the path, the second sets the path to /usr/sfw/bin
[23:50:22] <elektronkind> Red_Cloud: when /bin/sh is the shell "export VAR=blah" doesnt work, where is does in more modern bash.
[23:50:37] <Red_Cloud> Bartman007 does "export" add it to the end of the path?
[23:50:38] <elektronkind> Red_Cloud: so in /bin/sh, you must "VAR=blagh; export VAR"
[23:51:10] <Bartman007> Red_Cloud: oh! I didn't even notice that one used `export' and the other didn't.
[23:51:12] <elektronkind> the latter also works in bash, btw. you can do it both ways there
[23:51:32] <Bartman007> Red_Cloud: I was focusing on the differences in the values of the variable.
[23:52:00] <Bartman007> PATH=/usr/sfw/bin:$PATH  !=  PATH=/usr/sfw/bin
[23:52:04] <Red_Cloud> Bartman007 not to worry, I need to learn the differences in the values also.
[23:52:24] *** mritun_ has joined #opensolaris
[23:54:20] <_Megaf> secondlife on OpenSolaris?
[23:54:52] <_Megaf> http://www.opensolaris.org/jive/thread.jspa?messageID=144300&tstart=0
[23:55:02] <e^ipi> why not, the client's open-source
[23:55:05] *** jlc has quit IRC
[23:55:28] <e^ipi> why you'd want to use secondlife at all is a different story
[23:56:12] <Red_Cloud> Bartman007 and elektronkind I just located an instruction manual which might help me quite a bit:  http://www.decf.berkeley.edu/help/unix/searchpath.html
[23:56:28] <Fetch> hmm, Furries + OpenSolaris
[23:56:34] <Fetch> Enterprise furry stability
[23:56:39] <Bartman007> Red_Cloud: oh, so you didn't understand the purpose of the PATH variable?
[23:58:03] <Red_Cloud> Bartman007 yes I understand how it allows search in your path but the intricacies involved have me confused quite a bit.  Such as PATH=/blah/$PATH  and export PATH and so forth.
[23:58:17] <jbk> hello
[23:58:56] <Bartman007> Red_Cloud: well, that largely depends on what shell you are using.
[23:59:15] <Red_Cloud> I am about to chose my shell for my user and root and want to set my paths but the syntax involved varies so much that I really need to fully understand.  I am thinking of bash
[23:59:19] <Fetch> should there be any issue doing a LU from snv64a to snv69?
[23:59:53] *** sstallion has quit IRC





top