[00:00:17] <CIA-26> gd78059: Codemgr_wsdata/putback.cmt
[00:00:18] <CIA-26> dhain: 6567270 SUNWefcr cannot be removed if the efdaemon service is enabled
[00:04:08] *** hile_ has joined #opensolaris
[00:09:41] <Gman> gdamore, you should at least register your sun id now, so that your commits show up under your real name :)
[00:10:34] <alanc> but then we would taunt him more about using the wrong putback flag for his comments
[00:11:04] * Gman couldn't care less :)
[00:11:30] <jbk> evening
[00:11:47] <hile_> afternoon alanc
[00:12:11] <alanc> afternoon
[00:12:21] <hile_> my brain hurts... i was looking at xterm source code over the weekend to figure out htf work does something they do
[00:12:42] <hile_> they've got a modified xterm that does command-linei loggng
[00:13:04] <hile_> this source code hurts my brain
[00:13:30] *** bengtf_ has joined #opensolaris
[00:13:31] <flyingparchment> in format, i ran a 'compare' test, and it shows the status as numbers: 243/55/0.  what do they mean?
[00:13:36] *** bengtf_ is now known as bengtf
[00:14:15] <alanc> seems like a silly place to put command logging - why not just use auditing or put it in the shell?
[00:14:30] <hile_> oh there's logging in the shell too
[00:14:31] <alanc> especially since xterm's source code makes everyones head hurt
[00:14:50] <hile_> but when all the admins root window xterms are gained by a given wrapper, it's a best-case logging
[00:15:03] <hile_> turning out auditing and getting *just* commands entered isn't quite so easy
[00:15:42] <the-decider> hile_: its best to not try to make sense of the xterm source.
[00:15:47] <hile_> Personally, I'd  ratherdo it with rxvt
[00:15:54] <hile_> but I need to figure out where the hell it happens first :)
[00:16:03] <the-decider> xterm, and all of its derivitives, are rather horrendous.
[00:17:08] <kjetilho> hile_: I just read an old article in :login; about a modified script(1)
[00:17:25] <kjetilho> it could even send keypresses in realtime to syslog over UDP :)
[00:17:44] <delewis> the user logged in could easily kill script, though.
[00:17:49] <the-decider> ...or one of those nifty tty loggers ;)
[00:19:00] <sommerfeld> jim gettys once stated in my presence that portions of xterm are derived from vt100 firmware.
[00:19:06] <kjetilho> delewis: yes, but not until it had logged it had been started
[00:19:53] <kjetilho> here's the article: http://www.usenix.org/publications/login/2006-04/pdfs/langley.pdf
[00:19:59] *** sfire||mouse has joined #opensolaris
[00:20:09] <delewis> sommerfeld: I'm somewhat skeptical about that. Portions, possibly, but xterm seems to emulate vt100 better than a real vt100. :-)
[00:20:23] <delewis> anyone that's actually done a Solaris install using a real vt100 knows this.
[00:21:00] <kjetilho> Solaris install uses the "sun" terminal type
[00:21:21] <delewis> kjetilho: in this case, vt100 had been explicitly selected.
[00:21:33] <kjetilho> ok
[00:22:09] <delewis> and curses still looked god-awful. The fields weren't quite right either, even after playing around with the tab-delimiter and the number of spaces in a tab.
[00:22:16] *** vmlemon has quit IRC
[00:22:30] *** nrubsig has joined #opensolaris
[00:22:31] *** ChanServ sets mode: +o nrubsig
[00:22:49] <nrubsig> stevel: ping!
[00:22:52] * delewis had the 'pleasure' of using a vt100 last year
[00:23:00] *** jmcp_ has joined #opensolaris
[00:23:05] <nrubsig> delewis: congratulations!
[00:23:06] <delewis> I finally dragged in an old Wyse and used it, instead.
[00:23:36] <nrubsig> delewis: what was wrong with the VT100 ?
[00:23:46] <sfire||mouse> for the thumper, and ipmi, does it run off of the normal network ports, or off of the net mgt port?
[00:23:46] <delewis> nrubsig: ever done a Solaris install with a real vt100?
[00:24:01] * nrubsig giggles
[00:24:07] <nrubsig> delewis: ahhhhhhh
[00:24:08] <delewis> nrubsig: pretty much everything but a real vt100 does a better job of vt100 than a real vt100. :-)
[00:24:48] <nrubsig> delewis: try a VT220 next time - more keys, more fun!
[00:25:01] <alanc> I seem to have finally repressed the memories, but I think tvi925 was the terminal I hated most, for having break too close to backspace... make a typo, lose your connection
[00:25:07] <delewis> nrubsig: only had vt100s laying around and a few re-branded Wyse's that had no problem emulating vt100.
[00:25:21] <the-decider> I remember the actual vt100s having such miserable squeeky keys
[00:25:30] <delewis> our university had a PDP-11/70 back in the day (we were a huge DEC shop)... the PDP-11/70 went away and all the vt100s we purchased (about 100) went into a storage closet.
[00:25:32] <nrubsig> delewis: if you don't want the Vt100 anymore send it to me, please.
[00:25:34] <kjetilho> the vt100 is simply slow, you don't want to run it at much faster than 9600 since it can lose characters
[00:25:48] <delewis> I found myself needing a serial terminal last year to do an install on an old Ultra 1 with no framebuffer... asked around, and discovered the 'vt100 closet' :-)
[00:25:57] <kjetilho> and it big and bulky as well, compared to the nice little vt320's
[00:26:06] <delewis> nrubsig: do you know how heavy vt100s are?
[00:26:18] <nrubsig> delewis: yes
[00:26:25] <delewis> they aren't very good terminals, anyway. As I've said, almost everything emulates vt100 better than real vt100. :-)
[00:26:36] <kjetilho> if you think the vt100 is heavy, you should try to carry the vt05
[00:26:36] <nrubsig> delewis: that's why I want you to take the shipping costs.
[00:26:44] <delewis> nrubsig: :-)
[00:26:51] * the-decider thought it was rather functional for news-reading marathons
[00:27:09] <delewis> the-decider: in that case it might just be the Solaris termcap is buggy.
[00:27:28] <delewis> but anyway, the Solaris installer with vt100 emulation was not functional on a real vt100.
[00:27:30] <nrubsig> stevel: steeeeeeeeeeveeeeeeellllllllllll<plock>
[00:27:41] <the-decider> well, 'rn' wasn't exactly a big user of terminal functionality
[00:28:23] * nrubsig had the pleasure to play "GNAW!" on a vt320 ...
[00:28:39] * wesolows spends most of his time playing GNAW!
[00:28:40] <the-decider> what frustrated me for awhile was that the 'xterm' termcap that shipped in most linux distributions wasn't really functional on run-of-the-mill 'xterm's
[00:29:05] <nrubsig> wesolows: ?!
[00:29:08] <nrubsig> wesolows: really ?!
[00:29:13] <wesolows> yeah
[00:29:31] <nrubsig> wesolows: erm
[00:29:34] <nrubsig> wesolows: you know...
[00:29:35] <wesolows> I do it so I can look busy without really working
[00:29:47] <nrubsig> wesolows: ... you could sponsor some putbacks them.
[00:29:52] <nrubsig> s/them/then/
[00:29:55] <wesolows> no, I prefer to play GNAW!
[00:30:11] <nrubsig> ahhh,stubborn child...
[00:30:18] * nrubsig looks for candies and the whips
[00:30:21] <delewis> sponsor a ksh93-related putback that commits GNAW :-)
[00:30:30] <wesolows> GNAW! is fun; sponsoring putbacks, not so much
[00:30:33] <delewis> then you can have GNAW on every future release of Solaris!
[00:30:45] <wesolows> what future releases of Solaris?
[00:31:09] * wesolows is wondering what happens when we reach nv_(2**32-1)
[00:31:09] <delewis> 11, perhaps?
[00:31:16] <delewis> hah.
[00:31:23] <delewis> EOVERFLOW
[00:31:34] <wesolows> maybe then marketing will have to let us ship it
[00:31:47] <delewis> marketing is the hold-up?
[00:31:58] <alanc> I thought GNAW went in with ksh93?
[00:32:10] <the-decider> they're trying to decide if they want to re-name it the Solaris JAVA OS
[00:32:13] <wesolows> I always assumed they decide what gets shipped and when.  Maybe not - the whole thing is kind of a black box to me.
[00:32:26] <the-decider> and re-spin all of the packages with the JAVA prefix instead of SUNW
[00:32:33] <delewis> I always thought it was an engineering decision when it was time to ship the next minor release.
[00:32:45] <delewis> that would make sense, but then again, nothing makes sense in corporate America.
[00:33:07] <kjetilho> next release won't be 11 anyway.  it's well known that odd numbered Solaris releases aren't installed by customers, and the number 11 has been "tainted" by the Solaris Express releases.
[00:33:10] <alanc> more of management & marketing setting the desired ship date and feature set and engineering working to make that and seeing how far they slip
[00:33:12] <nrubsig> alanc: yes...
[00:33:14] <wesolows> Not really.  There's some influence, but we don't just get to say we're done with this release now go name, package, ship, and market it.
[00:33:21] <delewis> kjetilho: this is true.
[00:33:25] <stevel> nrubsig: pong
[00:33:40] * delewis sees very few Solaris 7 and 9 systems out in the field
[00:33:41] <nrubsig> stevel: did you see my email about XAUTHORITY ?
[00:33:46] <stevel> nrubsig: nope
[00:33:49] <delewis> there's another reason for not installing odd releases, though.
[00:34:00] <alanc> think some of our package scripts will need a little work to go past build 100 since they assumed two-digit build numbers in the package versions
[00:34:04] <wesolows> kjetilho: True; the odd-numbered releases are the good ones, but customers prefer to use the more broken even-numbered ones so that they can cost us more money in support.
[00:34:04] <flyingparchment> the next release will be Solaris XI
[00:34:08] <doc_nl> 11 is the silly number anyway
[00:34:23] <nrubsig> stevel: http://mail.opensolaris.org/pipermail/opensolaris-code/2007-September/006102.html
[00:34:24] <kjetilho> wesolows: hehe, a little bitter are we? :)
[00:34:34] <delewis> and it's not really they're tainted, but simply the time between an even release and an odd release is less than the time it takes for a customer to cycle through hardware to require a specific, newer release of Solaris or to get it certified for whatever application they're using.
[00:34:36] <wesolows> kjetilho: You want bitter, ask the S9 tech lead. :-)
[00:34:45] <kjetilho> lol
[00:35:06] *** SirFunk has joined #opensolaris
[00:35:09] <alanc> delewis: we fixed that by moving Solaris from releasing every year to 3+ years between releases
[00:35:15] <delewis> many customers still haven't moved off Solaris 8 *cough* GE *cough*
[00:35:18] * the-decider was pretty happy with 9 -- until 10 came out.
[00:35:51] <delewis> of course all of the hardware GE installs nowadays for a customers is now EOL'd.
[00:35:54] <kjetilho> I don't remember any killer features in 9.  was it SVM integration?
[00:36:01] <delewis> (typically, v480s or E450s)
[00:36:09] <delewis> and v240s, unless they've changed that in the last year.
[00:36:16] <alanc> you don't want to know how many customers paid extra for the DST change patches for Solaris 2.5.1 & 2.6...
[00:36:20] <nrubsig> kjetilho: some localisation stuff was better
[00:36:26] <delewis> alanc: doesn't surprise me a bit.
[00:36:36] <delewis> stubborn companies that pay stubborn vendors.
[00:36:45] <doc_nl> :X
[00:36:45] <kjetilho> nrubsig: okay, Solaris never supported my locale anyway, so I don't care
[00:36:46] <alanc> at least I haven't heard of any customers seriously using SunOS 4 in a couple of years
[00:36:47] <the-decider> SVM was the big thing.  Oh, and it installed faster -- that 8_Recommended set was getting huge.
[00:36:48] <nrubsig> stevel: where is the master source for that README ?
[00:36:55] <nrubsig> kjetilho: erm
[00:36:59] <nrubsig> kjetilho: which locale are you ?
[00:37:06] <kjetilho> nn_NO
[00:37:18] <stevel> http://opensolaris.org/os/community/on/install_quickstart/
[00:37:40] <nrubsig> kjetilho: huh ?
[00:37:55] <stevel> fixed
[00:37:58] <nrubsig> kjetilho: did you see nn_NO.UTF-8 yet ?
[00:38:00] <nrubsig> stevel: thanks!
[00:38:30] <kjetilho> nrubsig: it really doesn't matter.  localising output just breaks scripts.
[00:38:31] <movement> wesolows: ping
[00:38:34] <wesolows> movement: pong
[00:38:55] <kjetilho> Unix has never been good at separating data and presentation.
[00:38:57] <nrubsig> kjetilho: in which way does it break scripts ?
[00:39:15] <movement> wesolows: edquota has had a gcc build error for what looks like a long time. it seems that the shadow build doesn't kick in, since the Makefile directly builds edquota from edquota.c
[00:39:20] <movement> wesolows: does that sound reasonable?
[00:39:26] <kjetilho> e.g. numbers being printed as 23 443 564,78
[00:39:27] <wesolows> movement: yes, that's unsurprising
[00:39:42] <nrubsig> kjetilho: POSIX's LC_* model should be quote good in data and presentation seperation.
[00:39:50] <wesolows> movement: There's a bug filed against cw to make it compile in this case, but no one has done it.
[00:40:02] <movement> wesolows: OK, thanks, I'll dig that bug out.
[00:40:13] <kjetilho> nrubsig: if you set it the locale to C, yes.
[00:40:14] <nrubsig> kjetilho: could you please post a complaint to shell-discuss at opensolaris dot org witi i18n-discuss at opensolaris dot org in the CC: , please ?
[00:40:15] *** doc_nl has quit IRC
[00:40:36] <kjetilho> nrubsig: no, because I think it is broken by design to do this in the shell
[00:40:37] <wesolows> movement: Basically if the "normal" command would trigger linking, the shadow compiler is not invoked.  This is as true for cc -o foo foo.c as it is for cc -o foo foo.o
[00:40:44] <nrubsig> kjetilho: no, I mean set everything to en_US.UTF-8 and LC_MESSAGES=nn_NO.UTF-8
[00:41:31] <movement> 6407174
[00:41:31] <movement> thanks
[00:41:38] *** g4lt-mordant is now known as unbewont
[00:42:01] <kjetilho> some scripts will still do stuff like file * | grep "not stripped"
[00:42:02] <Auralis> unbewohnt
[00:43:22] <nrubsig> kjetilho: in such cases you use (LC_ALL=C file *) | ...
[00:43:29] <nrubsig> or LC_MESSAGES
[00:44:29] <kjetilho> I can't recall *ever* seeing a script do that
[00:44:46] <nrubsig> I did.
[00:44:55] <kjetilho> don't tell me, you wrote it :)
[00:45:21] * nrubsig adds an item to his shell style guide
[00:45:22] <kjetilho> well actually I remember doing it once to make tr(1) behave.  those intervals are just too weird
[00:47:06] *** yarihm has quit IRC
[00:47:15] <nrubsig> kjetilho: use /usr/xpg4/bin/tr instead - it will behave in all cases without the LC_ALL=C workaround.
[00:47:24] <nrubsig> kjetilho: /usr/bin/tr is junk.
[00:47:39] <nrubsig> kjetilho: or use ksh93 builtin operators for string manipulation.
[00:47:56] <kjetilho> so a-z and A-Z will actually have the same length in XPG4 tr?
[00:48:14] <nrubsig> kjetilho: AFAIK yes.
[00:48:29] <kjetilho> I guess there are no such guarantees for :alpha:
[00:49:23] <kjetilho> I meant :lower: and :upper: of course
[00:49:33] <nrubsig> kjetilho: /usr/bin/tr is a thing where noone really understands what is does and noone dares to fix it since it may break ancient scripts which rely on the bugs.
[00:51:41] <nrubsig> Does anyone know the ARC case number for the javaexec kernel module ?
[00:51:48] <sommerfeld> whoever thought using case-insensitive collation sequences for tr character ranges made sense wasn't thinking.
[00:53:28] <sommerfeld> nrubsig: 1997/123
[00:53:35] <nrubsig> sommerfeld: any URL ?
[00:53:51] *** perlmonk has quit IRC
[00:56:59] <sommerfeld> It's listed in http://www.opensolaris.org/os/community/arc/caselog/1997/
[00:57:35] <nrubsig> kjetilho: take a look at http://svn.genunix.org/repos/on/branches/ksh93/gisburn/scripts/xmldocumenttree1.ksh
[00:57:50] <nrubsig> kjetilho: function "attrstrtoattrarray"
[00:57:55] <sommerfeld> Follow the process at http://www.opensolaris.org/os/community/arc/arc-faq/arc-publish-historical-checklist/ to ask for it to be published
[00:58:52] *** sfire||mouse has quit IRC
[01:00:16] <CIA-26> akolb: 6598427 waitq_runall() doesn't like being interrupted
[01:00:41] *** Triskelios has quit IRC
[01:01:40] *** pfa3rh has quit IRC
[01:07:30] <nrubsig> is there any userland API to get the L1 cache line size ?
[01:09:05] <stevel> some CPUs report it via cpuid i thought
[01:09:25] <nrubsig> stevel: You did only fix the master file of http://dlc.sun.com/osol/on/downloads/b72/README.opensolaris , right ?
[01:09:39] <stevel> yeah - i'm not gonna go regenerate all the existing README.opensolaris files
[01:09:42] <stevel> it'll be fixed in the next delivery
[01:09:56] <nrubsig> stevel: well, I am looking for an API within Solaris userland...
[01:10:48] <nrubsig> stevel: preferably something "portable" (e.g. portabe between Solaris/x86, Solaris/SPARC, Solaris/PPC, Solaris/S390, Solaris/ARM, Solaris/SX-8 etc.)
[01:13:51] <nrubsig> Linux seems to provide such infos via /usr/bin/getconf ...
[01:14:17] * nrubsig wonders in which color Don Cragun's face would turn if he'd ask him for such an extension ...
[01:14:20] <nrubsig> pink ?
[01:14:21] <nrubsig> green ?
[01:14:25] <nrubsig> red ?
[01:14:45] * nrubsig hides under the couch and stops imagining that...
[01:15:41] <alanc> sounds like something only useful to people over-optimizing their code
[01:16:08] <stevel> alanc: HPC & scientific computing folks like to know
[01:17:22] <nrubsig> stevel: yes, but preferably not the way how Linux implemented it.
[01:18:47] <brendang> nrubsig: well,
[01:19:03] <brendang> nrubsig: there should be a user-land way to fetch L* cache sizes easily
[01:19:23] *** bondolo has quit IRC
[01:19:24] <brendang> nrubsig: I argued not that long ago for this to be in kstat - on one of the opensolaris threads
[01:19:44] <brendang> nrubsig: the old fashioned way was to dig around with prtdiag and adb/mdb
[01:19:59] <brendang> nrubsig: and was very un
[01:20:08] *** rikochet has quit IRC
[01:20:11] * alanc wonders what such programs would do if they migrated between cpus in a system with different cache sizes
[01:20:22] *** bondolo has joined #opensolaris
[01:21:10] *** yongsun has joined #opensolaris
[01:21:52] <nrubsig> alanc: erm... are there such systems where cache sizes differ (exlcuding UltraSPARC-3/4 machines) ?
[01:21:55] <brendang> alanc: I hope their performance would decrease, because I hope that the compiler has taken cache size into account. although, I don't know how good compilers are at doing that.
[01:22:03] *** plarsen has quit IRC
[01:22:30] <brendang> nrubsig: hardware cache sizes do differ a lot
[01:22:55] <alanc> I thought that mixing CPUs with different cache sizes happened in the same SPARC systems you could mix CPUs of different speeds
[01:22:58] <sommerfeld> nrubsig: i'm using one right now.
[01:23:03] <nrubsig> brendang: yes, on the other side compilers like Sun Studio have options like -xtarget=native
[01:23:22] <sommerfeld> one 600mhz with 4MB cache, one 750mhz with 8MB cache.  long story.
[01:23:55] <nrubsig> sommerfeld: well, at least you have a sparc with more than one CPU
[01:23:56] <brendang> nrubsig: they have options - but what does it actually do. considering that there isn't a standard user-interface to get these details, and the compilers run as non-root
[01:23:57] <Auralis> hehe, sound slike mine
[01:24:20] <alanc> I'd also wonder if CPU makers would start enabling/disabling cache on the fly on laptops to save power, much like they change CPU speed, but that's probably overkill
[01:24:35] *** pauliukas_ has joined #opensolaris
[01:24:46] <alanc> for the Sun compilers, I thought they just had builtin knowledge about the cache sizes of SPARC CPU's Sun had shipped
[01:24:47] <pauliukas_> Why should I use OpenSolaris instead of Linux?
[01:24:47] <nrubsig> brendang: I guess it checks the CPU name or something like that and then fetches a pre-configured set of options
[01:24:50] <brendang> alanc: interesting... we need power by CPU component metrics..
[01:24:59] <kjetilho> nrubsig: what do you want this information for?
[01:25:13] <brendang> nrubsig: ok. I guess/hope to. :)
[01:25:38] *** EtherNet_ has quit IRC
[01:25:42] <kjetilho> pauliukas_: look at the feature list
[01:25:57] <pauliukas_> Got any recommendations on where I could find it?
[01:26:06] <jmcp_> pauliukas_: sun.com
[01:26:11] <pauliukas_> oh...
[01:26:13] <pauliukas_> Hahah. okay.
[01:26:26] <jmcp_> http://developers.sun.com/sxde/
[01:26:27] <nrubsig> kjetilho: you mean cache size ?
[01:26:34] <kjetilho> nrubsig: yes.
[01:26:59] <kjetilho> in particular, L1 size
[01:27:00] *** derchris^ has joined #opensolaris
[01:27:19] *** derchris has quit IRC
[01:28:37] <pauliukas_> But more particularly about this open source aspect... I heard there are a few Solaris distros. Which one should I try out first? Any differences? Any comparison table?
[01:28:56] <kjetilho> it seems easier to make better decisions when implementing algorithms based on L2/L3 cache size (whichever is larger)
[01:29:09] <pauliukas_> Or should I just use the (real, orginal) Solaris?
[01:29:28] <jmcp_> pauliukas_: I can't comment on which is best - only you can make that choice. www.genunix.org is another good place to look
[01:29:49] <nrubsig> pauliukas_: I would try BeleniX first, then Martux
[01:30:36] <kjetilho> pauliukas_: first try official Solaris to get a feel for it, then when you feel like hacking, choose whichever fits the direction you want to go in
[01:30:50] *** Triskelios has joined #opensolaris
[01:33:50] *** EtherNet- has joined #opensolaris
[01:37:47] *** karrotx has joined #opensolaris
[01:41:41] *** myrkr has quit IRC
[01:45:47] *** FBdev has joined #opensolaris
[01:47:07] *** e^ipi has joined #opensolaris
[01:47:33] <e^ipi> heyas
[01:48:22] *** NikolaVeber has quit IRC
[01:57:52] *** uebayasi has joined #opensolaris
[01:58:23] <mick_work> anyone have issues w/ solaris not rendering text in qemu?
[02:01:03] <mick_work> CDE starts and everything - but the Console's text is all out of wack
[02:01:21] <mick_work> menu items render just fine
[02:03:18] *** charlesnw has quit IRC
[02:07:56] *** e^ipi_ has joined #opensolaris
[02:08:04] *** estibi_ has quit IRC
[02:08:31] <axisys> anycan vouche for this `A general rule of thumb is that load averages that are persistently above 4 times the number of CPUs will result in sluggish performance' according to this http://www.princeton.edu/~unix/Solaris/troubleshoot/cpuload.html
[02:08:46] <jmcp_> bollocks
[02:09:39] <axisys> where did he get that 4 times? i thought anything over 1 would be cpu utilization is high and if vmstat's r column shows `0' cpu saturation is not high
[02:09:52] <axisys> and we should be ok
[02:09:57] <jmcp_> if I've got a t2000 running a squillion apache instances, I would expect to see load averages well above 150 and still have plenty of interactive response
[02:10:02] <Tempt> Wow, spam of the morning's text says: Any clodhopper can secretly admire beyond omphalos, but it takes a real turkey to cough syrup beyond guardian angel.
[02:10:22] <e^ipi_> wiser words were never spoken
[02:10:25] <jmcp_> Tempt: that's gold
[02:10:27] <nrubsig> jmcp_: can I get the T2000 ?
[02:10:28] <axisys> jmcp_: 150% u mean?
[02:10:37] <jmcp_> axisys: no, 150. not a percentage
[02:10:42] <pauliukas_> lol
[02:10:45] <axisys> jmcp_: wow!!!
[02:10:52] <Tempt> jmcp_: There was one last night about mastodons.
[02:11:01] <jmcp_> nrubsig: no, I'm not sharing
[02:11:15] <jmcp_> Tempt: good to know that extinct animals are being brought back
[02:11:16] <pauliukas_> The question is... Why are you running a squillion apache instances.
[02:11:19] <nrubsig> jmcp_: not even if I snap your neck and hide the corpse ?
[02:11:24] <jmcp_> pauliukas_: it was an "if"
[02:11:33] <jmcp_> nrubsig: not even then. I hide my hardware very well
[02:11:36] <nrubsig> pauliukas_: server ? load testing ?
[02:11:37] <jmcp_> pauliukas_:
[02:11:42] <jmcp_> pauliukas_: a gedanken-experiment
[02:11:43] <pauliukas_> oh... I misread that.
[02:11:53] <pauliukas_> Gotcha. I thought your site was getting hammered or something.
[02:11:56] <brendang> nrubsig: BTW, here was the thread about CPU cache stats, and my suggestion to put it into kstat: http://www.opensolaris.org/jive/thread.jspa?messageID=136102&#136102
[02:12:05] <nrubsig> jmcp_: s/gedanken/mutenten/
[02:12:10] <nrubsig> brendang: ewwwwww
[02:12:15] <nrubsig> brendang: Jive
[02:12:17] <nrubsig> brendang: ewwwwww
[02:12:22] <nrubsig> brendang: Jive
[02:12:25] <nrubsig> brendang: ewwwwww
[02:12:26] <axisys> jmcp_: ok so over 100% would be still high utilization but with no latency and no cpu saturation nothing to worried ..correct?
[02:12:32] <pauliukas_> JIVE!?!?!?
[02:12:36] <pauliukas_> WHY OH WHY!
[02:12:56] <brendang> nrubsig: oops, I should really dig it from the archives
[02:13:08] <pauliukas_> http://www.jivesoftware.com/pricing/index.jsp
[02:13:13] <pauliukas_> OpenSolaris = OpenSource.
[02:13:14] <nrubsig> pauliukas_: "Jive", it's the forum software at opensolaris.org and I am NOT the biggest fan of that stupid junkyerd of software.
[02:13:17] <pauliukas_> Let's get SMF or something.
[02:13:28] <pauliukas_> nrubsig: I know... Me neither.
[02:13:32] <pauliukas_> I hate Jive.
[02:13:56] <pauliukas_> vBulletin is my favorite forum software, but it's also closed-source.
[02:13:56] * jmcp_ concalls
[02:14:10] <pauliukas_> But seriously, for an open source project... It should have been an open source forum.
[02:14:17] <nrubsig> Welcome to the JIVE HATERS CLUB!
[02:14:23] <Tempt> I found something even better than vBulletin
[02:14:26] <Auralis> heres your badge
[02:14:26] <Tempt> It's this forum, right
[02:14:30] <pauliukas_> Teknix: Are you serious?
[02:14:31] <Tempt> and you can choose how to access it
[02:14:32] <pauliukas_> Err...
[02:14:34] <pauliukas_> *Tempt
[02:14:36] <nrubsig> pauliukas_: send complains to stevel ...
[02:14:39] * nrubsig hides
[02:14:40] <Tempt> with a range of software for all sorts of different machnes
[02:14:44] <pauliukas_> who's stevel?
[02:14:53] *** nrubsig was kicked by stevel (stevel)
[02:15:00] *** nrubsig has joined #opensolaris
[02:15:01] *** ChanServ sets mode: +o nrubsig
[02:15:04] <Tempt> In fact, just about any OS from AIX to Xenix can participate in this "forum"
[02:15:10] <pauliukas_> Ops kicking ops?
[02:15:15] * pauliukas_ blinks a few times
[02:15:33] <pauliukas_> Tempt: So what forum is better than vBulletin?
[02:15:51] <nrubsig> pauliukas_: erm, I am ranting about Jive too much...
[02:16:06] <pauliukas_> Let's do a petition to get rid of Jive.
[02:16:10] <bda> ...
[02:16:20] <richlowe> nrubsig: no, you suggested stevel fixed jive.
[02:16:24] <richlowe> nrubsig: which is considerably worse.
[02:16:28] <richlowe> s/fixed/fix/
[02:16:32] <nrubsig> heh
[02:16:35] <bda> http://www.reuters.com/article/oddlyEnoughNews/idUSEIC47086020070904 # !!
[02:16:36] <pauliukas_> I wouldn't invest nor use a OS which uses Jive for it's forums!
[02:16:37] <stevel> nobody likes jive, we just don't have the resources to deploy something better
[02:16:51] <richlowe> and, for some reason, they won't just remove the damn thing.
[02:17:01] <stevel> pauliukas_: if you derive the quality of an OS from whatever they use as a forum software, then you have bigger issues
[02:17:02] <pauliukas_> So Sun is controlling opensolaris.org?
[02:17:11] <Tempt> pauliukas_: Well, it runs on different software, too, you can choose the software that suits your needs
[02:17:17] <stevel> would you stop using Apache because they use a closed-source bug tracker?
[02:17:18] <pauliukas_> stevel: Hahah. I'm extremely biased.
[02:17:25] <Tempt> pauliukas_: And the protocols are based on RFCs - NNTP and NNRP
[02:17:27] <pauliukas_> What? Apache uses a close....
[02:17:30] <pauliukas_> That's it, I'm switching to lighttpd.
[02:17:50] <stevel> ... and the sites undoubtedly run on closed-source hardware.  are you going to stop using that too?
[02:17:58] <pauliukas_> Ahhh!
[02:18:03] * pauliukas_ jumps off a bridge
[02:18:12] <pauliukas_> It's just that I hate particularly Jive.
[02:18:19] <delewis> actually, the Apache Foundation purchased a T2000 awhile back to replace one of their Itanium web servers.
[02:18:19] <alanc> pauliukas_: Sun runs the opensolaris.org webserver - opening up maintainence of the site software to the community is on the todo list
[02:18:19] <pauliukas_> If it were vB or IPB, I'd be happy.
[02:18:34] <richlowe> alanc: if anyone is crazy enough to want to.
[02:18:35] <pauliukas_> alanc: Gotcha. That would be a great thing.
[02:18:36] <stevel> that's fine. i hate it too - but separate that from the project itself
[02:18:42] <Auralis> well, in that case, he pretty much must run ultrasparc-t1 or -t2 systems to be truely free, the cpu is GPL after all :)
[02:18:49] <Tempt> In other words, why can't they just deploy INN and we'll all play nicely.
[02:19:09] * nrubsig wonders whether it would help to assign all Jive bug reports to stevel
[02:19:10] * nrubsig RUNS
[02:19:15] *** nrubsig was kicked by stevel (no. it wouldn't.)
[02:19:16] <pauliukas_> Jive is just... So expensive for such a P.O.S.
[02:19:25] *** nrubsig has joined #opensolaris
[02:19:26] *** ChanServ sets mode: +o nrubsig
[02:19:35] <alanc> richlowe: I imagine, like the first few community contributions to OpenSolaris itself, the first batch of community contributions to the website will be tiny fixes for personal annoyances
[02:19:41] <pauliukas_> It's as if they go sleep with companies to get them to use their software.
[02:19:52] <pauliukas_> when these companis/corporations could use other forum products!
[02:20:05] <alanc> like I might actually fix the bug I filed that RSS feeds should use the feed reader icon instead of the ancient orange xml box
[02:20:06] <sommerfeld> Or golf with them.
[02:20:11] <delewis> Jive is more than just forum software.
[02:20:12] <Tempt> Does nobody else like the idea of just using news and letting people choose their client and running one of the many NNRP<>forum gateways?
[02:20:17] <delewis> it's a tie-in to a mailing lists.
[02:20:23] <delewis> s/ a /  /
[02:20:32] <pauliukas_> Ah...
[02:20:38] <pauliukas_> But vB can do that too. Case closed.
[02:20:51] * nrubsig is away
[02:20:54] <delewis> vB isn't free, either.
[02:20:57] <pauliukas_> I know.
[02:21:03] <pauliukas_> But it doesn't cost 16K$!!!
[02:21:22] <delewis> I really don't care about web users being able to post to the mailing lists.
[02:21:32] <delewis> that should be a subscriber-only thing, using a mail client.
[02:21:32] <axisys> jmcp_: load avergae 150 would be 478%
[02:21:38] <axisys> jmcp_: on a t2000
[02:21:47] <stevel> pauliukas_: where does vB support mailing list gateways?
[02:21:51] <sommerfeld> bda: problem with the scsi chain on the 757 perhaps?
[02:21:52] <wesolows> you know, that debate's been going for 3 years now
[02:21:53] <stevel> i don't see that on their site..
[02:21:56] <pauliukas_> stevel: It's a third-party plugin.
[02:22:02] <pauliukas_> Look on www.vbulletin.org
[02:22:07] <axisys> jmcp_: of it does not stay for full 15 mins .. it becomes negligible
[02:22:08] <Tempt> delewis: NEWS READER!
[02:22:11] <pauliukas_> It's OSS and doesn't cost, but it's third party.
[02:22:23] <axisys> jmcp_: s/of/if/
[02:22:23] <delewis> in that case we could just use the standard pipermail and given web users access to the archives only.
[02:22:25] <bda> sommerfeld: You use chickens for SCSI, not goats.
[02:22:26] <pauliukas_> I've seen a few sites use the feature to make it seem like there's more posts.
[02:22:59] <delewis> if you're not 'savvy' enough to subscribe to a mailing list, why are you interested in OpenSolaris?
[02:23:10] <richlowe> alanc: like jive!
[02:23:24] <Tempt> If you're not 'savvy' enough to understand that a newsserver is the *correct* way to do this, you shouldn't be debating the issue.
[02:23:33] <Tempt> This isn't a closed discussion, why use a mailinglist?
[02:23:38] <sommerfeld> "SCSI is NOT magic. There are fundamental technical reasons why it is necessary to sacrifice a young goat to your SCSI chain now and then."
[02:23:49] <sommerfeld> heard it years ago
[02:23:58] <pauliukas_> I find that a forum is just more elegant and easier to use than a mailing list.
[02:24:05] <stevel> it's all subjective.
[02:24:06] <pauliukas_> We live in the 21st century, people.
[02:24:15] <stevel> and personal preference
[02:24:20] * stevel shuts down mailman AND jive
[02:24:20] <Tempt> Yes, the 21st century
[02:24:21] <stevel> go use smoke signals
[02:24:24] *** e^ipi has quit IRC
[02:24:28] <pauliukas_> lol
[02:24:32] <Tempt> Where everyone is meant to be as fucking dumb as possible to 'leave no child behind' or some such shit
[02:24:37] <stevel> or carrier pigeons
[02:24:38] <bda> sommerfeld: You're still using the purple book. You want the blue book. Goat market went bearish. Chickens are up for connective esanguination.
[02:24:39] <Tempt> C'mon guys, we're meant to be *technical* people
[02:24:54] <richlowe> Jive could be fixed to be less annoying
[02:24:59] <richlowe> (for people not actually using it, obviously)
[02:25:06] <alanc> I find that having to go visit forum sites regularly to see what's new is a waste of time and causes me to not remember to go - forums with RSS feeds are barely passable, but I still prefer e-mail to keep in touch with what's going on
[02:25:43] <alanc> RSS feeds don't thread as well as my mail reader does though
[02:25:54] <alanc> (or my news reader, though I don't use it as much as I used to)
[02:26:07] <bda> We should just replace it all with threaded IRC.
[02:26:25] <axisys> jmcp_: 468% actually .. my bad
[02:29:21] * jmcp_ wonders wtf all the fuss is about load averages
[02:29:43] <jmcp_> the "load average" concept is really irrelevant in these days of dma, multithreaded cpus, multithreaded IO ....
[02:30:33] <bda> When you're running a system where iostat lies to you, it may very well be the only useful mentric you have.
[02:30:45] <bda> For some unlikely value of useful.
[02:30:50] * jmcp_ snorts
[02:31:03] <wesolows> load average tells you something, but not a whole hell of a lot
[02:31:32] <bda> It's sort of like Arthur Dent in that regard.
[02:31:37] <bda> "Hello, something seems to be happening."
[02:33:16] <nrubsig> erm, is it usefull to write something like this in a request-sponsor email: "... this is only easy janitor/cleanup work... if you want to become an OS/Net sponsor this is your chanche to get an easy training item..."
[02:33:19] <nrubsig> ?
[02:33:48] <sommerfeld> indeed.  xload is sort of vaguely useful in spotting runaway processes on what should be an idle system (load asymptotic to N rather than to 0)
[02:33:50] <wesolows> not really
[02:34:02] <nrubsig> wesolows: why ?
[02:34:27] <wesolows> nrubsig: THe problem is that no one wants to be a sponsor, not that people are wishing they could become one but haven't found the right opportunity
[02:34:41] *** postwait has joined #opensolaris
[02:34:47] <nrubsig> wesolows: groan
[02:35:36] <nrubsig> erm
[02:35:38] <nrubsig> wait
[02:35:50] <nrubsig> wesolows: how was the sponsor list actually populated ?
[02:36:04] <nrubsig> wesolows: giant mantrap and everyone who fell inside got the "sponsor" stamp ?
[02:36:17] <wesolows> nrubsig: volunteers from among various senior engineers plus the advocates
[02:36:26] <Tempt> "sponsor or RIF, your choice"
[02:36:36] <stevel> a smattering few are junior engineers
[02:36:40] <nrubsig> Tempt: ouch
[02:36:44] <sommerfeld> according to microsoft, "unix" is inappropriate language: http://www.xbox-scene.com/xbox1data/sep/EEllllEZAAgEyoVOfY.php
[02:36:47] <richlowe> Tempt: good plan.
[02:36:50] <wesolows> true, there are actually a lot of sponsors
[02:36:52] <alanc> and even those who want to be sponsors don't have the time, since their management thinks it should be someone else's job to provide sponsors, not their teams
[02:37:24] <wesolows> alanc: well, to be fair, my management is somewhat justified there, given that it's not in any way a part of the Solaris organisation.
[02:37:44] <nrubsig> groan
[02:37:45] <wesolows> at this point I would sponsor only changes to code that I wrote
[02:38:02] <Tempt> sommerfeld: Ha!
[02:38:07] <wesolows> or someone removing wbem and agents
[02:38:10] <wesolows> I'd do that, too
[02:38:13] <tomww> alanc: +5
[02:38:15] <nrubsig> for chickengod's sake, I am sitting on a sh*tload of xx@@@!!! patches and can't get rid of them.
[02:38:25] <nrubsig> graaaaaaaahhh
[02:38:35] <nrubsig> (sorry, that needed to be said tonight)
[02:38:56] <richlowe> nrubsig: pretend they're all marketing related.
[02:39:07] <alanc> well, if you'ld stop trying to make stevel fix jive, maybe he'd have more time to work on getting the hg gate out in the open so you can putback directly
[02:39:11] <wesolows> alanc: What do you think about a fallback?  If no sponsor for N weeks, you get to login to some kind of restricted environment on the gate machine and putback directly?
[02:39:13] <nrubsig> wesolows: which code did you write ?
[02:39:19] <wesolows> nrubsig: I'll never tell!!!
[02:39:32] <alanc> (though then you'ld have to get webrti opened too)
[02:39:45] <richlowe> wesolows: it couldn't be restricted enough.
[02:39:48] <Tempt> Or maybe, just ban outside contributions altogether, pack up the bat and ball and go home.
[02:39:55] <richlowe> you'd have to be able to putback without being able to read the gate.
[02:40:02] <wesolows> richlowe: Then I guess whoever feels that way will have to provide a sponsor, eh? :-)
[02:40:06] <stevel> while the 'admitting defeat' option is always available, it's not my preferred choice of action
[02:40:15] <richlowe> I'm up for it.
[02:40:32] <richlowe> let the marketing folks urgently pretending this all works find a way out of the hole.
[02:40:35] <richlowe> they have the money.
[02:40:45] <alanc> wesolows: maybe once we get the open gate, we could try that...  not sure we could get a hole through the firewall for putbacks sooner than we could get the gate on the right side of it
[02:40:46] <sommerfeld> some combination of "rome wasn't thrown over the firewall in a day" and "changing engines in mid-flight"
[02:41:07] <richlowe> alanc: You could, if you just did it.
[02:41:14] <wesolows> the hg transition would have been a lot easier if it could have happened at the start of S12 as was originally intended
[02:41:25] <richlowe> wesolows: I agree entirely.
[02:41:36] <alanc> unfortunately, poking unapproved holes through the firewall is a firable offense, and I like my paycheck
[02:41:39] <richlowe> but Sun refuse to release anything.
[02:41:58] <wesolows> alanc: I wasn't suggesting an unapproved one, rather one which reflects a new official policy.
[02:42:05] <richlowe> this would all be easier if we said "fuck it", and move a gate outside, and let Sun decide whether they wanted to play along, or gtfo
[02:42:11] <nrubsig> alanc: like the holes in SWAN when everyone was running his/her own IPv6 tunnel for testing ?
[02:42:26] <wesolows> richlowe: I'm all for it, really.
[02:42:27] <alanc> nrubsig: I know nothing about those, and probably don't want to
[02:42:41] * wesolows doesn't know anything about those either
[02:43:05] *** myrkr has joined #opensolaris
[02:43:14] <alanc> so there's an hg gate outside right now, isn't there?    either pry it from stevel's clutches or make a clone and run with it...
[02:43:16] *** sommerfeld has quit IRC
[02:43:33] <e^ipi_> fork ON
[02:43:41] <e^ipi_> *shrug*
[02:43:44] <stevel> i don't have it in my clutches
[02:43:47] <nrubsig> e^ipi_: want to get votekicked ?
[02:43:58] <stevel> if someone wants to take it and fork, go for it
[02:44:03] <nrubsig> stevel: lets check: claws, teeths, two beagles...
[02:44:08] <stevel> that's the whole point of open source, right?
[02:44:12] <wesolows> someone could, if they wanted more legitimacy, file a grievance with the OGB
[02:44:15] <nrubsig> stevel: you quality as monster with two guard dogs!
[02:44:28] <stevel> they are useless guard dogs
[02:44:39] <pauliukas_> ExpertsExchange is addicting.
[02:44:40] <wesolows> not sure what the outcome would be - I know I'd want to see a clear plan with dates for everything needed for universal putback before I'd vote to dismiss the dispute, though.
[02:44:47] <alanc> it's not like there's a ON community that they can really appeal to, since there's no way of contacting them
[02:44:48] <nrubsig> stevel: didn't they scare some bugattis off ?
[02:44:57] <nrubsig> stevel: s/bugattis/burglars/
[02:45:21] <wesolows> alanc: Which is why they'd have to appeal to the OGB - the ON CG was unresponsive, they aren't making any apparent progress, my wads are stuck in limbo, etc.
[02:45:45] <stevel> nrubsig: yeah - but if that burglar came in with a steak they'd be happy and wagging their tails
[02:45:48] <nrubsig> what is a "CG" ?
[02:45:53] <wesolows> nrubsig: community group
[02:45:54] <alanc> should dust off the plan I had for poking communities with no contributors and add communities with no mailing lists to it...
[02:46:13] <richlowe> alanc: I doubt it would change anything.
[02:46:20] <richlowe> I fully believe ON would vanish before they'd fix anything.
[02:46:20] <wesolows> maybe we should just forcibly create a mailing list for ON with the Core Contributors on it
[02:46:32] <richlowe> wesolows: give ON to those that care about it.
[02:46:40] <richlowe> new C-Team, et al.
[02:46:47] <wesolows> richlowe: Let's give jbeck some time.
[02:47:00] <wesolows> I'm willing to wait a few weeks to see what he does.
[02:47:07] <nrubsig> wesolows: how long ? Until I have grey hair ?
[02:47:16] <richlowe> nrubsig: we've all seem jimg's photos.
[02:47:22] <richlowe> nrubsig: it looked grey to me already ;)
[02:47:23] <nrubsig> grumpf
[02:47:49] <richlowe> yet another indication that we shouldn't just sit around waiting :)
[02:47:52] <wesolows> nrubsig: I don't know, will your hair be gray in 3 weeks?
[02:48:01] *** ottom has joined #opensolaris
[02:48:12] <nrubsig> wesolows: I likely died before that date thanks to having no money for food.
[02:49:05] <wesolows> nrubsig: That's not something we can really help with, I'm afraid.
[02:49:12] <hile_> get a job?
[02:49:20] <nrubsig> hile_: thank you
[02:49:23] * nrubsig sulks
[02:49:49] <alanc> unfortunately, as 15 years of evidence shows, Sun doesn't want to pay anyone to work on ksh
[02:49:57] <wesolows> It's usually easier to rob orphans, grandmothers, and widows than to get a job, especially in western europe.
[02:50:36] <nrubsig> wesolows: or avoid study fees
[02:52:18] <wesolows> alanc: I'm not so sure Sun really wants to pay anyone to work on anything, if they can instead merely aggregate it from outside sources with a very small team of people from whatever country has the lowest wages at the moment.  But then, Sun's hardly unique in that regard.
[02:53:10] <wesolows> Though there are still a few people who understand what a giant steaming pile of fail results from that approach.
[02:53:55] <Tempt> Well, Wall St continually asks Sun to fire the entire R&D effort and become more like Dell
[02:54:08] <Auralis> wall st never was smart
[02:54:13] <Tempt> So trying to get everything done by people on the outside for nothing makes sense for them
[02:54:20] *** perlmonk has joined #opensolaris
[02:54:29] <Tempt> Perhaps they can get someone to design ROCK in their spare time, too.
[02:54:34] <nrubsig> Tempt: how would Sun be able to develop things like Niagara1/2 without R&D ?
[02:54:49] <Auralis> tooth fairy
[02:55:10] <alanc> Wall Street doesn't think anyone besides Intel needs to design CPUs
[02:55:14] <nrubsig> Auralis: well, technically Niagara was developed by Sun dissidents at Alfara
[02:55:15] <pauliukas_> woohoo. Solaris install almost done.
[02:55:21] <pauliukas_> What's the first thing that I should do?
[02:55:32] <nrubsig> pauliukas_: $ rm -Rf /*
[02:55:38] <nrubsig> as "root"
[02:55:39] <pauliukas_> hahahaha... not funny.
[02:55:39] <jmcp_> pauliukas_: turn off webconsole :-)
[02:55:44] * jmcp_ thwaps nrubsig
[02:55:46] <pauliukas_> jmcp_: But... Why?
[02:55:51] <alanc> useradd a non-root user
[02:55:56] <jmcp_> pauliukas_: it's a pile ... and you don't need it
[02:55:56] <pauliukas_> OP... kick that....
[02:56:00] <pauliukas_> oh wait...
[02:56:01] <Tempt> remove root from /etc/passwd
[02:56:02] <nrubsig> pauliukas_: webconsole sucks memory
[02:56:05] <jmcp_> and cycles
[02:56:08] <pauliukas_> Oh okay.
[02:56:08] <nrubsig> Tempt: /kick
[02:56:09] <pauliukas_> What is it?
[02:56:12] <Tempt> and rm -rf /bin/ksh*
[02:56:14] <pauliukas_> Is it only a Webmin clone or something?
[02:56:19] <richlowe> effectively.
[02:56:22] <nrubsig> Tempt: /kickbanflup
[02:56:29] <pauliukas_> So how do I uninstall or remove it
[02:56:36] <pauliukas_> Aswell, is 256MB of RAM okay/usable?
[02:56:37] <Tempt> followed by ln -s bash ksh ; ln -s bash ksh93
[02:56:38] <richlowe> with fire, preferably.
[02:56:38] * pauliukas_ ducks
[02:56:41] <nrubsig> pauliukas_: disable is sufficient for now
[02:56:48] <pauliukas_> But... How?
[02:56:50] <richlowe> svcadm disable webconsole
[02:56:51] <nrubsig> Tempt: /kickmanflup
[02:56:52] <e^ipi_> nrubsig: why?
[02:57:06] <nrubsig> e^ipi_: why what ?
[02:57:07] *** jcea has quit IRC
[02:57:19] <jamesd> nrubsig, will be sad to know that my copy of the sea turtle book (korn shell) has arrived.
[02:57:24] <e^ipi_> you don't need a root account... root is just a collection of privileges anyways
[02:57:29] <wesolows> webmin clone?  well, it does predate webmin by several years, but sure you can think of it however you like
[02:57:35] <wesolows> like webmin, it's a useless pile of garbage
[02:57:40] <e^ipi_> also, what's up with my nick
[02:57:43] *** e^ipi_ is now known as e^ipi
[02:57:50] <Tempt> remove the initial boot delay using the "init 0" tune command
[02:57:57] <pauliukas_> wesolows: ROFL!
[02:57:58] <nrubsig> jamesd: too bad that it covers almost no ksh93 features... ;-(
[02:58:23] <nrubsig> wesolows: wanna advertise SMC instead ?
[02:58:27] <wesolows> nrubsig: hell no
[02:58:34] <nrubsig> see ?
[02:58:48] <wesolows> smc, sunmc, webmin, webconsole all == complete waste of entropy
[02:58:58] <jamesd> nrubsig, but it only cost me about a quarter plus $3.99 shipping, and we do a lot of korn shell scripting at my work
[02:58:59] <nrubsig> you can't waste entropy
[02:59:09] <wesolows> waste of concentrated energy?
[02:59:12] <Auralis> you can only increase it
[02:59:14] <wesolows> they needlessly increase entropy
[02:59:17] <wesolows> is what I mean
[02:59:26] <wesolows> they waste order, I suppose
[02:59:35] * nrubsig throws a ZPM after wesolows
[02:59:45] <nrubsig> jamesd: URL ?
[02:59:48] * wesolows ducks behind a physicist
[02:59:54] <nrubsig> jamesd: where can I order it ?
[03:00:19] *** alanc is now known as alanc-away
[03:02:14] <jamesd> nrubsig,  amazon.com  be sure to check out the used book section, not sure its availible in other countries
[03:03:30] <nrubsig> Mhhh... "Steve Fosset abucted by aliens"
[03:04:01] <jamesd> nrubsig, vi book from $1  plus $3.99 shippng http://www.amazon.com/gp/offer-listing/0937175676/ref=lp_g_1/102-7280375-6712134?ie=UTF8&coliid=I2SCTVPKFP3O0H&colid=2XTJ8WD2PU3B5
[03:04:11] <pauliukas_> Xorg or Xsun?
[03:04:14] <pauliukas_> Which one is better?
[03:04:19] <pauliukas_> faster/whatever
[03:04:37] <nrubsig> pauliukas_: depends on what you want to do.
[03:04:41] <pauliukas_> ...
[03:04:46] <pauliukas_> Desktop computer and playing with the OS?
[03:04:55] <pauliukas_> Give me examples, if possible.
[03:05:01] <nrubsig> pauliukas_: depending on usage Xsun may be preferred or Xorg
[03:05:17] <pauliukas_> In what kind of conditions would it be preferential to use XSun?
[03:05:18] <jamesd> xorg has better support for nvidia and avi cards, xsun is better for sun framebuffers
[03:05:29] <pauliukas_> This has an Intel integrated crapo card.
[03:05:37] <hile_> if your framebuffer were not supported in Xorg
[03:05:51] <pauliukas_> I ran Ubuntu on a similar box.
[03:05:55] <pauliukas_> And that uses Xorg.
[03:06:16] <nrubsig> pauliukas_: Xsun has extensions fo shared memory transport, better support for Sun gfx cards, Display PostScript support and multiple support (e.g. 8+24bit) or special depths like 30bit (10bit per R,G,B channel)
[03:06:23] <nrubsig> s/fo/for/
[03:06:47] <nrubsig> pauliukas_: Xorg is better for "commodity" gfx cards
[03:06:51] <pauliukas_> gotcha
[03:06:56] <nrubsig> pauliukas_: if you use VMware use Xsun
[03:07:07] <pauliukas_> oh taht's just great... The whole screen turned orange.
[03:07:16] <pauliukas_> No, it's an IBM something.
[03:07:17] * steleman takes offense at his QuadroFX 4500 being labeled "commodity"
[03:07:26] <pauliukas_> lol
[03:07:40] * nrubsig snaps steleman neck and sneaks away with the Quadro
[03:08:00] <nrubsig> pauliukas_ killed steleman! I saw it! He took the quadro!
[03:08:05] <jamesd> steleman, you can buy a lot of sugar cane and  wheat for what that  4500  quatro cost.
[03:08:08] <pauliukas_> lol
[03:08:08] * nrubsig sneaks away
[03:08:15] * pauliukas_ shows his crappy Intel integrated
[03:08:20] <pauliukas_> Intel crapo =! quadro
[03:08:23] <nrubsig> 950 ?
[03:08:28] <steleman> jamesd: my people called nVidia's people :-)
[03:08:57] *** palowoda has quit IRC
[03:11:27] <wesolows> nrubsig: 5015982 may be more controversial than you think...I'm looking for the original docs when the feature was added.  I think the worry was about resource exhaustion triggered by unprivileged users.
[03:12:12] <nrubsig> wesolows: how fast can users generate core dumps ?
[03:12:19] *** jwk404 has quit IRC
[03:12:33] <richlowe> try it. :)
[03:12:46] <wesolows> really really fast
[03:13:23] <nrubsig> wesolows: and getting _one_ line per core dump in /var/adm/messages is really a problem ?
[03:13:46] <nrubsig> wesolows: note that I can use /usr/bin/logger to spam that log, too.
[03:13:59] <e^ipi> steleman: by the way, after me & martin fought with it for a week or two, we've finally decided that this ffb2+ will not be able to drive my monitor properly ever
[03:14:02] <wesolows> maybe not; I might be misremembering
[03:14:13] <wesolows> 1999/243 mentions it but seemed to converge
[03:14:19] <steleman> e^ipi: it just cannot do 1600X1200 ?
[03:14:23] <e^ipi> so when i've got some cash i'll try'n pick up an xvr-600 or xvr-100 or something
[03:14:28] <wesolows> you will never see that case, sadly, because it names a partner many times
[03:14:31] <e^ipi> steleman: oh, it can do 1600x1200 just fine
[03:14:43] <e^ipi> but not 1600x1000, which is the native res. of this monitor
[03:14:50] <nrubsig> wesolows: if I want to take a system down my last choice would be via syslog.
[03:14:52] <steleman> e^ipi: your monitor is weird :-)
[03:15:05] <e^ipi> it's an off the shelf widescreen
[03:15:08] <nrubsig> wesolows: IMO a thread-based fork-bomb would be more effective.
[03:15:10] <wesolows> as I said, I am probably misremembering
[03:15:15] <wesolows> never mind
[03:15:17] <steleman> e^ipi: my humble advice: i prefer the xvr-600 over the xvr-100
[03:15:56] <e^ipi> yeah, it seems a better card, but ultimately my choice is dictated by finance
[03:16:04] <Auralis> off the shelf widescreens do not have 1600x1000, the 22" onces have 1680x1050
[03:17:01] <e^ipi> which the ffb2+ still can't do, so it's irrelevant
[03:17:16] *** simford has joined #opensolaris
[03:18:00] * nrubsig checks whether simford can sponsor putbacks
[03:18:03] <Auralis> xv-100 will shititself at that res, it will be painfull slow and unstable display, ghosting and shit
[03:18:12] <brendang> system-log looks like it is still remote spammable,
[03:18:13] <e^ipi> i'm cool with that
[03:18:14] <Tempt> xvr600 will be nice
[03:18:17] <brendang> $ svcprop -p config/log_from_remote system-log
[03:18:17] <brendang> true
[03:18:30] <Tempt> good
[03:18:39] <e^ipi> if the price difference is only a few $ either way, i'll go with the 600
[03:18:39] <Tempt> all the better to annoy people with
[03:18:48] <Tempt> 600s are all over the place
[03:18:52] <e^ipi> but if it's much more, i'm buying the 100
[03:18:54] <Tempt> shit, i've got a couple on the shelf for spares
[03:18:55] <brendang> so why bother logging in and creating coredumps? don't need a username/password...
[03:19:25] <simford> nrubsig: what kind of putbacks?
[03:19:45] <Tempt> brendang: I've seen a syslog flood take a host down. A 6 CPU host with 6Gb of RAM
[03:19:58] <nrubsig> simford: OS/Net
[03:20:03] <e^ipi> ebay north america rarely has them, and when they do they go for a couple hundred $
[03:20:05] <pauliukas_> Hey guys...
[03:20:06] <brendang> Tempt: impressive. it's a pretty slow flood.
[03:20:08] <simford> nrubsig: nope, sorry
[03:20:14] <pauliukas_> None of the arrows nor backspace on my keyboard work
[03:20:15] <pauliukas_> What gives?
[03:20:28] <nrubsig> simford: Sorry, I am a bit desperate and now start to stalk potential sponsors at night... ;-(
[03:20:37] <Tempt> brendang: Ha, inn configured to extreme verbose logging and processing a 500k+ posts
[03:20:39] <simford> nrubsig: I can sponsor g11n related putbacks: http://www.opensolaris.org/os/project/nv-g11n/
[03:20:51] <Tempt> brendang: Remember, syslog calls sync() after every single line
[03:20:54] <simford> nrubsig: that's fine :)
[03:21:14] <brendang> Tempt: oh. 500k+ posts - silly syslogd.
[03:21:41] <Tempt> It screwed the I/O so much the kernel had trouble writing to swap
[03:21:53] <Tempt> b00m panic panic panic
[03:22:14] *** sahafeez has quit IRC
[03:22:23] <Tempt> I think it was SunCluster that actually panic'd the box in the end because it was so muddied it didn't get a packet on the wire for over 10 seconds
[03:22:57] <e^ipi> anyways, gtg
[03:23:04] *** e^ipi has quit IRC
[03:23:38] <nrubsig> Tempt: this access created a PANIC which created a PANIC which created a PANIC which created a PANIC which created a PANIC which PANIC which created a PANIC which creC which created a PANIC which created a PANIC which created a which created a PANIC which created a PICNICK which created a PANIC which created a PANIC which created a PANIC which created a P...
[03:23:45] <pauliukas_> anyone?
[03:23:48] *** Megaf has quit IRC
[03:25:07] <jmcp_> pauliukas_: you're not using bash, you're using /sbin/sh
[03:25:19] <jmcp_> which is traditional Bourne Shell, not GNU's Bourne Again SHell
[03:25:25] <pauliukas_> right... I now figured that out.
[03:25:28] <jmcp_> pauliukas_: exec /bin/bash
[03:25:35] <pauliukas_> I just wrote bash and it worked.
[03:25:40] <pauliukas_> How do I start the X server?
[03:25:52] <pauliukas_> And is there any tutorial for noobies or something?
[03:26:15] <jmcp_> svcadm enable cde-login
[03:26:16] *** stevel has quit IRC
[03:26:21] *** yongsun has quit IRC
[03:26:39] *** palowoda has joined #opensolaris
[03:26:50] <pauliukas_> How do I find out which IP it has been assigned?
[03:27:54] *** yongsun has joined #opensolaris
[03:28:13] *** unbewont is now known as g4lt-mordant
[03:31:13] <jmcp_> ifconfig -a
[03:32:17] <pauliukas_> okay. that found.
[03:32:26] <pauliukas_> Needed to edit sshd.conf to allow root login.
[03:32:30] <pauliukas_> Now, how do I restart sshd?
[03:32:48] <hile_> svcadm refresh ssh
[03:32:51] <hile_> or restart ssh?
[03:33:10] <hile_> or HUP the master sshd process :)
[03:33:32] <pauliukas_> Password:
[03:33:32] <pauliukas_> Last login: Tue Sep  4 21:18:56 2007
[03:33:33] <pauliukas_> Sun Microsystems Inc.   SunOS 5.10      Generic January 2005
[03:33:33] <pauliukas_> #
[03:33:34] <pauliukas_> I'm in!
[03:33:48] <pauliukas_> Alright. So what kind of neat stuff should I do?
[03:33:49] <postwait> Anyone here use Zimbra?
[03:35:08] <flyingparchment> hmm, this LSI card decided to rebuild the mirror for no reason
[03:35:25] <jmcp_> flyingparchment: it has a reason, it just isn't letting you in on the secret
[03:35:47] <flyingparchment> i was fiddling with scsi stuff earlier, i think i might've confused it
[03:38:32] <flyingparchment> i think i'll log an rfe that format's 'analyze' should have a progress bar ;-) (or at least show % complete, ETA)
[03:39:02] <jmcp_> flyingparchment: closed, will not fix
[03:39:07] <flyingparchment> :(
[03:39:09] <flyingparchment> why?
[03:39:28] <jmcp_> flyingparchment: /usr/sbin/format is a command line util - if you want a progress meter you'll need to implement a curses- or gui-based wrapper
[03:39:33] <flyingparchment> it doesn't even tell what units it reports progress in.. i'm assuming it's cylinders
[03:39:33] <jmcp_> don't futz with format
[03:39:43] <flyingparchment> jmcp: well, it could still do the latter
[03:39:45] <flyingparchment> (%/ETA)
[03:40:32] <jmcp_> that would be a better rfe
[03:41:03] <jmcp_> flyingparchment: RFE "how big is format's analyze?"
[03:41:43] <flyingparchment> it shouldn't be too hard, maybe i'll write the fix too
[03:47:37] *** Carl-H has joined #opensolaris
[03:49:44] *** Carl-H has left #opensolaris
[03:51:26] * jmcp_ lunches
[03:51:27] *** stukreit has joined #opensolaris
[03:51:37] *** Tekni has joined #opensolaris
[03:52:42] <Tempt> Yep, lunchtime it is.
[03:53:44] *** nrubsig has quit IRC
[03:55:13] *** coraline has left #OpenSolaris
[03:57:37] *** nachox has joined #opensolaris
[03:57:56] <nachox> hey people
[04:00:15] <CIA-26> th199096: 6563073 Fix cstyle in the sharefs code, 6575901 libc`sharefs() and ld.so have conspired to kill smdiskless
[04:10:01] *** migi has quit IRC
[04:10:23] *** migi has joined #opensolaris
[04:10:47] *** karrotx has quit IRC
[04:10:52] *** halton has joined #opensolaris
[04:14:01] <flyingparchment>  is there any reason not to build onnv-gate (nightly) on an NFS filesystem?
[04:14:45] <wesolows> yes
[04:14:48] <wesolows> it's fucking slow
[04:14:54] <flyingparchment> heh.  that's okay
[04:14:56] <wesolows> other than that, go ahead
[04:15:07] <flyingparchment> it's running in vmware, a bit mbore slowness won't amke much difference
[04:21:17] <flyingparchment> what about bfu, is it okay to do that from nfs, or should i copy the archives?
[04:22:04] <jbk> hmm.. do i fly into sjc or save $100 and fly into oakland and chance the traffic...
[04:24:09] *** ichigo has quit IRC
[04:26:50] *** Gman has quit IRC
[04:31:48] *** BatonT has joined #opensolaris
[04:32:02] *** Gman has joined #opensolaris
[04:32:44] <jmcp_> flyingparchment: I usedto bfu via nfs all the time, it didn't make much difference to me
[04:34:56] <jamesd> firing off make and going to bed has solved many slowness issues.
[04:37:16] <jmcp_> that's for sure :)
[04:37:40] <jbk> :)
[04:39:12] *** sfire||mouse has joined #opensolaris
[04:40:30] <pauliukas_> Oh man...
[04:40:35] <pauliukas_> Solaris zones seem so sexy.
[04:40:45] <pauliukas_> Are they useful? Or do they actually have a lot of disadvantages?
[04:41:09] <jmcp_> they are very very useful
[04:41:20] <pauliukas_> Because... This whole idea is so neat.
[04:41:23] <jmcp_> as to disadvantages - depends on what you class as good and/or bad
[04:41:28] <Triskelios> pauliukas_: they're used by many many people for serious things
[04:41:36] <pauliukas_> How about performance-wise?
[04:41:42] <jmcp_> at home I run global zone + two non-global zones: one for webserving, one for my vpn
[04:41:43] <jmcp_> they're fine
[04:41:56] <jmcp_> I don't see any noticeable difference in performance
[04:41:57] <flyingparchment> zone performance is identical to normal performance
[04:41:57] <pauliukas_> In a production environment, that is...
[04:42:00] <flyingparchment> there's no virtualisation
[04:42:01] <pauliukas_> sweet.
[04:42:03] <pauliukas_> Oh wow.
[04:42:04] <Triskelios> pauliukas_: the overhead is negligible
[04:42:15] <pauliukas_> Guys... I'm ashamed to say this...
[04:42:20] <pauliukas_> But I think that I'm in love with Solaris
[04:42:21] * pauliukas_ blushes
[04:42:22] <jmcp_> :-D
[04:42:29] <jmcp_> excellent!
[04:42:37] <Auralis> wait till he discovers dtrace
[04:42:38] * jmcp_ cracks open the Champagne
[04:42:39] <jmcp_> hehh
[04:42:54] <pauliukas_> Auralis: I already went a bit through dtrace training.
[04:43:01] <Auralis> ah
[04:43:07] *** ChanServ sets mode: +o Tpenta
[04:43:12] <jmcp_> pauliukas_: dtrace, zones, zfs and some very classy schedulers ... all killer features for Solaris imnsho
[04:43:23] <pauliukas_> Oh yeah, ZFS is my favorite of all.
[04:43:39] <pauliukas_> But some people are saying that it's not good.
[04:44:26] <jmcp_> pauliukas_: I use it, I love it but I'm clearly biased. My observation is that the people who say ZFS ain't so good are either Sun's competitors, or people who haven't been through the pain of having to manage *large* storage installations
[04:44:38] <jmcp_> and who haven't suffered enough data corruption with Reiserfs or XFS
[04:44:39] <pauliukas_> So... Tell me.
[04:44:51] <jmcp_> or indeed any other filesystem
[04:45:08] <pauliukas_> I use Linux with ext3 and hardware raid1 in a production environment for webservers.
[04:45:19] <pauliukas_> I heard about RAID-Z. Would that replace the need for hardware raid?
[04:45:30] <flyingparchment> raidz is raid5
[04:45:36] <pauliukas_> ah okay.
[04:45:36] <flyingparchment> if you use raid1 now, you want a zfs mirror
[04:45:41] <pauliukas_> Gotcha...
[04:45:44] <jmcp_> pauliukas_: hw raid does not have the data protection that ZFS provides
[04:45:54] <pauliukas_> jmcp_: I know... That's why I'm so excited about this,
[04:46:07] <pauliukas_> But ZFS doesn't fail... Does it? Or have there been nightmares?
[04:46:18] <jmcp_> I don't know of any nightmares with it
[04:46:28] <jmcp_> what do you mean by "fail", though>
[04:46:29] <jmcp_> ?
[04:46:30] <pauliukas_> I know stories of when hardware RAID just collapse and loses all your data.
[04:46:35] <movement> is there a way to stop pkgadd dumping out the 'copyright' file?
[04:46:41] <pauliukas_> Or when RAID5 can't recover a lost drive.
[04:46:46] <jmcp_> movement: not that I know of
[04:46:49] <nachox> zfs doesnt, the boot archive does :P
[04:47:05] <pauliukas_> boot archive?
[04:47:06] <jmcp_> pauliukas_: I've helped get customers back from corrupted hw raid
[04:47:13] <jmcp_> pauliukas_: grub's boot archive
[04:47:16] <pauliukas_> ah...
[04:47:17] *** cypromis_ has joined #opensolaris
[04:47:23] <flyingparchment> solaris was getting too reliable, so sun added the boot archive
[04:47:30] <pauliukas_> ROFL
[04:47:49] <movement> jmcp_: yet pkg-get etc. don't seem to?
[04:47:54] <pauliukas_> Oh man... Solaris sounds so exciting.
[04:47:59] <pauliukas_> If it works like advertised...
[04:48:20] <edp> pauliukas_, ZFS is nicely integrated with Zones as well
[04:48:26] <pauliukas_> good to know.
[04:48:32] <pauliukas_> My next training topic is Linux zones.
[04:48:34] <Triskelios> pauliukas_: memory requirements can be steep, and ZFS likes 64-bit. but that's most of the drawbacks...
[04:48:47] <pauliukas_> All my current hardware is 64-bit...
[04:48:53] <pauliukas_> And all my future will be 64-bit too.
[04:48:57] <jmcp_> movement: perhaps they don't include a copyright file in the pkg boilerplate?
[04:49:20] <pauliukas_> Triskelios: Is ZFS a performance problem on modern servers?
[04:49:28] <pauliukas_> As in... Does it use a lot of CPU resources?
[04:49:31] <movement> how does solaris install do it?
[04:49:49] <jmcp_> movement: no idea
[04:50:03] <Triskelios> pauliukas_: no, it's fairly lightweight CPU-wise
[04:50:09] <pauliukas_> wow.
[04:50:12] <jmcp_> pauliukas_: no, but it does chew address space like it's a ravenous hungry beast
[04:50:17] <jmcp_> hence the need for 64bit hw
[04:50:18] <jmcp_> :-)
[04:50:21] <flyingparchment> you might want to avoid zfs for OLTP, otherwise it's fine
[04:50:22] <pauliukas_> Does that mean I don't have to buy $700 RAID controllers?
[04:50:32] <pauliukas_> What's OLTP, if I might ask?
[04:50:45] <jmcp_> OnLine Transaction Processing
[04:50:50] <Auralis> yes it means you don't need raid controlers
[04:50:54] <pauliukas_> Never heard of that one before.
[04:51:04] <flyingparchment> OLTP is transaction-heavy database work
[04:51:12] <pauliukas_> So what are the disadvantages of chewing memory space.
[04:51:14] <flyingparchment> (like you might have with a web application)
[04:51:23] <flyingparchment> pauliukas_: it doesn't run so well on 32-bit systems.
[04:51:28] <pauliukas_> Oh, that makes sense... But could it be used just for casual webserving and stuff?
[04:51:29] <flyingparchment> 64-bit has effectively unlimited address space
[04:51:36] *** karrotx has joined #opensolaris
[04:51:38] <pauliukas_> Oh. So it works just fine on that?
[04:51:40] <jmcp_> yeah
[04:51:41] <flyingparchment> yes
[04:51:42] <pauliukas_> Neat.
[04:51:59] <Tempt> zfs is full of win. I've even got some small Oracle instances on zfs and the performance isn't a problem.
[04:52:08] <pauliukas_> heh.
[04:52:10] <flyingparchment> oltp work is being worked on too, i expect it'll be much better before long
[04:52:19] <pauliukas_> oh wow. this is amazing.
[04:52:55] <jmcp_> time for more coffee
[04:53:32] <flyingparchment> i think the only feature zfs doesn't have that people keep asking for is encryption
[04:53:34] *** slowhog has left #opensolaris
[04:53:36] <flyingparchment> and that's also being worked on
[04:53:51] <pauliukas_> Yeah. I heard about that some time ago. Encryption is not an issue for me.
[04:54:11] <pauliukas_> Would it seem far-fetched to run a Linux container with Apache and MySQL and everything for a shared hosting environment?
[04:54:24] <flyingparchment> why not run apache and mysql on solaris?
[04:54:37] <Jiko> my main issue with zfs isn't anything to do with Sun at all
[04:54:40] <flyingparchment> for mysql in particular, remember linux zones are 32-bit only
[04:54:42] <pauliukas_> I do web hosting. And the software to control all of that is only available for Linux.
[04:54:47] <Jiko> our backup product doesn't have support for it :-/
[04:55:11] <flyingparchment> what could be linux-specific about web hosting software?
[04:55:40] <pauliukas_> The panel which controls the resources used by the users and also provides a front-end to users.
[04:55:58] <pauliukas_> This one, in my case: www.directadmin.com
[04:56:22] <pauliukas_> I've seen some people asking them to have a Solaris version, but I doubt that this will ever happen.
[04:57:54] <flyingparchment> heh, having some software vendor install mysql for you sounds like a fairly bad idea :)
[04:58:04] *** sioraiocht has quit IRC
[04:58:16] <pauliukas_> Well, it does... Until you see that everything works and that everyone around you does the same.
[04:59:00] <Tempt> Are you running a hosting business?
[04:59:03] <pauliukas_> yes
[05:00:02] <Tempt> If you're running a hosting business and you have a working platform, don't mess with it
[05:00:09] <Tempt> Just milk the cash cow and leave it well alone.
[05:00:46] <pauliukas_> Thing is, my platform is really small. And I'm looking for the best way to develop the platform and make it ready for growth.
[05:01:08] <flyingparchment> i don't think i'd want to plan a long-term solution using linux zones.  seems like more of a transitional thing.
[05:01:15] <flyingparchment> i mean, if everything is linux, why not just use linux?
[05:01:28] <pauliukas_> To have features such as ZFS and dtrace.
[05:01:41] <pauliukas_> hmmm
[05:01:44] <delewis> flyingparchment: you can DTrace processes from within a branded zone.
[05:01:45] <pauliukas_> http://www.virtualmin.com/os-support.html
[05:01:57] <Tempt> Any hosting related software that is OS-specific is obviously shabby by design
[05:02:11] <delewis> referring to Zimbra?
[05:02:24] <Triskelios> I think the whole deal with directadmin having the "vendor" do the setup work is highly questionable...
[05:02:37] <Tempt> *anything* like that
[05:02:47] <delewis> Zimbra is a real POS.
[05:02:53] <Tempt> I mean, I'm looking at this directadmin shit, and it's 32bit x86 only
[05:02:54] <pauliukas_> Triskelios: It's like that in the hosting industry.
[05:03:01] <pauliukas_> Tempt: Nope. I run it on 64-bit.
[05:03:06] <Tempt> Why on earth would a hosting panel by contrained in that way?
[05:03:09] <pauliukas_> And, in fact... DirectAdmin is one of the best panels...
[05:03:16] <pauliukas_> Because it doesn't bloat up your system.
[05:03:22] <pauliukas_> http://www.cpanel.net/
[05:03:28] <Tempt> and I quote from http://www.directadmin.com/install.html - "Also, 64-bit OS's are not supported at this time."
[05:03:28] *** cypromis has quit IRC
[05:03:32] <pauliukas_> cPanel is the most popular panel on the planet. I hate it.
[05:03:47] <pauliukas_> Tempt: Their website is really out of date. Go on the forums if you want to see the latest.
[05:03:50] <Triskelios> pauliukas_: there is not enough technical knowledge for hosting people to do the setup themselves?
[05:03:53] <delewis> Virtuoso is gaining some popularity.
[05:03:58] <delewis> but then again, that's a bit more than just a panel.
[05:04:23] <Tempt> geez
[05:04:25] <pauliukas_> Triskelios: Sadly, most people who run hosting businesses don't even know what Linux is.
[05:04:37] <Tempt> I'd hate to be in the modern hosting business
[05:04:38] <pauliukas_> Most of them order a server with cPanel and never ever login into the shell.
[05:04:53] <Tempt> I'm glad the only hosting I do is free, and people don't mind emailing a request and waiting for a response.
[05:04:54] <pauliukas_> Well, it really is sad.
[05:04:59] <pauliukas_> Yeah...
[05:05:26] <pauliukas_> But, to be competitive, you have to offer a panel which allows the user to instantly see how much he's using. Or to create new domains/databases or perform other tasks.
[05:05:32] <Tempt> The people that host with me get a place to upload their files and a mysql database and a suggestion to help themselves.
[05:05:37] <flyingparchment> i think the sort of hosting where your customers ask for cpanel probably has too sliw margins to be fun :)
[05:05:53] <Tempt> I'd rather host zones for people or something like that.
[05:05:59] <jbk> well most webhosting is slim margins anyway
[05:06:02] <pauliukas_> flyingparchment: That's why I don't do cPanel. People who do cPanel mostly oversell their servers like crazy.
[05:06:09] <jbk> i've not seen any that aren't incredibly cheap
[05:06:10] <pauliukas_> I know some people who make crazy money.
[05:06:19] <flyingparchment> pauliukas_: that applies to any similar software
[05:06:33] <delewis> I think a lot of hosting providers nowadays are looking at Virtuozzo (or OpenVZ) over cPanel, as they're going the virtualization route, anyway.
[05:06:40] <delewis> GoDaddy uses Virtuozzo.
[05:06:43] *** Fullmoon has joined #opensolaris
[05:06:46] <pauliukas_> delewis: These are two totally different beasts.
[05:06:48] <delewis> and from what I've seen it's *much* nicer than cPanel.
[05:06:49] <pauliukas_> No. Stop. Let me explain.
[05:07:11] <pauliukas_> Virtuozzo is like VMWare. It's virtualization. It's used for VPS hosting. It's simply a root jail.
[05:07:23] <flyingparchment> a root jail is nothing like vmware...
[05:07:27] <pauliukas_> Virtuozzo only provides you with a panel to control your virtual machine.
[05:07:29] <delewis> pauliukas_: you missed the other half of my state 'as they're doing the virtualization route, anyway'
[05:07:34] <delewis> Virtuozzo does include a nice panel.
[05:07:37] <Tempt> It stops people uploading php that just rapes through /etc/system looking for 'sploits?
[05:07:39] <pauliukas_> flyingparchment: Technically, no. Because vmware is a hypervisor.
[05:08:13] <pauliukas_> Virtuozzo is backed by open source technology called OpenVZ.
[05:08:23] <pauliukas_> You can freely download and experiment with it.
[05:08:39] <Tempt> Does anyone here actually pay for their hosting?
[05:08:41] <pauliukas_> I think GoDaddy, as their panel, use their own custom-built one.
[05:08:49] <pauliukas_> Well, I buy servers and bandwidth. So I guess that I do :-P
[05:09:04] * flyingparchment doesn't have any hosting :)
[05:09:06] <boyd> Oh, FFS: checking size of char... configure: error: cannot compute sizeof (char), 77
[05:09:07] <Tempt> (apart from those who have their own infrastructure, haa)
[05:09:08] <flyingparchment> (except email)
[05:09:12] *** sioraiocht has joined #opensolaris
[05:09:28] <Tekni> Tempt: I don't pay for my colo, no
[05:09:31] <pauliukas_> Tempt: But there's also a very sad truth that most free hosting out there is total crap. Oversold servers. A lot of downtime.
[05:09:34] <boyd> Now I'm  no expert but I thought sizeof(char) was *defined* to be 1
[05:09:41] <pauliukas_> With the exception of getting hosting off a friend.
[05:09:51] <flyingparchment> i think tempt is talking about getting hosting from someone you know
[05:09:54] <flyingparchment> boyd: it is
[05:09:58] <pauliukas_> Or "colocating" and hiding your own server inside a huge corporate datacenter.
[05:10:18] <boyd> flyingparchment: So configure checking that and the failing is even more braindededness
[05:10:26] <boyd> s/the/then
[05:10:27] <pauliukas_> Alright. I should go back to viewing those Sun videos about Solaris.
[05:10:28] <flyingparchment> boyd: s/checking that and the failing//
[05:10:36] <boyd> flyingparchment: :)
[05:10:47] <Tempt> pauliukas_: I like the colocation option.
[05:11:06] <pauliukas_> The free/hiding one? :-P
[05:11:19] <Tempt> The three servers happily sitting in the DC option.
[05:11:27] <pauliukas_> heh.
[05:11:48] <pauliukas_> I have some friends who are hiding their servers from the corporate eyes in a DC and run a business off of them!
[05:11:53] <Tempt> Heh
[05:11:56] <pauliukas_> Personally, I would never do that for something in production.
[05:11:59] <Tempt> I'm sure that's pretty common these days
[05:12:02] <Tempt> Anything on the dodgy.
[05:12:04] <pauliukas_> Yeah...
[05:12:35] <boyd> You could hide a hellava lot in a corporate DC
[05:12:49] <pauliukas_> I don't have the connections nor balls to do that. I pay my stuff fair and square... And it's quite expensive too!
[05:12:53] <Tempt> getting and keeping IP ranges is the trickier part
[05:13:08] <pauliukas_> For shared hosting, you only need one IP per box.
[05:13:19] <pauliukas_> So that's easier to get.
[05:13:26] <flyingparchment> Tempt: wireless!
[05:13:30] <pauliukas_> LOL
[05:13:31] <Tempt> ha
[05:13:56] *** postwait has quit IRC
[05:13:59] <Tempt> I wouldn't want to be paying for three RU of colo in Australia
[05:14:24] <pauliukas_> It's not that expensive these days.
[05:14:41] <pauliukas_> If you rent your own cabinet, it comes down to about $30 per month per U without electricity nor bandwidth.
[05:14:44] <Tempt> I think I'll stick to my cost-free plan
[05:15:11] * delewis wonders if his E4500 and Photon would go un-noticed in a DC...
[05:15:20] *** tassiework has joined #opensolaris
[05:15:23] <delewis> I suspect not, but I'm an optimist.
[05:15:44] * jamesd saw something incredible at my datacenter today,  an ss10 with a pair of 711's attached.
[05:15:57] <delewis> jamesd: at Wells Fargo?
[05:16:03] <pauliukas_> A what with what what?
[05:16:04] <jamesd> yes
[05:16:07] <delewis> scary.
[05:16:07] *** tassiework has left #opensolaris
[05:16:09] * pauliukas_ is no Sun know-it-all
[05:16:30] <jamesd> delewis, its not attached to the network, just sitting there  waiting to be destroyed.
[05:16:32] <Auralis> 15 years old hardware
[05:16:39] <delewis> jamesd: know what it's being used for?
[05:16:42] <delewis> oh.
[05:16:50] <flyingparchment> destroyed?  save it
[05:16:55] <pauliukas_> hah
[05:17:04] <delewis> flyingparchment: might be worth it if it were at least an SS20.
[05:17:20] <jamesd> flyingparchment, corprate policy says everything has to go through the chipper unless its a leased box.
[05:17:27] <Tekni> heh, i was using an s20 in production up until 2004
[05:17:37] <delewis> jamesd: shame. Wells Fargo buys some nice gear.
[05:17:44] <Tempt> jamesd: Man, I know of a few SS2s with DLT4000s still in production and running
[05:17:48] <Tempt> 'backup server'
[05:18:02] <jamesd> we have a u2, that is being replaced soon...
[05:18:07] * Auralis still runs a ss10
[05:18:31] * jamesd has a ss5  that he turns on occausionally...
[05:19:04] <jmcp_> jamesd: utym "monitor stand"
[05:19:07] <Tekni> i should have kept an ipx
[05:19:08] <jamesd> and my tv stand is a fully functional  ipx
[05:19:36] <pauliukas_> lol
[05:21:15] *** rachel has quit IRC
[05:21:27] <Tempt> WickedWicky has an SS20 running his blog online
[05:21:59] <Tempt> I'm using a Classic for a monitor stand. They're just perfect.
[05:22:36] <jamesd> i am not a hosting company... i dont host any of my own web pages.. i have   contextshift.co.uk for that...   and blogger.com
[05:22:43] <pauliukas_> you guys are crazy
[05:23:06] <Tekni> nope.. just nostalgic
[05:23:07] <boyd> Uh oh, somebody noticed.
[05:23:11] <pauliukas_> LOL
[05:23:17] <pauliukas_> good sense of humor too
[05:24:10] <Tempt> I'm looking forward to putting my OpenVMS machine online
[05:24:19] <pauliukas_> lol
[05:24:34] <Tekni> i miss the pizza box.. they stacked well.  I wish Sun would cram something into that form factor again
[05:24:39] <Tempt> I'm just trying to weasel somewhere to put it
[05:25:08] <jamesd> i think the u2 is a nice form factor...   not quite  tiny but one hell of a box...
[05:25:50] <delewis> I never cared for the two rows of SBus cards in the Ultra 2.
[05:26:14] <delewis> it's quite difficult to use the left-most, upper slot (looking on from the front)
[05:26:48] <jamesd> i never used more than 2 cards in mine... though i do have to try the qfe that sitting waiting to be used... and if i put in a cg6 it would then have 3 cards.
[05:27:11] <delewis> I've got a QFE, 2 socal's, and 1 differential SCSI card in mine.
[05:27:31] <jmcp_> Tekni: we do, but generally they go into racks
[05:27:48] <flyingparchment> aren't rack machines deeper than the old pizza boxes?
[05:27:59] <delewis> flyingparchment: by far.
[05:28:04] <delewis> except the Netra X1s.
[05:28:06] <Tekni> jmcp_: i have plenty of the rackmount gear
[05:28:09] <jamesd> i have a sunswift, cg6 and soon a  qfe.  and a pair of 711's  (one 12 disk and one 6 disk) attached.
[05:28:10] <delewis> those are pretty short.
[05:28:34] <delewis> Netra X1s only go about 1/3 of the way back that traditional rack server does.
[05:29:03] <Tekni> jmcp_: i was thinking something more like a t2000 lite in the pizza box for home use with WAF
[05:29:14] <Tempt> Nah, you just need to be patience wedging the SBus cards into the U2. I really should pull the cards out of my Ultra-2 (monitor stand edition), it's got a pair of sunswifts, a qfe and a ge in there
[05:29:15] <jmcp_> WAF?
[05:29:21] <Tekni> wife acceptance factor
[05:29:45] <Tempt> Get an 880
[05:29:48] <Tempt> use it as a side table
[05:29:53] <Tempt> put a doily and a vase of flowers on it
[05:29:54] <Tempt> :)
[05:30:02] <pauliukas_> LOL
[05:30:23] <jmcp_> ah
[05:30:28] <delewis> why stop with an 880. Get a 6800.
[05:30:35] <delewis> and just cover it in flowers.
[05:30:36] <jamesd> Tekni, forget the t1000 or t2000, unless you have a basement... the t2000 is loud enough... and the t1000 is worse and whiney.
[05:30:38] <delewis> she'll never know.
[05:30:42] <Tempt> Delewis, you fool, you wouldn't be able to see the vase and doily!
[05:30:54] <delewis> Tempt: open up the closet and put those inside. :-)
[05:30:55] *** sahafeez has joined #opensolaris
[05:30:58] <delewis> somewhere...
[05:31:12] <Tekni> jamesd: hence the lite version.. it is supposed to be a "coolthreads" system, eh?
[05:31:16] <Tempt> p575
[05:31:20] <Tekni> why all the jet fans
[05:31:20] <Tempt> That's a good home server
[05:31:32] <delewis> p575's size varies.
[05:31:49] <delewis> it can be as small as 4U.
[05:31:50] <Tempt> Because people love noisy hardware I figure. I mean, everything is so loud now, I figure there must be a customer perception that "Teh louder teh fasta"
[05:32:09] <delewis> it depends on how many CPU/memory units you get with it (each are 4U)
[05:32:10] <sfire||mouse> after booting the thumper today I'd agree
[05:32:14] <delewis> at least if it's anything like the p570.
[05:32:27] <Tempt> Just give me more hardware
[05:32:49] * Tempt burgles delewis' house and steals all his hardware
[05:33:13] <Tekni> sfire||mouse: oh come on.. just buy some good ear plugs
[05:33:14] * delewis fumbles to get the machine gun out of the closet
[05:33:30] <Tekni> the only thing louder than a thumper is a room full of them :)
[05:33:43] <sfire||mouse> Tekni: I'd rather try and figure out how to configure the damn system
[05:33:49] <Tempt> You've never heard an ES47 with fans running a full speed
[05:33:58] <Tempt> or an IBM BladeCentre
[05:34:13] <boyd> A rack of t1000s is pretty loud.. all those 1U fans
[05:34:18] <pauliukas_> Indeed.
[05:34:23] <Tekni> sfire||mouse: what's the problem?
[05:34:25] <sfire||mouse> I finally got the net mgm port working, and I dont know if I messed up the preconfigured OS on it, or if it even had it
[05:34:26] <pauliukas_> I've seen a few t1000s at the datacenter...
[05:34:31] <pauliukas_> I'm like "The heck... Is this word war 3?"
[05:34:58] <Tempt> If I was in the US, I'd be able to not only have the mountain of hardware but an arsenal of weapons with which to protect it
[05:35:04] <Tempt> Like Delewis and his machinegun ;)
[05:35:12] <pauliukas_> indeed.
[05:35:14] <delewis> there's nothing worse than standing next to a rack loaded full of DS4800 disk units (14 disks each.. so 140 disks in a full rack)
[05:35:15] <sfire||mouse> Tekni: problem is, I've never installed Solaris really before, and I don't know if it came ready to be setup
[05:35:20] <delewis> of course, maybe a rack full of Thumpers..
[05:35:21] * delewis shudders
[05:35:22] <Tekni> sfire||mouse: it should have
[05:35:27] <boyd> Tempt: Or you could live in europe and rely on poisonous reptiles
[05:35:30] <pauliukas_> Alright. I should stop reading jmcp_'s blog and actually learn more solaris.
[05:35:35] *** DoYouKnow has joined #opensolaris
[05:35:36] <jamesd> they powered on a t2000 in our lunch room, big enough for 30-40 people.. and it was still loud  12 ft away.
[05:35:41] <Tekni> sfire||mouse: when you boot it up, it should be prompting you for system info
[05:35:51] <Tempt> boyd: Actually, I think they like the Galapagos Islands
[05:35:53] <delewis> T2000s put out a lot of heat from what I've heard.
[05:35:58] <DoYouKnow> installing solaris on qemu :)
[05:35:59] <sfire||mouse> pauliukas_: url for the blog? I only have benr's
[05:36:00] <delewis> which is why I question the power efficiency.
[05:36:01] <boyd> jamesd: the 1000 is louder than the 2000
[05:36:09] <pauliukas_> sfire||mouse: http://www.jmcp.homeunix.com/blog
[05:36:14] <boyd> Tempt: Komodo Island I think
[05:36:15] <jamesd> boyd, i know and whiney as well...
[05:36:16] <DoYouKnow> well, still dling
[05:36:16] <Tempt> jamesd: Man, surely it would have been easier just to fix the toaster
[05:36:17] <pauliukas_> http://www.linkedin.com/in/jamescmcpherson
[05:36:22] <pauliukas_> I'm good at googling people :-P
[05:36:26] <sfire||mouse> Tekni: would start /SP/console be a good place to start
[05:36:42] <edp> sfire||mouse, yes
[05:36:49] * boyd is getting linked in requests from people he doesn't know
[05:36:52] <jamesd> Tempt, the t2000 didn't get hot enough... the air felt just a touch warmer than room temp  even after being on for over an hour.
[05:37:00] <flyingparchment> ah, finally got a local copy of solaris docs.. no more will i have to wait for docs.sun.com
[05:37:04] <Tempt> jamesd: work it harder. They get hot with load.
[05:37:06] <jamesd> Tempt, they have windows boxes that they would use for a toaster.
[05:37:20] <boyd> flyingparchment: untill next week, when it's out of date
[05:37:23] <sfire||mouse> I've got root@unknown #
[05:37:30] <flyingparchment> boyd: heh.  not planning to upgrade for a while :)
[05:38:17] <jamesd> you know where you are in the datacenter when you walk behind the rack of xeons....  even at the start of the row where just 2 ft before you were getting blasted by frigid air.
[05:38:42] <pauliukas_> I'm presuming old Xeons.
[05:38:54] <sfire||mouse> hmm, have more now
[05:39:00] <pauliukas_> The new 3000 and 5100 series are considerably cooler.
[05:39:12] <pauliukas_> The 5000 and older were based on crappy Netburst.
[05:39:34] <jamesd> paul, i think they are 1.8ghz and faster...
[05:39:47] <pauliukas_> 1.8GHz... that must be old.
[05:40:28] <jamesd> pauliukas_, i'm not in charge of the windows boxes... so i don't keep track of what is in them...
[05:40:38] <pauliukas_> heh. I sure hope not.
[05:40:51] <pauliukas_> If you were a Windows sysadmin, I would have ceased talking to you!
[05:40:51] <pauliukas_> :-P
[05:41:07] <flyingparchment> funny, half this channel feels the same way about linux
[05:41:08] <flyingparchment> ;-)
[05:41:14] <pauliukas_> LOL
[05:41:18] <pauliukas_> hahah
[05:41:22] * pauliukas_ backs away slowly.
[05:41:36] <jamesd> pauliukas_, is it okay that i sit in the same aisle with them, they are less than 20 ft away, and in the other direction there is a MS sql  dba.
[05:41:37] <sfire||mouse> edp: what would start /SYS do in the ILOM?
[05:41:46] <flyingparchment> sfire||mouse: turn the computer on
[05:42:05] <sfire||mouse> would that start the system config if it hasn't be done yet?
[05:42:19] <flyingparchment> i don't know how sun pre-configures solaris
[05:42:24] <flyingparchment> but i would expect so, yes
[05:42:31] <sfire||mouse> eh, lets try :)
[05:42:39] <sfire||mouse> target already started
[05:42:42] <sfire||mouse> foo
[05:42:51] <flyingparchment> reset /SYS
[05:43:10] <pauliukas_> Hey... when is Solaris 11 going to be out?
[05:43:19] <sfire||mouse> yeah, trying that now
[05:43:20] <pauliukas_> Or is it one of these things that will come with time?
[05:43:39] <sfire||mouse> yay bios :)
[05:43:46] <pauliukas_> They have that "10 moves ahead of the competition" thing for Solaris 10.
[05:44:01] <pauliukas_> I wonder what they're planning for 11... "Just don't worry about them. They're gone"
[05:44:09] <flyingparchment> sun seems to be interested in extending the life of S10 as long as possible, so 11 might be a while
[05:44:19] <flyingparchment> (a lot of features from nevada are being backported)
[05:44:34] <pauliukas_> nevada being the codename of 11?
[05:44:39] <flyingparchment> yes
[05:44:50] <pauliukas_> Can you give me just a quick example of the most amazing feature in Nevada?
[05:45:02] <pauliukas_> I swear to you that I'll go on wikipedia after annoying the heck out of you.
[05:45:12] <jamesd> pauliukas_, zfs, dtrace, the list goes on...
[05:45:19] <pauliukas_> Oh... these were from 11?
[05:45:21] <Triskelios> jamesd: I think he meant not in S10
[05:45:22] <pauliukas_> didn't know that...
[05:45:38] <flyingparchment> i can't think of one off hand.  i guess nwam is nice.  better hardware support..
[05:45:48] <flyingparchment> (although some SATA stuff just went into s10u4)
[05:45:51] <Triskelios> decent desktop support
[05:46:19] <flyingparchment> ZFS came from nevada, dtrace was in the initial S10 release
[05:46:43] <flyingparchment> linux zones came from there too, and the iscsi target.. i think most of the interesting features got integrated in u4
[05:46:47] <pauliukas_> Ah man... This Linux container is so cool
[05:46:50] <flyingparchment> (which was just release like yesterday)
[05:46:51] <pauliukas_> I know what I'll do...
[05:46:59] <jamesd> the stuff in solaris 11 at moment is just enhancements to solaris 10,  perhaps next year some really cool stuff will be going in, crossbow, clearcase,  and ZFS boot for sparc and full ZFS support in the installer will  all go in hopefully.
[05:47:06] <pauliukas_> Install cPanel on a  container just for fun! And rm -rf it because I hate it that much
[05:47:15] <jamesd> oops  s/clearcase/clearview/
[05:47:29] <jbk> well some of it was stuff they had promised in 10 anyway
[05:47:41] <pauliukas_> Here's another question for everyone.
[05:47:42] <jbk> but took longer than planned to integrate
[05:47:58] <pauliukas_> Can ZFS support multiple-size stripes/disks for RAID-Z?
[05:48:07] <jbk> i know one person who still distrusts zfs 'because it was late -- thus it must have more problems than they admit'
[05:48:10] <flyingparchment> pauliukas_: raidz stripe size is entirely dynamic
[05:48:11] <pauliukas_> I want to create myself a poor man's NAS for my home.
[05:48:19] <jamesd> pauliukas_, no
[05:48:22] <pauliukas_> by just patching up all the spare HDDs I can find.
[05:48:28] <flyingparchment> but you can't mix disk sizes
[05:48:28] <pauliukas_> ah.
[05:49:01] <jbk> though if you have 3 of one size, and 3 of another, you could create 2 raidz vdevs
[05:49:09] <jamesd> pauliukas_, buy a bunch of 500GB sata drives...   $300 usable terabyte of raidz storage more if compressed... pretty awesome deal
[05:49:10] <Triskelios> pauliukas_: you can strip different disk sizes but each raid component has to be the same size (or larger)
[05:49:25] <Triskelios> *stripe, rather
[05:49:26] <pauliukas_> jamesd: Yeah... I think I'll be thinking of doing that.
[05:49:34] <pauliukas_> Now that I don't need to get expensive RAID controllers :-P
[05:49:50] <flyingparchment> if one disk was twice as large as the other, you could create two vdevs on it :)
[05:49:53] <pauliukas_> The only problem is that I have old computers here. Read, P3 with 256 of RAM.
[05:49:55] <flyingparchment> (you'd need raidz2 then though)
[05:50:10] <flyingparchment> zfs in 256MB will suck
[05:50:15] <pauliukas_> indeed.
[05:50:33] <jamesd> flyingparchment, yeah,   zfs needs 1GB minimum currently...
[05:50:56] <flyingparchment> i actually used it for a bit in a 256M system, and it liked to deadlock :)
[05:51:08] <flyingparchment> (that was around snv_70 though)
[05:51:20] *** Chihan has quit IRC
[05:51:52] <jamesd> my blade 1500 with  512MB was painful, it went from usable as a desktop, to  usable as a fileserver, not both and usually took 30-60 seconds for it to switch the roles.
[05:52:35] <jamesd> its much nicer with 1.5GB of ram
[05:52:58] <tsp> I'm running zfs on 256mb, and by some miracle its working better now than it was last week
[05:53:32] <flyingparchment> yeah, i've used it a bit on a 3GB system and it was good.. never really tried it with a production load  though
[05:54:18] <jamesd> its awesome even on my u2... 2x 300's and 2GB of ram
[05:55:36] <Tempt> So what we can really say is there is nothing particularly compelling in SX for a server that you don't get in Solaris 10 ;-)
[05:55:37] <Triskelios> ZFS works acceptably on on my U5 with 384MB RAM, 7 x 20G disk array...
[05:55:58] <moazamraja> works great on my SB 2000 with 2GB
[05:56:05] <moazamraja> <-- had to put in his 2 cents
[05:56:08] <flyingparchment> Tempt: until a month or two when the next batch of cool stuff goes in :)
[05:56:36] <jamesd> i guess the array helps, but  zfs sucks on my blade150 at work with   2x 80GB ide mirrored...  3-5 seconds to get a prompt back sucks.
[05:56:45] <pauliukas_> So how are Linux containers created?
[05:56:46] <flyingparchment> hmm, xen and IP virtualisation.. what else is coming  soon?
[05:56:56] <flyingparchment> pauliukas_: the same way as solaris containers, but you set the brand to lx
[05:57:00] <pauliukas_> Stupid video didn't give out any commands, unlike the other Solaris container video.
[05:57:08] <pauliukas_> flyingparchment: And what about the whole distro thign?
[05:57:10] <flyingparchment> pauliukas_: it automatically extracts/installs CentOS or RHEL from a CD
[05:57:16] <pauliukas_> NEAT!
[05:57:17] <flyingparchment> (you gave to give it the cd)
[05:57:25] <moazamraja> damn near everything sux on a Blade 150.
[05:57:36] <flyingparchment> i hear you can run other distributions on it, but not out of the box.  there are various how-tos around about that
[05:57:41] <Triskelios> pauliukas_: I have a Debian etch zone on my laptop
[05:57:52] <flyingparchment> (the main limitation is that it only emulates a 2.4 kernel - so RHEL/CentOS 4+ won't work)
[05:57:53] <pauliukas_> I'm a CentOS guy, so it's no problem for me.
[05:57:57] <pauliukas_> Oh...
[05:59:03] <flyingparchment> the last comment i saw from the brandz team was that there's "no active interest" in sun adding 2.6 emulation.. i'm not really sure what the future of brandz is given that.
[05:59:12] <pauliukas_> sad..
[05:59:33] <pauliukas_> What's the best place to find information on creating a container or Linux container?
[05:59:39] <flyingparchment> (i can't see 2.4 being competetive for long.. doesn't RHEL 3 reach EOL in a couple of years?)
[05:59:43] <flyingparchment> pauliukas_: docs.sun.com
[06:00:06] <pauliukas_> Already tried that, I can't seem to search like a human being,.
[06:00:08] <Triskelios> pauliukas_: google for brandz is usually fine
[06:00:38] <CIA-26> se146197: 6515554 sdpadm does incorrect check on args passed to it, 6519010 SIOSYSENABLESDP is misnamed, 6564594 "BAD TRAP: module "genunix" due to a NULL pointer dereference" on amd64 infinband systems in snv_66, 6599427 sdp needs to validate work queue entry before processing the mblk, 6599436 sdpib leaking memory during detach
[06:01:37] <boyd> flyingparchment: I agree, it seems like a short-term "lets get the RHEL users onto solaris" option to me.
[06:02:15] <pauliukas_> Is there any way to use DHCP for containers.
[06:02:47] <boyd> pauliukas_: Yes, if they are exclusive IP zones
[06:02:54] <pauliukas_> Meaning...
[06:02:59] <pauliukas_> Static IPs?
[06:03:01] <flyingparchment> pauliukas_: http://docs.sun.com/app/docs/doc/817-1592/6mhahup1g?a=view
[06:03:22] <boyd> no, they need their own interface (which must be GLDv3)
[06:03:37] <boyd> then they can do anything they want (change IP, adjust routes, etc)
[06:03:44] *** sahafeez has quit IRC
[06:03:59] <boyd> Basically they get their own IP stack instance, rather than sharing, as happens normally
[06:05:10] <boyd> "set ip-type=exclusive" in zonecfg then when you "add net" you only "set physical=xxx", not an address. You set all the network details up in the zone like you do in the global zone
[06:05:13] *** Fullmoon has quit IRC
[06:05:21] <Tempt> We really need to beat gdamore harder
[06:05:25] <Tempt> Get cassini done
[06:05:39] * boyd has no cassini, so he doesn't care :)
[06:06:48] * Tempt eats cassinis for breakfast
[06:06:55] * jmcp_ dislikes cassinis
[06:07:08] <boyd> Yay bge! Boo ce!
[06:07:10] <Tempt> I don't like cassini per se, I just like quad gig-e
[06:07:16] <Tempt> bge can get bent
[06:07:20] <Tempt> WHAT JUMBO FRAMES?
[06:07:21] <flyingparchment> repent cassini, and accept e1000g and your new overlord
[06:07:21] <Tempt> :)
[06:07:26] <mlh> brandz only does linux 2.4.x? that's crazy
[06:07:35] <Tempt> I've got some e1000gs as well, but only dual port, not quad
[06:07:50] <boyd> mmmmmmmm.... neeeeeptune...
[06:07:55] <Tempt> Yeah, the idea of mixing Linux in on your nice clean hardware. They should get rid of it, pronto
[06:08:05] <Tempt> boyd: Are you buying me neptune cards? Ooh, gimme, gimme.
[06:08:12] <flyingparchment> i quite like bge, but i've heard the driver is kind of screwey (seems to work okay though)
[06:08:23] <Tempt> boyd: I'll refrain from lolspeak for 1 week per port of neptune provided
[06:08:33] * boyd goes off to order 52
[06:09:12] * gdamore is not doing cassini.
[06:09:16] <Tempt> Make sure you don't order the PCI express version, I need PCI-X
[06:09:30] <Tempt> gdamore: The beatings will increase until cassini is finished!
[06:09:46] * gdamore directs Tempt at Joyce Yu.
[06:09:59] * Tempt pokes gdamore with a stick
[06:10:02] <flyingparchment> isn't 10GE faster than the PCI-X bus?
[06:10:09] <pauliukas_> yes
[06:10:19] <boyd> If you want a picture of the future, imagine gdamore's boot crushing a cassini adaptor. Forever.
[06:10:39] <gdamore> heh.
[06:10:40] <Tempt> Sun will probably just drop cassini support soon anyway
[06:10:47] <Tempt> like they did with the earlier sbus gig-e cards
[06:10:52] <Tempt> Suddenly dropped after 2.6
[06:10:53] <gdamore> no.  this is a major snafu right now.
[06:11:04] <gdamore> are you talking about "vge" ?
[06:11:11] <jbk> Tempt: probably not -- there's still a _lot_ of ce cards out there
[06:11:19] <boyd> There's a lot of ce still out there.... a lot... and they're still shipping them (or were until v recently)
[06:11:22] <gdamore> ce is the onboard nic on a number of machiens as well
[06:11:23] <Tempt> I'm sure there were a lot of those SBus gig cards as well
[06:11:30] <boyd> (on the gigaswifts)
[06:11:31] <gdamore> no, there weren't actually
[06:11:35] <Tempt> Didn't stop them vanishing overnight, and Sun even pulled the drivers
[06:11:38] <gdamore> Sbus gem is still supported.
[06:11:46] <Tempt> Gem is the later card
[06:11:49] <Tempt> There was one before it
[06:11:50] <gdamore> sbus vge was rare ... i've never seen one.
[06:11:55] *** bengtf has quit IRC
[06:11:58] <Tempt> I can't remember, I sold all my cards on ebay :)
[06:12:11] * gdamore might convert gem to GLDv3.
[06:12:24] <gdamore> because gem == eri + gigabit xcvr, and he already converted eri
[06:12:26] <Tempt> gdamore: Not a bad move, considering there are still GEM cards out there
[06:12:32] * boyd wonders what brendang is going to do about his dtrace bug.
[06:12:32] <Tempt> gdamore: like the 880 onboard
[06:12:55] <boyd> Tempt: the v880 has a gem?
[06:12:59] <gdamore> ce is a political mess, owned by "engineers" with too many lines of code and too little clue....
[06:13:03] <gdamore> v880 has a ce.
[06:13:26] <boyd> Ah... sorry.. stiched the wrong comments together
[06:13:54] <Tempt> boyd: Yes, the 880 has a GEM
[06:14:05] <gdamore> NSN got rankled when they heard I was working on a GLDv3 port... they claimed they (Joyce) were doing one, and I was directed to "back off"
[06:14:08] <Tempt> boyd: eri for the copper, ge for the fibre
[06:14:12] <gdamore> so I'm ignoring it.
[06:14:25] <gdamore> eri is not just copper, but 100BaseT ... not gigabit
[06:14:27] <Tempt> Everyone forgets about the daktari's poor little fibre port, hidden away
[06:14:32] <gdamore> gem is fiber only.
[06:14:41] <Tempt> yes, and the 880 has a cheery fibre interface
[06:14:56] <boyd> .... And so many things to plug it into
[06:15:24] <flyingparchment> even low-end ge switches tend to have a couple of fibre ports nowadays
[06:15:25] <Tempt> boyd: Given I've got about 20 Sun fibre cassini cards in my desk drawer, I think there's plenty to plug it into
[06:15:30] <Tempt> They make a handy cluster interconnect on 880s
[06:15:31] * gdamore decided it would be more fun to work with 3rd party drivers than with NSNs....
[06:15:46] <boyd> gdamore: I wish they'd have a community vote... who do you want to work on your ce driver? gdamore or some faceless bunch of people who seem to not do anything?
[06:16:06] <Tempt> The community could all chip in a few bucks.
[06:16:19] * boyd checks his pockets...
[06:16:26] <flyingparchment> is ce open?
[06:17:08] <boyd> I have 1020 Indian Rupees, 5 Aus dollars and 4 Singapore bucks.
[06:17:14] <delewis> flyingparchment: the specs are :-)
[06:17:23] <boyd> I think that adds up to about $1.50 US.. is that enough?
[06:17:32] *** sahafeez has joined #opensolaris
[06:17:46] <pauliukas_> uhh... Any help?
[06:17:46] <pauliukas_> bash-3.00# zonecfg -z test
[06:17:47] <pauliukas_> Sorry, I don't know anything about your "xterm-color" terminal.
[06:17:50] <gdamore> ce is definitely *not* open.
[06:18:03] <boyd> pauliukas_: export TERM=xtermc
[06:18:04] <flyingparchment> pauliukas_: try "dtterm" or "xterm"
[06:18:13] <gdamore> I wouldn't want it open in its current state... I'd be *embarrassed*  but I was going to open it as part of the GLDv3 work
[06:18:14] <boyd> .. or just xterm
[06:18:14] <pauliukas_> And what does that so?
[06:18:15] <flyingparchment> or xtermc, but that describes a version of xterm you probably aren't using
[06:18:16] <pauliukas_> *do
[06:18:27] <delewis> gdamore: at this point would it be easier to just use the Cassini specifications Sun has provided and re-implement ce(7D)?
[06:18:40] <pauliukas_>  Well, I'm logged in via SSH on bash.
[06:18:46] <delewis> last I heard, the ce code was in pretty bad shape.
[06:18:47] <boyd> Hah! a rebel ce driver
[06:18:50] <gdamore> probably.  there is a *BSD driver that is a lot cleaner than the crap from NSN
[06:18:52] <boyd> pauliukas_: from a mac?
[06:18:58] <pauliukas_> boyd: Yes.
[06:19:08] <pauliukas_> CTCP does wonders, eh?
[06:19:16] <boyd> pauliukas_: No, I recognise that msg
[06:19:22] <gdamore> well, since about 11,000 lines (at least) of the 22KLOC of ce.c are swill, I'd say yes, pretty bad shape.
[06:19:24] <pauliukas_> Ah...
[06:19:31] <pauliukas_> Anything I can do about that?
[06:20:01] <boyd> pauliukas_: the mac has a terminal type that solaris doesn't understand. Tell solaris you have either an "xterm" or an "xtermc" and it shoul;d work
[06:20:06] <pauliukas_> okay.
[06:20:09] * gdamore has a hard time believing *any* NIC driver should be over 10KLOC.  nxge is 33 (or 36?) KLOC
[06:20:22] <pauliukas_> yay. it worked. thanks.
[06:20:34] <mlh> you could also add xterm-color as an alias for xtermc if you wanted to
[06:20:41] <mlh> in the terminfo database
[06:20:48] <pauliukas_> Do I need to do this at every login?
[06:21:01] <boyd> pauliukas_: unless you put it in your .profile or simialr.
[06:21:04] <boyd> similar
[06:21:05] <pauliukas_> gotcha.
[06:21:10] <mlh> gosh, raise a bug in opensolaris for it .. it'd get back to solaris eventually
[06:21:26] <Tempt> The question is, do we chip in a few bucks each for gdamore to write the code, or chip in a few bucks for someone to beat gdamore until the job is done?
[06:21:36] * boyd does (on the mac) TERM=xterm ssh otherbox to avoid having to worry about it
[06:21:54] <boyd> Tempt: How much extra to get the video on YouTube?
[06:22:00] <gdamore> Well the *former* won't work, because I can't work due to conflict-of-interest considerations.  (I'm not allowed to touch NSN's code.)
[06:22:09] <Tempt> boyd: Free with the beatings, I'd assume
[06:22:12] <mlh> it'd be nice for it to be solved rather every mac user fiddle with their .profiles
[06:22:15] <delewis> NSN?
[06:22:18] <gdamore> THe latter *won't* work either, for the same reasons, but it will piss me off. :-)
[06:22:27] <boyd> gdamore: what about out-of-hours, when you have your community hat on?
[06:22:33] <gdamore> NSN is the group that owns the NICs, etc.
[06:22:38] <delewis> ah.
[06:22:39] <Tempt> (That's the pointy one)
[06:22:42] <delewis> that's an internal Sun group?
[06:22:47] <gdamore> boyd: nope.  as ce.c is closed, and as it relates to Sun's business, I can't.
[06:22:54] <boyd> oh, yeah...
[06:23:10] <delewis> so could we just dump the closed ce and import the BSD-licensed ce?
[06:23:11] <gdamore> I already asked about off-hours helping of a 3rd party 10GbE vendor, and was told it represented conflict-of-interest.
[06:23:12] <Tempt> There's the classic problem with cassini
[06:23:17] <Tempt> The only people with cassini cards are Sun customers
[06:23:24] <Tempt> It is against the current spirit to do anything for customers
[06:23:30] <Tempt> Hence nothing shall be done; case closed.
[06:23:31] <boyd> Doesn't that imply that *any* work that a sun person does on opensolaris is off-limits?
[06:23:41] <gdamore> boyd: no.
[06:23:52] <gdamore> but in this case, it doesn't suit NSN, so its a no-no.
[06:24:24] <gdamore> and working on 3rd party would be seen as competitive with NSN.  and possibly conflict of interest with my day job... because my group already has some IHV relationships.
[06:24:35] <gdamore> (we encourage IHVs to write their own drivers....)
[06:24:50] <boyd> We should all file RFEs against ce.. "ce *must* be nemoized"
[06:25:09] <delewis> but wouldn't NSN see ce maintenance as making what they produce more marketable?
[06:25:12] <gdamore> and, as my position is funded in part by Intel, anything that competes with Intel NICs is also off-limits for me.
[06:25:19] <delewis> it's a win-win situation for everyone involved.
[06:25:22] <boyd> gdamore: So, basically emplying you was a way for sun to stop you working on community net drivers? (in effect, if not intention)
[06:25:24] <gdamore> delewis: stop trying to be rational!
[06:25:40] <delewis> you're willing to fix NSN code, thus not requiring any NSN people, and the ce consumers benefit from the improved code...
[06:25:42] <delewis> gdamore: :-)
[06:25:45] <gdamore> employing me in *current* position, yes, to a limited extent.
[06:25:54] <gdamore> but I couldnt' get funding elsewher.e
[06:26:15] * boyd wonders if it's Intell trying to stop competitor's NICs getting used :)
[06:26:28] <delewis> well, we at least have ce specs now... http://www.sun.com/processors/manuals/cs_plus.pdf
[06:26:33] <Tempt> Nah, it'd be Sun trying to give their customers another fistful of fail.
[06:26:36] <delewis> just no one to work on the code :-(
[06:26:42] <gdamore> NSN are already unhappy with me for "fixing" hme, dmfe, qfe, and eri.  I guess they were using all those lines of code for job justification
[06:26:50] <flyingparchment> gdamore: doesn't any nic work compete with intel?
[06:26:51] <boyd> lol
[06:26:58] <delewis> this sounds like SPARC graphics pt. deux.
[06:27:06] <boyd> gdamore: You need to get into the gnome team so you can work on NICs :)
[06:27:16] <pauliukas_> What does Solaris use instead of wget?
[06:27:21] <Tempt> I guess the gnome team has the most budget
[06:27:22] <gdamore> the 100Mbps work isn't seen as competitive.  and I can help to a certain extent, but ... writing full drivers, probably not.
[06:27:23] <boyd> pauliukas_: wget
[06:27:23] <delewis> pauliukas_: wget.
[06:27:24] <pauliukas_> And is there some sort of package manager thing?
[06:27:24] <Tempt> pauliukas_: We use wget
[06:27:29] <pauliukas_> bash-3.00# wget
[06:27:30] <boyd> pauliukas_: /usr/sfw/bin/wget
[06:27:30] <pauliukas_> bash: wget: command not found
[06:27:30] <Tempt> pauliukas_: pkgadd, pkgrm
[06:27:31] <pauliukas_> What now?
[06:27:32] <gdamore> GUI work is boring, and I suck at it.
[06:27:37] <Tempt> boyd: spoilsport
[06:27:38] <pauliukas_> oh...
[06:27:47] <delewis> pauliukas_: pkgadd(1)
[06:27:57] <pauliukas_> Any way to make it work like the good ol' wget without any path in front og it?
[06:28:04] <Tempt> yeah
[06:28:05] <gdamore> but yes, the NSN mindset is much the same as the SPARC graphics mindset... both are groups that IMO Sun should just ditch.
[06:28:06] <Tempt> set your PATH
[06:28:09] <Tempt> (ffs!)
[06:28:26] <Tempt> I'm sure Sun will ditch them soon
[06:28:31] <Tempt> along with those pesky SPARC wankers
[06:28:38] <boyd> gdamore: so, I'm confused... you basically can't work on any NIC drivers... what are you meant to be doing?
[06:28:38] <Tempt> (and pesky irritating customers)
[06:28:42] <delewis> gdamore: does NSN own Neptune, as well?
[06:28:45] <boyd> (apart from IRC, obv)
[06:28:48] <gdamore> delewis: yes.
[06:28:55] <Tempt> Don't we all get paid to IRC all day?
[06:29:03] <delewis> they won't be dying anytime soon, then, unless Neptune gets transferred elsewhere.
[06:29:04] <jbk> Tempt: i can't irc from work :(
[06:29:09] <jbk> can't even ssh out
[06:29:18] <gdamore> boyd: I have my own NIC drivers, but really the work is much more varied, core platform support, power management, i2c, i8042, all kinds of stuff....
[06:29:21] <boyd> jbk: Geez... where are you, Guantanamo?
[06:29:29] <boyd> gdamore: Ah
[06:29:30] <jbk> no, investment bank
[06:29:36] <Tempt> jbk: https?
[06:29:43] <boyd> Here we go...
[06:29:55] <Tempt> Sounds like a job for ...
[06:29:56] <Tempt> SSGD!
[06:29:57] <jbk> i'm pretty sure i could get around it, but it's one of those things not worth the job
[06:30:02] <jbk> +risking
[06:30:07] <gdamore> some of the work I've done, like the rtls fixups for suspend/resume, are seen as supportive and so makes mgmt happy.  my mgmt is also happy about the work I've done for afe and mxfe.
[06:30:18] <gdamore> (afe was committed earlier today, btw.  yay.)
[06:30:23] <g4lt-mordant> jbk hell, insomnia got out and she worked at a bank.  of course, she used her crackberry more often than not, but hey
[06:30:42] * boyd wishes he could log calls as a customer that are not bugs or rfes but kudos.
[06:30:43] <pauliukas_> "mkdir: Failed to make directory "paulius"; Operation not applicable"
[06:30:44] <jbk> investment bank, or just commercial bank?
[06:30:48] <nachox> gdamore, all those are being opensourced?
[06:30:57] <flyingparchment> boyd: log a request for an already-implemented feature :)
[06:31:07] <boyd> Ugh... IRC on crackberry...
[06:31:10] <Tempt> boyd: Hey, I wish I could log faults, bug and RFEs without getting sent to the black call centre of calcutta
[06:31:19] <nachox> pauliukas_, let me guess, you were trying to create that at /home
[06:31:27] <jbk> Tempt: you need to be one of their stragetic accounts :)
[06:31:28] <pauliukas_> you assumed that right.
[06:31:28] <flyingparchment> i wish i could buy any support to begin with.
[06:31:33] <boyd> Tempt: I suspect that wasn't meant to sound quite like it did :)
[06:31:33] <jbk> you get a separate dial in number
[06:31:34] <Tempt> I've taken to declaring all Solaris bugs to be unfixable by Sun
[06:31:35] <gdamore> nachox: well, I'm tring to get rtls opensourced, but afe and mxfe already were....  I get my name in the list of OpenSolaris copyright holders now... heh.
[06:31:42] <Tempt> I can't be bothered logging calls
[06:31:46] <g4lt-mordant> time to trot out jamesd's blog...
[06:31:54] <Tempt> I'd be better off buying a couple of piglets and trying to explain the faults to them
[06:31:55] <nachox> pauliukas_, /home in solaris is managed by the automounter, that is why you get that error
[06:32:00] <Tempt> At least I could eat them afterwards
[06:32:04] *** Tekni has quit IRC
[06:32:20] <nachox> gdamore, and voting privileges? :)
[06:32:25] <jamesd> pauliukas_, http://www.google.com/url?sa=t&ct=res&cd=2&url=http%3A%2F%2Fuadmin.blogspot.com%2F2005%2F02%2Fyoure-never-far-from-home.html&ei=RDHeRtSULpnAiAGAioStAg&usg=AFQjCNEUrS_FBL-qulX_6rK7KOKkWir9pA&sig2=mL4CY5BMVcyQC1zq7BTgeA
[06:32:40] <nachox> tinyurl....
[06:32:49] <pauliukas_> thanks.
[06:32:51] <gdamore> I already have opensolaris voting privs.  :-)  but that means jack all within Sun....
[06:32:53] <jamesd> http://uadmin.blogspot.com/2005/02/youre-never-far-from-home.html
[06:33:00] <Tempt> jbk: Yes, when I moved from my last location (Which was one of Sun's biggest customers) to here, suddenly my support calls get ignored, nobody cares and my problems can go do in a fire
[06:33:26] <Tempt> argh
[06:33:41] *** bengtf_ has joined #opensolaris
[06:33:42] *** bengtf_ is now known as bengtf
[06:33:46] <nachox> Tempt, have to love the market economy :) nobody gives a shit if you dont have the money
[06:34:05] <Tempt> Hey, they still charge for gold support
[06:34:10] <Tempt> you just don't get anything in return
[06:34:33] <Tempt> Sun would do better having a walk in spare parts barn
[06:34:38] <gdamore> your support calls should *not* be ignored.
[06:34:41] <Tempt> swap over your dead components
[06:34:53] <Tempt> gdamore: They aren't *ignored*, they just circulate around indian call centres
[06:35:05] <Tempt> With a weekly "please be sending of another explorer"
[06:35:10] <nachox> Tempt, and soon to Argentina :)
[06:35:17] <gdamore> oh, sounds like you have Sun's internal ITops helpdesk instead of the one customers are supposed to get....
[06:36:13] <gdamore> Tempt: is your problem bad hardware, bad software, or you don't know?
[06:36:32] <Tempt> gdamore: My problem is long gone
[06:36:40] <gdamore> heh.
[06:36:41] <pauliukas_> How do I install a C compiler on this thing?
[06:36:52] <flyingparchment> pauliukas_: /usr/sfw/bin/gcc
[06:36:59] *** Fullmoon has joined #opensolaris
[06:37:00] <nachox> pauliukas_, there is gcc if you installed all of solaris
[06:37:01] <flyingparchment> pauliukas_: or download studio, www.sun.com/software/products/studio/
[06:37:03] <Tempt> The Sun line is pretty much like Windows now
[06:37:07] <Tempt> retry, reboot, reinstall, repeat
[06:37:15] <jmcp_> pauliukas_: if you did a Solaris Express Developer Edition installation, then you should have Sun Studio in /opt/SUNWspro/bin
[06:37:16] <nachox> pauliukas_, otherwise, install sun studio, it's free
[06:37:19] *** perlmonk has quit IRC
[06:37:20] <Tempt> To be honest, Oracle support isn't much better
[06:37:37] <Tempt> The difference is Oracle aren't trying to position themselves as a free-software sell-the-support company
[06:37:38] <pauliukas_> How can I set the thing to run my binaries from /usr/sfw/bin?
[06:37:44] <gdamore> Everyone outsources call centers, and they have brain damaged "scripts" ... I hate that.
[06:37:49] <pauliukas_> So that I don't have to do the full path at every time.
[06:37:53] <Triskelios> pauliukas_: edit PATH in /etc/default/login
[06:37:54] * flyingparchment inserts obligatory compliment for mysql support -- much better than the actual product they support..
[06:37:57] *** Fullmoon has quit IRC
[06:38:02] <nachox> of course it isnt, they are diverting all the engineers to supporting their in house version of linux
[06:38:39] <Triskelios> pauliukas_: I use PATH=/usr/bin:/sbin:/usr/sbin:/usr/ccs/bin:/usr/dt/bin:/usr/openwin/bin:/usr/X11/bin:/usr/sfw/bin:/usr/sfw/sbin:/opt/sfw/bin:/opt/sfw/sbin:/opt/csw/bin:/opt/csw/sbin:/opt/jdsbld/bin:/opt/onbld/bin:/opt/SUNWspro/bin
[06:38:53] <pauliukas_> thanks a lot.
[06:39:06] <Tempt> /bin:/usr/bin:/sbin:/usr/sbin:/usr/local/bin:/usr/ccs/bin:/usr/dt/bin:/usr/openwin/bin:/usr/sfw/bin:/usr/sfw/sbin:/opt/csw/bin:/opt/SUNWspro/bin:/usr/local/bin:/global/oracle/product/10.2.0/Db_1/bin:/opt/PCOWtools/bin on this particular host
[06:39:13] <flyingparchment> don't you need to know this stuff to run a Linux web hosting company? :)
[06:39:22] <nachox> Triskelios, you have all that in /etc/default/login ?
[06:39:28] <Triskelios> nachox: yup
[06:39:29] <Tempt> ouch, there's two /usr/local/bins in there
[06:39:30] <pauliukas_> Linux normally preconfigures these kinds of profile stuff.
[06:39:34] <Tempt> script must be having a picnic
[06:39:36] <g4lt-mordant> good god
[06:39:37] <nachox> that is sick
[06:39:49] <g4lt-mordant> /bin:/usr/bin:/opt/SUNWspro/bin:/opt/csw/bin:/usr/sfw/bin:/etc
[06:39:52] <delewis> pauliukas_: doesn't work that way on everything else.
[06:39:52] <Tempt> It'll be oraenv playing funny buggers
[06:39:55] <jbk> that is one thing i'm hoping will be addressed at the opensolaris summit
[06:40:07] <pauliukas_> Hah.
[06:40:10] <jbk> they default shell environment after an install leaves much to be desired
[06:40:17] <Tempt> It's a test
[06:40:24] <Tempt> If you can't configure a PATH, you don't get to play
[06:40:28] <Tempt> and you can go back to Windows
[06:40:30] <Tempt> :)
[06:40:31] <pauliukas_> lol
[06:40:37] <Triskelios> nachox: it's actually necessary for development work
[06:40:38] <flyingparchment> doesn't work that way on linux a lot of the time, there are a few things i install that use their own directories
[06:40:44] <g4lt-mordant> Tempt, what about CBA?
[06:40:56] * pauliukas_ bangs head on keybaord
[06:41:08] <pauliukas_> It opens up the login file as read-only.
[06:41:16] <Tempt> g4lt-mordant: CBA?
[06:41:21] <nachox> Triskelios, not really, there is a script that sets the environment or a custom profile file
[06:41:23] <g4lt-mordant> can't be arsed
[06:41:54] <Triskelios> nachox: nobody wants to reproduce that for every account
[06:41:58] <nachox> Tempt, you use both sunfreeware and blastwave?
[06:42:21] <Tempt> and PCOWtools
[06:42:22] <nachox> Triskelios, you have skel files to do that
[06:42:23] <Tempt> that's on my workstation
[06:42:34] <Tempt> my profile goes hunting for things to play with
[06:42:39] <Triskelios> nachox: ... not all the accounts were created locally
[06:43:04] <Tempt> I'm slowly phasing out blastwave
[06:43:07] <Tempt> bit-by-bit
[06:43:13] <Triskelios> Tempt: yeah, I'm doing the same...
[06:43:31] <Tempt> The dep trees are killing me
[06:43:51] <nachox> unless you're sharing your remotely mounted home dirs with different unices it shouldnt be a problem
[06:44:04] <Triskelios> nachox: we are, actually
[06:44:36] <flyingparchment> i do that -- found an interesting problem today, one of the options in my ~/.ssh/config is recognised by openssh 4.2, but not sun ssh
[06:44:37] <Tempt> My profile works with Solaris, AIX, FreeBSD, HPUX and Tru64
[06:44:39] <Triskelios> nachox: and I use the same thing on my laptop, don't want to have a custom ~/.profile everywhere
[06:45:03] <Tempt> When I was about 14 I discovered this tool
[06:45:04] <Tempt> called
[06:45:05] <Tempt> "uname"
[06:45:18] <boyd> Tempt: don't forget "if" and "case"
[06:45:33] <moazamraja> anyone here been able to compile libevent on solaris?
[06:45:36] <Tempt> Well, yes.
[06:45:44] <Tempt> if, case, uname.
[06:45:45] <Triskelios> moazamraja: yeah, I use it for transmission
[06:45:49] <nachox> then you're supposed to be using profiles that set the environment according to the result of uname, such a nasty thing as default is not very tidy, specially because all the shell scripts inherit that and there are tons of scripts that are not properly written and do not set PATH
[06:45:52] <flyingparchment> moazamraja: find strsep() from NetBSD and add it to the source tree
[06:45:56] <g4lt-mordant> boyd, your csh is showing ;P
[06:45:56] <moazamraja> i'm getting an error during compile
[06:45:57] <Triskelios> moazamraja: (the bittorrent client)
[06:46:03] <Tempt> You can even search for directories and stuff
[06:46:05] <moazamraja> cc -DHAVE_CONFIG_H -I. -I. -I..   -I.. -I../compat  -g -c regress.c
[06:46:06] <moazamraja> "regress.c", line 62: cannot find include file: "regress.gen.h"
[06:46:07] <Triskelios> moazamraja: try SFElibevent
[06:46:07] <boyd> g4lt-mordant: What!? where? ugh!
[06:46:13] <Triskelios> moazamraja: or check its patches, if any
[06:46:26] <g4lt-mordant> hint: try finding case in bourne shells
[06:46:41] <moazamraja> Triskelios: SFElibevent URL?
[06:46:52] <nachox> in any case, not my workstation, not my problem :P
[06:46:58] <flyingparchment> galt: what's wrong with the bourne shell case?
[06:46:59] <Triskelios> moazamraja: http://pkgbuild.sourceforge.net/spec-files-extra/
[06:47:18] <g4lt-mordant> flyingparchment, other than it's a godawful hack, nothing, nothing at all
[06:47:29] <boyd> Since libevent explicitly supports solaris event ports, I'm guessing it shouldn't be *that* hard to get going
[06:47:42] <flyingparchment> boyd: amusingly, the first release that supported ports didn't compile on solaris :)
[06:48:21] <boyd> Of course
[06:48:28] <Triskelios> moazamraja: ahhh, that package is forcing use of gcc
[06:48:47] <moazamraja> shit.
[06:48:54] <boyd> g4lt-mordant: ummm... it's there... complete with "esac" for the end
[06:48:55] <moazamraja> i'm trying to use sun compilers
[06:49:14] <moazamraja> ok..lemme contact someone in the compiler team and see if they have any tips
[06:49:22] <pauliukas_> Alright. I'm out to sleep.
[06:49:25] *** linux_user400354 has joined #opensolaris
[06:49:25] <pauliukas_> See y'all tomorrow
[06:49:30] <jmcp_> ciao
[06:50:00] <nachox> off to sleep, i need to be up in 5 hours
[06:50:01] <g4lt-mordant> boyd, and that, in a nutshell is the precise part I call a godawful hack.  cases should fall through
[06:50:12] *** laca has quit IRC
[06:50:13] *** pauliukas_ has quit IRC
[06:51:03] <boyd> g4lt-mordant: I'm thinking that this connection you make "boyd says case" -> "csh" is getting a little.. stretched.
[06:52:07] * Tempt yawns
[06:52:08] <Triskelios> g4lt-mordant: case is an extremely useful and accepted construct in bourne sh, if you have a complaint about its semantics... too bad
[06:55:22] <moazamraja>  pkgtool build --download SFElibevent
[06:55:22] <moazamraja> ERROR: SFElibevent not found
[06:55:22] <moazamraja> INFO: No spec files specified: nothing to do.
[06:55:23] <moazamraja> wtf....
[06:55:25] <moazamraja> grrr.
[06:55:45] <boyd> s/$/.spec/   IIRC
[06:56:42] <moazamraja> tried that.
[06:57:16] <Triskelios> moazamraja: you need the spec files actually there (svn co ... )
[06:57:27] <Triskelios> pkgtool only grabs the tarballs
[06:57:43] *** sahafeez has quit IRC
[06:57:58] <moazamraja> poop on a stick. :)
[07:05:38] *** uebayasi has quit IRC
[07:09:22] *** nachox has quit IRC
[07:21:32] *** BatonT has quit IRC
[07:24:50] *** estibi has joined #opensolaris
[07:30:01] *** BatonT has joined #opensolaris
[07:38:30] *** libkeise1 has joined #opensolaris
[07:47:46] *** sfire||mouse has quit IRC
[07:50:54] *** libkeiser has quit IRC
[07:53:33] *** estibi has quit IRC
[07:59:05] <boyd> Hey, does anyone know offhand where to get the OpenSolaris presentation templates?
[08:00:05] <boyd> Ah... here: http://opensolaris.org/os/community/advocacy/mktgdownloads/
[08:00:43] <Gman> http://www.opensolaris.org/os/community/advocacy/files/
[08:01:36] <boyd> Wow... lots there.. thanks
[08:04:38] <WickedWicky> hey all :)
[08:05:10] *** e^ipi has joined #opensolaris
[08:09:36] *** bengtf has quit IRC
[08:09:43] *** jmcp_ has quit IRC
[08:10:49] <flyingparchment> is there a file transfer program in S10 that can handle large files?  (not wget)
[08:11:22] <boyd> are you allergic to wget?
[08:11:29] <flyingparchment> oh, i have lftpget installed
[08:11:34] <flyingparchment> boyd: lftpget doesn't do large files
[08:11:50] <flyingparchment> it thinks it's 800MB, it's actually 25GB
[08:11:50] * boyd is surprised
[08:11:55] * flyingparchment was too
[08:12:03] *** linma has joined #opensolaris
[08:12:06] <flyingparchment> i mean, wget _does_ support it, it's just not compiled with support
[08:12:10] <flyingparchment> (at least i think it does..)
[08:16:58] *** e^ipi_ has joined #opensolaris
[08:21:40] *** Chihan has joined #OpenSolaris
[08:29:57] *** nostoi has joined #opensolaris
[08:33:26] *** e^ipi has quit IRC
[08:36:33] *** BatonT has quit IRC
[08:37:50] *** bengtf_ has joined #opensolaris
[08:37:51] *** bengtf_ is now known as bengtf
[08:39:35] *** WifiJane has quit IRC
[08:40:08] *** BatonT has joined #opensolaris
[08:44:12] *** e^ipi_ has quit IRC
[08:55:19] <WickedWicky> lol
[08:55:33] <WickedWicky> sometimes the stuff temp posts on his site makes me crack up
[08:57:35] * boyd thinks he has a better way to make Patrick's web server faster...
[08:59:11] <WickedWicky> totaly different machine?
[08:59:19] <WickedWicky> I was talking about this spam quote on www btw
[08:59:21] <WickedWicky> oh and check
[08:59:24] <WickedWicky> http://gizmodo.com/gadgets/google-phone/
[09:02:03] <ofu> what exactly is ds? Sep  4 20:08:00 caleb ds: NOTICE: ds@1: invalid message length, received 4128 bytes, expected 37536
[09:02:14] <ofu> i cant figure out what this may be
[09:02:34] *** Erwann has joined #opensolaris
[09:09:51] <trochej> Coffee?
[09:11:37] <quasi> trochej: waaaay ahead of you there
[09:12:16] <razrX> ofu: is that on 71 ?
[09:12:28] * quasi is already on the second cup but suspects it will be 3rd or 4th before waking up
[09:12:35] <razrX> 2nd here too
[09:13:13] <WickedWicky> same here
[09:13:43] <WickedWicky> traveling to work is usefull for catching up on the news, reading the email mess (thus knowing what hell is waiting for you) plus caffeinating yourself up
[09:16:24] <trochej> quasi: I only just got to job. My daughter broke and didn't wake me up before seven :)
[09:16:53] *** tsoome has quit IRC
[09:17:23] <ofu> razrX: no, U4
[09:17:28] *** bnitz has joined #opensolaris
[09:17:29] *** cydork has joined #opensolaris
[09:18:38] <quasi> trochej: I was on the train headed for work at 7
[09:18:48] *** e^ipi has joined #opensolaris
[09:20:19] *** Cyrille has joined #opensolaris
[09:21:02] *** e^ipi has quit IRC
[09:21:37] <Tempt> boyd: Does this involve stapping a rocket to it?
[09:21:54] <boyd> Tempt: :) replace with $100 hand-me-down PC :)
[09:22:05] <Tempt> Oh man
[09:22:06] <WickedWicky> haha
[09:22:08] <Tempt> That's harsh
[09:22:11] <WickedWicky> he didnt even know I was patrick
[09:22:14] <Tempt> I'd rather have the SPARCstation
[09:22:17] <WickedWicky> how's that for mocking someone from behind
[09:22:47] <WickedWicky> he still gave me his advise even after I told him I am Patrick though, that shows character :P
[09:22:58] <WickedWicky> and yes, Sparcstation > pentium
[09:23:17] <WickedWicky> Tempt, where did you get this turkey quote from?
[09:24:01] <Tempt> I literally recieved it in a spam
[09:24:07] <Tempt> one of the ones that made it past filtering, it was the only text
[09:24:19] <Tempt> There was another line about mastodons in a similar spam
[09:24:37] <WickedWicky> gotta love intertrans or babelfish
[09:28:39] <Tempt> alright, home time
[09:29:04] <moazamraja> you know...Sun should take the 24.1" monitor and just make it an all-in-one workstation/desktop
[09:29:05] <WickedWicky> be good
[09:29:20] <moazamraja> DVD on the side
[09:29:35] <moazamraja> HD inside, 4-8GB RAM max, opteron CPU
[09:29:39] <moazamraja> or some intel jobby.
[09:34:57] *** NikolaVeber has joined #opensolaris
[09:35:48] *** hspaans has quit IRC
[09:37:31] *** mikefut has joined #opensolaris
[09:38:35] *** tsoome has joined #opensolaris
[09:39:44] <quasi> moazamraja: Smac?
[09:40:23] *** myrkr has quit IRC
[09:42:14] <moazamraja> hehe
[09:42:15] <moazamraja> yeah
[09:42:30] <moazamraja> never gonna happen
[09:47:29] <quasi> I'm sure we'll soon be seeing howtos for rolling opensol onto imacs
[09:48:07] <moazamraja> that's happened already, i believe
[09:49:09] <quasi> I've seen it for minis and *books, but don't recall seeing any in the new setup
[09:49:53] *** MattMan has joined #opensolaris
[09:50:35] <moazamraja> i'll try it eventually
[09:50:42] <moazamraja> gotta backup my iMac HD first tho
[09:51:19] <moazamraja> i hve the nVidia based model, shouldn't be too bad
[09:54:17] <moazamraja> i'm running b70 via Parallels just fine though
[09:55:46] *** alfred has joined #opensolaris
[09:56:24] *** timsf has joined #opensolaris
[09:56:39] <timsf> morning all
[10:00:12] *** dmarker has quit IRC
[10:00:24] *** dmarker has joined #opensolaris
[10:02:26] *** mikefut has quit IRC
[10:02:37] *** mikefut has joined #opensolaris
[10:04:09] *** peteh has joined #opensolaris
[10:07:17] *** The-spiki has quit IRC
[10:18:19] <asyd> \_o<
[10:20:15] *** mikefut_ has joined #opensolaris
[10:22:39] *** mikefut_ has quit IRC
[10:22:44] <trochej> asyd: I always wonder, why are you running aroung holding a penis above your head? :P
[10:22:54] <asyd> lol
[10:23:39] *** otep has joined #opensolaris
[10:29:42] <WickedWicky> quack quack
[10:30:17] <trochej> :)
[10:31:29] <Tempt> duck season!
[10:31:53] *** nostoi has quit IRC
[10:32:45] *** nivox has joined #opensolaris
[10:36:27] *** ICU has joined #opensolaris
[10:36:44] *** mikefut has quit IRC
[10:37:59] <ICU> hi, is it possible to create a zpool as a raid0?
[10:38:02] <Tempt> man
[10:38:16] <Tempt> If you thought compiling stuff with studio was bad when it was written for gcc, try IBM's compiler!
[10:38:55] *** iMax has quit IRC
[10:39:33] *** iMax has joined #opensolaris
[10:39:56] <trochej> ICU: I believe not
[10:40:20] <Tempt> ICU: ZFS will automatically stripe I/O to a zpool
[10:40:42] <ICU> Tempt: i wan't max performance no reliability
[10:40:45] *** LLcoolM has joined #opensolaris
[10:40:56] <Tempt> ICU: so zpool create poolname c1t0d0s0 c1t1d0s0 (for example) will have no redundancy
[10:41:05] <Tempt> ICU: but writes will alternate across devices
[10:41:23] <ICU> Tempt: ah that would be too easy
[10:41:49] <Tempt> too easy?
[10:41:53] *** maxpil has quit IRC
[10:42:14] <trochej> Tempt: hmm. I haven't thought of that. Too obvious. :)
[10:42:23] <ICU> to easy to configure - thats too obvious
[10:42:27] <Tempt> That's why I'm here. To answer the obvious questions.
[10:45:34] <ICU> Thanks a lot :)
[10:46:06] <trochej> Tempt: Now, that you explained, I remember that I translated a part in ZFS guide that mentioned the striping :)
[10:50:09] *** boro has joined #opensolaris
[10:50:15] <boro> hello
[10:50:51] <timsf> ICU, have you looked at the ZFS Best Practices wiki ?
[10:51:04] <timsf> http://www.solarisinternals.com/wiki/index.php/ZFS_Best_Practices_Guide
[10:51:17] <ICU> timsf: yes but if its there i didn't see it
[10:51:25] <timsf>  - in some cases, a mirrored configuration will give you the best performance (according to that doc)
[10:51:41] <Tempt> At the expense of capacity.
[10:51:48] <timsf> raid0 ? A concatenation ? Nope.
[10:52:51] <boro> when i add "File System Management" profile to a user, profiles -l user will show just commands defined with "solaris" policy in exec_attr, no "suser" policy commands, is that a standard behaviour ?
[10:53:05] *** Fullmoon has joined #opensolaris
[10:55:24] <LLcoolM> is there a solaris distribution for desktop use?
[10:56:14] <Tempt> SXCE
[10:56:20] <Tempt> Full of desktop goodness
[10:56:58] <trochej> Truu
[10:57:49] <LLcoolM> where can i get it
[10:58:16] <Tempt> www.opensolaris.org
[10:58:18] <LLcoolM> or what does sxce stand for maybe its on opensolaris.org
[10:58:29] <LLcoolM> which one is it the express or the community edition
[10:58:34] <Tempt> Solaris Express Community Edition
[10:58:42] <LLcoolM> ah ok
[10:58:57] <trochej> LLcoolM: www.opensolaris.org/sxce_dvd or sxce_cd
[10:59:46] <LLcoolM> trochej: i assume the dvd version contains more packages
[11:00:07] <tsoome> lol, why it should?
[11:00:30] <richlowe> the DVD version has the DE bits.
[11:01:18] <tsoome> and cd version does not have it?
[11:01:35] <tsoome> somehow i doubt...
[11:02:29] <LLcoolM> hmm, it wont let me log in to the download area
[11:02:42] <tsoome> damn it seems the whole world is downloading s10 now:(
[11:02:57] <trochej> tsoome: Nope, at least me did it already :)
[11:03:05] <tsoome> 14KB/s transfer rate:(
[11:03:09] <trochej> LLcoolM: You need to register
[11:03:13] <LLcoolM> Error: Sorry, your Username and password could not be found. Please check the spelling and try again, or click "Register" to set up your account. For help with forgotten Usernames or passwords, click the "Forgot Username or Password?".
[11:03:20] <LLcoolM> i have an account
[11:03:30] <LLcoolM> but i do not get the chance to enter the account data
[11:03:46] <LLcoolM> ok, i could register again, but that wouldn't help i suppose
[11:04:37] <LLcoolM> somehow adblock must have killed the usernam/password field
[11:06:03] <ICU> anyone here running a news service on solaris 10 / zfs?
[11:06:49] <Tempt> checking for AIX... yes
[11:06:56] <Tempt> haha, fun and games ;)
[11:07:34] <LLcoolM> hmm, the download of the dvd is kinda slow. what is the difference between cd and dvd version anyway?
[11:07:53] <LLcoolM> will i be able to download the packages missing on the cd easily?
[11:07:54] <Tempt> One is designed to be burned to a DVD
[11:07:57] <tsoome> from dvd image you can burn.... an dvd
[11:07:58] <Tempt> The other is designed to be burned to CDs
[11:08:07] <LLcoolM> ah there are more than 1 cd
[11:08:11] <tsoome> cool, isnt it? ;)
[11:08:11] <LLcoolM> ok
[11:08:12] <Tempt> Yes
[11:08:17] <Tempt> For people with no DVD drive.
[11:08:30] <LLcoolM> well download is extremly slow
[11:08:41] <Tempt> That's because there's just been a new Solaris release
[11:08:46] <Tempt> and everyone on the planet is suckin' it down
[11:08:49] <LLcoolM> downloading part 1 will take me >24h
[11:08:58] <LLcoolM> isnt there a torrent for that?
[11:09:03] <Tempt> No torrent.
[11:09:06] <LLcoolM> :-(
[11:09:15] <Tempt> Just chill, it'll pick up speed again soon enough
[11:09:18] <LLcoolM> ok, i have to postpone it
[11:09:41] *** yarihm has joined #OpenSolaris
[11:10:21] <LLcoolM> for it wont be finished until my next daily disconnect
[11:10:33] <Tempt> Use Sun Download Manager
[11:10:38] <Tempt> it'll handle the resumes nicely for you.
[11:10:52] <LLcoolM> Tempt: assuming it works on netbsd
[11:10:55] <timsf> Or you could just drop $30 and buy the media
[11:10:56] <Tempt> It's Java
[11:11:01] <Tempt> If you can run Java, you can run SDM
[11:11:22] <Tempt> If you can't run Java, you've .. got plenty of free memory not being wasted by Java apps
[11:11:27] <LLcoolM> does it require x11? because if not i could download it on my linux server
[11:11:35] <Tempt> It's an X app
[11:11:49] <Tempt> I just use Opera anyway
[11:11:54] <Tempt> it supported HTTP resumes just fine
[11:12:06] *** alanbur has joined #opensolaris
[11:13:24] *** ghatak has joined #opensolaris
[11:14:02] <ghatak> Hi, is there a diagnostics CD from sun to find hardware faults on Sun Fire 2100 boxes?
[11:14:37] <asyd> do you have a sun contract support ?
[11:14:44] <ghatak> asyd: i do
[11:14:50] <asyd> if yes, download explorer, run it, and sent the report to sun
[11:15:09] <iMax> you could also run SunVTS
[11:15:17] <LLcoolM> *grrml* the sun download manager comes in rpm for linux
[11:16:15] <ghatak> iMax: that is what i am looking for,
[11:16:18] <ghatak> thanks
[11:17:28] *** KermitTheFragger has joined #opensolaris
[11:20:13] *** ecc_ has quit IRC
[11:21:43] *** infidel has joined #opensolaris
[11:23:54] <LLcoolM> can the download manager download more than just 1 file at once?
[11:24:04] <LLcoolM> because it seems to put them into a queue only
[11:24:13] <ghatak> iMax: would SunVTS work ong 2100 boxes, the doc for support does not mention that
[11:25:00] <iMax> I am not sure, I have only used it on Sparc machines I think
[11:26:12] <iMax> if you just want to test the memory, then I think any x86 based memory tester should do
[11:26:28] *** SnakeUK has joined #opensolaris
[11:26:43] *** SnakeUK has quit IRC
[11:26:47] <ghatak> iMax: I wanted to run diagnostics as the box crashed three times in two days, with nothing in messages
[11:26:53] *** SnakeUK has joined #opensolaris
[11:27:13] *** SnakeUK has quit IRC
[11:27:32] <LLcoolM> ouch, now it stopped downloading at all
[11:29:55] <iMax> ghatak: google says that apparently 2100 M2 is/was not supported but SunVTS seems to run on it
[11:30:20] <ghatak> iMax: Ok i am going to give it a try,
[11:44:02] *** Fullmoon has quit IRC
[11:45:31] *** Snake007uk has joined #opensolaris
[11:45:32] *** infidel has quit IRC
[11:47:09] *** Gman is now known as GmanaFK
[11:49:52] *** manish has joined #opensolaris
[11:50:04] <ofu> http://www.opensparc.net/cgi-bin/goto.php?w=/pubs/preszo/07/HC19.sphillips.v1.pdf sounds really interesting
[11:50:25] *** manish is now known as monzie
[11:51:31] <Tempt> Alright, time to upgrade the workstation to u4
[11:52:22] *** JWheeler has quit IRC
[11:57:24] *** Dar_HOME is now known as Dar
[12:00:32] *** simford has quit IRC
[12:01:34] *** infidel has joined #opensolaris
[12:02:31] *** Pietro_S has joined #opensolaris
[12:10:29] *** Gekkko has joined #opensolaris
[12:12:10] *** calumb has joined #opensolaris
[12:22:11] *** Fullmoon has joined #opensolaris
[12:28:07] <ghatak> How do i mount a CDROM/DVDROM connected to a USB port ? It is detected by BIOS because i can but a boot disk in it and the system boots from CDROM
[12:28:58] *** ichigo has joined #opensolaris
[12:29:27] <Berny> rmformat -l to find out which device the drive has got
[12:29:41] *** Megaf has joined #OpenSolaris
[12:31:18] <tomww> look also into  /media  maybe it is automounted by rmvolmgr  (on recend Nevada/Opensolaris releases)
[12:31:32] <Tempt> or look for it in /cdrom
[12:31:39] <Tempt> USB CDs get automounted
[12:31:59] <ghatak> running rmformat -l says " No removeable deices found"
[12:32:31] <ghatak> however, running cfgadm shows that USB-STORAGE connected and configure on port USB1/8
[12:32:56] *** KermitTheFragger has quit IRC
[12:33:29] <Tempt> odd
[12:33:33] * Tempt shrugs
[12:35:20] <Berny> svcadm restart hal; svcadm restart rmvolmgr  sometime does the trick
[12:46:41] <WickedWicky> b00
[13:02:58] *** boro has quit IRC
[13:06:39] *** deather_ has joined #opensolaris
[13:07:17] *** jafari has quit IRC
[13:07:24] *** nettie has left #opensolaris
[13:10:34] *** The-spiki has joined #opensolaris
[13:10:44] *** deather has quit IRC
[13:10:46] *** deather_ is now known as deather
[13:15:10] *** Megaf has quit IRC
[13:20:16] *** KermitTheFragger has joined #opensolaris
[13:22:15] <Tempt> WickedWicky: boo
[13:22:33] <Tempt> WickedWicky: upgrading the blade... waaaiiiting.
[13:23:19] <dlg> mmmblade
[13:23:24] <Tempt> So .. much .. waiting .. when .. I .. want .. workstation.
[13:23:40] <Tempt> I need my Blade-1000
[13:23:53] * dlg has none
[13:23:56] <Tempt> I *neeeed* it.
[13:24:21] <Tempt> dlg: You'd run BSD on it! :)
[13:24:45] <dlg> i should get two then
[13:25:07] <Tempt> three
[13:25:14] <dlg> nv?
[13:25:26] <Tempt> One for Solaris, one for OpenSolaris, one for BSD.
[13:26:20] <dlg> would be nice
[13:26:24] * dlg pretty poor though
[13:26:37] <dlg> apparently the m4k is splittable into two domains
[13:26:43] <dlg> one for nv and one for solaris
[13:26:44] <dlg> :D :D
[13:26:45] <Tempt> It is
[13:26:48] *** derchris^ has quit IRC
[13:26:51] <Tempt> the m5k is 4 domains
[13:26:54] <dlg> yeah
[13:27:01] <Tempt> Someone should buy me an m4k
[13:27:04] *** derchris has joined #opensolaris
[13:27:34] <Tempt> Someone should give me an m4k, then I could give dlg my blade.
[13:27:39] <Tempt> It all cycles down the chain
[13:27:48] <trochej> :)
[13:28:59] <Tempt> Perhaps M9000.
[13:29:07] <Tempt> That'd be good enough.
[13:30:28] *** halton has left #opensolaris
[13:32:16] *** kaiwai has joined #opensolaris
[13:32:30] <kaiwai> anyone here from NZ?
[13:32:38] <renihs> m5000 is heavy enough already
[13:33:54] <kaiwai> it could be heavier
[13:34:03] *** alanbur has left #opensolaris
[13:34:11] <renihs> pff
[13:34:13] <renihs> :p
[13:35:43] <kaiwai> if it can be moved without a fork lift, its not a real computer
[13:36:20] <renihs> we call them ants
[13:36:21] <renihs> :p
[13:37:25] <kaiwai> damn I'm getting crap speed from dlc - a huge 7KB/ps
[13:38:00] <renihs> awesome :p
[13:38:05] <renihs> wires must be glowing hot
[13:38:17] *** infidel has quit IRC
[13:38:48] *** perlmonk has joined #opensolaris
[13:38:49] <kaiwai> I know, I've got a cool kit hooked up to them
[13:41:07] *** fbo has joined #opensolaris
[13:41:15] <Tempt> kaiwai: I thought you were a PC guy?
[13:44:08] <PerterB> you?
[13:44:19] <PerterB> oops, wrong window
[13:47:35] *** loke_ has joined #opensolaris
[13:49:08] *** loke has quit IRC
[13:55:33] *** Sup3rkiddo has joined #opensolaris
[14:01:32] *** MattMan is now known as MattAFC
[14:01:40] *** jcea has joined #opensolaris
[14:02:23] <kaiwai> Tempt: I am a kinda-pic guy
[14:02:36] <kaiwai> then again, I've made the mistake once, back to the Mac next time around
[14:04:20] <kaiwai> bugger that for a joke, damn sun has pathetic servers
[14:04:54] <Tempt> highly pathetic
[14:05:07] <Tempt> especially compared to the awesome power of .. the Mac.
[14:05:38] <kaiwai> no, not the server but the dlc server and the mind blowing 5K I'm getting off it
[14:06:46] <Tempt> aah, bandwidth problems
[14:06:51] <Berny> thats not the server i'd say ;-)
[14:07:08] <kaiwai> either that or Xtra internet sucks
[14:07:09] <Tempt> I pulled down studio 12 today at about 900k/sec, so they've obviously got some sort of restriction in place
[14:07:16] <kaiwai> which is highly likely
[14:07:33] <Tempt> I noticed I got much, much better performance grabbing it from a host in the US compared to in .au
[14:07:50] <Tempt> Then again, internet performance has been markedly shabby all-round this week in au
[14:07:54] <holcomb> i either get 50K/s or 2M/s from dlc
[14:08:47] <kaiwai> hmm, I'll try something
[14:08:49] *** kaiwai has quit IRC
[14:09:13] <Tempt> Excellent.
[14:10:30] <Berny> hehe
[14:11:57] <madhatter> Has anybody an url to documentation about upgrading SXCE with two identical sized slices? Somebody mentioned such a method to me few days ago, but I'd like to read more about it.
[14:12:16] <Tempt> No matter how hard anyone whines about slow downloads, they aren't going to improve until the current demand abates.
[14:12:59] <Tempt> ooh, blade-1000 rebooting ...
[14:13:02] *** master_of_master has joined #opensolaris
[14:13:41] *** postwait has joined #opensolaris
[14:15:47] <Berny> madhatter: jmcp had some hint... i think there was a blog about that... though i can't remember where
[14:15:57] <jmcp> Berny: huh?
[14:16:08] <jmcp> live-upgrade, perhaps?
[14:16:19] <Berny> jmcp: that issue with identical disks and lu?
[14:16:31] <Berny> biosdev not giving anything usefull
[14:16:33] *** alfred has left #opensolaris
[14:16:43] <jmcp> oh, right
[14:16:49] *** kaiwai has joined #opensolaris
[14:16:53] <kaiwai> hmm, interesting
[14:17:03] <madhatter> If there are better solutions I'd go for those then ;)
[14:17:04] <jmcp> run /sbin/biosdev -d, save the output, then write a shell script to emit the 0x8? /path/to/device info
[14:18:02] <Berny> is there a fix for biosdev on the way by  now?
[14:18:08] *** mega has joined #opensolaris
[14:18:11] <jmcp> dunno
[14:18:15] <Berny> that issue must annoy a lot of people ;-)
[14:18:22] <kaiwai> is buisdev broken?
[14:18:29] <kaiwai> *biosdev
[14:18:46] <Berny> been broken for ages... least if you have identical disks
[14:19:12] <madhatter> jmcp: What about live-upgrade? I saw an application here but it froze when I started it
[14:19:40] <jmcp> madhatter: what are you asking?
[14:20:31] <madhatter> jmcp: If there really is something like a comfortable live-upgrade? I have been wondering because you mentioned that name. But when trying that here it did not work
[14:21:35] <jmcp> it's worked for me
[14:21:43] <jmcp> once I figured out the /sbin/biosdev issue
[14:21:53] <hile_> jmcp - speaking of LU.. is my bug fixed yet?
[14:22:00] <jmcp> probably not :(
[14:22:07] <Tempt> Bah, coherent-console doesn't seem to work
[14:22:16] <kaiwai> yeap, sun is limiting the download speed
[14:22:22] <madhatter> jmcp: Any details on that issue?
[14:22:23] <hile_> what's the bugID again? I forget?
[14:23:29] <jmcp> madhatter: on which issue?
[14:23:49] <madhatter> jmcp: < jmcp> once I figured out the /sbin/biosdev issue
[14:23:52] <jmcp> right
[14:23:56] <jmcp> .... let me check sunsolve
[14:24:05] *** Gekkko has quit IRC
[14:25:08] <jmcp> 6536905 biosdev 1.4,1.5 changes render SATA disks under old framework invisible to LU
[14:25:12] *** master_o1_master has quit IRC
[14:25:25] *** JWheeler has joined #opensolaris
[14:25:40] <jmcp> which is not visible in b.o.o. because it's targeted against the bios for the u20m2, rather than biosdev
[14:26:27] <madhatter> Hm, I will dive into that
[14:26:50] <Tempt> wha?
[14:26:58] <Tempt> U20 bios updates being rolled into Solaris?
[14:27:31] *** spoown has joined #opensolaris
[14:27:55] <kaiwai> are microcode updates going to get included with solaris?
[14:27:57] <jmcp> Tempt: no
[14:28:20] *** postwait has quit IRC
[14:28:25] <jmcp> Tempt: /sbin/biosdev operates correctly, as far as we can see, it's the bios in my u20m2 which doesn't behave properly
[14:29:54] <Berny> jmcp: add the x2200m2 to the list...
[14:30:09] <Berny> i see that issue on that two machines as well
[14:30:54] *** myrkr has joined #opensolaris
[14:31:31] <jmcp> Berny: you should log a separate bug against the bios in those systems - different teams handle their bioses. reference 6536905 when you log it
[14:32:06] <Berny> log via boo or via my spectrum account?
[14:34:01] <hile_> what's the bugid on my LU bug, jmcp?
[14:35:35] <jmcp> hile_: can't recall offhand
[14:35:39] <jmcp> need to search through my logs
[14:35:43] <jmcp> Berny: spectrum
[14:36:34] <Berny> .oO(guess that will run in a dead end, since i run nevada on those machines... but i'll give it a shot)
[14:38:34] <Berny> is there a way to get the serial number of the box remotely? (i.e. via software)
[14:40:25] <Berny> ha smbios spits something out
[14:40:42] <jmcp> perhaps ipmitool, on the sp
[14:41:20] <PerterB> it also affects some non-Sun biosses, which are going to be a lot harder to get fixed
[14:41:53] <jmcp> yeah
[14:42:44] <Berny> oh my... osc is crawling again
[14:43:00] <kaiwai> hmm, smbios seems to work for me
[14:47:17] *** nachox has joined #opensolaris
[14:48:09] *** linux_user400354 has quit IRC
[14:48:15] <nachox> sun's servers are suffering,  s10u4 downloads must be killing them
[14:50:04] <Berny> right got a ticket on that one...
[14:50:34] <Gekkko[PDA]> bittorrent should be implemented.;
[14:50:39] <Pietro_S> where pidgin(gaim) store it's configuration? I tried to delete ~/.gaim and ~/.pidgin but it still ask me on passwords for my accounts :-(
[14:50:40] *** DoYouKno has joined #opensolaris
[14:51:16] <asyd> PerterB: probably in ~/.gnome2 or ~/.gnome2_private
[14:53:54] <jmcp> I think it's in ~/.purple, actually
[14:53:56] <jmcp> nfi why
[14:54:54] <timeless> heh
[14:55:03] <timeless> i know a pidgin dev, i could probably ask :)
[14:55:27] <Gekkko[PDA]> purple is the library
[14:55:30] <Gekkko[PDA]> libprpl
[14:55:33] <timeless> anyway, i hope things are going well for everyone, i'm about to start my summer vacation travelling through europe :)
[14:55:51] *** sstallion has quit IRC
[14:56:20] *** sstallion has joined #opensolaris
[14:57:10] *** Umarzuki has joined #opensolaris
[14:57:17] <nachox> just dtrace it or truss it, look what files it opens
[14:57:30] <Umarzuki> hullo
[14:57:46] <Umarzuki> hello everyone
[14:58:08] <Pietro_S> not found in .gnome2 nor .gnome2_private, but I will delete then to be sure ... nothing can happend when I delete them completely when I use xfce, right?
[14:58:09] <nachox> downloading at a steady 20k/s :)
[14:58:42] <Umarzuki> i'm thinking of installing opensolaris..
[14:59:38] <Umarzuki> tell me why it's better than ubuntu?
[15:00:24] <nachox> i have a better idea, install an opensolaris distribution, use it for a few months and then you tell us why it is better
[15:00:56] <Umarzuki> aha..good point
[15:01:06] <Umarzuki> but the time it took to downlod it
[15:01:41] <kaiwai> nachox: thats nothing, I'm downloading at an awesomely mind blowing 7k/s
[15:02:00] <nachox> if you feel confortable with ubuntu then you will probably like nexenta, try it inside a vmware
[15:02:33] *** cypromis has joined #opensolaris
[15:02:33] <Umarzuki> ok..
[15:03:36] <nachox> kaiwai: downloading both segments at the same time makes it better?
[15:03:48] *** toblun has quit IRC
[15:03:49] <nachox> or you get the same agregate?
[15:03:56] <kaiwai> nachox: I'm using that gawd aweful SDM
[15:04:05] <nachox> you suck
[15:04:06] <kaiwai> 7k/s constant speed
[15:04:25] <Gekkko[PDA]> lol nachox
[15:04:26] <kaiwai> I know I do, I don't hear any guys complain yet :)
[15:04:27] <nachox> i learnt my lesson and now use wget
[15:04:39] <Gekkko[PDA]> no personal attacks.
[15:05:28] *** toblun has joined #opensolaris
[15:05:39] *** DoYouKnow has quit IRC
[15:05:43] *** postwait has joined #opensolaris
[15:09:22] *** pjlv has quit IRC
[15:10:57] <Pietro_S> ~/.purple was the right directory ...
[15:13:17] *** calumb has quit IRC
[15:14:37] <Pietro_S> hmm, but problem wasn't solved - it still doesn't remember my passwords
[15:14:45] *** Umarzuki_ has joined #opensolaris
[15:15:39] <Pietro_S> I hope that saving passwords works for someone ...
[15:15:43] <timsf> works for me -
[15:15:51] <timsf> is gnome-keyring-daemon running ?
[15:16:09] <timsf> gnome-keyring-manager seems to be what pidgin uses to remember passwords.
[15:16:12] *** DoYouKno has quit IRC
[15:16:47] <Umarzuki_> opop
[15:18:46] *** cypromis_ has quit IRC
[15:19:01] *** DoYouKno has joined #opensolaris
[15:22:56] <Pietro_S> timsf: does it have svc(smf) manifest on your machine, or is it started other method?
[15:23:46] *** spoown has quit IRC
[15:24:03] <Pietro_S> I have such running process ...
[15:24:03] <Cyrille> I don't think it's started through SMF.
[15:24:12] *** Umarzuki has quit IRC
[15:24:22] <Cyrille> It might be from the default gnome-session startups
[15:25:04] *** calumb has joined #opensolaris
[15:25:30] <Pietro_S> question is how to check if it works ...
[15:26:58] <timsf> just run gnome-keyring-manager
[15:27:26] <timsf> (right, it's not started via SMF - it seems to be a member of the contract Xorg is started in, ctstat -av)
[15:29:57] <timsf> (I should say, I'm running snv_68 here - a bit old...)
[15:31:09] * kaiwai hopes phase 1 of nwam is merged with gnome 2.20
[15:32:46] *** laca has joined #opensolaris
[15:33:28] *** Umarzuki_ has quit IRC
[15:38:13] <Pietro_S> hmm, when I logged to JDS, it worked, but in xfce not :-(
[15:40:17] <Pietro_S> I even can't run gnome-keyring-manager fom xfce ...
[15:40:54] <laca> is xfce started with dbus-launch?
[15:40:55] *** pfa3rh has joined #opensolaris
[15:42:49] <Pietro_S> don't know, how can I inspect it?
[15:44:40] <laca> you should see something like /usr/bin/dbus-launch --exit-with-session /usr/bin/xfce4-session in your process list
[15:44:53] *** ghatak has left #opensolaris
[15:45:13] <laca> and the DBUS_SESSION_BUS_ADDRESS env variable should be set in your shells
[15:46:17] <Pietro_S> laca: that pkgmk core dump, is fixed in sxce70, it will write error message correctly only pkgbuild/pkgtool don't put that message to it's log ... (as we spoke about it yesterday)
[15:46:45] <laca> so what's the error message?
[15:47:38] <Pietro_S> don't remember exact, but it cause file with " " in path ...
[15:48:57] *** calumb is now known as calLNCH
[15:49:15] <WickedWicky> when I do vi /etc/shadow and i get the rmmount.conf displayed
[15:49:19] <WickedWicky> that aint good... is it...?
[15:50:57] *** millhouse has joined #opensolaris
[15:51:31] <timsf> no, definitely not. what's the bets you're thinking about fsck at the moment?
[15:52:37] <Pietro_S> dbus looks like working, that process is there (same arguments) and evn variable is set to unix:path=/tmp/dbus-1fYp13DV6J,guid=e19e63947bb760b01a5de10046deb0a9
[15:52:38] <kaiwai> neat, the cap is gone
[15:53:43] <kaiwai> nope, there it is again
[15:54:32] <WickedWicky> timsf: this is after an fsck
[15:54:43] * timsf looks nervous
[15:55:00] <WickedWicky> the machine went down during a power failure
[15:55:08] <WickedWicky> came up, didnt find its metadbs
[15:55:27] <nachox> kaiwai: still 7k/s?
[15:55:31] <WickedWicky> someone fscked /dev/rdsk/c1t0d0s0 , mounted as /
[15:55:38] <WickedWicky> rebooted
[15:55:39] <timsf> ow
[15:55:40] <WickedWicky> and poof
[15:55:43] <nachox> ouch
[15:55:56] <WickedWicky> so I guess the inode structure is blown to pieces
[15:56:03] <kaiwai> nachox: nope, 30K/s
[15:56:06] <kaiwai> still shit though
[15:56:16] <kaiwai> why not a mirror in the pacific?
[15:56:17] <nachox> better than my 20
[15:56:43] <kaiwai> geeze, I can download porn faster than that
[15:56:54] <WickedWicky> the coolest thing is: nobody knows what the machine was suposed to be running, application wise
[15:57:03] <WickedWicky> so they'll bring in the server tomorrow and we'll have a look
[15:57:06] <Pietro_S> well, can be my xfce broken, cause of lu? Should I rebuild it from spec files?
[15:57:11] <millhouse> for ZFS, I've got some errors on a zpool.  When I run 'zpool status -v' I get a list of data errors.  Can I translate that into actual files?
[15:57:14] <millhouse> and if so, how
[15:57:25] <laca> Pietro_S: what exactly is broken?
[15:58:53] <nachox> kaiwai: of course, p0rn is distributed world wide, solaris isnt
[15:58:56] <timsf> millhouse, yes - on recent versions of nevada it tells you what files are broken
[15:58:56] <nachox> :P
[15:59:36] <timsf> otherwise, zdb -vvv <pool>  > output.txt will tell you more than you want to know abuot your pool structure
[15:59:45] <timsf> or is it -vvvvvv, I forget.
[15:59:50] <nachox> WickedWicky: is there anything in /lost+found?
[16:00:20] <timsf> (assuming the problems are in actual files, rather than other bits of zfs metadata)
[16:00:48] <Pietro_S> laca: gnome-keyring-daemon, pidgin doesn't remember my passwords (in JDS it works)
[16:00:49] <WickedWicky> cant look right now, they shut down the machine for shipment to here
[16:01:32] <nachox> WickedWicky: see? ruined all the fun....
[16:01:48] <WickedWicky> we can have fun tomorrow
[16:01:53] <WickedWicky> the machine will be under my desk :)
[16:03:32] <Pietro_S> laca: also gnome-keyring-manager doesn't connect to keyring daemon, even it's running
[16:04:34] *** jafari has joined #opensolaris
[16:05:36] <millhouse> timsf:  This box is running solaris 10u3 (version 3 of zfs)
[16:05:45] <millhouse> I have some opensolaris b68 machines too (ver 7 of zfs)
[16:07:17] *** calLNCH has quit IRC
[16:07:45] *** lisppaste3 has joined #opensolaris
[16:10:47] <timsf> if you take the disks from the s10 box and pop them into the b68 box, it might tell you more information, but you should still be able to get what you need from the zdb output.
[16:11:12] <timsf> (even on s10)
[16:13:47] *** alanc_away has joined #opensolaris
[16:21:46] <kaiwai> anyone running zfs for their boot drive?
[16:22:02] <millhouse> well, i ran zdb -vvv <pool> and it says that the pool is in a "consistent" state.  ZFS resilvered the volume and says that it looks "good" but had (31) data errors
[16:22:29] <millhouse> it's a backup/test box anyway, so i might just reinstall the system with OpenSolaris b68 and test--this is an iscsi setup.
[16:23:09] <millhouse> but on b68, doing 'zfs status -v' would give me the list of actual files that are bad?
[16:24:00] <kaiwai> hmm, AMD has released ATI drivers for Linux - i hope they go bankrupt before they gain any friends
[16:24:20] <millhouse> kaiwai:  why do you say that?
[16:24:41] <kaiwai> millhouse: screwing end users over for many years for instance
[16:24:57] <kaiwai> crap windows drivers, crappier linux ones and none for solaris or freebsd
[16:26:26] <elektronkind> dear god. s10u4 downloads are creeping along. Anyone know of a torrent for it?
[16:26:48] <millhouse> kaiwai:  Ah..  Well as far as video drivers go, I've been an NVidia fan for years..  As far as CPU's go, I've liked AMD CPU's vs Intels'
[16:27:02] <millhouse> elektronkind:  when was s10u4 released?
[16:27:04] * elektronkind beats "Fetch" the network dog with a newspaper
[16:27:08] <elektronkind> millhouse: yesterday
[16:27:22] <elektronkind> you can get it from sun.com/solaris
[16:27:50] <Pietro_S> wait few dayes and you will enjoy fast speed again ;-)0
[16:27:52] <ofu> nvidia still has no freebsd/64 driver, is that better than ati stuff?
[16:27:59] <kaiwai> millhouse: AMD and ATI are now one - if AMD can't get their act together, both their GPU and CPU business need to be punished; for me, Nvidia all the way
[16:28:57] <millhouse> yeah...  I think AMD should have bought NVidia instead of ATI...  we'll see..  I would like to see their CPU's get back in the game and rival Intel a lot more...if for nothing else than healthy competition
[16:30:12] <kaiwai> Or Matrox
[16:30:16] *** xOmega has joined #opensolaris
[16:30:18] *** hile_ has quit IRC
[16:30:30] <kaiwai> Matrox had some nice gpu's at one point; G550 was nice with their opensource drivers
[16:30:49] *** MattAFC is now known as MattMan
[16:31:03] <nachox> only nvidia probably costs as much as amd does
[16:32:02] <kaiwai> it would have to be a merger of mutual benefit for it to work
[16:33:43] <kaiwai> given the current financial state
[16:33:49] <kaiwai> then again, I call that karma
[16:33:56] <kaiwai> fuck users and get fucked yourself
[16:36:33] <millhouse> elektronkind:  Thanks!  I'm downloading the x86 dvd now :)
[16:38:01] <timsf> Hey millhouse - sorry I didn't get back to you
[16:38:19] <timsf> http://opensolaris.pastebin.ca/682356 - that's from snv_69
[16:38:39] <timsf> it shows zpool status showing a file that's broken in a pool.
[16:38:43] <nachox> AMD has to pull out one good product to get over this, it is all it takes... in the short run
[16:38:58] <nachox> look what apple did with the ipod
[16:39:07] <timsf> Beneath is the zdb -vvv output that allows you to map from object ids to filenames (if you're running an earlier version of zfs)
[16:39:08] <elektronkind> millhouse: how fast is the dvd download going for you?
[16:39:20] <millhouse> timsf:  no worries..  I'm at work myself, so my attention gets diverted ;)
[16:39:26] <kaiwai> nachox: one difference, at the time of the ipod they were making a profit; AMD is losing money hand over fist
[16:39:33] *** hile_ has joined #opensolaris
[16:39:43] <millhouse> elektronkind:  at this rate, ~20 hours to go...
[16:39:51] <gdamore> good morning *
[16:39:54] <elektronkind> ugh
[16:40:04] <elektronkind> I'm ripping along at 7K/s
[16:40:04] <kaiwai> I'd purchase an AMD laptop but their ATI support is crap, their wireless is shit and their chipset is as dodgy as a whore on K road
[16:41:52] <kaiwai> you end up stuck with a broadcom wireless made by a company of assholes unwilling to provide specifications to the opensource community
[16:44:09] *** karrotx has quit IRC
[16:47:38] <trs81> ati's about to release decent linux drivers, I wonder if they'll get to work on solaris drivers next?
[16:48:34] <millhouse> elektronkind:  what were the major changes between this and u3?
[16:48:45] <kaiwai> I doubt it, IIRC Alan has tried to get Ati to give him some specs so far ati has refused
[16:48:54] <elektronkind> millhouse: http://docs.sun.com/app/docs/doc/817-0547/getjd?l=en&a=view
[16:49:06] <elektronkind> that's the "What's new" doc for it
[16:49:21] <kaiwai> trs81: I'd like to see more competition but ultimately AMD/ATI is killing themselve slowly with their arrogance
[16:50:52] <millhouse> elektronkind:  Thanks!
[16:53:16] <trs81> millhouse: s10u4 gives the filenames in zpool status -v - http://docs.sun.com/app/docs/doc/819-5461/6n7ht6qsq?l=en&a=view
[16:54:29] <timsf> thanks trs81 - I lose track sometimes (you'd think I should know this stuff!)
[16:54:31] <millhouse> trs81:  awsome, that's what I was looking for specifically (zfs improvements).
[16:55:02] <ofu> filenames?
[16:55:39] <timsf> ofu yeah, a list of what files are affected by any errors in a zpool.
[16:55:52] <ofu> ah, thats cool
[16:56:04] <elektronkind> before it would only give you the file's inode
[16:56:19] <timsf> kindof.
[16:56:19] <ofu> what about overflows if many files are affected?
[16:56:41] <timsf> why shuold it overflow if it's coded properly?
[16:57:20] <elektronkind> the only thing such a long list would overflow is my patience and hope ;)
[16:58:59] <millhouse> trs81:  do you know what version of zfs this has?
[16:59:48] <kaiwai> is it possible to adjust the brightness of the screen in solaris?
[17:00:05] *** stevel has joined #opensolaris
[17:00:09] *** ChanServ sets mode: +o stevel
[17:05:57] *** manish_ has joined #opensolaris
[17:06:44] <trs81> millhouse: no idea sorry, I'd guess 6 or so, but http://www.opensolaris.org/os/community/zfs/version/6/ (and other numbers) only lists support for nevada and update 3
[17:08:31] <timsf> remember those version numbers are really just on-disk format changes
[17:08:49] <timsf> so you could have a perf bugfix that affects ZFS, but doesn't change the on-disk format.
[17:15:09] <DoYouKno> this ZFS stuff is pretty awesome
[17:15:30] *** pauliukas has joined #opensolaris
[17:15:37] <DoYouKno> how good is broadcom support on solaris with ndiswrapper?
[17:15:37] <pauliukas> Goooood morning
[17:15:52] <DoYouKno> does it crash a lot?
[17:16:08] <pauliukas> http://www.intuitive.com/solaris/
[17:16:18] <pauliukas> Man, they got Solaris 9 for dummies, but no Solaris 10 for dummies.
[17:16:20] <pauliukas> Not fair.
[17:18:22] *** alred has joined #opensolaris
[17:19:06] <timeless> you tend to get much better errors from it
[17:19:34] *** monzie has quit IRC
[17:23:19] <movement> DoYouKno: I never had it crash, but it's extremely flaky
[17:23:42] <movement> pauliukas: watch out, at least one "solaris 10" book is nothing of the sort
[17:23:55] <pauliukas> what?
[17:24:21] *** mp3ters has joined #opensolaris
[17:24:58] <mp3ters> Anybody konw of some good tutorials for configuring ftpd in Sol 10?? The man pages just don't do it for me
[17:25:09] <timeless> oops, wrong window
[17:25:15] *** micken has joined #opensolaris
[17:25:17] <elektronkind> mp3ters: I just install and use proftpd
[17:25:26] <elektronkind> it's far more flexible
[17:25:30] <mp3ters> yea?
[17:25:50] <mp3ters> GUI?
[17:25:57] <elektronkind> no, no gui
[17:26:01] <mp3ters> even if its cli thats fine but just need a tutorial
[17:26:19] <elektronkind> proftpd has good documentation
[17:26:31] <mp3ters> hmmm
[17:26:36] <DoYouKno> movement, on linux, my broadcom only works when I'm not around other windows laptops
[17:26:40] <DoYouKno> err
[17:26:43] <DoYouKno> yeah
[17:26:48] <DoYouKno> when I'm not around other windows laptops
[17:27:06] <DoYouKno> something with the TX power
[17:27:14] <Pietro_S> wasn't there any change in packaging gtk-2.0? xfce failed to build because gtk+-2.0 isn't found ...
[17:27:16] <mp3ters> 1.3.1rc3  ?
[17:27:19] <micken> whoa
[17:27:25] <DoYouKno> movement, that's with the native broadcom driver
[17:29:49] <pauliukas> Anyone know a great vi tutorial? Or some get started thing?
[17:30:23] <mp3ters> oreilly has a good one, i'm gonna buy that book
[17:30:26] <mp3ters> its like 20bucks
[17:30:37] <mp3ters> has a lil weizel on the cover
[17:30:44] *** tsoome has quit IRC
[17:30:50] *** charlesnw has joined #opensolaris
[17:31:08] <FBdev> mp3ters vimtutor is a cool command
[17:31:24] <pauliukas> oh, cool.
[17:31:36] <pauliukas> A whole book for vi? Can I say overkill?
[17:31:44] <mp3ters> no its a lil paper back
[17:31:49] <mp3ters> its not intimidating at all
[17:31:54] <mp3ters> and its rich
[17:32:07] <FBdev> pauliukas vimtutor will show you the basics
[17:33:29] <pschow> mp3ters:  how about docs.sun.com, http://docs.sun.com/app/docs/doc/816-4555/wuftp-17?l=en&q=ftpd+solaris+10&a=view
[17:34:12] <mp3ters> no shit...
[17:34:15] <mp3ters> nice
[17:34:24] *** nachox has quit IRC
[17:34:50] <mp3ters> prior, i was only finding man pages for each file in etc/ftpd
[17:34:54] <mp3ters> thx
[17:35:39] <movement> DoYouKno: it's simplest to just get one with an atheros chipset.
[17:35:45] <movement> they're not expensive...
[17:36:46] <millhouse> trs81:  sorry i didn't respond earlier.  thanks for the info
[17:36:58] <millhouse> DoYouKno:  Yeah, I love ZFS!
[17:37:16] <Pietro_S> hmm, it looks like pkg-config don't like me, what outpuut gives you this? pkg-config --print-errors gtk+-2.0
[17:41:45] <LLcoolM> millhouse: zfs is one of the best features solaris has to offer
[17:42:04] <laca> Pietro_S: you need to create x11.pc manually
[17:42:16] <LLcoolM> linux is better with hardware support, bsd is better for routing and solaris has zfs :)
[17:42:56] <laca> Pietro_S: http://forum.java.sun.com/thread.jspa?threadID=5204312&tstart=0
[17:43:02] <LLcoolM> what keeps me from installing solaris on real hardware is mostly the lack of disc encryption
[17:43:36] <millhouse> LLcoolM:  Yeah, it's why we decided to go with Solaris/OpenSolaris for our iSCSI storage system
[17:44:00] <Pietro_S> laca: thanks
[17:45:15] *** mega has quit IRC
[17:45:39] <LLcoolM> and btw what keeps me from installing solaris on a vm is it's poor download speed
[17:46:07] <millhouse> yeah, hardware support has been somewhat of an issue, but at the same time, we're building iscsi/storage setup, so the hardware doesn't need to be all that fancy (i.e. usb, firewire, etc.)
[17:47:01] *** tsoome has joined #opensolaris
[17:47:51] <LLcoolM> my home-server's hardware is also supported by solaris, but i do not have time to migrate and solaris lacks disc-encryption, so i'll have to wait.
[17:48:28] <LLcoolM> btw, does solaris support isdn?
[17:48:50] <LLcoolM> because i also have a fax (with an active avm isa card)
[17:48:53] *** Sup3rkiddo has quit IRC
[17:49:48] <millhouse> we're trying to move away from windows as much as we can (though much of our apps are dependent on it!).  Solaris/OpenSolaris -> iSCSI/ZFS "storage"; Linux (ubuntu) for most other "general" servers (ftp, etc.)
[17:50:22] <LLcoolM> millhouse: i hate ubuntu, too much broken packages.
[17:50:26] <millhouse> i'd *love* to deploy a bsd router/firewall/packet-shaper but my boss doesn't want too many linux/unix systems floating around....:(
[17:50:54] <LLcoolM> millhouse: what do you use as router instead?
[17:50:58] <LLcoolM> windows?
[17:51:04] <LLcoolM> because that would be hell of a mistake
[17:51:18] <pauliukas> Yeah...
[17:51:22] <pauliukas> Windows NT router.
[17:51:31] <pauliukas> Don't you love a network that crashes every 10 packets?
[17:51:36] <millhouse> LLcoolM:  I've become a fan of it over time (mostly as a desktop distro--we're looking at it possibly replacing windowsXP/vista on most desktops).  As a server, it's not bad...  I'd love to have gentoo or something else, but ubuntu as a server gives us (my boss mainly) more "uniformity"
[17:51:36] <LLcoolM> haha, that was a good one :)
[17:51:51] <millhouse> i.e. no gentoo for this, bsd for that, solaris here, debian there, and redhat over yonder ;-)
[17:52:08] <millhouse> we have a "FireBox" (linux) router/firewall.
[17:52:23] <LLcoolM> millhouse: well, better than windows, thats for sure
[17:52:35] <millhouse> it's pretty secure...  but...honestly, I want to use bsd 1/2 so i can find a good reason to play with it
[17:52:38] <millhouse> oh yeah!!!
[17:52:57] <millhouse> it has a lot of overhead though--it inspects EVERY network packet and right now the rules are very rigid
[17:53:01] <pauliukas> So now that we have some Solaris experts in here...
[17:53:08] <LLcoolM> millhouse: linux is good as a router/firewall, as long as you do not employ 10000+ iptables rules.
[17:53:09] <pauliukas> Can anyone recommend a book or resource for a total newbie?
[17:53:27] <LLcoolM> pauliukas: i am a solaris newbie :)
[17:53:27] *** KermitTheFragger has quit IRC
[17:53:48] <pauliukas> LLcoolM: Problem is... I don't want to idle the IRC all day. I want to actually use Solaris.
[17:53:53] <LLcoolM> i have been using solaris for some years but never administered it
[17:53:56] <pauliukas> Instead of just telling everyone how good it is :-P
[17:53:57] <onga> a solaris newbie or a UNIX newbie all together?
[17:54:12] <pauliukas> onga: I use Linux and am a Linux sysadmin
[17:54:25] <pauliukas> So mostly the differences in Solaris or anything to help.
[17:54:28] <millhouse> i just want bsd...
[17:54:30] <millhouse> heh
[17:54:32] <LLcoolM> onga: only solaris newbie. but i have used it for years at the university (on those good old sun rays)
[17:54:35] <onga> I'm sorry to hear that :)
[17:54:58] <onga> jokes aside... there's a good book, which is actually solaris 9 based
[17:55:10] <LLcoolM> onga: link?
[17:55:13] <onga> for sysadmins... "Solaris solutions for system administrators"
[17:55:14] <pauliukas> I've heard there's Solaris 9 for dummies
[17:55:59] <pauliukas> Where is vimtutor located, btw?
[17:56:06] <onga> which covers a lot of different problems, but from a sol 9 perspective... alot has chnaged since then.. but a lot has stayed the same :).. its a good start before you start to hook into some of the new cool stuff in 10
[17:56:09] <LLcoolM> pauliukas: i'd like to have solaris and gnu tools, but nexenta is yet to much unstable
[17:56:46] <onga> other to that, bigadmin has a reasonable amount of good articles
[17:56:51] <onga> to tackle specific tasks
[17:57:43] <onga> http://sun.com/bigadmin/ and http://www.amazon.com/SolarisTM-Solutions-System-Administrators-Time-Saving/dp/0471348104
[17:58:12] <LLcoolM> ah and a friend of mine did solaris administration and he liked solaris for its good "on heavy load"-behavior, not realizing, that this was only because of the sparc architecture and not because of solaris mostly. he was very disappointed when he installed solaris x86. but he still keeps it because of zfs - so zfs must be quite good :)
[17:58:56] <LeftWing> LLcoolM: That sounds like an exceptionally subjective appraisal.
[17:59:24] <LLcoolM> LeftWing: well thats what he told me. i never cared about load when using solaris :)
[17:59:28] <onga> umm no... Solaris is good for heavy workloads because it is well designed, and has a stack of work, testing and what not for code to be commited
[17:59:43] <onga> not because of SPARC
[17:59:54] <onga> driver support on x86 till 10 was poor at best, sure
[18:00:09] <onga> but that has nothing to do with core solaris design or engineering.
[18:02:34] *** JSRJ has joined #opensolaris
[18:02:49] *** Atomic-Punk has quit IRC
[18:04:28] *** _bernie has joined #opensolaris
[18:06:53] *** Sup3rkiddo has joined #opensolaris
[18:07:43] *** hile_ has quit IRC
[18:07:54] *** Fullmoon has quit IRC
[18:08:13] *** Atomic-Punk has joined #opensolaris
[18:10:50] *** Fullmoon has joined #opensolaris
[18:11:15] <pauliukas> Did anyone every try that: http://www.sun.com/tryandbuy/ ?
[18:11:22] <onga> yeah
[18:11:24] <onga> we do it all the time
[18:11:38] <pauliukas> I'm betting there's some huge protection fees or something
[18:11:45] <onga> nah
[18:11:49] <pauliukas> really?
[18:11:57] <onga> you get the hardware for up to 60 days
[18:12:02] <onga> takes about 6 weeks to get it
[18:12:07] <pauliukas> ah
[18:12:10] <onga> if you want it
[18:12:13] <onga> you pay the invoice
[18:12:14] <onga> if not
[18:12:15] <onga> send it back
[18:12:30] <pauliukas> I probably wouldn't be approved anyways. I'm just a random dude.
[18:12:37] <onga> heh
[18:12:40] <pauliukas> Would be neat to be able to play with an actual Sun machine.
[18:12:41] <onga> who knows.
[18:12:51] *** Triskelios has quit IRC
[18:12:56] <pauliukas> When buying servers, I was looking at what Sun had. Their prices are, sadly, very hgih.
[18:13:03] <onga> do you work for a big company that could get one for you?
[18:13:10] <pauliukas> I work for just myslef
[18:13:15] <pauliukas> *myself
[18:13:23] <onga> their AMD kit isn't too bad... in fact for what you get in T1.. cheap as chips
[18:13:35] <onga> T2 will be... even better.. very interested in what the price point will be
[18:13:45] <onga> well there's always ebay :)
[18:13:46] <pauliukas> Yeah, but the new Intel Xeons kick T1's ass.
[18:14:01] <onga> buy some older SPARC kit if you want SPARC... Ultra 10's are usually cheap
[18:14:24] *** ICU has quit IRC
[18:14:27] <onga> not that fast.. but cheap... get one with enough cache on the cPU and they're not too bad for $50 out of pocket if you're learning sun stuff
[18:14:40] <onga> new intel xeons kick T1's ass?
[18:14:41] <onga> heh
[18:14:43] <onga> are you serious?
[18:14:44] <onga> :P
[18:14:57] <pauliukas> There were benchmarks in the past, too.
[18:14:58] <onga> what intel seon has 32 hardware threads?
[18:15:04] <pauliukas> I remember that when they started the try and buy thing.
[18:15:05] <onga> *xeon
[18:15:08] *** MattMan has quit IRC
[18:15:14] <onga> heh
[18:15:15] *** Fullmoon has quit IRC
[18:15:18] <pauliukas> They said that if a blog posted benchmarks of the T1, they could keep a server for free.
[18:15:20] <pauliukas> So a few did.
[18:15:37] <pauliukas> And the benchmarks showed that the T1 was more expensive and could handle less web connections.
[18:15:44] <onga> the T1's are designed for massive, threaded workloads
[18:16:03] <onga> a xeon will never keep up with them on highly threaded tasks
[18:16:04] <pauliukas> Well, I'm assuming. But they were pushing the web serving purpose back then.
[18:16:05] <onga> single threaded
[18:16:06] <onga> sure
[18:16:14] <pauliukas> Still...
[18:16:15] <seanmcg> pauliukas: uh ? less web connections that what ?  show us the data :)
[18:16:28] <pauliukas> If you get 2 x Clovertown you get 8 really fast cores.
[18:16:36] <onga> a single threaded application would probably run faster on a xeon
[18:16:37] <onga> but other to that
[18:16:47] <pauliukas> seanmcg: Was a long time ago. Lost the link. Sorry.
[18:16:51] <onga> Xeon's chip design is bollocks
[18:17:01] <pauliukas> X86 isn't pretty... But it works.
[18:17:01] <onga> it's 2 dual cores banged onto one chip
[18:17:20] <onga> heh
[18:17:29] <onga> T1 doesn't work?
[18:17:31] *** nachox has joined #opensolaris
[18:17:39] <pauliukas> lol
[18:17:40] *** slowhog has joined #opensolaris
[18:18:08] <onga> T1's are the fastest machines we've put into production
[18:18:12] <onga> they twiddle their thumbs alkl day
[18:18:17] <onga> even under heavy loads
[18:18:23] <onga> sure, you don't want to do floating point on them
[18:18:31] <pauliukas> So... From what I understand...
[18:18:37] <onga> but for application server, or web server loads, they are insanely great
[18:18:37] <pauliukas> They're 6 cores 1.6GHz?
[18:18:42] <onga> 8 core
[18:18:44] <onga> 32 threads
[18:18:46] <pauliukas> ah...
[18:18:48] <onga> 4 threads per core
[18:19:01] <pauliukas> Threads? Does Xeon even have that?
[18:19:01] <onga> running at 1.0ghz or 1.2ghz
[18:19:07] <onga> no
[18:19:10] <pauliukas> lol
[18:19:10] <onga> not like that
[18:19:14] <pauliukas> Okay. T1's rock.
[18:19:17] <LLcoolM> is the good old intel bx chipset supported (including the intel IDE)?
[18:19:18] *** sfire||mouse has joined #opensolaris
[18:19:33] <onga> in a xeon "quad core" (that is 2 dual core chips bange don one cause they banged it out the door quick), you get 4 threads
[18:19:34] <onga> at a time
[18:19:37] <onga> T1 runs 32
[18:19:40] <onga> T2 runs 64
[18:19:52] <pauliukas> geezz...
[18:19:59] <onga> on one chip
[18:20:04] <pauliukas> That's fucking (sorry my language) crazy.
[18:20:08] <onga> yes
[18:20:10] *** estibi has joined #opensolaris
[18:20:16] <pauliukas> Now, tell me.
[18:20:17] *** Erwann has quit IRC
[18:20:21] <onga> so I think you may of misread your xeon statement on benchmarks.
[18:20:26] *** RaD|Tz has joined #opensolaris
[18:20:29] <pauliukas> Why are all these idiots out there using Linux with Intel processors?
[18:20:34] <onga> especially at the time you refer to, which I gather dual core wa the latest thing on intel
[18:20:42] <onga> while Sun has 32 threads, they had 2
[18:21:01] <onga> btu don't get me wrong, they're not for everything
[18:21:10] <pauliukas> no? :-(
[18:21:12] <onga> floating point math isn't the best, if you do alot of it.. T2 will fix that
[18:21:28] <pauliukas> But that's mostly used for rendering or scientific work, right?
[18:21:36] <pauliukas> But for databases or web serving... Not so much?
[18:21:46] <onga> outside of that, T1 really flys. We support a mail system, with 40,000 users
[18:21:51] <onga> 2 T1 front ends
[18:21:52] <onga> they idle
[18:21:54] <onga> all day
[18:21:55] <onga> every day
[18:22:03] <pauliukas> hah
[18:22:16] <onga> thats running Sun Java Messaging Server
[18:22:26] <onga> web services on them, absolutely hammer
[18:22:35] <Pietro_S> reinstalled  xfce, but still gnome-keyring-daemon doesn't work :-(
[18:22:37] <onga> our web proxy servers, running sun java proxy server, hammer on t1
[18:22:48] <pauliukas> hammer = good, right?
[18:22:53] <onga> absolutely
[18:22:54] <pauliukas> Not as in... Getting hammered? heh
[18:22:58] <_bernie> movement: may I ask your help on an oprofile issue?
[18:23:18] <onga> they hit no more than a load of around 10% cpu use per box
[18:23:27] <onga> and support thousands of concurrent users, accessing the web
[18:23:32] <pauliukas> I've been looking at the whole Solaris system for a few days now.
[18:23:37] <pauliukas> Gosh I wish that I could use it.
[18:23:48] <pauliukas> Unfortunately, the business that I do is not too fond of Solaris.
[18:23:51] <_bernie> movement: the the OLPC laptops, we can't seem to get stack traces out of oprofile... the same version of oprofile works fine on PCs with Fedora 7.
[18:24:03] <onga> pauliukes: why can't you?
[18:24:05] <_bernie> movement: any idea why this could happen?
[18:24:11] <pauliukas> onga: I'm in the shared web hosting market.
[18:24:22] <pauliukas> The software that's used is really closely tied into Linux
[18:24:32] <pauliukas> Or *gulp* Windows.
[18:24:36] <bda> onga: How much mail?
[18:24:53] <sfire||mouse> pauliukas: ensim? or cPanel?
[18:25:01] <pauliukas> sfire||mouse: I use DirectAdmin
[18:25:02] <onga> pauliukes: when you say tied to linux, do you mean poorly coded and won't compile on a strict compiler? :P even so most GNU stuff (some with minor mods) will run on solaris
[18:25:09] <pauliukas> I despise ensim and cPanel :-P
[18:25:16] <pauliukas> They take over the system too much and cost too much and are very unreliable
[18:25:21] <onga> bda: when you say how much mail? sotrage or daily incoming/outgoing?
[18:25:25] <bda> onga: As in stored, and daily throughput?
[18:25:43] <pauliukas> onga: The panels used in web hosting are just tied down to Linux. They use quotas to calculate disk space.
[18:25:57] <onga> bda: stored we have 10 message stores @300GB each, utilise about 1/3 of each store
[18:25:58] <sfire||mouse> pauliukas: we have 2 ensim boxes here, one for use to use, one that one of our clients wanted.  Then we have a hosting controller box
[18:26:09] <pauliukas> And they're mostly closed source. So you can't go in there and make it somehow work with Solaris.
[18:26:19] <pauliukas> sfire||mouse: Ensim works on Solaris?
[18:26:32] <sfire||mouse> pauliukas: nope
[18:26:36] <pauliukas> Gotcha.
[18:26:40] <onga> daily traffic, from memory can be upwards of 200,000, but has been upto a million on the gateways when we had some 0wn3d machines spamming away
[18:26:41] <onga> :P
[18:26:44] <sfire||mouse> pauliukas: I'm configuring my first sun box now
[18:26:49] <bda> onga: Nice.
[18:26:57] <pauliukas> I've just heard so many hell stories with Ensim. I really love DirectAdmin. Gives me more power to setup the server how I want.
[18:27:01] *** Cyrille has quit IRC
[18:27:11] <pauliukas> Since yesterday, I've been looking for panels that could support Solaris.
[18:27:14] <onga> the gateways aren't T1 though, they're v240s
[18:27:15] <pauliukas> One that I found is VirtualMin.
[18:27:26] <onga> runn sun messaging server too, and Sophos PureMessage
[18:27:27] <pauliukas> I'll need to play with it and see. But I don't like their licensing costs so much
[18:27:47] <onga> cpanel?
[18:28:00] <pauliukas> cPanel is Linux-only and I hate it!!
[18:28:11] <pauliukas> It's a collection of loosely-nit-togethor scripts.
[18:28:18] <pauliukas> And it only supports apache 1.3, on top of that.
[18:28:21] <bda> Feh, Sophos.
[18:28:39] <onga> since when was it only linux?
[18:28:51] <pauliukas> onga: Okay... Fine. Linux and BSD
[18:29:14] <pauliukas> onga: http://www.cpanel.net/products/cPanelandWHM/linux/sys_requirements.htm
[18:29:22] <bda> onga: What are your storage backends?
[18:29:31] <bda> I'm looking at rebuilding that bit of our infra.
[18:29:51] *** alanc-away is now known as alanc
[18:30:32] <onga> bda: Sun 3510 and thats moving to an EMC (sigh) SAN
[18:30:44] <onga> with 4gb FC
[18:30:53] <onga> but the disks only need to be SATA
[18:31:14] <bda> Yea, I figured it was outside my price range. :)
[18:31:15] <onga> if I were to do it again
[18:31:21] <onga> I'd change the 240s for T1s
[18:31:22] <wesolows> wow, 4gb.  That's only about 3 times slower than first-generation SAS
[18:31:23] *** julesa has joined #opensolaris
[18:31:36] <onga> 4GB FC is slower than SAS?
[18:31:37] <onga> :P
[18:31:45] <wesolows> and only about 3x as expensive, too.  What a steal!
[18:32:03] <wesolows> SAS == 3gb x 4 per port
[18:32:07] <onga> areyou sure you don't want to check your figures of FC drives VS SAS drives
[18:32:10] <pauliukas> http://www.sun.com/servers/coolthreads/t1000/index.xml
[18:32:20] <pauliukas> $4000 for a server with a non-hotswappable 80GB SATA drive!
[18:32:24] <pauliukas> what the f**
[18:32:29] <wesolows> he doesn't care about the drives, they're SATA
[18:32:45] <onga> and you have a SAS based SAN switch too ?
[18:32:46] <onga> :P
[18:33:06] <onga> FC is expensive though, yes.
[18:33:07] <wesolows> SAN is just another word for boondoggle
[18:33:41] <onga> sure, why not.
[18:33:44] <wesolows> if your vendor can afford to take you to strip clubs for buying their gear, you ought to think about what their margins must look like...
[18:34:02] <bda> O_o
[18:34:03] <wesolows> and wonder why their company keeps all of your company's money
[18:34:04] <pauliukas> lol
[18:34:05] <onga> don't get me wrong. Storage vendors are a rort
[18:34:07] <monkey_squad> or they are just an awesome vendor
[18:34:09] <onga> and disks are over priced
[18:34:27] <monkey_squad> lap dances are money well spent
[18:34:44] <millhouse> how stable is opensolaris b71 (in comparison w/build68)?
[18:34:48] <onga> but if you look at a few basic stats, and then a few real world stats, you'll find SAS isn't so great. you won't be filling 3gb/s pipe anytime soon with a single SAS drive.
[18:35:01] <monkey_squad> the monkey squad throws money at strippers just like how they throw rocks at monkeys : fast, hard, and often
[18:35:13] <wesolows> huh?  when did I ever say you would?  no single disk can sustain 3gb/s
[18:35:27] <stevel> millhouse: it fixed some bugs. and introduced new ones.  ostensibly it would have benefited from some of the attempt to stabilise b70.  you can read however much into that you want though... :)
[18:35:54] <LLcoolM> wesolows: and 640k will be enough for everyone :)
[18:36:00] <pauliukas> LOL
[18:36:39] <wesolows> ok, either you guys are talking nonsense or I'm still suffering the aftereffects of last night's hookers and blow til dawn party
[18:36:44] <onga> wesolows: how do you connect SAS drives?
[18:36:55] * wesolows assumes it's the latter and wanders away
[18:37:08] <wesolows> I usually use a cable.
[18:37:22] <onga> a direct cable
[18:37:29] <wesolows> No, a cable to an expander.
[18:37:39] <wesolows> A direct cable from an HBA to a disk drive is a waste of a phy.
[18:37:58] <wesolows> It's ok for disks inside a server but makes no sense for arrays.
[18:38:52] <movement> _bernie: almost missed that altogether. try #oprofile on OFTC
[18:39:03] <onga> thats right
[18:39:54] *** hile_ has joined #opensolaris
[18:39:55] <onga> so when you're connecting to a SAN, with multiple tiers of storage, and types of disks, can you do that with a SAS connection?
[18:40:14] <pauliukas> http://linux.slashdot.org/article.pl?sid=07/09/04/0028230
[18:40:17] <pauliukas> I wish they do!
[18:42:06] <bda> http://www.ibm.com/us/ # ru ready?
[18:44:28] <bda> IBM has some college slacker and IM speak. HP has .. Gwen Stefani?
[18:44:38] <bda> Weird.
[18:44:51] <millhouse> stevel:  thanks.  I will..  Mostly I'm curious about iscsi and zfs support in it...
[18:46:10] *** sommerfeld has joined #opensolaris
[18:46:11] *** ChanServ sets mode: +o sommerfeld
[18:48:22] <pauliukas> Hey guys/
[18:48:29] <pauliukas> Many corporations and companies use Solaris, right?
[18:48:39] <pauliukas> Is there any official page showing the biggest users and some stats?
[18:50:55] <PerterB> that's not the kind of thing large corporations like to publish
[18:51:01] <loke_> pauliukas: yes
[18:51:03] <nachox> pauliukas: you can start with http://www.sun.com/software/solaris/success_stories.jsp
[18:51:21] <loke_> PerterB: actually, if you're a Sun customer you can ask for references
[18:51:38] *** bengtf has quit IRC
[18:51:50] <pauliukas> Friends say that Linux is more popular among companies than Solaris.
[18:51:54] <loke_> PerterB: I can tell you that many of the largest banks are using Solaris
[18:52:02] <pauliukas> for some reason, I thought that every corporation ran Sun.
[18:52:06] <loke_> sorry, that was for pauliukas
[18:52:08] <pauliukas> For some service or something.
[18:52:18] <pauliukas> loke_: Awww. You're thinking of me. Thanks.
[18:52:30] <loke_> pauliukas: but why is this relevant for you? Isn't it better to evaluate it on its merits?
[18:52:38] <monkey_squad> your friends are right paul
[18:52:40] *** migi has quit IRC
[18:52:57] <pauliukas> Well, I thought that I'd ask some people who do Linux adminning every day and see their opinion.
[18:53:03] <loke_> pauliukas: it depends on wat you mean by "popular"
[18:53:18] *** sahafeez has joined #opensolaris
[18:53:27] <pauliukas> How many companies/corporations/enterprises use the Solaris OS.
[18:53:36] <loke_> pauliukas: if all they do is linux, they'll be frustrated initially by Solaris. Not becuase it's worse, but because some fundamental tasks are different
[18:53:49] *** cypromis has quit IRC
[18:53:50] <pauliukas> Right. I'm also a Linux user, actually.
[18:53:51] <loke_> pauliukas: many... very many
[18:54:10] <pauliukas> If you have any resources that would help me learn Solaris, that'd be great too.
[18:54:29] <delewis> it's hard to use Linux in an environment when it fails to provide mechanisms to answer simple questions.
[18:54:46] <loke_> pauliukas: Not really... It's been a long time since I needed such docs
[18:54:53] <loke_> and I don't do Solaris training anymore
[18:54:56] <delewis> pauliukas: http://docs.sun.com
[18:55:03] <stevel> pauliukas: http://www.opensolaris.org/os/community/advocacy/immigrants/
[18:55:10] <pauliukas> I was thinking of books. Or something that I could read from A to Z.
[18:55:25] <delewis> that generally doesn't work.
[18:55:30] <pauliukas> stevel: Thanks. that looks helpful.
[18:55:38] <delewis> grab yourself a generic 'howto use UNIX' text and then start using Solaris.
[18:55:42] <nachox> docs.sun.com has downloadable books in pdf
[18:55:44] <stevel> go to that URL, open up every page. go to File->Print. sort them alphabetically. then bind them into a book.
[18:55:45] <delewis> fill in your knowledge gaps by consulting docs.sun.com
[18:55:46] <monkey_squad> Im pretty sure the man pages and admin guides are alphabetically sorted, paul :)
[18:55:52] <pauliukas> hah.
[18:55:59] <pauliukas> I've been doing that for years on Linux.
[18:56:09] <pauliukas> But I thought that there was a better way to learn.
[18:56:11] <delewis> pauliukas: you're obviously in need of a 'howto use UNIX text'..
[18:56:18] <delewis> that might be the best place to start.
[18:56:24] <pauliukas> UNIX for Dummies? :-P
[18:57:02] <delewis> well, considering that you lack knowledge like how to set a PATH, such a text might prove beneficial.
[18:57:13] <monkey_squad> realistically you should just get some broad stroke advice on what to look at, and maybe the people here can do that
[18:57:19] <delewis> such knowledge gaps can be huge stumbling blocks in your experience with Solaris.
[18:57:57] <pauliukas> okay. thanks for the advice
[18:58:02] *** alred has quit IRC
[18:59:14] *** JSRJ_ has joined #opensolaris
[18:59:57] *** spiekey2 has joined #opensolaris
[19:00:43] * loke_ is browsing this into to Solaris thing... While a lot of it is good, the language is terrible. The guy can't spell...
[19:01:37] <delewis> who's the author?
[19:01:43] <spiekey2> hi
[19:02:10] <spiekey2> i was send here from #solaris since i have a "libcmd.so.1 file not found" error when i boot
[19:02:26] <spiekey2> it floods my whole screen and i can only boot into single user mode.
[19:02:37] <spiekey2> i was hoping to get some hints/tips?!
[19:02:38] <loke_> delewis: ben rockwood
[19:05:30] <mp3ters> ben rockwood is solid
[19:05:45] <palowoda> no doubt
[19:06:35] *** Plazma_Work has joined #opensolaris
[19:07:31] <julesa> Is there a ports project like the bsd's, or apt-*, or is it all roll your own?
[19:08:10] <sommerfeld> julesa: there are several
[19:08:26] <palowoda> blastwave.org pkg-get
[19:08:48] <spiekey2> anyone? :)
[19:08:48] *** sahafeez has quit IRC
[19:09:10] *** sahafeez has joined #opensolaris
[19:09:11] <monkey_squad> what build of sxce are you running spiekey2!
[19:09:20] <julesa> sommerfeld, palowoda, thanks.
[19:09:38] *** GmanaFK is now known as Gman
[19:09:42] <sickness> is it normal that both the snv70 and 71 install in vmware hangs at "Setting up Java. Please wait..." ?
[19:09:48] <sickness> (tried 1 and 4 options...)
[19:09:59] <palowoda> vmware isn't normal with solaris.
[19:10:07] <spiekey2> monkey_squad: 5.10
[19:10:28] <monkey_squad> 71 installed fine for me, sickness
[19:10:36] <sickness> monkey_squad: tnx
[19:10:58] <monkey_squad> naturally stuff takes a long time to happen though
[19:11:11] <monkey_squad> so your hang might just be normal inaction action? :)
[19:12:01] <spiekey2> monkey_squad: is 5.10 the answer you were waiting for?
[19:12:06] *** infidel has joined #opensolaris
[19:12:08] <spiekey2> i am not sure what you meant ;)
[19:12:36] <palowoda> Why did they send you here if your using 5.10?
[19:12:57] <spiekey2> i dunno :(
[19:13:05] <palowoda> Them slimballs.
[19:13:22] <spiekey2> they said #solaris is just for dicsussion, not for helping
[19:13:38] <palowoda> What does that mean?
[19:13:40] <monkey_squad> clearly slimeballs
[19:14:02] <spiekey2> i am coming from the linux world and i am sure i can solve this if i get some hint or tips or instructions.
[19:14:36] <palowoda> Try opensolaris instead of solaris.
[19:15:11] <palowoda> Solaris is kind of old and outdated.
[19:15:21] <spiekey2> thats unfortunately not possible since i have to use solaris 5.10 due to the customer :(
[19:15:22] <monkey_squad> if you are just playing with yourself spiekey, definitely use solaris express community edition, not solaris 1 through 10
[19:15:30] <PerterB> palowoda: you mean "tried and tested"
[19:15:41] <monkey_squad> in that case your customer can pay for real tech support for their old os :)
[19:15:54] *** bengtf_ has joined #opensolaris
[19:15:55] <palowoda> PerterB: In the land of free support what do you want?
[19:15:56] <spiekey2> oh well :(
[19:15:56] *** bengtf_ is now known as bengtf
[19:16:33] <PerterB> palowoda: pardon?
[19:17:24] <palowoda> PerterB: Same as it ever was.
[19:17:33] *** JSRJ has quit IRC
[19:18:07] <spiekey2> could someone please at least give me the default content of ld.config?
[19:18:33] *** cypromis has joined #opensolaris
[19:18:43] *** cypromis has quit IRC
[19:18:44] *** Murmuria has joined #opensolaris
[19:19:24] *** cypromis has joined #opensolaris
[19:19:39] *** DannyJ has joined #opensolaris
[19:20:07] <monkey_squad> Id help you out but I really know nothing about solaris, I'm just here because I made a typo when joining #opensuse
[19:20:32] <stevel> that's a helluva typo
[19:20:46] <micken> heh
[19:20:56] <palowoda> spiekey2: man crle
[19:21:01] <flyingparchment> crle :((((
[19:21:03] <stevel> no honey, i totally typo'd. i meant to type 'cnn.com' and instead i typed 'goatse.cx'
[19:21:14] <flyingparchment> use LD_RUN_PATH or -R instead of crle, you'll be much happier
[19:21:27] <pauliukas> monkey_squad: rofl
[19:21:29] <pauliukas> Nice typo...
[19:21:51] <flyingparchment> sickness: i had that problem too.  the caiman installer worked though.
[19:22:30] <sickness> flyingparchment: bumping the virtual ram from 512mb to 768mb solved the problem...
[19:22:41] <flyingparchment> sickness: for the old installer?
[19:23:24] <sickness> for snv71 I think...
[19:23:42] <sickness> it still starts up java, wasn't caiman meant to be only gnome/gtk2 and no java? :)
[19:25:10] <Berny> christ whats wrong with osc-eu again?
[19:25:53] *** Murmuria has quit IRC
[19:26:47] *** bondolo has quit IRC
[19:28:03] *** Murmuria has joined #opensolaris
[19:28:14] *** libkeise1 is now known as libkeiser
[19:33:00] *** bondolo has joined #opensolaris
[19:33:11] *** bubbva has joined #opensolaris
[19:36:49] *** xOmega has left #opensolaris
[19:37:54] *** spiekey2 has quit IRC
[19:47:58] *** julesa has left #opensolaris
[19:48:25] *** Mdx4 has joined #opensolaris
[19:52:00] *** cypromis has quit IRC
[19:58:18] *** WifiJane has joined #opensolaris
[20:00:50] <_bernie> movement: thanks
[20:02:01] *** DoYouKno has quit IRC
[20:02:27] <Berny> .oO(someone go and kick the osc server please)
[20:02:55] *** Emmedics4 has joined #opensolaris
[20:03:02] *** Mdx4 has quit IRC
[20:03:15] *** Dar is now known as Dar_HOME
[20:06:26] <Berny> .oO(lemme guess sun switches osc to the new portal today and right now?)
[20:06:45] *** cydork has quit IRC
[20:08:42] *** JSRJ has joined #opensolaris
[20:08:53] <pauliukas> What's osc?
[20:09:04] <flyingparchment> online support centre
[20:09:05] <Berny> online support centre
[20:09:08] <pauliukas> oh
[20:09:36] <Berny> .oO(another one writing english english :-))
[20:09:57] * flyingparchment is english :)
[20:11:40] <Berny> how convenient
[20:12:34] <LLcoolM> what happened: suddenly i can download slowaris fast
[20:13:02] <flyingparchment> saying slowaris makes you look like a tard
[20:13:05] <Berny> they switched osc's bandwidth to dlc?
[20:13:08] *** pschow has quit IRC
[20:14:04] <LLcoolM> Berny: dunno, probably. flyingparchment that was a stylistic device
[20:14:34] <Berny> wait till you installed it... then think again about slow :)
[20:15:44] <Berny> hooo, doesn't time out any more... now it gives a java exception
[20:15:48] *** Sup3rkiddo has quit IRC
[20:16:02] *** Emmedics4 has quit IRC
[20:16:06] *** fbo has left #opensolaris
[20:16:17] <LLcoolM> Berny: i want to get rid of this silly sun download manager as soon as possible so the faster it downloads the better :)
[20:16:41] *** Snake007uk has quit IRC
[20:17:02] <Berny> sdm isn't all that bad... runs just fine via webstart
[20:17:07] *** perlmonk has quit IRC
[20:17:30] <LLcoolM> Berny: i hate the sun ads. i guess i will compile them out :)
[20:17:39] <LLcoolM> at least if i intend to use it again
[20:17:40] <pauliukas> hah
[20:17:41] <pauliukas> what sun ads
[20:17:54] <Berny> get sdm 1.x didn't have ads :-)
[20:18:02] <flyingparchment> why use sdm anyway?
[20:18:14] <LLcoolM> flyingparchment: because download was awfully slow this morning
[20:18:16] <Berny> pauliukas: try sdm 2 it features sun headlines/ads
[20:18:24] <pauliukas> oh. neat.
[20:23:30] <axisys> I am getting this on my sol b70 x86 "template for bge0 incorrectly configured change to cipso type" .. fix is to remove solaris truxted extension.. how do I remove it? I do not have a /var/sadm/tx dir either
[20:23:41] *** Emmedics4 has joined #opensolaris
[20:24:02] *** xOmega has joined #opensolaris
[20:24:42] *** visc has joined #opensolaris
[20:24:54] *** xOmega has left #opensolaris
[20:25:11] <visc> Hey guys. is there a solaris 9 x86 download for sun explorer?
[20:25:13] *** JSRJ_ has quit IRC
[20:25:48] *** hile_ has quit IRC
[20:29:03] <sickness> how can I disable those port in a default install of snv71? *.111, *.49357, *.55080 ?
[20:29:59] <asyd> rpcbind can be disabled yes, but you lost lot of things
[20:30:26] <axisys> sickness: netservices limited is probably what u need
[20:30:35] <sickness> axisys: oh
[20:31:45] *** hile_ has joined #opensolaris
[20:31:59] <sickness> nope
[20:32:03] <sickness> those 3 ports are still opened :/
[20:32:10] <axisys> sickness: not really open
[20:32:13] *** monkey_squad has quit IRC
[20:32:17] <axisys> sickness: to the outside world
[20:32:18] *** bengtf has quit IRC
[20:32:22] <Emmedics4> sickness: fuser them :)
[20:32:25] *** Emmedics4 is now known as Mdx4
[20:32:25] <Berny> open only locally
[20:33:25] <sickness> really?
[20:33:29] *** fbo has joined #opensolaris
[20:33:33] <axisys> sickness: really :-)
[20:33:40] <Berny> thats at least my understanding ;-)
[20:33:42] <sickness> http://rafb.net/p/do1a4c29.nln.html  <- doesn't * mean every address? like 0.0.0.0?
[20:33:49] <sickness> I'd prefer to see 127.0.0.1.port =)
[20:34:00] <visc> are you able to telnet to your IP from these ports?
[20:34:06] <axisys> sickness: there is an article on it.. i cant find it
[20:34:14] <sickness> I don't know if 111 would be telnettable anyway :/
[20:34:27] <Berny> axisys: wasn't that on bigadmin somewhere?
[20:34:39] <visc> to check if your machine is listening at this port.
[20:35:01] <Berny> sickness: it's telnettable it's just the question if you can then type the protocol correctly :-P
[20:35:08] *** DoYouKnow has joined #opensolaris
[20:35:31] *** bnitz has left #opensolaris
[20:35:37] <sickness> axisys: http://blogs.sun.com/gbrunett/?entry=solaris_secure_by_default_part2 <- maybe this?
[20:35:42] <axisys> sickness, Berny : http://www.google.com/search?q=rpcbind+netservices+limited
[20:35:43] <DoYouKnow> Does the ATI Mobility on the HCL encapsulate ATI Radeon Xpress 200M?
[20:35:43] <Mdx4> sickness: 111 is portmap or i'm wrong ?
[20:35:44] <axisys> 3rd link
[20:35:47] <sickness> Berny: lol
[20:36:00] <axisys> sickness: exactly!
[20:36:12] <axisys> sickness: so u r fine
[20:36:16] <sickness> Mdx4: yeah, should be, but I don't want to just shutdown that service, I know that it will broke a lot of other little things...
[20:36:22] <sickness> axisys: tnx, I'll read...
[20:36:53] <Mdx4> if pr0n works is all fine.
[20:37:00] <Pietro_S> DoYouKnow: depend what you want to do, ATI Express 200M works fine in 2D, but no 3D at all
[20:37:17] <pauliukas> Hey guys.
[20:37:20] <pauliukas> How do you update solaris?
[20:37:24] <Berny> axisys: i'm getting old :-P
[20:37:36] <sickness> Mdx4: yeah, that's the point of it all :)))
[20:37:37] <Berny> pauliukas: solaris 10 or nevada?
[20:37:43] <pauliukas> 10
[20:37:44] <palowoda> pauliukas: What build are you running?
[20:37:49] <pauliukas> x86?
[20:37:51] <pauliukas> 11/06?
[20:37:59] <palowoda> Oh you need money for that.
[20:38:11] <Berny> pauliukas: live upgrade or regular upgrade...
[20:38:16] <Berny> or fresh install
[20:38:29] <pauliukas> Just to update to the latest version.
[20:38:34] <pauliukas> Like you'd update linux.
[20:38:42] <pauliukas> To get new binaries, new kernel, etc.
[20:38:43] <palowoda> Just select the upgrade option.
[20:39:00] <Berny> dvd in, click through option, upgrade
[20:39:06] <pauliukas> dvd? what?
[20:39:09] <pauliukas> But I don't have it.
[20:39:14] <palowoda> download it
[20:39:15] <Berny> dwonload it
[20:39:16] <pauliukas> Doesn't it fetch the updated stuff from the net?
[20:39:20] <Berny> nope
[20:39:22] <pauliukas> You're kidding me...
[20:39:25] <palowoda> that is pay for stuff
[20:39:36] <pauliukas> So I need to burn a DVD whenever I want to simply update a few things?
[20:39:42] <Berny> no
[20:39:46] <palowoda> you should have load opensolaris
[20:39:54] <palowoda> s/load/loaded
[20:40:02] <Berny> download the image, loopback mount it and do a live upgrade :-)
[20:40:05] <pauliukas> opensolaris does that kind of stuff better?
[20:40:10] <palowoda> yep
[20:40:15] <pauliukas> good. it better.
[20:40:22] <Berny> palowoda: bfu? :-)
[20:40:30] <palowoda> which build?
[20:41:10] <palowoda> Berny: which build?
[20:41:55] <Berny> palowoda: i was just asking if you referred to bfu when saying nevada does upgrades better
[20:42:26] <palowoda> No bfu isn't really an upgrade option now is it.
[20:42:35] <DoYouKnow> Pietro_S, in solaris 10 and opensolaris?
[20:43:11] <Berny> palowoda: well then you would still need and dvd image for upgrading nv don't you?
[20:43:18] <Berny> s/and/an/
[20:43:25] <palowoda> Oh wow download the image than.
[20:43:55] * Berny might have missed a more elgeant way though...
[20:44:32] <Berny> .oO(do it like a mathematician... reduce problem to a known problem and adapt that solution :-))
[20:44:39] <flyingparchment> upgrading nevada and S10 are pretty much identical..
[20:44:46] <flyingparchment> main difference is you have to upgrade nv because there are no patches
[20:44:55] <palowoda> Oh sure Berny we are leaning how to dance here.
[20:45:23] <Berny> what dance? :-)
[20:45:28] <palowoda> The boring dance.
[20:45:29] <millhouse> has anyone used filesync?
[20:46:58] <Pietro_S> DoYouKnow: in opensolaris, I never tried installation Solaris 10 on notebook ...
[20:47:31] <sickness> I'm just seeing that "PasswordAuthentication no" in /etc/ssh/sshd_config is not obeyed by sshd :/
[20:48:33] <axisys> sickness: by default it is yes
[20:48:49] <axisys> sickness: if u change it u need to restart the ssh
[20:48:50] <DoYouKnow> brb
[20:48:53] *** DoYouKnow has quit IRC
[20:48:57] <sickness> axisys: ok but I changed to no, issued an svcadm refresh, and even rebooted...
[20:49:00] <sickness> uhm
[20:49:03] <axisys> svcadm restart ssh
[20:49:24] <Berny> reboot should have fixed that...
[20:50:24] <sickness> lemme try again...
[20:51:34] *** tiger11 has joined #opensolaris
[20:51:55] <sickness> 111/tcp  open     rpcbind
[20:52:07] <sickness> nmap also clearly states that 111 tcp is still opened from the outside... :/
[20:52:35] *** jafari has quit IRC
[20:52:55] <tiger11> ISCSI backing stores - zvols vs. files?
[20:53:02] *** stevel has quit IRC
[20:53:23] *** Murmuria has left #opensolaris
[20:53:29] *** bengtf_ has joined #opensolaris
[20:53:32] *** bengtf_ is now known as bengtf
[20:56:09] <tiger11> does anybody know if one is definitively better than the other?
[20:58:20] <quasi> zvols sounds nicer
[21:00:07] <flyingparchment> i don't see why you'd use a file if you could use a zvol
[21:00:08] *** WickedWicky has quit IRC
[21:00:16] <CIA-26> daemon@elpaso: Added tag onnv_73 for changeset 1fec74abebba
[21:01:16] <palowoda> Hmm that was a weird CIA report.  I thought it was based on bugids?
[21:01:38] <tiger11> quasi/flyingparchment: here is my current situation. I am using zvols and it seems to function well, except that usage stats differ from host to initiator.
[21:01:55] <tiger11> is this a known issue?
[21:02:13] <flyingparchment> what does usage mean for iscsi?
[21:03:23] <movement> palowoda: it's based on commits to the open tree
[21:03:25] <sommerfeld> potentially less overhead involved in zvols
[21:03:28] *** millhouse has quit IRC
[21:03:57] <flyingparchment> palowoda: all onnv commit messages are just a list of CRs fixed in the commit
[21:04:25] <palowoda> But changeset 1fec74abebba refers to what CR?
[21:04:33] <sommerfeld> that'a mercurial changeset id
[21:04:36] * quasi wonders if using files as backing doesn't go with the same warning as when doing so in zfs
[21:04:51] <flyingparchment> that refers to no CR.. as sommerfeld says, it's a mercurial revision number
[21:05:35] <tiger11> flyingparchment: sorry, let me clarify. I create a zvol on the target and share it. I mount it on the initiator and start creating files.....
[21:06:11] <flyingparchment> tiger11: you missed a step.  you create a filesystem on the device, and mount that :)
[21:06:48] <tiger11> sorry....trying to hit the high points.
[21:06:57] *** perlmonk has joined #opensolaris
[21:07:52] <flyingparchment> well, iscsi doesn't just have file overhead, there's fs too.
[21:08:23] *** RaD|Tz has quit IRC
[21:10:04] <tiger11> This is what I am doing on the target
[21:10:05] <tiger11> 1. zfs create -V 3g vluns/lun0
[21:10:05] <tiger11> 2. zfs set shareiscsi=on vluns/lun0
[21:10:05] <tiger11> This is what I do on the initiator:
[21:10:05] <tiger11> 1. iscsiadm add discovery-address <some-address>
[21:10:05] <tiger11> 2. devfsadm -Cv -i iscsi
[21:10:07] <tiger11> 3. format (to find the device)
[21:10:08] <tiger11> 4. zpool create storage <device>
[21:10:16] <tiger11> does that good so far?
[21:10:25] *** Jondice has quit IRC
[21:11:12] *** pauliukas_ has joined #opensolaris
[21:12:00] <Marv|LG> docs.sun.com down today?
[21:12:19] <pauliukas_> I guess Sun doesn't want me to discover Solaris.
[21:13:12] <tiger11> I create an empty file with something like 'dd if=/dev/zero of=./file.dat bs=2048 count=665600'. When I create this file, host and initiator see it. I then delete this file, and the target still reports the same about of disk usage while the initiator is back to the same amount as before I created the file.
[21:13:24] *** sahafeez has quit IRC
[21:13:32] *** JSRJ has left #opensolaris
[21:13:34] <flyingparchment> that's because you can't "free" space on a block device
[21:13:36] *** theRealballchalk has quit IRC
[21:14:19] *** theRealballchalk has joined #opensolaris
[21:14:48] <tiger11> so as long as the initiator usage is lower than the zvol size, am I ok?
[21:14:59] <flyingparchment> yes.
[21:15:07] <flyingparchment> it's only a problem if you're using a sparse zvol anyway
[21:15:19] <tiger11> that was where I was eventually leading this :-D
[21:15:28] <sommerfeld> Or at least, you can't yet free space on a block device.
[21:15:45] <sommerfeld> if you enable compression *and* overwrite blocks with zeros, the zvol will free space when you zeroize blocks.
[21:16:11] <flyingparchment> what about if you disable compression?  is there anything like sparse files for zvols?
[21:16:32] <flyingparchment> hmm, actually, i guess sparse files don't work like that either..
[21:16:49] <sommerfeld> if you disable compression you still get sparse file behavior.  compression is currently the only way to turn allocated blocks into holes.
[21:17:59] *** bengtf has quit IRC
[21:18:11] *** Fullmoon has joined #opensolaris
[21:18:12] <elektronkind> general device driver question: Is it safe from winthin a driver (say, a fs driver) to open and manipulate something such as /dev/ip ?
[21:18:52] *** Fullmoon has quit IRC
[21:19:24] *** Plazma_Work has quit IRC
[21:19:52] <elektronkind> the driver needs to get a list of interfaces and their IP addresses, netmasks and whatnot - and do so in a portable way (ie, Solaris 10 pre- and post- IP Instances)
[21:20:08] <tiger11> so with sparse volumes, would I still be alright as long as the zpool was expanded to accomodate? or I am I missing something?
[21:20:22] <flyingparchment> tiger11: as long as you make sure to add space before you run out, you're fine
[21:22:23] <tiger11> flyingparchment: I appreciate the info.
[21:23:06] *** vmlemon has joined #opensolaris
[21:26:05] <sickness> someone is able to compile this on snv71? perl -MCPAN -e 'install Digest::SHA1'
[21:26:20] <sickness> it invariably exits with an error... :/
[21:26:44] *** stevel has joined #opensolaris
[21:26:45] *** ChanServ sets mode: +o stevel
[21:28:00] <axisys> sickness: rpcinfo -p <IP> should fail
[21:28:03] <Berny> .oO(can i open a ticket, that osc is broken?)
[21:28:06] <elektronkind> wow, SDLC download speeds have really picked up from this morning
[21:28:17] *** Lucifer has joined #opensolaris
[21:28:31] <tiger11> I am looking at bug id #6457694 for iscsi tuning. Are there any other tips/tricks for performance?
[21:28:58] <elektronkind> tiger11: jumbo frames
[21:29:09] *** Lucifer has quit IRC
[21:31:23] <flyingparchment> Berny: use the phone? :)
[21:32:00] * Berny grins
[21:32:22] <Berny> it's out of hours for my contract :-\
[21:32:56] <palowoda> Berny: You need more money.
[21:33:02] *** pauliukas has quit IRC
[21:33:22] <Berny> who doesn't? ;-P
[21:34:29] <palowoda> Did you expect someone to open a ticket for you on an opensolaris irc channel?
[21:34:38] <Berny> nope
[21:34:52] <flyingparchment> it's quite common for people to complain here about various bits of sun's website being down
[21:35:21] <palowoda> It maybe common but useless.
[21:35:24] <sommerfeld> unless it's opensolaris.org, the chance of someone here being able to do something about it low
[21:35:37] <flyingparchment> it's not useless, it's relaxing
[21:35:42] <flyingparchment> you can talk to all the other people with the same problem :)
[21:35:43] <Berny> indeed it is
[21:35:52] <Berny> very true
[21:36:44] <Berny> maybe someone from isnide sun knows if today is the day that sun switches from the old osc to new portal thing...
[21:38:03] <tiger11> elektronkind: good point. Last time I tried that, my NIC wasn't supported though. I will need to check that again.
[21:38:55] <sickness> axisys: ok it fails
[21:43:24] <Marv|LG> smf scripts are so easy to write...
[21:44:38] *** fbo has quit IRC
[21:46:03] *** micken has quit IRC
[21:50:20] *** nrubsig has joined #opensolaris
[21:50:21] *** ChanServ sets mode: +o nrubsig
[21:50:54] <nrubsig> stevel: ping!
[21:51:06] <nrubsig> stevel: where can I find the B72 source release annoucement email ?
[21:51:29] <stevel> i didn't send one
[21:51:34] <nrubsig> erm
[21:51:39] <nrubsig> stevel: grumpf... ;-(
[21:52:01] <nrubsig> stevel: that makes it difficult to write an annoucement with links to that email... ;-(
[21:52:37] <stevel> i haven't written build announcements for quite a while now
[21:52:38] <nrubsig> offtopic: is there any way to upload images (jpeg) temporarily somehow ?
[21:52:47] <nrubsig> stevel: ... ;-(
[21:53:14] <nrubsig> stevel: you fail as minister of propaganda^Wmarketing ... =:-)
[21:53:36] <stevel> i never held that position, so that's fine by me
[21:54:55] <nrubsig> http://www.nrubsig.org/ksh93_ray_test20070904_c001.jpg
[21:55:09] * nrubsig giggles madly
[21:55:34] *** yarihm has quit IRC
[21:56:18] *** vmlemon has quit IRC
[21:56:21] <Auralis> don't tell me you wrote a raytracer in ksh
[21:56:31] <nachox> nrubsig: did you call vt?
[21:56:38] <nrubsig> nachox: vt ?
[21:56:44] <nrubsig> nachox: ah
[21:57:00] <nrubsig> nachox: not yet, planned for tomorrow afternoon
[21:57:16] <nrubsig> nachox: I've been busy with babysitting the last two days... ;-/
[21:58:51] <nrubsig> Auralis: the problem is now to get texture&&bumpmap support working
[21:58:58] <Auralis> lol
[22:00:18] <nrubsig> erm
[22:00:22] <CIA-26> gd78059: 6297020 support for ADMtek ethernet chipsets (Linksys LNE100TX v4/v5) desired (fix packaging)
[22:00:36] <nrubsig> The big problem is to find a description how the physics work.
[22:01:36] *** laca has quit IRC
[22:03:05] <nrubsig> Auralis: and I found a new, clever way how complex datatypes can help with raytracing.
[22:03:09] <nrubsig> =:-)
[22:03:18] *** timsf has quit IRC
[22:03:22] *** migi has joined #opensolaris
[22:03:42] <nrubsig> but as a result I learned that |complex double/float| is half-busted in Studio 11
[22:04:11] *** bengtf has joined #opensolaris
[22:04:24] <kjetilho> ksh93 has support for complex numbers?
[22:04:47] <sommerfeld> it's the new FORTRAN
[22:05:00] <nrubsig> kjetilho: not yet
[22:05:12] <kjetilho> sommerfeld: :-)
[22:05:16] *** IvanR_ has quit IRC
[22:05:21] <nrubsig> kjetilho: but that's why "typeset"'s "-c"/"-C" options are currently reserved.
[22:05:22] <nachox> yeah, ksh93 will suck till it can handle complex numbers
[22:05:38] <nachox> becuse it is SO usefull
[22:05:58] <Auralis> does it handles imaginaery numbers yet? :)
[22:06:20] <kjetilho> nachox: marketing will LOVE IT for benchmarking
[22:06:26] <kjetilho> all the imaginary numbers they need
[22:07:10] <sommerfeld> And all they need to make them real is to rotate the rack 90 degrees.
[22:07:11] *** _bernie|home has joined #opensolaris
[22:07:26] <nrubsig> nachox: well, it gives it an edge over perl who don't have builtin complex suppot... :-)
[22:07:54] <nachox> but ksh has poorer networking support
[22:08:05] <nachox> and it has no classes!
[22:08:25] *** vmlemon has joined #opensolaris
[22:08:46] <Stric> .. and it's still shellscripting
[22:09:20] <kjetilho> classes in Perl4 were a hack, you can use the same trick in shell, I'm sure
[22:10:56] <axisys> sickness: there we go.. just because it sees port 111 does not mean anything.. i guess u realized by now
[22:18:04] *** xOmega has joined #opensolaris
[22:19:06] *** hile_ has quit IRC
[22:19:22] *** hile_ has joined #Opensolaris
[22:19:25] *** stevel has quit IRC
[22:19:35] <delewis> stop giving marketing ideas. Next thing you know, they *will* start publishing benchmarks that use complex numbers.
[22:20:40] <delewis> 'Niagara T2 sets record for SPECweb -- 130+56i'
[22:21:54] <hile_> *snort*
[22:22:21] <delewis> oh, wait, that's what they're doing with Performance/Watt and SWaP.
[22:22:26] <delewis> it can't be!
[22:22:29] <palowoda> T2 has to start making money at some point in time.
[22:23:17] *** yarihm has joined #OpenSolaris
[22:23:21] <sommerfeld> actually, wait.  the racks in the blackbox *are* rotated 90 degrees from the expected direction....
[22:23:56] <sommerfeld> (systems don't face into the aisle; they face the doors at the end)
[22:24:51] *** DannyJ has quit IRC
[22:25:47] * delewis would hate to walk into a Blackbox full of T2000s
[22:25:48] *** bobbyz has joined #opensolaris
[22:26:07] <delewis> that'd be an inferno.
[22:26:24] <palowoda> But where all hardware sluts.
[22:27:06] <trochej> Hmm
[22:27:14] <trochej> I'd love to walk in such a blakbox :)
[22:27:30] <delewis> no, you wouldn't. Well, you wouldn't want to stay for any length of time.
[22:27:41] <delewis> It's my understanding that T2000s put out *a lot* heat.
[22:28:05] <trochej> delewis: Considering my actual work accomodations, it wouldn't realy make a difference :)
[22:28:36] <Berny> wanna visit my server room? :-)
[22:28:45] <Stric> delewis: nominal power: 275W.. that's not too much
[22:28:55] <trochej> Nah, all I need is my room
[22:29:07] <trochej> 38 Degrees at morning
[22:29:08] <trochej> Typical
[22:29:16] <trochej> Or 18 at winter
[22:29:20] <trochej> :)
[22:29:36] <trochej> during winter?
[22:29:49] <delewis> SunSolve says roughly 1400BTUs/hr which isn't that bad.
[22:29:49] <Berny> server room has 25deg C all year long
[22:30:03] <delewis> most people I've talked to, though, say it puts a lot more heat than a system should it's size, though.
[22:30:05] <Stric> a v240 can do 442W
[22:30:06] *** mp3ters has left #opensolaris
[22:30:49] <Berny> prop some sf4800 in ;-P
[22:30:54] <trochej> Berny: You have to have a cool server room :)
[22:32:16] <Berny> trochej: i need to have the a/c fixed :-)
[22:35:02] *** emiel1 has joined #opensolaris
[22:35:23] *** emiel1 has quit IRC
[22:36:29] <nachox> cool, there is another openssh
[22:36:38] <nachox> a new version
[22:37:29] <nrubsig> nachox: don't worry, I don't intend to write a ksh93 version.
[22:37:34] <ofu> http://www.theregister.co.uk/2007/09/05/netapp_sues_sun_over_zfs/
[22:38:50] <nrubsig> Is docs.sun.com down again ?
[22:39:06] *** xOmega has quit IRC
[22:43:31] <nachox> all the servers are being used to transfer solaris 10, there was no time for docs.sun.com
[22:43:34] <nachox> :P
[22:45:50] *** migi_ has joined #opensolaris
[22:48:45] <nachox> ouch, that is an uncool thing to happen
[22:48:48] *** Mdx4 has quit IRC
[22:48:53] *** sfire||mouse_ has joined #opensolaris
[22:49:11] <nachox> luckily CDDL protect us :)
[22:49:45] <kjetilho> not really, CDDL can't grant you a patent license unless Sun holds the patent
[22:50:06] *** sfire||mouse has quit IRC
[22:52:27] <delewis> I'm not really satisified with Netapp at the moment, but if it's true that Sun contacted them with the listed demands, then that's 'not good' (TM).
[22:52:36] <delewis> that's really going to hurt Sun's image.
[22:52:43] <delewis> hypocrisy++
[22:53:01] <richlowe> indeed.
[22:53:07] <delewis> 'you can use the technology we've open sourced as long as you're not a competitor'
[22:53:10] <richlowe> "software patents are bad, unless they're ours"
[22:53:29] <richlowe> that's how I'd read it, anyway, if there isn't more to it.
[22:53:51] <Auralis> there is always more
[22:54:05] <delewis> I'm sure there is, but you know how the first blood letting is. The Register read will hit Slashdot and that will be that.
[22:54:50] <delewis> 4-years of FUD confirmed. News at 10.
[22:54:57] <wesolows> delewis: Let me be very careful to say that I'm not commenting specifically on any litigation.  But the CDDL only grants patents if you take the code under the terms of that license (which in turn means keeping it open).
[22:55:05] <wesolows> *patent rights
[22:55:44] <nachox> kjetilho: 2.1 b
[22:55:58] <kjetilho> ?
[22:56:05] <nachox> in cddl
[22:57:01] <delewis> wesolows: I'm really not so much as worried about the legal situation than I am with Sun's image, and thus the image of OpenSolaris (the public still associates Sun intimately with OpenSolaris, as they should at this point)
[22:57:14] *** laca has joined #opensolaris
[22:57:35] *** ALVAN has joined #opensolaris
[22:57:40] <delewis> and you know how people are. They don't care about the legal specifics. This is 4-years of FUD confirmed to them, and gives them a solid foundation to create more FUD upon.
[22:57:41] <ALVAN> hi all
[22:58:02] <ALVAN> the boot loader from opensolaris can boot a fakeraid ?
[22:58:03] *** DoYouKnow has joined #opensolaris
[22:58:10] *** mp3ters has joined #opensolaris
[22:58:15] *** mp3ters has left #opensolaris
[22:58:20] <nachox> delewis: i really doubt that, it will happen what always happens, some noise at first and then nothing
[22:58:46] <delewis> we'll see. Netapp is pretty buddy-buddy with the Lunix and FreeBSD community.
[22:58:50] <nrubsig> alanc: ping!
[22:58:56] *** migi has quit IRC
[22:58:57] <kjetilho> nachox: hmm.  but can it be sustained?  sounds weird to me
[22:58:57] <flyingparchment> zfs is dead!  see, reiser4 was better all along
[22:59:14] <delewis> flyingparchment: I wouldn't be surprised to hear that sort of nonsense.
[22:59:20] <wesolows> ZFS: The murder-free alternative to reiser4?
[22:59:29] <richlowe> alleged murder.
[22:59:31] <delewis> I wonder if Jonathan will have the balls to say anything about it.
[22:59:38] <wesolows> ZFS: No wives were harmed in the making of this product?
[22:59:42] <richlowe> delewis: I doubt it.
[22:59:46] <delewis> I doubt anything as interesting as 'Yeah, we fucked up. Big.'
[22:59:50] <nrubsig> alanc: I am going to /op you as preparation for possible ZFS-is-dead-OpenSolaris-is-evil-lets-spam-the-channel-trolls
[22:59:53] *** nrubsig sets mode: +o alanc
[22:59:55] <Berny> when it comes to it he has the biggest balls ;-)
[23:00:04] <richlowe> delewis: any sane lawyer would lock him up some place and gag him, if he showed any indication of wanting to.
[23:00:10] <delewis> richlowe: yeah.
[23:00:13] <CIA-26> gww: 6598910 zonecopy function in labeld should validate pointer offsets, 6598913 Label builder door service should validate size of memory passed by caller, 6598969 datasize for obsolete stobclear calculation not clear.
[23:00:14] <holcomb> what were the patents netapp was "violating"?
[23:00:49] <delewis> holcomb: we won't know that until it goes to court, and even if then, that depends on the court proceedings being public, which probably won't be the case.
[23:01:22] <Berny> lawyers should just be burnt anyway
[23:01:28] <holcomb> i'm just wondering if they're more grounded in reality than netapp's "we patented snapshots" idea
[23:01:38] <delewis> that's how most patent lawsuits go, anyway. Company Foo settles with Company Bar for an undisclosed amount of money and conditions.
[23:01:58] <sommerfeld> The story at http://www.theregister.co.uk/2007/09/05/netapp_sues_sun_over_zfs/ above links to a .pdf of the complaint that includes patent numbers.
[23:02:05] <nachox> delewis: welcome netapp solaris 1.0 :)
[23:02:21] <kjetilho> holcomb: that claim is ungrounded?
[23:03:12] <holcomb> dunno.  seems fishy
[23:03:14] <kjetilho> I think snapshots by simply keeping uberblock was introduced by Netapp?
[23:03:34] <ofu> http://www.netapp.com/go/ipsuit/ has the whole story
[23:03:45] <wesolows> well, one side of it, maybe
[23:03:49] <richlowe> The old wafl papers indicate that they use a copy of the 'old' root inode, yes.
[23:04:09] <wesolows> pointless to speculate about legal proceedings
[23:05:16] *** jafari has joined #opensolaris
[23:05:38] <elektronkind> interesting
[23:05:58] <elektronkind> esp. since o'grady is there now, right?
[23:05:59] <delewis> I wonder if this has anything to do with Apple postponing full ZFS integration in Leopard.
[23:06:07] *** zeroxis has joined #opensolaris
[23:06:10] <richlowe> wesolows: doesn't seem it'll stop people.
[23:06:19] <wesolows> richlowe: It never seems to, no.
[23:06:39] <holcomb> http://www.theonion.com/content/node/29130
[23:06:42] <Stric> nice to see sessionids in legal papers
[23:06:45] <kjetilho> hmm, how did the WORM based filesystem in Plan9 work?
[23:06:57] <DoYouKnow> kjetilho, everything is a device
[23:07:03] <DoYouKnow> it unified the namespace
[23:07:40] <kjetilho> yes, but the WORM must have forced them to do something similar
[23:08:17] *** zeroxis has quit IRC
[23:08:18] <sommerfeld> just don't free blocks
[23:08:32] *** nivox has quit IRC
[23:10:35] <DoYouKnow> everything is a file, I mean. For example, most of the functionality of /proc in linux was developed within plan9 first before it was ported
[23:11:38] <nrubsig> can we just agree to sue Linux and then cancel this discussion, please ?
[23:11:46] <kjetilho> http://cm.bell-labs.com/who/seanq/p9fs.html or http://plan9.bell-labs.com/who/seanq/cw.pdf
[23:11:52] <nrubsig> that wuld save lots of time.
[23:12:12] <DoYouKnow> I think it's interesting how far behind Microsoft has been lately
[23:12:25] *** nostoi has joined #opensolaris
[23:12:36] <DoYouKnow> Windows is still a solid OS, but they have been behind in administrative features
[23:12:51] <sommerfeld> I'd love to see a serious attempt to argue that the as-built patent system is unconstitutional because it fails to "promote the progress of science and useful arts"
[23:13:51] *** kFuQ has quit IRC
[23:14:18] <the-decider> hmm, doesn't look like its hit /. yet
[23:14:22] <DoYouKnow> partitioning ability/zfs like stuff, integration of objects with more visible forms of the OS, Microsoft-manufactured debugging equipment (sysinternals was ahead of Microsoft in this regard). While other Operating systems, such as Leopard and Solaris seem to be aiming towards more integrated functionality
[23:14:52] *** kFuQ has joined #opensolaris
[23:15:23] <DoYouKnow> I heard Microsoft is revamping the partition manager in the next major release, but they will still probably be behind in other areas
[23:15:40] <DoYouKnow> apple was well known for a while to be behind in the very things they are ahead of in now
[23:16:57] <DoYouKnow> Microsoft definitely has the capability
[23:17:09] * DoYouKnow has a feeling they are bogged down in trying to support every type of hardware
[23:20:34] <DoYouKnow> the average user won't notice that stuff, but the average administrator will
[23:20:46] <DoYouKnow> if it's done right
[23:21:03] <elektronkind> WinFS still has yet to see the light of day
[23:21:54] <elektronkind> I still think that, after MS announced WinFS and sbsequently saw ZFS, they flinched. Back to the drawing board.
[23:23:55] <DoYouKnow> error recovery in NTFS is *slow*
[23:24:16] <DoYouKnow> if you are talking about serious damage
[23:24:39] <DoYouKnow> I was watching the demo on ZFS, they claim it is very fast
[23:25:26] <DoYouKnow> they [Sun]
[23:26:11] <kjetilho> haha, the Plan9 WORM Cache filesystem was implemented on a VAX-750 with a whopping 6 MiB of RAM.  those were the days.
[23:26:35] <flyingparchment> wasn't that one old enough to be called VAX-11/750? :)
[23:27:05] <kjetilho> flyingparchment: I actually typed that, but then I saw the paper called it just VAX-750
[23:28:08] <kjetilho> the idea is *very* close to prior art for patent 5819292, I'd say.  it is complicated by the addition of the cache layer.
[23:28:17] *** aptr has joined #opensolaris
[23:30:48] <brendang> does anyone know if the OpenSolaris development servers exist yet? ie, servers running different builds of OpenSolaris and Solaris for us to work on?
[23:30:55] *** nachox has quit IRC
[23:31:03] *** ALVAN has left #opensolaris
[23:31:27] <sommerfeld> brendang: i've seen some evidence that it's being worked on
[23:31:42] <brendang> sommerfeld: cool
[23:31:46] <sfire||mouse_> is there a best practice to add to PATH on logging in for /sbin/sh on solaris?
[23:32:14] <kjetilho> sfire||mouse_: yes, "don't"
[23:32:51] <kjetilho> I do "exec bash" as soon as I log in, and has the interactive setup in .bashrc
[23:33:15] <kjetilho> you may of course want to choose a different shell
[23:33:33] <richlowe> brendang: the test farm is in beta (see request-sponsor)
[23:33:37] <richlowe> brendang: I don't know anything about a build-farm.
[23:34:04] <sfire||mouse_> thats what I would probably do, I like bash :)
[23:34:16] <sommerfeld> of course, given a test farm running solaris...
[23:34:33] <tomww> richlowe: but would be a nice thing to have a build-farm :-)
[23:35:48] <sfire||mouse_> now to get this system uptodate
[23:37:21] <nrubsig> brendang: for which project ?
[23:37:32] <brendang> I actually want a test farm of both x86 and SPARC, running every customer release of Solaris 10
[23:37:38] <nrubsig> brendang: e.g. for which project do you need machine access ?
[23:37:39] <brendang> nrubsig: DTraceToolkit
[23:38:01] <nrubsig> brendang: well, there is an option but it's B65 only
[23:39:13] *** infidel has quit IRC
[23:39:40] * coffman votes for a build farm
[23:39:53] <coffman> any way to get some zones? :P
[23:40:48] <brendang> I'd need root and full DTrace access
[23:40:55] *** Tpenta has quit IRC
[23:42:34] <elektronkind> weird. The new s10u4 ISOs lack the UpgradePatches directory :/
[23:42:47] <richlowe> I thought that was by intent.
[23:42:56] <richlowe> wasn't that only needed for the old ashanti zone upgrade bits?
[23:43:12] <elektronkind> they were darn useful
[23:43:32] *** jlc has quit IRC
[23:43:42] <the-decider> well, how are we supposed to "upgrade" our u3 boxes to u4? :(
[23:43:43] <the-decider> :sniff:
[23:46:34] *** vmlemon has quit IRC
[23:47:34] *** ocr has joined #opensolaris
[23:47:37] <delewis> sure you don't have the 7/07 HW media? :-)
[23:47:37] *** bobbyz has quit IRC
[23:48:00] <the-decider> don't you mean 09/07 :P
[23:48:19] <delewis> no, I meant the HW release for the Mx000 hardware.
[23:48:48] *** migi_ has quit IRC
[23:48:57] <the-decider> don't be thinkin' so.  Dale's talking about the ISOs just d/ld from sun.com
[23:49:03] <the-decider> er dale == elektronkind
[23:49:04] * holcomb runs dial and waits for 6pm
[23:49:43] <elektronkind> oh no. it's sol-10-u4-ga-x86-dvd.iso
[23:51:15] *** migi_ has joined #opensolaris
[23:52:00] *** EtherNet- has quit IRC
[23:52:25] *** tiger11 has left #opensolaris
[23:53:34] *** bobbyz has joined #opensolaris
[23:53:37] *** nostoi has quit IRC
[23:53:42] *** BatonT has quit IRC
[23:57:49] *** bobbyz has quit IRC





top