November 14, 2006  
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30

[00:00:21] <gisburn> Beasts with rabies shouldn't run around freely.
[00:00:38] <gisburn> (disclaimer: watch the smileys dancing)
[00:00:45] <sickness> I don't even know if "personalities" like rms still code something these days...
[00:01:00] <sickness> they became more PR people than programmers...
[00:01:06] <g4lt-mordant> they don't
[00:01:09] <sickness> like james gosling :/
[00:01:17] <Auralis> gisburn: Steyr IWS2000 FTW :)
[00:01:18] <gisburn> sickness: we were ever PR people and no pogrammers
[00:01:29] <g4lt-mordant> rms actually admitted to me in a mail over four years ago that he doesn't coed at all any more
[00:01:40] <sickness> g4lt-mordant: oh, that's sad
[00:01:45] <gisburn> Auralis: nah. I suggest to do it from a secure range. The RMS cult followers may be a nasty problem.
[00:01:49] *** stevel changes topic to "opensolaris.org currently down | Latest builds: SXCR: 51a (sparc) / 51 (x86) | ON build: 51 | ON nightly: 20061106"
[00:02:04] <gisburn> stevel: erm... what changed ?
[00:02:07] <Auralis> gisburn: well, 40mm RHS at 1klick is nothing to sneez at
[00:02:19] *** stevel has left #opensolaris
[00:02:49] <gisburn> Auralis: I prefer the AMR since I can a) handle it and b) it can take out a Leopart-1 main battle tank easily if you hit the tracks
[00:02:52] <Auralis> gisburn: but for rapid cultists surpresseion: http://www.heckler-koch.de/core.php?dat=Y29tcG9uZW50PWFydGljbGVzJmFjdGlvbj1zaG93JnhJRD1wcm9kdWN0QXJ0aWNsZURldGFpbHMmYXJ0aWNsZUlEPTIyNiZjYXRJRD0xMzA1JmxhbmdJRD0xJnBhcmVudElEPTY0NyZuYXZpZ2F0aW9uSUQ9NjUxJnVzZUZsYXNoPTE=
[00:03:24] <gisburn> Auralis: nope.
[00:03:24] <Auralis> gisburn: the AMR is jsut the first prototype of the iws
[00:03:27] <gisburn> Auralis: nothing for me.
[00:03:41] <gisburn> Auralis: oh... I left the boat at that stage... ;-(
[00:03:44] <g4lt-mordant> nuke them from orbit, it's the only way to be sure
[00:03:50] <sickness> lol
[00:04:07] <gisburn> g4lt-mordant: ever heart of surgical attacks ?
[00:04:17] <Auralis> gisburn: for the iws they uped the calibrer to 15.2 from the original 12.7
[00:04:20] <sickness> what about the good old cs weapons? m4a1 / ak47 anyone? :P
[00:04:33] <gisburn> g4lt-mordant: I don't declare war on entire countries like Mr. bush to take out one person
[00:05:11] <gisburn> sickness: AK-47 ? Hello ? That thing is as precise as throwing bricks from the same distance.
[00:05:21] <g4lt-mordant> gisburn, it IS surgical.  surgery can be done with chainsaws as well
[00:05:38] *** Burana has quit IRC
[00:05:38] <g4lt-mordant> bah, BMG.  the only way to fly
[00:06:15] <Auralis> bmg is crap compared to the iws
[00:06:21] <gisburn> yup
[00:07:01] <gisburn> g4lt-mordant: AMR is a special anti material rifle, sort of a special purpose built sniper gun to take out high-value equipment from ranges up to 3km
[00:07:33] <sahafeez> ok, get this - the job of the military is to kill people and break thing.
[00:07:34] <gisburn> actually not designed for use against humans
[00:07:43] <Auralis> with the IWs you can shot down a Apache with a single shot
[00:07:44] <gisburn> sahafeez: actually no.
[00:07:48] <g4lt-mordant> gisburn, and the .50 BMG just starts to get it's wind up at 5km
[00:08:08] <gisburn> sahafeez: the idea behind the AMR is to avoid collateral damage
[00:08:09] <Auralis> g4lt-mordant: that is bullshit
[00:08:10] <g4lt-mordant> and yes, the BMG can take out a tank
[00:08:11] <sahafeez> gisburn, and what unit did you serve in?
[00:08:23] <sahafeez> war is won through fear
[00:08:30] <g4lt-mordant> Auralis, no, you hav a 5" drop with the .50 BMG at 5kyards
[00:09:02] <gisburn> g4lt-mordant: a .50 snipfer rifle lacks both punch, velocity and the special amor-piercing dart of the AMR
[00:09:12] <gisburn> sahafeez: no unit. Technical consultant.
[00:09:25] <gisburn> "civilist"
[00:09:27] * gisburn ducks
[00:09:50] <sahafeez> ah, well i am a graduate of the "john wayne school for wayward boys"
[00:11:00] <gisburn> sahafeez: as I said the AMR is not designed to be used against humans (directly). Usually the SF can use it to take out things like radar stations etc.
[00:11:21] *** laca has joined #opensolaris
[00:11:22] <gisburn> (e.g. you hit the emitter or receiver and you're done)
[00:11:29] <sickness> uhm, so is the amr the "AWP" that's found in cs? :P
[00:11:29] <sahafeez> sounds cool. after my time. i got out in 1993
[00:11:52] <Auralis> g4lt-mordant: 5" drop at 5 klicks with the BMG? lol, what are you smooking?
[00:11:57] <gisburn> sahafeez: it IS cool since noone really wants to mess with you if you have that thing.
[00:12:10] <sahafeez> back in my day we did not have that fancy gps bombs..no we had to hold a laser on the target from 2 clicks aways. these kids and their toys
[00:12:17] <g4lt-mordant> gisburn, ever hear of .50 AP?
[00:12:25] <gisburn> g4lt-mordant: AP=?
[00:12:32] <g4lt-mordant> armor piercing
[00:12:41] <g4lt-mordant> tungsten dart and he jacket
[00:13:15] <Auralis> still has only half the muzle energie of the IWS
[00:13:18] <g4lt-mordant> http://www.biggerhammer.net/barrett/faq.html
[00:13:43] <sahafeez> i loved the word light in the name. the barret light 50. that fucker is heavy
[00:14:41] <gisburn> sahafeez: depends. Wet silencer is around 3/4 of the wheight of the AMR itself.
[00:14:51] <gisburn> and it uses ceramics.
[00:15:04] <gisburn> making it very sentitive during transportation
[00:15:41] <sahafeez> well i have never used an AMR so..
[00:16:27] <sommerfeld> bah.  nothing beats orbital mind control lasers.
[00:16:54] <gisburn> heh
[00:16:54] <sahafeez> when i was in it was the, hum, M24SWS
[00:17:52] <g4lt-mordant> well, the real power is the 5"/54, the .50 BMG was really for popgun targets
[00:20:51] <PerterB> anyone have a working example of a grub menu.lst to dual boot two solaris instances in different slices of the same partition? Not really getting anywhere with the example in http://www.sun.com/bigadmin/features/articles/grub_boot_solaris.html (which makes sense, one of their instances is sol9 so they do the chainloader thing)
[00:24:37] *** miffe has left #opensolaris
[00:25:17] <sommerfeld> PerterB: live upgrade will set this up for you automatically
[00:25:37] <sommerfeld> PerterB: but give me a sec to dig up an example
[00:26:13] <PerterB> oh, ta... I didn't think to use LU, just installed nevada in a spare slice
[00:26:32] *** rmorse has joined #opensolaris
[00:28:18] <PerterB> presumably LU would handle the existing metadbs too, which the nevada installer doesn't well... doesn't give me the option to preserve the existing metadb slice(s) for example
[00:29:26] *** PPRO has joined #opensolaris
[00:29:32] <PPRO> hi all
[00:29:39] <jamesd> hello
[00:29:48] <gisburn> sommerfeld: ping!
[00:29:51] <sahafeez> http://slashdigg.com/ now if i could get slashdiggreddit.com
[00:30:09] <gisburn> sommerfeld: is there any workaround (ctf*-switch) to work around the CTF problem ?
[00:30:24] <sommerfeld> PerterB: http://pastebin.com/823713
[00:30:30] <PPRO> a little question, how can i check the wwn hba on sol8/sol9 (without fcinfo/hbacmd)
[00:30:45] * gisburn purrs to get sommerfeld's attention...
[00:30:46] *** slowhog has quit IRC
[00:30:50] <sommerfeld> eww.
[00:30:59] <gisburn> uhm
[00:31:05] <sommerfeld> gisburn: so, there are a few options
[00:31:06] <gisburn> sommerfeld: don't you like cats ?
[00:31:26] <sommerfeld> you aren't one.
[00:31:34] <gisburn> sommerfeld: prove it
[00:31:34] <sommerfeld> i dislike fakes
[00:31:36] <gisburn> =:-)
[00:32:08] <sommerfeld> gisburn: in answer to your question: I can think of several workarounds.
[00:32:09] <PerterB> sommerfeld: thanks! thought I tried something along those lines already, but that was maybe before I figured the (hdxxx) numbering out
[00:32:17] <sickness> hi PPRO :)
[00:32:28] <PPRO> sick
[00:32:31] *** stevel has joined #opensolaris
[00:32:31] *** ChanServ sets mode: +o stevel
[00:32:32] <sommerfeld> gisburn: 1) avoid use of the variable-size automatic array construct temporarily (use alloca instead)
[00:32:42] *** stevel changes topic to "Latest builds: SXCR: 51a (sparc) / 51 (x86) | ON build: 51 | ON nightly: 20061106"
[00:32:47] <djgregor> w00t!
[00:32:55] <twincest> stevel: any word on Hg?
[00:32:56] <sommerfeld> gisburn: 2) avoid using SS11
[00:33:13] <sommerfeld> gisburn: 3) temporarily disable use of ctf* for your builds.
[00:33:25] <djgregor> question: anyone have complaints if I were to work on adding "-print0" support to find(1) and "-0" support to xargs? :-)
[00:33:28] <sommerfeld> gisburn: 4) fix the ctfmerge code to not choke on the new debug syntax
[00:33:58] <sommerfeld> djgregor: I have this vague recollection that PSARC already approved of a proposal to do exactly that.
[00:34:38] <djgregor> sommerfeld: if you happen to have a reference handy, that would be cool, otherwise, I'll be hunting for it. :-)
[00:35:18] * djgregor has taken ideas from Solaris features and implemented them in OpenBSD, so he might was well go the (approximate) other way
[00:35:23] <sommerfeld> djgregor: give me a minute to dig it out
[00:36:01] <djgregor> I still need to debug my SX51 panics... whew, so much to do
[00:37:08] <gisburn> sommerfeld: based on the detail that the putback goes into OS/Net and is almost imminent [1], [2] and [3] are no options, leaving only [4] as viable path... ;-(
[00:37:38] <sommerfeld> djgregor: 2006/024 Find print0 Option Addition; was approved in January 2006
[00:37:56] *** sahafeez has quit IRC
[00:38:00] <stevel> twincest: i need to regenerate the repository; it's not trivial unfortunately
[00:38:03] <sommerfeld> djgregor: apparently has not yet integrated
[00:39:00] <twincest> stevel: ah, that's okay, as long as it hasn't been lawyered
[00:39:06] <djgregor> sommerfeld: okay.  thanks!  did you happen to notice if there was anything similar for "-0" on grep and friends?
[00:39:08] <stevel> no no, it's blocked on me at the moment
[00:41:11] <sommerfeld> the proposal as approved included find -print0, xargs -0, and cpio -0
[00:41:47] <sommerfeld> grep -0 seems like an obvious extension
[00:42:01] <sommerfeld> especially if there's prior art on one or more of the *BSD's
[00:42:04] <djgregor> it's one of the standard GNUish things
[00:42:08] <djgregor> and likely also SD
[00:42:11] <djgregor> er, BSD
[00:42:51] <gisburn> WTF does the -0 do ?
[00:42:55] <djgregor> hmm... looks like it's -Z
[00:43:04] <djgregor> gisburn: uses a null instead of a newline as the line separator
[00:43:13] <gisburn> erm
[00:43:27] <djgregor> so you can do fun things like "find / -type f -print0 | xargs -0 ..." and have it operate sanely no matter what is in a filename
[00:43:31] <gisburn> and this is usefull... now ?
[00:43:43] *** miffe has joined #opensolaris
[00:43:44] <sommerfeld> lest someone name a file "foo\nrm -rf /"
[00:44:13] <gisburn> oh, wair
[00:44:20] <gisburn> you mean '\0' ?
[00:44:25] <gisburn> that is silly
[00:44:31] <sommerfeld> no, it's very sensible
[00:44:48] <gisburn> the POSIX shell spec AFAIK forbids the usage of '\0' in variables
[00:44:57] <sommerfeld> if your threat model includes file names in the filesystem picked by adversaries
[00:45:15] <gisburn> sommerfeld: and the shell should be able to use this... how ?
[00:45:33] <sommerfeld> not the shell.  pipelines between shell commands
[00:45:49] <djgregor> gisburn: no shell variables involved, just command arguments (argv[])
[00:45:52] *** craigmohrman has joined #opensolaris
[00:46:17] <craigmohrman> Hello Roland.
[00:47:56] *** _william_ has left #opensolaris
[00:48:25] *** rmorse has quit IRC
[00:49:05] <PerterB> sommerfeld: that worked first time, guess I was being dumb with partition numbers before... thanks for th sanity check
[00:49:23] *** stevel has left #opensolaris
[00:50:15] *** koolniczka has quit IRC
[00:51:00] <djgregor> actually, xargs uses whitespace to separate arguments on standard input, not newlines... ouch
[00:52:12] <PerterB> now if only I could get my nge NIC working under SXCR
[00:57:42] *** miffe has left #opensolaris
[01:04:14] *** doownek has quit IRC
[01:05:27] *** estibi has left #opensolaris
[01:09:12] *** PPRO has quit IRC
[01:09:36] *** craigmohrman has quit IRC
[01:13:15] *** sil3nt|warri0r has quit IRC
[01:21:54] *** doownek has joined #opensolaris
[01:23:03] *** uebayasi has joined #opensolaris
[01:26:14] *** stevel has joined #opensolaris
[01:26:14] *** ChanServ sets mode: +o stevel
[01:26:19] *** stevel has left #opensolaris
[01:31:07] <twincest> ick, extendedFILE(5) seems like a nasty hac
[01:31:08] <twincest> +k
[01:32:37] <gisburn> yup
[01:32:41] <gisburn> use libast
[01:32:42] <gisburn> :-)
[01:32:48] <sommerfeld> well, it's the same sort of thing as 0 at 0 dot so.1
[01:33:15] <twincest> sommer: does that put a valid 0 word at address 0?
[01:33:24] <sommerfeld> build 64-bit, or use enable_extended_FILE_stdio(3C)
[01:33:41] <sommerfeld> yes, it maps a read-only page at virtual address 0
[01:34:09] *** stevel has joined #opensolaris
[01:34:09] *** ChanServ sets mode: +o stevel
[01:34:15] <twincest> hm.. no manpage?
[01:34:24] <sommerfeld> unfortunately, we were painted into a corner by the ABI; nasty hacks were the only way out
[01:34:41] <sommerfeld> twincest: hmm.  I see a manpage on snv_49
[01:35:04] <twincest> named 0 at 0 dot so.1?  i don't see it on snv_50
[01:35:20] <sommerfeld> oh, no, i thought oyu were referring to  enable_extended_FILE_stdio(3C)
[01:35:25] <twincest> oh
[01:36:31] <sommerfeld> 0 at 0 dot so.1 is documented in the ld.so.1 manpage
[01:37:05] <sommerfeld> "the user compati-
[01:37:05] <sommerfeld>      bility  library is intended neither to enable the generation
[01:37:05] <sommerfeld>      of such applications, nor to endorse  this  particular  pro-
[01:37:05] <sommerfeld>      gramming practice."
[01:37:08] <twincest> ah
[01:43:03] <Gman> stevel, welcome back to the land of the living :)
[01:43:21] *** mjf has quit IRC
[01:43:39] <stevel> gman: thanks :)
[01:58:59] *** migman has joined #opensolaris
[01:59:21] *** karrotx has joined #opensolaris
[02:00:04] *** peteh has quit IRC
[02:00:15] *** peteh has joined #opensolaris
[02:02:02] *** rpaulo_ has joined #opensolaris
[02:03:14] *** Kronuz has joined #opensolaris
[02:03:18] <Kronuz> hey are there any books you can recommend about Solaris?
[02:03:44] <Kronuz> I was thinking about Solaris Internals and the other companion book
[02:03:56] <Kronuz> but also I'd like something in the administrative realm
[02:06:03] <dlg> man -k
[02:07:37] *** rpaulo has quit IRC
[02:11:20] *** Yamazaki-kun_ has joined #opensolaris
[02:14:48] *** miffe_ has quit IRC
[02:19:07] *** Yamazaki-kun has quit IRC
[02:19:29] *** nwf has joined #opensolaris
[02:24:06] <tga> hello
[02:24:20] <tga> did anyone here played with WanRays by any chancE?
[02:24:36] <Kronuz> dlg: hehe man ?
[02:24:47] <dlg> Kronuz: sure
[02:24:52] <Kronuz> :P
[02:24:57] <Kronuz> why didn't I think of it!
[02:25:21] <Kronuz> dlg: well, actually I prefer to read from a printed source
[02:25:35] <Kronuz> but man is alright too I guess
[02:25:49] *** thesix1 has joined #opensolaris
[02:31:42] *** thesix1 has quit IRC
[02:33:20] <alanc> tga: there is no such thing as a Wan Ray - just Sun Rays running over lower bandwidth
[02:33:43] <Kronuz> sad to see the T2000 doesn't have RAID5 or RAID 1+0 support :(
[02:33:54] *** sartek has joined #opensolaris
[02:34:12] *** jtmuzix has joined #opensolaris
[02:34:25] <sartek> hi, (how) can i get solaris "nevada"?
[02:35:02] <alanc> download Solaris Express from sun.com or Solaris Express: Community Edition from opensolaris.org - both are snapshots of the Nevada release
[02:43:36] *** jtmuzix has quit IRC
[02:44:08] <sartek> alanc: ty, are major diffrences between the two versions for non-commercial users, i mean for non-server, desktop users?
[02:44:24] <dlg> can the sun compiler deal with macros with varargs?
[02:45:03] <boyd> sartek: the community release is more bleeding edge... less tested (still pretty good though)
[02:45:04] <alanc> no - Solaris Express: Community Edition is just released sooner, with less testing first, and more often - some of the same ISO's go out later as the full Solaris Express, once they've had more testing
[02:45:25] <boyd> Yeah, alanc said it better :)
[02:45:31] <alanc> (SX:CR is every two weeks, SX is once a month)
[02:45:34] <jbk> dlg: should be able to
[02:45:39] <jbk> however
[02:45:45] <dlg> however?
[02:45:46] *** sommerfeld has quit IRC
[02:45:59] <jbk> i believe the C99 std is slightly different than what gcc also allows
[02:46:23] <dlg> #define DPRINTFN(m, v...)       do { if ((m) & mfidebug) cmn_err(CE_NOTE, v); } while (0)
[02:46:26] <boyd> gcc allows somehitn other than the standard!?! I'm shocked!
[02:46:27] <dlg> would you expect that to work?
[02:47:56] <jbk> not quite
[02:48:06] <sartek> boyd, alanc thanks
[02:48:13] <dlg> jbk: educate me please
[02:49:04] <jbk> dlg: #define DPRINTFN(m,v, ...) do { if ((m) & mfidebug) cmn_err(CE_NOTE, v ## _VA_ARGS__); } while (0)
[02:49:54] <dlg> damnit, that doesnt let my macro fit nicely  in 80 chars
[02:50:07] <jbk> well you can continue it
[02:50:12] <jbk> and make it multiline
[02:50:29] <dlg> "mfi.c", line 40: invalid token in #define macro parameters: ...
[02:51:16] <jbk> [jking13@sunkc10 ~/work/pam_modules/pam_sprint/src]$ cc -c log.c
[02:51:34] <karrotx> opensolaris is for sissys
[02:51:36] * karrotx hides
[02:51:39] <jbk> grep '#include' log.c
[02:51:41] <jbk> #include <syslog.h>
[02:51:41] <jbk> #include <stdio.h>
[02:51:41] <jbk> #include <setjmp.h>
[02:51:41] <jbk> #include "log.h"
[02:52:04] <jbk> [jking13@sunkc10 ~/work/pam_modules/pam_sprint/src]$ grep "#define" log.h
[02:52:09] <jbk> #define VERROR(fmt, ...) \
[02:52:16] <jbk> works here
[02:53:05] <dlg> ugh, pam
[02:53:12] <jbk> heh.. i know :)
[02:53:42] *** avr_ has joined #opensolaris
[02:54:07] <jbk> thankfully that was just some prototype stuff that never had to go in :)
[02:54:21] <jbk> error handling & cleanup in pam modules is annoying :)
[02:54:46] <dlg> oh wait, i suck
[02:54:52] <dlg> pam modules are annoying
[02:56:16] <jbk> i know :)
[02:57:20] <dlg> do i need a header for that macro?
[02:57:39] <jbk> shouldn't
[02:57:55] <jbk> what version of sun's compiler are you using?
[02:58:10] <dlg> cc: Sun C 5.8 2005/10/13
[02:58:23] <jbk> and just to be sure, what does 'which cc' show?
[02:58:31] <dlg> /opt/SUNWspro/bin/cc
[02:58:46] <jbk> ok.. it should work then
[02:58:55] <jbk> unless i'm brain farting on something
[02:59:04] <dlg> no, it looks right to me
[03:02:00] *** yippi has joined #opensolaris
[03:03:51] <gisburn> !seen kupfer
[03:03:53] <Drone> kupfer (kupfer!i=kupfer@nat/sun/x-4f49b394ec6833e8) was last seen in #opensolaris on Wed 08 Nov 2006 23:13 GMT, saying 'hi Rich.  thanks for forwarding me the tools list.'.
[03:04:00] <gisburn> grumpf
[03:05:43] *** yippi has quit IRC
[03:10:21] *** swoolley has joined #opensolaris
[03:11:20] <swoolley> stevel: congrats on gpl for java -- just read about gpl for opensolaris (possibly) !!!
[03:11:33] *** yippi has joined #opensolaris
[03:11:46] <yippi> is srirama around?
[03:11:51] <boyd> swoolley: You read what?
[03:12:00] <swoolley> boyd: /.
[03:12:08] <boyd> Ah... must be true then :)
[03:12:12] <swoolley> :)
[03:12:26] <dlg> jbk: i had a compiler opt that was make it upset it seems
[03:12:31] <jbk> ahh
[03:12:32] <swoolley> don't you implicitly trust /.?
[03:12:32] <jbk> what option?
[03:12:45] <jamesd> boyd, some clueless manager said they would look at making opensolaris  GPL to get reporters off there a$$
[03:12:52] <boyd> swoolley: Trust it to be what? :)
[03:13:35] <jbk> assuming that turn's off c99 support (i don't know all the flags off the top of my head) that could definately do it
[03:13:42] <swoolley> jamesd: supposedly prompted by jschwartz himself
[03:13:53] <swoolley> boyd: fill in the blanks ;)
[03:14:08] <swoolley> whatever you want it to be
[03:14:10] <boyd> indeed... they have mentioned that before actually
[03:14:24] <jamesd> swoolley, it can't be done because sun can't releasee all of the modules under the gpl, so  that makes it imposible.
[03:14:57] <swoolley> jamesd: I've been told as much, so I thought you guys were getting closer to getting the old modules out, or at least removed from the build.
[03:15:22] <swoolley> I understand some of those modules are needed for certain enterprise customers, though
[03:15:32] <dlg> hrm
[03:15:44] <jamesd> swoolley, there are also sun partners and people embeding solaris that can't use a gpl'd OS thus sun would be hurting its customers to make it GPL.
[03:15:46] <dlg> now that im attaching disks to my driver, it wont let me unload it
[03:15:51] <boyd> There are third party drivers for sun-branded hardware, e.g.
[03:15:55] <richlowe> jamesd: they could dual license.
[03:15:56] <dlg> how do i detach the disks first?
[03:16:26] <alanc> parts of Solaris are already GPL'ed - more could be
[03:16:31] <swoolley> oh, I seem to have begun a debate (evidenced by the others chiming in)
[03:16:38] <alanc> but not yet all
[03:16:48] <dlg> alanc: i think thast usually a case of solaris integrating software that is gpl
[03:17:04] <dlg> rather than sun writing gpl code and releasing it
[03:17:31] <alanc> OpenOffice and now Java being the 2 main counter examples
[03:17:59] <alanc> though significant parts of some GNOME apps & libraries were written by Sun and released as GPL/LGPL
[03:18:16] <dlg> hrm
[03:18:23] <swoolley> Well, if linus keeps it at gplv2 and doesn't upgrade, I'll be looking to ditch it if it becomes a liability to still have gplv2 code around
[03:18:31] <dlg> i guess i spend too much time looking at system code
[03:18:31] <stevel> swoolley: thanks :)
[03:18:43] <alanc> no clue if more of it actually will be GPL'ed though - if that's under discussion, it's still at a level way above my pay grade 8-)
[03:18:59] <swoolley> stevel: played ultimate with will on sunday.
[03:19:05] <stevel> alanc: mine too... but that's like 90% of the company ;-)
[03:19:25] <stevel> swoolley: how's the new job?
[03:19:45] <swoolley> stevel: much better
[03:20:07] <swoolley> will seems happy in his too
[03:21:05] <boyd> delewis: ping
[03:21:14] <Kronuz> hey, what's a good SAS controller card with RAID5 support?
[03:21:15] <swoolley> it's not really stretching my skills, but they needed something fast, and I seem to be able to fill in the gaps they didn't have the expertise to cover
[03:21:28] <Kronuz> 'cause the T2000 doesn't support RAID5 (only 0 and 1)
[03:21:59] <Kronuz> and I was thinking if I could add a good disk controller with RAID5 support that would be it
[03:22:15] <Kronuz> I'd just need it for four drives on RAID5
[03:22:15] <jamesd> t2000 only has 4 drives... and  1 or 2 of them should be made for booting and root.. so raid 5 is not a possible.
[03:22:32] <Kronuz> you only need three drives for RAID5
[03:22:45] <swoolley> stevel: on the bright side though, I might go back to using java now that it's gpl.
[03:22:48] <Kronuz> and with four it should be more than enough
[03:22:49] <jamesd> Kronuz, right but what good is raid5 for data if your / drive fails.
[03:23:13] <Kronuz> jamesd: what do you mean?
[03:23:21] <swoolley> afaict, sun is now the only company that "gets" open source.
[03:23:22] <Kronuz> RAID5 is by definition redundant
[03:23:33] <Kronuz> stripped and redundant
[03:23:36] <jamesd> Kronuz, you allocate one drive for / ... and 3 for raid5...  if / drive fails the system is down.
[03:23:41] <swoolley> big company, that is
[03:23:42] <boyd> swoolley: No no... IBM get that they should claim to get it
[03:23:57] <Kronuz> jamesd: oh, I see what you mean
[03:24:07] <swoolley> heh, IBM isn't releasing the crown jewels
[03:24:16] <boyd> That's my point
[03:24:26] <jamesd> Kronuz, and even the most advanced raid5 controllers aren't usable for booting.
[03:24:47] <Kronuz> yeah, that's going to be a problem I hadn't thought about :(
[03:24:52] <stevel> swoolley: we definitely are :)
[03:25:10] <Kronuz> damn...
[03:25:24] <swoolley> yeah, sun seems to have the openness down to the rank and file
[03:25:28] <swoolley> and up to the top
[03:25:29] <Kronuz> I'll boot from a USB
[03:25:30] <Kronuz> lol
[03:25:31] <jamesd> Kronuz, really to make a t2000 happy your going to have to connect to external storage... 32 threads  is very data hungry.
[03:25:53] <Kronuz> jamesd: yeah, but I was hoping to manage without it for some time
[03:26:08] <dlg> jamesd: i boot from raid5s all the time
[03:26:09] <Kronuz> (to get the 32GB version of it)
[03:26:10] <swoolley> stevel: when was the last time you talked to will or done an ultimate game with him?
[03:26:23] <jamesd> jf you can live with  about 300GB you can live for a while thanks to  146GB sas drives.
[03:26:37] *** yippi has quit IRC
[03:26:48] <dlg> how can i reload a module for a scsi driver?
[03:27:05] <Kronuz> dlg: oh, you do? what controller card do you have?
[03:27:18] <dlg> megaraids
[03:27:21] <twincest> a T2000 should be good platform for a program with many independant threads which spend most time moving network data around, right? (little computation involved)
[03:27:21] <Kronuz> jamesd: yeah, that's what I said... I'll have less than 300GB at the start
[03:27:26] <dlg> unless you're talking about booting a sparc off one
[03:27:47] <Kronuz> dlg: ??
[03:27:56] <Kronuz> dlg: what do you mean?
[03:28:05] <stevel> swoolley: it's been a while, i think at his last newark game
[03:28:13] <dlg> i dont know of a raid5 controller that has fcode on it
[03:28:35] <Kronuz> dlg: I have no idea what fcode is :S
[03:28:43] <dlg> Kronuz: what machine do you want raid5 in?
[03:28:49] *** yippi has joined #opensolaris
[03:28:52] <twincest> kronuz: openprom needs fcode for a device it will boot from
[03:29:05] <Kronuz> dlg: it would be the T2000 w/UltraSPARC T1
[03:29:10] <swoolley> stevel: you still doing the newark games?
[03:29:23] <dlg> Kronuz: tough luck then
[03:29:23] <twincest> you can't put a random PC HBA in and boot from it
[03:29:31] <dlg> Kronuz: i dont know why you'd want to do that though
[03:29:48] <swoolley> I've been biking nine miles a day and doing amtrak for 40 miles.
[03:30:04] <Kronuz> dlg: I just wanted to avoid buying external storage at the moment
[03:30:10] <swoolley> oh, 40 each way, so 80 miles
[03:30:25] <Kronuz> seems like I'll have to go for the 1.0Ghz and get an external array
[03:30:31] <dlg> Kronuz: you cant fit much into a t2000 anyway
[03:31:05] <Kronuz> yeah, I know... I know I'd have to eventually get external storage, but for now I was thinking of just buying the T2000 servers
[03:31:13] <jbk> hmm
[03:31:15] <Kronuz> (the big ones)
[03:31:22] <Kronuz> and then later add a SAN or something
[03:31:32] <Kronuz> budget is the problem
[03:31:34] <Kronuz> :P
[03:31:44] <stevel> swoolley: sporadically, every friday or so
[03:32:05] <dlg> so noone knows how i can unload a module for a hba driver?
[03:32:39] <jamesd> dlg, man modunload
[03:32:54] <jamesd> man modload,  man modinfo
[03:34:21] <dlg> that looks like its about rereading config
[03:34:40] <Kronuz> hmm
[03:34:45] <Kronuz> fcode is just the firmware?
[03:35:15] <jbk> no, it'll unload the driver if the driver will allow it
[03:35:18] <dlg> dlg@vavaea mfi$ modinfo |grep mfi
[03:35:18] <dlg> 208 fffffffffba74e70   3518 203   1  mfi (LSI Logic MegaRAID SAS)
[03:35:18] <dlg> dlg@vavaea mfi$ sudo rem_drv mfi
[03:35:18] <dlg> Driver (mfi) not installed.
[03:35:18] <dlg> dlg@vavaea mfi$ sudo modunload -i mfi
[03:35:19] <dlg> Invalid id mfi
[03:36:08] <jbk> try modunload -i 208
[03:36:23] *** nwf has quit IRC
[03:36:26] <dlg> dlg@vavaea mfi$ sudo modunload -i 208
[03:36:26] <dlg> can't unload the module: Device busy
[03:36:32] <dlg> how od i make it unbusy?
[03:37:22] <jamesd> umount all filesystems in use on it,  unload any modules that depend/lock it into memory
[03:37:47] <dlg> no fses yet
[03:37:53] <dlg> i havent got the io path right
[03:37:56] <dlg> but sd is attached to it
[03:38:14] <dlg> how can i detach sd from my hba?
[03:38:16] <jamesd> sounds like you may have to reboot to get rid of it.
[03:38:23] <dlg> damnit
[03:38:29] <dlg> these boxes take forever to cycle
[03:39:15] <jamesd> well there is always  deep magic involving  kdb   but i have no clue how that would work
[03:39:38] <dlg> that sounds too hard
[03:40:20] <swoolley> stevel: well, will and I first went to this sunday 10:30 game at shepard canyon park, in montclair
[03:40:35] <swoolley> level of play was better than newark
[03:40:45] <swoolley> (that's not saying much)
[03:41:25] <swoolley> been three months though since my last game and my legs are dying
[03:42:55] *** solarisjon has left #opensolaris
[03:42:59] *** Yamazaki-kun_ is now known as Yamazaki-kun
[03:43:36] *** nwf has joined #opensolaris
[03:45:37] *** yippi has quit IRC
[03:46:16] <stevel> swoolley: EBWL started registration, you should signup?
[03:46:19] *** yippi has joined #opensolaris
[03:47:03] *** migman has quit IRC
[03:47:06] *** stevel has quit IRC
[03:51:35] *** astinus has quit IRC
[03:51:44] *** rodrickbrown has joined #opensolaris
[03:53:47] *** yippi has quit IRC
[03:54:19] *** yippi has joined #opensolaris
[04:08:49] *** peteh has quit IRC
[04:10:10] *** peteh has joined #opensolaris
[04:12:44] *** Kronuz has quit IRC
[04:17:35] *** Kronuz has joined #OpenSolaris
[04:18:46] <Kronuz> hey, if I add FC to a server, and I want to have redundant connections, should I use a dual FC HBA or two single FC HBA per server? what do you think?
[04:19:00] <Kronuz> I was thinking two would be better but I'm not sure...
[04:21:14] <jbk> well
[04:21:17] <jbk> depends
[04:21:37] <jbk> i mean, if the cards aren't hot pluggable (or at least if the server doesn't support it)
[04:21:52] <jbk> if you have a spare lying around, is it that much of a difference?
[04:21:53] <Kronuz> hmm
[04:22:25] <Kronuz> yeah, I suppose a single one with dual connections would be better
[04:22:38] <jbk> i mean if the card dies, either way, you're taking an outage
[04:22:50] <Kronuz> but two single and one dual are just 100 different in price
[04:23:06] <Kronuz> and if one brakes down, I'd just have to change that single one (half the price)
[04:23:13] <Kronuz> I guess
[04:23:14] <jbk> so how much does an outage cost?
[04:23:21] <jbk> no hardware contract?
[04:23:26] <Kronuz> what do you mean?
[04:23:46] <jbk> what type of server are you looking at?
[04:23:55] <Kronuz> oh, the Sun T1000
[04:23:56] <jbk> whitebox? something from sun? dell? hp?
[04:23:58] <jbk> ahh
[04:24:05] <jbk> are you planning to get a hardware support contract?
[04:24:20] <Kronuz> not sure yet... are those expensive?
[04:24:30] <jbk> i think they even quote them on the site
[04:24:35] <Kronuz> also, are they really worth it?
[04:24:36] <jbk> it depends what you want
[04:24:43] <jbk> well
[04:24:50] <jbk> it's like insurance
[04:24:56] <jbk> if you never have to use it, no
[04:24:57] <jamesd> most places that depend on the hardware think they are.
[04:25:10] <Kronuz> I'm quickly going out of our budget :P
[04:25:11] <jbk> but sometimes all it takes is one or two failures (even minor)
[04:25:22] <jbk> but there are options
[04:25:28] <jbk> depending on what sort of response you want, etc.
[04:25:39] <jamesd> if hardware failure costs $1,000 a minute, the coverage is well worth  it...
[04:25:44] <jbk> you can get 2 hour response and all sorts of frills
[04:25:54] <Kronuz> yeah, how quick
[04:25:54] <jbk> you can get next day response as well
[04:26:15] <Kronuz> that I'd get... we're getting everything redundant to go for the next day
[04:26:46] <jbk> (that doesn't necessairly mean they'll sit and wait exactly 24 hours before delivering, but it does impact their priority in relation to any other calls)
[04:26:48] <Kronuz> unless major shit happens we wouldn't be needing faster service at any given time
[04:27:17] <jbk> so maybe, if the use of two slots doesn't present an issue (in terms of future expandability)
[04:27:20] <jbk> get two cards
[04:27:28] <jbk> if one dies, you can pop it out while you wait for a replacement
[04:27:44] <Kronuz> we'll always have sufficent power to run without one of anything
[04:27:47] <jbk> (which if you use sun hbas, they should be covered by the hardware contract i.e. no additional cost to replace)
[04:27:54] <Kronuz> (exept for the SAN ^_^ atm)
[04:28:03] <Kronuz> if the SAN dies, I get fired
[04:28:08] <Kronuz> hehe
[04:28:10] <kleppari> heheh
[04:28:25] <kleppari> what kind of SAN?
[04:28:30] <Kronuz> it's just that it's too darn expensive to have two of those
[04:28:41] <Kronuz> Sun StorageTek 6140
[04:28:46] <jbk> though I would of course be sure to explicitly ask about that, but i believe that is the normal policy for any sun branded addons put into a sun server under contract
[04:29:04] <kleppari> Kronuz, you should have had a support contract on that ;-)
[04:29:07] <jbk> well it all comes down to how much downtime you can tolerate
[04:29:43] <Kronuz> yeah, well it says 3 Years on site
[04:29:48] <Kronuz> warranty
[04:29:53] <jbk> at my last place of employment, the owner of the company liked to try to squeeze blood from a turnip
[04:30:03] <jbk> and didn't buy any hardware contracts for the servers we had
[04:30:26] <jbk> well for some other cheapness, batteries powering datacenter drained
[04:30:48] <jbk> in the process blew the power supplies on one of the E3000s we had
[04:30:57] <Kronuz> O_O
[04:31:29] <jbk> the cost to have sun replace it was as much as it would have been to put the box under contract
[04:31:51] <Kronuz> but it could have been worse
[04:32:04] <jbk> yes, much worse
[04:32:05] <Kronuz> and then that would have meant trouble, I guess
[04:32:27] <Kronuz> I'll have to look into those hardware contracts
[04:32:50] <Kronuz> it's just that the SAN is so expensive
[04:32:53] <jbk> but i've often found when someone says they can tolerate a large amount of downtime (trying to be cheap), it's usually up until there's an actual outage :)
[04:33:00] <Kronuz> and not to mention the switches and HBAs
[04:33:05] <jbk> it is
[04:33:21] <jbk> of course do you need that level of performance vs. say NAS?
[04:33:23] <Kronuz> the cheap one uses SATA :(
[04:33:52] <jbk> I mean supposedly NFS4 over gige is supposed to not be too shabby (though I can't vouch for it myself)
[04:33:53] <Kronuz> jbk: well, I was just figuring, if later we need more power we'll have to throw the NAS and get a SAN
[04:34:02] <Kronuz> so I was thinking of paying for it upfront
[04:34:10] <Kronuz> (still it's expensive)
[04:34:35] <jbk> i'm somewhat spoiled at the moment :)
[04:34:50] <Kronuz> if it's a NAS which one would tou recommend?
[04:34:52] *** rodrickbrown has quit IRC
[04:34:53] <Kronuz> you*
[04:35:17] <jbk> the project i'm on, the hardware costs are dwarfed by the software costs, so it's been easy to get little things that were overlooked
[04:35:26] <Kronuz> 'cause the SAN it's 48 grands just with 5 146 drives
[04:35:51] <Kronuz> jbk: oh, you got oracle ? ;)
[04:35:52] <Kronuz> :P
[04:35:57] <jbk> haha
[04:36:00] <jbk> no comment :)
[04:36:01] <kleppari> hahah
[04:36:12] <Kronuz> ROTFL
[04:36:42] <Kronuz> honestly, I was checking the license costs
[04:36:50] <Kronuz> $40,000 per CPU!
[04:37:02] <Kronuz> and cores count as .75 CPUs
[04:37:15] <kleppari> yep
[04:37:16] <jbk> look at some of the addons :)
[04:37:35] <jbk> like RAC
[04:37:38] <Kronuz> so a T1000 would be $240,000 just for Oracle
[04:37:44] <Kronuz> not the optional RAC, for instance
[04:37:58] <Kronuz> (and for that single T1000 with 8 cores)
[04:38:07] <Kronuz> man that's crazy!
[04:38:23] <Kronuz> they should just make OpenOracle available
[04:38:28] <Kronuz> ^_^
[04:38:41] <jbk> why i don't get why some places go cheap on the hardware *cough*dell+linux*cough* then run oracle on it...
[04:38:59] <Kronuz> lol
[04:39:06] <jbk> kinda the whole $1000 computer protecting a $10 surge protector
[04:39:07] <Kronuz> it's a joke
[04:39:18] <kleppari> heh, yeah
[04:39:23] <edp> sun+solaris is similar in cost to dell+linux nowadays
[04:39:38] <Kronuz> edp: it is, really
[04:39:52] <Kronuz> I was even afraid to look at Sun (for the prices)
[04:40:16] <edp> the dells look cheaper at first until you bump them up to 8/16/32 gb ram like the suns
[04:40:20] <Kronuz> but as it turns out it's even cheaper than DELL
[04:40:20] *** ajchac1 has joined #opensolaris
[04:40:23] <jbk> yes, but sun's sales and marketing sucks unfortunately
[04:40:38] <jbk> if you're not a fortune 500 company, you don't exist
[04:40:51] <Kronuz> yeah, I've realized ;)
[04:40:52] <jbk> (i *really* wish they'd do something about that)
[04:40:57] <Kronuz> sales really really suck
[04:41:12] <Kronuz> sucks big, BIG time
[04:41:29] <Kronuz> well, all I can hope support is better
[04:42:09] <edp> i've been happy with their sales so far
[04:42:26] <jbk> i thankfully don't have to deal with sales, but i know people that do at smaller companies
[04:42:29] <edp> though i haven't dealt with them a ton
[04:42:54] <Kronuz> edp: in the US it might not be as bad as in other countries, and still is not nowhere near as good as DELL's or IBM's for instance
[04:43:36] <jbk> though in their defense
[04:43:56] <Kronuz> in other countries it just is for all practical accounts non-existent (sales and marketing service)
[04:44:08] <jbk> someone did bring this up with someone at sun, and as an example, they used the edmonton, alberta sun office
[04:44:17] <jbk> including the intern, there was like 14 employees there
[04:44:21] <jbk> IBM had like 700
[04:44:50] <jbk> so it does put them at a bit of a disadvantage
[04:45:05] <Kronuz> yeah, and you can tell, just by calling they don't have much personnel
[04:45:25] <edp> is the support good at least?  like if you get 4 hr support they actually respond in 4 hours?
[04:45:38] <Kronuz> I suppose as long as the technical support and service responce is good sales is not really as important
[04:46:05] <jbk> edp: that is part of the contract they sign
[04:46:30] <jbk> i've generally not had any issues
[04:46:39] <Kronuz> edp: and if not, you get it for free (like a pizza)
[04:46:42] <jbk> though you do sometimes hit a snag here or there
[04:46:55] <jbk> i mean, i one time had a 280 lock up hard
[04:47:08] <jbk> given that we had several hundred 280's
[04:47:15] *** avr_ has quit IRC
[04:47:26] <jbk> i can tell you without even looking if it's locking up hard (can't break, etc.)
[04:47:33] <jbk> your best bet is 1) memory, 2) cpu
[04:47:37] <Kronuz> several hundreds!? O_O
[04:47:44] <Kronuz> where do you work?
[04:48:05] <jbk> a large company :)
[04:48:08] <Kronuz> hehe
[04:48:12] <Kronuz> you bet
[04:48:16] <Kronuz> several hundreds!
[04:48:26] <kleppari> this is starting to sound like an X-files episode..
[04:48:28] <jbk> but the night guy i was talking to wouldn't dispatch a hw guy out with both
[04:48:31] <kleppari> are you one of them?
[04:48:48] <jbk> <_<
[04:48:52] <jbk> >_>
[04:48:53] <jbk> no
[04:48:54] <jbk> :P
[04:48:57] <Kronuz> lol
[04:49:16] <edp> i love it when you know exactly what's wrong but can't convince support of it
[04:49:40] <jbk> but that's not the typical case
[04:49:52] <Kronuz> really, what a controversy, I hate that position
[04:50:03] <Kronuz> :P
[04:50:07] <jbk> usually, send them core file, explorer
[04:50:14] <edp> in most of the sun x64 systems it doesn't seem like there's that many parts that could even break
[04:50:17] <jbk> and they'll send a guy out with a replacement part
[04:50:42] <Kronuz> core file?
[04:50:45] <jbk> or if it's a software issue, either tell you a patch that should resolve it, or lodge a bug that you can track
[04:50:49] <jbk> crash dump
[04:50:54] <Kronuz> oh
[04:51:05] <Kronuz> can one actually do something with those things?
[04:51:08] <jbk> yes
[04:51:09] <Kronuz> <.<
[04:51:10] <Kronuz> >.>
[04:51:31] <kleppari> lol :)
[04:51:36] <kleppari> what did you think they were for?
[04:51:47] <edp> i thought they were just to waste bandwidth :)
[04:51:53] <Kronuz> dunno
[04:51:55] <Kronuz> my ignorance is abysmal :(
[04:52:09] <jbk> i mean, if you want to learn, you could do the analysis yourself using mdb
[04:52:24] <jbk> but they already have lots of people that are much better at it
[04:52:41] <jbk> and probably have some handy tools they've developed to assist
[04:52:53] <Kronuz> I've read a lot about mdb... is that really much better than gdb?
[04:53:04] <Kronuz> (not that I've used gdb that much)
[04:53:29] <jbk> i think the adb compatibility makes it a bit more difficult to use
[04:53:37] <jbk> and i don't believe it has any source browising capability
[04:53:47] <jbk> but it has some pretty neat features
[04:54:06] <Kronuz> adb, gdb, mdb, what's next? zdb? :P
[04:54:08] <jbk> and sun liked it enough to have it replace adb & crash
[04:54:15] <jbk> well
[04:54:26] <jbk> not sure what adb stood for (assembly debugger?)
[04:54:31] <jbk> gdb == gnu debugger
[04:54:56] <jbk> mdb == modular debugger (or mike's debugger as i been told it was called before marketing got involved)
[04:54:57] <Kronuz> (you'll have to forgive my ignorance, as I said I'm new to *nix and even newer to Solaris)
[04:55:27] *** yippi has quit IRC
[04:55:28] <Kronuz> but I'm a Windows guru...
[04:55:33] <Kronuz> well, not really... that either
[04:55:38] <Kronuz> :-(
[04:55:47] <Kronuz> :P
[04:56:21] <jbk> but one of the other advantages of the support contracts is that you get people that can do all the mdb stuff for you :)
[04:56:26] <Kronuz> but I'm good playing videogames
[04:56:47] <Kronuz> jbk: I definitely have to look closer to those contracts
[04:57:04] <jbk> but they can often tell you the reason for a crash
[04:57:21] <jbk> though sometimes if it's a hardware issue, it can be limited by how much information the hardware provides
[04:57:36] <edp> in my dealings with MS support, most answers tend to be "we know it's broken but we're not going to fix it" or "it's that way by design"
[04:57:48] <Kronuz> lol
[04:57:56] <Kronuz> don't you just love MS?
[04:58:02] <jbk> if it's a software issue, you can have a bug filed
[04:58:03] <edp> broken in that case meaning bluescreening a server every day
[04:58:12] <Kronuz> hehe
[04:58:19] <jbk> and it's of course assigned a priority and severity
[04:58:34] <jbk> and also to some degree who's asking for it to be fixed
[04:59:02] <Kronuz> most blue screens I've seen are related to a hardware issue (these days, it wasn't like that on the Win95 days)
[04:59:23] <jbk> that's what most of the crashes i see..
[04:59:34] <jbk> typically ram followed by cpu
[04:59:35] <Kronuz> still I wouldn't use Windows for a server
[04:59:37] <Kronuz> no way
[04:59:50] <edp> mine was a bug in the services for macintosh software they ship with the OS.. the bug was in NT4 and 2K
[05:00:02] <jbk> just due to # of components
[05:00:36] <Kronuz> jbk: what about the NAS you were saying?
[05:00:47] <Kronuz> jbk: in case I can't buy the SAN...
[05:00:55] <Kronuz> http://store.sun.com/CMTemplate/CEServlet?process=SunStore&cmdViewProduct_CP&catid=153482
[05:01:00] <jbk> i don't really have a whole lot of experience there.. just what i've read
[05:01:17] <Kronuz> (that's the 6140 I told you about)
[05:01:43] <Kronuz> I wonder if the four systems there are all 6140
[05:02:02] <Kronuz> why the first one says it has just two FC cables to host
[05:02:03] <jbk> i beleive so, just different configurations
[05:02:17] <Kronuz> and what's that cache per array
[05:02:42] <Kronuz> do you think it's a drive thing or something different about the SAN
[05:02:51] <Kronuz> (even if they're the same model)
[05:03:17] <jbk> well the first two have different numbers of drives & cache
[05:03:34] <jbk> the last two have more cache as well plus faster drives
[05:03:36] <Kronuz> 'cause I could probably get the cheap one easier and later add the drives... if after adding drives I could end up with the "expensive" solution at a later time
[05:04:00] <jbk> you'd have to ask your local rep, but i suspect you can upgrade
[05:04:01] <Kronuz> that cache is probably a SAN thing, right?
[05:04:09] <jbk> well it's with any disk array
[05:04:37] <jbk> they tend to have cache in front of the drives to increase performance
[05:04:42] <Kronuz> I could probably get by with SATA drives for start (for a few months)
[05:05:06] <jbk> though i would find out what the upgrade path is
[05:05:21] <Kronuz> jbk: the thing I've realized with Sun is they don't like to do upgrades (it seems)
[05:05:32] <jbk> for example to go from 2gb cache->4gb, can you just add 2gb, or do you have to replace all the ram with larger dimms?
[05:05:39] *** nwf has quit IRC
[05:05:46] <jbk> i think it depends
[05:06:01] <jbk> in some ways they make it a lot easier than other vendors
[05:06:16] <Kronuz> I asked them if I get the T1000 with 16 GB I could later upgrade to 32GB and they said no :P
[05:06:18] *** nwf has joined #opensolaris
[05:06:28] <jbk> umm
[05:06:33] <jbk> i think that's incorrect
[05:06:43] <edp> I know with other vendors you have to pay a premium for that upgrade path
[05:06:44] <jbk> though, it may be the case that you might have to replace all the dimms
[05:06:49] <Kronuz> which it probably means I'll have to change all the DIMMs, but that's not what they said
[05:06:54] <jbk> and can't just drop in another 16gb
[05:07:12] <Kronuz> they said "no, you can't"
[05:07:17] <Kronuz> but we all know how sales suck
[05:07:19] <Kronuz> ^_^
[05:07:41] <Kronuz> also are those DIMMs special Sun branded DIMMs?
[05:07:47] <Kronuz> or any DIMM would do?
[05:07:56] <jbk> if you can get it with either size, then you should be able to upgrade
[05:08:17] <Kronuz> ('cause HP and IBM used to be like that... only their brand)
[05:08:31] <jbk> well yes and no
[05:08:40] <jbk> sun will want you to use their ram to stay under contract
[05:08:50] <Kronuz> oh
[05:08:53] <jbk> though you can often get 3rd party ram that is compatible
[05:08:59] <jbk> but ymmv
[05:09:04] <Kronuz> and their 3000 FC drives
[05:09:12] <Kronuz> even when I can get those for 600
[05:09:16] <Kronuz> I guess
[05:09:20] <jbk> one of the nice thing w/ the sparc stuff at least
[05:09:35] <jbk> is all the frus on the new stuff have electronically readable serial numbers
[05:09:53] <jbk> so say, samsung discovers that a certain batch of ram was bad
[05:10:15] <Kronuz> oh, they tell you right away?
[05:10:18] <jbk> what'll eventually happen is if this was ram that sun sold, there'll be a fco
[05:10:28] <jbk> and
[05:10:40] <jbk> basically if you send explorers to sun periodically
[05:10:40] <kleppari> then you get planned downtime..
[05:10:42] <jbk> w/ fru info
[05:11:01] <jbk> basically they can say 'dimms x,y,z on these servers are bad and hsould be replaced'
[05:11:02] *** ircminer03 has quit IRC
[05:11:24] <edp> does sun's x64 hardware not have those serial numbers?
[05:11:39] <jbk> hmm i don't know honestly.. haven't messed with them too much
[05:11:45] <jbk> if you have one, try prtfru :)
[05:12:06] <edp> hopefully in a month i'll have one
[05:12:27] <Kronuz> 16 x 146GB @15KRPM FC drives = $42,695 ($2,670 per drive) whereas you can get the same type of drive for as little as $500
[05:12:33] <jbk> i don't know enough about the components to know if it's something that sun does w/ their stuff, or if it's commonly available on the dimms, just maybe not as easy to get the info on other OSes
[05:12:34] <Kronuz> that's what sucks
[05:12:54] <delewis> Kronuz: yes, but are the $500 drives warranted?
[05:12:54] <jbk> yeah you do pay a bit
[05:13:04] <jbk> but one of the other things you get with that
[05:13:05] <delewis> (I would guess not)
[05:13:15] <Kronuz> damn, for 2600 I can get 5
[05:13:19] <jbk> is that sun will guarantee the availability of spares for X years (under contract)
[05:13:23] <Kronuz> how many times do I have to change the drive!
[05:13:32] <jbk> well
[05:13:34] <Kronuz> (even without warranty)
[05:13:37] <edp> Kronuz, i wouldn't want to have to wait for a manufacturer's warning if a drive dies
[05:13:38] <jbk> if they're IBM drives...
[05:13:39] * jbk runs
[05:13:42] <Kronuz> I mean , it's just too much :P
[05:13:51] *** laca has quit IRC
[05:14:05] <Kronuz> jbk: they are IBM... what's with those? :P
[05:14:21] <Kronuz> well, EMC sells those for about $900
[05:14:32] <Kronuz> still a lot cheaper than 2600
[05:17:35] <jbk> well may want to be sure you compare all the costs, emc might be cheaper, but usually they aren't )
[05:17:39] <jbk> :)
[05:17:46] <jbk> my guess is they'll get you on support
[05:18:35] <jbk> but if they do come out cheaper, hey..
[05:18:44] <Kronuz> I hate having to be checking for the price tags but this time shit is just too expensive... hopefully in a c ouple years that won't be a problem anymore
[05:19:10] <jbk> though you might also want to see if maybe sun will knock some off the price for the whole bundle
[05:19:26] <Kronuz> I certainly hope so
[05:20:07] <Kronuz> I'm also applying to the Startup Essentials Offering where they say there'll be "deep, scary-deep, discounts"
[05:20:21] <Kronuz> hehe, we'll see about that
[05:20:22] * Stric just compared a 2-socket opteron/xeon machine w/ 8G ram.. sun vs dell, came out about the same
[05:20:26] <jbk> while i've not heard anything bad about their storage, they don't sell as much as EMC, IBM, or hitachi, so they might be willing to deal a bit to move some storage
[05:21:07] <Kronuz> what about the Hitachi SANs they have
[05:21:15] <Kronuz> nah... if this one costs this...
[05:21:18] <jbk> they just resell them
[05:21:21] <Kronuz> I'de better not even look
[05:21:26] <jbk> the hitachi stuff is more high end :)
[05:21:28] <jbk> heh
[05:21:31] <Kronuz> yeah
[05:21:34] <Kronuz> that's what I mean
[05:21:39] <Kronuz> "high end"
[05:21:45] <Kronuz> ;)
[05:22:04] <jbk> the hitachi stuff is nice
[05:22:15] <Kronuz> those solutions are just too darn "high end" for our budget
[05:22:16] <jbk> though apparently they don't have as much software as EMC does
[05:22:25] <jbk> but what i liked was
[05:23:00] <jbk> hitachi: 'here, come over and I'll show you how to carve up storage on the array.  We can come out and do it, or you can do it yourself, whatever you like'
[05:23:12] *** gisburn has quit IRC
[05:23:13] <jbk> emc: 'dont even look at array or we'll freak'
[05:23:38] <Kronuz> yeah, that's what I noticed at EMC
[05:23:43] <kleppari> What is Sun Startup Essentials?
[05:23:44] <kleppari> Time to market is critical, especially in emerging markets such as Web 2.0, but so is your cashflow!
[05:24:01] <Kronuz> http://sun.com/startupessentials/
[05:24:02] <kleppari> heh, obviously more people than me see another bubble coming
[05:24:12] <Kronuz> <.<
[05:24:13] <Kronuz> >.>
[05:24:18] <jbk> hehe
[05:24:25] <Kronuz> "pretty bubbles in the air!"
[05:24:56] <Error_404> kleppari: I think it's all a big bubble
[05:24:57] <jbk> though it doesn't seem as outrageous as the last one
[05:24:58] <Kronuz> we hope to get out before it bursts ;)
[05:25:03] <jbk> so hopefully the popping won't be as bad
[05:25:07] <Error_404> only this time, i want to explit it
[05:25:10] <Error_404> exploit
[05:25:20] <Error_404> wanna start a web20 company w/ me?
[05:25:29] <jbk> last thing i need is another idiotic CIO coming in and almost fatally crippling the company
[05:26:32] <Kronuz> Error_404: let's do it! I'm starting one right now (and I'll even use my boss's equipment)!!
[05:26:37] <jbk> hehe
[05:26:44] <Kronuz> :P
[05:26:56] <Kronuz> what? it's a start...
[05:26:59] <Kronuz> ^_^
[05:27:14] <kleppari> Kronuz, what are you starting?
[05:27:27] <Kronuz> a web company!
[05:27:32] <Kronuz> ROTFL
[05:27:35] <Kronuz> for the new bubble
[05:27:57] <Kronuz> and, as I said, I'll get out before it bursts... that's the trick ;)
[05:28:06] <Kronuz> to sell it before shit happens
[05:28:13] <kleppari> heh heh
[05:28:21] <Kronuz> hehe
[05:29:02] <Kronuz> did you know about that bunch of companies Amazon bought (the pile of crap they wasted millions on...)
[05:29:22] <Kronuz> lol, well, I'll be one of those companies for the new amazon.com
[05:29:34] <boyd> I don't want to be too much of a downer, but do you, like, have a product?
[05:29:43] <boyd> :)
[05:29:46] <Kronuz> no need!
[05:29:51] <Kronuz> it's the bubble comming!
[05:29:58] <kleppari> then why do you need a SAN?
[05:30:05] <Kronuz> oh, that's my boss
[05:30:14] <jbk> in flight magazine syndrome?
[05:30:16] <Kronuz> and the equipment I'll use :)
[05:31:09] <Kronuz> without his knowledge, hehe... but don't say, it's our evil plan (Error_404 and mine)
[05:31:34] *** sartek has left #opensolaris
[05:32:08] <Kronuz> don't listen to me... I start talking gibberish when I'm tired...
[05:32:28] <Kronuz> :P
[05:32:34] *** bobbyz has quit IRC
[05:32:39] *** ircminer03 has joined #opensolaris
[05:32:42] <kleppari> I'm thinking about selling pet rocks online
[05:32:52] <Kronuz> oh, yeah
[05:32:55] <Error_404> I'm not kidding around actually, I have an idea & a domain name to go with it
[05:33:09] <Error_404> i want to build something & sell it to google as fast as possible
[05:33:18] <Kronuz> kleppari: or pets food! (no wait, that was the idea amazon.com bought)
[05:33:49] *** nwf has quit IRC
[05:33:57] <Kronuz> Error_404: like what?
[05:34:01] <kleppari> A Web 2.0 website typically features a number of the following techniques:
[05:34:02] <kleppari>     * Ajax-based rich Internet application techniques
[05:34:02] <kleppari>     * Non-Ajax-based rich Internet application techniques
[05:34:04] <kleppari> it's perfect.
[05:34:24] <Kronuz> lol
[05:34:26] <Error_404> collaboration + ajax + gradients = web2.0
[05:35:09] <Kronuz> Error_404: I have a few ideas myself, I think we should try to put something together
[05:35:39] *** ajchac1 has left #opensolaris
[05:38:40] <Error_404> yum
[05:38:47] <Error_404> palak paneer
[05:40:55] *** bobbyz has joined #opensolaris
[05:43:59] <dwc-> yum
[05:48:43] <dlg> if i panick a box
[05:49:03] *** spike723_ has quit IRC
[05:49:04] <dlg> how do i fix it?
[05:49:10] *** ircminer03 has quit IRC
[05:49:19] <elektronkind> depends on whhat caused the panic
[05:49:37] <jbk> got the crashdump?
[05:49:38] <richlowe> You should have a crash dump in /var/crash/$(uname -n)
[05:49:43] <dlg> well, a module im working on
[05:49:57] <richlowe> You can also boot with -k to load kmdb
[05:50:01] <Triskelios> geez, ufsrestore is taking ages
[05:50:03] <dlg> so its nothing interesting to you guys
[05:50:15] <dlg> however, it is saying i need to run fsck on /
[05:50:24] <dlg> and all the stuff devfsadm relies on is bad
[05:50:25] *** spike723_ has joined #opensolaris
[05:50:30] *** ircminer03 has joined #opensolaris
[05:51:31] *** slowhog has joined #opensolaris
[05:52:49] *** nwf has joined #opensolaris
[05:58:24] <Kronuz> okay guys, thanks for the help
[05:58:44] <Kronuz> thanks jbk
[05:59:15] <Kronuz> I have to go... I'm tired
[05:59:20] <jbk> np
[05:59:22] <Kronuz> g'night
[06:00:29] *** Kronuz has left #OpenSolaris
[06:01:11] *** Gman is now known as GmanAFK
[06:08:55] *** lloy0076 has joined #opensolaris
[06:10:41] *** alanc is now known as alanc-away
[06:10:42] <Error_404> jesus hell people are greedy
[06:10:43] <Error_404> http://blogs.zdnet.com/Burnette/index.php?p=201
[06:11:08] <lloy0076> Is Gaim the only instant messenger that understands MSN that's easily usable on OpenSolaris ON?
[06:11:13] <Error_404> Sun: "here, have Java under the GPL" Press: "give us solaris"
[06:11:25] <Error_404> lloy0076: aMSN?
[06:11:33] <lloy0076> oops
[06:11:41] <lloy0076> I meant the protocol that Microsoft Messenger uses...
[06:11:56] <Error_404> and i mean "amsn" the application
[06:12:01] <lloy0076> Error_404: Realistically, the GNU world wants everything for free.
[06:12:02] <Error_404> which speaks MSN
[06:12:15] *** whaq has quit IRC
[06:12:16] <lloy0076> Error_404: As in beer and other freedoms as well.
[06:12:18] * lloy0076 sigh
[06:12:42] *** whaq has joined #opensolaris
[06:13:00] <Error_404> I don't actually really understand why they'd want GPL'ed solaris
[06:13:11] <Error_404> so that they can half-implement the features in linux?
[06:13:17] <jbk> i don't get why people don't understand that they can't
[06:13:53] <Error_404> Linux ZFS: "it's like solaris' ZFS, but sometimes it obliterates your data :D "
[06:13:54] <richlowe> jbk: but you have to admit, it must be so pleasant not living in the real world...
[06:14:03] <jbk> i don't think it's a question on if they'd like to
[06:14:52] <lloy0076> I actually wonder if the GPL really matters too much for large projects such as Linux now.
[06:15:13] <lloy0076> Let's say that somehow a huge coalition of Linux copyright holders manage to revoke the GPL on the Linux source code base.
[06:15:22] <jbk> just need to wait until the next big thing his the in flight magazines, and linux will lose its luster :)
[06:15:28] <lloy0076> I suspect everyone would hack it and crack it and go on like that, just as they do with Windows...
[06:15:32] <lloy0076> ...except they'd have the source.
[06:15:39] <jbk> err hits
[06:16:00] <lloy0076> (anyone who claims that there are no pirate copies of Windows is living in la la land)
[06:17:43] <Error_404> it's partly how MS built their empire
[06:18:08] <lloy0076> I know.
[06:18:29] <lloy0076> If MS actually wiped out piracy, they'd probably take themselves out at the same time.
[06:18:48] <lloy0076> Suddenly the "free" operating systems (with less onerous licenses) would become more attractive.
[06:21:15] <lloy0076> [Not to mention the bad press they'd get: I can see Today Tonight (a local crappy current event TV program) showing pictures of 1984 like agents taking away some granny's STOLEN copy of Windows and then blaming Microsoft. It's like people who complain about getting caught speeding. If they weren't speeding, they wouldn't get caught. It's as simple as that, but still it's the POLICEMAN'S fault they were speeding. Huh?]
[06:21:22] * lloy0076 gets off the soap box
[06:22:37] *** cyril_plisko has left #opensolaris
[06:31:02] * lloy0076 gets confused by dual licenses with GPL and other licenses...
[06:41:29] <g4lt-mordant> lloy0076, it is like blaming the cops for speeding issues if the cops are the worst offenders and they only enforce speeding laws on people who cannot properly defend themselves
[06:43:09] <dwc-> Error_404: deletes your data? reiser?
[06:43:22] <dwc-> (and sometimes offs your wife?)
[06:46:09] <Error_404> lol
[06:47:46] *** rachel_ has joined #opensolaris
[06:48:29] *** [1]Cyl has joined #opensolaris
[06:52:39] *** nwf has quit IRC
[06:59:47] *** karrotx is now known as xtorra
[07:06:04] *** Cyl has quit IRC
[07:06:05] *** [1]Cyl is now known as Cyl
[07:07:45] *** karrotx has joined #opensolaris
[07:09:49] *** karrotx has quit IRC
[07:09:58] *** karrotx has joined #opensolaris
[07:10:42] *** bougie has joined #opensolaris
[07:11:25] *** equanimity has joined #opensolaris
[07:15:30] *** equanimity has left #opensolaris
[07:26:35] *** lloy0076 has left #opensolaris
[07:27:01] *** Yamazaki-kun has quit IRC
[07:30:28] *** tga has left #opensolaris
[07:40:14] *** nyati_ has quit IRC
[07:46:02] *** milek has joined #opensolaris
[07:46:15] <milek> hi
[07:46:15] *** jcea has quit IRC
[07:49:58] *** bougie has quit IRC
[07:53:45] *** nyati has joined #opensolaris
[07:57:35] *** whaq|away has joined #opensolaris
[08:00:23] *** whaq has quit IRC
[08:06:16] *** rpaulo has joined #opensolaris
[08:12:28] *** milek has quit IRC
[08:21:20] *** rpaulo_ has quit IRC
[08:25:16] *** mjf has joined #opensolaris
[08:29:52] *** whaq|away has quit IRC
[08:29:57] *** whaq|away has joined #opensolaris
[08:34:54] *** |tsoome| has quit IRC
[08:40:21] *** whaq|off has joined #opensolaris
[08:42:19] *** whaq|away has quit IRC
[08:45:35] *** |tsoome| has joined #opensolaris
[08:47:24] *** whaq|off has quit IRC
[08:52:56] *** timofonic has joined #opensolaris
[08:52:57] <timofonic> Hello
[08:53:02] <timofonic> Are there a OpenSolaris distribution for SPARC with the good classic taste of Solaris 8-9? I mean not linuxified...
[08:55:27] <timofonic> Classic solaris taste but newer stuff
[08:55:28] <trygvis> uh? linux-ified how?
[08:55:28] <timofonic> Full of GNU bullshit, KDE...
[08:55:28] <trygvis> solaris doesn't come with KDE
[08:55:28] <timofonic> But Gnome
[08:55:28] <trygvis> you can choose between sun's java desktop (which is gnome-based) or cde
[08:55:34] <timofonic> But old skool Solaris people prefer Solaris 9...
[08:55:37] <trygvis> and there is no gnu stuff in your PATH by default
[08:55:38] *** Risky has quit IRC
[08:55:48] <trygvis> why would they do that? solaris 10 >> solaris 9
[08:56:13] <Triskelios> timofonic: um, wtf?
[08:56:23] <timofonic> Some old skool Solaris sysadmins say me Solaris 10 is too linux-ified
[08:56:46] <trygvis> that is a pretty lol statement
[08:57:10] <timofonic> One of them is a good friend, worker on telecom networking stuff and a lot years experience in computers and servers/networking
[08:57:41] <timofonic> It can be funny to me, but one of those guys said me that is not a UNIX n00b...
[08:58:06] <trygvis> that is just a silly statement
[08:58:11] <timofonic> Why?
[08:58:30] <timofonic> Why he said that? What do you think
[08:58:46] <trygvis> I just told you what I think ..
[08:58:47] <Triskelios> nobody would agree with him
[08:58:55] <timofonic> He said something like "Solaris 10 is too much linux-ified, the older ones are better"
[08:58:57] <Triskelios> especially no one at sun
[08:59:26] <timofonic> I don't trust on home's answers :)
[08:59:30] <ofu> sol10 is about 25-30% faster than sol8 or 9
[08:59:40] <timofonic> It's like asking Microsoft is Linux is better than Windows...
[08:59:49] <timofonic> is=if
[09:00:08] <timofonic> But what about the environment? It has been linux-ified?
[09:00:21] <trygvis> dude, wtf does linux-ified mean?
[09:00:50] <trygvis> the new (often useful) GNU tools are not in your path by default
[09:01:57] <timofonic> Uhm, I will ask him about that when he has spare time :)
[09:02:46] <timofonic> Solaris 10 is faster than 8 or 9, but even in "old" 64bit SPARC machines?
[09:03:13] <trygvis> I would assume so
[09:03:23] <timofonic> Or now is too centered in the x86 and newer SPARCs world?
[09:04:06] <Triskelios> it should definitely be faster on ultrasparc iiis and such
[09:04:21] <ofu> timofonic: our e420 were about 30% faster on sol10
[09:04:40] <timofonic> I Have two T1 machines, one of them a giant Netra i with raid5
[09:04:44] <ofu> if you do enough tcp/ip stuff
[09:04:54] <richlowe> Hm.
[09:05:11] <timofonic> I'm starting in the SPARC world with "vintage" machines :)
[09:05:34] <timofonic> Cheap, easy to get (for me) and looks nice retro stuff
[09:05:52] <richlowe> I thought the Netra i's were sun4m.
[09:05:52] <timofonic> I'm too por for high-end SPARCs
[09:05:56] <timofonic> por?poor
[09:06:35] <timofonic> richlowe: Not sure, I must look more at the specs on those machines. I'm quite newbie and lately I'm tired all time
[09:07:00] <timofonic> And I have problems powering on those machines, a friend must help me...
[09:07:05] <richlowe> oh, the i 150 wasn't.
[09:07:23] <timofonic> Both ones did run nicely, but the other day stopped to run
[09:07:52] <timofonic> This big Netra i uses SBUS only, 512mbytes of ram...
[09:09:10] *** Cyrille has joined #opensolaris
[09:09:11] <richlowe> I suspect the Netra i's were all sun4m or US-I
[09:09:22] <richlowe> but I can't seem to reach sunsolve, so I maybe misremembering bits :)
[09:10:46] <timofonic> uhm, I will investigate that after my friend helps me on fixing the two machines :P
[09:11:44] <timofonic> One of them locks up shortly after booting and using tbe openbios, the other seems to start but not display output or keyboard leds on/off when pushing block shift...
[09:11:45] <ofu> netra t1 105 ist ultrasparc-IIi
[09:12:08] *** rachel_ has quit IRC
[09:12:43] <timofonic> uhm
[09:12:53] <dlg> i heart the t1 105
[09:13:09] <timofonic> I did read T1 on the boot...
[09:16:01] *** peteh has quit IRC
[09:16:45] <richlowe> Yeah, I was talking about the Netra i referenced above.
[09:17:30] <timofonic> About serial null cable... just an "average" db25 to db9 null modem cable or a custom one?
[09:17:47] <timofonic> I did read about sparc serial pinouts and got scared...
[09:18:04] <rydis> Suns are DCE, PC:s are DTE, typically, so you don't want a null modem cable, just a regular serial cable.
[09:18:46] <Doc> umm.. no
[09:18:49] <timofonic> Can a HDD damage lock a SPARC machine or avoiding to showing output in the display?
[09:18:57] <dlg> anyone familiar with drivers for scsi controllers?
[09:19:05] <timofonic> Doc: No to what? :P
[09:19:06] <rydis> No? Hmm. I must be seriously confused, then.
[09:19:11] <Doc> sun to PC needs x-over
[09:19:24] <timofonic> Me too, and spared (of crashing one of the computers)
[09:19:37] <Doc> http://www.sunhelp.org/unix-serial-port-resources/ is your friend
[09:19:42] <ofu> timofonic: if the scsi bus hangs, especially format takes really long
[09:19:50] <timofonic> Doc: What do you mean?
[09:20:05] <timofonic> Doc: That page confused me even more, too many different sun serial pinouts
[09:20:12] <Doc> no, there are 2
[09:20:29] <timofonic> Doc: Can I damage something if using the wrong one?
[09:20:36] <Doc> (well, 3 if you include the ones that support 2 serial ports over a single DB25, but you can largely ignore them)
[09:20:46] <richlowe> dlg: jmcp would probably be your best bet.
[09:21:19] <dlg> richlowe: he's flying around at the moment, and im suffereing because of it
[09:21:21] *** peteh has joined #opensolaris
[09:21:28] *** slowhog has quit IRC
[09:22:25] <timofonic> Netra X1, 1280, t1 Model 105/AC200/DC200, Sun Fire V100/ V120/ V480/ V1280: [Pin ID] These machines have RJ45 connectors for serial ports. <---- NO, mine has two db25 ones, A and b
[09:25:24] <trygvis> uh? my SF v100 has the RJ45 connector
[09:25:29] <asyd> a netra with db25 ?!
[09:25:56] *** whaq|off has joined #opensolaris
[09:25:57] <timofonic> This Netra i has steel wheels and is more than 50kg... But externally not showed more than the front schematic about parts and the Netra i logo
[09:26:01] <timofonic> asyd: yep
[09:26:06] <rydis> The page mentions several Netra models that have DB25:s, doesn't it?
[09:26:23] <timofonic> I'm afriad of pinouts
[09:26:38] <timofonic> afraid
[09:26:58] <richlowe> the large Netra i would be pretty much an E150.
[09:33:24] <timofonic> Any manner of check it without booting? Now not boots correctly :(
[09:39:38] *** whaq|off is now known as whaq
[09:40:42] <timofonic> "Finally, the E150 is one of the worst bits of kit every shipped by Sun; the inside is mounted in foam blocks (!?!?!!) and if you shut it down (init 5) you can only switch it back on by either (a) opening the unit up or (b) attaching a Sun keyboard and using the 'power' button."
[09:40:47] <timofonic> WTF?
[09:42:28] *** hspaans has quit IRC
[09:44:09] <luxh> hehe
[09:44:24] <timofonic> This is my system http://sunsolve.sun.com/handbook_pub/Systems/E150/E150.html
[09:45:06] <timofonic> What's wrong with E150? It's that kind of rock solid retro design, hehe
[09:46:39] *** alobbs has joined #opensolaris
[09:48:23] <timofonic> What one of them is that? http://www.sunhelp.org/unix-serial-port-resources/serial-pinouts/
[09:48:27] <timofonic> alobbs: Are you Spanish?
[09:49:48] <alobbs> timofonic: yes, I am..
[09:50:07] <alobbs> why? :)
[09:59:10] *** whaq|away has joined #opensolaris
[09:59:20] *** estibi` has joined #opensolaris
[09:59:31] *** whaq has quit IRC
[10:04:32] *** pikapika has quit IRC
[10:10:06] *** GmanAFK has quit IRC
[10:10:06] *** alanc_away has quit IRC
[10:10:52] *** alanc_away has joined #opensolaris
[10:11:03] *** GmanAFK has joined #opensolaris
[10:11:42] *** ienup has joined #opensolaris
[10:12:42] *** ienup has quit IRC
[10:21:09] *** whaq|away is now known as whaq
[10:21:51] *** Burana has joined #opensolaris
[10:25:36] *** y___schil has joined #opensolaris
[10:27:58] *** sil3nt|warri0r has joined #opensolaris
[10:31:05] <timofonic> Are there some photo gallery about "vintage" sparc machines with their names? I need to identify the other machine
[10:31:27] <asyd> maybe sunstuff ?
[10:34:05] <timofonic> too few photos :P
[10:34:08] *** ___schily has quit IRC
[10:35:01] <Doc> bill was setting one up, although not sure if he ever did...try looking around on sunhelp.org
[10:35:24] <timofonic> Corrupt? http://sunstuff.org/hardware/systems/sun4/
[10:35:52] <timofonic> thanks to all for the help
[10:36:03] *** coffman_ has joined #opensolaris
[10:36:30] *** coffman_ has quit IRC
[10:38:25] <timofonic> I like this kind of gallery: http://www.auxio.org/cgi-bin/display
[10:39:14] <timofonic> It can be better, butl ooks interesting
[10:40:32] *** Burana has quit IRC
[10:41:23] <timofonic> SUN Ultra Enterprise 1
[10:41:25] <timofonic> Not found on most galleries :P
[10:41:35] <timofonic> I'm stupid, it's on the chasis :D
[10:48:58] *** dunc has joined #opensolaris
[10:52:35] <Berny> morning
[10:54:33] <estibi`> morning
[10:55:13] <estibi`> does anyone know when the snv_52 will be?
[10:56:24] <Berny> http://www.opensolaris.org/os/community/on/onnv_schedule.txt
[10:56:29] <raph_ael> hello
[10:57:34] <richlowe> SX:CR tends to appear between a week and two weeks after a build is delivered to the WOS.
[10:57:45] <richlowe> I assume a source snapshot will appear when stevel gets time to generate it.
[11:01:07] <GmanAFK> hey hey hey
[11:01:32] <richlowe> Hey GmanAFK.
[11:02:33] <Berny> while talking about when things appear... does anyone have any detail on "punchin"
[11:02:40] <Berny> morning GmanAFK
[11:03:01] *** damienc has joined #opensolaris
[11:03:02] <Berny> .oO(isn't it time for bed on your side of the world? ;-))
[11:03:06] <Berny> morning damienc
[11:11:02] *** calumb has joined #opensolaris
[11:14:30] *** GmanAFK is now known as Gman
[11:14:42] <Gman> Berny, yep, though tuesday nights are usually team meeting nights :)
[11:15:18] <Berny> .oO(meaning a social in the nearest club/pub? ;-))
[11:17:45] *** LordKing has joined #opensolaris
[11:18:16] *** richlowe has quit IRC
[11:24:05] *** kloczek has quit IRC
[11:25:18] *** kloczek has joined #opensolaris
[11:25:46] *** richlowe has joined #opensolaris
[11:26:08] * richlowe mutters about postrun
[11:30:20] <Gman> Berny, funny you should say that
[11:30:29] <Gman> seemingly i've managed to polish off a 12 pack of stubbies :)
[11:30:35] <Gman> richlowe, suffer in your jocks!
[11:31:07] <richlowe> I honestly don't see the benefit of the install/upgrade being faster if it means waiting so long before you can log in afterward.
[11:31:12] <richlowe> it's not like the desktop is *usable* any faster.
[11:31:28] <Gman> if you have any better ideas, i'd love to hear it
[11:31:56] <Gman> we have the sucky situation of having all these post install scripts really slowing down install :(
[11:33:40] <richlowe> Yeah, but I don't get why slowing down install is worse than having it run reasonable, but not let anyone log in while they run.
[11:33:46] <richlowe> Is there any real time saving there at all?
[11:34:02] <Gman> i don't honestly know
[11:34:20] <Gman> but laca's a smart guy - so i doubt he created the whole postrun without a real need
[11:34:26] *** PosixC has joined #opensolaris
[11:35:45] <PosixC> tftp question : I running get myFile.tx from the tftp>prompt ; I do get that file yet in sniffing on port 69 I don't see a thing
[11:35:58] <PosixC> "snoop -v port 69" is my sniffing
[11:36:19] <PosixC> shouldn't it be port 69 fot tftp sniffer by default?
[11:43:38] <Doc> what host are you tftp'ing from?
[11:46:11] *** triplah has joined #opensolaris
[11:48:28] <PosixC> I tried both the ip on the LAN, and 127.0.0.1
[11:52:00] <PosixC> also tried tftp> get zzz.xx and snoop -v | grep zzz.xxx
[11:52:11] <PosixC> yet did not saw a thing in snoop
[11:52:14] <PosixC> any ideas ??
[12:00:36] *** triplah has quit IRC
[12:00:40] <PosixC> strangely , when I tftp and get  from within another client to this ip , it does succeed !
[12:01:13] *** triplah has joined #opensolaris
[12:05:41] <Stric> when you connect to yourself, you go over the loopback and you're sniffing what's coming in on the cable
[12:07:28] *** sparc-kly_ is now known as sparc-kly|WORK
[12:07:30] *** sil3nt|warri0r has quit IRC
[12:15:50] *** coolvibe has joined #opensolaris
[12:16:21] *** OnkelSchorsch has joined #opensolaris
[12:18:40] *** paul has quit IRC
[12:21:24] *** LordKing has quit IRC
[12:27:45] *** mega has joined #opensolaris
[12:29:25] *** tarzeau has joined #opensolaris
[12:29:34] <tarzeau> it really might get GNU GPLd?
[12:32:30] <Gman> you never know :)
[12:32:46] <Gman> though really that was just schwartz putting the shits up rich green ;)
[12:34:01] <Gman> though word on the street
[12:34:02] <Gman> is that apparently
[12:34:08] <Gman> sun took a bribe from microsoft
[12:34:13] <Gman> to make it gpl incompatible
[12:34:16] <Gman> imagine that? ;)
[12:34:22] <Gman> *cough* bullshit *cough*
[12:39:19] <whaq> PosixC, when you put a sniffer on 127.0.0.1... it's as though you're smelling your own behind, no?
[12:41:00] *** Capricorn^80 has joined #opensolaris
[12:41:37] <Capricorn^80> hi need some help . i m buidling grass on solaris 9 n getting this error
[12:41:38] <Capricorn^80> checking for location of JPEG library... /usr/local/lib
[12:41:38] <Capricorn^80> checking for jpeg_start_compress in -ljpeg... no
[12:41:38] <Capricorn^80> checking for jpeg_start_compress in -ljpeg... no
[12:41:38] <Capricorn^80> configure: error: *** Unable to locate JPEG library.
[12:42:38] <dlg> it probably cant find libjpeg
[12:43:43] <Capricorn^80> well i have mentioned  the path of JPEG
[12:43:49] <Capricorn^80> n all other libraries path
[12:44:22] <Capricorn^80> n i have searched but there is no file with name -ljpeg
[12:44:30] <Capricorn^80> wat should i do ?
[12:45:05] *** paul has joined #opensolaris
[12:45:10] <Gman> it'll be libjpeg.so*
[12:45:12] <Berny> maybe it's called libjpeg.so.*
[12:46:28] <Capricorn^80> i got one file  i.e /var/Sun/mps/bin/base/jre/lib/sparc/libjpeg.so
[12:47:13] <Gman> you should have it in /usr/lib
[12:47:34] <Gman> look for SUNWjpg
[12:49:03] <Capricorn^80> .. /var/sadm/pkg/SUNWjpg
[12:49:46] <lasseoe> isn't libjpeg in /usr/sfw/lib ?
[12:49:58] <Berny> pkginfo|grep SUNWjpg does that list that package?
[12:49:59] <Capricorn^80> let me check
[12:50:18] <Berny> ll /usr/lib/libjpeg.so
[12:50:18] <Berny> lrwxrwxrwx   1 root     root          17 Oct 16 17:06 /usr/lib/libjpeg.so -> libjpeg.so.62.0.0
[12:50:33] <richlowe> lasseoe: not in recent Nevada
[12:50:46] <lasseoe> richlowe: ok, good to know :)
[12:51:07] <Berny> even on sol10 it's in /usr/lib
[12:52:03] <lasseoe> hm I never knew that, but I don't use it often
[12:52:24] <Capricorn^80> yes
[12:52:27] <Capricorn^80> its there
[12:53:27] <Capricorn^80> but this library is in my path
[12:53:29] <Capricorn^80> crle -c /var/ld/ld.config -l /usr/lib:/usr/sfw/lib:/usr/sfw/include:/usr/local/lib
[12:53:34] <Capricorn^80> then y its not working
[12:54:03] <andersmo> Capricorn^80: Read config.log - you'll find a lot more details on what configure did and why it failed to find the library there.
[12:54:14] <Capricorn^80> i did that
[12:54:31] <Capricorn^80> its telling me same error about -ljpeg
[12:55:04] <Gman> look at config.log
[12:55:05] <Gman> see what it says
[12:56:18] <Capricorn^80> wait i tell u
[12:56:32] <Capricorn^80> ./configure --with-jpeg-includes=/usr/local/include --with-jpeg-libs=/usr/local/lib --with-jpeg-includes=/usr/sfw/include --with-jpeg-includes=/usr/sfw/lib
[12:56:39] <Capricorn^80> now it giving me the error
[12:56:50] <Capricorn^80> of another header fine
[12:57:05] <Berny> why don't you leave out the --with-jpeg stuff?
[12:57:17] <Capricorn^80> configure:7411: checking for jpeglib.h
[12:57:17] <Capricorn^80> configure:7419: gcc -E  -I/usr/sfw/lib  conftest.c >/dev/null 2>conftest.out
[12:57:17] <Capricorn^80> configure:7415:21: jpeglib.h: No such file or directory
[12:57:17] <Capricorn^80> configure: failed program was:
[12:57:17] <Capricorn^80> #line 7414 "configure"
[12:57:17] <Berny> it should find it ok in /usr/include and /usr/lib
[12:57:18] <Capricorn^80> #include "confdefs.h"
[12:57:20] <Capricorn^80> #include <jpeglib.h>
[12:57:49] <Berny> you're forcing it to look in /usr/sfw/include where it doesn't exist
[12:58:20] <Capricorn^80> wat should i do
[12:58:32] <Berny> just a ./configure
[12:58:33] <Capricorn^80> if i dont force it will not work
[12:58:37] <Capricorn^80> ok
[12:58:41] <Capricorn^80> let me do that
[12:59:47] <Capricorn^80> heres the output
[12:59:48] <Capricorn^80> checking for location of External PROJ.4 data files...
[12:59:48] <Capricorn^80> configure: warning: *** External PROJ.4 directory not specified
[12:59:48] <Capricorn^80> checking for /epsg... no
[12:59:49] <Capricorn^80> configure: warning: *** Unable to locate PROJ data files.
[12:59:49] <Capricorn^80> checking for nad2bin... /usr/local/bin/nad2bin
[12:59:49] <Capricorn^80> checking whether to use Readline... no
[12:59:51] <Capricorn^80> checking whether to use JPEG... yes
[12:59:53] <Capricorn^80> checking for location of JPEG includes...
[12:59:55] <Capricorn^80> checking for jpeglib.h... no
[12:59:58] <Capricorn^80> configure: error: *** Unable to locate JPEG includes.
[13:00:26] <damienc> grep jpeglib.h /var/sadm/install/contents
[13:00:31] <damienc> will tell you where the file is.
[13:01:13] <Berny> .oO(i hope he has SUNWjpg-devel installed)
[13:02:50] <Capricorn^80> its in /usr/sfw/include/jpeglib.h
[13:03:03] <Capricorn^80> thats y i was forcing it
[13:03:33] <Berny> are you sure that file is really there?
[13:03:42] <Berny> .oO(and you have permission to read it?)
[13:03:49] <Capricorn^80> i m pasting u from my terminal
[13:03:56] <Capricorn^80> let me check it permission
[13:04:31] <andersmo> Capricorn^80: Why are you telling it to look for _headers_ in the lib and not the include directory? "--with-jpeg-includes=/usr/sfw/lib"
[13:04:37] <Capricorn^80> i think i dont
[13:04:46] <andersmo> That's what you pasted. =)
[13:06:04] <andersmo> From the configure line you pasted: "--with-jpeg-includes=/usr/sfw/include --with-jpeg-includes=/usr/sfw/lib" - try something like "--with-jpeg-includes=/usr/sfw/include --with-jpeg-libs=/usr/sfw/lib" instead? (assuming those are the parameters expected by configure)
[13:06:41] <Capricorn^80> still getting error
[13:06:43] <Capricorn^80> checking for location of JPEG includes...
[13:06:43] <Capricorn^80> checking for jpeglib.h... no
[13:06:43] <Capricorn^80> configure: error: *** Unable to locate JPEG includes.
[13:06:53] *** paul_ has joined #opensolaris
[13:07:29] <Capricorn^80> if i have added these lib in my path
[13:07:32] <Capricorn^80> i.e
[13:07:44] <Capricorn^80> crle -c /var/ld/ld.config -l /usr/lib:/usr/sfw/lib:/usr/sfw/include:/usr/local/lib
[13:07:46] *** paul has quit IRC
[13:07:50] <Capricorn^80> then y i m getting error
[13:08:51] <Capricorn^80> . /usr/sfw/include is there
[13:09:25] <damienc> is /usr/sfw/include/jpeglib.h?
[13:09:38] <damienc> what did you get with?  grep jpeglib.h /var/sadm/install/contents
[13:09:57] <sickness> i'm back
[13:10:27] <Capricorn^80> this ... /usr/sfw/include/jpeglib.h f none 0644 root bin 46205 19783 1015056292 SUNWjpg
[13:10:46] <andersmo> Capricorn^80: header files (typically in include directories) and libraries (typically in lib directories) are very different beasts, and telling the run-time linker to look for libs in include directories is likely to do you no good. Step slowly and make sure you understand everything before moving on, and you'll finish sooner and learn more. =)
[13:10:55] <damienc> cool, then the --with-jpeg-includes=/usr/sfw/include should work
[13:11:27] <damienc> if it doesn't, then you should look at config.log
[13:11:49] *** miffe has joined #opensolaris
[13:12:09] <Capricorn^80> so tell me in short wat should i do now ?
[13:12:16] <damienc> it will show you the command that failed.
[13:12:39] <damienc> search through config.log looking for jpeglib.h
[13:13:01] <damienc> look at the gcc calls made near that point
[13:13:05] <damienc> and which one failed.
[13:14:24] <regx> hey im getting a failed msg durring install of a BrandZ zone does any one has any EXP on this?
[13:14:47] <regx> gtar: ./usr/X11R6/lib/modules/libcfb32.a: Cannot write: No space left on device
[13:14:48] <regx> gtar: Archive contains obsolescent base-64 headers
[13:14:48] <regx> gtar: Skipping to next header
[13:14:48] <regx> gtar: ./usr/X11R6/lib/modules/liblayer.a: Cannot write: No space left on device
[13:14:49] <regx> gtar: Skipping to next header
[13:14:49] <regx> gtar: A lone zero block at 157311
[13:14:50] <regx> gtar: Error exit delayed from previous errors
[13:14:58] <regx> fuck SORRY
[13:15:03] *** calumb is now known as calAFK
[13:15:09] <damienc> "no space left on device" looks kinda critical
[13:15:10] *** eugene has joined #opensolaris
[13:15:12] <Doc> umm.. No space left on device"
[13:15:15] *** axxl has joined #opensolaris
[13:15:26] <regx> omg
[13:15:28] * andersmo thinks he regx suddenly spotted it himself. =)
[13:15:31] <regx> seriuosly
[13:15:34] <regx> im a fuckin retard
[13:15:37] <regx> sorry for waisting your time
[13:15:43] <andersmo> s/he //
[13:16:14] *** Snake007uk has joined #opensolaris
[13:16:40] <regx> i wonder if i can specify where i want it to install..
[13:17:57] <lasseoe> hm sunsolve still doesn't work
[13:19:00] *** Gman is now known as GmanAFK
[13:25:25] *** OnkelSchorsch has quit IRC
[13:25:34] *** paul_ is now known as paul
[13:25:56] <Capricorn^80> ok
[13:27:13] <Capricorn^80> this is config.log
[13:27:15] <Capricorn^80> configure:7365: checking whether to use JPEG
[13:27:15] <Capricorn^80> configure:7385: checking for location of JPEG includes
[13:27:15] <Capricorn^80> configure:7411: checking for jpeglib.h
[13:27:15] <Capricorn^80> configure:7419: gcc -E   conftest.c >/dev/null 2>conftest.out
[13:27:15] <Capricorn^80> configure:7415:21: jpeglib.h: No such file or directory
[13:27:16] <Capricorn^80> configure: failed program was:
[13:27:17] <Capricorn^80> #line 7414 "configure"
[13:27:19] <Capricorn^80> #include "confdefs.h"
[13:27:21] <Capricorn^80> #include <jpeglib.h>
[13:29:08] *** calAFK is now known as calumb
[13:29:16] <Capricorn^80> <damienc> : i pasted it for u
[13:30:37] <damienc> it is strange that the gcc command line does not have -I/usr/sfw/include
[13:31:03] <Capricorn^80> may be
[13:31:09] <Capricorn^80> as gcc is installed
[13:31:16] <damienc> try compiling a similar file yourself. The file only needs one line - "#include <jpeglib.h>"
[13:31:34] <damienc> cmd line: gcc -E -I/usr/sfw/include yourfile.c
[13:32:31] <damienc> in an earlier example that you posted, there was a -I item on the gcc line.
[13:33:40] *** movement has quit IRC
[13:34:28] *** movement has joined #opensolaris
[13:37:18] *** sil3nt|warri0r has joined #opensolaris
[13:38:59] <Capricorn^80> bash-2.05# gcc -E -I/usr/sfw/include yourfile.c
[13:39:00] <Capricorn^80> gcc: yourfile.c: No such file or directory
[13:39:00] <Capricorn^80> gcc: no input files
[13:40:51] <sil3nt|warri0r> hello folks
[13:41:24] <sil3nt|warri0r> i installed the "entire distribution", but i am not getting the x window
[13:41:28] <damienc> Capricorn^80: did you create a file called yourfile.c with the one line in it?
[13:41:39] <sil3nt|warri0r> every time i reboot i get a CLI login
[13:41:45] <Capricorn^80> no
[13:42:00] <damienc> 10 mins ago I said:
[13:42:05] <Capricorn^80> wat line i should put in that
[13:42:07] <damienc> try compiling a similar file yourself. The file only needs one line - "#include <jpeglib.h>"
[13:42:10] <Capricorn^80> sorry
[13:42:27] <Capricorn^80> ok
[13:42:53] <sil3nt|warri0r> i look at the xorg log file
[13:44:34] <sil3nt|warri0r> but dont get much :(
[13:45:25] <sil3nt|warri0r> how do i start my X server
[13:48:32] <eugene> sil3nt|warri0r: see /etc/inittab
[13:48:38] <eugene> sil3nt|warri0r: is the default runlevel 3 or 5?
[13:49:32] <sil3nt|warri0r> right now i am in linux
[13:49:41] <sil3nt|warri0r> but i gues its 3
[13:49:50] <sil3nt|warri0r> i tried "init 5"
[13:50:05] <sil3nt|warri0r> but it shutdown the box
[13:50:17] <Auralis> default is 3, 5 is shutdown and poweroff
[13:50:22] <sil3nt|warri0r> ahh
[13:50:24] <sil3nt|warri0r> k
[13:50:51] <sil3nt|warri0r> so how i boot with Xwindow login
[13:50:56] <sil3nt|warri0r> i mena the x server
[13:51:43] <Auralis> you just boot, X gets started automaticaly as part of RL2
[13:51:57] <Auralis> if it does not then X has a configuration problem
[13:52:12] <triplah> sil3nt|warri0r: i had major problems getting X to work - i actually dropped sol for the time being because of it
[13:52:21] <triplah> amd 3200+ with nvidia 6600GT
[13:52:42] <triplah> used to spend a bit of time trying to start X, then would drop to tty
[13:52:56] <Auralis> for nvidia cards you want to check on the official nvidia drivers
[13:53:07] <sil3nt|warri0r> i also run the "gdm" command
[13:53:13] <sil3nt|warri0r> but nothing comes up
[13:53:16] <sil3nt|warri0r> and no error
[13:53:36] <triplah> Auralis: yeah  i did that :P
[13:53:43] <triplah> ok bed
[13:53:46] <triplah> *runs*
[13:53:47] <Cyrille> you should enable the gdm service. svcs -a | grep gdm to find its exact name then svcadm enable -s <name>
[13:53:50] <sil3nt|warri0r> just the screen goes black sometime and then back where it was (cli)
[13:54:15] <Cyrille> name should be something like gdm2-login with fluff around
[13:54:16] <Auralis>  /var/dt/Xerrors if you use Xsun, no clue where xorg puts its logs
[13:54:37] <sil3nt|warri0r> /var/log/X.org.log i guess
[13:54:41] <triplah> var/log/Xorg.0.log
[13:54:48] <sil3nt|warri0r> k
[13:54:52] <triplah> something like that :)
[13:54:57] *** tsoome has joined #opensolaris
[13:55:14] <sil3nt|warri0r> where do i find the xorg conf file
[13:55:27] *** damienc is now known as dcFood
[13:55:30] <triplah> /etc/X11/xorg.conf ?
[13:55:41] <sil3nt|warri0r> <Cyrille> its gdm-login
[13:56:24] <Cyrille> ah well, I was close (and it's gdm2-login indeed on Solaris 10)
[13:57:13] <Capricorn^80> <damienc> : i did that
[13:57:35] <Capricorn^80> bash-2.05# gcc -E -I/usr/sfw/include myfile.c
[13:57:35] <Capricorn^80> # 1 "myfile.c"
[13:57:38] <Capricorn^80> # 1 "<built-in>"
[13:57:38] <Capricorn^80> # 1 "<command line>"
[13:57:38] <Capricorn^80> # 1 "myfile.c"
[13:57:38] <Capricorn^80> "#include <jpeglib.h>"
[13:57:38] <Capricorn^80> bash-2.05#
[13:58:03] <Cyrille> seems pretty empty...
[13:59:02] <Capricorn^80> Cyrille : u talking to me ?
[13:59:16] <Cyrille> did you actually put quotes around #include <jpeglib.h>?
[13:59:21] <Cyrille> Capricorn^80: yes
[13:59:28] <Capricorn^80> yes
[13:59:33] <Cyrille> don't
[13:59:53] <Capricorn^80> u mean quotes should be there in file ?
[13:59:57] <Cyrille> no
[14:00:17] <Cyrille> I mean the file should just contain #include <jpeglib.h>
[14:00:25] <Cyrille> with no quotes or nothing
[14:00:35] <Capricorn^80> ok
[14:01:06] <Capricorn^80> i did that .. got some long programming code
[14:01:29] <Capricorn^80> just pasting last few lins
[14:01:30] <Capricorn^80> lines
[14:01:32] <Capricorn^80> extern void jpeg_abort (j_common_ptr cinfo);
[14:01:32] <Capricorn^80> extern void jpeg_destroy (j_common_ptr cinfo);
[14:01:32] <Capricorn^80> extern boolean jpeg_resync_to_restart (j_decompress_ptr cinfo, int desired);
[14:01:32] <Capricorn^80> # 2 "myfile.c" 2
[14:01:39] <Capricorn^80> wat next to do
[14:02:24] *** |tsoome| has quit IRC
[14:04:53] <Cyrille> well, it seems to show that part of the problem is that you don't have -I/usr/sfw/include in your CFLAGS. Someone suggested you try running 'CFLAGS="-I/usr/sfw/include" LDFLAGS="-L/usr/sfw/lib" ./configure' and see if that changes the outcome, you might want to give that a try.
[14:05:29] *** peteh_ has joined #opensolaris
[14:05:55] <Cyrille> another suggestion is to run ./configure --help or something like that to get the list of options and check if there's not one such as --with-jpeg=<path>
[14:07:26] <Capricorn^80> bash-2.05# gcc -E -I/usr/sfw/include myfile.c ... can u little explain wat this command is doing ?
[14:07:27] <sil3nt|warri0r> which runlevel is for GUI boot
[14:08:07] <Cyrille> Capricorn^80: it was doing a preprocessing pass on myfile.c and printing the result to stdout.
[14:08:38] <Cyrille> since your file only contained an include, it basically amounted to try and find libjpeg.h and output it (and its own includes)
[14:11:21] *** peteh has quit IRC
[14:21:28] *** Capricorn^80 has quit IRC
[14:21:59] *** Capricorn^80 has joined #opensolaris
[14:22:05] <Capricorn^80> sorry got d/c
[14:22:35] <Capricorn^80> Cyrille: did u get my last msg's ?
[14:22:44] <Cyrille> depends which one it was.
[14:23:01] <Cyrille> last I saw from you was a question about the meaning of the gcc -E ... line
[14:23:02] <Capricorn^80> the log i paste u
[14:23:07] <Cyrille> no then.
[14:23:12] <Capricorn^80> oh no
[14:23:14] <Capricorn^80> wait
[14:23:26] <Cyrille> don't paste large amounts of log...
[14:24:07] <Capricorn^80> its not large . i m just pasting error lines
[14:24:10] <Capricorn^80> i tried bash-2.05# CFLAGS="-I/usr/sfw/include" LDFLAGS="-L/usr/sfw/lib" ./configure
[14:24:11] <Cyrille> fair enough
[14:24:28] <Capricorn^80> n got error
[14:24:29] <Capricorn^80> configure: warning: *** External PROJ.4 directory not specified
[14:24:29] <Capricorn^80> checking for /epsg... no
[14:24:29] <Capricorn^80> configure: warning: *** Unable to locate PROJ data files.
[14:24:29] <Capricorn^80> checking for nad2bin... /usr/local/bin/nad2bin
[14:24:29] <Capricorn^80> checking whether to use Readline... no
[14:24:30] <Capricorn^80> checking whether to use JPEG... yes
[14:24:33] <Capricorn^80> checking for location of JPEG includes...
[14:24:34] <Capricorn^80> checking for jpeglib.h... no
[14:24:36] <Capricorn^80> configure: error: *** Unable to locate JPEG includes.
[14:24:41] <regx> he did it anyways lol
[14:24:52] *** calumb is now known as calLNCH
[14:25:05] <Capricorn^80> n then tried
[14:25:06] <Capricorn^80> bash-2.05# ./configure --help | grep jpeg
[14:25:06] <Capricorn^80>   --with-jpeg             support JPEG functionality (default: yes)
[14:25:06] <Capricorn^80>   --with-jpeg-includes=DIRS
[14:25:06] <Capricorn^80>   --with-jpeg-libs=DIRS   JPEG library files are in DIRS
[14:25:06] <Capricorn^80> bash-2.05#
[14:27:49] <Cyrille> right, and though it seems you had that conversation before, I imagine running ./configure --with-jpeg-includes=/usr/sfw/include --with-jpeg-libs=/usr/sfw/lib didn't help? Do you at least see the gcc command line in the log featuring these includes?
[14:28:44] *** dcFood is now known as damienc
[14:29:03] <Capricorn^80> u mean should i check /opt/grass.6.20/config.log file ?
[14:29:21] <Cyrille> possibly
[14:29:42] <Capricorn^80> there is goes
[14:29:44] <Cyrille> it should have detailed output about what configure tries to do and how exactly it fails
[14:29:45] <Capricorn^80> there it goes
[14:29:55] <Capricorn^80> onfigure:7411: checking for jpeglib.h
[14:29:55] <Capricorn^80> configure:7419: gcc -E   conftest.c >/dev/null 2>conftest.out
[14:29:55] <Capricorn^80> configure:7415:21: jpeglib.h: No such file or directory
[14:29:55] <Capricorn^80> configure: failed program was:
[14:29:55] <Capricorn^80> #line 7414 "configure"
[14:29:56] <Capricorn^80> #include "confdefs.h"
[14:29:58] <Capricorn^80> #include <jpeglib.h>
[14:30:21] <Cyrille> and we're back to the gcc command not having -I/usr/sfw/includes...
[14:30:34] <Cyrille> where did you find the code you're building?
[14:30:49] <damienc> I told him to try build the one liner
[14:31:01] <damienc> because that was the code from config.log
[14:31:06] <Capricorn^80> damienc: i did that
[14:31:09] <Cyrille> the one liner builds fine *if you specify -I/usr/sfw/include*
[14:31:19] <damienc> that was part of my instructions too
[14:31:22] *** mrdeviant has joined #opensolaris
[14:31:24] <Cyrille> the problem is getting that into the command the configure script is trying to use
[14:31:27] <regx> woo
[14:31:34] <damienc> Cyrille: agreed
[14:31:43] <regx> ive got all 6 ISO's burned labeled and in hard cd cases
[14:31:56] <regx> quad XEON box here i come!
[14:32:09] <regx> are all XEON's 64bit?
[14:32:12] <damienc> Capricorn^80: where did you get the 'grass' tarball? I'll have a quick look.
[14:32:21] <damienc> though I only have Solaris 10 handy
[14:32:28] <Cyrille> damienc: and I'm starting to get the distinct feeling there might be an issue with the configure script or something so I'd rather have a direct look at it than keep playing let's put random characters on the command line and see what happens ;-)
[14:32:45] <Capricorn^80> ok wait let me tell u
[14:32:55] <damienc> Cyrille: true.
[14:33:50] <Cyrille> (because in the amount of time that's been spent we presumably could have written the f*$%ing makefiles ourselves ;-))
[14:34:00] <Capricorn^80> http://wgbis.ces.iisc.ernet.in/grass/index.php
[14:34:42] <damienc> What version? 6.2.0RC2 ?
[14:34:58] <Cyrille> latest from download page?
[14:34:59] *** onlinebacon has joined #opensolaris
[14:35:37] <Cyrille> wow, that server's apparently running from a 128k modem...
[14:35:37] <Capricorn^80> http://wgbis.ces.iisc.ernet.in/grass/download/index.php#g62x
[14:35:42] <Capricorn^80> check this
[14:35:43] *** eugene has quit IRC
[14:35:45] <damienc> got it.
[14:35:49] <Capricorn^80> in stable table
[14:35:56] <damienc> Cyrille: yeah, 5.3k/s
[14:36:00] <damienc> and falling.
[14:36:12] <Cyrille> well I was getting 8k until you started ;-)
[14:36:20] <onlinebacon> Capricorn^80: My dad uses that
[14:36:40] <onlinebacon> arcgis and grassgis
[14:36:49] <Capricorn^80> onlinebacon: wat ?
[14:36:52] <Capricorn^80> oh really
[14:37:00] <onlinebacon> yeah
[14:37:02] <Capricorn^80> did he use arcgis on solaris ?
[14:37:19] <onlinebacon> I think he did at work, I'll ask him
[14:37:21] <Capricorn^80> i love to have arcgis on solaris
[14:37:22] <Cyrille> Capricorn^80: so is it the stable version (6.0.2) or the testing one (6.2.x)?
[14:37:24] <Capricorn^80> but not getting it
[14:37:27] *** loke has joined #opensolaris
[14:37:31] <Capricorn^80> stable
[14:37:37] <Cyrille> right so.
[14:37:48] <damienc> buggeer
[14:37:53] <damienc> bugger
[14:38:00] <Cyrille> ?
[14:38:07] <onlinebacon> Capricorn^80: Nah they used windows and linux at his lab :(
[14:38:15] <damienc> I was well into  6.2.0RC2
[14:38:20] <Cyrille> mwahahahah
[14:38:24] <Cyrille> sorry.
[14:38:44] <damienc> 8.5MB of source code!!
[14:38:47] <damienc> wow
[14:38:52] <Cyrille> at 8k...
[14:38:59] <Capricorn^80> ok
[14:39:11] <damienc> 17mins
[14:39:15] <onlinebacon> http://www.crn.com/sections/breakingnews/breakingnews.jhtml;?articleId=193600331
[14:39:26] <onlinebacon> I know it is a few days out of date but is that happening?
[14:39:29] <onlinebacon> GPL for java?
[14:39:32] <Cyrille> yup
[14:39:36] <mrdeviant> sadly, yes.
[14:39:39] <damienc> Capricorn^80: will get back to you in 20 mins
[14:39:50] <Cyrille> if the server holds up
[14:39:50] <damienc> mrdeviant: why?
[14:39:56] <onlinebacon> yeuch
[14:40:03] <Capricorn^80> hmm
[14:40:06] <mrdeviant> why what?
[14:40:10] <Capricorn^80> ok damienc
[14:40:15] <damienc> Cyrille: http://wgbis.ces.iisc.ernet.in/grass/mirrors.php
[14:40:15] <Capricorn^80> i m also leaving to home
[14:40:15] <Cyrille> why sadly I'd imagine
[14:40:17] <Capricorn^80> i will talk u later
[14:40:24] *** axisys has quit IRC
[14:40:29] <onlinebacon> cya Capricorn^80
[14:40:48] <Capricorn^80> ok take care all .. catch u later
[14:41:26] <Cyrille> damienc: from the Irish mirror, the stable version is 6.2.0...
[14:41:37] <damienc> mmm, yeah, other mirrors are the same
[14:41:55] <Cyrille> how interesting that the mirrors are more up to date than the "official" site...
[14:41:59] <damienc> Capricorn^80: What version? 6.2.0RC2 ? or 6.0.2?
[14:42:13] <Cyrille> well he said he was trying to build 6.0.2
[14:42:34] <damienc> typo perhaps?
[14:42:39] <Capricorn^80> 6.0.2
[14:42:42] <Cyrille> I specifically asked between the two
[14:42:51] <onlinebacon> but he thought that was stable
[14:43:03] <Capricorn^80> cool_capricorn80 at hotmail dot com .. thats my id
[14:43:05] <damienc> only one of the sites says it is stable
[14:43:06] <Cyrille> which is how it was presented on the site he pointed out
[14:43:08] <onlinebacon> so wouldnt he want the newest stable one?
[14:43:14] <damienc> rest have 6.2.0 as stable
[14:43:25] <Cyrille> I guess he would want any which actually compiles and runs...
[14:43:44] <damienc> 12.5MB for 6.2.0 source!!
[14:43:49] <Capricorn^80> onlinebacon: if u get any info about arcgis for solaris do mail me
[14:43:55] <onlinebacon> I will :)
[14:43:59] <Cyrille> anyway the problem we've been spending some time on is on 6..0.2 so let's have a look at this one.
[14:44:09] <Cyrille> damienc: that's inflation for you...
[14:44:11] <onlinebacon> Going to ask  him if they had any solaris machines at work
[14:44:12] <Capricorn^80> damienc n cyrille : thx for your support .. but work is still to be done :P
[14:44:14] <Capricorn^80> thx bro
[14:44:18] <Capricorn^80> catch u later
[14:44:23] <onlinebacon> Capricorn^80: wait a sec
[14:44:25] <onlinebacon> I'll ask him now
[14:44:34] <Capricorn^80> ok
[14:47:16] <mrdeviant> what's going on with the hg repository? I can't get code out of it.
[14:47:25] <onlinebacon> nah, they had one solaris machine there that was a lot slower than the rest of the other machines, so they didnt run it on it
[14:49:27] <Capricorn^80> well just ask him if i can get link of arcgis for linux
[14:49:43] <Capricorn^80> n it got that source code so that i can built it on solaris
[14:50:35] <onlinebacon> he says they bought it from somewhere
[14:51:04] *** sniffy has quit IRC
[14:51:15] *** sniffy has joined #opensolaris
[14:52:02] <Capricorn^80> oh
[14:52:08] <Capricorn^80> it means arcgis is not free
[14:52:15] <Capricorn^80> anyways thx
[14:52:22] <Capricorn^80> i will talk u later
[14:52:31] *** Capricorn^80 has quit IRC
[14:52:52] <damienc> I think I'll give up - configure 6.2.0 cannot find "External PROJ.4 includes"
[14:53:25] *** axisys has joined #opensolaris
[14:53:32] <damienc> I dont want to go downloading and building tons of other modules.
[14:54:59] *** axisys has quit IRC
[14:55:45] <Cyrille> well capricorn's logs also complained about proj.4 stuff missing
[14:56:12] <damienc> but it exits for me.
[14:56:36] <Cyrille> it apparently was just a warning then
[14:57:03] <damienc> still another 12 mins for 6.0.2
[14:57:05] <Cyrille> anyway I don't actually want to build it, I just want to have a look at the configure script and find out whence it gets its options for checking the jpeg stuff..
[14:57:06] <damienc> <4k/s
[14:57:23] <Cyrille> I'm 84% at the moment, but I have a meeting in three minutes...
[14:57:26] *** nyati has quit IRC
[14:57:45] <damienc> configure.in has LOC_CHECK_INCLUDES
[14:57:54] <onlinebacon> if you turn hyperthreading will your processor go faster or slower?
[14:57:55] <damienc> not the standard AC_CHECK_HEADERS
[15:00:41] <myrkraverk> if I want to compile my own xorg - do I just grab the official source, or should I get something else?
[15:02:13] *** Somethingelse has joined #opensolaris
[15:03:52] *** jafari has joined #opensolaris
[15:04:08] <jafari> good morning all
[15:05:33] <onlinebacon> mornign jafari
[15:06:59] <myrkraverk> morning? what time is it at your place?
[15:07:24] <damienc> NYC = 9am
[15:07:47] <onlinebacon> here it is 2pm
[15:07:57] *** eboutilier has joined #opensolaris
[15:08:04] <jafari> yea 9am for me too
[15:08:18] <jafari> what part of town you from onlinebacon
[15:09:25] <onlinebacon> uk :)
[15:09:42] <Cyrille> quite a large town, isn't it?
[15:09:57] <onlinebacon> Cyrille: I meant the UK, as in, United Kingdom
[15:10:15] <Cyrille> as opposed to what?
[15:10:19] <jafari> oh cool
[15:10:21] <onlinebacon> oh, sorry
[15:10:31] <onlinebacon> I thought you thought it was part of the town that jafari lives in :)
[15:10:38] <lasseoe> UK is just a suburb to the US ;-)
[15:10:41] *** |ReIkO| has joined #opensolaris
[15:10:42] <|ReIkO|> hi all
[15:10:46] <onlinebacon> :o you came from here :)
[15:10:47] * damienc is in north of Dublin city, Ireland.
[15:11:04] <Cyrille> onlinebacon: no, I was making a poor attempt at a joke ;-)
[15:11:20] <onlinebacon> ah :)
[15:11:29] * onlinebacon laughs to make Cyrille feel better
[15:11:45] <Cyrille> thank you, thank you, I'll be here all week (unfortunately)
[15:11:53] <regx> whats a good site to grab some packages from? i wanna try installing a package for the first time
[15:11:53] <damienc> onlinebacon: you were asked "what part of town you from".
[15:11:55] <|ReIkO|> some knows how can I pass data from client ssh to server sshd in the source code?
[15:12:13] <damienc> regx: just install one from your Solaris media.
[15:12:15] <onlinebacon> oh, i thought that was a saying
[15:12:29] <regx> damienc, where are they located and what type of file are they?
[15:12:53] <damienc> regx: what part of installing a package are you trying to 'try' ?
[15:13:06] <regx> i guess that pkgadd thing
[15:14:01] <damienc> do you have Solaris installed somewhere?
[15:14:07] <regx> ya
[15:14:13] <damienc> where did you install it from?
[15:14:21] <regx> a dvd
[15:14:39] <damienc> cd to the DVD
[15:14:50] <onlinebacon> how many people use opensolaris in the world?
[15:14:53] <regx> i dont knwo my friend gave it to me
[15:14:57] <damienc> then 'cd Solaris_11/Product' (change 11 as appropriate)
[15:14:58] <|ReIkO|> many
[15:14:59] <twincest> online: 7
[15:15:19] <damienc> regx: pkgadd -d . SUNWAnyPackageYouWant
[15:15:31] <regx> what is this SUNW i keep seeing?
[15:15:38] <twincest> regx: sun's stock symbol
[15:15:43] <regx> when do i use it?
[15:15:46] <twincest> (Sun Worldwide)
[15:15:58] <twincest> use it?  well, you could call your broker...
[15:15:59] <Cyrille> Stanford University NetWork
[15:16:01] <damienc> Solaris package names begin with 'SUNW'
[15:16:12] <regx> ah
[15:16:25] <damienc> try:
[15:16:30] <regx> i was setting up brandZ zones and it said somthign about SUNWlx
[15:16:35] <damienc> pkgadd -d . SUNWimagick
[15:16:54] <twincest> solaris packaging guidelines suggest prefixing the package name with a unique four character identifier, such as a stock symbol
[15:17:02] <twincest> so you see LTRSpackagename quite a lot
[15:17:26] <regx> anyone familar with BrandZ zones etc?
[15:18:00] <myrkraverk> regx: I'm trying myself now ;P
[15:18:36] <regx> myrkraverk, lemme know if you can figure out how to install the "Centos package in my case" to a different location manually..
[15:18:44] <regx> cuz im getting "no disc space left"
[15:19:01] <regx> for some reason i only have 50mb on my "/" and it fills it up instantly upon install request
[15:19:18] <twincest> oh dear, you used the default partitioning scheme
[15:19:20] <damienc> use the -R parameter with pkgadd
[15:19:28] <regx> twincest, :( ya
[15:19:45] *** Ezechial has quit IRC
[15:19:46] <twincest> i've found thats usually fairly useless
[15:19:47] <regx> i like how it auto mounts cd's :)
[15:20:00] <twincest> (hopefully the new installer will fix it)
[15:20:01] <myrkraverk> regx: erm - I'm using EXCR b50 - I arleady have brandz
[15:20:10] <regx> same here
[15:20:17] <regx> im talking about when your installing the ZONE itself
[15:20:18] <Cyrille> damienc: I'm looking at the configure script, and this is not the same one as capricorn's, it doesn't have the same line numbers...
[15:20:20] <twincest> myrk: you don't have a linux installation though
[15:20:24] * myrkraverk is creating a centos zone from the tarball right now
[15:20:35] <myrkraverk> twincest: erm?
[15:20:42] <damienc> Cyrille: "Unable to locate External PROJ.4 includes." - in 6.0.2. Guess it's not optional
[15:20:43] <twincest> regx: brandz still uses zonecfg to create the zone, doesn't it?
[15:20:47] <regx> ya
[15:20:50] <twincest> myrk: he's running out of space installing Linux in the zone
[15:21:00] <Cyrille> damienc: we're not looking at the same code as he is.
[15:21:01] <myrkraverk> http://www.opensolaris.org/os/community/brandz/install/ <----- here
[15:21:17] <myrkraverk> Cyrille: ah - I made a zfs fs for my zone
[15:21:18] <damienc> Cyrille: why am I not surprised.
[15:21:29] <twincest> regx: i think setting zonepath elsewhere should be enough
[15:21:39] <twincest> regx: (it's been a while since i uesd brandz though, so..)
[15:21:40] <myrkraverk> twincest: sorry, you ^^^^^^^^^^^
[15:21:44] <damienc> Cyrille: it's strange to me that configure sets CPPFLAGS, not CFLAGS.
[15:21:49] * damienc gives up.
[15:21:52] <twincest> myrk: what about it?
[15:21:56] <regx> twincest, ill try it again
[15:22:01] <regx> do you know the syntax?
[15:22:07] <Cyrille> damienc: well if it's just trying to gcc -E the code, it's only looking for the preproc flags
[15:22:10] <twincest> zonecfg -z lxzone
[15:22:16] <twincest> set zonepath=/other/location
[15:22:18] <damienc> Cyrille: ah.
[15:22:19] <regx> AH
[15:22:20] <regx> ok
[15:22:40] <Cyrille> damienc: at least that would be the expected behaviour...
[15:22:46] <regx> well i have my zone path set to /export/centosinstall  and df -h shows 22GB and durring the zone install it installs to "/"
[15:23:07] <twincest> hmm
[15:23:17] <Cyrille> but obviously something's not taken into account, and since we don't have the same code as he does, this is kind of a pointless exercise so I'll leave it at that...
[15:23:18] <regx> i have to be missing somthign
[15:23:27] <myrkraverk> twincest: erm - why I'm not running out of disc space :P
[15:23:36] <twincest> myrk: i don't understand
[15:24:51] *** onlinebacon has left #opensolaris
[15:24:52] <lasseoe> if it installs any packages in the global zone, then it expands them in /var/spool
[15:24:55] *** Burana has joined #opensolaris
[15:25:06] <lasseoe> probably why his filesystem fills up, assuming /var is on /
[15:25:17] *** coffman has joined #opensolaris
[15:26:15] <regx> it is trying to install to /var/ somthing
[15:26:31] <regx> so i should prolly reinstall?
[15:26:46] <regx> and repartition
[15:28:16] *** flowtech has joined #opensolaris
[15:28:26] <myrkraverk> erm, how do I tell solaris to route traffic for 192.168.3.1 through lo0 ?
[15:28:29] *** mega has quit IRC
[15:28:34] <myrkraverk> I'm not getting the route man page ;/
[15:29:32] <regx> route add default hostname -interface ?
[15:29:54] <myrkraverk> regx: I don't want it to be the default gw (if possible)
[15:30:16] <regx> odd that you dont have man route...
[15:30:31] <flowtech> man -M /usr/man route should do it
[15:30:33] <myrkraverk> route add 192.168.3.1 (I have it, I just don't understand it)
[15:30:41] <flowtech> oh :D
[15:30:44] <regx> route add IP -interface?
[15:31:03] <myrkraverk> both
[15:31:11] <myrkraverk> route add 192.168.3.1 127.0.0.1
[15:31:14] <myrkraverk> and
[15:31:25] <myrkraverk> route add 192.168.3.1 127.0.0.1 -interface
[15:31:31] <myrkraverk> are giving me errors ;/
[15:31:49] <regx> pkgadd -? what flag?
[15:33:15] <myrkraverk> damn, sorry guys, I'm just stupid - I was trying this in my centos zone ;P
[15:34:11] <regx> goll
[15:34:13] <regx> :)
[15:34:29] <myrkraverk> damn, I seem to have frozen the box ;/
[15:34:34] <regx> yay
[15:34:41] <myrkraverk> yes
[15:35:05] <regx> man i dont wanna reinstall
[15:35:07] <myrkraverk> and it even rebooted on its own
[15:35:08] <regx> that took like an hour
[15:35:34] <regx> i install FBSD on my 1u last night.. 4 mins.. fastest bsd install ive ever seen
[15:35:35] <myrkraverk> (didn't know solaris migh do that)
[15:35:57] *** triplah has quit IRC
[15:38:20] <regx> any suggestions on what partition sizes i should use?
[15:40:44] *** rachel_ has joined #opensolaris
[15:43:52] <myrkraverk> hmm, having my lx zone net on lo0 (I don't have a working NIC driver yet) seems to be a bad idea - pinging it simply reboots
[15:45:06] <myrkraverk> I guess I'm going to address that on the brandz list ;/
[15:45:27] *** karrotx has quit IRC
[15:46:51] *** Burana has quit IRC
[15:46:52] *** Burana has joined #opensolaris
[15:48:28] <regx> lol wow
[15:50:03] *** Kernel86|Laptop has quit IRC
[15:50:53] <coolvibe> woo, java became open source
[15:50:55] <coolvibe> kick ass
[15:51:12] * coolvibe congratulates Sun
[15:51:44] *** axisys has joined #opensolaris
[15:52:40] *** mega has joined #opensolaris
[15:53:11] <loke> coolvibe: "whatever"
[15:55:07] <myrkraverk> coolvibe: already?
[15:55:26] *** calLNCH is now known as calumb
[16:02:02] *** laca has joined #opensolaris
[16:03:09] *** Dar_ is now known as Dar_HOME
[16:05:06] *** Burana has quit IRC
[16:05:55] *** RaD|Tz has joined #opensolaris
[16:09:02] <myrkraverk> coolvibe: url?
[16:09:17] <coolvibe> myrkraverk: look at www.sun.com
[16:11:01] <quasi> myrkraverk: it has been announced, but is going to be taking a while before it happens
[16:11:57] *** crash| has joined #opensolaris
[16:12:00] *** solarisjon has joined #opensolaris
[16:12:31] <myrkraverk> quasi: ah, just like opensolaris ;)
[16:13:50] <PosixC> I am having problems with tftp on solaris : I have in /etc/inetd.conf the following lines:
[16:13:50] <PosixC> tftp    dgram   udp6    wait    root    /usr/sbin/in.tftpd      in.tftpd -s /tftpboot
[16:13:50] <PosixC> tftp    dgram   udp     wait    root    /usr/sbin/in.tftpd      in.tftpd -d -s /tftpboot
[16:14:03] <PosixC> yet I try from another machine on the same subnet
[16:14:23] <PosixC> the follwing : tftp  => getting to the tftp prompt; then
[16:14:28] <PosixC> tftp> get a.a
[16:14:38] <quasi> myrkraverk: I've heard someone saying v7 would be the gpl one
[16:14:41] <PosixC> where a.a is a small file under /tftpboot on the server
[16:14:48] <PosixC> and I am getting
[16:15:01] <PosixC> Transfer timed out.
[16:15:20] <jbk> PosixC: solaris 10?
[16:15:23] <PosixC> (I did not mention it but i am running "tftp IP"
[16:15:27] <PosixC> yes
[16:16:06] <PosixC> and I **DO** see the TFTP request when snooping on port 69
[16:16:26] <jbk> try inetconv -i /etc/inetd.conf
[16:16:56] <jbk> then inetadm | grep tftp
[16:17:03] <PosixC> a moment (I tried it earlier but I will try again)
[16:17:25] <PosixC> inetadm | grep tftp
[16:17:25] <PosixC> enabled   online         svc:/network/tftp/udp6:default
[16:17:25] <PosixC> enabled   online         svc:/network/tftp/udp:default
[16:17:27] *** Shadow^of^LUV has joined #opensolaris
[16:17:36] <vaneth> Hm, what type of installation should I use for shell/www/mail server? Reduced? Core?
[16:18:15] *** rachel_ has quit IRC
[16:18:22] <PosixC> jbk,again Transfer timed out.
[16:18:29] <PosixC> I am quite despereate with it
[16:18:39] <PosixC> a.a is a small file
[16:18:44] <jbk> what are the permissions of /tftpboot ?
[16:19:19] <jbk> and the files in there?
[16:20:19] *** solarisjon has quit IRC
[16:20:25] <PosixC> jnk, a moment ; I have an idea
[16:20:29] *** gnu2it2 has joined #opensolaris
[16:20:32] <trygvis> niks
[16:20:34] <trygvis> woops
[16:20:40] *** eboutilier has quit IRC
[16:20:59] *** solarisjon has joined #opensolaris
[16:21:28] *** king has joined #opensolaris
[16:22:17] *** tsoome has quit IRC
[16:22:34] *** vmhobbes has joined #OpenSolaris
[16:22:44] *** king has quit IRC
[16:22:52] <PosixC> jbk, it was a problem of firewall
[16:22:57] <PosixC> in the client
[16:23:04] <jbk> ahh
[16:25:00] *** solarisjon has quit IRC
[16:25:40] *** solarisjon has joined #opensolaris
[16:26:07] *** Shadow^of^LUV has quit IRC
[16:39:53] *** timofonic has quit IRC
[16:40:38] <PosixC> jbk, oooommmmm, I still have problems
[16:40:58] <PosixC> now : I get "Timed out waiting for TFTP reply"
[16:41:34] <PosixC> and I am trying to get not a big file :  214,328 bytes
[16:41:51] <jbk> does running snoop on the box show it getting/sending the packets?
[16:42:20] <axisys> PosixC: hmm.. i thought solaris tftp has a built in file size limit
[16:42:56] <axisys> PosixC: u have a very small size file.. nm
[16:43:14] <PosixC> When I snoop I see that there are packets of size 524 bytes (it seems to me that it should be 512)
[16:43:37] <PosixC> axisys, this is simply tring to perform netboot
[16:43:44] <PosixC> through tftp
[16:44:32] <PosixC> I did such things on linux (also with DHCP) and it worked very good
[16:45:25] *** djgregor has quit IRC
[16:45:31] <axisys> PosixC: so did u say u have a tcpwrapper and/or ipf?
[16:45:48] <PosixC> axisys, currently it is down
[16:45:51] *** calumb has quit IRC
[16:46:08] <axisys> so u r tftping from another unix box or router?
[16:46:17] <PosixC> yes
[16:46:32] <axisys> yes means router or unix box?
[16:46:40] <PosixC> it is a unix box
[16:46:44] <PosixC> diskless
[16:46:53] <axisys> oh ok..
[16:46:57] *** calumb has joined #opensolaris
[16:47:36] <axisys> u dont have anything in /etc/hosts.allow or /etc/hosts.deny that may be blocking it?
[16:48:42] <PosixC> don't think so but i will check in a moment
[16:50:14] <PosixC> axisys, no,nothing
[16:50:58] <axisys> stop the tftp svcs and then run it like this /usr/sbin/in.tftpd -d /tftpboot to see what it is doing
[16:51:20] <PosixC> ok
[16:51:23] *** axxl has quit IRC
[16:52:35] <PosixC> axisys, the thing is : when I try the same from a diffenet unix (I mean get that file) it DOES work ok
[16:52:52] <PosixC> but when I try it with a diskless station it does NOT !
[16:53:12] *** vmhobbes has quit IRC
[16:54:50] *** calumb has quit IRC
[16:55:00] *** bengtf has joined #opensolaris
[16:55:26] *** calumb has joined #opensolaris
[16:56:10] <axisys> so tftp server works in general.. good
[16:56:28] <axisys> so it could be a IP related..
[16:57:12] <PosixC> the thing is this:
[16:57:21] <axisys> if u run in debug mode like i suggested it will give you a lot of info about what the diskless client doing
[16:57:27] <PosixC> the diskless station has a MAC address
[16:57:34] <PosixC> in /etc/ethers
[16:57:56] <PosixC> there is a line like this:
[16:57:56] <PosixC> 00:03:BA:D8:B8:DA ni
[16:58:26] <PosixC> and in /etc/hotst there is the follwing mapping to ni
[16:58:28] <PosixC> 192.168.0.120 ni
[16:58:46] *** Burana has joined #opensolaris
[16:58:57] <axisys> sounds like u r trying to jumpstart a diskless client
[16:59:16] <PosixC> coneceptually this is what I try to do
[16:59:35] <LeftWing> Why would you JumpStart a diskless client..?
[16:59:44] <PosixC> and snoop -v port 69 shows
[16:59:44] <PosixC>  Source address = 192.168.0.120, ni
[17:00:16] <PosixC> LeftWing, in is pseudo diskless; I simply boot it with boot net - install
[17:00:37] <LeftWing> Pseudo.. diskless..
[17:00:54] <LeftWing> It's either diskless or it isn't.  =P
[17:02:12] <LeftWing> Ahh, I'm going to bed.
[17:02:15] <LeftWing> Night folks.
[17:02:17] * LeftWing &
[17:03:12] *** Burana has quit IRC
[17:03:13] *** Burana has joined #opensolaris
[17:11:17] <lasseoe> hm
[17:11:20] *** Burana has quit IRC
[17:11:30] <lasseoe> wonder when SunSolve starts working properly again
[17:12:51] <quasi> lasseoe: seems to be working now
[17:13:30] <Stric> I got 'Patch Club Report Newsletter' from 10/14/2006 up until now today :)
[17:14:04] <quasi> Stric: me too
[17:14:07] *** laca is now known as lacaAFK
[17:21:22] *** loke has quit IRC
[17:26:48] *** whaq has quit IRC
[17:30:33] *** sommerfeld has joined #opensolaris
[17:30:33] *** ChanServ sets mode: +o sommerfeld
[17:31:00] *** ndroux has quit IRC
[17:32:30] *** ndroux has joined #opensolaris
[17:37:22] <myrkraverk> hmm, I haven't checked, but in case someone here knows: does sunstudio have different options for amd64 and em64t code generation?
[17:37:51] <regx> what is OVERLAP in the partitioner mean?
[17:38:12] *** stevel has joined #opensolaris
[17:38:13] *** ChanServ sets mode: +o stevel
[17:40:16] *** coolvibe has quit IRC
[17:42:30] <richlowe> mornin' stevel
[17:43:12] <stevel> g'morning
[17:43:32] <jbk> regx: that's normal for slice 2
[17:44:48] <regx> can you give me a suggestion as to what my partition sizes should be for a 40gb hdd? last time i screwed it up some how
[17:46:09] *** deather__ has joined #opensolaris
[17:47:10] <jbk> what do you plan to do with the system?
[17:47:40] <regx> setup BrandZ zones with a few Zones and learn off it and maybe setup like apache and DNS
[17:48:21] <Error_404> why bother with brandz?
[17:49:57] *** djgregor has joined #opensolaris
[17:50:24] <hile_> why use brandz unless you've got something like an application that ties you to linux
[17:50:31] <regx> well
[17:50:48] <regx> i have my reasons
[17:50:52] <regx> :-/
[17:50:56] <Error_404> no you don't
[17:51:06] <regx> lol
[17:51:15] <hile_> eg, i've only got one POWER machine, so  the only choice i have to learn NIM is "buy another POWER machine" or NIMOL
[17:51:38] <hile_> grabbing RH9 ISOs and sticking a RH9 image in VirtualPC on my laptop is the much cheaper option.
[17:51:54] <regx> so... back to my original question.. is there like a place i can get a good idea on my partition sizes?
[17:52:19] <PosixC> question: are the source of packages available ? I am looking for the sources of SUNWiscsir for exmaple
[17:52:31] <PosixC> looked in cvs.opensolaris.org - could not fins
[17:52:36] <PosixC> s/fins/find
[17:52:53] * regx is eating M&Ms
[17:53:42] *** mega has quit IRC
[17:54:03] *** PosixC has quit IRC
[17:54:38] *** PosixC has joined #opensolaris
[17:54:43] <richlowe> if SUNWiscsir is the initiator, it'd be in NWS
[17:54:53] <richlowe> which is open, but I suspect will have vanished from cvs.opensolaris.org
[17:55:25] *** vmhobbes has joined #OpenSolaris
[17:55:28] <richlowe> ah, nope, it's there.
[17:55:34] <hile_> regx: leave slice 7 unmounted (at about 32 or 64 MB) for mddb data; slice 1 is swap at, say 2x physical RAM
[17:55:41] <hile_> and the rest of your rootdisk in s0 as /
[17:58:29] <regx> thanks
[17:58:43] *** slowhog has joined #opensolaris
[17:59:05] <regx> 3.4GB install lol
[18:02:32] *** deather_ has quit IRC
[18:03:00] *** tsoome has joined #opensolaris
[18:05:47] *** vmhobbes has quit IRC
[18:09:41] *** dunc has quit IRC
[18:09:58] *** Burana has joined #opensolaris
[18:14:09] *** bengtf has quit IRC
[18:15:26] <sickness> finally at home :)
[18:15:32] *** Cyrille has quit IRC
[18:15:37] <regx> dude thats sick
[18:16:22] *** bengtf has joined #opensolaris
[18:19:26] *** LordKing has joined #opensolaris
[18:20:29] *** bougie has joined #opensolaris
[18:21:07] <bougie> hello :)
[18:21:40] *** coffman has quit IRC
[18:22:34] *** Somethingelse has quit IRC
[18:26:15] <regx> HI
[18:28:44] *** nwf has joined #opensolaris
[18:31:53] *** craigmohrman has joined #opensolaris
[18:32:59] *** craigmohrman has left #opensolaris
[18:34:04] *** craigmohrman has joined #opensolaris
[18:34:26] *** craigmohrman has left #opensolaris
[18:39:05] *** dunc has joined #opensolaris
[18:43:57] *** Snake007uk has quit IRC
[18:48:28] *** damienc has quit IRC
[18:50:51] *** solarisjon has quit IRC
[18:53:48] *** deather__ is now known as deather
[18:55:42] <rpaulo> folks, the solution to install SX on Parallels Desktop is to pass install_media=cdrom,atapi-cd-dma-enabled=0 to the kernel
[18:56:25] <rpaulo> someone said he/she was having this problem yesterday, but I can't recall the nick.. :-/
[18:56:55] *** pikapika has joined #opensolaris
[18:57:02] <g4lt-mordant> rpaulo, what if it's already that way, because you have a SCSI-only system?
[18:57:34] <rpaulo> what do you mean? Parallels doesn't emulate SCSI AFAIK.
[18:58:07] <rpaulo> I don't know if you know, but Parallels is a virtual machine.
[18:58:07] *** ssam has joined #opensolaris
[18:59:28] <gdamore> rpaulo: glad to hear to you figured out the parallels thing.
[19:00:52] <rpaulo> :)
[19:00:54] *** ssam has quit IRC
[19:01:23] *** alobbs has quit IRC
[19:01:39] *** |joni| has quit IRC
[19:02:11] *** lacaAFK is now known as laca
[19:02:24] <pikapika> hello
[19:04:26] *** Kronuz has joined #OpenSolaris
[19:04:39] <Kronuz> hello
[19:04:52] <Kronuz> hey, do you know if Sun sells UPS systems?
[19:05:25] <gdamore> as far as i know, they do not
[19:05:31] <Kronuz> oh
[19:05:36] <dunc> but i bet the website knows either way
[19:05:41] <Kronuz> so where would you get one of those?
[19:05:53] <Kronuz> dunc: I looked there, but I couldn't find them
[19:06:02] <gdamore> small systems you can pick up from places like fry's or compusa.
[19:06:05] <Kronuz> but neither as gigabit ethernet switches
[19:06:14] <gdamore> APC makes good UPS'
[19:06:21] *** |joni| has joined #opensolaris
[19:06:29] <Kronuz> gdamor: but what about for larger systems
[19:06:40] <gdamore> like what class of system?
[19:07:17] <Kronuz> gdamor: well, it's just that it isn't a single computer server... I'm buying three T1000 and a SAN
[19:07:17] <gdamore> up to 2 kVA is no problem for APC.
[19:07:36] <dunc> we've  got a UPS the size of a small room including the batteries at work
[19:07:46] <dunc> i think it's by powerware
[19:07:59] <Kronuz> O_O
[19:08:05] <Kronuz> size of a small room!
[19:08:08] <gdamore> actually, APC has up to 1.6MW :-)
[19:08:12] <gdamore> http://www.apc.com/products/category.cfm?id=13
[19:08:17] <Kronuz> I suppoese 2 kVA is alright
[19:08:28] <Kronuz> how can you calculate how long it will last?
[19:08:49] <Kronuz> (a good thing is the T1000 are energy efficient)
[19:09:11] <gdamore> you can usually do something like take the kVA * .7 / Watts
[19:09:37] <dunc> we've also got a   cos phi  guage in the switchgear room :)
[19:09:38] <gdamore> i'm not sure that gives you time though... there is a missing factor.
[19:09:59] <dunc> cos(phi)
[19:10:02] *** astinus has joined #opensolaris
[19:10:25] <dunc> Kronuz, u need to know the Ah rating of your combined batteries
[19:10:34] <gdamore> check the website.  it gives you calculations and also product selection.
[19:11:55] <Kronuz> do you usually get two UPS or just one... I suppose for redundancy I'd want two
[19:13:15] <gdamore> to get redundancy you really need dual redundant power supplies on your hardware.  not sure if the T1000 includes that or not.
[19:13:23] <hile_> The T1000 does not
[19:13:35] <Kronuz> oh.. hmm
[19:13:38] <hile_> you get redundant powr supplies
[19:13:54] <gdamore> is it an option on the T1000?
[19:14:01] <Kronuz> but still if one power supply fails down it goes
[19:14:02] <hile_> and you run physically diverse circuits
[19:14:11] <hile_> gdamore: no that i know of
[19:15:02] <g4lt-mordant> I think that's how schwartznegger beat it was power issues ;P
[19:15:07] *** LordKing has quit IRC
[19:15:27] <Kronuz> how much discount you think I can get from Sun if I push them?
[19:15:33] <Kronuz> 10, 20% ?
[19:15:41] <hali> Kronuz: 60% of so .. if you buy sixty racks of t1000s
[19:15:48] <g4lt-mordant> hold out for 100% ;)
[19:15:51] <gdamore> heh.
[19:15:54] <Kronuz> lol
[19:16:04] <sniffy> Yay. Lets not feed @sun.com people ;)
[19:16:16] <Kronuz> hali: really, you've seen up to 60% discounts!?
[19:16:32] <Stric> gdamore: no, t1k doesn't have redundant psu option (doesn't fit)
[19:16:41] <hali> Kronuz: yes, even more on storage... but you need to buy a lot of kit ... i mean... A LOT!
[19:16:45] <gdamore> i see that now on the website
[19:17:13] *** bondolo has quit IRC
[19:17:20] <hali> Kronuz: i remember this insurance company who bought 8 at the time maxed out top end SANs and they got silly discount.. the total order value was still 10s of millions
[19:17:22] <Kronuz> hali: I'd guess... but for mere mortals getting a couple servers and a SAN
[19:17:50] <hali> you can probably get 20% without to much bitching
[19:18:02] <gdamore> fyi: the T2000 has dual redundant power supplies.
[19:18:09] <hali> and take advantage of the developer programs if you are buying any dev server
[19:18:25] <Kronuz> hali: oh... dev programs
[19:18:26] <hali> most dev programs give ~30% right off hardware prices
[19:18:27] <Kronuz> hmm
[19:18:34] <gdamore> 20% is probably believable.  especially for sparc h/w
[19:18:36] <hali> but you have to use the servers for development
[19:18:45] <hali> internal development only
[19:18:49] <Kronuz> hali: what do you know about the startup program?
[19:18:50] *** bondolo has joined #opensolaris
[19:18:55] <hali> Kronuz: no
[19:19:01] <gdamore> at the low end the margins are smaller and you can't get as good a discount as you can on higher end h/w
[19:19:02] <hali> nothing
[19:19:24] <Kronuz> I already applied to the startup program, I hope we get the discounts
[19:19:35] <Kronuz> but for what you say I could guess it'll be about 30%
[19:19:40] <Kronuz> (at the most)
[19:19:41] <gdamore> good luck. :-)
[19:20:02] * gdamore wants a T1000 or T2000 for doing massively parallel compilation.
[19:20:08] * gdamore wants a pony, too. :-)
[19:20:12] <Kronuz> lol
[19:20:18] * Kronuz wants the world
[19:20:39] <Stric> gdamore: http://blog.madism.org/images/pony.jpg
[19:20:56] <gdamore> boo hoo. :-(
[19:21:21] <gdamore> i'll settle for the T1000 though.  :-/
[19:21:43] <sniffy> One really needs more storage for it though.
[19:22:01] <sniffy> It's easy to run out of the mere 80GB I got it with (which I expected too though).
[19:22:02] <Kronuz> I'm trying to get the T1000 without hard drives
[19:23:32] <gdamore> heh.
[19:23:41] <Kronuz> didn't know they couldn't have dual power supply
[19:23:47] <gdamore> i'm still waiting for my Via based system to arrive so I can set it up as a NAS server. :-0
[19:23:56] <gdamore> go for the T2000 if that is important to you.
[19:24:13] <Kronuz> gdamore: yeah, I'll probably will have to
[19:24:19] <Kronuz> tho' it's 4000 more
[19:24:28] <Kronuz> (almost for the same thing)
[19:25:52] <Kronuz> darn SAN is taking most of our budget
[19:25:52] <gdamore> you get more cores, don't you?  or at least the ability to add more (more sockets)
[19:25:58] <Kronuz> not really
[19:26:19] <Kronuz> T2000 and 8 x 1.0Ghz Cores with 16GB of RAM = 18545
[19:26:24] <Kronuz> T1000 and 8 x 1.0Ghz Cores with 16GB of RAM = 14445
[19:26:30] <Kronuz> same thing
[19:26:34] <Kronuz> 4000 more
[19:26:47] <Kronuz> except for the 2 psu option and a DVD drive
[19:26:57] <gdamore> its a 2U system, though.
[19:27:01] <Kronuz> (and two more holes for disks)
[19:27:11] <Kronuz> yeah, I won't use the holes :P
[19:27:20] <Kronuz> if I go for it it'd be just for the 2psu
[19:27:27] *** |ReIkO| has quit IRC
[19:27:48] <svoboda> the new x64 ILOM is amazing
[19:27:52] <svoboda> sun rocks
[19:28:05] <hile_> amazing compared to what?
[19:28:23] <gdamore> you get a 2nd disk drive, too. :-)
[19:28:26] <svoboda> amazing compared to anything AMD/Intel based in the market
[19:28:46] <svoboda> you can redirect the monitor, keyboard, cdrom, etc. over ssh
[19:28:57] <Stric> Kronuz: t2k has 16 dimm slots (vs 8) too.. etc
[19:29:09] <gdamore> and extra PCI-E and PCI-X slots.
[19:29:19] <gdamore> t1k only has 1 PCI-E slot, and no legacy PCI
[19:29:23] <mrdeviant> sounds like what i can do with my ibm bladecenter :)
[19:29:25] <Kronuz> Stric:: yeah, but still the same ammount of RAM there
[19:29:26] <Stric> but yeah, you can configure them for similar specs for a higher price.
[19:29:38] <Kronuz> yep
[19:29:44] <Stric> just like other machines.. maxed out low-end vs non-maxed higher end..
[19:29:53] <Kronuz> :P
[19:30:11] <Kronuz> I think they should let you have those machines in a more configurable way
[19:30:22] <sommerfeld> Kronuz: The niagara CPU doesn't do multi-socket SMP.
[19:30:29] <Kronuz> it's like "that's all we offer, on the website"
[19:30:40] <gdamore> hmm.... no indications of USB ports.   don't these systems have USB expansion on them?
[19:30:49] <Auralis> Kronuz: give your thanks to ebay for the liited config options
[19:30:49] <Stric> Kronuz: then you call/mail your sun representative and say what you really want..
[19:31:02] <Kronuz> and then they say they don't have that option
[19:31:07] <Kronuz> I've already tried
[19:31:12] <Stric> what option are you trying to do?
[19:31:31] <Auralis> Kronuz: go through a small reseller, they are much more flexible
[19:31:37] <Kronuz> well, for instance the T1000 without hard drives
[19:31:47] <sommerfeld> looks like T1000 doesn't have usb, but the T2000 does.
[19:32:04] <hile_> hi, sommerfeld
[19:32:18] <Kronuz> I still don't get the Performance Packs options... it's cheaper with better support?!
[19:32:45] <hile_> i think it's one of those "you must choose a support contract with this system"
[19:33:00] <hile_> so they discount the hardware cost, but the listed price doesn't include hte contract cost?
[19:33:01] <Kronuz> yeah, but the 1 year silver is included!
[19:33:04] <hile_> (just a guess)
[19:33:15] <hile_> silver is a pain in the ass
[19:33:16] <sommerfeld> (T2000 has 2 front, 2 back)
[19:33:20] <Kronuz> so if you don't change that you get it cheaper
[19:33:22] <hile_> we had that at my last job.
[19:33:34] <hile_> my current job is much nicer
[19:33:39] <Kronuz> hile_: really? how come?
[19:33:45] <Kronuz> 'bout the silver
[19:33:48] <gdamore> guess you can only netboot a t1k then.  no dvd, no USB....
[19:33:56] <hile_> 8a-5p onsite support
[19:34:09] <hile_> next day (at best) for parts
[19:34:21] <Kronuz> oh
[19:34:24] <Stric> gdamore: I heard some rumors about t1k designers wanted to leave out disk altogether (use SAN), but that didn't play well with others..
[19:34:33] <Kronuz> hile_: what about the warranty?
[19:34:46] <Kronuz> with no contract
[19:34:53] <gdamore> interesting.  well, it only has room for two drives, so just a mirrored boot drive...
[19:34:59] <myrkraverk> hmm, any X - pert here? or would anyone here know what my best source for Xorg source is?
[19:35:00] <hile_> that sucks worse ;)
[19:35:01] <Kronuz> (I've heard that's not a good option tho')
[19:35:03] * gdamore is not a big fan of SAN
[19:35:14] <sommerfeld> t1k is clearly the "rack and stack for horizontal scalability", with individually expendable nodes..
[19:35:24] <gdamore> myrkraverk: later in the day you can find alanc around.  he's the Xorg expert.
[19:35:38] <Kronuz> sommerfeld: only if it had dual psu
[19:35:47] <myrkraverk> gdamore: later in the day? it's about 18:30 here :P
[19:35:56] <Stric> +1h here
[19:35:59] <gdamore> ib 10:35 am on the west cost usa.
[19:36:05] <gdamore> s/ib/its/
[19:36:30] <myrkraverk> gdamore: hmm, that seems awfully slow ;P
[19:37:48] <gdamore> Kronuz: maybe the intent is to stick these in a rack that provides redundant external power?
[19:38:01] <myrkraverk> . o O (on other things, I've heard Slowlaris and Windoze, but is there something similar for Linux - or is GNU just a swear word on its own?)
[19:38:04] <Kronuz> hmm
[19:38:11] <gdamore> Loonix.
[19:38:13] <Kronuz> gdamore: what rack would that be?
[19:38:14] <Stric> myrkraverk: Lunatix?
[19:38:17] <Kronuz> Loonix sucks
[19:38:22] <myrkraverk> Stric: ah ;)
[19:38:26] <twincest> myrk: linsux
[19:38:30] <gdamore> Kronuz: no idea. just a guess.
[19:38:34] <myrkraverk> twincest: ;)
[19:39:01] <myrkraverk> twincest: personally, I just use gnu as a swear word ;)
[19:39:16] <sommerfeld> i've more frequently seen "lunix".
[19:39:22] <twincest> i think that kind of naming is lame beyond words, but you did as
[19:39:24] <twincest> +k
[19:39:31] <hile_> or loonix
[19:40:23] <myrkraverk> twincest: yes - until I don't think I've ever come across such names for linux (and, quite honestly, gnu seems fine for a swear word)
[19:40:36] <myrkraverk> + now in the middle
[19:40:45] <gdamore> loonix is used in the netbsd community.
[19:41:14] <myrkraverk> ;)
[19:41:17] <gdamore> there is also "flea" for freebsd and "ogre" for openbsd
[19:41:44] <myrkraverk> are they meant to be degrading?
[19:41:52] <gdamore> ogre certainly is.
[19:41:59] <gdamore> dunno about flea.
[19:42:00] <mrdeviant> my favorite derogatory term is what some postgresql users call mysql : ImSOL.
[19:42:01] <myrkraverk> it is? :P
[19:42:21] <myrkraverk> ImSOL ?
[19:42:28] *** nwf has quit IRC
[19:42:34] <mrdeviant> as in, I'm S-O-L
[19:42:53] <hile_> I like that mrdeviant
[19:42:54] <Kronuz> myrkraverk: lol "a swear word" ... you bet it is
[19:42:58] 
[19:43:01] <myrkraverk> as in "I'm so lame" ?
[19:43:08] <mrdeviant> shit otta luck
[19:43:12] <myrkraverk> ah
[19:43:39] <g4lt-mordant> of course, some names just lend themselves to derision, like WinCE
[19:43:46] <myrkraverk> ;)
[19:43:49] <g4lt-mordant> no changes necessary
[19:44:06] *** calumb has quit IRC
[19:44:33] *** sunnyDay has joined #opensolaris
[19:44:41] <Kronuz> myrkraverk: like: "if you don't like it, GNU you!"
[19:44:55] <myrkraverk> Kronuz: excactly ;)
[19:44:58] <Kronuz> lol
[19:45:20] <Kronuz> yeah, that's my feeling too
[19:45:21] <PerterB> "GPL me harder"
[19:45:29] <Kronuz> rotfl
[19:45:33] *** jcea has joined #opensolaris
[19:45:40] <myrkraverk> PerterB: now, that's bordering on indecency ;)
[19:45:48] * PerterB smiles
[19:47:38] <Kronuz> or "go and GPL your mother!"
[19:48:15] <Kronuz> personally, I've always seen GNU as a cancer...
[19:48:40] *** RaD|Tz has quit IRC
[19:48:50] *** LordSky has joined #opensolaris
[19:49:12] <LordSky> hi all
[19:49:21] <Kronuz> hi
[19:55:33] *** jcea has joined #opensolaris
[19:57:19] *** LordKing has joined #opensolaris
[19:57:19] *** flowtech has quit IRC
[19:59:22] <Kronuz> hey, has anyone seen benchmarks for the T2000 1.2 Ghz vs. 1.0 Ghz ?
[19:59:36] <trygvis> anyone using openldap and are changing user's passwords with a ldif?
[19:59:37] <Kronuz> you think it's worth it? (the 0.2 difference)
[20:01:57] <Kronuz> 'cause other than that... is RAM memory that expensive?
[20:02:00] <Stric> grmbl. the N+2 redundant power supply in v445.. if I connect two each to two different power sources, that should be ok, right?
[20:02:13] <Kronuz> as to go from 16GB to 32 GB for $10,000
[20:02:25] <Stric> Kronuz: 2GB ECC dimms aren't cheap
[20:02:32] <Kronuz> but 10 grand!?
[20:03:00] <Kronuz> (for 16 GB)
[20:06:14] <Stric> 50% higher than what dell takes for instance..
[20:07:04] <Stric> wait.. that's not ecc I think
[20:07:20] <Stric> oh, it is..
[20:07:26] <Kronuz> you see ;)
[20:07:32] <Kronuz> it's expensive
[20:07:35] <Kronuz> Sun is expensive
[20:08:04] *** jcea has joined #opensolaris
[20:08:24] <Stric> about $5k for a single 4G dimm at dell..
[20:08:46] <Stric> there's no volume up there.. thus higher costs.
[20:10:08] <Kronuz> Stric: It's been really hard to decide... 32 vs 16 GB
[20:10:23] <Kronuz> at least in one of our systems I guess we could ave 32GB
[20:10:30] <Kronuz> (maybe for the database)
[20:10:35] *** yakov has joined #opensolaris
[20:10:38] <yakov> yo brazas
[20:11:16] <Kronuz> Stric: do you think the extra 16GB can really be useful? :P
[20:11:20] *** yarihm has joined #OpenSolaris
[20:11:26] <Stric> Kronuz: so you just have to decide whether 2* 8x1GHz/16G  is better or worse than 1* 8x1.2GHz/32G for your workload.
[20:13:08] <Kronuz> Stric: but how :S
[20:13:15] <Kronuz> try and buy!!!
[20:13:16] <Kronuz> :P
[20:13:19] <Stric> Why are you looking at 32G machines?
[20:13:28] <Stric> Do you need it or "it looks cool"?
[20:13:44] <Stric> if the first, then you need it and it's worth it for you.
[20:13:47] <Stric> if the latter, you don't.
[20:13:48] <trygvis> you can have a looot of appservers running with 32GB memory
[20:13:53] <Kronuz> dunno, 'cause it looks cool, and we'll need it eventually
[20:14:02] <Stric> trygvis: but the question is if he needs it in a single machine
[20:14:03] <Kronuz> and since you can't upgrade those systems memory...
[20:14:23] <Kronuz> (without throwing away the 16 GB)
[20:14:41] <Stric> either you have 32G in one machine with 8 cores or you have 16+16 with 8+8 cores
[20:14:43] <Kronuz> I was thinking it might be worth to get the 32 GB version from start
[20:14:47] <Stric> for roughly the same price
[20:15:26] <Kronuz> yeah, that's the dilemma
[20:15:43] <Stric> not much dilemma, do you need lots of memory in -one- machine or not?
[20:15:58] *** solarisjon has joined #opensolaris
[20:16:21] <Kronuz> we won't be needing the 32 GB right away, but we might soon
[20:16:21] <Stric> or will two machines do the task equally good.. if so, you have twice the performance
[20:16:34] <Kronuz> also 'cause I'm not sure how to calculate how much memory I'll need
[20:16:43] *** yakov has quit IRC
[20:16:46] <Kronuz> (exactly) I just know roughly
[20:17:30] <Kronuz> Stric: it's probably better two 1Ghz with 16 GB than one 1.2Ghz with 32
[20:17:38] <Kronuz> I would guess...
[20:17:53] <trygvis> most likely, but it will depend on your application
[20:17:54] <Kronuz> but the 32 GB... I might need them in the future, not sure
[20:18:08] <Stric> and adding two more 16G machines will not help?
[20:18:11] <trygvis> you could probably always sell your old 16GB
[20:18:22] <Kronuz> yeah, that's probably the best
[20:18:26] <Kronuz> to stick with the 16GB
[20:18:44] <Kronuz> adding more with 16GB would probably solve the issue too
[20:19:03] <Kronuz> but 32 sounds cool too
[20:19:05] <Kronuz> hehe
[20:19:06] <Stric> "probably" :P
[20:19:15] *** Wez has quit IRC
[20:19:22] <Stric> "it sounds cool, so I got half the performance instead"
[20:19:24] <Kronuz> and there's also the 1.0 vs 1.2 Ghz...
[20:19:40] <Stric> no, there's 16*1.0GHz vs 8*1.2GHz
[20:20:09] <Kronuz> well, yeah, actually
[20:20:22] <Kronuz> should be better, I hope
[20:20:24] <Kronuz> :P
[20:20:39] <Kronuz> hey what's with the Opteron systems?
[20:20:40] <Stric> if your stuff is single-threaded, then t2k is the wrong machine anyway
[20:20:41] <Kronuz> the two dual cores
[20:20:57] <Kronuz> they're like twice as fast aren't they?
[20:21:10] <Stric> twice as fast as... what?
[20:21:17] <Kronuz> the cpu speed
[20:21:24] <Kronuz> 2.6, 2.8 Ghz
[20:21:38] <Stric> yes
[20:21:55] <Kronuz> I'm sort of inclied to the UltraSPARC for that single benchmark I saw yesterday
[20:22:04] <regx> woohoo i install BIND9 on my t2000.. hopefully it wont be TO much
[20:22:08] <Kronuz> but I still have doubts... I've never used SPARC before
[20:22:23] <Stric> Kronuz: the T1 cpu is good at some things, opteron at others.. pick depending on what you're gonna do
[20:22:26] <Kronuz> regx: too much why?
[20:22:43] <sommerfeld> it's not meaningful to compare clock speed between differing processor implementations.
[20:22:44] <regx> im kidding. was hoping some one would be like "HAH like you have a t2000"
[20:22:45] <Stric> if you're gonna do floating point stuff, ditch t2k
[20:22:57] <sommerfeld> even intel vs amd
[20:23:02] <Kronuz> Stric: yep, I'll be doing web/database/mail serving.... and the UltraSPARC T1 is supposedly designed for that... so
[20:23:09] *** jacotton has joined #opensolaris
[20:23:13] <sommerfeld> compare benchmarks you care about
[20:23:31] <Auralis> get a try and buy machine and test YOUR workload on it
[20:23:49] <Stric> sommerfeld: well.. latest intel and amd ones are kinda ok to compare per clock speed.. don't mix netburst though..
[20:23:57] <Kronuz> Auralis: yeah, that's what the Sun guys keep telling me...
[20:24:00] <Stric> or T1
[20:24:18] <sommerfeld> it's like comparing two sports cars based on their peak rpm's alone
[20:24:26] <Kronuz> but I was thinking what's the point if there have been benchmarks (lots of them) already
[20:25:12] <Auralis> the point is that you know exactly how it behaves with your workload, which in return makes you feel better and leading to a informed buy
[20:25:16] <sommerfeld> no measure of power or work or what have you, just how fast something in the guts is spinning.
[20:27:12] *** delewis has quit IRC
[20:27:33] <Kronuz> Stric: *or* I can always get one of each (32GB and 16GB)
[20:27:52] <Kronuz> for the extra 10
[20:28:14] *** Tpenta has quit IRC
[20:29:31] <sunnyDay> does anyone knows why there is no b52 of ***opensolaris*** yet? was something changed ? according to schedule,(http://www.opensolaris.org/os/community/on/onnv_schedule.txt)  it was to be released at 11/06/2006
[20:29:48] <sunnyDay> there are snapshots though
[20:30:00] *** bougie has quit IRC
[20:30:05] <sunnyDay> but no b52 in http://dlc.sun.com/osol/on/downloads/
[20:30:42] *** bougie has joined #opensolaris
[20:31:07] *** whaq has joined #opensolaris
[20:34:24] <edgy> I noticed the same thing sunnyDay
[20:34:40] <edgy> I had to settle for only upgrading my machine to 51, instead of 54
[20:35:26] <sunnyDay> edgy, 54 seems to me still a bit far....
[20:35:50] <sunnyDay> edgy, according to schedule http://www.opensolaris.org/os/community/on/onnv_schedule.txt
[20:36:05] <sunnyDay> 12/04/2006 is the date for 54
[20:36:18] <sunnyDay> onnv_54 i mean to be precise
[20:36:49] *** _william_ has joined #opensolaris
[20:37:32] *** solarisjon has quit IRC
[20:37:53] <_william_> siur
[20:38:00] <_william_> hi all
[20:44:16] *** sommerfeld has quit IRC
[20:44:31] *** sommerfeld has joined #opensolaris
[20:44:31] *** ChanServ sets mode: +o sommerfeld
[20:44:59] *** solarisjon has joined #opensolaris
[20:45:32] *** sahafeez has joined #opensolaris
[20:59:06] <kleppari> uhm..
[20:59:11] <kleppari> using collect..
[20:59:19] <kleppari> Creating experiment database test.4.er ...
[20:59:20] <kleppari> Warning: this experiment is being recorded to a file system
[20:59:20] <kleppari> of type "zfs", which may distort the measured performance.
[20:59:22] <kleppari> anything to this?
[20:59:31] *** bengtf has quit IRC
[21:01:25] <sil3nt|warri0r> where do i find the xchat binary for solaris
[21:02:20] <gdamore> www.blastwave.com
[21:02:33] <gdamore> (pkg-get -i xchat)
[21:06:53] <sil3nt|warri0r> k
[21:07:05] *** GmanAFK is now known as Gman
[21:07:06] <Gman> hey hey hey
[21:07:18] <hile_> good morning Gman
[21:07:19] <jbk> hi
[21:07:26] <twincest> lo gman
[21:07:33] <Gman> sil3nt|warri0r, i have xchat binaries
[21:07:33] <sil3nt|warri0r> is the ctrl+alt+F1, ctrl+alt+F2 ...... works under solaris ??
[21:07:42] <Gman> [if you're still looking]
[21:07:45] *** brianwilson has joined #opensolaris
[21:07:52] <sil3nt|warri0r> how do i get
[21:08:42] <Gman> http://www.gnome.org/~gman/xchat.tar.bz2
[21:08:49] <sil3nt|warri0r> i search at blastwave , i think i didnot get xchat, but i get gaim
[21:08:52] <Gman> unpack that, and there should be binaries
[21:09:12] <Gman> though i must warn you, there is a chance it might not work, depending on what version of gnome you have installed
[21:09:15] <sil3nt|warri0r> k
[21:09:15] <sil3nt|warri0r> thx man
[21:09:33] <Gman> no worries dude
[21:09:35] <sil3nt|warri0r> i just installed solaris 10
[21:09:52] <sil3nt|warri0r> the default pkges
[21:09:58] * Gman on solaris nevada
[21:10:07] <Gman> [hence you might experience some issues]
[21:11:48] <sil3nt|warri0r> is the ctrl+alt+F1, ctrl+alt+F2 ...... works under solaris ?? like in linux
[21:12:00] <sil3nt|warri0r> the v terminal
[21:12:07] <sil3nt|warri0r> s
[21:12:22] <Stric> no
[21:13:20] <Gman> http://www.opensolaris.org/os/project/vconsole/
[21:13:42] <Gman> there's some bfu archives for beta development available
[21:13:53] <Gman> but you'd probably need to be on solaris express
[21:13:59] <sil3nt|warri0r> so how i enable it ?
[21:15:24] <Gman> you'd have to upgrade to solaris express
[21:15:44] <Gman> download the tarball, and follow the installation guide on http://www.opensolaris.org/os/project/vconsole/Downloads/
[21:17:29] <sil3nt|warri0r> k
[21:17:44] <sil3nt|warri0r> how do i upgrade to express
[21:19:01] *** hell` has joined #opensolaris
[21:19:04] <Stric> or you can use the multiplatform command screen for similar and much more functionality
[21:20:16] <sil3nt|warri0r> <Stric> how ?
[21:20:18] <hell`> anyone here ever used JRocket 1.5 for sparc compared to sun jdk
[21:20:33] <Stric> sil3nt|warri0r: by installing it.. it's probably in blastwave too
[21:20:44] <sil3nt|warri0r> k
[21:23:27] *** crash| has quit IRC
[21:24:43] *** laca is now known as lacaAFK
[21:27:00] <Kronuz> hey, what do you think of a software based array of disks vs a SAN?
[21:27:12] <whaq> what's the weakest machine that anyone is running solaris on? Pentium III like?
[21:27:20] <Kronuz> like using a Sun Fire X4500 Server with 48 SATA drives
[21:27:49] <Kronuz> 'cause the guys at Sun are offering me that configuration too
[21:28:15] <whaq> Kronuz, what's the question again? that sounds mighty nice
[21:29:14] <Kronuz> a Sun Fire X4500 with 16GB of RAM, two dual core 2.6GHz Opteron processors, and 48 500GB SATA disks
[21:29:33] <Kronuz> to be used as a storage solution using ZFS and RAID-Z
[21:30:02] <regx> OMG brandz is hott
[21:30:07] <regx> so easy too
[21:30:41] <hile_> Kronuz: that could work well depending on your requirements..
[21:31:08] <hile_> Does your backup vendor support ZFS yet?
[21:31:12] <Kronuz> hile_ but it would never match the performance a SAN could give, I suppose
[21:31:13] <regx> anyone know how to create your own OS tarball for brandz?
[21:31:30] <Stric> TSM "supports" zfs.. you can backup and restore, but ACLs are lost
[21:31:30] <hile_> Personally I prefer a SAN as well
[21:31:42] <hile_> Stric: know that one from expereince
[21:31:52] <hile_> Kronuz, I prefer a SAN as well
[21:32:10] <hile_> then it's easier for me to deport diskgroups to another host if i'm doing maintenance, or if something crashes.
[21:32:20] <hile_> or (in work's case) SRDF failover
[21:32:27] <Auralis> Kronuz: a x4500 can deliver over 2GB/sec disk performance
[21:32:49] <Kronuz> hile_: yeah, me too... tho' probably the 48 drives would have a better performance than a SAN with 5 146GB @ 15KRPM
[21:33:10] <Kronuz> 'cause I was thinking of getting a SAN with just 5 drives (to start with)
[21:33:25] <Kronuz> and the solution they're offering me has 48 SATA drives
[21:33:34] <whaq> wouldn't that be considered a san too when you set it up as iscsi target?
[21:33:41] <Auralis> whaq: yes
[21:33:42] <regx> how do i check system performance like process info specs/
[21:33:54] <Kronuz> well, I meant the "traditional" SAN
[21:34:06] <Stric> regx: psrinfo -v;prtdiag -v
[21:34:07] <Kronuz> with FC
[21:34:20] <Kronuz> tho the 6140 has 4Gb/s
[21:34:30] <whaq> Auralis, so what're the big differences? more storage(specific) related features and automation.. and perhaps better interfaces for it?
[21:34:51] <whaq> oh, the interface
[21:35:06] <Kronuz> not sure... it's pprobably better to have the StorageTek 6140...
[21:35:09] <Kronuz> aghrr!
[21:35:13] <regx> Stric, TOP shows your the PID stuff like utilization etc.. how about that...?
[21:35:14] <hile_> fuck me, did sunsolve shit itself again?
[21:35:50] <Stric> regx: either install top or try prstat
[21:35:54] <quasi> Kronuz: one very big advantage of a simple NAS compared to SAN is that you can use network jockeys to run the net rather than having someone dedicated to messing with san switches
[21:35:58] <regx> ok thanks
[21:36:13] <regx> prstat is perfect thanks man
[21:37:07] <Kronuz> quasi: would that x4500 solution be considered a NAS?
[21:37:35] <Kronuz> I've yet to fully understand the concepts of SAN vs. NAS
[21:37:46] <quasi> Kronuz: sort of - at least it wouldn't make much sense to wire it into a traditional SAN
[21:38:22] <Kronuz> quasi: yeah, it would be like having that instead of a FC more traditional SAN
[21:38:34] <whaq> san = block device, mostly not shared... nas = filesystem, mostly for multiusers
[21:38:40] <Kronuz> proly without the FC switches and HBA cards
[21:38:55] <quasi> right
[21:39:10] <quasi> did sunsolve just die?
[21:39:41] <Kronuz> whaq: if a SAN is not shared (mostly) what's the benefit vs. having an internal array?
[21:40:01] <quasi> sharing
[21:40:02] <Kronuz> whaq: or vs. having an external array that a single system/server uses
[21:40:21] <Kronuz> quasi: but whaq said they'r emostly not shared
[21:40:34] <twincest> the LUNs aren't shared
[21:40:40] <twincest> you can have many luns on one san device
[21:41:35] <Kronuz> I guess the only benefit I can see on a SAN is that you could change/move the assigned allocated spaces more easlily between servers
[21:42:06] <Kronuz> (whereas if you have drives on a per server basis more space could end up being wasted)
[21:42:33] <Kronuz> is that it? is that what a SAN is all about?
[21:42:47] <Kronuz> vs. internal or per server external storage
[21:42:54] <whaq> yea, the LUNs, not the device itself. NAS is shared amongst (end) users, whereas SAN is shared amongst servers(farm)..
[21:43:16] <quasi> twincest: clusters often share luns
[21:43:41] <quasi> whaq: I don't agree
[21:44:30] <whaq> quasi, well, those were broad generalizations...
[21:45:06] <quasi> nas can be used for the exact same as san
[21:45:27] <lasseoe> uhm not really
[21:45:40] <quasi> iscsi
[21:45:46] <lasseoe> you're not likely to get raw disks over NAS :)
[21:46:18] <whaq> san can directly replace your local harddrives..
[21:47:07] <Kronuz> quasi: what information is shared among clusters?
[21:47:20] <whaq> ..main feature over NAS may be: boot drive (without PXE or other kungfu), database access.. (what else?)
[21:47:44] <Kronuz> quasi: I was checking that the other day, but for PostgreSQL databases they can't share the files, for the operating system, they can't either, for the most part...
[21:49:01] <whaq> i think the drive sharing is not on application level, it's probably on (clustered) OS.. course there are exceptions.
[21:49:04] <quasi> Kronuz: with cluster you can
[21:49:27] <Kronuz> quasi: not the database, at least
[21:49:47] <quasi> Kronuz: well, you'd have one active node and one passive node
[21:50:16] <Kronuz> quasi: with postgresql?!
[21:50:32] <Kronuz> but it wouldn't be using the same files, would it?
[21:50:45] <Kronuz> I'm sure I read somewhere that was impossible with pgsql
[21:50:58] <quasi> Kronuz: you'd stop one instance and start the other
[21:51:01] <Kronuz> (without using a snapped FS view)
[21:51:08] <Kronuz> oh
[21:51:13] <Kronuz> yeah... or that
[21:51:22] <Kronuz> I could guess 48 x 500GB SATA drives together would have a better performance than 5 x 146GB FC @ 15K RPM
[21:51:25] <Kronuz> right?
[21:51:32] <Kronuz> but I'm thinking about the future too
[21:51:41] *** nwf has joined #opensolaris
[21:52:05] <Kronuz> 'cause eventually, don't know with how many 15K RPM drives, it will reach the performance 48 SATA can give
[21:52:14] <Stric> Kronuz: depends on how you use them.. http://blogs.sun.com/roch/entry/when_to_and_not_to for raidz for instance
[21:52:56] <quasi> yes, 48 drives would be faster, but generally I'd recommend FC for a database (and other random access) and SATA for sequential access
[21:53:21] <whaq> FC? you meant SCSI?
[21:53:23] <Kronuz> quasi: but even for database, 48 would beat the 5 15K RPM
[21:53:42] <quasi> Kronuz: possibly
[21:53:44] <Kronuz> whaq: well, they don't call them SCSI nor SAS... it says FC-AL
[21:54:01] <Kronuz> quasi: then again there's the I/O limit
[21:54:09] <Kronuz> which with the X4500 would be 2Gb/s
[21:54:23] <whaq> FC-AL? That's a drive interface?
[21:54:38] <Stric> whaq: Fibre Channel.
[21:54:42] <Kronuz> whereas in the SAN it would be 4Gb/s (even if 5 don't fill the 4Gb/s in a DB environment)
[21:54:45] <quasi> Kronuz: 5 FC-AL disks aren't likely to deliver 2Gb/s
[21:54:45] <Stric> (Arbitrated Loop)
[21:54:54] <Kronuz> yep
[21:55:01] <Kronuz> Arbitrated Loop
[21:55:05] <Kronuz> there's also the FC-SW
[21:55:19] <quasi> Kronuz: just because the fc delivers that much, it doesn't mean that the disks will
[21:55:22] <Kronuz> (and the other one, which I haven't seen in any SAN so far)
[21:55:33] <Kronuz> quasi: that's what I say
[21:55:36] <Stric> and before that, you had regular FC :)
[21:55:46] <Kronuz> quasi: probably (most likely) the 5 drives won't
[21:55:53] <libkeiser> "command queueing" in sata world is a joke. it doesn't work right because the protocol spec is fundamentally broken. as quasi said, if you need good random access, use FC
[21:55:55] <whaq> but the FC is further abstraction of the drive.. whereas the drive itself is connected to the SAN hardware thru either SCSI or SATA\PATA, right?
[21:56:30] <quasi> Kronuz: but that's for raw sequential IO - random access could well perform differently
[21:56:48] <Kronuz> whaq: that's a question for which I think we'll never have an answer... like how high is the sky... or why the sea is blue
[21:56:50] <Stric> whaq: http://en.wikipedia.org/wiki/Fibre_Channel
[21:57:05] <Kronuz> :P
[21:57:43] <whaq> Kronuz, actually.. there are answers to those questions..
[21:57:48] <Kronuz> <.<
[21:57:50] <Kronuz> >.>
[21:57:51] <Stric> think of FC as "network scsi", kinda.
[21:57:52] *** Wez has joined #opensolaris
[21:58:55] <Kronuz> quasi: yep, I'm aware of that... tho' eventually, enough FC drives would saturate the 4Gb/s
[21:59:12] <Kronuz> (the number of which would depend on the use of the drives)
[21:59:15] <whaq> Stric, exactly.. I didn't think it should be compared w/ SATA. Maybe w/ iscsi or aoe..
[21:59:25] <regx> finally some inteligent chatter in an IRC channel im in.. :)
[21:59:52] <Stric> whaq: FC disks exist (a bunch of Sun models has had them internally too), so why wouldn't it?
[22:00:15] <whaq> Stric, oh cool. I didn't know that.
[22:00:30] <libkeiser> whaq: you're oversimplifying.  it's actually very useful to compare FCP versus straight-up scsi, or sata because the protocol flow control and switch latency, fc-al overhead, etc. have a significant impact on storage system performance.
[22:00:33] * Auralis hugs her fc-al drives
[22:00:57] <Stric> fcal disks are getting uncommon nowadays and thus expensive..
[22:01:19] * Stric had to look around real good to find fcal disks for some b1ks :/
[22:01:23] <whaq> libkeiser, that makes sense. Thanks!
[22:01:33] <jbk> heh..
[22:01:43] <jbk> what was the quote about fibre channel..
[22:01:51] <jbk> 'networking designed by disk firmware writers'
[22:01:54] <whaq> Auralis, a little hot though, aren't they :P
[22:01:59] <libkeiser> jbk: hah
[22:02:16] <whaq> heh
[22:17:35] *** axisys has quit IRC
[22:18:48] <hile_> fcal drives rock
[22:23:38] *** sunnyDay has quit IRC
[22:25:25] <Kronuz> difficult decisions
[22:26:06] <hile_> hey libkeiser
[22:26:20] <Kronuz> three T2000 with 8 1.0Ghz cores and 16 GB or four with 8GB...
[22:26:36] *** [1]Cyl has joined #opensolaris
[22:26:59] <Stric> depends if you need cpu or ram more
[22:27:12] <Kronuz> exactly... and I don't know what I'll need the most :S
[22:27:35] <Kronuz> are databases more memory or cpu hungry (postgresql, that is)
[22:27:41] <Stric> IO
[22:27:56] <Kronuz> Stric: and after that?
[22:28:05] <Stric> probably cpu to drive the IO
[22:28:54] * Stric looks at a postgresql server.. <10MB ram..
[22:29:37] <Stric> maybe you can configure it to use more ram.. for something.. dunno..
[22:30:07] <Kronuz> they suggested me to leave the memory mostly to the OS and just allocate 5-10% to postgresql
[22:30:22] *** bougie has quit IRC
[22:30:29] <Stric> 10% of 16GB is about 100x more than what I'm seeing..
[22:30:38] <Kronuz> but still if I get four with 16 MB I don't know how postgresql could use that distributed power
[22:30:45] <Kronuz> the four with 8, I mean
[22:30:58] <Kronuz> (and 8GB that is)
[22:31:00] <Kronuz> :P
[22:31:14] <Stric> maybe not at all.. dunno. :)
[22:31:17] * Stric wanders off
[22:31:28] <Kronuz> Stric: why did you start talking about megabytes again! :P
[22:31:30] <Kronuz> hehe
[22:31:40] <Stric> I believe you did..
[22:31:54] <Kronuz> [15:29] * Stric looks at a postgresql server.. <10MB ram..
[22:31:59] <Stric> 22:26 (Kronuz) three T2000 with 8 1.0Ghz cores and 16 GB or four with 8GB...
[22:32:08] <Stric> :)
[22:32:13] <Kronuz> GB
[22:32:16] <Kronuz> it says GB
[22:32:27] <Kronuz> ;)
[22:32:50] <Stric> oh.. you mean that.. blaming me for being unable to differ between MB and GB.. nice :P
[22:32:57] <Kronuz> hehe
[22:33:01] <Kronuz> <.<
[22:33:04] <Kronuz> >.>
[22:33:13] <Stric> ^.\ .. what?
[22:33:22] <Kronuz> o_O ?
[22:33:45] <Stric> oh, trying to make weird smileys..
[22:36:55] <Kronuz> lol
[22:38:06] *** arnabold has joined #opensolaris
[22:38:15] *** arnabold has left #opensolaris
[22:40:38] *** regx has quit IRC
[22:46:23] <Kronuz> has anyone around used one of those Sun Application Switches ?
[22:46:35] <Kronuz> like the N1400 or the N1216 ?
[22:48:38] <Burana> Sorry for this...: Let's have a look at...
[22:48:39] <Burana> what Sun delivers...
[22:48:39] <Burana> Take:
[22:48:39] <Burana> 1 bite of an OpenSource CPU (Niagara) under GPL
[22:48:39] <Burana> 1 portion of an OpenSource OS (Solaris) under CDDL
[22:48:40] <Burana> 1 cup of java OpenSource programming language under GPL
[22:48:41] <Burana> 1 fish in a glass OpenSource Application Server under GPL
[22:48:43] <Burana> Spread it all onto the Gnome Deskop (GPL) and add
[22:48:45] <Burana> 1 dip of OpenOffice (GPL)
[22:48:47] <Burana> Wholy crap! That's the whole stack!
[22:49:49] <sommerfeld> for some applications
[22:49:52] <myrkraverk> Burana: except the e1000g NIC driver - that I can't fix myself ;/
[22:50:15] <Burana> I forgot postgresql is also included
[22:50:48] <andersmo> Sun has developers working on db.apache.org/derby/ as well, which is APL-licensed. =)
[22:50:54] <Burana> and the whole softwarestack (management software etc.) was promised to be also opensauced , if I remember correctly
[22:52:49] <Burana> at least you can download almost everything from sun for free and free use. you pay only for support
[22:53:27] *** nwf has quit IRC
[22:53:59] <dwc-> anyone else have a problem with lockd not registering with portmapper?
[22:54:16] <dwc-> (mostly on boot)
[22:54:23] <dwc-> (but sometimes when the services are restarted)
[22:55:08] *** gnu2it2 has quit IRC
[22:55:31] *** hile_ has quit IRC
[22:55:38] *** Cyl has quit IRC
[22:55:39] *** [1]Cyl is now known as Cyl
[22:59:46] *** arnabold has joined #opensolaris
[23:00:10] *** rarruda has joined #opensolaris
[23:01:31] *** djgregor has quit IRC
[23:03:49] * myrkraverk is wondering if his usb hdd is cursed ;/
[23:04:05] *** rpaulo has quit IRC
[23:04:18] <myrkraverk> after finding out how to update_drv it to work, now in wont show up in dmesg anymore ;/
[23:05:30] <myrkraverk> ah, reboot fixed that (lucily)
[23:08:36] <myrkraverk> is it possible to comment out lines in driver_aliases?
[23:08:48] *** eboutilier has joined #opensolaris
[23:09:32] <boyd> Morning, all
[23:09:36] <twincest> lo
[23:11:01] *** mrdeviant has quit IRC
[23:13:41] *** gisburn has joined #opensolaris
[23:17:00] *** LordKing has quit IRC
[23:17:25] *** nwf has joined #opensolaris
[23:18:52] <myrkraverk> alanc-away: you there?
[23:19:16] <myrkraverk> alanc-away: can you tell me where I should grap X sources to try and compile?
[23:20:08] <myrkraverk> hmm, now my usb hdd is flashing red (when it's doing stuff) is that perhaps normal?
[23:20:56] *** bobbyz has quit IRC
[23:21:23] <boyd> How on earth are we supposed to know what is normal for your USB drive?
[23:22:04] <myrkraverk> boyd: like, how on *earth* am I supposed to know? :P
[23:22:36] <myrkraverk> boyd: I'm a programmer, all this stuff called "hardware" just gets in my way ;P
[23:23:12] <myrkraverk> maybe I should stop using it, and just program with pencil and paper?
[23:24:06] *** peteh_ has quit IRC
[23:24:37] <Error_404> if you don't know anything about the hardware you're speaking to, you're a terrible programmer
[23:25:30] <myrkraverk> Error_404: as it is, I'm trying (and failing a bit) to know as little about pc hardware as I can
[23:25:50] <Error_404> fantastic, what project do you work on?
[23:26:17] *** brianwilson has quit IRC
[23:26:37] *** astinus has quit IRC
[23:26:54] <myrkraverk> I'm working on a delta compressor, makefile/build-stuff generator and a text editor atm (all sorta intertwined)
[23:27:36] <Error_404> no, i mean what's the  name?
[23:27:39] *** Burana has quit IRC
[23:28:10] <myrkraverk> Error_404: erm, I'm not too fixed on a name yet
[23:28:49] <myrkraverk> I'm not part of any publicly visible stuff (anymore)
[23:28:52] <eboutilier> myrkraverk: Re: Building X: I've never tried to build X from scratch, but the X Community webpage on OpenSolaris does talk about it. Not sure if that's exactly what you're looking for though...
[23:28:56] <eboutilier> http://www.opensolaris.org/os/community/x_win/
[23:29:05] <myrkraverk> eboutilier: thanks ;)
[23:29:10] <eboutilier> np
[23:29:11] <Gman> hey eboutilier
[23:29:13] *** djgregor has joined #opensolaris
[23:29:18] <eboutilier> Hi Gman
[23:29:27] *** djgregor is now known as djgregor_class
[23:31:42] <elektronkind> man, I'm here at SC06 and got to put my hands on the new X8000-P
[23:32:01] <elektronkind> it is indeed smaller. 3 fit in a rack.
[23:32:12] *** deedaw has joined #opensolaris
[23:32:58] <gisburn> !seen kupfer
[23:33:00] <Drone> kupfer (kupfer!i=kupfer@nat/sun/x-4f49b394ec6833e8) was last seen in #opensolaris on Wed 08 Nov 2006 23:13 GMT, saying 'hi Rich.  thanks for forwarding me the tools list.'.
[23:33:05] <Doc> X8000?
[23:33:17] <zarathustra> !seen alobbs
[23:33:18] <Drone> alobbs (alobbs!n=Alvaro at alobbs dot com) was last seen in #opensolaris on Tue 14 Nov 2006 08:52 GMT, saying 'why? :)'.
[23:33:28] <Doc> at least try and keep up with the name changes - it's an SB8000 now!
[23:33:28] <Error_404> Doc: almost twice as good as the x4100
[23:33:31] <Error_404> ;)
[23:33:59] <elektronkind> X8000/X8000P is the sun opteron blade chassis
[23:34:09] *** sahafeez has quit IRC
[23:35:03] <elektronkind> the P version is shorter than the original X8000 which sacrifices the additional row of hot-plug PCIe slots for more rack density
[23:35:04] <Doc> no, Sun Blade 8000 is the Sun Opteron blade chassis
[23:35:24] <sommerfeld> electronkind: amazing what you can do with a hydraulic press...
[23:35:41] <elektronkind> heh heh
[23:36:16] *** TpentaNB has joined #opensolaris
[23:36:28] *** TpentaNB is now known as Tpenta
[23:36:33] <elektronkind> Doc: ah right, SB, not X
[23:36:38] *** ChanServ sets mode: +o Tpenta
[23:36:47] <elektronkind> I'm just used to calling anything Opteron "X" now :)
[23:37:02] <elektronkind> http://www.sun.com/servers/blades/8000p/
[23:37:29] <Doc> ahh.. so the P has been announced? cool
[23:37:37] <elektronkind> yup. today.
[23:38:00] <Doc> biggest problem with them is that they are only N+1 power
[23:38:15] <Doc> SB8000 is N+N
[23:38:29] *** LordSky has quit IRC
[23:39:03] <elektronkind> considering the P is aimed at HPC, it probably still a consideration, but one of lower priority.
[23:39:33] *** arnabold has quit IRC
[23:40:15] *** Axposf has joined #opensolaris
[23:40:25] <Axposf> hi all
[23:40:52] <elektronkind> The only opensolaris presence here was in the form of a small pile of opensolaris stickers
[23:40:53] <elektronkind> :/
[23:40:58] *** hell` has quit IRC
[23:41:37] *** pikapika has quit IRC
[23:42:14] *** rpaulo has joined #opensolaris
[23:42:43] *** rpaulo has quit IRC
[23:44:15] *** rpaulo has joined #opensolaris
[23:51:02] *** pikapika has joined #opensolaris
[23:54:10] <Gman> stevel, i saw your account request for jds svn - i'm not allowed to approve them apparently, but i'll ping dermot
[23:54:21] <stevel> my account request?
[23:54:51] * stevel doesn't need to request access...
[23:54:58] * stevel just grants himself access ;-)
[23:55:05] <stevel> (not that i'm high on power or anything)
[23:55:40] <Gman> anyone who affiliates with the jds project basically comes up on our list :)
[23:56:09] <stevel> odd. when did i request affiliation?
[23:56:24] <Gman> dunno :/
[23:56:36] <stevel> huh. well indeed i am an observer
[23:56:37] <Gman> oh, any chance you have time to prod grommit?
[23:56:44] <stevel> sure. what's wrong?
[23:57:00] <Error_404> stevel: not power, just pot
[23:57:05] <Gman> pos is failing to update
[23:57:13] <Gman> stevel, looks like a permissions problem
[23:57:26] *** onlinebacon has joined #opensolaris
[23:57:29] <Gman> find : cannot get the current working directory
[23:57:30] <onlinebacon> hey
[23:57:34] <Gman> hi onlinebacon
[23:57:45] *** rarruda has quit IRC
[23:57:53] <onlinebacon> would samba let my dad be able to browse the files on my computer from his windows machine?
[23:58:11] <twincest> yes
[23:58:16] *** asdf__ has joined #opensolaris
[23:58:22] <onlinebacon> excellent
[23:58:24] <twincest> samba is (another other things) a CIFS server, Windows' version of NFS
[23:58:38] <onlinebacon> is there any way he would be able to control my computer from within windows?
[23:58:43] <onlinebacon> my comp is running solaris
[23:58:49] <twincest> you could use VNC
[23:58:57] <twincest> or just ssh
[23:59:08] <onlinebacon> lol, I forgot about that
[23:59:19] <onlinebacon> thanks
[23:59:31] <onlinebacon> but i gota go now, have to be off the comp by 11 and its 10:59
[23:59:32] <onlinebacon> thanks again
[23:59:33] *** onlinebacon has quit IRC

top