[00:02:02] *** stevel has joined #opensolaris
[00:02:02] *** ChanServ sets mode: +o stevel
[00:05:05] <onbot> commit by Tom Erickson:  6510395 Add "-target 1.5" to javac build
[00:06:18] <Error_404> that's part of O/N?
[00:07:20] <richlowe> No.
[00:07:31] *** ttw has joined #opensolaris
[00:07:34] <richlowe> it's a crappy synopsis.
[00:07:55] <richlowe> Error_404: "to javac arguments", is what it actually means.
[00:08:18] *** Tpenta changes topic to "Latest builds: SXCR: 54 | ON: 55"
[00:08:34] <ttw> my netra had NetBSD on it and when trying to install OpenSolaris I constantly get invalid disklabel messages. tried formatting but doing something wrong.  can anyone assist?
[00:09:01] <stevel> happy birthday alan :)
[00:09:08] <stevel> how old are you now?
[00:09:21] <tomww> is asking allowed?
[00:09:36] <Tpenta> 42
[00:09:51] <Tpenta> it was actually yesterday; ok it's still the 9th in the US, but it's the 10th here
[00:11:23] <stevel> well happy meaning of life birthday to you then
[00:12:25] <Tpenta> :)
[00:12:39] <Tpenta> oh stevel, i put my b55 stuff up the other day
[00:12:59] <stevel> yup, saw that
[00:14:34] <sommerfeld> ttw: still there?
[00:14:51] <ttw> somerfeld: yup, and struggling
[00:15:08] <ttw> ... trying to 'dd' the disk with zeros but keeps reading the backup copy
[00:15:22] <ttw> ... think i'm going to have to do the whole disk
[00:15:27] <Tpenta> bill: sorry couldn't resist the horrid "rings a bell" pun in the keyboard tones case
[00:15:57] <ttw> ... unless you have a brilliant idea ... ? ? /
[00:16:24] <richlowe> Tpenta: would that happen to be from 2007?
[00:17:14] <ttw> sommerfeld: you were just taunting me ... weren't you ? ... that's not nice ;)
[00:17:30] <Tpenta> richlowe: 2007/019
[00:17:36] <richlowe> Thanks.
[00:18:27] <Auralis> ttw: boot the system from the install media to single user mode, boot -s, boot net -s  etc. then run format, select the drive, label it, yes, save and all that jazz, and exit the shell and let the boot continue
[00:18:52] <richlowe> Tpenta: now I know there *are* cases, I'll ping plocher to find out why there's no 2007 index in the ARC community :)
[00:19:03] <Tpenta> I was just thinking that
[00:19:28] <ttw> auralis: tried that ... the partitions are all wrong ... or something ... will try again ... one sec
[00:19:29] *** pitty has joined #opensolaris
[00:20:25] * Tpenta just posted the question to arc-discuss
[00:20:50] <ttw> auralis: nope, no luck ... just got a partition with blank stuff and system that doesn't recognise it probably need to set slice 2 to the entire disk, eh?
[00:21:11] <jamesd> ttw, you may need to run   format -e  to get rid of a non solaris label or one created with  zfs
[00:21:30] <ttw> jamesd: ok, that sounds very promising ... executing ...
[00:21:57] <richlowe> Tpenta: doh, I'd intended to Cc arc-discuss and missed.
[00:22:04] <Tpenta> :)
[00:23:30] *** ojpitre has joined #opensolaris
[00:23:31] <ttw> jamesd: ok, looks like i had to switch to SMI label to erase and reset but appears to have worked. any opinion on whether i should switch back to EFI?
[00:23:58] <Tpenta> the case in question is about implementing KIOCMKTONE and the mentioned that it takes an arg of teh frequency, which brought to mind 2006/624, which was about setting beeper frequency. My question (which was answered satisfactorally) was about the two thing working together. Then I made the horrid pun "just wanted to make sure the everything works together,  seeing stuff about beeps and tones rang a bell with me for that previous  case"
[00:24:01] <jamesd> ttw, the solaris installer doesn't like the efi label...   smi is what you need
[00:24:24] <sommerfeld> ttw: sorry, too many things attracting my attention righ tnow.
[00:24:29] <ttw> oh, ok ... need to look up the difference on that stuff ... i guess the EFI label came from netbsd install
[00:24:47] <ttw> sommerfeld: no, worries, i appreciate the thought anyway
[00:24:58] <sommerfeld> in general, EFI labels are only used with (a) large disks or (b) disks used for ZFS alone.
[00:25:08] <sommerfeld> where large is "over 1 TB"
[00:25:25] <ttw> ... looks like jamesd sorted me out anyway.  thanks
[00:25:26] <_syphilis_> i'd like to see EFI become standard, personally
[00:25:32] <sommerfeld> which really means "large virtual disks implemented by intelligent storage arrays"
[00:25:45] <jamesd> ttw, efi labels ar new...   not sure if  netbsd is using them or if they are just using a label that is foreign enougg to solaris to be considered non-smi that causes the problem
[00:25:48] <sommerfeld> but real >1TB disks are coming.
[00:25:54] <Tpenta> well, didnt the story break the other day about the first terabyte disk?
[00:26:00] <ttw> and the installer won't read them??  ... the EFI labels, that's weird isn't it?
[00:26:21] <sommerfeld> I think NetBSD attempts to use nativ SMI labels but it may be just wrong enough to cause trouble.
[00:26:22] <jamesd> Tpenta, yeap... hitachi is going to release one soon and it will be $400 usd  each.
[00:27:06] <jamesd> ttw, the solaris installer is as old as the hills...   so its kind of stuck in its ways...  its slated for a major overhall shortly
[00:27:08] <Tpenta> about 40c/gb
[00:27:37] <ttw> like most *nix installers unfortunately
[00:27:54] <kjetilho> jamesd: does that mean the pkg-tools will be overhauled, too?
[00:28:09] <kjetilho> (it seems to me they are to blame for the slowness of the install.)
[00:28:22] <jamesd> kjetilho, no, that is not on the list.. its going to get a speed increase and will support ZFS on /
[00:28:43] *** lloy0076 has joined #opensolaris
[00:28:44] <kjetilho> ok, that's good :)
[00:29:19] * dlg want solaris with an openbsd like installer
[00:29:32] * jamesd can't wait to see what all becomes possible when  zfs on / is supported... it may bring a whole new meaning to live upgrades...
[00:29:51] <dlg> it should also make installs on the x4500 more fun
[00:30:04] * Tpenta can't wait, thos 6gb root partitions I have are starting to look small
[00:30:34] <jamesd> point the installer at a clone of / and tell it to upgrade... when done.. reboot the cloned /   and  have a new version...  upgrades will only require a reboot then no booting solaris install media..
[00:31:09] * jamesd is only dreaming but it does sound realistic.
[00:31:19] <asyd> :)
[00:31:35] <kjetilho> and obviously very useful for patches which should be installed in single-user
[00:32:45] <jamesd> yeap and backing out bad patches becomes a 2 second process.    zfs rollback  root@patch1234567
[00:32:48] <kjetilho> or perhaps not -- since you lose other changes made before reboot.
[00:33:06] <kjetilho> you'll need to be very careful about what you snapshot
[00:33:37] <dlg> pkg tools are a better solution for patch mgmt
[00:33:54] <jamesd> kjetilho, the root filesystem should not change normally besides patches....  you won't be able to install new software and /etc should be a seperate filesystem... but the rest should be very stable.
[00:34:10] <lloy0076> What's the major block on enabling ZFS as a root file system for Solaris? (Apart from lack of time and resources)
[00:34:25] <dlg> getting all the bits of the root fs on bootable disks
[00:34:29] <jamesd> lloy0076, patches, the installer,  packages, etc...
[00:34:30] <dlg> and making the bootloader understand it
[00:34:33] <dlg> im guessing
[00:34:37] <kjetilho> dlg: well, I kind of like the incremental (differential) nature of patches
[00:34:56] <kjetilho> in Linux, I have to download 40 MB of data when a single file in OpenOffice.org needs changing
[00:34:59] <dlg> kjetilho: true, but fs snapshots are incremental
[00:35:18] <dlg> what if the 3rd patch of 5 is borked and you need to back it out independantly of the last two?
[00:35:28] <dlg> i mean, fs snaps are linear
[00:35:39] * Auralis waits for the first person snapshoting / and doing a upgrade install in place
[00:36:14] <jamesd> Auralis, couldn't be that much more difficult that  lupgrades.
[00:36:29] <lloy0076> This isn't going to be in terribly correct technical terms, but...
[00:36:32] <jamesd> except its on a cloned filesystem instead of a sepearate slice.
[00:36:48] <lloy0076> Is it possible to "give" an X server running in a Solaris Zone access to the real hardware?
[00:36:51] * lloy0076 ponder
[00:37:08] <lloy0076> ...without using the X protocol (i.e. as though it is the root X server)
[00:37:11] <jamesd> lloy0076, yes, but its a lot of hassle...
[00:37:17] <Auralis> jamesd: yes and i bet people will want to do exactly that
[00:37:34] <kjetilho> jamesd: you may perhaps zfs send the diff made since the patch snapshot was made so you can reconstruct if someone _did_ change /usr?
[00:37:45] <jamesd> lloy0076, its better just just to make a remote connection to a Xserver running in that zone.
[00:38:20] * Auralis has the feeling lloy0076 is thinking more along the lines of a multi seated use of a single box
[00:38:35] <jamesd> kjetilho, that sounds painful, i think the system would have to lock out changes made to /usr while an upgrade is occuring
[00:38:43] <lloy0076> I'm trying to work out if one could basically use the global zone as a "dumb host".
[00:38:57] <lloy0076> And use another zone with a real X running in it as the one you use predominantly.
[00:39:13] <jamesd> lloy0076, i read in a mailing list that its possible, but the process did seem a bit painful.
[00:39:23] <lloy0076> Ok :)
[00:39:46] *** hspaans has joined #opensolaris
[00:39:56] <hspaans> g'day to all
[00:40:08] <ttw> where's the best place to get an overview of zone and their config? docs.sun.com ??
[00:40:36] <sommerfeld> dlg: well, last I checked, zfs for "/" was going to require that the boot/root pool consist of a single mirror
[00:40:48] <sommerfeld> so you'd likely end up with two pools on an x4500
[00:40:55] <kjetilho> ttw: BigAdmin first, I guess
[00:40:59] <jamesd> ttw, http://uadmin.blogspot.com/2005/03/zones-links.html   links to the major cool documentation
[00:41:06] <ttw> cheers
[00:41:19] <sommerfeld> one for boot/root, the other for vast tracts of data.
[00:41:25] <hspaans> btw some sun studio gods arround?
[00:41:28] *** mihaic has joined #OpenSolaris
[00:49:40] *** comay has quit IRC
[00:52:41] *** Mazon is now known as mazon
[00:53:15] *** BadKarma has quit IRC
[00:53:26] *** BadKarma has joined #opensolaris
[01:05:27] <_syphilis_> can vacation(1) handle UTF-8 encoded messages?
[01:05:49] *** ojpitre has quit IRC
[01:12:29] *** sahafeez has quit IRC
[01:12:52] *** ttw has quit IRC
[01:21:42] *** mrdeviant has joined #opensolaris
[01:21:47] *** uebayasi has joined #opensolaris
[01:25:03] *** dclarke has joined #opensolaris
[01:25:23] *** ShadowHntr has joined #opensolaris
[01:27:47] *** bbtm has quit IRC
[01:28:55] *** jsubl2 has joined #opensolaris
[01:29:41] <Theoden-Nexenta> Hey jsubl2
[01:29:52] <jsubl2> howdy Theoden-Nexenta
[01:31:08] <jsubl2> I work from home on Wednesday.. ssssoooo I am relaxing with a few cold ones
[01:34:30] <mihaic> I like them alive myself... but meh, whatever floats your boat ;P
[01:35:32] *** comay has joined #opensolaris
[01:36:10] *** comay has quit IRC
[01:36:41] *** comay has joined #opensolaris
[01:37:07] *** comay has quit IRC
[01:38:28] <_william_> gn
[01:38:29] *** _william_ has left #opensolaris
[01:40:49] *** Kmays has joined #opensolaris
[01:41:34] *** hspaans has quit IRC
[01:41:49] *** comay has joined #opensolaris
[01:53:09] *** comay has quit IRC
[02:00:02] *** darrenm has quit IRC
[02:00:03] *** BadKarma has quit IRC
[02:00:05] *** peteh has quit IRC
[02:00:51] *** peteh has joined #opensolaris
[02:01:22] *** karrotx has joined #opensolaris
[02:01:48] *** adam_bin has joined #opensolaris
[02:01:54] *** karrotx has quit IRC
[02:01:57] *** karrotx has joined #opensolaris
[02:05:13] *** BadKarma has joined #opensolaris
[02:08:06] * Tpenta wonders just how long a complete jds build will take on his ferrari
[02:09:12] <mihaic> Everything takes just as long overnight :P
[02:09:38] <kjetilho> what's your codage on that beast in kLOC/s ?
[02:10:01] <Tpenta> no idea
[02:10:07] <Tpenta> i dont normally use it for building
[02:10:11] <Tpenta> well not stuff of that size
[02:10:50] <mihaic> Expect some 10 hours :]
[02:10:59] *** dclarke has quit IRC
[02:11:14] <mihaic> IO may not be great on the Ferrari. That, and if you're building Gnome, Firefox, Evolution and Thunderbird...
[02:12:42] <Tpenta> might be a while, ... o well, what I really want uis the fixed gst-python stuff and gstreamer stuff so I can try to play with jokosher now that they've fixed it to use a more generic output device
[02:12:45] <jamesd> everything can be instaneous if you have enough boxes tied together with dmake and enough ram and bandwidth.
[02:13:00] <_syphilis_> whee
[02:13:08] * _syphilis_ liberates a v40z to learn jumpstart with
[02:16:49] <jamesd> _syphilis_, you could just as easily liberate a  ss5 and learn jumpstarting, or  a u2 if you want 100mbit.
[02:17:08] <_syphilis_> we don't have any ss5's or u2's
[02:21:43] *** simford has joined #opensolaris
[02:24:18] *** ShadowHntr has quit IRC
[02:31:41] *** adam_bin has quit IRC
[02:49:25] *** IvanR_ has joined #opensolaris
[02:53:55] *** Theoden-Nexenta has left #opensolaris
[02:57:42] <_syphilis_> odd, pfinstall(1M) refers to "Solaris 2.10"
[03:02:13] *** comay has joined #opensolaris
[03:06:17] *** lacaAFK is now known as laca
[03:19:41] *** yongsun has joined #opensolaris
[03:20:05] *** Theoden has joined #opensolaris
[03:20:34] <Theoden> Can someone tell me how to remove a directory the refuses to delete?
[03:20:34] <Theoden> rmdir <dir>  replies - menu-xdg/: File exists
[03:20:35] <Theoden>
[03:20:52] <_syphilis_> is the directory empty?
[03:23:25] <Theoden> yes
[03:23:28] <AbeFroman> one does not simply march into #opensolaris!
[03:23:37] <AbeFroman> oh wait, wrong guy
[03:24:35] *** comay has quit IRC
[03:26:10] *** MikeTLive has joined #opensolaris
[03:26:34] <Theoden> ????
[03:27:45] <MikeTLive> RA
[03:29:27] *** ShadowHntr has joined #opensolaris
[03:30:18] <ShadowHntr> i got a question re: Sun Ultra 5/10
[03:30:27] <ShadowHntr> i'm reading the sun system handbook
[03:30:41] <ShadowHntr> and the processor support on the 5 vs. 10 differ
[03:30:45] <ShadowHntr> can the Ultra 5 desktop support a 440MHz proc module?
[03:30:52] <ShadowHntr> it lists as compatible with the Ultra 10.
[03:31:15] <pitty> ShadowHntr: sure it can
[03:31:26] <Tpenta> it's the same board
[03:31:46] <ShadowHntr> i just wanna make sure i've done my research before i start buying loads of shit for this system
[03:31:51] <delewis> when it doubt compare the FRU numbers.
[03:31:51] <ShadowHntr> i'm sure you understand. :)
[03:32:14] <delewis> not that I've looked, but I imagine the FRU numbers on the procs for the Ultra 5 and Ultra 10 match up 1-1.
[03:32:16] <kjetilho> the main difference is that you need to remove the floppy on the 5 to fit big DIMMs
[03:32:27] <ShadowHntr> yeah i saw that
[03:32:32] <ShadowHntr> when i replaced some DIMMs in it
[03:32:35] <ShadowHntr> but that's alright
[03:32:41] <ShadowHntr> i don't use the bugger anyway lol
[03:33:28] <Theoden> Well - nice that someone gets help
[03:33:49] <ShadowHntr> appreciate the help =)
[03:34:00] <kjetilho> Theoden: it's not empty, it's just your imagination :)
[03:34:08] <ShadowHntr> actually i've seen some pretty good deals on ebay for sun fire v120s... :)
[03:34:25] <Theoden> kjetilho: No - it's empty - I know when a dir is empty.
[03:34:25] <kjetilho> touch ". " is a nice way of confusing people
[03:35:03] <ShadowHntr> LOL
[03:35:08] <kjetilho> try stripping the trailing slash.  oooold GNU rm mishandled that.
[03:35:15] <sommerfeld> ShadowHntr: most likely the configuration hasn't been tested by sun, and the hw folks are reluctant to say something works until they actually try it :-)
[03:35:20] <ShadowHntr> gotcha.
[03:35:25] <Theoden> kjetilho: I did that too.
[03:35:38] * ShadowHntr hugs his Ultra 5 that he got for such a great deal... :)
[03:35:41] <ShadowHntr> at the time
[03:35:48] <ShadowHntr> ultra 5s were selling for almost 200 on ebay
[03:35:50] <ShadowHntr> i got mine for 40
[03:35:52] <ShadowHntr> untested
[03:35:55] <ShadowHntr> tested it fine =)
[03:36:09] <pitty> Theoden: do you have lsof installed? you can lsof /dir to see if any stale inodes are still referencing it
[03:36:10] <delewis> it's been awhile since Ultra 5s sold for $200 :-)
[03:36:16] <sommerfeld> ShadowHntr: 440 could work, or it could suck more juice/put out more heat than the U5 enclosure could cope with..
[03:36:18] <kjetilho> just yesterday the last few Ultra 5 hit the dumpster ...
[03:36:28] <kjetilho> way overdue, really
[03:36:43] <ShadowHntr> sommerfeld: it can handle the extra power, but not sure of the heat...
[03:36:59] <pitty> Theoden: otherwise, while not recommeneded alway, you can run 'unlink $dirName'
[03:37:24] *** jsubl2 has quit IRC
[03:38:04] <kjetilho> Theoden: run: perl -e 'opendir(D, "empty"); print join("|", "", readdir(D), ""), "\n"'
[03:38:18] <kjetilho> replace "empty" by the directory name
[03:39:15] <Theoden> k
[03:40:21] *** jsubl2 has joined #opensolaris
[03:40:35] <pitty> talk about overkill
[03:42:14] *** stevel has quit IRC
[03:44:56] <Theoden> unlink did it.
[03:45:18] *** stevel has joined #opensolaris
[03:45:18] *** ChanServ sets mode: +o stevel
[03:45:19] <Theoden> pitty: Does that remove it permanently? or will itr com back at reboot?
[03:45:32] <jsubl2> its gone
[03:45:42] <Theoden> jsubl2: thx.
[03:46:33] <jsubl2> unlink does no error checking
[03:47:33] *** stevel has quit IRC
[03:48:00] <alanc> unlink directories on UFS usually results in you needing to fsck
[03:50:03] <kjetilho> but the log will be clean, so Solaris will think no fsck is needed
[03:50:44] <Theoden> How do I force that in opensolaris?
[03:51:02] <kjetilho> umount it and fsck manualy
[03:51:18] <Theoden> Can you force it at reboot - like in linux?
[03:52:21] <jsubl2> init 1   then unmount disk and fsck
[03:52:55] <jsubl2> df -h to see your disk if you need a reminder
[03:53:04] *** uncertainty_ has joined #opensolaris
[03:53:22] <jsubl2> control d when done brings sys back up
[03:55:35] *** adam_bin has joined #opensolaris
[03:56:52] *** slowhog has quit IRC
[03:56:58] *** deedaw has quit IRC
[03:57:22] <_syphilis_> 'kernel /multiboot' when using sun-supplied pxegrub always gives 'Error 13: Invalid or unsupported executable format', but i'm fairly sure the multiboot file is valid.. what might cause that?
[03:57:31] <_syphilis_> (loading the file from tftp)
[03:58:12] <kjetilho> are you sure it's the right architecture?
[03:58:33] <_syphilis_> 'file' identifies it as an x86 executable (and i don't believe sparc uses multiboot..)
[03:59:29] *** sommerfeld has quit IRC
[03:59:47] <_syphilis_> hmm, google suggests grub might dislike the tftp server
[04:02:01] *** jsubl2 has quit IRC
[04:02:09] *** Griffous has joined #opensolaris
[04:02:12] *** Gr|ffous has quit IRC
[04:07:42] *** peteh has quit IRC
[04:07:58] <_syphilis_> aha, -B option to tftpd broke it
[04:08:31] *** peteh has joined #opensolaris
[04:09:21] *** uncertainty has quit IRC
[04:10:34] *** yongsun has quit IRC
[04:12:54] *** jsubl2 has joined #opensolaris
[04:13:03] *** simford has quit IRC
[04:13:23] *** simford has joined #opensolaris
[04:13:55] <jsubl2> to fsck the root disk you have to boot cd#1 right?
[04:14:14] <_syphilis_> no, you can do it from single user
[04:14:27] <_syphilis_> reboot immediately if it changes anything, and have / mounted ro while you do it
[04:15:33] <jsubl2> how do you have / mount ro
[04:16:06] <_syphilis_> mount -oremount,ro / should work
[04:16:15] <_syphilis_> if your / is damaged it may well be ro already
[04:16:33] <_syphilis_> (if it dropped you into maintenance on boot - look at 'mount' to see)
[04:16:34] <jsubl2> oo.  that was the command i could not seem to come up with:  mount -oremount,ro /
[04:16:43] <jsubl2> will have to try that.  thanks _syphilis_
[04:17:19] <jsubl2> the man page is kinda funky in the remount area on my sys.  sxcr 54
[04:17:29] *** xinkeT has quit IRC
[04:17:46] *** mihaic has quit IRC
[04:17:57] *** mihaic has joined #OpenSolaris
[04:19:08] <_syphilis_> hmm, apparently the default installer option on a serial console is to start a GUI install :)
[04:21:21] *** mihaic has quit IRC
[04:22:16] <lloy0076> :P
[04:22:24] <lloy0076> X Over Serial Cable
[04:22:32] *** xinkeT has joined #opensolaris
[04:26:15] *** mihaic has joined #OpenSolaris
[04:28:30] *** Error_404 has quit IRC
[04:31:07] <jmcp> lloy0076: you've never run an XTerm over 9600 baud then
[04:31:14] *** Theoden-Nexenta has joined #opensolaris
[04:31:23] <lloy0076> Nope
[04:31:41] <jmcp> painful, but doable :)
[04:32:56] * boyd has a serial terminal at home that's capable of being driven by an X server. He doesn't have the X server though
[04:37:09] *** broadcast has joined #opensolaris
[04:38:11] <_syphilis_> i have an old svr3 system with a GUI that works on both the graphical console and a text console.. it's sort of neat
[04:38:20] *** yongsun has joined #opensolaris
[04:39:16] *** jsubl2 has quit IRC
[04:43:14] *** jsubl2 has joined #opensolaris
[04:46:23] *** Error_404 has joined #opensolaris
[04:46:39] *** miffe has quit IRC
[04:55:27] *** jsubl2 has quit IRC
[05:08:18] *** freakazoid0223 has quit IRC
[05:09:32] *** g4lt-U60 has quit IRC
[05:09:42] *** g4lt-U60 has joined #opensolaris
[05:11:26] *** freakazoid0223 has joined #opensolaris
[05:12:39] *** alanc is now known as alanc-away
[05:27:28] *** adam_bin has quit IRC
[05:30:03] *** laca has quit IRC
[05:41:40] *** mrdeviant has quit IRC
[05:42:31] *** prg3 has quit IRC
[05:46:05] *** IvanR_ has quit IRC
[05:46:58] *** prg3 has joined #opensolaris
[05:53:56] *** sparc-kly_ has quit IRC
[05:54:10] <_syphilis_> Fun Fact: 'sconadm register' crashes on systems with ipv6 addresses configured
[05:55:57] <Error_404> ugh, this UNIX beard is getting out of hand, i gotta shave
[06:03:20] *** abackos has joined #opensolaris
[06:06:42] *** dlynes_laptop has joined #opensolaris
[06:21:28] <jmcp> Doc: ping!
[06:31:07] *** abackos has quit IRC
[06:34:14] <_syphilis_> hmm, sun manuals can be a bit long winded at times
[06:42:48] <lloy0076> _syphilis_: Maybe so, but they're terribly useful.
[06:43:07] <_syphilis_> i agree, i just wish i could spend less time reading the manual than actually doing what i want to do :)
[06:43:41] <lloy0076> heh
[06:43:52] <lloy0076> I wish that regardless of how long or how well the manuals are.
[06:49:53] *** IvanR_ has joined #opensolaris
[06:50:09] <dlg> woo, mfi seems stable at last!
[06:50:46] <jmcp> dlg: congrats!
[06:51:31] <dlg> https://svn.itee.uq.edu.au/viewvc/mfi/mfi.c?r1=72&r2=74
[06:51:37] <dlg> i think thats the fix
[06:52:50] <jmcp> that's it? geez :)
[06:53:03] <dlg> yeah
[06:53:04] * dlg cry
[06:53:08] <dlg> what a waste of time
[06:56:10] <_syphilis_> huh, nice, free solaris media
[06:58:20] *** janus_man has quit IRC
[06:58:49] *** janus_man has joined #opensolaris
[07:08:44] *** bougie has joined #opensolaris
[07:08:58] <Error_404> _syphilis_: that program's been going a while
[07:09:38] <Error_404> i got my copy about a month ago
[07:13:02] *** simford has quit IRC
[07:13:24] *** simford has joined #opensolaris
[07:24:43] *** alo has quit IRC
[07:33:50] *** loke_ has joined #opensolaris
[07:37:23] *** loke has quit IRC
[07:43:44] *** lloy0076 has quit IRC
[07:48:11] *** lloy0076 has joined #opensolaris
[07:52:50] *** bougie has quit IRC
[07:53:34] *** estibi has quit IRC
[07:54:51] *** lloy0076 has quit IRC
[07:55:12] *** whaq has quit IRC
[07:57:40] *** dlynes_ has joined #opensolaris
[08:04:05] *** Starless_ has joined #opensolaris
[08:05:20] *** dunc has quit IRC
[08:05:30] <Starless_> I have a hard drive with a solaris partition and a fat32 partition; can I convert the fat32 partition to zfs without reinstalling everything; or do I have to reinstall everything and make a zfs filesystem inside of a primary solaris partition?
[08:06:50] <Starless_> This is on sxcr 53, x86, if it matters
[08:07:27] <tsoome> reinstall...
[08:09:06] <Starless_> I was afraid that was the case. Oh well. Thank you for your help
[08:10:17] *** Starless_ has left #opensolaris
[08:10:42] <Error_404> ? he doesn't need to reinstall
[08:10:59] <Error_404> just nuke the dos partition
[08:14:14] <Tpenta> has anyone here managed to build jds as outlined in the jds project? I keep getting errors trying to find SUNWlibpopt-manpages-0.1.tar.gz
[08:14:41] *** dlynes_laptop has quit IRC
[08:19:55] <tsoome> hm? 2 solaris partitions are supported now?
[08:24:52] *** ShadowHntr has quit IRC
[08:30:02] <Error_404> ?
[08:30:15] <Error_404> does zfs care what partition type it is?
[08:30:20] <Error_404> i've never bothered to check
[08:38:07] <Error_404> I feel dirty....
[08:38:20] <Error_404> I just applied at a "PC Technician" firm
[08:38:26] <Error_404> specializing in spyware removal
[08:38:46] * Error_404 is worth 10x as much as they're capable of offering him
[08:39:57] <tsoome> :D
[08:40:10] * tsoome is priceless
[08:40:11] <tsoome> :D
[08:40:20] <Error_404> I'm broke & hungry
[08:43:39] <Error_404> also, if i get the job i'll have to break my 7 year record for not having touched windows
[08:45:30] <dwc-> a lot has changed since nt4
[08:45:46] <dwc-> and 98se
[08:46:21] <Error_404> I've had no reason to use it this far
[08:46:36] <Error_404> it can change all it likes, doesn't make it useful to me
[08:47:17] <dwc-> I didn't say it would
[08:50:12] <Error_404> in the application i made sure to highlight the hell out of my unix experience, so maybe they'll decided to diversify & hire a cheap unix hand (me)
[08:50:39] <Error_404> or perhaps i'm just hoping i won't have to remove spyware from someone who doesn't know how to properly surf pr0n's computer
[08:55:11] *** Cyrille has joined #opensolaris
[08:56:42] *** dunc has joined #opensolaris
[08:58:26] *** gombadi has joined #opensolaris
[08:59:35] <dwc-> a futile hope, if there is one
[09:00:14] *** gombadi has quit IRC
[09:00:31] <Error_404> old windows-hands tend not to understand UNIX, nor do they want anything to do  with it
[09:01:56] *** tsoome has quit IRC
[09:02:15] *** tsoome has joined #opensolaris
[09:03:29] <dwc-> the same could be said of unix-hands and windows
[09:05:41] *** gombadi has joined #opensolaris
[09:08:03] *** mazon is now known as Mazon
[09:09:04] <Error_404> yeah, but that's because it's rubbish
[09:09:05] <Error_404> ;)
[09:11:58] *** tsoome_ has joined #opensolaris
[09:20:30] *** lisppaste3 has quit IRC
[09:24:07] *** lisppaste3 has joined #opensolaris
[09:31:05] *** gombadi has left #opensolaris
[09:34:25] *** LordKing has joined #opensolaris
[09:39:24] *** pikapika has quit IRC
[09:39:54] *** alobbs has joined #opensolaris
[09:42:33] *** Belgar has joined #opensolaris
[09:47:23] *** MattMan has joined #opensolaris
[09:59:06] *** xushi_ has joined #opensolaris
[10:01:42] *** bunker_work has joined #opensolaris
[10:10:06] *** alanc_away has quit IRC
[10:10:27] *** alanc_away has joined #opensolaris
[10:10:47] *** logic has joined #opensolaris
[10:13:46] *** Dar_HOME is now known as Dar
[10:16:47] *** smellyfish has joined #opensolaris
[10:18:54] *** mega has joined #opensolaris
[10:19:27] *** smellyfish has quit IRC
[10:26:36] *** Fish has joined #opensolaris
[10:27:48] <raph_ael> hello
[10:28:08] <Fish> hello
[10:30:09] <raph_ael> :)
[10:30:30] <Belgar> ehlo
[10:37:38] *** xushi_ has quit IRC
[10:44:55] <Error_404> Hmm... in theory there's a way to pull deltas off a zfs, correct.... just what's changed since last snapshot
[10:45:37] <PerterB> zfs send?
[10:50:57] *** simford has quit IRC
[11:01:32] *** yongsun has quit IRC
[11:05:55] *** |tsoome| has joined #opensolaris
[11:05:55] *** tsoome has quit IRC
[11:08:01] *** gallium has joined #opensolaris
[11:13:08] <Peanut> If a motherboard is listed as 'supported' on the Solaris/Express HCL, and it has an onboard Nvidia, what are the changes of getting OpenGL to work on it in Nevada and get a decent framerate out of Stellarium? (MB is Asus M2NPV-VM). Or is OpenGL-support something that just depends on Xorg?
[11:14:09] <Griffous> there is an incremental mode for zfs send I believe
[11:14:28] <Griffous> Peanut, I would think so
[11:14:37] <Berny> Peanut: if you use the nvidia xorg driver, chances re pretty good
[11:14:49] *** mega has quit IRC
[11:15:00] <Griffous> really, onboard or not doesn't matter all that much, especially for video cards. Nics are sometimes a bit different, but I imagine you'll be ok
[11:15:22] <Peanut> *nod* I don't expect there are any true open-source opengl video-drivers out there.
[11:15:39] <Griffous> at worst, you may need to manually add the pci ids of your onboard card to the /etc/drive_aliases file so that the driver will recognise it
[11:15:58] <Griffous> though confusing at first, it's not actually that hard
[11:16:07] <Peanut> *nod* that doesn't scare me much :)
[11:18:38] * Peanut has sort of pieced together what he wants - Asus M2NPV-VM, Athlon 5200+, 2GB DDR2-800, 2x 400GB Sata-300, box, psu - Eur. 1008. A somewhat comparable U20 is twice that amount :(
[11:18:52] <Griffous> Error_404, it's a sad day for me too, I'm in the process of pulling ubuntu off my parent's PC, and putting winxp back on again. gutted
[11:18:57] <Griffous> at least I tried though eh?
[11:20:20] *** broadcast has quit IRC
[11:20:43] *** broadcast has joined #opensolaris
[11:21:22] * Peanut finds the HCL next to useless though :-( Sure, 'reported to work'. - Does sound work? Does SATA-300 work, or actually only PATA? OpenGL? Do you need any nifty resolution tricks because the video bios is buggered? Have fun finding out...
[11:22:34] <mlh> www.sunsource.net needs attention
[11:23:07] <mlh> latest news is almost a year old
[11:23:17] <Griffous> Peanut, yeah it's a bit tough
[11:23:48] <Griffous> though is it just the video that you are worried about? That seems a smaller concern in the overall picture of this new machine, surely?
[11:24:14] <Griffous> I remember going through the process building a NAS, so if my 8 disks wouldn't work, now THAT would be a problem.
[11:25:08] * Peanut likes the way FreeBSD set up their hardware-compatibilty list - for each driver, they list which chipsets they support etc.
[11:25:18] <Griffous> In the end, they didn't work, that wasn't a solaris issue. The IBM bios can't handle that many disks, and would just reboot, lol. I had to pull the bios chip in the sata card out, so the real bios couldn't see them. Thankfully solaris could still directly address the disks though :)
[11:27:04] <Peanut> Good one, Griffous *grin* just rip the damn thing out alltogether - I've often felt that way about scsi/raid/network-card bios. They all want their 15 seconds of fame in the boot-screen, and that tends to add up.
[11:29:45] <Error_404> I had to completely disable my SATA disks in my motherboards' bios
[11:29:50] <Error_404> because EFI was screwing it up
[11:30:06] <Error_404> lucky for me, solaris initializes the SATA chip & reads the data off it just fine
[11:34:09] *** MattMan has quit IRC
[11:34:48] *** MattMan has joined #opensolaris
[11:38:54] *** mega has joined #opensolaris
[11:39:30] *** Odin-LAP has quit IRC
[11:42:51] *** rachel has quit IRC
[11:46:24] *** logic has quit IRC
[11:50:07] *** damienc has joined #opensolaris
[11:58:41] <Griffous> only an issue if you want to boot of it I guess
[11:58:56] <Griffous> and when zfs boot comes along, I guess I'll have to rethink that one
[12:01:11] *** uebayasi has quit IRC
[12:24:39] *** calumb has joined #opensolaris
[12:29:26] *** mega has quit IRC
[12:31:02] *** sartek has joined #opensolaris
[12:31:37] *** axxl has joined #opensolaris
[12:34:38] *** logic has joined #opensolaris
[12:38:31] *** logic has quit IRC
[12:39:08] *** logic has joined #opensolaris
[12:39:40] <logic> NICKLIST
[12:40:55] <PerterB> gesundheit
[12:41:23] *** logic has quit IRC
[12:42:34] *** sartek has quit IRC
[12:44:26] *** logic has joined #opensolaris
[12:46:32] *** bougie has joined #opensolaris
[12:46:45] *** MattMan has quit IRC
[12:47:00] *** MattMan has joined #opensolaris
[12:47:12] <bougie> hello :)
[12:49:36] <cmihai> MS guy asked about SSGD
[12:50:06] <cmihai> Told him it's Old' SCO/Tarantella. He went: HA, do you know who "OWNS" SCO? :P
[12:50:25] <cmihai> Pretty funny reaction really.
[12:50:33] <GoodKarma> haha
[12:51:30] <cmihai> Then he told me how Novell singned the MS deal not to pay 40 million USD / year or something for SCO patents.
[12:52:01] *** calumb has quit IRC
[12:52:06] <cmihai> Didn't know it wasn't the EVIL SCO I was talking about, but still funny :)
[12:52:18] <gallium> who is that clown?
[12:52:22] *** calumb has joined #opensolaris
[12:52:23] <GoodKarma> speaking of SCO....google 'SCO'... Did you mean 'TESCO'?
[12:52:43] <cmihai> Doesn't do that here.
[12:52:57] <quasi> so what's the deal with SSGD? free month trial or?
[12:52:58] <GoodKarma> maybe they've fixed it already
[12:53:13] <cmihai> quasi: pretty much.
[12:53:45] <quasi> cmihai: any idea what it costs after that?
[12:53:52] <cmihai> An arm and a leg
[12:54:08] <GoodKarma> haha
[12:54:26] *** bunker_work has quit IRC
[12:54:46] <lasseoe> jus have to find a donor then :)
[12:54:59] <quasi> even if it didn't cost an arm and a leg, knowing sun it is probably very hard to buy a single license
[12:55:02] <asyd> http://www.crazyapplerumors.com/?p=736
[12:55:05] <asyd> hello quasi
[12:55:12] <quasi> hey asyd
[12:55:35] <cmihai> lol wtf
[12:55:37] <cmihai> http://www.sun.com/software/products/sgd/get.jsp
[12:55:44] <cmihai> With Opera there is a price and buy button
[12:55:47] <cmihai> With firefox there isn't.
[12:56:09] <Berny> morning
[12:56:43] <cmihai> lasseoe: lol, 5000$?
[12:57:08] <cmihai> Sun Secure Global Desktop Starter Kit Promo - 50 Concurrent User License (Key) for Microsoft Windows, AS/400, Unix and Mainframe. Includes 1 year Basic Support (updates and upgrades, no phone support). For Education Only. $4,950.00
[12:57:30] <cmihai> 1 license is 200$
[12:57:54] <cmihai> From the Education Software Discount List..
[12:57:57] <cmihai> http://www.sun.com/products-n-solutions/edu/promotions/software/index.html
[12:57:59] *** kFuQ has quit IRC
[12:59:11] <quasi> cmihai: 200 for windows or unix exclusively - 345 for both on the normal price - but I seriously doubt they will sell it like that
[12:59:48] *** kFuQ has joined #opensolaris
[13:00:20] <cmihai> Can't afford that.
[13:00:43] <quasi> and $24/year for single users
[13:01:05] <cmihai> In addition to those 345?
[13:01:12] <quasi> yeah
[13:01:19] <quasi> you can't buy it alone
[13:01:24] <cmihai> It says here license is per socket
[13:01:30] <cmihai> So if you have 2CPU's...
[13:01:31] <quasi> add to that a windows license
[13:01:38] <cmihai> Hehe
[13:01:43] <cmihai> I get Windows free.
[13:01:50] <cmihai> All MS products anywho
[13:01:51] <quasi> wait for the quad cores to go out
[13:03:42] <cmihai> They are out.
[13:03:55] <cmihai> Intel is already shipping quad core Xeons
[13:06:18] <quasi> bah, I want a real proc
[13:06:29] <cmihai> Yeah, Xeons suck.
[13:06:37] <cmihai> Now quad core opterons... mmm
[13:06:59] <cmihai> But Cray will probably buy them all :P
[13:07:06] <cmihai> They're makeing a 24000 quad Opteron supercomputer
[13:07:38] <cmihai> And they're building it it's own power station :)
[13:07:41] <quasi> the nice thing about quad amd is that they will fit current dual core setups with a simple bios upgrade
[13:07:50] <cmihai> 400TB RAM..
[13:07:57] <quasi> for AM2 and socket F
[13:08:17] <cmihai> 11 Petabytes of disks...
[13:08:22] <Belgar> ahmm
[13:08:24] <cmihai> Cray sure knows how to make it big
[13:09:04] *** dunc_ has joined #opensolaris
[13:09:08] <jteo> doesn't one always get better prices via one's locl Sun rep?
[13:09:13] <jteo> s/locl/local
[13:09:30] <quasi> jteo: not always
[13:09:31] <Belgar> compared with list prices? always :)
[13:09:39] <cmihai> I hate how big companeis (Cisco, Sun) don't care about end users
[13:10:03] <Belgar> you forgot m$ :)
[13:10:08] <cmihai> You're alaways supposed to find some weird reseller and representative and such. It's just too much a hassle to get one for personal use.
[13:10:10] <jteo> quasi: you speak from experience i presume.
[13:10:17] <cmihai> Belgar: bah, MS will sell you anything.
[13:10:19] <jteo> cmihai: i concur.
[13:10:20] <Stric> quasi: you know that Core2duo are faster than Athlon64 per MHz, right?
[13:10:27] *** triplah_ has quit IRC
[13:10:39] <cmihai> Stric: that means nothing if the architecture sucks
[13:11:06] <cmihai> For heavy enterprise usage Opterons beats any intel hands up, pants down.
[13:11:08] <Belgar> stric :)
[13:12:27] <Stric> It all depends on what you're looking for and how you're measuring it
[13:12:41] <quasi> jteo: it tends to be a bit of a fight and if you're not buying much, some resellers don't even want to sell
[13:12:42] <jteo> i for one would appreciate at least a one time developer discount off a Sun workstation. but maybe I'm cheap. -sigh-
[13:13:05] <cmihai> quasi: had those issues even with companies
[13:13:14] <cmihai> We're forced to hold an auction for ANYTHING over 5000$
[13:13:19] <cmihai> So you're supposed to invite Sun.
[13:13:22] <cmihai> To buy 1 server.
[13:13:23] <cmihai> HAH
[13:13:33] <cmihai> So naturally the worst scum shows up.
[13:13:38] <cmihai> You know, stuff like HP :P
[13:14:21] <quasi> cmihai: not making it easy for yourself
[13:14:57] <cmihai> It's the way stuff works.
[13:15:07] <cmihai> You're supposed to write specs and invite companies to bid.
[13:15:12] <cmihai> Lowest price gets the deal.
[13:15:23] <jteo> hmm. kupfer is rounder than I thought.
[13:15:28] <Berny> does /me knows that
[13:15:31] <cmihai> Usually 50.000-100.000$ deals so Sun won't even flinch for that.
[13:15:46] <Berny> cmihai: just choose your required specs the way that only sun can win 8-)
[13:15:49] <cmihai> Or even Cisco, so you're stuck with Allied Telesyn
[13:15:53] <quasi> cmihai: make the specs so that others have a harder time winning
[13:16:24] <cmihai> quasi: when you're saying "48 disks 16TB RAM 2 quad core CPU's Integrated Lights Out" you are doing that ;P
[13:16:41] <jteo> it's no wonder Dell has a whole business model built around actually selling you something.
[13:16:50] <cmihai> Heh
[13:17:12] <quasi> jteo: not that it has worked all that well recently
[13:17:45] <Stric> jteo: I tried buying a rack mounted server from dell. No can do. I had to be/have a company.
[13:18:45] <jteo> maybe it's something to do with size. the larger you are, the more silly a company gets.
[13:19:18] <cmihai> Stric: what does that have to do with anything?
[13:19:45] <cmihai> Oh, right, planning to rip you off on support and service plans
[13:19:58] <cmihai> And upgrades and spare parts and services and all that
[13:20:01] <Stric> cmihai: why is a rack mounted machine special? I can buy regular machines without problems, but rack mounted, nooo..
[13:20:55] *** dunc_ has quit IRC
[13:21:14] <Error_404> Stric: probably because they sell rackable kit to businesses
[13:21:41] <cmihai> They sell plenty more desktops to businesses
[13:21:58] <Peanut> Reminds me of having to get a price for a Juniper router. But noo, Juniper wouldn't talk to me and sent me off to their distributor. Who then passed me on to their partner. Who came up with a very nice offer for a Foundry switch.
[13:22:09] <cmihai> Hah
[13:22:14] <cmihai> Awesome :)
[13:22:27] <cmihai> Peanut: you have no idea what we have to go through to get prices on Cisco hardware
[13:22:32] <cmihai> They don't actually give you the price.
[13:22:37] <cmihai> You have to "negotiate"
[13:22:53] <cmihai> With a ton of resellers and authorised parteners and all that crap.
[13:23:00] *** logic has quit IRC
[13:23:45] <cmihai> Of course you'll never be able to find any of that stuff in a store, goodness forbid
[13:23:50] <quasi> having only one sun reseller in the country is no fun either - especially when they're as bad as this one
[13:24:01] <cmihai> quasi: having none is no fun either.
[13:24:18] <quasi> cmihai: I'd almost prefer that
[13:24:49] <Error_404> I just hope sun extends startup essentials to canada sometime in the next year
[13:25:14] <cmihai> Yeah, this kind of stuff is no fun if you're not in the US
[13:25:15] <Peanut> And this wasn't just some SOHO switch, we need to switch/route 32 1 Gbit/s and a 10Gbit/s connection, and it must switch/forward 16Gb/s sustained.
[13:25:40] <cmihai> Heh
[13:25:47] <cmihai> So they wouldn't sell you something that's about 10k?
[13:26:03] <cmihai> Oh... 10Gbit?
[13:26:16] <Peanut> You'd be surprised how many 'non-blocking' switches, well, aren't.
[13:26:18] <cmihai> That is damn expensive. And they wouldn't sell you that?
[13:26:39] *** kjetilho has left #opensolaris
[13:26:48] *** Snake007uk has joined #opensolaris
[13:32:07] <Peanut> Actually doing a gigabit over a 1Gb/s connection is something that seems to scare and/or surprise vendors of networking equipment. Finding a switch/router that isn't overcommitted is a surprising amount of work - but fun ;-)
[13:33:27] <sickness> i'm back
[13:33:28] <quasi> Peanut: a bit like looking at ibms blade chassis - something like a 3 or 4 gbit bus for 7 blades with gigabit
[13:33:50] <cmihai> All do that
[13:34:25] <quasi> not as bad as that
[13:37:55] <Stric> I wonder when a fix for this month old security hole will be released for sol10.. http://sunsolve.sun.com/search/document.do?assetkey=1-26-102724-1
[13:38:37] *** uncertainty_ is now known as uncertainty
[13:38:49] <Peanut> Q: I want to do some TCP tests, and I need to make a box that simulates a 300ms RTT on a 1Gb/s link. I've found dummynet (FreeBSD, seems a bit dated) and NistNet (Linux, even more dated, doesn't do bridging). Would there be any way to build a delaying bridge in OpenSolaris?
[13:39:40] <andersmo> Peanut: linux can do it with tc.
[13:39:49] <Stric> It was easy to delay packets with IRIX 5.3.. just give it some cpu load and delay scaled with it ;)
[13:39:57] <andersmo> (which is built-in in the mainline kernel nowadays)
[13:40:35] <quasi> darn, http://dlc.sun.com/osol/on/downloads/b55 seems to be down (from http://www.opensolaris.org/jive/thread.jspa?messageID=82016&tstart=0#82016)
[13:42:45] <Peanut> andersmo: I've been going trough the tc docs today, but tc seems to only be able to specify a maximum latency, not a fixed latency?
[13:44:25] <andersmo> Peanut: http://linux-net.osdl.org/index.php/Netem
[13:45:52] <Peanut> Ah, I bow to your superiour Google-fu :-) Thanks
[13:48:04] <Peanut> Guess it'll be Linux then
[13:48:35] *** kFuQ has quit IRC
[13:49:05] *** kFuQ has joined #opensolaris
[13:50:06] *** hile_ has joined #opensolaris
[13:50:42] <andersmo> Heh, Google-fu... I'm actually wearing a Google "I'm feeling lucky." t-shirt now, which I got when I interviewed for the job I'm starting in the coming summer. Maybe it's the t-shirt. ;)
[13:51:04] <Peanut> Ah, congratulations :-)
[13:51:15] <jteo> it's good to have a job you love. (not me)
[13:51:26] *** |tsoome| has quit IRC
[13:51:55] *** tsoome_ is now known as tsoome
[13:52:08] <tsoome>  /msg NickServ IDENTIFY koleuss
[13:52:22] <tsoome> damit:P
[13:52:41] * gallium steals tsoome 's account
[13:52:44] <gallium> </joke>
[13:53:01] <tsoome> ;)
[13:57:27] <raph_ael> you'd better quick change you passwd :)
[13:58:03] <tsoome> my account is so ugly, nobody wants it;)
[14:02:05] <cap_> I hope you have been a good boy and used a unique password for freenode...
[14:02:18] <cmihai>  tsoome very nice.
[14:02:21] * cap_ tries tsoome@$(whois ... ):pw
[14:02:44] <cmihai> tsoome: you're not using a clock either
[14:03:01] <cmihai> So now everyone has your IP and your ssh server :)
[14:03:15] <tsoome> clock?
[14:03:34] <cmihai> cloak
[14:03:39] <cmihai> Typo
[14:03:54] <cmihai> Eg: 15:03 -!- tsoome [n=tsoome at 194 dot 106.105.145]
[14:04:02] <cmihai> 15:04 -!- cmihai [n=cmihai@unaffiliated/cmihai]
[14:04:04] <tsoome> :)
[14:04:23] <cmihai> It's free you know :P
[14:04:44] <tsoome> incoming ssh is closed anyhow
[14:18:08] *** dunc_ has joined #opensolaris
[14:27:23] *** uebayasi has joined #opensolaris
[14:31:33] <cmihai> tsoome: I learned to always type / commands in the server window, not on channels :)
[14:31:34] *** jcsmith has quit IRC
[14:31:42] <cmihai> Just in case :)
[14:31:53] *** jcsmith has joined #opensolaris
[14:32:14] *** MattMan has quit IRC
[14:36:35] <Belgar> don't push your luck :-)
[14:36:48] *** kFuQ has quit IRC
[14:37:05] <gallium> you must be using a window manager without focus follows mind
[14:37:16] <cmihai> Hehe
[14:37:52] <cmihai> Nothing more fun than GAIM stealing focus ;]
[14:38:00] <cmihai> Especially when typing root credentials ;P
[14:38:16] *** |tsoome| has joined #opensolaris
[14:38:37] <cmihai> Good thing is with my password, most people think I've said something about their mothers in Klingon.
[14:40:20] <cap_> better yet, use plain english words so no one ever suspects it to be a password ;-)
[14:40:21] <gallium>  /msg nickserv identify idiot
[14:40:28] <gallium> d'oh!
[14:40:33] <gallium> :-)
[14:40:58] <tsoome> :)
[14:43:18] <Doc> i think nickserv is pointing at you right now :)
[14:43:46] * tsoome knows....
[14:43:58] *** MattMan has joined #opensolaris
[14:44:02] * tsoome is just another blondy...
[14:45:07] <jteo> true.
[14:45:50] *** tsoome has quit IRC
[14:46:01] *** |tsoome| is now known as tsoome
[14:51:02] *** gallium has quit IRC
[14:52:54] <elektronkind> oh dear
[14:53:09] <elektronkind> http://video.google.com/videoplay?docid=-4024457809967804989
[14:53:14] <elektronkind> Immolate Me Elmo
[14:55:08] <lasseoe> hm :-/
[14:56:01] <cap_> does anyone know if belenix is available for x86_64? ..all I can find is that it was a priority for 0.3 (belenix is now at 0.5.1)
[14:58:40] <cap_> I'm very impressed by the amount of stuff that just works when booting the 32-bit livecd but I'm a bit surprised that there was no x86_64 iso on the download page.
[14:59:35] <Kmays> need a livecd of Nexenta Alpha6
[14:59:41] <cmihai> There isn't one yet.
[14:59:51] <cmihai> Or I don't think there is one, check the status on the page.
[15:02:55] *** calumb is now known as calLNCH
[15:08:03] *** jlc has joined #opensolaris
[15:11:21] *** bunker_work has joined #opensolaris
[15:12:43] *** kFuQ has joined #opensolaris
[15:12:46] *** tsoome_ has joined #opensolaris
[15:14:12] *** tsoome has quit IRC
[15:15:00] *** tsoome_ is now known as tsoome
[15:16:00] *** Fish has quit IRC
[15:17:01] <cmihai> GOD, I hate how Ctrl Backspace kills X
[15:17:06] <cmihai> No need for ALT.
[15:17:50] <cmihai> It's a bug :(
[15:18:42] <jlc> anyone use dbx?
[15:18:59] <jlc> http://developers.sun.com/solaris/articles/traceback.html
[15:19:04] <jlc> i'm reading that
[15:19:15] <cmihai> Hah, it only works with NumLock on.
[15:19:18] *** kFuQ has quit IRC
[15:19:19] <cmihai> This is dumb.
[15:19:26] <jlc> when i do dbx core I get: dbx: File `core' is not executable
[15:19:35] <jlc> do i have to make it +x ?
[15:19:42] *** kFuQ has joined #opensolaris
[15:20:13] <jlc> I do that, then I get: dbx: File `core' is not ELF executable
[15:20:16] <jlc> argg
[15:20:40] *** kFuQ has joined #opensolaris
[15:20:55] <hile_> i thought it was dbx executable core
[15:21:06] <hile_> where executable is the path to the executable that created the core
[15:21:35] *** kFuQ has quit IRC
[15:21:38] <jlc> ah, so you have to give full path
[15:22:02] <hile_> I've always given the full path
[15:22:14] <jlc> cod228a::>dbx /export/home/cod228a/MyNextel/atlspold/mustgather/core
[15:22:14] <jlc> dbx: File `/export/home/cod228a/MyNextel/atlspold/mustgather/core' is not executable
[15:22:27] *** hile__ has joined #opensolaris
[15:22:30] <jlc> (dbx) where >> traceback.txt
[15:22:30] <jlc> dbx: program is not active
[15:22:33] <jlc> then i get that
[15:22:42] *** hile__ has quit IRC
[15:22:42] <hile_> Did you read what I stated?
[15:22:51] <hile_> dbx /path/to/program core
[15:22:58] <jlc> oh
[15:22:58] <tsoome> jlc dbx is correct
[15:23:07] *** hile__ has joined #opensolaris
[15:23:08] <tsoome> or use pstack core
[15:23:13] <jlc> path to program, missed that part
[15:23:24] *** hile__ has quit IRC
[15:23:31] <jlc> i did a pstack, does it give the same info?
[15:23:36] <tsoome> yes
[15:23:43] <jlc> oh, forget dbx then ;)
[15:24:09] *** aramdune has joined #opensolaris
[15:24:12] <tsoome> of course with dbx you can do more
[15:24:20] <jlc> guess if I read the url better you wouldn't have had to answer hile_
[15:24:25] <tsoome> browse a stack etc
[15:24:32] <jlc> it says dbx /path/executable core
[15:24:37] <jlc> lol
[15:24:42] * jlc goes for more coffee
[15:25:02] *** dunc_ has quit IRC
[15:25:07] <jlc> when you do dbx /blah core does that do anything to the running server, any harm in doing that?
[15:25:26] <jlc> production server so I wouldn't want to run something that will make it fall down go boom
[15:25:28] *** kFuQ has joined #opensolaris
[15:25:33] *** Belgar has left #opensolaris
[15:28:20] <jlc> is there any good info out there on reading a pstack? other than seeing appserver and ../bin/sparc/native_threads/java -s
[15:28:25] <jlc> gets all crazy looking ;)
[15:28:37] <jlc> or a tool that can do it
[15:29:14] *** kFuQ has quit IRC
[15:30:48] <jlc> http://rafb.net/p/MUfPhc11.html
[15:31:15] <jlc> there are another 46 pages of stuff that looks like that, how do you decrypt it?
[15:33:19] *** Odin- has quit IRC
[15:35:47] *** laca has joined #opensolaris
[15:35:53] <jlc> gm laca
[15:36:02] <laca> hey jlc
[15:36:19] <jlc> I got a question about core's since you've told me stuff before, your my guru ;)
[15:36:50] <jlc> http://rafb.net/p/MUfPhc11.html
[15:36:53] <jlc> at the top of that
[15:36:56] <jlc> /bin/sparc/native_threads/java
[15:37:03] <jlc> is that considered the executable
[15:37:41] <jlc> would a dbx give me different info that a pstack, cause all the pstack gave me was 46 pages of HEX or something else
[15:39:54] <laca> the problem is that the binary is stripped, i.e. doesn't have symbol name information in it
[15:40:14] <laca> so neither pstack nor dbx can match the hex values to function names
[15:41:05] <jlc> is there anything I can do to figure out how this core was generated?
[15:41:26] *** Kmays has quit IRC
[15:42:44] *** UnixTitan has joined #opensolaris
[15:44:37] <laca> jlc: well, not much
[15:45:08] <jlc> thats bunk :)
[15:45:13] <jlc> or sucks
[15:47:30] <jlc> Does workshop or Data Display Debugger help any, have you ever used those?
[15:49:34] <jteo> DDD is a frontend to gdb.
[15:50:36] <jlc> workshop is sun studio isn't it
[15:50:59] <jlc> jteo: does it just show a gui of what I'm looking at in a txt file, or show anything else?
[15:51:40] <jteo> it's the same problem really: you can't get useful stack traces without a binary with symbols (your binary is stripped)
[15:55:25] <jlc> so if I do dbx /path/to/executable core, would that give any better info or is it still the problem with the binary not having symbols
[15:55:40] <jteo> jlc: same problem.
[15:56:11] <jlc> so some how I have to put debug symbols into websphere and hope it cores again :)
[15:56:14] <jlc> is that right
[15:59:24] *** Odin- has joined #opensolaris
[16:00:40] *** alobbs has quit IRC
[16:02:21] <jteo> jlc: that's one way of doing it.
[16:02:32] *** mrdeviant has joined #opensolaris
[16:04:06] *** alobbs has joined #opensolaris
[16:04:55] <cap_> cmihai, re. 64-bit nexenta livecd, I can't even find a non-livecd nexenta. also 64-bit wasn't mentioned in alpah6 hilights. any pointers?
[16:05:31] *** Odin- has quit IRC
[16:09:07] *** Snake007uk has quit IRC
[16:09:24] *** Snake007uk has joined #opensolaris
[16:10:29] *** RaD|Tz has joined #opensolaris
[16:14:59] *** Statia has joined #opensolaris
[16:19:38] *** calLNCH is now known as calumb
[16:20:32] *** MikeTLive has quit IRC
[16:22:43] *** uebayasi has quit IRC
[16:24:45] *** darrenm has joined #opensolaris
[16:25:47] <jteo> wb darrenm
[16:25:50] *** Mazon is now known as mazon
[16:35:56] *** Statia has quit IRC
[16:37:13] *** mega has joined #opensolaris
[16:37:30] *** hile_ has quit IRC
[16:43:14] *** alobbs has quit IRC
[16:45:25] *** hile_ has joined #opensolaris
[16:46:51] *** broadcast has quit IRC
[16:46:52] *** darrenm has quit IRC
[16:46:52] *** peteh has quit IRC
[16:47:18] *** darrenm has joined #opensolaris
[16:50:16] *** gabriele has joined #opensolaris
[16:50:56] <gabriele> hello , i need help i have:  CPU states: 48.1% idle,  0.2% user, 51.7% kernel,  0.0% iowait,  0.0% swap .. and loggin the kernel i have : http://pastebin.com/855983 i have acipica driver problem
[16:51:01] <gabriele> how can i stop this driver?
[16:51:28] <darrenm> ignore the stats, what is the real problem ?
[16:51:45] <gabriele> 51 % kernel
[16:51:47] <gabriele> stuck
[16:51:54] <gabriele> solaris real slow :(
[16:52:01] <darrenm> okay now we are getting some where
[16:52:04] <jamesd> darrenm, see the pastebin   its from    tick-1234hz { @[stack()] = count();}
[16:52:07] <jamesd> in dtrace.
[16:52:20] <darrenm> being 51% in the kernel isn't the problem it is just a statistic
[16:52:40] <darrenm> jamesd: pastebin ?
[16:52:47] <gabriele> but the machine is impossible to use.. yes pastebin http://pastebin.com/855983
[16:52:50] <jamesd>  http://pastebin.com/855983
[16:54:23] <jamesd> i know there is a work around to that problem and i don't remember it, and i'm about to leave... just waiting for the wife.
[16:55:17] <gabriele> jamesd:  i need only some wild fix.. some hard reboot that fix.. something that kill acipica
[16:55:36] <gabriele> its very urgent.. on why i have stucked i have time to understand.. but now i must have this machine online
[16:55:43] <darrenm> acpica is pretty low level ;-) killing it might not be what you want ;-)
[16:55:44] <gabriele> there is no way?
[16:55:46] <jamesd> gabriele, i think it involves changing a setting in the bios to disable apic or something.
[16:55:50] <gabriele> azz
[16:56:07] <gabriele> but why only now this problem.. i have change nothing
[16:56:11] <jamesd> okay i'm gone now...
[16:56:16] <gabriele> byeeee
[16:56:51] *** LordKing has quit IRC
[16:57:00] <darrenm> what build you running and what kind of hardware ?
[16:57:05] <gabriele> 5.10
[16:57:07] <gabriele> solaris machine
[16:57:10] <gabriele> sun machine
[16:57:12] <zdzichuBG> ecision to bring on the market the Total HD.
[16:57:13] <zdzichuBG> Essentially, what the Total HD does is it plays on one side the HD DVD and on the other the Blu Ray format
[16:57:14] <gabriele> dual cpu ..
[16:57:18] <zdzichuBG> sorry
[16:57:46] <darrenm> which release of Solaris 10 ?
[16:58:06] <gabriele> yes
[16:58:19] <sickness> gabriele: did you try latest sxcr snv54 too?
[16:58:34] <gabriele> no :(
[16:59:27] <gabriele> i have rbeooted
[16:59:32] <gabriele> and always 51
[16:59:49] <gabriele> % on kernel
[17:01:14] *** laca is now known as lacaMtg
[17:01:39] *** mega has quit IRC
[17:03:03] <gabriele> no idea?
[17:04:50] <gabriele> load averages:  0.83,  0.50,  0.21                                     17:03:27
[17:04:51] <gabriele> 46 processes:  45 sleeping, 1 on cpu CPU states: 48.3% idle,  0.0% user, 51.7% kernel,  0.0% iowait,  0.0% swap Memory: 2047M real, 1765M free, 123M swap in use, 2037M swap free
[17:06:12] <jteo> machine type, and what did you change last?
[17:06:23] <gabriele> a HD
[17:06:27] <gabriele> scsi =)
[17:06:30] <gabriele> but is not mounted now
[17:07:16] <gabriele> is a sun machine..
[17:08:03] <darrenm> which Sun machine there are quite a few x86 machines
[17:08:19] <gabriele> there is no way to see it?
[17:08:22] <gabriele> a banner? a command?
[17:08:32] <Triskelios> prtdiag
[17:08:45] <Triskelios> gabriele: ^
[17:08:59] <darrenm> btw if this is Sun hardware and Solaris 10 rather than SXCR your best bet is calling Sun if you have a support contract and/or the harware is still under warranty
[17:09:38] <gabriele> prtdiag
[17:09:40] <gabriele> http://pastebin.com/855993
[17:09:54] <gabriele> i have no support :(
[17:10:39] <Triskelios> that's a fairly new machine...
[17:10:48] <gabriele> yes but dont work
[17:10:54] <gabriele> kernel is stucked 51%
[17:10:59] <gabriele> and is to much load
[17:12:42] <gabriele> no idea on how can i fix it?
[17:12:43] <jteo> no recent patches other than your (hardware) scsi hard disk addition?
[17:12:53] <gabriele> jteo:  correct no pachtes
[17:13:11] <darrenm> which release of Solaris 10 are you running ?
[17:13:16] <gabriele> there is a command?
[17:13:22] <jteo> gabriele: uname -a
[17:13:36] <gabriele> SunOS pegasus 5.10 Generic_118855-15 i86pc i386 i86pc
[17:13:55] <darrenm> contents of /etc/release please
[17:14:46] <tsoome> gabriele, I would start with patches;)
[17:15:01] <tsoome> current kernel is 118855-19 for instance
[17:15:02] <gabriele> but it have always worked fine
[17:15:32] <tsoome> also this will be first suggestion from support;)
[17:15:51] <gabriele> understand.. there is a automatic patcher?
[17:16:01] <tsoome> smpatch
[17:16:10] <tsoome> or get pca
[17:16:21] <tsoome> pca will do better job...
[17:16:39] <darrenm> update connection if you are already running a recent enough vintage of Solaris 10
[17:16:42] <gabriele> content release: http://paste.lisp.org/display/34772
[17:16:59] <darrenm> but note that if you dont' have a contract you won't get all patches until the next update release is made
[17:17:04] <quasi> pca risks bringing in a couple of patches you really don't want if you're on a release prior to 11/06
[17:17:22] <tsoome> ?
[17:17:26] <quasi> darrenm: but pca gives you all the patches
[17:17:41] <jteo> gabriele: paste the output of "cat /etc/release"
[17:17:50] <gabriele> jteo:  http://paste.lisp.org/display/34772
[17:17:51] <darrenm> how can it give you all the patches since they aren't all available externally from Sun unless you have a contract
[17:18:13] <quasi> tsoome: http://soulfood.dk/archives/2006/11/19/T15_31_02/index.html has a couple of issues I've seen
[17:18:14] <gabriele> i have no contract for this machine
[17:18:15] *** mega has joined #opensolaris
[17:18:27] *** osdummy has joined #opensolaris
[17:18:33] <quasi> darrenm: not the contract patches, but a whole lot more than what smpatch gives you
[17:18:46] <darrenm> so not all patches then!
[17:18:51] <quasi> (like 100+ more on 6/06)
[17:18:58] <gabriele> i only need that the kernel work fine.. and not raise at 51 cpu
[17:19:01] <quasi> should have stuck an almost all
[17:19:06] <darrenm> the /etc/release is a FILE not a directory
[17:19:09] <darrenm> contents of the file
[17:19:18] <gabriele> ah oki sorry darrenm
[17:19:33] <gabriele> now i am rebooting the machine for the 20 times.. when she come up.. i give you
[17:19:53] *** axisys has joined #opensolaris
[17:20:12] <jteo> gabriele: the scsi HD is plugged in?
[17:20:19] <gabriele> jteo:  now yes
[17:21:10] <gabriele> ash-3.00# cat /etc/release
[17:21:12] <gabriele>                         Solaris 10 6/06 s10x_u2wos_09a X86            Copyright 2006 Sun Microsystems, Inc.  All Rights Reserved.                         Use is subject to license terms.                             Assembled 09 June 2006
[17:22:53] *** MattMan has quit IRC
[17:23:28] <tsoome> quasi, as for 120051-04 - are you sure the libc patch is recent? ;)
[17:23:33] <axisys> i have zpool created on c1d0s7, i put a second disk c2d0 on ultra 20, install sol u3 on the second disk and booted from the second disk.. now how do i access the pool sitting on first disk?
[17:24:17] <gabriele> :(
[17:24:50] *** bunker_work has quit IRC
[17:24:59] <darrenm> okay so you have update 2.  Update 3 is out and that includes an updated acpica binary in it - not sure if any specific bugs were fixed in it though.
[17:25:26] <jteo> axisys: zpool import -f yourpool ?
[17:25:41] *** lacaMtg is now known as laca
[17:25:46] <gabriele> darrenm: do you have a command that see if my HD scsi is installed correctly?
[17:26:05] <darrenm> iostat -E
[17:26:17] <darrenm> you should also see it in format
[17:26:42] <gabriele> darrenm: http://paste.lisp.org/display/34772#1
[17:26:55] <quasi> tsoome: that one happened with smpatch as well on a machine with a contract
[17:26:58] <gabriele> its ok?
[17:27:17] <quasi> tsoome: so yes, I'm sure
[17:27:29] <darrenm> do you see the disk you expect to see in the list ?
[17:27:30] <tsoome> well, version `SUNW_1.22.2' not found means you have a too old libc
[17:27:39] <onbot> commit by Sangeeta Misra:  6508701 ire_add_v4() often adds unresolved IREs even when told not to; 6469271 misspelled collision related kstat names in GLDv3
[17:27:39] <gabriele> darrenm: yes
[17:27:45] <gabriele> =)
[17:27:58] <quasi> tsoome: I remember lasseoe having the exact same on sparc
[17:28:13] <darrenm> if you want to test it you can do so destrucitvely to the data on the disk using the analyze purge command from format
[17:28:41] <gabriele> i will only that % kernel load go out
[17:28:48] <axisys> jteo: cool that did it
[17:28:53] <tsoome> 118345-19 seems to be current for libc
[17:29:45] <axisys> jteo: hypothetically speaking.. if i were to forget the pool name.. how would i go by importing it?
[17:30:02] <jteo> axisys: remember to export the zpool first next time. otherwise, i don't think zpool import will show it
[17:30:17] <jteo> (not entirely sure)
[17:30:25] <oxygene> axisys: zpool import -f -a?
[17:30:48] * richlowe despairs.
[17:31:01] <jteo> there's a -a for zpool import?
[17:31:02] <axisys> jteo: i were able to import w/o doing export.. it worked.. thnx
[17:31:03] <tsoome> ok it's obsoleted by 118855 :)
[17:31:08] <axisys> oxygene: thnx
[17:31:18] <darrenm> just run zpool import with no other arguments it will tell you all the available pools that can be imported
[17:31:26] <darrenm> if you have deleted the pool then run zpool import -D
[17:31:40] <jteo> darrenm: it lists zpools that haven't been exported as well?
[17:31:50] <oxygene> jteo: according to zpool --help..
[17:31:51] <darrenm> zpool import does not
[17:32:03] <axisys> darrenm: excellent.. i did not delete the pool btw .. but good to know
[17:32:22] <jteo> oxygene: ah. i just scrolled down the man page.
[17:32:23] <jteo> :)
[17:32:38] <axisys> i got my pool back.. yeah
[17:33:10] <tsoome> yeah and SUNW_1.22.1 is latest version of libc:) so someone in sun must be kicked:)
[17:33:35] <darrenm> why should someone be kicked ?
[17:34:23] <richlowe> I presume the useradd patch bug.
[17:34:24] <tsoome> because provided  usermod expects SUNW_1.22.2 and it's not released...
[17:34:33] <richlowe> (ie, the patch that prevents user* working because of libc versioning screwups caused by TX)
[17:35:34] <tsoome> usermod is compiled against too recent libc;)
[17:36:06] <Triskelios> I'm getting a sense of deja vu about the lib versioning problems with the user* tools..
[17:40:41] *** mega has quit IRC
[17:40:52] *** calum_ has joined #opensolaris
[17:41:34] *** Netwolf has joined #opensolaris
[17:41:59] *** stevel has joined #opensolaris
[17:41:59] *** ChanServ sets mode: +o stevel
[17:42:40] <richlowe> Triskelios: the number of recent-ish 10 patches that broke things annoyed a bunch of people.
[17:45:02] *** pseudoXh4 has joined #opensolaris
[17:45:49] *** alanc-away is now known as alanc
[17:46:13] *** damienc has quit IRC
[17:46:18] <sickness> gabriele: can I pm you?
[17:46:21] *** vipe has joined #opensolaris
[17:46:27] <gabriele> sickness: yes
[17:46:59] <gabriele> sickness: devi registrarti per querare:  /ns register pass
[17:47:08] <sickness> yeah
[17:47:11] <vipe> would it be possible to set multiple mountpoints on zfs volumes (global zone, no clustered) ?
[17:47:12] <sickness> I'm registered
[17:47:17] <darrenm> patching is the wrong release mechanism for features
[17:47:20] <sickness> are you registered too?
[17:47:33] <darrenm> vipe: no that is what lofs is for
[17:47:43] <stevel> richlowe: hrm, the 'wx edit' i started last night of every file under SCCS control is still going :-P
[17:47:44] <darrenm> and zonecfg already has that capability
[17:47:45] <richlowe> darrenm: Yes, for exactly these reasons.
[17:47:53] <richlowe> stevel: I didn't expect it to be fast. :)
[17:48:08] *** comay has joined #opensolaris
[17:48:09] <richlowe> stevel: it maybe faster to use find and awk to spit out a active list directly.
[17:48:12] <richlowe> stevel: (but I doubt it, honestly)
[17:48:13] *** gabriele is now known as gciao
[17:48:16] <darrenm> stevel: why are you doing a wx edit of every file ?
[17:48:22] <richlowe> darrenm: my benifit.
[17:48:32] * stevel points at richlowe
[17:48:33] <richlowe> darrenm: since I can't, and I'd like to see copyright checks run over all of ON
[17:48:36] <stevel> he made me do it ;-)
[17:48:46] <richlowe> (so I can compare with my output)
[17:48:49] <darrenm> ah
[17:49:08] <jteo> politicians. ;)
[17:49:10] <darrenm> since I guess you (richlowe) are working in a mercurial workspace ?
[17:49:44] <darrenm> have we sorted out keywords yet for the swtich over ?
[17:50:02] <darrenm> ie what replaces "%Z%%M% %I%     %E% SMI"
[17:50:05] <darrenm> as the ident value
[17:51:03] <richlowe> darrenm: not yet.
[17:51:21] <richlowe> darrenm: Yes, and 'wx copyright' is implemented directly in wx.
[17:51:33] <richlowe> darrenm: and I feel hacking it out of wx to generate a baseline for testing invalidates said testing.
[17:54:51] *** kFuQ has joined #opensolaris
[17:55:06] *** hile_ has quit IRC
[17:57:08] *** hile_ has joined #opensolaris
[17:57:17] <jteo> wb hile_.
[17:57:28] <hile_> morning
[17:57:56] *** sommerfeld has joined #opensolaris
[17:57:56] *** ChanServ sets mode: +o sommerfeld
[18:00:08] *** slowhog has joined #opensolaris
[18:01:16] *** whaq has joined #opensolaris
[18:01:19] <Theoden-Nexenta> http://imgs.xkcd.com/comics/regular_expressions.png
[18:02:30] <Teknix> I'm trying to get the qlc driver from SXCR b54 to load on a b53 opensolaris kernel.  Is there any reason that shouldn't work?
[18:03:38] <jteo> sometimes the code changes?
[18:04:14] <axisys> i am looking for a `how to' on securing a sol 10 server? i know it depends on what the use of a server.. by looking for a general guideline.. for example i never use telnet/ftp/smtp..
[18:04:15] <Teknix> maybe I should try 11/06
[18:04:28] *** gciao has quit IRC
[18:06:22] <darrenm> Teknix loading things built on newer releases on older releases is not supported and often does not work
[18:06:25] *** calum_ has quit IRC
[18:06:38] <darrenm> if you want stuff from SXCR 54 go to SXCR 54
[18:06:38] <Teknix> darrenm: that's kind of what I thought
[18:07:08] *** deather_ has joined #opensolaris
[18:07:30] <darrenm> axisys: install Solaris 10 11/06 (aka update 3) it has all remote services except ssh disabled by default if you ask for this behaviour during install
[18:07:59] <zdzichuBG> axisys: and you can install Trusted Solaris on 11/06
[18:08:05] <axisys> darrenm: oops i didn't :P
[18:08:06] <Teknix> darrenm: well, I'm trying to load it on nexenta which is now up to b55
[18:08:24] <darrenm> ah
[18:08:25] <Teknix> although this particular machine is still on b53..
[18:09:02] <Teknix> i'll try it on b55, or using a slightly older one
[18:09:05] <axisys> zdzichuBG: hmm.. i heard that might be tooooo retrictive.. might be wrong on that
[18:09:28] <axisys> also is JASS any good?
[18:09:44] <_syphilis_> apparenty, not all services are disabled by 'secure by default': online          5:16:33 svc:/system/webconsole:console
[18:09:49] <darrenm> axisys: yes
[18:10:05] <darrenm> _syphilis_: it should only be listening on localhost though
[18:10:13] <_syphilis_> ah, that's not so bad then
[18:10:21] <darrenm> there are some services that are not disabled but instead changed to listen only on a local transport
[18:10:25] <darrenm> rpcbind is like this
[18:10:42] <axisys> i did that for sendmail only listening to localhost
[18:10:47] <darrenm> if you nmap scan an SXCR or S10u3 box you will see ssh and rpcbind but rpcbind won't actually respond to you
[18:12:20] <axisys> darrenm: is there a way to achieve the `secure by default' now since i failed to pick it during install of u3?
[18:12:37] <darrenm> yes
[18:12:43] <darrenm> # netservices limited
[18:12:52] <axisys> darrenm: where ?
[18:12:59] <darrenm> # is your root prompt
[18:13:04] <darrenm> run the netservices command
[18:13:07] <axisys> darrenm: heh
[18:13:07] <darrenm> with the argument limited
[18:13:21] <hile_> morning darrenm
[18:13:44] <_syphilis_> errr, netservices was introduced in 6/06 (or 1/06?) and its manpage already claims its interface is "Obsolete"
[18:13:59] <comay> one issue with the webconsole service is it always starts up a jvm
[18:14:30] <darrenm> _syphilis_: so what
[18:14:44] <_syphilis_> don't things usually last a bit longer before they become obsolete?
[18:14:51] <darrenm> Obsolete interfaces don't get removed in patch releases and all S10 updates are patch releases
[18:14:58] <darrenm> born Obsolete is allowed
[18:15:03] <darrenm> this command was born Obsolete
[18:15:06] <axisys> darrenm: just ran it.. it restarted couple services and restarted dtlogin as well..
[18:15:12] <darrenm> yep
[18:15:59] *** alobbs has joined #opensolaris
[18:17:29] <darrenm> it restarted dtlogin to put the X server into local listen mode
[18:17:34] <axisys> darrenm: so i should be in save level secure as if i installed u3 with limited network services?
[18:17:52] <darrenm> as far as the Solaris provided services yes
[18:18:00] *** deather has quit IRC
[18:18:03] *** deather_ is now known as deather
[18:18:05] <axisys> darrenm: correct
[18:18:40] <axisys> darrenm: to make more secure next step would be jass?
[18:19:07] <Peanut> axisys: or unplugging ;)
[18:19:21] <axisys> Peanut: no kidn'
[18:19:33] *** alobbs has joined #opensolaris
[18:39:27] *** peteh has joined #opensolaris
[18:41:19] *** hile_ has quit IRC
[18:43:02] *** Snake007uk has quit IRC
[18:55:15] *** Theoden-Nexenta has left #opensolaris
[18:55:19] *** hile_ has joined #opensolaris
[18:56:39] *** Cyrille has quit IRC
[18:57:09] <onbot> commit by svemuri:  6494657 FP emulator has to handle the fact that quad fp ops may generate illegal_instruction traps
[19:01:43] *** tsoome has quit IRC
[19:10:14] *** dunc has quit IRC
[19:13:07] *** abackos has joined #opensolaris
[19:14:39] *** aramdune has quit IRC
[19:15:21] *** pikapika has joined #opensolaris
[19:15:54] *** pitty has left #opensolaris
[19:16:05] <pikapika> hello
[19:17:52] *** abackos has quit IRC
[19:21:27] <richlowe> darrenm: for what it's worth, unless I'm missunderstanding what plocher is saying, I think his scheme would quickly lead to madness.
[19:21:32] * richlowe has given up trying to make that point, however.
[19:22:24] *** tsoome has joined #opensolaris
[19:23:27] *** alobbs has quit IRC
[19:25:08] <onbot> commit by David Marx:  6498366 dadk should use d_severity instead of d_action for GDA_RECOVERABLE and GDA_FATAL in updating kstats.
[19:26:11] <trygvis> anyone know of a good resource on how to tune postgresql on solaris and sun hardware?
[19:28:18] *** wizard123 has joined #opensolaris
[19:28:20] <wizard123> hi
[19:28:58] <wizard123> is default jds theme available for download?
[19:29:30] <wizard123> ( i couldn't ask more stupid question :P )
[19:29:37] <quasi> trygvis: http://www.solarisinternals.com/wiki/index.php/Application_Specific_Tuning#PostgreSQL
[19:29:38] <mihaic> trygvis: try starting here: http://www.sun.com/software/solaris/howtoguides/postgresqlhowto.jsp
[19:30:12] <trygvis> thanks!
[19:30:35] *** abackos has joined #opensolaris
[19:30:45] <mihaic> Pretty cool that Sun supports PostgreSQL :)
[19:32:40] <trygvis> yeah, too bad I can't get it to perform
[19:33:09] <mihaic> Well, what are you using it for?
[19:33:45] <quasi> trygvis: how many processors?
[19:33:54] <trygvis> 8, 16 cores
[19:34:11] *** bp has joined #opensolaris
[19:34:17] <trygvis> I have a single application inserting 15 million rows every 10 second :)
[19:34:57] *** mazon is now known as Mazon
[19:35:02] <mihaic> Ugh...
[19:35:33] <mihaic> And what are you compareing performance to? Oracle? MySQL?
[19:35:53] <trygvis> oracle
[19:36:14] <dwc-> 1.5 million rows a second? uh... faster disk, more memory
[19:36:39] <trygvis> but neither was pushing the IO which was a bit weird
[19:37:36] <trygvis> it's a v890 with 12 drives and 16GB memory IIRC
[19:37:37] *** aqquariuss has quit IRC
[19:37:54] <mihaic> Hm.. so compared to Oracle performance is crap, or both perform poorly?
[19:38:12] <mihaic> If so, you could be hitting some other bottleneck then disk speed or RAM..
[19:38:14] <trygvis> oracle was crap, postgres was even more crap
[19:38:34] *** bondolo has quit IRC
[19:38:35] <trygvis> yeah, that's what I was wondering about too, but I have no idea what the problem should be
[19:38:53] <trygvis> but, we're just getting started on the application so there are still a few points to look at
[19:39:02] <trygvis> I just wanted to get the basic tuning right
[19:39:11] <mihaic> Well, unless you manage to isolate the problem somehow there's little more you can do.. throwing hardware at the issue might help, but i will cost :)
[19:39:32] <cap_> dtrace!
[19:39:35] <mihaic> Hell, give MySQL a spin just to see what's going on, if there's some other kind of bottleneck..
[19:39:53] <trygvis> no way I'm infecting the server with mysql
[19:40:17] <mihaic> Hahaha, glad to see MySQL is just as loved as Linux around here :)
[19:40:21] <quasi> trygvis: both postgres and mysql run into problems on a 16 core box
[19:40:22] <cap_> trygvis, just curious, what was the diff between oracle and postgres (roughly)
[19:41:39] <andersmo> trygvis: there are lots of reasons why a database would perform... suboptimally with such data rates. any indexes on the table? what's transaction rate? how are the tables partitioned, if at all?
[19:42:06] <andersmo> "what's transaction rate" - uhm, rather, "how often do you commit"?
[19:44:09] <mihaic> Here's a realy beaut: what PostgreSQL version are you using?
[19:44:21] <mihaic> Give PostgreSQL 8.2  spin. It has embedded DTrace probes in the src.
[19:46:02] *** whaq has quit IRC
[19:47:29] <mihaic> http://pgfoundry.org/docman/view.php/1000163/230/PostgreSQL-DTrace-Users-Guide.pdf
[19:47:58] <mihaic> http://pgfoundry.org/projects/dtrace/
[19:52:16] *** sahafeez has joined #opensolaris
[19:52:41] <Triskelios> something's weird with opengrok... http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/pkgdefs/SUNWagp/postinstall isn't displaying the latest revision (compare with http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/pkgdefs/SUNWagp/postinstall?r=3236 )
[19:56:15] *** natlwalker has joined #opensolaris
[19:57:42] <natlwalker> anyone here?
[19:58:10] <sahafeez> no
[19:59:57] <quasi> natlwalker: haven't seen anyone for a while - maybe he doesn't do irc anymore ;)
[20:01:16] *** estibi has joined #opensolaris
[20:01:23] <natlwalker> lol
[20:02:52] *** calumb has quit IRC
[20:02:56] *** natlwalker has quit IRC
[20:04:05] <stevel> triskelios: what's wrong?
[20:04:08] <stevel> looks right to me
[20:04:17] *** jsubl2 has joined #opensolaris
[20:08:31] <alanc> Triskelios: would it be you who just filed the agpgart bug then?
[20:11:57] <Triskelios> uh, yeah. and I just rebuilt agpgart/agpmaster/agptarget/i915 with the PCI IDs added on my laptop a couple of minutes ago and it worked
[20:12:58] <alanc> heh
[20:12:59] <Triskelios> stevel: just wondering why it was displaying the previous revision by default (I had to get the current one by specifying the revision)
[20:14:33] *** axxl has quit IRC
[20:14:43] *** pip` has joined #opensolaris
[20:16:17] <stevel> triskelios: it should be displaying the current rev
[20:16:20] <stevel> what diffs do you see?
[20:16:39] <onbot> commit by bm42561:  6461633 DR cleanup
[20:16:46] *** broadcast has joined #opensolaris
[20:17:51] <Triskelios> stevel, http://src.opensolaris.org/source/diff/onnv/onnv-gate/usr/src/pkgdefs/SUNWagp/postinstall?r2=3236&r1=2171 - it's displaying r=2171
[20:19:02] <Triskelios> alanc: I'm not sure why the i810 driver now grabs 12MB memory first and reloads int10 before noticing VideoRAM is 32768 here
[20:19:11] <Triskelios> also it apparently doesn't have DRI support
[20:19:25] <alanc> I don't know
[20:19:46] <alanc> the one in Solaris should have DRI support if the kernel dri driver recognized the card
[20:20:07] <alanc> I don't think i945 is recognized by the kernel DRI yet though
[20:20:12] <Triskelios> alanc: my modified i915 is loaded and /dev/dri/card0 exists
[20:20:40] <stevel> triskelios: odd... i see it now
[20:20:47] <stevel> can you email opengrok-discuss with that and report it as a bug?
[20:20:51] *** _william_ has joined #opensolaris
[20:20:58] <alanc> hmm - dont' remember if we tried DRI on the i945 or were just testing basic Xorg 2D on it
[20:21:37] <_william_> hi all
[20:22:42] *** dunc has joined #opensolaris
[20:23:58] *** pseudo_ has joined #opensolaris
[20:25:26] <Triskelios> alanc: http://www.opensolaris.org/os/community/on/flag-days/pages/2006101501/ - this was merged a while back afaik..
[20:25:35] *** wizard123 has left #opensolaris
[20:26:24] <alanc> right - I know DRI's there for i845->i915, just thought they hadn't pulled in the i945 & i965 yet
[20:30:09] <alanc> I filed http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6497788 about them having too old of a DRI version for the Xorg 7.2 i810 driver to use
[20:30:50] <Triskelios> ah. I'm using Xorg 6.9 circa b53 fwiw
[20:32:50] <Triskelios> it seems to have I810DRIScreenInit, I assumed it wasn't being called because of the lack of video memory.. it might still be failing the check if it's looking at the 12M it initially allocates
[20:34:22] <Triskelios> s/I810/I830/
[20:37:22] *** xerror has left #opensolaris
[20:40:13] *** noyb_ has joined #opensolaris
[20:46:51] *** pseudoXh4 has quit IRC
[20:48:40] <Triskelios> stevel, I'll post to og-discuss in a minute
[20:49:52] *** jlc has quit IRC
[20:55:17] *** noyb has quit IRC
[20:55:20] *** TrogL has joined #opensolaris
[20:55:55] <TrogL> I've got a kstat question.  Is it valid to have empty members of a kstat chain?
[20:56:17] <TrogL> I've found instances in the hme0 driver and segmap
[20:56:42] <TrogL> netstat -k is not dealing with it gracefully
[20:56:56] *** broadcast has quit IRC
[20:56:56] *** stevel has quit IRC
[20:56:56] *** slowhog has quit IRC
[20:56:56] *** peteh has quit IRC
[20:56:56] *** darrenm has quit IRC
[20:56:57] <TrogL> and it's causing me havoc in net-snmp
[20:57:17] *** stevel has joined #opensolaris
[20:57:18] *** ChanServ sets mode: +o stevel
[20:58:23] *** Theoden-Nexenta has joined #opensolaris
[20:59:34] <TrogL> I can post sample code of my kstat problem if anybody's interested.  I think it's a bug
[20:59:55] <TrogL> failing that, which opensolaris mailing list do I post it to?
[21:00:48] * richlowe would guess networking-discuss or opensolaris-bugs
[21:02:10] <elektronkind> do you have to do something extra when installing SUNWCall to get a working en_US.UTF8 locale?
[21:02:34] <richlowe> Specify it when you're selecting locales, I think.
[21:02:35] <elektronkind> locale -a doesn't list it, but the l10 lib for it appears to be there.
[21:04:08] *** pitty has joined #opensolaris
[21:04:41] *** TrogL has left #opensolaris
[21:04:58] <elektronkind> funny thing is that en_CA.UTF8 is listed
[21:09:06] *** logic_ has joined #opensolaris
[21:09:11] *** RaD|Tz has quit IRC
[21:12:58] *** kFuQ has quit IRC
[21:14:47] *** uncertainty has quit IRC
[21:14:56] *** uncertainty has joined #opensolaris
[21:15:33] *** msupino has joined #opensolaris
[21:24:55] *** dlynes_laptop has joined #opensolaris
[21:46:38] <cmihai> I'm having major issues trying to mount my "always just worked" iPod.
[21:47:04] <cmihai> rmformat reports it as /dev/rdsk/c2t0d0p0 after a LOOONG time
[21:47:53] <cmihai> But with mount -F pcfs -o hidden it just doesn't work.
[21:48:04] <cmihai> mount: /dev/dsk/c2t0d0p0 no such device
[21:49:40] <cmihai> And even that just takes forever
[22:00:59] *** __william__ has joined #opensolaris
[22:02:41] <trygvis> cmihai: perhaps you have to run devfsadm -v or something
[22:02:49] <trygvis> or the device might have changed, check dmes
[22:02:52] <trygvis> dmesg
[22:03:08] <trygvis> quasi: problems how? there is only one thread insterting
[22:03:21] <trygvis> cap_: oracle was about 10 times faster
[22:03:50] <trygvis> andersmo: once for the entire set of rows .. tried to split it up into smaller bunches with not difference. and it really have to be a single batch
[22:04:52] <cap_> trygvis, then how did they both suck? (if the diff was 10x)
[22:05:19] <trygvis> neither managed to use any of the IO capacity
[22:05:49] <cap_> newbie question: how do I get something similar to linux "cat /proc/cpuinfo" in opensolaris?
[22:05:52] <jmcp> cmihai: I think you need to use the c2t0d0p2 device instead - p0 is the entire disk which .... amazingly enough ... isn't always the slice the vold (spit!) wants to use
[22:06:03] <trygvis> cap_: psrinfo -v
[22:06:16] *** logic_ has quit IRC
[22:06:21] <cap_> thank you =)
[22:08:07] *** _william_ has quit IRC
[22:08:43] *** logic_ has joined #opensolaris
[22:11:16] *** william__ has joined #opensolaris
[22:12:35] <andersmo> trygvis: no indexes or partitioning?
[22:12:59] <trygvis> no, but no need either
[22:13:48] <andersmo> trygvis: generally I'd suspect internal lock contention on data pages and similar to hinder scalability here.
[22:14:30] <trygvis> why? there is only a single thread, the total cpu usage is 2% and I'm getting 2MB/s
[22:14:37] <trygvis> with pgsql
[22:17:00] <andersmo> no network roundtrips?
[22:18:55] <cap_> time to bring out dtrace I say :-)
[22:19:03] <trygvis> used ~70kB/s
[22:19:04] <quasi> trygvis: I'm not quite sure, I've just heard that from someone who has run really, really big datamarts on postgres
[22:19:18] *** __william__ has quit IRC
[22:19:31] <quasi> trygvis: but I wouldn't be surprised if there was something in the locking like mysql suffers from
[22:19:55] <trygvis> but still, it wouldn't block as there is only a single thread inserting
[22:21:49] <quasi> so maybe that thread can't keep up with the load?
[22:21:56] <andersmo> trygvis: generally, inserts into a database aren't as cheap as just writing to disk. There's logging, there's concurrency control (page latches as well as row/table locks - which is done even without other threads). General databases aren't very good at handling lots of streaming data. BerkeleyDB has a "queue" access method which is optimized for serial inserts and access.
[22:22:23] <trygvis> I'm fully aware of that, but that is not really the issue here
[22:22:37] <andersmo> then what is really the issue? =)
[22:22:47] <trygvis> would I be asking here if I kew? ..
[22:23:09] <andersmo> Then how are you sure it isn't an issue? ;)
[22:23:10] <trygvis> my laptop outperforms a $120k machine
[22:23:43] *** logic_ has quit IRC
[22:23:43] <andersmo> oracle on you laptop outperforms oracle on the big iron?
[22:23:48] <trygvis> no, postgres
[22:24:03] *** bondolo has joined #opensolaris
[22:24:20] <quasi> trygvis: I helped a guy benchmark mysql on a t2000 - he had the exact same performance description as you
[22:24:39] <quasi> trygvis: which turned out to be locking that went bonkers with many procs
[22:25:04] <trygvis> yeah, mysql in particular is known for that. but I only have a single inserting thread/process
[22:25:13] *** william__ has quit IRC
[22:25:25] *** _william_ has joined #opensolaris
[22:25:30] <quasi> trygvis: but it should be fairly simple to get a better ide - try to disable some of the processors and see if it helps
[22:25:31] <andersmo> trygvis: yeah, but locking is _still_ costly. Because it's pessimistic - you acquire and release locks anyway.
[22:25:40] *** logic_ has joined #opensolaris
[22:25:51] *** pip` has left #opensolaris
[22:26:16] <trygvis> quasi: ok
[22:26:23] <trygvis> andersmo: but then it would use either CPU or memory
[22:27:34] <quasi> trygvis: psradm
[22:27:57] <trygvis> I'll try it out tomorrow
[22:28:36] <andersmo> trygvis: and acquiring a lock is usually done using an atomic memory access (like test-and-set, or compare-and-swap). On multi-processors those are expensive because the hardware has to make sure no other procs have touched the memory while it is working - with the associated pains with cache invalidations and bus activity, etc. But anyway, it isn't necessarily related to your case. =)
[22:28:56] <trygvis> andersmo: it's definitely not
[22:29:25] <quasi> trygvis: getting just as bad a performance with just one proc?
[22:29:38] <trygvis> can't try until tomorrow, I'm home now
[22:30:03] <andersmo> trygvis: have you taken a look at the dynamic tracing guide's chapter on the sched provider, and tested the sample scripts there? They can help you look at how processes and threads spend their time - running, being preempted, waiting for kernel- or user level locks, etc.
[22:30:34] <andersmo> trygvis: or just look at the microstate accounting info from prstat, which generally collects the same data.
[22:31:03] <trygvis> as I said before, there are other obvious stuff I have to look at which is more likely to be the issue
[22:32:09] <quasi> if you fiddle with the number of procs, try looking at lockstat while you do
[22:32:45] <trygvis> will do
[22:33:14] *** pseudo_ has quit IRC
[22:34:33] *** bondolo has quit IRC
[22:35:39] <Teknix> qlc is working now with the vanilla qlogic hba, but mpxio is not.  Should it be able to?
[22:35:53] <Teknix> I find conflicting information
[22:36:19] <logic_> blist
[22:36:44] <Auralis> mpxio works when your device path has a fcp device, not a socal, qlc etc
[22:36:55] <jmcp> Teknix: what's the model number of the hba?
[22:37:07] <jmcp> x86 or sparc, btw?
[22:37:08] <Teknix> 2340
[22:37:09] <Teknix> x86
[22:37:39] <jmcp> mpxio should be on by default then ... have a look for "mpxio-disable" in your /kernel/drv/fp.conf file
[22:38:22] <Teknix> yeah, it's enabled by default in fp.conf, but it doesn't seem to load (nothing in dmesg about it).. and there are still multiple paths to the same LUNs
[22:38:34] <jmcp> what is the exact line in fp.conf ? cut-n-paste please
[22:38:51] *** bondolo has joined #opensolaris
[22:38:59] <Teknix> mpxio-disable="no";
[22:39:22] <jmcp> and you ran    /sbin/bootadm update-archive   too?
[22:39:27] <Teknix> yeah
[22:39:52] <jmcp> use rafb.net/paste and show me your output from "iostat -En"
[22:39:53] <Teknix> on s10 6/06, fcinfo hba-port -l shows the card info
[22:39:57] <Teknix> on nexenta it's not
[22:40:01] <jmcp> oh !!
[22:40:08] <jmcp> that could be a problem then
[22:40:23] <jmcp> are you running nexenta now, or SX or S10 ?
[22:40:25] <Teknix> i would assume so, but i'm trying to understand why
[22:40:47] <Teknix> the machine i'm trying to get this to work on is now running nexenta with a b55 kernel
[22:40:55] <Teknix> the driver is SXCR b54
[22:41:18] <jmcp> what's the output of       modinfo | grep SunFC    show ?
[22:41:34] <Teknix> 141 fffffffff0065000  d2eb0  98   1  qlc (SunFC Qlogic FCA v20060807-2.17)
[22:41:34] <Teknix> 146 ffffffffeff2b000   91d8  77   1  fcip (SunFC FCIP v20061207-1.56)
[22:41:34] <Teknix> 147 fffffffff00c8000   a138   -   1  fctl (SunFC Transport v20061207-1.64)
[22:41:34] <Teknix> 148 fffffffff00d1000  19a30  78   1  fcp (SunFC FCP v20061207-1.154)
[22:41:34] <Teknix> 151 fffffffff00bf000   56e0  96   1  fcsm (SunFC SAN Management v20061207-)
[22:41:35] <Teknix> 155 fffffffff010d000  18c08  64   1  fp (SunFC Port v20061207-1.88)
[22:41:52] <jmcp> that looks up to date to me
[22:42:05] <jmcp> what storage are you attached to?
[22:42:13] <Teknix> it's an infortrend array
[22:43:08] <Teknix> x4100 with two of the qlogic HBAs, attached to two separate qlogic sanbox 5200s
[22:43:25] <Teknix> the array has two FC ports, and each of those attach to a port on the two 5200s
[22:44:20] <jmcp> ok, here's how it works - if your device is recognised by the scsi_vhci driver as using the T10 TPGS or ALUA behaviours, then it should "just work" (ie, be enumerated) with mpxio. If it is not recognised, then it won't be enumerated with mpxio
[22:44:53] <Teknix> the only thing i had to do was add the PCI ID to /etc/driver_aliases
[22:44:56] <jmcp> which model infortrend?
[22:45:02] <jmcp> that's the hba, not the array
[22:45:33] <Teknix> yeah.. hang on, i'll pull up the model # for the array
[22:47:04] <jmcp> ff-r/s2021-{4|6} ?
[22:47:37] <Teknix> A16FG2422
[22:48:00] <cap_> jmcp, how does the scsi_vhci driver figure that out? static list of devices or can it be probed?
[22:48:39] <onbot> commit by narayan:  6422488 vdc should read/write VTOC & GEOM to a disk on file; 6483975 vDisk server should bypass lofi to access disk image files; 6492690 vds unable to open zfs volumes on boot; 6505765 vntsd crashes while in a tight loop of bind/unbind a guest domain
[22:49:22] <Teknix> jmcp: http://www.infortrend.com/main/2_product/a16f-g2422.asp  if it helps
[22:49:41] <jmcp> ta much
[22:49:51] <jmcp> cap_: I don't think I'm allowed to say :(
[22:50:04] <cap_> allowed?
[22:50:10] <jmcp> correct
[22:50:15] <cap_> explain
[22:50:32] <jmcp> I would if I could
[22:51:21] <cap_> the scsi_vhci driver is available in source form isn't it?
[22:51:34] <jmcp> Teknix: that web page is annoying - all their text is underlined!
[22:51:48] <Teknix> yeah, it is annoying indeed
[22:53:06] <Teknix> these arrays use a similar firmware to what's in the 3511
[22:53:42] <jmcp> Teknix: I don't see anything in the docs available on infortrend.com for that unit which indicate whether it uses active/active or active/passive for failover ... if it does failover at all
[22:54:41] <sommerfeld> cap_: actually, it wasn't part of the early releases.  i believe the code is being refactored to allow the bulk of it to be open-sourced (not sure if that has happened yet..)
[22:55:47] <Teknix> jmcp: failover for the FC ports you mean, or for the controller?  This unit only has one controller, with two FC ports on it
[22:56:43] <jmcp> failover for the ports
[22:57:03] <jmcp> sommerfeld: it's still in progress ... a slooooow thing, too, unfortunately
[22:57:20] <Teknix> well, i'm still trying to figure that out myself.. I'll contact the vendor
[22:57:48] *** mrdeviant has quit IRC
[22:58:11] <jmcp> Teknix: you can log an RFE via your sun support contract or via bugs.opensolaris.org requesting that support be added for that array and others from infortrend
[22:58:27] <Teknix> ok
[22:58:28] <cap_> sommerfeld, but scsi_vhci is in opensolaris (including eg. belenix) I though that meant it was open (as in source available), seems like I could learn something here, time to explain?
[22:58:59] <jmcp> no, it's not open at all. I think what you're seeing there is the binary-only bit known as "closed-bins"
[22:59:51] <sommerfeld> so, there are (at least) three distinct types of code: 1) open source.  source and binaries freely available.  2) redistributable binary.  you can redistribute the binary, but source isn't available.  3) not redistributable.  binary only; only found in sun's solaris distribution.
[22:59:56] <sommerfeld> closed-bins is #2.
[23:00:00] <cap_> jmcp, see, there's an explaination you could provide =)
[23:00:01] <sommerfeld> we're working to move stuff out of there
[23:00:17] <sommerfeld> but it took many many years to pull all this code together in the first place
[23:00:27] <cap_> got it
[23:00:45] <cap_> is there some nice overview of what is #1 and what is #2 in opensolaris currently?
[23:01:09] <jmcp> cap_: clearly, *I* am #1 in OpenSolaris ... if Time mag. is to be believed :)
[23:01:14] <jmcp> ask a better question ....
[23:01:17] <Teknix> jmcp: thanks for the help
[23:01:23] <jmcp> Teknix: you're welcome
[23:01:32] <cap_> jmcp, hehe, obviously
[23:01:53] <cap_> jmcp, #1 and #2 referred to sommerfeld's 1, 2 and 3 types of software
[23:01:58] <Teknix> is there a reason the card wouldn't show up via fcinfo on nexenta, given that it does on 6/06 ?
[23:02:24] *** pikapika has quit IRC
[23:02:27] <jmcp> Teknix: there probably is, but I don't what that reason might be
[23:02:33] <cap_> the weather here is truely fubar *sighs*
[23:02:52] <sommerfeld> and in general we don't release code you won't be usefully buildable, so as things are better pared apart more code will be put out there..
[23:03:17] <Peanut> We've got an official 'weather alert' over here, cap - major storms over Europe. (Netherlands here)
[23:03:34] <sommerfeld> cap_: last I checked, if you wanted winter, you had to go to Colorado.
[23:03:40] *** pikapika has joined #opensolaris
[23:03:56] *** abackos has quit IRC
[23:04:09] *** jsubl2 has quit IRC
[23:04:10] <cap_> sommerfeld, well don't come here, the winter we ordered got stuck in customs or something
[23:04:11] <jmcp> Teknix: have a look @ http://www.jmcp.homeunix.com/roller/resources/jmcp/Getting_to_know_the_SAN_stack.pdf - there's some debugging stuff in there which you might find useful
[23:04:33] <jmcp> Teknix: I don't work on the san stuff anymore btw
[23:05:53] *** estibi has quit IRC
[23:06:32] *** estibi has joined #opensolaris
[23:08:07] <Teknix> thanks
[23:09:50] *** estibi has quit IRC
[23:10:19] <hile_> morning jmcp
[23:10:33] *** bougie has quit IRC
[23:11:10] <jmcp> hi hile_
[23:11:33] <hile_> he lives ;)
[23:11:41] <jmcp> :)
[23:11:49] <jmcp> I think J is still getting over the jetlag
[23:15:33] <hile_> I'm not looking forward to that part of our upcoming trip
[23:15:58] <jmcp> yeah ... it's a helluva thing
[23:16:00] <hile_> on the plus side my passport finally showed the fuck up
[23:16:06] <jmcp> that's handy
[23:16:08] <hile_> did your parcel show up yet?
[23:16:13] <jmcp> nope :(
[23:16:15] <hile_> waht the fuck?
[23:16:34] <hile_> admittedly, i don't write all that neatly, but damn....
[23:16:41] <_syphilis_> is there a particular version of X that's recommended for building realvnc on solaris?
[23:16:55] <_syphilis_> (XFree86 4.3.0 doesn't seem to build successfully)
[23:17:13] <jmcp> when we were waiting to checkin @ SFO some bloke showed up without a passport. we heard the staff member behind the counter say "Sir, if you want to travel internationally, you *must* have a passport" .... and he+wife+bazillion pieces of baggage had to go away
[23:17:29] <Tpenta> o dear
[23:17:39] *** nrubsig has joined #opensolaris
[23:17:39] *** ChanServ sets mode: +o nrubsig
[23:17:50] <nrubsig> !seen kupfer
[23:17:53] <Drone> kupfer (kupfer!i=kupfer@nat/sun/x-1321d2c8f47a88d1) was last seen in #opensolaris on Thu 21 Dec 2006 19:56 GMT, saying 'Though I can't remember which driver at the moment...'.
[23:17:55] <jmcp> Tpenta: this couple didn't look like they'd always lived in the USA, either
[23:20:16] <Tpenta> roland, he said he'll be around shortly
[23:21:02] <sahafeez> jmcp: sound like someone i would think about sterilizing before they breed.
[23:21:13] <nrubsig> Tpenta: ok
[23:21:37] <sommerfeld> jmcp: on the other hand, it's generally easier to tell that by sound (accent) than by sight...
[23:21:56] <nrubsig> sahafeez: ?!
[23:22:06] *** kupfer has joined #opensolaris
[23:22:06] *** ChanServ sets mode: +o kupfer
[23:22:23] <nrubsig> kupfer: Hi!
[23:22:26] <jmcp> sahafeez: they looked like they were in their mid-50s or so, heading back to $country_of_origin to visit family
[23:22:29] <kupfer> hi Roland!
[23:22:35] <Tpenta> wow, this has to be some kind of record for #chanops
[23:22:38] <Tpenta> in here anyway
[23:22:47] <jmcp> sommerfeld: the addresses on their luggage gave it away
[23:22:53] <sommerfeld> now, if $country_of_origin was Canada I'd understand.
[23:23:32] * Tpenta wonders what kind of travel agent would over look "Oh you need a passport to travel internationally"
[23:23:49] <sommerfeld> probably booked online
[23:23:51] <richlowe> afternoon kupfer, Alan (though it's not afternoon there).
[23:23:57] <sahafeez> great, that means they had kids.
[23:24:05] <kupfer> hi Rich
[23:24:15] <jmcp> sommerfeld: try a little further west of Australia .... had some US involvement during the 50s, 60s and 70s .....
[23:24:25] <jmcp> hi rich, kupfer
[23:24:58] <sahafeez> i think that the human race is getting dumb to do regulation. we are protecting all the dumb people to breading age and passing on the dumb genes.
[23:26:01] <quasi> Tpenta: we get lots of joy with that around .eu too - a passport is not required to fly between eu countries, but a passport is the only kind of id that the airlines accept
[23:26:32] <sahafeez> when i lived in the uk they took my drivers card
[23:26:47] <jmcp> did they give it back?
[23:26:56] <sahafeez> ha. yes.
[23:27:45] <richlowe> I'm told the UK is fussy with drivers license stuff.
[23:28:14] <sommerfeld> Can't have been too fussy.  they let me drive a rental car with a massachusetts license.
[23:28:23] <richlowe> well, everywhere is fussy, but I'm told the UK is fussier.
[23:28:26] <Theoden-Nexenta> Tpenta: Probably one on the Mexican/American border
[23:28:32] <richlowe> sommerfeld: nice to visitors, mean to people moving there.
[23:28:33] <sahafeez> now the last time i was in the EU (2001) it was interesting because they had line for EU passports, USA/AU/CA passports and all others.
[23:28:37] *** estibi has joined #opensolaris
[23:28:56] <sahafeez> the all others was long. they looked at my US passport for 1 sec and waved me on
[23:29:06] <richlowe> sommerfeld: the US lets you drive on the old british non-photo licenses, though trying to use them as ID takes the better part of 20 minutes explanation per-time.
[23:37:09] <sahafeez> yah.
[23:37:31] <sahafeez> i really wish we would just have a national licenses id hear in the states
[23:43:09] <onbot> commit by mb158278:  6411129 ssc050_get_port_bit does not handle all return values of ddi_get_soft_state





top