[00:03:11] <clergyman> Is there a problem with this method that I'm not seeing: http://ritwikghoshal.blogspot.com/2008/04/today-i-mount-my-ntfs-partitions-in-my.html ?
[00:04:13] <clergyman> Other sources seem to suggest far more introcate and complex solutions to mounting an NTFS volume, that's like four steps compared to this http://blogs.sun.com/gerhardhofweber/resource/files/how_i_did_ntfs-3g-15012-on-opensolaris.txt
[00:05:58] *** niq has quit IRC
[00:05:59] *** nrubsig has joined #opensolaris
[00:05:59] *** ChanServ sets mode: +o nrubsig
[00:06:10] <nrubsig> !seen gman
[00:06:38] <nrubsig> Anyone with adminstrative access to opensolaris-announce@ around ?
[00:08:16] *** tek-ops has joined #opensolaris
[00:09:49] <tek-ops> is there an upgrade path to go from snv_93 to opensolaris 2008.5?
[00:09:59] <tek-ops> 2008.05, rather
[00:10:25] <clergyman>  /topic
[00:10:26] <clergyman> ops
[00:10:44] <Stric> tek-ops: from sxce to osol2008? I'm guessing no..
[00:11:04] <tek-ops> why is that?
[00:11:16] <Stric> different packaging system etc
[00:11:29] <tek-ops> cant you use ips with sxce?
[00:11:46] <Stric> not that I know of.. it's pure svr4
[00:13:45] <Auralis_> no ips on sxce
[00:14:00] *** vmlemon_ has quit IRC
[00:14:06] <hile_> heya Auralis
[00:14:12] <Auralis_> hi there
[00:14:47] <tek-ops> and you cant do a liveupgrade from sxce?
[00:14:53] *** vmlemon_ has joined #opensolaris
[00:16:56] *** nrubsig has quit IRC
[00:20:33] <CosmicDJ> there's no lu in 2008.xy
[00:22:52] *** Tobbe is now known as Tobbe|autoaway
[00:23:08] *** MeP3aBeu has quit IRC
[00:25:53] *** wewek has left #opensolaris
[00:26:45] *** artiflo has quit IRC
[00:28:36] <clergyman> http://rafb.net/p/VFpVcp75.html
[00:28:40] <clergyman> opensolaris made a boo boo :(
[00:30:02] <clergyman> Or the packager for all I know
[00:30:49] *** art-vandelay has joined #opensolaris
[00:34:28] *** mustang has quit IRC
[00:34:58] <Auralis_> the pkg manager
[00:36:33] <clergyman> Seems to be only on the IPSmysql5 package, the other two installs fine separately
[00:37:31] *** blistov3 has quit IRC
[00:39:36] *** Kernel86_ has joined #OpenSolaris
[00:40:02] *** Kernel86 has quit IRC
[00:41:27] *** dripcoffeemaker has joined #opensolaris
[00:43:57] *** art-vandelay has quit IRC
[00:45:34] <dripcoffeemaker> I've got an Opensolaris b93 system that I've been having network problems with. I decided to try and fix them by replacing an old 3com card with an Intel Pro 1000 card, and it is detected as an e1000g0 by the system, but it refuses to activate the network connection.
[00:46:21] <dripcoffeemaker> I'm trying to setup a static IP so I can connect to it easily from other PCs on my lan, so I enter in the information on 192.168.1.0 network, but it just says it's  misconfigured (this is with the network gui)
[00:47:07] <dripcoffeemaker> It may be because on bootup, the system is still looking for my old network card, but I can't see it in the GUI anywhere to remove it
[00:47:21] <dripcoffeemaker> any ideas?
[00:49:06] <kleppari> try ifconfig e1000g0 plump
[00:49:11] <kleppari> or whatever the interface name is
[00:49:15] *** cypromis has joined #opensolaris
[00:49:17] <kleppari> err, plumb, not plump :P
[00:49:29] <sickness> http://blog.wired.com/photos/uncategorized/71101.jpg
[00:50:03] <clergyman> heh
[00:50:44] <dripcoffeemaker> hmm, I get a blank line
[00:51:45] *** stux|work has quit IRC
[00:51:57] *** uebayasi has quit IRC
[00:52:48] <clergyman> I must say I'm very impressed with the available repositories, not missing a single program I've needed so far :)
[00:58:02] *** stux has joined #opensolaris
[00:59:41] *** vmlemon_ has quit IRC
[01:00:35] *** Kernel86_ is now known as Kernel86
[01:01:35] <clergyman> What's the command to see what files an installed package contains?
[01:01:44] <clergyman> Like I want to list all files installed by IPSmplayer
[01:01:51] *** mega has quit IRC
[01:02:16] *** jmp242 has joined #opensolaris
[01:02:47] <jmp242> anyone know anything about debugging CIFS shares on opensolaris?
[01:02:58] <jmp242> I used sharemgr show -vp
[01:03:16] <jmp242> which shows the share as expected. I can ping between my OpenSolaris computer and my WinXP computer
[01:03:28] <jmp242> but when on XP I try and access the share I get network path not found ...
[01:03:55] *** robert3 has joined #opensolaris
[01:06:18] *** robert3 has quit IRC
[01:06:26] <TomJ> try browisng to the server with \\<IP>
[01:06:28] <TomJ> from XP
[01:06:32] <jmp242> I did
[01:06:32] <TomJ> does it show shares?
[01:06:58] *** robert2 has joined #opensolaris
[01:07:11] <jmp242> specifically, I did \\192.168.1.97 and no it doesn't show shares
[01:07:25] <jmp242> it says: an error occured reading folder:
[01:07:47] *** comay has joined #opensolaris
[01:07:52] <jmp242> says filename, directory name, or volume label syntax is incorrect
[01:08:02] *** dripcoffeemaker has quit IRC
[01:08:02] *** ChanServ sets mode: +o comay
[01:08:36] <jmp242> the xp system is 192.168.1.100 if that matters
[01:08:47] <jmp242> and they are connected to a  netgear 10/100/1000 switch
[01:09:09] <jmp242> so there should not be any network issues I could figure, especially as they can ping each othre
[01:09:26] *** Gekz has joined #OpenSolaris
[01:11:49] *** not-me-guv has joined #opensolaris
[01:12:01] <jmp242> I wonder is there is some othre network setup for solaris that is still screwed up
[01:12:12] <jmp242> I can't connect via ssh either... connection refused
[01:12:50] <TomJ> jmp242: I've only setup CIFS once. this is the guide I used to do it :  http://blogs.sun.com/timthomas/entry/solaris_cifs_in_workgroup_mode
[01:16:26] <TomJ> type svcs ssh to check what's going on with ssh
[01:16:40] <TomJ> if it says disabled, do: svcadm enable ssh
[01:16:43] <TomJ> if it says offline, do svcs -xv
[01:16:50] <TomJ> that will list all services with problems
[01:16:54] <jmp242> ok
[01:17:05] <clergyman> Whoa, I just learn something inadvertently.. The error I got before (pkg dying and creating a snapshot in /tmp/blah).. I changed working directory to /tmp/blah and then ignored that for a while, and continued to install other packages.. and it turns out all the packages I installed after that was installed in the snapshot, not in my running systems lol
[01:17:10] <clergyman> system*
[01:19:05] <jmp242> ok, looks like a bunch of services aren't starting - maybe because of the network card change
[01:20:34] <jmp242> can I just try and start the services/
[01:20:35] <jmp242> ?
[01:22:19] <TomJ> what was the driver for the old card?
[01:22:38] <jmp242> hmmm, it was a 3com card
[01:22:47] <jmp242> elexl0?
[01:22:48] <TomJ> ls -al /etc/hostname.*
[01:23:00] <TomJ> do you have a /etc/hostname.elewhatever ?
[01:23:20] <jmp242> yes, elxl0
[01:23:29] <TomJ> ok, mv /etc/hostname.elxl0 /etc/hostname.e1000g0
[01:23:31] <TomJ> then reboot
[01:23:56] <TomJ> (I assuem you're using the same IP/hostname etc on the new card as you had on the old)
[01:24:00] *** Fish has quit IRC
[01:24:01] <jmp242> no
[01:24:16] <TomJ> ok. did you reboot yet?
[01:24:17] <jmp242> I've decided to not do a loopback connection to my XP machine anymore and use the switch
[01:24:20] <jmp242> no reboot yet
[01:24:27] <jmp242> can I just delete the elxl0
[01:24:36] <TomJ> ok, edit /etc/inet/hosts and change the IP listed there against your hostname
[01:24:37] <jmp242> and create one manually for the e1000g0
[01:25:03] <TomJ> then save that and reboot
[01:25:15] <jmp242> so I don't remove the entry for elxl0
[01:25:21] <TomJ> how do you mean?
[01:25:29] *** maxote has left #opensolaris
[01:25:36] <TomJ> what does the hosts file say?
[01:26:01] <jmp242> hosts file says nothing
[01:26:08] <jmp242> blank excepted for commented out lines
[01:26:36] <TomJ> what did /etc/hostname.elxl0 say?  (now hostname.e1000g0 after the mv)
[01:26:46] *** jtmuzix has joined #opensolaris
[01:27:02] <jmp242> 192.168.2.1
[01:27:14] <TomJ> fine, just edit that then and change it there
[01:27:15] <jmp242> I want to change it to 192.168.1.97
[01:27:20] <jmp242> ok
[01:27:43] <TomJ> very odd that hosts is blank
[01:27:52] <TomJ> when back at a prompt, trying ping localhost
[01:30:30] <robert2> I think I'm having a permissions problem with gnome-terminal and xterm.  If I open a gnome-terminal window, all the stty settings are trashed/lost.  If I then try to run xterm, I get "Error 32, errno 13: Permission denied Reason: get)pty: not enough ptys".  But if I become su (after entering my password, I have to type ^J as eol), then I can run xterm or gnome-terminal and all is good.  Can anybody tell me what's going on, and how to fix this?
[01:30:58] <jmp242> is there any text editor installed on opensolaris save vi/
[01:31:14] <jmp242> I can't get it to do anything but  type gibberish - nothing like linux vi at all...
[01:31:15] <TomJ> there's vim (or vi may just be a link to vim)
[01:31:24] <TomJ> try vim file
[01:31:55] <Auralis_> opensolaris vi is vim in vi mode
[01:32:15] <jmp242> well, back errow gives me d
[01:32:29] <jmp242> I can't delete anything in a file, it just ignores me when I use i for insert mode
[01:32:41] <TomJ> strange
[01:32:44] <TomJ> sort that out later
[01:32:45] <TomJ> quit out of it
[01:32:51] <TomJ> echo "192.168.1.97" > /etc/hostname.e1000g0
[01:32:52] <jmp242> backspace does nothig... there doesn't actually seem to be a way to edit the internet files
[01:32:59] <TomJ> echo "127.0.0.1  localhost loghost" >> /etc/inet/hosts
[01:33:04] <TomJ> reboot
[01:33:31] <Dominic> jmp242: erm, x/d?
[01:33:46] <TomJ> (dont actually type reboot, type init 6)
[01:34:56] <clergyman> I have that vim issue as well
[01:35:21] <TomJ> what is your TERM set to? echo $TERM
[01:35:21] <jmp242> hrm
[01:35:32] <clergyman> xterm
[01:35:37] <jmp242> the last echo gives me /etc/init/hosts not a directory
[01:35:46] <TomJ> jmp242: /etc/inet/hosts
[01:35:51] <TomJ> e not i
[01:35:52] <jmp242> nevermind
[01:36:01] <jmp242> grr
[01:36:19] <jmp242> mine is xterm as well
[01:36:30] <jmp242> ok. so now I init 6
[01:36:42] <TomJ> yes
[01:36:47] <jmp242> ok
[01:37:25] <jmp242> This would have been easier (I might have been able to fix this), if vi wasn't being weird for me... thanks for the help, I never thought about echo >> to a file lol
[01:38:54] <clergyman> pkg install IPSnano
[01:39:07] <clergyman> nano doesn't have the same issue for me
[01:39:12] <jmp242> great - and link to pico/
[01:39:13] <jmp242> ?
[01:39:21] <jmp242> just for my memory of the name
[01:39:49] <jmp242> alias?
[01:39:54] <jmp242> does that work on solaris
[01:40:02] <TomJ> alias is a function of the shell not the OS
[01:40:03] <TomJ> so yes
[01:40:12] <TomJ> no need to rename nano to pico though
[01:40:14] <TomJ> just run nano
[01:40:40] <jmp242> I know, it's just that I always run pico on my linux boxes . . .
[01:41:01] <clergyman> nano is very similar to pico
[01:41:19] <jmp242> yea, I run nano on Mac OS X
[01:41:26] <jmp242> I just want to call it the same thing, hence alias
[01:42:28] <jmp242> I've rebooted
[01:43:24] <jmp242> where's dns stored outside of the GUI
[01:44:40] <jmp242> YES - tomJ, that fixed it
[01:44:52] <jmp242> will solaris remember this on reboots with the settings in those files/
[01:44:54] <jmp242> ?
[01:45:15] *** cypromis_ has joined #opensolaris
[01:45:35] <victori> nano is for little girls ;-X
[01:46:09] <Auralis_> cut, echo, paste, grep :)
[01:46:29] <TomJ> jmp242: yes, the settings are permanent
[01:46:33] <clergyman> victori: I'm forced to use it until I can figure out why vim is acting up :/
[01:46:47] <victori> don't worry I am just trolling
[01:47:35] <TomJ> clergyman: how are you accessing it?  ssh? local gnome-terminal ?
[01:48:07] <jmp242> hmm
[01:48:14] <jmp242> one other weirdness
[01:48:56] *** Disorganized_ has quit IRC
[01:49:32] *** not-me-guv has quit IRC
[01:50:07] <jmp242> I added dns now, but I'm getting host unreachable from gateway 192.168.1.97
[01:50:24] <jmp242> I tried putting my router into /etc/defaultrouter
[01:50:26] <jmp242> no go
[01:50:35] <TomJ> you did that before or after the reboot?
[01:50:35] <jmp242> nslookup is working now
[01:50:43] <jmp242> after the reboot now
[01:50:44] *** derchris has quit IRC
[01:50:49] *** derchris has joined #opensolaris
[01:50:53] <TomJ> route add default <routerIp>
[01:51:00] <TomJ> defaultrouter is only read on startup
[01:51:21] <jmp242> thank you bunches
[01:51:26] <jmp242> works now to the ne
[01:51:28] <jmp242> net
[01:51:35] <c00p> TomJ: It is read if you restart network/physical:default I believe as well
[01:51:36] *** aveteam has quit IRC
[01:51:42] <TomJ> c00p: ah ok, never tried that
[01:51:46] <c00p> :)
[01:51:53] <jmp242> hey, maybe I can get IPSnano installed
[01:52:01] <jmp242> for the next time I break something accidentally
[01:53:12] *** kimc has quit IRC
[01:53:36] *** cypromis has quit IRC
[01:54:03] <jmp242> hmmm, says unable to assemble image plan for that install . . .
[01:55:19] *** fr4g has joined #opensolaris
[01:55:42] <throwt> what happened to 102?
[01:55:54] <throwt> did it rape cats?
[01:56:03] <Stric> or corrupt zfs pools
[01:59:44] <houst0n-> I've not gotten around to removing it yet ... so far so good
[01:59:46] <houst0n-> ;)
[02:00:02] * houst0n- keeps fingers crossed
[02:00:03] *** cmihai` has quit IRC
[02:01:37] *** cmihai` has joined #OpenSolaris
[02:01:51] *** cypromis_ is now known as cypromis
[02:03:22] *** Kernel86 has quit IRC
[02:03:28] *** Kernel86 has joined #OpenSolaris
[02:05:40] *** fr4g has quit IRC
[02:08:44] *** fr4g has joined #opensolaris
[02:16:37] *** Kernel86 has quit IRC
[02:16:42] *** Kernel86 has joined #OpenSolaris
[02:17:30] *** syamajala has quit IRC
[02:18:05] *** jmp242 has left #opensolaris
[02:25:25] *** mustang has joined #opensolaris
[02:27:06] *** not-me-guv has joined #opensolaris
[02:31:46] *** cypromis has quit IRC
[02:37:08] *** cypromis has joined #opensolaris
[02:38:21] *** cypromis has quit IRC
[02:40:30] *** AxeZ has quit IRC
[02:42:33] *** clyons has quit IRC
[02:42:38] *** estibi_ has quit IRC
[02:42:53] *** clyons has joined #opensolaris
[02:56:52] *** Teo`` has quit IRC
[03:16:29] <clergyman> Hm, I think that's it.. I have a fully functional opensolaris desktop now
[03:16:39] <hile_> You poor soul
[03:16:47] <clergyman> hehe how come ?
[03:17:21] <clergyman> ooh I tell lies.. vim is still acting up and my delete key gives me tildes yaay :)
[03:32:47] *** not-me-guv has quit IRC
[03:46:59] *** jtmuzix has quit IRC
[03:47:22] *** jtmuzix has joined #opensolaris
[03:48:31] *** stux is now known as stux|away
[03:49:46] *** Kvsh- has joined #opensolaris
[03:50:01] <Kvsh-> you guys know of a good svn gui for open/solaris?
[03:51:57] <Auralis_> just grab the source for the one you like mos and see if it builds
[03:54:13] <clergyman> I give up :/ Does anyone have a fix for the delete key printing out a tilde (~) ?
[03:55:49] <clergyman> in gnome-terminal I can set compatibility mode and get it to act as backspace, but that's not good enough
[03:57:50] <clergyman> I've tried using stty erase .. but that seems to affect the behaviour of backspace as well
[03:59:35] *** Kvsh- has quit IRC
[04:10:12] *** T_B_ has joined #opensolaris
[04:14:45] *** T_B has quit IRC
[04:23:03] *** bits45 has joined #opensolaris
[04:32:32] *** servnhim has joined #opensolaris
[04:33:11] *** gewt has joined #Opensolaris
[04:33:16] <servnhim> ok...got a question...Is Sun making opensolaris to compete in the workstation market?
[04:33:28] <gewt> i managed to rbeak glib...
[04:33:32] <gewt> break*
[04:33:42] <servnhim> And if so why Solaris and not Linux?
[04:34:06] <servnhim> anyone?
[04:34:11] <gewt> ld.so.1: firefox-bin: fatal: libdbus-glib-1.so.2: open failed: No such file or directory
[04:34:25] <gewt> I did nothign to delete ti to my knowledge
[04:35:06] *** servnhim has quit IRC
[04:35:19] <gewt> any ideas as to what i cna do?
[04:35:23] <gewt> other than reinstall?
[04:35:27] *** jgracin has joined #opensolaris
[04:36:55] *** LordIllpalazo has quit IRC
[04:37:00] <gewt> reinstalling is teh only way to fix thsi isn't it?
[04:53:18] *** Odin- has joined #opensolaris
[04:58:22] <throwt> how did you break it?
[05:09:37] *** freakazoid0223 has joined #opensolaris
[05:11:56] *** tek-ops has quit IRC
[05:13:27] *** Kernel86_ has joined #OpenSolaris
[05:13:53] *** Kernel86 has quit IRC
[05:15:39] *** Kernel86 has joined #OpenSolaris
[05:17:50] *** jgracin has quit IRC
[05:17:51] *** Kernel86 has quit IRC
[05:18:10] *** Kernel86 has joined #OpenSolaris
[05:18:15] *** Kernel86_ has quit IRC
[05:20:17] *** sah-work has joined #opensolaris
[05:22:22] *** ShadowHntr has joined #opensolaris
[05:33:39] *** Odin- has quit IRC
[05:44:21] *** wesw has joined #opensolaris
[05:45:00] *** wesw has quit IRC
[05:48:23] *** wesw has joined #opensolaris
[05:55:34] *** Gekz has quit IRC
[05:58:10] *** master_of_master has quit IRC
[05:58:11] *** ranit has joined #opensolaris
[05:58:18] *** ninjaslim has joined #opensolaris
[05:59:00] *** ninjaslim has quit IRC
[05:59:28] *** ninjaslim has joined #opensolaris
[06:00:12] <ninjaslim> hey guys how does linux emulation work on solaris
[06:02:07] *** master_of_master has joined #opensolaris
[06:04:37] <e^ipi> like this: http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/uts/common/brand/lx/
[06:05:13] <ninjaslim> oh i see branding similar to bsd, i'm wondering about this because of matlab on solaris
[06:06:15] <e^ipi> not quite
[06:06:21] <e^ipi> you'd run a linux branded zone
[06:08:18] *** kovert has joined #opensolaris
[06:09:25] *** sah-work has quit IRC
[06:09:39] *** xinkeT has quit IRC
[06:11:08] *** bits45 has quit IRC
[06:12:36] *** wesw has quit IRC
[06:13:05] <bedlam> I'm getting around 2 MB/sec file copy speeds using ZFS, on opensolaris 05/08 x86. any suggestions on finding the cause?
[06:14:17] <jamesd> iostat   ... zpool iostat 2
[06:14:28] <jamesd> vmstat    ; mpstat
[06:20:57] <ninjaslim> e^ipi: i see that's interesting
[06:21:15] *** setsoru has joined #opensolaris
[06:24:19] *** Teknix has joined #opensolaris
[06:24:59] <ninjaslim> e^ipi: do linux zones generally work, i mean i know that they're most likely never necessary but for occasions like these do they work well
[06:25:39] *** setsoru has left #opensolaris
[06:25:48] *** jtmuzix has quit IRC
[06:26:00] <e^ipi> i dunno, they worked with Maple back in first year calculus for me *shrug*
[06:26:06] *** The-spiki has quit IRC
[06:26:13] <e^ipi> give it a whirl, worst that happens is it doesn't work
[06:29:54] *** ashner has joined #opensolaris
[06:35:58] <ninjaslim> i will eventually when i get my hands on matlab
[06:36:05] <ninjaslim> i actually need to try it on freebsd, supposedly it works on it
[06:36:21] <ninjaslim> it's documented in the handbook
[06:41:49] *** syamajala has joined #opensolaris
[06:48:41] *** aveteam1 has quit IRC
[06:52:35] *** fr4g has quit IRC
[06:55:51] *** syamajala has quit IRC
[07:00:51] *** gerard13 has joined #opensolaris
[07:00:54] *** Gekz has joined #OpenSolaris
[07:08:03] *** ashner has left #opensolaris
[07:11:05] *** ShadowHntr has quit IRC
[07:19:01] *** Fullmoon has joined #opensolaris
[07:21:24] *** blistov2 has joined #opensolaris
[07:21:25] <Fullmoon> When using SMF, is it even smart / necessary to daemonize?
[07:23:40] *** edgy has joined #opensolaris
[07:25:26] *** nprice_ has quit IRC
[07:33:15] *** edgy has left #opensolaris
[07:58:53] *** gerard13 has quit IRC
[08:02:38] *** pjama has quit IRC
[08:11:01] *** praks has joined #opensolaris
[08:12:06] *** praks has quit IRC
[08:14:43] *** gottadoit has quit IRC
[08:18:02] *** Gekz has quit IRC
[08:18:07] *** abisen has joined #opensolaris
[08:19:15] *** Gekz has joined #OpenSolaris
[08:31:15] <robert2> Problem: I start with a clean VirtualBox, install opensolaris, and then do the package update instructions and virtualBox guest addtions.  Everything is more or less good.  Then I upgrade to 101a using the great script at http://morettoni.net/files/updateos.  Apparently successful, but when I open a gnome-terminal, stty shows that pretty much nothing is set (so it tends to garble output because it doesn't output eol correctly.  If I try to run xterm, I
[08:35:00] *** abisen has quit IRC
[08:47:48] *** bigjohnto has quit IRC
[08:53:30] *** jamesd__ has joined #opensolaris
[08:53:53] *** tombhadAC has joined #opensolaris
[08:54:38] *** jamesd has quit IRC
[09:07:00] *** jamesd__ has quit IRC
[09:09:16] *** uebayasi has joined #opensolaris
[09:09:50] *** ranit has quit IRC
[09:18:29] *** jamesd__ has joined #opensolaris
[09:26:31] *** comay has quit IRC
[09:26:39] *** wewek has joined #opensolaris
[09:26:55] *** kovert has quit IRC
[09:31:14] *** jamesd__ is now known as jamesd
[09:31:19] *** not-me-guv has joined #opensolaris
[09:31:20] *** ChanServ sets mode: +o jamesd
[09:42:49] *** ninjaslim has quit IRC
[09:47:54] *** hsp has joined #opensolaris
[09:49:33] *** blistov2 has quit IRC
[09:56:03] *** pumpkin has quit IRC
[09:56:29] *** Trident has quit IRC
[09:56:39] *** Trident has joined #opensolaris
[10:07:49] *** Cripps has quit IRC
[10:20:38] *** abisen has joined #opensolaris
[10:22:14] <boyd> Damn... another build that I'm not gonna use...
[10:27:47] *** Openfree has quit IRC
[10:31:19] *** luna1 has joined #opensolaris
[10:31:55] *** Fish has joined #opensolaris
[10:33:40] *** Kernel86_ has joined #OpenSolaris
[10:35:42] *** Kernel86 has quit IRC
[10:35:47] *** Kernel86 has joined #OpenSolaris
[10:37:03] *** Kernel86_ has quit IRC
[10:46:37] *** Tobbe|autoaway is now known as Tobbe
[10:52:41] *** hsp has quit IRC
[10:54:42] *** tombhadAC has quit IRC
[10:58:45] *** estibi has joined #opensolaris
[11:01:08] *** Tobbe is now known as Tobbe|autoaway
[11:09:21] *** Zplay has joined #opensolaris
[11:10:15] *** Tobbe|autoaway is now known as Tobbe
[11:10:45] <Zplay> Hi, I'm a new user of opensolaris, and actually I'm trying to configure the ssh server, but I get this message when I try to connect
[11:10:46] <Zplay> no common kex alg: client 'diffie-hellman-group-exchange-sha1,diffie-hellman-group1-sha1', server 'gss-group1-sha1-toWM5Slw5Ew8Mqkay+al2g=='
[11:19:13] *** not-me-guv has quit IRC
[11:25:52] *** The-spiki has joined #opensolaris
[11:38:10] *** abisen has quit IRC
[11:39:40] *** netj has joined #opensolaris
[11:44:24] <Zplay> Now I tried to use keys to authenticate on the ssh server, but I get this message: Permission denied (gssapi-keyex,gssapi-with-mic,publickey,keyboard-interactive).
[11:45:06] <CosmicDJ> Zplay: correct permission on the keys?
[11:46:56] *** fr4g has joined #opensolaris
[11:47:20] *** netj_ has joined #opensolaris
[11:47:43] *** netj has quit IRC
[11:49:22] <Zplay> CosmicDJ: Yes I think
[11:51:34] <Zplay> This is my sshd_config http://pastebin.com/mc08c4e0
[11:58:50] <Tobbe> Is there any website that helps me choose the right OpenSolaris download?
[11:59:47] <Tobbe> I'm thinking I should get OSol 08.11RC1b, but I'm not sure
[12:00:37] <Zplay> Tobbe, you can get a look at genunix.org
[12:00:55] <Tobbe> the topic mentions ON102 and IPS101a, what are those? (I know what sxce is). The genunix page also lists SXDE, what is that?
[12:01:15] *** not-me-guv has joined #opensolaris
[12:01:58] <Zplay> sxde is developer edition whereas sxce is community edition
[12:02:13] <Tobbe> and what are the differences?
[12:02:37] <Zplay> no idea
[12:02:51] <Tobbe> and how do they relate to ON102, IPS101a and the 2008.5 and 2008.11 releases?
[12:03:51] *** Cripps has joined #opensolaris
[12:04:32] <rewolf-> from my understanding, ON refers to the kernel, and IPS to the packages.
[12:06:08] <rewolf-> (and 102 and 101a are revisions, but i guess no surprise there ;) )
[12:12:11] *** clyons has quit IRC
[12:13:47] <Tobbe> so neither ON nor IPS are a complete operating system?
[12:14:13] <trygvis> IPS is complete
[12:14:35] *** Trident has quit IRC
[12:14:43] *** Trident has joined #opensolaris
[12:15:13] <Tobbe> ok, where can I download a livedvd/install dvd based on IPS?
[12:15:18] <CosmicDJ> SXDE is deprecated
[12:15:44] <CosmicDJ> want solaris10 but have no money for support, download/install sxce
[12:15:56] <CosmicDJ> want smth for your desktop, install 2008.xy
[12:16:38] *** Cripps has quit IRC
[12:18:08] <CosmicDJ> 2008.05 and 2008.11 are supported OpenSolaris releases targeted add endusers
[12:18:10] <rewolf-> CosmicDJ: crap. so I have to reinstall my 2008.xy servers? :(
[12:18:11] <rewolf-> :)
[12:18:50] <CosmicDJ> IPS is the bleeding edge for OpenSolaris (2008.xy) users, only recommended if you know what you're doing and prepared to do some beta testing
[12:19:25] <CosmicDJ> rewolf-: well, since windows is also abused as a server os, mh no you don't ;)
[12:20:15] <rewolf-> CosmicDJ: heh, so you just compared 2008.xy to Windows? :)
[12:20:51] <CosmicDJ> hm right, I should compare OpenSolaris 2008.xy with Ubuntu Desktop Edition
[12:21:00] <rewolf-> :D
[12:21:26] <CosmicDJ> can also be used as a Server, but it's more suited for a desktop (think of gnome and all that stuff)
[12:22:16] *** Rakhun|away has joined #opensolaris
[12:22:50] <rewolf-> svcs disable *gdm* - first thing :)
[12:23:16] <rewolf-> but yes, easier with sxce on server, agreed.
[12:23:54] <rewolf-> but we still like opensolaris on the servers here :)
[12:23:56] <Rakhun|away> Hi all :) I'm trying to boot a computer with an old screen that doesn't support the update frequency that is used by the liveCD by default, is there a kernel flag or so to make it use another frequency?
[12:25:05] <CosmicDJ> rewolf-: I still hope it'll be the foundation for an X/gome/openoffice/etc.-less server edition :)
[12:26:11] <rewolf-> i think I still like /etc to be part ;)  but yes, that would be awesome!
[12:28:38] <Stric> Ok. Got an NFS server that's doing about 100MB/s disk io, no local apps responsible and it's doing about 10MB/s NFS.. any DTrace folks that know how to find out what's generating this io?
[12:29:48] <CosmicDJ> 1 gigabit ethernet?
[12:30:28] <Stric> yes.. and 1gig fc
[12:30:54] <Stric> normally no problem, but tonight (for the second time), it flipped over to "max out disks, and keep net at max 10MB/s"
[12:31:00] <Stric> no idea where the other 90MB/s goes
[12:31:41] <CosmicDJ> but you can get more than 10mb/s out of your network? did you check that with some network performance apps?
[12:31:51] <Stric> http://www.acc.umu.se/technical/statistics/ftp/monitordata/backend.html.en
[12:31:56] <Stric> Yes, I can get more
[12:32:48] <throwt> solaris 10 has x/oo/etc/gnome...
[12:33:25] <CosmicDJ> throwt: sure, but you don't have to install them; 2008.xy just installs everything
[12:37:32] <rewolf-> Stric: graph shows ftp service, you're talking about nfs ?
[12:38:08] *** hsp has joined #opensolaris
[12:38:41] <throwt> what kind of reads is it serving?  maybe they're small reads and readahead or large block sizes are f'ing you
[12:39:17] <Stric> rewolf-: the machine that has problems is the NFS backend for our FTP cluster
[12:39:47] <Stric> And I just turned "enabled" zfs_prefetch_disable, now it's doing 50-80MB/s of NFS instead
[12:45:19] <throwt> good
[12:45:42] *** clyons has joined #opensolaris
[12:46:01] <CosmicDJ> I heard that NFS may benefit from jumbo frames...
[12:46:28] <Stric> if I go back to zfs_prefetch_disable=0, then it starts doing 10MB/s NFS and 100MB/s disk again.. and this is after 21 days of _not_ working like that
[12:46:49] <Stric> CosmicDJ: It's not network related, it's ZFS
[12:46:52] <Stric> it seems
[12:47:12] <throwt> im guessing there are lots of writes and a fair amount of deletes
[12:47:16] <Stric> Nope
[12:47:18] <Stric> All reads
[12:47:42] <Stric> (right now)
[12:48:02] <Stric> Some prefetch algorithm in ZFS seems to have gone b0nkers
[12:49:39] *** Andrew2 has joined #opensolaris
[12:49:43] <Stric> So for 21 days, it has been just fine having zfs file level prefetch on.. then *snap*, it craps itself..
[12:49:51] *** The-spiki has quit IRC
[12:49:55] *** Rarok has joined #opensolaris
[12:50:20] <Stric> http://www.acc.umu.se/technical/statistics/ftp/monitordata/backend.html.en <- see spike.. just from zfs_prefetch_disable=1
[12:56:25] *** vertigo- has quit IRC
[12:56:37] <rewolf-> there is not any change in the number of clients no your system?
[12:56:53] <throwt> isnt the point 100mb of disk for 10mb of nfs?
[12:57:04] <Stric> throwt: exactly
[12:57:12] <Stric> rewolf-: nope.
[12:57:41] <Stric> if it's reading 100MB/s from disk, and only using 10MB/s of it (sending over nfs).. where does the other 90MB/s go?
[12:57:50] <Stric> 8GB of ram in the machine
[12:57:50] <rewolf-> how is the memory use in general, not any of the i/o is swap activity ?
[12:58:30] <Stric> http://pastebin.com/m79edc46b <- ::arc
[12:59:01] <Stric> http://pastebin.com/m74541f24 <- ::memstat .. no idea why it has 1.3GB free.. no need for that..
[12:59:31] <Stric> about 2.4M files on the filesystem.. so I guess a whole lot of kernel mem goes into DNLC cache etc
[12:59:53] <Stric> although only about 700k in dnlc_nentries
[13:00:04] <trygvis> DNLC cache?
[13:00:14] <Stric> directory name lookup cache
[13:00:20] <trygvis> ah
[13:00:20] <Stric> name<->inode stuff
[13:00:29] <trygvis> right
[13:07:51] <throwt> File does not exist: /var/opt/SUNWwbsvr7/admin-server/config/server.xml at /opt/SUNWwbsvr7/lib/upgradeServer.pl line 143 sjsws 7.0->u3 upgrade fails because i dont have an admin server in the global zone
[13:08:07] <Stric> Anyone know how to figure out why I have 1.3GB free? it should go into ZFS cache I guess.. c_max is 7GB and it's currently using 3GB.. so there's 4 more to claim..
[13:09:08] <Stric> this is snv98 on os2008.xx btw
[13:10:16] <CosmicDJ> did you just say you only 1GB ram left free?
[13:10:20] <CosmicDJ> +n't
[13:10:47] <Stric> http://pastebin.com/m74541f24 <- ::memstat
[13:11:03] <CosmicDJ> Free (freelist)            347245              1356   17%
[13:11:25] <Stric> why would I want 1.3GB free memory?
[13:11:36] <Stric> instead of having it in ZFS cache
[13:12:43] <CosmicDJ> how many files + size do you have?
[13:13:44] <throwt> gotta tune it, it wont use it all by default...  saves a gig or something
[13:13:49] <Stric> the archive itself is ftp.se.debian.org, se.archive.ubuntu.com, ftp.gnome.org, parts of ftp.mozilla.org etc.. so it's a mix of small tarballs and dvd isos.. about 3.7TB right now
[13:14:18] <rewolf-> Stric: have you tried the tools here: http://brendangregg.com/psio.html ? they are based on dtrace
[13:15:02] <rewolf-> (iotop and psio10)
[13:15:24] <rewolf-> probably they just list the nfsd though ;/
[13:15:47] <CosmicDJ> hm there's also smth called nfswizard or smth in the dtrace tk
[13:15:51] <Stric> I've run various stuff out of DTraceToolkit, but there is a whole lot of IO that's happening that isn't shown
[13:16:20] *** jgracin has joined #opensolaris
[13:16:55] <Stric> CosmicDJ: seems to be for the nfs client only
[13:18:05] *** Rarok has quit IRC
[13:22:03] <Berny> hmm, hmm whats "ERROR: /pci@780/SUNW,qlc@0,1: Last Trap: Fast Data Access MMU Miss" supposed to mean, when i try to boot a t1000 off an fc/al disk on a qlc hba?
[13:23:38] <Zplay> Can someone show me a working sshd_config ?
[13:24:19] *** kimc has joined #opensolaris
[13:24:32] <kimc> how can i get 'pkg' ?
[13:25:06] <CosmicDJ> kimc: install opensolaris 2008.xy
[13:25:13] <kimc> ah ok
[13:25:59] <Berny> Zplay, (open)solaris comes with a working sshd_config
[13:27:29] <CosmicDJ> Zplay: http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/ssh/sshd_config
[13:29:41] <CosmicDJ> Berny: latest Solaris release?
[13:30:22] <Zplay> Thanks Berny and CosmiDJ, but I don't understand why I get this error "no common kex alg: client [....]", I tried the original sshd_config and also tried to make mine but no more chance :(
[13:31:19] <Berny> CosmicDJ, running sol10u4 right now (fullz) patched... lucreated a second be, but cannot boot off the external disk with the second be. that error comes right from openboot, no kernel loaded yet
[13:31:46] <CosmicDJ> Berny: firmware up2date?
[13:31:52] <Berny> Zplay, you may want to check ssh_config on the client
[13:32:35] <Berny> CosmicDJ, openboot etc on the box itself are. trying to find out whats the current rev for the hba though... sunsolve is a bit laggy right now for me :-\
[13:32:50] <rewolf-> Stric: which of the tools have you been playing with, since there is something not shown ?
[13:33:59] <Stric> rewolf-: rfsio.d, rfileio.d, iostat.d, ... they just show which userland applications are reponsible.. not kernel stuff..
[13:34:24] <Zplay> Beryn, If I connect from the ssh server, it doesn't work better than another client :/ I also tried few kinds of ssh_config. And from a windows computer putty tells me "couldn't agree a key exchange algorithm......"
[13:34:45] <Berny> oh wait... how do i find out what label (smi or efi) i have on a disk?
[13:36:37] <CosmicDJ> format
[13:39:47] *** xRaich[o]2x has joined #opensolaris
[13:42:09] <Berny> where in format -e ? label only lets me choose a type not telling the current type
[13:42:56] <CosmicDJ> just look at your part. table
[13:43:28] <Berny> looks smi
[13:43:33] <Berny> (slice 0-7)
[13:43:39] <CosmicDJ> so not more than 7 parts?
[13:44:06] <Berny> nope... must be smi
[13:44:11] <CosmicDJ> yep
[13:44:18] <Berny> so it's not the efi label causing trouble than
[13:48:28] *** init07 has joined #opensolaris
[13:50:39] <rewolf-> Stric: what kind of granularity are you looking for? too deep its hard to know anyway?
[13:50:49] *** derchris has quit IRC
[13:50:53] *** derchris has joined #opensolaris
[13:51:28] *** robert2 has quit IRC
[13:51:48] *** Zplay has quit IRC
[13:55:15] <rewolf-> (eg. dtrace -n zfs\*:entry'{@[probefunc]=count();}' to get all zfs* operations)
[14:01:30] *** erflungued has quit IRC
[14:01:50] *** erflungued has joined #opensolaris
[14:03:43] *** MeP3aBeu has joined #opensolaris
[14:06:55] *** xRaich[o]2x has left #opensolaris
[14:07:31] <Stric> rewolf-: well.. hard to tell.. on a normal day with prefetch and currently without prefetch, what is read from disk is "used".. but currently (with prefetch enabled), it's reading 10x from disk than what is used..
[14:07:43] <wereHamster> this morning my opensolaris box wasn't responding anymore (not even to pings). So I reset/rebooted it and now it was responding to pings but the ssh server didn't start up (connection refused). How can I find out what went wrong?
[14:09:23] <throwt> go to the console
[14:09:34] *** vertigo- has joined #opensolaris
[14:11:13] <wereHamster> ah great, now it won't even boot up at all, just hangs after 'SunOS Release 5.11 ...\nCopyright ...\nUse is subject to license terms.'
[14:11:28] *** init07 has quit IRC
[14:12:02] <throwt> add '-kv' to the kernel line
[14:12:27] *** Rakhun|away has left #opensolaris
[14:13:52] <wereHamster> the last line is 'kb80420 is /isa/i8042@...', I don't see any apparent errors
[14:15:04] *** Kernel86_ has joined #OpenSolaris
[14:15:54] *** Kernel86 has quit IRC
[14:16:16] <wereHamster> ah well. If I just want a server (no desktop/gnome stuff at all), what should I install? solaris10, opensolaris, sxce?
[14:18:40] <codestr0m> wereHamster: I think SXCE has a cli installer which is more flexible (never used.. just see others recommend) opensolaris can be configured to use less services and disable various graphical things you need
[14:19:38] <codestr0m> uninstalling the unwanted apps in opensolaris may be a bit challenging.. you'll first have to uninstall a meta package called slim_install and from there remove what you don't want
[14:21:08] <wereHamster> yeah, I had my opensolaris installation stripped up pretty well (no more gnome packages etc).
[14:22:35] <codestr0m> wereHamster: in general for a production server.. I would say Sol 10 or SXCE.. os2008* is probably more bleeding edge than you want, but not sure if it's testing or actual production you're going to push this
[14:22:56] *** diggs-ita has joined #opensolaris
[14:23:58] <wereHamster> os2008 is more bleeding edge then sxce? "Solaris Express Community Edition is Sun's binary release... It is built from the latest OpenSolaris source and additional technology that has not been published in the OpenSolaris source base."
[14:25:21] *** CyberBlue has joined #opensolaris
[14:29:33] *** lixx has joined #opensolaris
[14:29:43] *** nachox has joined #opensolaris
[14:40:33] *** LordIllpalazo has joined #opensolaris
[14:44:14] *** Kernel86 has joined #OpenSolaris
[14:44:32] *** Kernel86_ has quit IRC
[14:47:36] <Andrew2> wereHamster: That sentence you quote is tripping over the confusing two different meanings of "OpenSolaris" in the same sentence.
[14:49:34] *** not-me-guv has quit IRC
[15:00:28] *** IvanR__ has quit IRC
[15:02:31] *** shl_ has quit IRC
[15:05:39] *** gonzzor has joined #opensolaris
[15:10:42] *** IvanR_ has joined #opensolaris
[15:11:23] *** AxeZ has joined #opensolaris
[15:13:55] *** Odin- has joined #opensolaris
[15:18:31] *** cypromis has joined #opensolaris
[15:19:22] *** Kernel86 has quit IRC
[15:19:29] *** Kernel86 has joined #OpenSolaris
[15:20:23] *** Fish has quit IRC
[15:23:10] *** netj_ has quit IRC
[15:23:16] *** netj has joined #opensolaris
[15:23:40] <gonzzor> I could use some help on solving a Netowrk problem with 2008.05.
[15:23:53] *** Kernel86_ has joined #OpenSolaris
[15:24:18] *** Kernel86 has quit IRC
[15:24:46] <gonzzor> I have just installed it but I can't get a working network. If I do svcadm enable  network/physical:nwam I get the rge0 and rge1 devices, but no IP from my DHCP (The cable and DHCP works with another computer.)
[15:24:53] *** Kernel86_ has quit IRC
[15:25:02] *** Kernel86 has joined #OpenSolaris
[15:26:27] *** gewt has quit IRC
[15:28:11] <gonzzor> If I then assign an IP manually I can ping my DHCP/gw..
[15:29:10] *** luc^ has quit IRC
[15:30:19] <gonzzor> But only for 1 packet is received, all others are lost.
[15:30:35] *** blistov2 has joined #opensolaris
[15:30:39] <gonzzor> However all packages shows up in Wireshark on the machine I ping.
[15:30:42] <gonzzor> packets
[15:30:52] <gonzzor> Any know problems I have missed?
[15:32:17] <gonzzor> I do see checksum errors for all DNS packets sent by Opensolaris
[15:33:41] *** nemoego has quit IRC
[15:33:55] <oholiks> checksum errors are probably caused by offloading functions of the nic, funny pings may have something to do with dns, or lack of
[15:34:09] *** syamajala has joined #opensolaris
[15:34:09] <oholiks> gonzzor: try ping -n
[15:34:34] <gonzzor> oholiks: Then ping works
[15:34:48] <gonzzor> I use ping -s -n 192.168.0.1, it works, but without -n it doesn't
[15:36:55] <oholiks> I dunno why, but that's how it is
[15:37:28] <oholiks> if you wait long enough, replys will probably show
[15:38:34] <gonzzor> Thanks..
[15:38:58] *** CyberBlue has quit IRC
[15:39:17] *** vertigo- has quit IRC
[15:44:43] *** xRaich[o]2x has joined #opensolaris
[15:47:22] *** jgracin has quit IRC
[15:48:07] *** jgracin has joined #opensolaris
[15:54:42] *** FrostCS has quit IRC
[15:55:12] *** Asako has joined #opensolaris
[16:02:42] *** Discordian has joined #opensolaris
[16:02:58] *** Kernel86_ has joined #OpenSolaris
[16:03:34] *** Kernel86 has quit IRC
[16:05:57] *** cypromis has quit IRC
[16:13:02] <th> gonzzor: that sounds like you dont have dns setup for PTR records for your 1.0.168.192.in-addr.arpa
[16:16:46] <CosmicDJ> or your /etc/nsswitch.conf is wrong
[16:17:40] *** dustman has quit IRC
[16:17:45] *** dustman has joined #opensolaris
[16:22:35] <gonzzor> I don't have the reverse dns for the IP, but I fixed the main issue by turining of the hw checksum for rge driver.
[16:23:51] *** netj has quit IRC
[16:29:41] *** sponix has joined #opensolaris
[16:30:01] <spiff> is there a way to create a local repository or cache for ips packages? I have to provision a lot of zones, and really dont want to download the same packages more than once..
[16:30:04] *** e1kg has quit IRC
[16:32:55] <kleppari> I think /var/pkg/download is used as a cache when installing zones
[16:33:33] *** qiyong has quit IRC
[16:35:18] *** mib_eho8mx7v has joined #opensolaris
[16:35:25] <spiff> yeah, only way I see it is to mount it i all the zones...which is a pain.
[16:38:35] <spiff> there is a lot of missing packages in the default ipkg zone install... java, less and perl to name some..
[16:38:49] <spiff> is there a bigger metapackage to grab the lot?
[16:44:31] *** vmlemon_ has joined #opensolaris
[16:48:06] <Asako> think I found a bug in virt-manager
[16:48:43] <Asako> when I use an http install source it looks for images/linux which doesn't exist, should be images/xen
[16:49:43] *** mib_eho8mx7v has quit IRC
[16:52:12] <Asako> iscsi seems a bit slow
[16:54:59] *** Fish has joined #opensolaris
[16:59:31] <wereHamster> where do I configure the default route (which file under /etc)?
[17:00:24] <consanguinity> /etc/defaultrouter
[17:01:24] <wereHamster> just the ip there?
[17:01:29] <consanguinity> or hostname
[17:03:16] *** [RIT]Rawn027 has joined #opensolaris
[17:03:42] <wereHamster> well, that didn't work. I'm using nwam btw (with a static ip)
[17:05:42] <wereHamster> 'route -p show' sais 'No persistent routes are defined' but when I do 'route add default 192.168.0.1' I get 'add net default: gateway 192.168.0.1: entry exists'
[17:11:50] *** dustman has quit IRC
[17:13:27] <wereHamster> wtf? Now I can't even log into the computer through ssh
[17:13:45] *** h3sp4wn has joined #opensolaris
[17:13:46] *** luna1 has quit IRC
[17:15:08] *** Kernel86_ has quit IRC
[17:16:41] <[RIT]Rawn027> wereHamster do a 'netstat -rn'
[17:16:56] *** dustman has joined #opensolaris
[17:17:01] <wereHamster> the ssh server is accepting the connection (as I see 'failed publickey for ??? from ???' on the console), but on the ssh client I never get the password prompt
[17:18:31] <CosmicDJ> wereHamster: did you change anything in sshd_config ?
[17:19:28] <wereHamster> no, I haven't touched that. And besides, the ssh server worked just fine ten minues ago (actually before I rebooted)
[17:20:21] <wereHamster> hm, I eventually get the p/w prompt, but after a couple minutes. Any idea what's taking the ssh server so long?
[17:20:37] *** stevel has joined #opensolaris
[17:20:37] *** ChanServ sets mode: +o stevel
[17:20:50] <TomJ> wereHamster: slow ssh login is usually caused by no DNS resolution
[17:21:12] <[RIT]Rawn027> has anyone gotten netatalk to compile on opensolaris or solaris 10?
[17:21:19] <TomJ> check if you can resolve names on the server, if not check /etc/resolv.conf, /etc/nsswitch.conf,  check that you can connect to the DNS server / any intenret host/ etc
[17:22:32] <wereHamster> TomJ: I can, dns/client:default is online, /etc/resolv.conf has nameservers in it, `dig` works
[17:24:39] *** kleppari has quit IRC
[17:27:50] <Asako> is there a way to see iscsi stats?
[17:28:34] <Asako> I seem to have a bit of lag
[17:29:06] *** kleppari has joined #opensolaris
[17:30:01] <wereHamster> how can I force the ethernet card to 100mbit (instead of gbit)?
[17:30:25] <spiff> what exactly is installed in a full ipkg zone? It sure isn't the 5G from the global zone...
[17:31:14] <Asako> http://pastebin.ca/1257744 does that seem right for an iscsi disk?
[17:32:30] <nachox> wereHamster, what driver?
[17:33:20] *** Odin- has quit IRC
[17:35:01] <wereHamster> nachox, nge. I think the ssh server is causing problems with the card. I never got more then one reply packet from ping (which lead to '13 packets transmitted, 1 packets received, 92% packet loss' etc). After I restarted the ssh service (svc disable/enable; restart wasn't enough) it works fine now, both login and pinging other hosts
[17:37:31] *** axisys has joined #opensolaris
[17:38:07] *** FrostCS has joined #opensolaris
[17:38:44] <spiff> does anyone know what is being installed in ipkg zones?
[17:42:05] <nachox> read nge(7D), that man page show what you needed to know
[17:49:35] *** kleppari has quit IRC
[17:50:56] <[RIT]Rawn027> so does one typically have more success compiling on OpenSolaris than Solaris 10?
[17:51:30] *** vmlemon_ has quit IRC
[17:51:34] *** vmlemon_ has joined #opensolaris
[17:52:52] *** jteo has quit IRC
[17:56:39] *** ahe has joined #opensolaris
[17:56:45] *** Odin- has joined #opensolaris
[17:58:00] *** stevel has quit IRC
[17:58:50] <CosmicDJ> [RIT]Rawn027: IIRC opensolaris introduced err() and friends, so IMHO source requires fewer modification (if any) in OSOL
[17:59:38] <[RIT]Rawn027> Thanks, because I have having a world of problems trying to get netatalk to compile on sol 10u6
[18:00:26] *** mikl has joined #opensolaris
[18:05:29] *** cypromis has joined #opensolaris
[18:06:31] *** ShadowHntr has joined #opensolaris
[18:08:03] <holcomb> i have learned a valuable lesson.  do not use the native solaris package versions of anything JES
[18:08:38] <spiff> please, someone point me to information regarding ipkg zones...
[18:09:16] <houst0n-> spiff: they're so small because they share /usr with the global zone. Try and touch /usr/bleh on one of the zones as root
[18:09:23] <houst0n-> they should contain the same packages
[18:09:53] <codestr0m> spiff: normally you inherit the packages from the global zone to answer your question. are you worried about size or in general want less installed?
[18:10:03] <spiff> oh, so I shouldnt have removed inherit-pkg-dir=/usr
[18:10:20] <spiff> I've read that ipkg zones cant be sparse, so I removed that from zonecfg
[18:10:38] <spiff> together with /lib, /platform, etc
[18:10:49] <spiff> ...not /etc :)
[18:11:26] <spiff> yet /usr is RO in the zone right?
[18:14:26] <spiff> codestr0m: I just want to be able to install packages in zones.
[18:14:26] <codestr0m> spiff: I think the defaults for the ipkg brand should be sane.. give those a try and then tweak from there
[18:14:57] <spiff> pkg install mysql in sparse zone dies on read only filesystem...
[18:15:22] <TomJ> spiff: well that would be normal for a sparse zone
[18:15:29] <spiff> exactly.
[18:15:29] <TomJ> you cant install packages inside a sparse zone, you install them from the global
[18:15:56] <spiff> Of course...but how do you then create a non-sparse zone?
[18:16:15] <spiff> and ipkg is not compatible with sparse zone, afaik
[18:16:24] <axisys> this article http://sunsolve.sun.com/search/document.do?assetkey=1-61-204591-1 recommends to play with ecache scrub value in case of cbi events on US-II cpu.. but not sure if sol 10 needs it..
[18:16:28] <axisys> any suggestion?
[18:16:29] <codestr0m> spiff: dont' inherit from the global zone. so you were on the right track at first
[18:16:49] <spiff> no I wasnt...I don't get anything installed if I don't inherit
[18:16:59] *** kleppari has joined #opensolaris
[18:17:04] <spiff> it downloads about 80 MB and calls it a day
[18:17:06] <consanguinity> axisys: did you actually read it?  "All versions of Solaris more recent than 8 have the scrubber already installed."
[18:17:11] *** TT has joined #opensolaris
[18:17:59] <codestr0m> cypromis: where's the release notes which explain how to update from snv_89 to 1xx I know it's dangerous.. I thought you had a link handy the other day and I'm having a blank on it
[18:18:01] <axisys> consanguinity: i read it.. but the article might be old.. so not sure if sol 10 would be included.. i was in doubt, that is why asking.. :-)
[18:18:20] <cypromis> codestr0m: LU ?
[18:18:47] <codestr0m> cypromis: there's release notes or a special readme on this update..
[18:18:51] <consanguinity> axisys: http://src.opensolaris.org/source/search?q=&defs=ecache_scrub_enable&refs=&path=&hist=&project=%2Fonnv
[18:19:26] <axisys> consanguinity: thnx
[18:19:30] <cypromis> you men for the os200811 upgrade ?
[18:19:31] <cypromis> ok
[18:19:35] <cypromis> cause for nevada not
[18:19:54] <codestr0m> cypromis: correct os2008*
[18:20:07] <cypromis> http://www.opensolaris.org/os/project/indiana/resources/relnotes/200811/x86/
[18:23:05] *** not-me-guv has joined #opensolaris
[18:23:14] *** galen has joined #opensolaris
[18:23:31] *** The-spiki has joined #opensolaris
[18:25:37] *** stux|away is now known as stux
[18:27:01] <galen> i feel like i'm going crazy here, but it seems my OS 2008.11 RC1 system is freezing periodically
[18:27:23] <galen> i was trying to troubleshoot why NFS was moving rapidly, then stopping completely, then moving rapidly again, then stopping completely
[18:27:28] <codestr0m> cypromis: thanks
[18:27:41] <galen> then i went to the machine and observed that the cursor was *also* freezing during these times
[18:28:13] *** luke has joined #opensolaris
[18:28:15] <galen> I switched kernel SMB server... same results.
[18:28:20] <galen> I'm really mystified what could be causing this
[18:28:21] <codestr0m> galen: iirc there's a bug report out with similar behavior. try searching for that and see if it affects you
[18:28:37] <jbit> galen: do you use ZFS compression?
[18:28:44] <galen> jbit: yes
[18:28:48] <galen> is that a potential cause?
[18:29:03] <jbit> i've heard people complain about similar behaviour when using it
[18:29:45] <jbit> but i have no idea if this is the cause or just coincidence
[18:34:24] <kleppari> I'm also having some very weird behavior
[18:34:37] <kleppari> seems to be related to gtk programs and keyboard input
[18:34:56] <galen> kleppari: with compression? or in general?
[18:35:02] <kleppari> no zfs compression
[18:35:04] <galen> the odd part is this doesn't happen if i'm just "using" the terminal
[18:35:16] <jbit> i generally don't see much point in using filesystem compression for general use, i only use it for volumes containing lots of source code
[18:35:47] <jbit> http://mail.opensolaris.org/pipermail/zfs-discuss/2007-May/040141.html
[18:35:54] <kleppari> i.e., I have pidgin open, and type a few messages quickly
[18:35:55] <galen> well, in theory, it does not add enough cpu overhead to have a negative effect
[18:35:56] *** sah-work has joined #opensolaris
[18:36:03] <kleppari> it hangs for a while
[18:36:11] <galen> well, yeah, gzip
[18:36:22] <galen> that's a lot, lot slower than then default compression
[18:37:02] <galen> i believe their gzip implementation is also not multi-threaded
[18:37:29] *** Disorganized has joined #OpenSolaris
[18:38:27] *** ahe has quit IRC
[18:42:00] *** kleppari has quit IRC
[18:44:05] <holcomb> for awhile the lzjb wasn't multithreaded
[18:44:43] *** The-spiki has quit IRC
[18:44:58] *** The-spiki has joined #opensolaris
[18:45:08] *** galen has quit IRC
[18:45:30] *** fr4g has quit IRC
[18:45:54] *** vmlemon_ has quit IRC
[18:45:58] *** vmlemon_ has joined #opensolaris
[18:47:31] *** galen has joined #opensolaris
[18:47:50] *** galen has quit IRC
[18:52:21] *** galen has joined #opensolaris
[18:52:45] <galen> ok, it seems that my machine freezes periodically while doing *any* IO activity
[18:52:51] <galen> not the hardware, the OS
[18:53:32] <nachox> boot with the kernel debugger enabled and check why it freezes with it
[18:55:46] *** cypromis has quit IRC
[18:55:46] *** galen has quit IRC
[18:57:12] <sponix> why was 102 Yanked ?
[18:57:36] <consanguinity> sponix: a zfs bug could corrupt root pools on some operations
[18:58:13] <sponix> nasty
[18:58:25] *** galen has joined #opensolaris
[18:58:35] <galen> hmm, well, i found another bug (not solaris)
[18:58:52] <galen> but i can say, opensoalaris 2008.11 RC1 freezes periodically whenever doing intensive IO
[18:58:59] <galen> boot disk. zpool, whatever
[18:59:23] <xRaich[o]2x> galen: what hardware?
[18:59:32] <xRaich[o]2x> i have the same problem on a notebook
[18:59:32] <galen> AMD64
[18:59:37] <galen> desktop
[18:59:48] <xRaich[o]2x> diskcontroller?
[18:59:52] <galen> previously had numerous variants of solaris without these issues
[19:00:04] <galen> numerous; nvidia, sil3112, sil3114, etc.
[19:00:04] *** cmihai` has quit IRC
[19:00:14] *** sah-work_ has joined #opensolaris
[19:00:20] *** ipfw has joined #opensolaris
[19:00:49] *** not-me-guv has quit IRC
[19:01:08] <xRaich[o]2x> galen: http://defect.opensolaris.org/bz/show_bug.cgi?id=3690
[19:01:52] *** cmihai` has joined #OpenSolaris
[19:01:58] <galen> great. been open for over a month without a fix.
[19:01:59] <clergyman> http://bugs.mysql.com/bug.php?id=34866
[19:02:03] <galen> "Severity:	critical"
[19:02:03] *** abisen has joined #opensolaris
[19:02:04] <clergyman> Anyone know if there's a fix for that ?
[19:02:21] <xRaich[o]2x> galen: they could not reproduce it
[19:02:30] <galen> grr. i can reproduce it here.
[19:02:43] <galen> i will give a suitable person remote access if necessary to explore the issue.
[19:02:54] <xRaich[o]2x> galen: i opened the bug. it would be good if you could provide some info here
[19:03:21] <galen> xRaich[o]2x: I'm not an expert, i will try something if you want, but i don't know i can fully isolate the core issue
[19:03:31] <xRaich[o]2x> i also opened a thread on the zfs mailing list
[19:03:35] <xRaich[o]2x> wait a sec
[19:03:50] <galen> It seems like solaris is out to get me.
[19:04:02] <galen> I want to have a simple, efficient NAS basically.
[19:04:06] <galen> I have yet to achieve that.
[19:04:18] <galen> I've been running various contortions of solaris for over a year.
[19:04:35] <xRaich[o]2x> galen: http://www.opensolaris.org/jive/thread.jspa?threadID=82726&tstart=0
[19:04:57] <galen> um?
[19:05:31] <xRaich[o]2x> it would be good if you could provide a scanpci output
[19:05:45] <galen> i'm rebooting it.
[19:05:56] <galen> this issue *did not* happen in older installations of SXCE.
[19:06:05] <xRaich[o]2x> yes i know
[19:06:13] <galen> i will run a scanpci output as soon as it finishes rebooting.
[19:06:20] <CosmicDJ> sxce 102? ;)
[19:06:24] <galen> damn thing won't even reboot itself reliably from the gui.
[19:06:28] <galen> 101x
[19:06:31] <galen> *101a
[19:06:52] <xRaich[o]2x> I'm not a developer myself but if you want a bug to get fixed you need to help me to get heard ;)
[19:07:13] <codestr0m> xRaich[o]2x: same bug as yours?
[19:07:19] <xRaich[o]2x> i already talked with a sun developer yesterday and he helped me to provide more info
[19:07:25] <xRaich[o]2x> codestr0m: the symptoms look alike
[19:08:05] *** pat_ has joined #opensolaris
[19:08:05] <xRaich[o]2x> galen: you are the first one that reports the same issue here
[19:08:28] <galen> well, it's agonizing. i hate to say it, but i'm dying for the day zfs shows up on another platform that's not... plagued with issues.
[19:08:48] *** clergyman has quit IRC
[19:09:26] <xRaich[o]2x> rule no.1 if you want something to work, provide info ;) i have the same issue on another box and i've been trying to provide as much information as possible
[19:09:40] <galen> it just powered on, i would have run scanpci sooner otherwise
[19:10:00] <xRaich[o]2x> galen: i know that bug is VERY annoying
[19:10:10] <galen> of potential interest, this issue is partially mitigated by turning off zpool compression
[19:10:38] <galen> it still hangs, but not as bad, not as hard, and not as long
[19:12:35] <galen> gahhh, this is not my day. pastebin.com is down.
[19:12:41] <galen> or my ISP is screwed up, you pick.
[19:12:56] <xRaich[o]2x> galen: post it at the thread i gave you
[19:13:03] <nachox> hang implies an irreversible state save a  harc reboot in my dictionary. it comes back, then it doesnt hang
[19:13:37] <galen> in my book, a freeze is when you have to reboot
[19:13:51] <galen> hanging is... temporary. though if you want to talk kernel, yes, hanging is usually dead
[19:13:54] <xRaich[o]2x> let's not argue about definitions here
[19:14:05] <galen> xRaich[o]2x: working on it, i have to register gahh
[19:14:21] <xRaich[o]2x> the issue is there no matter what you call it
[19:14:49] <galen> more freaking annoying password requiements
[19:15:40] *** sah-work has quit IRC
[19:17:18] <Asako> dang, iscsi kind of makes my host lag
[19:17:40] <xRaich[o]2x> galen: maybe you could provide a ::threadlist -v output yourself
[19:18:10] <galen> does that need to happen while the issue is happening, or just anytime?
[19:18:25] <xRaich[o]2x> when the issue is happening
[19:18:35] <xRaich[o]2x> galen: i reproduced it from a livecd
[19:18:56] <xRaich[o]2x> don't forget to make a snapshot of your rpool/ROOT if you are using opensolaris
[19:19:44] <galen> xRaich[o]2x: huh? why do i need to snapshot that for this purpose?
[19:20:04] <galen> xRaich[o]2x: scanpci output posted
[19:20:20] <xRaich[o]2x> galen: just trying to play safe ;) i messed up the zfs boot when trying to trace the bug down
[19:20:36] *** The-spiki has quit IRC
[19:21:10] *** sponix has quit IRC
[19:21:15] <galen> oh :)
[19:21:22] <xRaich[o]2x> rollback saved the day ^^
[19:21:44] <galen> xRaich[o]2x: I don't know what you mean by doing a threadlist?
[19:21:45] *** sbahra[] is now known as Samy
[19:21:54] <xRaich[o]2x> su to root
[19:22:07] *** madwizard has joined #opensolaris
[19:22:10] <xRaich[o]2x> echo "::threadlist -v" | mdb -k > dump
[19:22:15] *** clergyman has joined #opensolaris
[19:22:40] <xRaich[o]2x> and then attach the created file to your post
[19:24:01] <galen> waiting for the stupid opensolaris site to respond...
[19:25:21] <xRaich[o]2x> galen: if it's the same bug you zfs is waiting for a transaction that is not happening. we don't know if zfs is just the victim of a driver issue
[19:26:04] <galen> i think it's deep, because i see the issue across multiple controllers
[19:26:28] <galen> i am inexplicably unable to access opensolaris.org
[19:26:28] <xRaich[o]2x> galen: if it's the same bug we are one step further
[19:26:51] <xRaich[o]2x> what's happening?
[19:28:12] <galen> this is NOT my day. i think i'm going for a hike shortly.
[19:28:17] <galen> i cannot even ping opensolaris.org
[19:28:22] <xRaich[o]2x> odd
[19:28:39] <xRaich[o]2x> galen: you could reproduce the bug and email me the threadlist
[19:28:45] <xRaich[o]2x> i would post it
[19:29:03] <galen> cannot ping opensolaris.org from my secondary connection either
[19:29:09] <xRaich[o]2x> maybe you could set up a live cd and let me handle it via ssh
[19:29:54] <xRaich[o]2x> just be sure to back up your files ;)
[19:30:53] *** Niosop has joined #opensolaris
[19:32:01] <Niosop> Hello.  I'm trying to start up a Django based non-web script via SMF, but it throws an error:  django.core.exceptions.ImproperlyConfigured: Error loading psycopg2 module: ld.s
[19:32:02] <Niosop> o.1: python2.4: fatal: relocation error: file /opt/coolstack/lib/python2.5/site-
[19:32:02] <Niosop> packages/psycopg2/_psycopg.so: symbol _PyArg_ParseTupleAndKeywords_SizeT: refere
[19:32:02] <Niosop> nced symbol not found
[19:32:08] *** TT has left #opensolaris
[19:32:11] *** Tobbe is now known as Tobbe|autoaway
[19:32:14] *** syamajala has quit IRC
[19:32:20] *** vmlemon_ has quit IRC
[19:32:24] *** vmlemon_ has joined #opensolaris
[19:32:29] <Niosop> Anyone have any pointers on where I could start looking for a solution?  The same service loads fine via command line.
[19:33:22] <Niosop> Only thing I can think of is a difference in enviornment between when it's launched via SMF and via CLI
[19:34:09] <Asako> I love all the linux fanboys talking about Sun on slashdot
[19:34:22] <galen> xRaich[o]2x: http://filebin.ca/dsekkc/threadlist.txt.bz2
[19:34:37] <galen> i would love sun if solaris didn't hate me.
[19:35:23] <Asako> have you tried SXCE?
[19:35:56] <galen> yes, numerous versions of it. i especially love the whole part where it demands static DNS entries and won't get them off DHCP unless you do extensive additional configuration.
[19:36:17] <Asako> hmm
[19:36:26] <Asako> I always use the same resolvers any way :P
[19:36:45] <galen> xRaich[o]2x: opensolaris.org is not available to me on both links again. was able to post.
[19:37:00] <galen> xRaich[o]2x: anything else i can get before I leave for the day? clearly, sun hates me.
[19:37:13] <xRaich[o]2x> galen: you took the threadlist while the bug was happening?
[19:37:18] <galen> xRaich[o]2x: yes
[19:37:26] <xRaich[o]2x> k
[19:37:29] <xRaich[o]2x> looking now
[19:37:32] <galen> i mean, it pauses, then it resumes
[19:37:37] <galen> when it's paused, the window is very short
[19:37:48] <galen> i grabbed it while i was writing data rapdily
[19:38:17] *** luc^ has joined #opensolaris
[19:38:34] *** vertigo- has joined #opensolaris
[19:38:46] *** vertigo- is now known as vertigo
[19:39:02] <galen> i still can't believe how hard it is to get... simply a fully working file server.
[19:39:15] *** Juanito has joined #opensolaris
[19:39:25] <Asako> nah, it's easy
[19:39:33] <Asako> I use Nexenta for that though
[19:40:04] <Niosop> galen, I've had pretty good luck.  Although NFS performance on ZFS blows, gonna have to move everything over to iSCSI me thinks.
[19:40:38] <galen> Niosop: I prefer file level sharing, as it gives me a bit more flexibility with compression and better end to end file integrity
[19:40:52] *** luc^ has quit IRC
[19:41:01] <galen> also, as my friend discovered, there are weird bugs with iSCSI between OS X and Solaris
[19:41:12] <galen> he still can't get his write speeds over 25 MB/sec
[19:41:29] <Niosop> galen, Ahh, yeah, different initiators are so different.
[19:41:43] <galen> well, it's unclear, the initiator can handle good performance on other platforms
[19:41:48] <Niosop> galen, Surprisingly, the Windows one seems to work best, although the built in VirtualBox one works pretty good
[19:42:05] <Asako> linux iscsi sucks too
[19:42:10] <xRaich[o]2x> galen: hmm doesn't look the same
[19:42:16] <Niosop> yeah, loves to kernel panic for me in linux
[19:42:21] <Asako> in centos 5 it kernel panics when you access the disk
[19:42:28] <galen> I've also concluded that Apple *did* fix their NFS client under OS X - or at least the performance issue - but this new I/O issue screws up my file transfers by pausing them every few seconds.
[19:42:33] <Asako> so I switched to another solaris box, hehe
[19:42:46] <galen> Oh, and I also found a new bug in OS X's NFS client. I can reliably kernel panic my OS X box now!!
[19:42:52] <galen> it's great. technology wants to destroy me.
[19:42:55] <Asako> my whole xen solution is built on SXCE
[19:43:21] *** kleppari has joined #opensolaris
[19:43:25] <Asako> running bonnie on my centos vm now
[19:43:47] <Asako> some reason ssh keeps timing out on my file server
[19:44:06] <galen> the frustrating part is that i'm sure i could resolve the OS X client bug with apple engineering involvement, but it would require at least 20 hours of my time
[19:44:08] <galen> It's agonizing
[19:44:20] <Asako> yeah
[19:44:24] <galen> resolve = find out why it's happening then cross my fingers while i wait for them to fix it
[19:44:29] <Asako> I deal with broken systems all day
[19:44:49] *** [RIT]Rawn027 is now known as [RIT]Rawn027|awa
[19:45:14] <Asako> iscsi access shouldn't make my server slow down this much
[19:46:08] <Asako> does it really hog the NIC that much?
[19:46:08] <kimc> i'm going to run more iscsi tests if my build of on current has COMSAT bits
[19:46:23] <galen> is there any way to swap out my kernel under OS 2008.11?
[19:46:36] <galen> i want to fix the problem, i don't care how, I just want to stop wasting time on it.
[19:47:02] <Asako> how can I monitor NIC traffic?
[19:47:19] <xRaich[o]2x> galen: posted your threadlist on the mailing list. maybe someone knows what's going on i doubt that this has something to do with zfs since there is no thread hanging somewhere doing something with zfs as far as i can see
[19:47:23] <galen> there's a % utilization GUI
[19:47:49] <Asako> I'm in ssh
[19:48:02] <Pummel> ssh -X or -Y? :p
[19:48:03] <Asako> would iptraf work in solaris?
[19:48:12] <Asako> x11 forwarding is on
[19:48:18] *** luc^ has joined #opensolaris
[19:48:22] <Pummel> I was kidding
[19:48:49] <Asako> I can ping the box, but it's really unresponsive
[19:48:59] <Asako> Timeout, server not responding.
[19:49:23] *** [RIT]Rawn027|awa is now known as [RIT]Rawn027
[19:49:34] <Asako> bah, I'll have to add a switch
[19:50:37] <Asako> Nov 16 13:49:52 sctest1.liquidweb.com mac: [ID 486395 kern.info] NOTICE: rtls0 link down
[19:50:37] <Asako> Nov 16 13:49:54 sctest1.liquidweb.com mac: [ID 435574 kern.info] NOTICE: rtls0 link up, 100 Mbps, full duplex
[19:50:37] <Asako> Nov 16 13:49:55 sctest1.liquidweb.com mac: [ID 486395 kern.info] NOTICE: rtls0 link down
[19:50:37] <Asako> Nov 16 13:49:57 sctest1.liquidweb.com mac: [ID 435574 kern.info] NOTICE: rtls0 link up, 100 Mbps, full duplex
[19:50:41] <Asako> hah
[19:51:01] <Asako> damn rtls sucks
[19:51:40] <Asako> any significant amount of traffic makes the nic flap
[19:52:36] <xRaich[o]2x> galen: what process was hanging while you where making the threadlist?
[19:52:42] <galen> I do not know
[19:52:44] <galen> everything hangs
[19:52:46] <galen> down to the cursor
[19:52:51] <galen> well
[19:52:54] <xRaich[o]2x> hrm
[19:52:54] <galen> sometimes
[19:53:00] <galen> sometimes, it's just the i/o that hangs
[19:53:21] <xRaich[o]2x> on my box it's just the io that hangs
[19:53:44] <galen> well, what causes cursor hangs is intensive I/O to a raidz2 with the default lightweight compression on
[19:53:51] <galen> turning it off reduces/eliminates cursor freezes
[19:54:22] <xRaich[o]2x> how much ram?
[19:59:35] *** CardFire has quit IRC
[19:59:43] *** CardFire has joined #opensolaris
[20:02:26] <Asako> I've had to replace the NIC on every box that has a realtek installed
[20:03:19] *** CardFire_ has joined #opensolaris
[20:04:39] <Asako> libvirt.libvirtError: virDomainCreate() failed POST operation failed: (xend.err "Boot loader didn't return any data!")
[20:04:44] <Asako> hmm, that's not good
[20:08:23] *** [RIT]Rawn027 is now known as [RIT]Rawn027|awa
[20:11:37] <Asako> now the disk is hosed
[20:14:50] *** [RIT]Rawn027|awa is now known as [RIT]Rawn027
[20:14:57] *** vmlemon_ has quit IRC
[20:15:01] *** vmlemon_ has joined #opensolaris
[20:16:10] *** [RIT]Rawn027 has quit IRC
[20:17:24] *** comay has joined #opensolaris
[20:18:26] *** CardFire has quit IRC
[20:23:02] *** ChanServ sets mode: +o comay
[20:24:12] *** abisen has quit IRC
[20:24:19] <Doc> you should try not to get your disks wet, so hosing them probably isnt a good idea
[20:24:42] <clergyman> In /usr/mysql/bin/mysql_config --cflags there are quite a few options that neither suncc or g++ is willing to take.. what compiler is used for SUNWmysql5 ?
[20:24:45] <sickness> what means "yanked"?
[20:24:59] *** luna1 has joined #opensolaris
[20:25:02] <clergyman> sickness: pulled
[20:27:44] *** aksyn has joined #opensolaris
[20:31:13] <codestr0m> clergyman: which version of sun cc. the options change between the express version which comes with os2008* and that of SS12 which is probably what whoever packaged it used
[20:31:30] <clergyman> Ah, I'm using os 200811
[20:31:41] <clergyman> It still compiles, but it whinges a bit
[20:32:16] <clergyman> Can I safely assume that it'll compile fine with SS12 if it compiles in os2008?
[20:33:19] *** CardFire_ has quit IRC
[20:34:27] <kimc> just started an ON build with b102
[20:34:57] <kimc> b102 has now has 'top'
[20:35:31] <holcomb> oh boy.
[20:35:42] <kimc> 95 processes: 85 sleeping, 7 running, 1 zombie, 2 on cpu
[20:37:50] <Asako> so do I need to create a disk slice before an xen domain can see the drive?
[20:38:01] <Asako> if I create a zpool on the drive and then destroy it I'm fine
[20:42:51] *** kim0 has joined #opensolaris
[20:47:46] <kimc> looks like maybe On won't build right now..
[20:48:03] *** comay has quit IRC
[20:49:27] *** galen has quit IRC
[20:49:50] <h3sp4wn> Asako: If its a seperate controller then you should be able to do pci passthrough otherwise just create a zpool then a single zvol
[20:50:18] *** galen has joined #opensolaris
[20:51:38] <Asako> it's an iscsi target
[20:51:48] <Asako> I just want to use the raw disk
[20:53:08] <Asako> putting a zvol on top of a zvol seems kind of dumb
[20:54:42] <h3sp4wn> so just use ufs
[20:55:06] *** dustman has quit IRC
[20:55:12] <Asako> it's for a xen instance
[20:55:30] <Asako> I got it to work, just wondering if there's a better way
[20:57:15] <h3sp4wn> [samw@titan] (~)% svcs -x
[20:57:15] <h3sp4wn> svc:/system/filesystem/zfs/auto-snapshot:daily (ZFS automatic snapshots)
[20:57:38] <h3sp4wn> Asako: On a different machine right ?
[20:58:02] <Asako> yeah, the xen host accesses it through iscsi
[20:58:15] <Asako> the actual disk is on my AVS cluster
[20:59:34] <Asako> libvirt.libvirtError: virDomainCreate() failed POST operation failed: (xend.err "(2, 'Invalid kernel', 'xc_dom_find_loader: no loader found\\n')")
[20:59:41] <Asako> I just finished the install
[21:00:04] <Asako> where does xvm store the actual config files any way?
[21:00:38] <h3sp4wn> somewhere in either /var or /lib I forget but fortunately you can get the actual config with - xm list -l domain
[21:01:09] *** gonzzor has left #opensolaris
[21:01:23] <h3sp4wn> (I did know but don't bother anymore because you can edit that file then undefine and then do - xm new -F foo.sxp
[21:01:49] <Asako> yeah
[21:01:59] <Asako> I'm just used to creating a file and using xm create
[21:02:19] <h3sp4wn> There is some stuff virsh doesn't do anyway
[21:03:47] <h3sp4wn> (changing on_xend_start and adding the driver paravirtualised part) unless there is something I still don't get
[21:04:01] *** abisen has joined #opensolaris
[21:04:18] <Asako> seems like it's not finding the kernel
[21:04:21] <h3sp4wn> The driver paravirtualised part likes to disappear randomly as well
[21:04:22] <Asako> grub comes up
[21:04:50] <Asako> how do I actually delete a vm with virsh?
[21:04:57] <h3sp4wn> virsh undefine
[21:05:07] <Asako> thanks
[21:05:11] <h3sp4wn> well it doesn't destroy the disks etc
[21:05:20] <Asako> that's ok, I'll reformat it
[21:06:12] *** ninjaslim has joined #opensolaris
[21:07:13] *** dvz has joined #opensolaris
[21:10:31] <Asako> virt-manager seems to work better
[21:10:45] *** Niosop has quit IRC
[21:11:13] *** sactodave has joined #opensolaris
[21:11:49] *** axisys has quit IRC
[21:12:15] *** lixx has quit IRC
[21:12:35] *** MeP3aBeu has quit IRC
[21:15:04] *** osladil has joined #opensolaris
[21:15:22] <clergyman> Can I somewhere change the URL for a repository? I'm using sunfreeware.com.. and I think I'm using the "main" URL, and I'd like to use an UK mirror
[21:15:41] <h3sp4wn> Asako: I think that is a horrible app I would rather have to work around bugs in xm / virsh than use that crap
[21:18:03] <Asako> maybe, at least it finds the right install media
[21:18:39] *** clyons_ has joined #opensolaris
[21:19:10] <Asako> I can see some benefit to virsh but I'd rather just write my own configs
[21:19:34] <h3sp4wn> There is nothing stopping you doing that
[21:19:47] <Asako> I gotta find where they're stored
[21:21:00] <h3sp4wn> why ? if you can define and undefine one from an sxp I don't see what point there is to that
[21:21:14] <h3sp4wn> (Other than you have to stop xend or its overwritten by the old one)
[21:22:36] <Asako> guess that would be best, trying to automate all of this
[21:23:14] <Asako> create a kickstart config under /tmp, run through the install automatically
[21:24:34] *** Tobbe|autoaway is now known as Tobbe
[21:35:18] *** The-spiki has joined #opensolaris
[21:36:28] *** sah-work has joined #opensolaris
[21:38:40] *** osladil has quit IRC
[21:38:57] *** dustman has joined #opensolaris
[21:42:29] *** axisys has joined #opensolaris
[21:43:16] <axisys> how do I use preload libumem on a running process? i need to check memory leaks
[21:43:42] <axisys> http://sunsolve.sun.com/search/document.do?assetkey=1-65-208721-1 shows how to call .. but not PID
[21:45:05] *** scrash08 has joined #opensolaris
[21:45:07] <jbk> umm
[21:45:12] <jbk> you're talking in contradictions
[21:45:27] <jbk> 'preload' something implies its loaded before it's running
[21:45:45] <jbk> you have to restart the process
[21:46:02] <jbk> with LD_PRELOAD=libumem.so set
[21:46:10] <Discordian> indeed
[21:46:20] <jbk> and then you can set the umem flags
[21:46:26] <scrash08> Hi.  I understood that no text installer (for headless installs) was available in opensolaris 2008.5 release.  is there one yet in the 2008.11 dev builds?
[21:47:42] *** sah-work_ has quit IRC
[21:47:47] *** sah-work_ has joined #opensolaris
[21:48:05] <axisys> jbk: let me then rephrase my question
[21:48:24] <axisys> jbk: is there a way to find memory leaks on a running process using libumem?
[21:48:46] *** sah-work has quit IRC
[21:48:47] <CosmicDJ> axisys: was it linked/started with libumen?
[21:49:02] <Discordian> Look at the process footprint over time
[21:49:32] <jbk> if pldd pid doesn't show libumem loaded, then no
[21:49:34] <axisys> CosmicDJ: no
[21:49:43] <CosmicDJ> axisys: then you can't
[21:49:48] <Discordian> Then use dtrace?
[21:49:59] *** diggs-ita has quit IRC
[21:50:37] <axisys> Discordian: i know it leaks.. prstat -s vsz shows it increases over time ..
[21:51:02] <axisys> Discordian: so dtrace could help find function is the culprit ?
[21:51:10] <axisys> CosmicDJ: thnx
[21:51:31] <CosmicDJ> axisys: well, kill it and restart it with libumem :)
[21:51:38] <Discordian> Yes dtrace might help
[21:51:50] <axisys> prtstat -s size that is
[21:51:58] <axisys> CosmicDJ: heh
[21:52:01] <jbk> is it an bundled binary or 3rd party?
[21:52:09] *** Juanito_ has joined #opensolaris
[21:52:09] <axisys> just leaked one meg again
[21:52:21] <clergyman> a meg? Sheesh
[21:52:22] <axisys> jbk: netcool/nervcenter app
[21:52:36] <CosmicDJ> and you're sure it's not just doing... stuff?
[21:53:45] <axisys> CosmicDJ: i was asked to kill it when it went up from 151M to 3G .. not sure about the time diff
[21:54:01] <axisys> CosmicDJ: now it is back to 151M at restart
[21:54:32] <Discordian> It's java?
[21:54:58] <Asako> this is weird, I can't make http requests from my vm
[21:55:21] <axisys> Discordian: C
[21:55:43] <CosmicDJ> axisys: http://wikis.sun.com/display/SunStudio/Memory+Checking+Tools
[21:55:51] <Discordian> set ulimits accordingly
[21:56:15] <jbk> then before you restart it, set LD_PRELOAD and the UMEM_DEBUG flags
[21:56:36] <jbk> also, can you do a clean shutdown?
[21:56:37] <jbk> if so
[21:56:47] <jbk> after you restart it
[21:56:58] <Discordian> I do hope thisn't productiom
[21:57:08] <Discordian> s/m/n
[21:57:16] <jbk> the next time it grows, what i'd suggest is a small dtrace script that triggers on a call to exit(), have it pause hte process and run gcore
[21:57:30] <jbk> that might be more accurate
[21:57:56] <jbk> but even if not, just be aware it'll report on all allocations that haven't been freed
[21:58:15] <axisys> jbk: i found a dtrace (after googling) script http://blogs.sun.com/sanjeevb/ that is suppose to find leaks
[21:58:19] *** syamajala has joined #opensolaris
[21:58:22] *** cmihai` has quit IRC
[21:58:39] <axisys> CosmicDJ: never used valgrind
[21:59:13] *** cmihai` has joined #OpenSolaris
[21:59:16] <Asako> now I'm getting annoyed
[21:59:18] *** scrash08 has left #opensolaris
[21:59:36] <Asako> I can ping the box, but it can't wget anything and I can't ssh to it
[22:00:15] <Discordian> Then you're networking isn't setup right
[22:00:19] <Discordian> your
[22:00:20] <CosmicDJ> Asako: portscan?
[22:00:31] <Asako> yum works, which is weird
[22:00:33] <CosmicDJ> Asako: are you sure there is smth listening in the ports?
[22:00:49] <Discordian> nmap it
[22:00:54] <Asako> it's a centos 4 vm
[22:01:00] <Discordian> nmap it
[22:01:05] <Asako> ok
[22:01:16] <Asako> PORT    STATE SERVICE
[22:01:16] <Asako> 22/tcp  open  ssh
[22:01:38] <CosmicDJ> Asako: and you tried to ssh in as... ?
[22:01:46] <Asako> root
[22:01:51] <Asako> ssh -v just sits
[22:01:55] <CosmicDJ> Asako: could be disabled
[22:01:57] <Discordian> So conclusion is it's networking is wrong
[22:02:05] <CosmicDJ> PermitRootLogin...
[22:02:05] <Asako> debug1: Connecting to 72.52.245.30 [72.52.245.30] port 22.
[22:02:06] <Asako> debug1: Connection established.
[22:02:06] <Asako> debug1: identity file /export/home/wattersm/.ssh/identity type -1
[22:02:07] <Asako> etc.
[22:02:14] <Discordian> if it sits might be dns
[22:03:13] <CosmicDJ> "debug1: Connection established." that doesn't sound like you can't connect ;)
[22:03:15] <Asako> no idea, I've done this hundreds of times
[22:03:34] <Asako> just never ran xen on solaris before
[22:04:26] *** cky has quit IRC
[22:04:29] <Asako> oh yeah, it's something to do with ipv6
[22:04:42] <Asako> Did not receive identification string from ::ffff:69.16.222.240
[22:05:19] *** sah-work_ has quit IRC
[22:06:36] *** cky has joined #opensolaris
[22:07:25] *** nrubsig has joined #opensolaris
[22:07:25] *** ChanServ sets mode: +o nrubsig
[22:07:37] <nrubsig> alanc: yt ?
[22:07:57] *** Juanito has quit IRC
[22:08:02] *** Juanito__ has joined #opensolaris
[22:08:07] *** Zplay has joined #opensolaris
[22:08:55] *** Juanito_ has quit IRC
[22:09:17] <Asako> write(2, "HTTP request sent, awaiting resp"..., 40HTTP request sent, awaiting response... ) = 40
[22:09:17] <Asako> select(4, [3], NULL, NULL, {900, 0}
[22:09:20] <Asako> and there it sits
[22:09:56] *** nrubsig has quit IRC
[22:12:20] *** vertigo is now known as vertigo-
[22:13:23] <Asako> it makes no sense
[22:13:37] <consanguinity> Asako: i've seen this behaviour when a disk fails
[22:13:48] <Asako> (xVM) mm.c:625:d8 Non-privileged (8) attempt to map I/O space 000000e0
[22:13:52] <Asako> like that?
[22:14:22] *** The-spiki has quit IRC
[22:14:30] <Asako> that's just xm dmesg
[22:15:28] <Asako> doubt the disk failed, I just installed on it
[22:16:24] *** kleppari has quit IRC
[22:17:27] *** ggeecko has joined #opensolaris
[22:18:59] *** swankier has quit IRC
[22:19:27] <Asako> how do I tell which disk device goes to which iscsi target?
[22:19:34] *** swankier has joined #opensolaris
[22:20:21] <Asako> found it
[22:23:13] *** kleppari has joined #opensolaris
[22:23:23] * Asako creates a windows machine
[22:23:55] <Asako> hah, it didn't find any hard drives
[22:25:15] <Asako> sigh, I <3 linux
[22:25:25] <Asako> EXT3-fs error (device dm-0) in start_transaction: Journal has aborted
[22:26:21] <Asako> no wonder I can't log in
[22:26:35] <throwt> i see those often enough i thought they're normla
[22:28:43] <Asako> me too, almost
[22:31:29] <Asako> fixing them is part of my job
[22:31:49] <CosmicDJ> Asako: you're working sunday evening?
[22:31:55] <Asako> yup
[22:32:25] <Asako> it's quiet
[22:32:34] <Asako> I can actually get stuff done
[22:32:36] <CosmicDJ> get yourself a girlfried ;)
[22:32:54] <CosmicDJ> then it won't be quiet anymore
[22:33:23] <Asako> yeah, probably
[22:34:51] <Asako> no big deal, been this way for years
[22:35:48] *** mikl has quit IRC
[22:41:41] <Asako> it's cool that creating a zpool on top of a zvol doesn't confuse the system
[22:44:13] <jbk> heh.. i'm glad i don't have to deal with ext3 crappiness anymore :)
[22:45:24] <Asako> we'd have to quit using centos first
[22:45:24] * jbk got sick of stuff like that causing outages all the time
[22:45:47] <Asako> your server is down?  hold one while I fsck a few terabytes
[22:45:47] <jbk> and expecting me to somehow fix the problem (in the sense of preventing it from happening again)
[22:46:08] <Asako> prevention is easy, use a better FS
[22:46:26] <jbk> redhat doesn't support other FSes :)
[22:46:31] <Asako> or disable journaling, fat lot of good it does any way
[22:46:43] <jbk> and these people were too cheap to buy veritas
[22:46:56] <Asako> I know redhat doesn't, which is dumb
[22:47:04] <jbk> which was amusing
[22:47:09] <Asako> they should offer xfs or jfs
[22:47:17] <jbk> they were writing a trading platform
[22:47:25] <delewis> XFS and JFS both have issues.
[22:47:30] <jbk> and would nickel and dime their hardware
[22:47:40] <Asako> same here, we've always been cheap
[22:47:51] <delewis> JFS is largely unmaintained these days and XFS has some rather serious data integrity problems.
[22:47:56] <hali> Vertias Foundation Basic is even free
[22:48:02] <hali> so easy to test
[22:48:06] <Asako> so I'm trying to build a cloud computing cluster on top of solaris
[22:48:10] <jbk> well this was at a very large bank (that hasn't taken nearly as many writedowns as its competitors)
[22:48:18] <e^ipi> oh fuck off... "cloud computing"
[22:48:21] <Asako> lol
[22:48:22] <e^ipi> it's a cluster...
[22:48:29] <Asako> I know
[22:48:39] <Asako> I just do what they ask
[22:48:42] <consanguinity> e^ipi: a "cloud" is a cluster built with linux: it's more likely to disintegrate...
[22:48:44] <e^ipi> you should synergize
[22:48:49] <jbk> haha
[22:48:56] <Asako> lol
[22:49:23] <jbk> of course i still question how many things really work well in a 'cloud'
[22:49:30] <Asako> at least with virsh I can script most of the build stuff
[22:49:38] <delewis> jbk: just because they haven't taken writedowns doesn't mean they won't. The problem with this crisis was that banks didn't writedown when they should've.
[22:49:43] <Asako> web servers
[22:49:51] <Asako> instant load balancing
[22:49:55] <jbk> delewis: but the point was, they weren't hurting for cash
[22:50:10] <delewis> jbk: until the depositors get scared and run the bank.
[22:50:15] <Asako> was this Wells Fargo?
[22:50:26] <jbk> and in fact, the bonus any one of the developers could have paid for licenses for all their machines
[22:50:31] <jbk> no
[22:50:51] <jbk> they were just being cheap
[22:50:56] <Asako> typical
[22:51:13] <Asako> I keep telling one of our customers they need a file server, nope, won't do it
[22:51:25] <jbk> but these were supposedly geniuses from goldman sachs (they jumped ship to where i was)
[22:51:38] <jbk> i still have to wonder if gs wasn't secretly thrilled to get rid of them
[22:51:41] *** hsp has quit IRC
[22:52:02] *** TT has joined #opensolaris
[22:52:02] <delewis> we've been interviewing a load of LEH candidates lately for a position.
[22:52:20] <delewis> sigh.
[22:52:23] *** mikefut has joined #opensolaris
[22:52:34] <delewis> guys that applying for a sysadmin position and can't tell you what information is stored in an inode.
[22:52:42] <jbk> haha
[22:53:16] <delewis> it's a security-related sysadmin position, and some of them can't even tell you the fundamental difference between telnet/rlogin and SSH.
[22:53:21] <hali> fuck that, i spend 50% of my time _supporting_ sysadmins in banks who are suposed to manage the applications we sell them
[22:53:26] <jbk> not surprised
[22:53:28] <Asako> it's hard to remember what's stored there
[22:53:33] <delewis> the ones that do know what SSH is, always think that it requires keys for authentication.
[22:53:37] <Asako> but I can google it
[22:53:44] <hali> references to blocks or referecences to other inodes
[22:53:57] <delewis> Asako: you've clearly never used stat().
[22:54:05] <hali> more in to scat
[22:54:06] <Asako> I have in linux
[22:54:23] <Asako> I know the diff between sym links and hard links
[22:54:24] <codestr0m> delewis: did you place an ad in a local paper or a tech website? seems odd to get responses from candidates that clueless
[22:54:28] <delewis> so you should know what data stat returns and where stat gets that information from.
[22:54:28] <jbk> i was kinda surprised
[22:54:38] <delewis> codestr0m: of course we did.
[22:54:42] <jbk> i got a cold call about a presales engineer position in a large software company
[22:54:45] <Asako> yeah, I've had to debug a few file systems before
[22:54:50] <jbk> i wouldn't think places would be hiring really..
[22:55:03] <Asako> debugfs is kind of cool
[22:55:11] <delewis> lots of places are hiring
[22:55:16] <delewis> just not in the financial sector.
[22:55:24] <delewis> healthcare and human staples are doing some hiring.
[22:55:28] <Asako> we've slowed down but still growing
[22:55:32] <delewis> (WalMart, specifically)
[22:55:51] <jbk> so i'm kinda conflicted.. the place i'm at now is really nice
[22:55:53] *** sponix2ipfw has joined #opensolaris
[22:56:04] <jbk> but i'm wondering how much longer i really want to do the sysadmin thing..
[22:56:09] *** ipfw has quit IRC
[22:56:23] <Asako> know the feeling
[22:56:34] <Asako> I ended up staying here, got too much time invested to leave
[22:56:38] <clergyman> I know what's stored in an inode, but I can't for the life of me get my keyboard to behave properly lol
[22:56:51] <clergyman> Maybe I started this whole computer thing in the wrong end :)
[22:56:54] <Asako> speaking of keyboard, how do I make backspace work properly?
[22:56:58] <Asako> and my home/end keys
[22:57:07] <delewis> stty(1M)
[22:57:08] <clergyman> My backspace is working, not delete though
[22:57:26] <Asako> ok, now to see if this file system corrupts itself
[22:57:44] <clergyman> Asako: http://rafb.net/p/2LrHIN61.html
[22:57:51] <clergyman> That sorted my home/end key problem
[22:57:59] <Asako> thanks
[22:58:29] <clergyman> Np.. I've currently got my delete key acting as backspace.. and in vim neither backspace OR delete does anything but moving the cursor heh
[22:58:31] *** bahumbug has joined #opensolaris
[22:58:41] <clergyman> Shouldn't be rocket science :/
[22:59:36] <Asako> I'm sick of my shell showing ^H
[22:59:46] <delewis> so use stty(1M)
[22:59:51] <Asako> I will, hehe
[23:00:30] <clergyman> delewis: What would stty be for delete? 'erase' only applies for the character to the left of the cursor
[23:01:08] *** Fish has quit IRC
[23:01:50] <clergyman> Whatever compatibility setting I set in gnome-terminal, it either prints ^H or ~ or acts as backspace
[23:02:54] <codestr0m> clergyman: I have a blog entry on this. let me grab the link as it could possibly help
[23:03:22] <Asako> argh
[23:03:30] <Asako> I can telnet right to port 22 but ssh doesn't work
[23:03:42] <clergyman> codestr0m: I think I've stubmbled upon your blog at some point
[23:03:48] <codestr0m> http://www.codestrom.com/wandering/2008/09/fix-stty-xterm.html
[23:04:46] * Asako waits for journal to abort
[23:05:00] <Asako> maybe we should just run zones, lol
[23:05:23] <Asako> linux zones are pretty cool
[23:05:36] *** _coredump_ has joined #opensolaris
[23:05:45] *** syamajala has quit IRC
[23:05:52] <Asako> that's an experiment for tomorrow I guess
[23:06:08] *** pumpkin_ has joined #opensolaris
[23:06:10] <clergyman> codestr0m: Ah yeah, that seems to fix it if I start an xterm.. Will I need to restart gnome-terminal for it to take effect here as well ?
[23:06:32] <codestr0m> clergyman: probably. I think it should put you on the right track
[23:07:18] <clergyman> Yaay :) If you set "Escape sequence" in compatibility mode in gnome-terminal it works there as well
[23:07:22] <clergyman> Thanks a bunch!
[23:07:39] <clergyman> And that fixed my vim issue as well
[23:08:01] <codestr0m> clergyman: my entry on it was helpful?
[23:08:24] <codestr0m> or just figured it out
[23:08:26] <clergyman> Or hm.. did it
[23:08:41] <clergyman> Let me restart gnome-terminal.. some tabs behave, some don't o.O
[23:08:44] *** clergyman has quit IRC
[23:09:32] <kleppari> have you experienced..  hmm..  choppy multitasking?
[23:09:42] *** Gman has joined #opensolaris
[23:09:42] *** ChanServ sets mode: +o Gman
[23:09:51] *** clergyman has joined #opensolaris
[23:10:07] <Asako> kleppari, not much, but this is a dual core
[23:10:10] <clergyman> I am such a tool.. only set the inputrc for one user.. :) Of course it won't work for other users heh
[23:10:26] <kleppari> Asako: yeah, I have a quadcore and 4G RAM
[23:10:32] <kleppari> and I experience it alot
[23:10:34] <clergyman> codestr0m: Yes, all sorted using your blog entry :)
[23:11:04] <Asako> only thing I notice is text printing on a terminal uses a lot of cpu
[23:11:08] <kleppari> i.e., switch workspace to firefox, hit C-t to open a new tab, type in a new address quickly and firefox seems to hang for a few seconds
[23:11:19] <Asako> so if I have a big rsync running, X11 gets choppy
[23:11:52] <clergyman> Asako: May be if your graphics card isn't accelerated
[23:11:55] <kleppari> the desktop is responsive(i.e. I can keep switching workspaces etc), but all gnome/gtk programs seem to hang for a few seconds
[23:11:56] <codestr0m> clergyman: good to hear it
[23:11:58] <clergyman> It'll have to render X using CPU
[23:12:19] <Asako> it's nvidia
[23:12:36] <Asako> doesn't matter what card I use, it's just X
[23:13:36] *** pat_ has quit IRC
[23:15:48] *** ninjaslim has quit IRC
[23:16:59] <kleppari> hm, it seems to be related to keyboard input
[23:17:02] <kimc> why was sxce 102 yanked ?
[23:17:39] <e^ipi> because of a bug.
[23:17:47] <sponix2ipfw> kimc:  I asked that earlier, it was a zfs bug that could cause corruption with root pools (rpool) when certain features where used
[23:17:55] <TT> has zfs support of jbod?
[23:17:58] *** Juanito__ has quit IRC
[23:18:07] <e^ipi> TT: malformed question.
[23:18:17] <kimc> ahh see
[23:18:35] *** sponix2ipfw is now known as sponix
[23:18:54] <TT> k
[23:19:22] *** ninjaslim has joined #opensolaris
[23:19:41] <jbk> ?SYNTAX ERROR
[23:19:43] <jbk> :)
[23:26:11] <throwt> im waiting...patiently for s2d, s2r laptop support
[23:26:24] *** Tobbe is now known as Tobbe|autoaway
[23:26:32] <TT> Can I create a zpool where I have 100% disk capacity and 0% redundance and if one device has a failure that the data on the other devices is correct?
[23:26:49] *** syamajala has joined #opensolaris
[23:26:50] <Asako> no
[23:27:01] <e^ipi> TT: no, everything's a stripe
[23:27:14] <TT> hm thats not cool
[23:27:21] <TT> =/
[23:27:28] <e^ipi> why, just use redundancy like you should be doing
[23:27:42] <consanguinity> actually, i thought zfs already did that
[23:27:44] <TT> dont have enogh disks
[23:27:49] <consanguinity> since metadata is replicated even on a non-redundant configuration
[23:28:25] <TT> its possible to create various zpools with on mount point?
[23:31:18] *** piwi has joined #opensolaris
[23:31:45] <TomJ> you can change the mountpoint of any zfs filesystem to be whatever you want
[23:32:10] <TomJ> so if you have zpool1 and zpool2 they will be /zpool1 and /zpool2  but   zpool1/data could be  /usr/local/data and zpool2/user could be /usr/local/data/user or whatever
[23:32:39] <TomJ> This can get confusing with regard to property inheritance though, so don't go overboard with interspersing them
[23:34:14] <sickness> I'm back
[23:34:34] <throwt> bl
[23:35:56] *** jwit has quit IRC
[23:37:24] *** pumpkin_ is now known as pumpkin
[23:37:57] *** tombhadAC has joined #opensolaris
[23:39:18] *** Teo` has joined #opensolaris
[23:40:34] *** scrash08 has joined #opensolaris
[23:40:39] <Asako> there has to be something weird that xen is doing
[23:40:53] <Asako> there's no reason I shouldn't be able to download files
[23:41:57] <scrash08> Can 'gui-install' be launched remotely via "ssh -X tunnel"?  I'm trying, but stuck at "cannot open display" messages.  Dunno if it's me, or gui-install, etm.
[23:43:20] <TT> so i could create a zpool with various mirrors and in every mirror is a dummy disk like an file? so if one mirror failed other mirrors are correct?
[23:43:48] <sickness> scrash08: there's a blog post about that, the environment, or the ssh binary on the installer live image, lacks the X-auth forwarding capability
[23:44:05] <sickness> scrash08: so you'd need to put a binary from an installed system or something like that...
[23:44:15] <sickness> scrash08: then it can work
[23:45:59] <scrash08> sickness: Hi. New to opensolaris ... I DL'd the 2008.11 latest iso ... is that the "live image"?  Atm, it's mounted in a Xen DomU ... I can ssh/login witho no trouble, just no renote app launch.
[23:46:08] *** TT has quit IRC
[23:46:25] <Asako> you can just boot the cd
[23:47:50] <sickness> scrash08: yeah that's the live iso
[23:48:05] <scrash08> Asako: SUre.  Shouldn't make a difference re: ability to  "ssh -X" launch remote apps, though ... or?
[23:48:26] <scrash08> sickness: aha ... investigating the X-auth forwarding there.  Thanks for the hint.
[23:48:30] *** jgracin has quit IRC
[23:49:02] *** elektronkind has quit IRC
[23:49:34] *** AxeZ has quit IRC
[23:52:40] <kleppari> is the desktop discussion forum moderated on opensolaris.org?
[23:54:54] <scrash08> sickness: I always get the 'direction' in tunneling bass-ackwards :-/   Is it the "ssh binary", i.e., the client, on the opensolaris installer that needs to have the Forwarding Agent enabled?
[23:55:23] <sickness> scrash08: yeah, I think it's that, lemme scavenge the blog post about that issue..
[23:55:44] <scrash08> sickness: Appreciated ... I'm wallowing in google results ...
[23:58:03] *** Asako has quit IRC
[23:58:45] <CIA-34> Brent Paulson <Brent.Paulson at Sun dot COM>: 6738808 c2audit routines set{u,s}mask can reference processes they shouldn't





top