[00:01:36] *** yarihm_ has quit IRC
[00:02:04] <wesolows> I probably can't say
[00:02:28] <wesolows> the driver is closed source
[00:02:39] <wesolows> consider that tale of woe #1
[00:02:48] *** Megaf has quit IRC
[00:05:23] <dlg> i was going to try and get one and write a driver for it
[00:08:49] *** plocher has quit IRC
[00:08:50] <jbk> evening
[00:11:00] *** alanbur has joined #opensolaris
[00:11:31] *** cmihai has joined #OpenSolaris
[00:13:11] *** ashner has joined #opensolaris
[00:14:02] *** bob_johnson has left #opensolaris
[00:17:10] *** dpn` has quit IRC
[00:17:30] * stevel throws a barrel of OpenID docs at alanbur
[00:17:59] <wesolows> the unit of documentation is now barrel instead of grey wall?
[00:18:09] <stevel> sorry - been thinking about kegs all afternoon
[00:18:12] <stevel> i've got barrels on my mind
[00:18:13] * alanbur stuffs stevel's head down the toilet and pulls the handle.  Repeatedly
[00:18:42] * stevel rips said toilet out of the floor and drops it on alanbur's head
[00:19:01] <alanbur> Just coz he escaped the asylum, he thinks he can taunt the other inmates...
[00:19:15] * nrubsig prepares a BOSSLER 40mm frontloader shotgun to bring peace to stevel && alanbur
[00:19:22] <stevel> the exit interview didn't say anything about not doing that
[00:19:48] <wesolows> no but it is ungentlemanly
[00:19:49] <alanbur> You just aren't strong enough - the guy in the film was a *big* bugger...
[00:20:02] <stevel> wesolows: you're right... i'm forgetting the gentlemanly rules of #opensolaris
[00:20:04] * nrubsig starts lifting the bossler thing and aims in alanbur's direction...
[00:20:07] <stevel> wesolows: wait. no. there aren't any
[00:20:11] <wesolows> oh, right
[00:20:12] <wesolows> carry on then
[00:20:18] <nrubsig> BAAAM
[00:20:23] <nrubsig> wesolows: done.
[00:20:30] * alanbur doesn't even bother to duck.
[00:20:37] *** dsch04_ has joined #opensolaris
[00:20:43] * stevel ties alanbur's shoelaces together
[00:20:57] <dsch04_> wesolows, sorry - disconnected, and my nick got locked
[00:21:11] *** _mw46_ has quit IRC
[00:21:15] <alanbur> Yeah figures, after all that practice you should be good at it...
[00:21:24] <stevel> ;-)
[00:21:25] *** Marv|LG has quit IRC
[00:22:05] <alanbur> That 'edit your ssh keys' form on oso kinda sucks...
[00:22:35] <stevel> kinda?
[00:22:40] <stevel> :-P
[00:22:43] *** dsch04_ has quit IRC
[00:22:46] <stevel> the webapp engineers should clearly be shot
[00:22:46] <alanbur> well, I was being kind ;-)
[00:23:04] * alanbur grabs bug gun thing and starts loading
[00:23:28] <alanbur> bug? somehow still oh so appropriate...
[00:23:29] <stevel> such love between the two former colleagues ;-)
[00:23:49] <jbk> :)
[00:23:59] *** _mw46_ has joined #opensolaris
[00:24:06] <alanbur> getting a textfield to wrap a continuous string isn't possible afaik
[00:24:38] *** jhawk has joined #opensolaris
[00:24:50] *** dsch04_ has joined #opensolaris
[00:25:09] <FireflyST> hmm
[00:25:14] <alanbur> I wonder how best to do it...
[00:26:12] <dsch04_> So, where was I?
[00:26:28] *** stevel has left #opensolaris
[00:26:29] <FireflyST> ok so, what does this mean when I run uvccapture?
[00:26:31] <FireflyST> Taking snapshot every 0 seconds
[00:26:31] <FireflyST> Taking images using mmap
[00:26:31] <FireflyST> Resetting camera settings
[00:26:32] <FireflyST>   control unsupported by this device
[00:26:34] <FireflyST>   control unsupported by this device
[00:26:36] <FireflyST>   control unsupported by this device
[00:26:39] <FireflyST>   control unsupported by this device
[00:26:40] *** stevel has joined #opensolaris
[00:26:40] *** ChanServ sets mode: +o stevel
[00:26:41] <dsch04_> Ah yes - 5 disks on a SAT2-MV8 (Marvell chipset)
[00:26:41] <FireflyST>   control unsupported by this device
[00:26:44] <FireflyST>   control unsupported by this device
[00:26:47] <FireflyST>   control unsupported by this device
[00:26:50] <FireflyST> Camera brightness level is -1
[00:26:52] <FireflyST>   control unsupported by this device
[00:26:54] <stevel> bah. cmd-w in the wrong window
[00:26:55] <FireflyST> Camera contrast level is -1
[00:26:55] <tenex> FireflyST could you stop pasting that
[00:26:56] <claws_bugs> fire
[00:26:57] <FireflyST>   control unsupported by this device
[00:27:00] <FireflyST> Camera saturation level is -1
[00:27:02] <claws_bugs> can you paste that few more times
[00:27:02] <FireflyST>   control unsupported by this device
[00:27:05] <FireflyST> saving to snap.jpg
[00:27:08] <claws_bugs> im not sure what u said
[00:27:17] <FireflyST> sorry
[00:27:18] <flyingparchment> FireflyST: http://rafb.net/paste/
[00:27:18] <stevel> claws_bugs: something about a control being unsupported i think ;-)
[00:27:23] <tenex> thank you
[00:27:24] <FireflyST> I probably shoulf have pastebinnned that
[00:27:28] <tenex> no biggy
[00:27:30] <claws_bugs> lord poseidon keep me calm
[00:27:31] <dsch04_> Ah yes - 5 disks on a SAT2-MV8 (Marvell chipset)
[00:27:44] <dsch04_> Not seen in the OS
[00:27:49] <kjetilho> FireflyST: you should've taken a pic and uploaded to FLickr
[00:28:01] <dsch04_> format sites there saying "searching for disks..."
[00:28:09] <dsch04_> s/sites/sits/
[00:28:13] <FireflyST> nah, looks like the jpg has no data
[00:28:32] *** sponix has quit IRC
[00:29:15] *** dsch04 has quit IRC
[00:31:29] <SplasPood> hrm... any recent issues post U4 with disk timeouts (specifically with the X4500)?
[00:31:43] <stevel> splaspood: i remember hitting that with the os.o thumpers
[00:31:46] <stevel> fix went into snv_73?
[00:31:52] <stevel> i forget the bugid
[00:32:16] *** jhawk has quit IRC
[00:32:52] <SplasPood> stevel: hrm
[00:33:39] <stevel> SplasPood: 6587133
[00:33:40] *** yippi has quit IRC
[00:33:59] <stevel> http://bugs.opensolaris.org/view_bug.do?bug_id=6587133
[00:34:02] <stevel> is that what you're seeing?
[00:34:50] <FireflyST> here is my output
[00:34:53] <FireflyST> http://rafb.net/p/9rHC8y78.html
[00:35:00] <FireflyST> What do you guys make of that
[00:35:01] <timeless> hi stevel
[00:35:13] * timeless frowns
[00:35:19] *** estibi has joined #opensolaris
[00:35:23] <timeless> which people am i playing tag w/ on tools-discuss?
[00:35:27] * stevel frowns back at timeless
[00:35:28] <SplasPood> stevel: yep, looks pretty darn close
[00:35:35] <FireflyST> looks like my jpgs do have different byte sizes thoughm but all round 4k
[00:35:38] *** theRealBallchalk has quit IRC
[00:36:16] * timeless rotfl
[00:36:25] *** theRealBallchalk has joined #opensolaris
[00:36:29] <timeless> ... and we have an odd complement of individual platforms listed
[00:36:30] <timeless> as well, though those are (sadly) all implemented by project code name
[00:36:30] <timeless> so basically nobody knows what any of them mean.  ;-}
[00:37:02] <stevel> yeah i always loved that
[00:37:13] <SplasPood> stevel: not being too familiar with the os/sol release cycles and what not...  would that fix wind up as a patch to U4, or... ?
[00:37:26] <timeless> stevel: you guys should try the nokia process
[00:37:27] <stevel> possibly as a patch to U4, or maybe just rolled into U5
[00:37:36] <stevel> not sure, and no way for me to find out anymore
[00:37:39] <timeless> the only people who seem to know what code names mean are the people on the outside
[00:37:46] <timeless> since we leak more information than we remember
[00:38:18] <SplasPood> stevel: considering how long I waited for U4 after U3 that'd kinda hurt
[00:38:29] <stevel> timeless: go google for "sun proprietary/confidential" :)
[00:38:31] <ottom> SplasPood: fixed in U5, no indication of it being in a patch
[00:38:32] <stevel> 40,500 results
[00:38:38] <SplasPood> ottom: :(
[00:38:41] <wesolows> I know what the code names mean.  They're easier to remember than the model numbers which are all confusingly alike.
[00:38:57] <dsch04_> OK, I found that there was/is a bug with the driver for the MV8 - http://bugs.opensolaris.org/view_bug.do?bug_id=6566207
[00:39:02] <SplasPood> so its either disable NCQ or just deal with it for.. what, a year?
[00:39:16] <timeless> stevel: i only found 180 for "nokia confidential"
[00:39:20] <dsch04_> It says it was fixed in "snv71"
[00:39:47] <SplasPood> dsch04_: what does that equate to?
[00:39:58] <dsch04_> Is the "71" the same number as in: sol-nv-b74-x86 ?
[00:40:18] <alanbur> yes, except no
[00:40:19] <wesolows> yes
[00:40:24] <SplasPood> Since I'm not running osol in production
[00:40:25] <dsch04_> Heh
[00:40:28] <SplasPood> just 'proper' solaris
[00:40:29] <wesolows> yes except 71 != 74 :-)
[00:40:36] <dsch04_> Ah, I see
[00:40:38] <alanbur> egzakerly :-)
[00:40:53] <dsch04_> I have 74, so I should have the fix
[00:41:02] <dsch04_> not that then
[00:41:35] *** oninoshiko has quit IRC
[00:43:03] <alanbur> For those of you on the inside, there is a list of all the product names on sunsolve internal
[00:43:35] <SplasPood> What svn# does U4 equate to?
[00:43:37] * elektronkind covets SWAN access ;)
[00:43:48] <SplasPood> snv even
[00:43:53] * SplasPood types 'svn' way too often
[00:44:13] <alanbur> I meant HW products, not releases ;-)
[00:44:30] <SplasPood> yea my question was unrelated
[00:44:37] <elektronkind> SplasPood: u4 doesn't equate to any specific nevada release. the new features/changes in u4 are more or less cherry-picked from nevada... features that are ready to be in a production solaris release
[00:44:38] <sommerfeld> SplasPood: not directly comparable.
[00:44:39] *** victori_ has quit IRC
[00:44:47] <alanbur> And there isn't a direct relationship anyway
[00:44:48] <SplasPood> gotcha
[00:44:51] <SplasPood> haha
[00:44:52] <dsch04_> So, is there some other way to list disk devices, other than "format" ?
[00:45:19] <wesolows> elektronkind: some features that are ready, but not the ones you want, and a bunch that aren't ready but marketing has decided need to be in
[00:45:29] *** victori_ has joined #opensolaris
[00:45:36] <elektronkind> dsch04_: iostat -nE
[00:45:37] <bda> Heh.
[00:46:20] <elektronkind> wesolows: hehe that's about right ;) although iscsitgtd can be marked off my wish list :)
[00:46:59] <dsch04_> elektronkind, thnks, but that hangs too. :(
[00:47:20] <elektronkind> something's up with your drive/buss
[00:47:44] <elektronkind> scsi inquiry commands not being received/returned or something like that
[00:48:06] <elektronkind> are there warnings or errors being reported by the kernel in /var/adm/messages ?
[00:48:16] * dsch04_ looking
[00:48:33] <SplasPood> yea
[00:48:41] <SplasPood> iscsitgtd was what I wanted for in U4
[00:48:56] <alanbur> kstat -c disk
[00:49:01] <SplasPood> only to start experiencing this DMA timeout issue
[00:49:14] <SplasPood> s/wanted/waited
[00:49:16] <flyingparchment> did pfil.ap finally go away in S10U4?
[00:49:29] <elektronkind> dsch04_: you can google for a copy of the dtrace scripted called "scsi.d" which is kinda like a snoop for scsi traffic. You can use that to see your buss or a drive is being wacko
[00:49:55] <dsch04_> device on port n still busy after reset
[00:51:23] <elektronkind> that error seems vaguely familiar to me. this is on a thumper ?
[00:52:36] <dsch04_> No. What's a thumper?
[00:52:45] <stevel> dsch04_: x4500
[00:53:00] <dsch04_> It's a home-built NAS device (or at least it will be when the disks work!)
[00:53:23] <wesolows> see?  code names are easier :-)
[00:53:56] <elektronkind> dsch04_: what kind of controller do you have in it?
[00:54:31] <dsch04_> Supermicro SAT2-MV8
[00:55:18] <dsch04_> It's using the marvell88sx0 driver
[00:55:56] <timeless> mixed
[00:55:59] <timeless> oops
[00:56:06] <alanbur> nrubsig: got one of these? http://www.youtube.com/watch?v=LsV50T5uEyw
[00:57:37] <wesolows> nrubsig: not sure if you were aware of this, but it looks like the kernel has been built with -xstrconst since may
[00:57:43] <elektronkind> dsch04_: hmm, this could be one of those situations where the rev of marvell chip on that card is a rev that the solaris marvell driver doesn't completely jive with (yet) ... this is just a guess on my part from past experiences with 3rd party HBAs
[00:58:01] <dsch04_> :(
[00:58:13] <nrubsig> wesolows: I know about this from the beginning.
[00:58:27] <wesolows> ok...I just happened to notice it and thought I'd mention it
[00:58:59] <elektronkind> the best way to figure this out is to physically look at the chip on the card and make a post about it to the storage and/or drivers list with any chip model number and revision info you find on the chip
[00:59:18] <SplasPood> hehe seems like even the 1st party HBAs aren't fully supported ;)
[00:59:31] <elektronkind> since the marvell driver is closed source it'll be hard to tell without someone who can look at the code
[00:59:59] <elektronkind> sucks that it's not working for you right now :/
[01:00:12] *** AtomicPnk has quit IRC
[01:00:36] *** AtomicPnk has joined #opensolaris
[01:00:47] *** sniffy_ has joined #opensolaris
[01:01:03] *** FBdev has quit IRC
[01:01:27] <elektronkind> hmm, and it looks like marvell doesn't provide their own solaris driver... they only provide one for their 1Gb ethernet chip :/
[01:01:36] <nrubsig> alanbur: I have a GAU-12 with powered armor for that purpose.
[01:01:59] <dsch04_> Yeah - I need that too, this board has marvell ethernet
[01:02:00] *** AtomicPnk has quit IRC
[01:02:09] * dsch04_ thinking about getting a new board
[01:02:24] <elektronkind> how many SATA busses do you need?
[01:02:29] <dsch04_> But the SAT2-MV8 is v. cheap
[01:02:35] *** AtomicPnk has joined #opensolaris
[01:03:05] <dsch04_> I'm planning on 2x(small) SATA drives for system (mirrorer) and 8x(big) drives for data
[01:03:35] <dsch04_> I've got a system running this set up currently on another server running linux
[01:04:14] <dsch04_> I'll put the system drives on the mobo ports and the 8x(big) drives on the controller
[01:07:45] *** Megaf has joined #OpenSolaris
[01:08:24] *** sommerfeld has quit IRC
[01:11:56] *** pablomh has joined #OpenSolaris
[01:12:52] *** alanbur has left #opensolaris
[01:14:31] *** pfa3rh has quit IRC
[01:15:05] *** victori_ has quit IRC
[01:15:57] *** victori_ has joined #opensolaris
[01:16:43] *** l1s has quit IRC
[01:16:59] *** sniffy has quit IRC
[01:20:34] *** Mdx4 has quit IRC
[01:23:06] *** SYS64738 has joined #opensolaris
[01:24:29] <dsch04_> elektronkind, which are the storage/drivers list?
[01:32:59] *** theRealBallchalk has quit IRC
[01:33:10] *** theRealBallchalk has joined #opensolaris
[01:33:50] *** rmorse has joined #opensolaris
[01:37:46] *** Gman has quit IRC
[01:40:35] *** vmlemon has quit IRC
[01:42:03] *** rmorse has left #opensolaris
[01:42:03] *** rmorse has quit IRC
[01:47:52] *** stevel has quit IRC
[01:48:28] *** gentry- has quit IRC
[01:56:08] *** theRealBallchalk has quit IRC
[01:56:21] *** theRealBallchalk has joined #opensolaris
[01:56:51] *** medar has quit IRC
[01:57:33] *** uebayasi has joined #opensolaris
[01:58:35] <moazamraja> anyone know what package libiconv.so is a part of?
[01:58:59] *** Megaf has quit IRC
[01:59:07] *** NikolaVeber_ has quit IRC
[01:59:08] *** wnorrix has joined #opensolaris
[02:00:02] *** medar has joined #opensolaris
[02:01:38] *** UberDuper has left #opensolaris
[02:03:49] *** yongsun has joined #opensolaris
[02:05:12] *** yongsun has quit IRC
[02:05:27] *** yongsun has joined #opensolaris
[02:15:00] <nrubsig> moazamraja: the |iconv()|-APIs are part of libc.so.1
[02:16:24] *** JBeck has quit IRC
[02:21:40] *** sleepcat has joined #opensolaris
[02:22:19] *** _mw46_ has quit IRC
[02:25:21] *** yippi has joined #opensolaris
[02:26:55] *** ChanServ sets mode: +o Tpenta
[02:27:05] *** yippi has quit IRC
[02:27:20] *** jhawk has joined #opensolaris
[02:29:22] *** jhawk has quit IRC
[02:29:38] *** comay has joined #opensolaris
[02:31:19] *** ChanServ sets mode: +o comay
[02:41:29] *** gcmandrake has joined #opensolaris
[02:42:50] <nrubsig> comay: comaaaaaaaay!
[02:43:30] * nrubsig runs around and meows to get comay's attention...
[02:44:19] <bda> O_o
[02:48:05] * nrubsig meows harder...
[02:50:25] *** phrost is now known as phrost6
[02:51:09] <bda> Dude, you are weird. :)
[02:51:50] <nrubsig> MEOOOOWWWW!!
[02:52:04] <jbk> haha
[02:53:24] * nrubsig makes a mental note: "MEEEOOWWWW ME-OO-WW-OO-<mouse>-MCMOMAMYMPMHMOMNMUMBMEMR!!"
[02:55:42] <nrubsig> (now you people have to figure out what the note is... :-) )
[02:56:18] <ashner> Mouse-hole C?
[02:56:35] <nrubsig> ashner: no
[02:56:43] <ashner> awwww
[02:57:11] <nrubsig> try $ ksh93 -c 'x="MEEE<from-above>" ; print -- "${x//M}"'
[02:57:23] <moazamraja> nrubsig: that's odd cuz one of my machines doesn't have the iconv so files
[02:57:30] <moazamraja> at least not anywhere in /usr/
[02:57:40] <timeless> help
[02:57:44] *** tenex has quit IRC
[02:57:57] <timeless> has anyone managed to get a bugzilla account from steven's bugzilla?
[02:58:19] <nrubsig> moazamraja: erm... the iconv functionaliy was folded into libc, the indivitual locale/charset stuff is located in different packages
[02:58:56] *** gcmandrake has quit IRC
[03:00:32] <jbk> timeless: lemme try..
[03:00:37] *** gcmandrake has joined #opensolaris
[03:00:58] <jbk> hmm.. seems to be taking a while after submitting my email address
[03:01:05] <jbk> is that where you're getting hung up?
[03:03:05] *** sv2_ has left #opensolaris
[03:03:45] *** nachox has joined #opensolaris
[03:04:37] <jbk> hmm it finally completed.. but no email
[03:06:27] *** simford has joined #opensolaris
[03:06:42] * jmcp is still waiting for the bugzilla signup email
[03:06:50] <timeless> oh good, glad to hear i'm not alone
[03:06:52] *** Chihan has joined #OpenSolaris
[03:06:54] <jbk> yeah
[03:07:01] <Doc> hrmmm
[03:07:08] <sleepcat> O_o
[03:07:18] <sleepcat> ;-P
[03:07:21] <jbk> anyone have access to do a mailq in the zone it's running in? :)
[03:07:22] <sleepcat> :-0
[03:07:41] <dlg> FrostCS?
[03:09:10] *** victori_ has quit IRC
[03:09:15] <jbk> is there a page for all the bug tracking eval stuff, or is it all being done via email?
[03:09:55] *** victori_ has joined #opensolaris
[03:10:00] <sleepcat> how will solaris admin salaries go if solaris becomes TOO easy to use?
[03:10:10] <sleepcat> has anyone thought of that?
[03:11:03] <elektronkind> too easy? there's always complex stuff being added :)
[03:11:23] <sleepcat> elektronkind: it is getting easier
[03:11:27] <jbk> heh.. if you're paying bottom dollar for someone to admin something like an E25k, you'll get what you deserve :)
[03:11:45] <sleepcat> how many people have E25ks though
[03:11:52] <elektronkind> lots
[03:11:54] <sleepcat> most businesses are like 50 people
[03:12:09] <flyingparchment> E25k is so 1990's
[03:12:16] <flyingparchment> horizontal scaling is in
[03:12:20] <sleepcat> exactly
[03:12:29] <jbk> depends :)
[03:12:33] <sleepcat> that's why i bet sun's x86 business is where most of the money is
[03:12:36] <elektronkind> true, but they're big investments and the machines do have legs for oltp
[03:12:37] <jbk> not every workload scales that well
[03:12:41] <jamesd> any vxvm guru's around?   ... how do i  make veritas re-use a disk marked "FAILING"   i fixed the problem, but veritas doesn't want to use it.. and it wont replace the disk because its only marked failing  not failed... and i don't want to reboot the box.
[03:13:02] *** victori_ has quit IRC
[03:13:12] <jbk> jamesd: I think it's buried somewhere in vxedit
[03:13:13] <elektronkind> jamesd: vxdiskadm, remove the disk, add it back ?
[03:13:38] <jbk> though i don't have access to the vxedit manpage to look
[03:14:19] <jbk> i've heard lots of horror stories about people that drink oracle's kool-aid about rac and think that everything can scale horizontally
[03:14:38] <jbk> except, the minute their interconnect gets saturated, their performance goes downhill _fast_
[03:14:47] *** Kitty has quit IRC
[03:14:47] <jamesd> elektronkind, it refueses because the disk is only "failing" not failed.
[03:14:48] <jbk> (ignoring all the other issues with RAC, especially on linux)
[03:14:49] <nachox> meh, i bought a dosk for my laptop to install nexenta and it didnt exactly came empty, it had a windows vista installed...
[03:15:26] <elektronkind> jamesd: I take it that a 'vxdctl enable' doens't put sense back into it?
[03:15:32] <jbk> jamesd: vxedit -g diskgroup set failing=off diskname
[03:15:48] <elektronkind> jkb with the sniper shot
[03:15:56] <elektronkind> er jbk ;)
[03:15:59] <jamesd> elektronkind, haven't tried it....  not a veritas expert, and this is a prod box... wanted advice from an expert... first
[03:16:01] <timeless> jamesd/jbk: did someone email steven saying it isn't working?
[03:16:08] <jbk> not yet..
[03:16:49] <nachox> jamesd, when the problem is too big there is no such thing as a production box :)
[03:17:22] <jamesd> nachox, the box is running and no impact to customers....  just 1/2 of the root mirror is off line...
[03:17:23] <jbk> jamesd: that command should do it
[03:17:40] <jbk> i just had to find the correct syntax
[03:17:54] <jbk> i've run into it, just not often enough to memorize the cmd
[03:18:04] <nachox> weird, i thought people actually avoided using veritas to mirror the root filesystem
[03:19:17] <jbk> naah
[03:19:27] <jbk> my last place did it on all their sun boxes
[03:19:36] <jbk> because disksuite sucks for SANs
[03:19:44] *** WifiJane has quit IRC
[03:19:46] <jbk> and didn't feel the benefit in having two different toolsets
[03:20:46] <jbk> though upgrading veritas encapsulated rootdisks is tricky
[03:21:24] <jbk> even more so if you have more than 5 volumes on your root disk
[03:22:00] <jbk> though it can be done
[03:22:17] <jbk> (just don't ask how :P it's messy)
[03:23:32] <jamesd> okay  vxdctl enable is not working...  appears to have done nothing...   disk is still listed as FAILING  and no io is going to that disk
[03:24:10] <jbk> jamesd: did you do the vxedit?
[03:24:32] <jamesd> jbk no i missed it...
[03:24:43] <jbk> vxedit -g diskgroup set failing=off diskname
[03:26:20] <jamesd> jbk how do i get it to reattach the volumes  that are detached
[03:26:36] <jbk> it detached them?
[03:27:03] <jamesd> yeah.. when it marked the disk failing.. it detached that 1/2 of all the mirrors
[03:27:14] *** Kitty has joined #opensolaris
[03:28:06] <jamesd> disk was marked failing,  and it has IOFAIL on all the affected plexes...
[03:28:19] <jbk> probably a vxvol command
[03:28:21] *** Kitty has quit IRC
[03:29:30] *** aka_druid has quit IRC
[03:29:33] <jbk> vxmend might also work..
[03:29:48] <jbk> there is a vxvol attach command, but i can't remember the exact syntax
[03:29:58] <jbk> the gui might be a bit easier
[03:30:06] *** Kitty has joined #opensolaris
[03:30:13] *** aka_druid has joined #opensolaris
[03:31:02] *** comay has quit IRC
[03:31:26] *** tenex has joined #opensolaris
[03:32:27] *** gcmandrake has quit IRC
[03:34:22] <jbk> jamesd: btw, unless it was a less than useful install, the manpages should be in /opt/VRTS/man
[03:34:33] <jbk> they're decent
[03:34:40] <jamesd> yeah they are there....  i'
[03:34:51] <jamesd> m using google for man pages anyway..
[03:35:06] <nachox> dont we all?
[03:35:09] <jbk> i don't suppose the site has a veritas support contract? :)
[03:35:18] <jbk> they could probably give you the command off the top of their head..
[03:36:08] <jamesd> yeah,  but now i'm away from the office.. and  don't want to call my co-workers to get that info... this was supposed to be a simple.. replace disk...  with the same disk and have vxdiskadm repair the damage
[03:36:30] <jbk> why replace it with the same disk?
[03:37:24] <jamesd> yes.. it was a bone head error ( yes mine) that caused the problem...
[03:47:51] *** sleepcat has quit IRC
[04:10:57] *** WifiJane has joined #opensolaris
[04:16:22] *** dclarke has joined #opensolaris
[04:22:03] *** hile_ has joined #opensolaris
[04:23:18] *** master_o1_master has joined #opensolaris
[04:32:16] <dclarke> evening folks
[04:32:26] *** diomedes has joined #opensolaris
[04:33:12] *** victori_ has joined #opensolaris
[04:33:31] <dclarke> jmcp around ?
[04:33:46] *** master_of_master has quit IRC
[04:33:57] <jbk> was earlier
[04:34:50] <dclarke> he had a nice blog entry that was talking about the AMD Opteron BIOS issue .. the erratum 109 thing
[04:35:11] <dclarke> I think I may be stuck booting windows on this box to run the AMIBIOS motherboard scanner
[04:38:15] *** diomedes has quit IRC
[04:38:42] *** lisppaste3 has quit IRC
[04:39:29] <jmcp> dclarke: http://blogs.sun.com/jmcp/entry/where_oh_where_is_my
[04:39:43] <dclarke> yes Sir
[04:39:54] <dclarke> hello there and thank you for suffering the same plight as I :-)
[04:39:57] <jmcp> :)
[04:40:01] <jmcp> I guess you're welcome
[04:40:13] <dclarke> be damned if I can figure out where to get the BIOS update ..
[04:40:14] <jmcp> not that I really want to suffer from that issue
[04:40:24] <dclarke> these mystery meat AMD Opteron's are no savings long term
[04:40:39] <dclarke> well .. Solaris 10 Update 4 runs fine on it .. thus far
[04:40:50] <dclarke> # uptime
[04:40:51] <dclarke>  10:40pm  up 33 min(s),  1 user,  load average: 0.00, 0.00, 0.02
[04:41:29] <dclarke> I'm just going to shutoff graphical login totally and then switch GRUB over to ttya
[04:41:38] <Doc> jmcp: that text would make much more sense if your hostname wasnt "broken"
[04:41:43] <dclarke> then I'll see if that works .... then go hunt for a BIOS update
[04:41:46] <Doc> and/or if you worked out how to put linebreaks into your blog
[04:42:08] <dclarke> I was gonna say .. a few <pre> </pre> would be nice
[04:43:45] <jmcp> Doc: it would also make more sense if I could be arsed fixing the css for the blog. which I can't. deal
[04:44:35] <dclarke> well .. it was one of the few meaningful pages that came up via a google search
[04:44:41] <dclarke> so .. I'm happy with it
[04:45:23] *** lisppaste3 has joined #opensolaris
[04:47:36] <dclarke> wow .. the world seems so quiet once an AMD Opteron rack server shuts down
[04:48:59] *** ashner has quit IRC
[04:49:00] <nrubsig> dclarke: wow... the world seems to quiet once all nuclear missels have been detonated.
[04:49:18] <nrubsig> groan
[04:49:29] <nrubsig> each time I make a joke I mess it up with a typo
[04:49:35] <nrubsig> s/to quiet/so quiet/
[04:49:40] <dclarke> I just spent the last four hours with a 1U dual Opteron box .. on my desk
[04:49:49] <dclarke> or at least .. three feet away
[04:49:52] <nrubsig> let me guess:
[04:50:00] <nrubsig> <rooaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaarrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr>
[04:50:01] <nrubsig> ?
[04:50:07] *** nachox has quit IRC
[04:50:09] <dclarke> woo hoo .. my new null modem works !
[04:50:14] <dclarke> GRUB works on ttya
[04:50:19] <dclarke> happy me :-)
[04:50:33] <dclarke> I love it when I get the pin outs right
[04:50:40] * nrubsig uses a knife and stabs dclarke's null modem
[04:50:50] * dclarke arghhhhh
[04:50:55] <nrubsig> dclarke: luck only lasts a few seconds before it becomes sour
[04:51:02] <wesolows> he didn't stab anything...it was a null modem after all
[04:51:19] <dclarke> so long as it survives boot ... I'm happy
[04:52:17] <dclarke> I always wonder what the hell good is the IPMI ports .. I never used them
[04:52:29] <dclarke> the baseboard stuff .. whatever that is
[04:52:38] <dclarke> must be SNMP crud
[04:55:03] <dclarke> after a fresh install .. I wonder why /sbin/dhcpagent is running .. when this machine uses a static IP
[04:56:54] <dclarke> my definition of a secure web server is one where there are no network access services at all .. period .. just HTTP running and a serial console
[04:56:55] <jbk> nwam?
[04:57:14] <dclarke> nwam ?  never heard of it
[04:58:21] <dclarke> looks like even S10u4 still has a minor issue with svc:/application/print/rfc1179
[04:59:03] *** halton has joined #opensolaris
[05:00:24] <dclarke> pardon me while I mutter to myself via a keyboard .. but what the hell is svc:/application/opengl/ogl-select
[05:01:04] <rootard> dclarke: looks like a service to me
[05:01:15] <dclarke> uh huh
[05:01:31] <dclarke> I'm killing off those services I see no use for .. on this machine
[05:01:41] <dclarke> like svc:/application/management/wbem
[05:04:02] *** tenex has quit IRC
[05:04:18] *** Megaf has joined #OpenSolaris
[05:11:26] *** rootard has left #opensolaris
[05:13:17] <dclarke> so I'm curious what the general philosophy is with regards to SVM mirrors and the location of the metadb slice .. I tend to use slice 4 in the middle of the disk geometrically .. but does it really matter where one locates it ?
[05:13:55] <jbk> probably not
[05:14:13] <dclarke> that's what I figured
[05:15:39] <jbk> it will be nice when such considerations are no longer necessary...
[05:16:42] <dclarke> I think that we are going to wait another year before that is in the mainstream Solaris world
[05:17:20] *** stukreit has joined #opensolaris
[05:17:30] <jbk> probably
[05:17:47] <jbk> since i think it's still a few more builds before newboot sparc integrates
[05:18:25] <dclarke> hrmmm
[05:18:30] * dclarke nod
[05:18:45] <dclarke> well .. I'm happy with this config thus far .. ps -ef | wc -l == 24
[05:18:53] <dclarke> that is a nice quiet machine
[05:19:16] <dclarke> of course with Solaris 8 I can get that down to about 9 processes
[05:19:25] <dclarke> but that is ancient history now
[05:23:54] *** Megaf is now known as _Megaf
[05:24:13] *** sfire||mouse has joined #opensolaris
[05:26:41] *** _Megaf is now known as drupal
[05:28:22] *** drupal is now known as _Megaf
[05:29:53] *** sleepcat has joined #opensolaris
[05:32:36] *** uebayasi has quit IRC
[05:37:42] *** nasser has joined #OpenSolaris
[05:40:12] <e^ipi> I wonder what this monsooned malabar will taste like
[05:41:13] <jbk> dirt? :)
[05:41:26] <e^ipi> *shrug*
[05:41:28] <e^ipi> maybe
[05:41:45] <e^ipi> if so, no big loss... $4 worth of beans
[05:43:16] *** slowho1 has joined #opensolaris
[05:44:35] *** slowho1 has left #opensolaris
[05:44:40] *** sfire||mouse has quit IRC
[05:44:47] *** SirFunk has joined #opensolaris
[05:44:50] <e^ipi> but I'll see in a couple days
[05:45:02] <jbk> slow metabolism?
[05:45:07] <jbk> or just not eating it tonight?
[05:47:27] *** _Megaf is now known as [_gutsy]_Megaf
[05:48:31] <e^ipi> roasted coffee takes a couple days to degas before it starts being tasty
[05:48:45] <jbk> ahh
[05:48:49] <jbk> <-- not a coffee drinker
[05:49:36] *** [_gutsy]_Megaf is now known as Megaf
[05:50:46] *** slowhog has quit IRC
[05:52:02] *** friendly12345 has joined #opensolaris
[05:57:34] <e^ipi> I've started roasting my own, primarily due to the best coffee in town being starbucks
[05:57:46] <e^ipi> which is a pretty damning critique of this town actually
[05:57:47] <jbk> heh
[05:57:52] <jbk> no tim hortons?
[05:58:01] <e^ipi> tim hortons is disgusting
[05:58:13] <e^ipi> they serve crap stale robusta from a can
[05:59:00] <jbk> isn't that like heresy for a canadian to say such a thing? :)
[05:59:05] <e^ipi> *shrug*
[05:59:20] <e^ipi> I'm not going to let jingoism force me in to drinking bad coffee
[06:00:26] *** wnorrix_ has joined #opensolaris
[06:02:23] *** karrotx has joined #opensolaris
[06:04:00] *** Gman has joined #opensolaris
[06:04:00] *** ChanServ sets mode: +o Gman
[06:04:07] <e^ipi> hey gman
[06:04:20] <jbk> evening Gman
[06:04:45] <Gman> hey e^ipi, jbk
[06:08:00] <e^ipi> So, I caught some bloody cold or something on the airplane home
[06:08:05] <jbk> ack
[06:08:07] <jbk> i hate that :(
[06:08:14] <Gman> ugh
[06:08:17] <e^ipi> yeah... stupid recycled air
[06:08:40] <jbk> i just had to deal with a monsoon over houston that delayed landing (and made the drive home interesting)
[06:08:47] *** nasser has left #OpenSolaris
[06:08:48] <Gman> i usually get sick when i get home too, because i've put my body through all sorts of stresses
[06:08:53] *** wnorrix has quit IRC
[06:09:41] <jbk> i still don't get that myself -- I spend hours sitting, yet feel exhausted afterwards when i fly
[06:09:52] <e^ipi> my flight from LAX to vancouver was delayed because the stewardess quit her job on the spot
[06:10:01] <e^ipi> so they had to find a new one
[06:10:03] <jbk> damn
[06:10:10] <jbk> what airline?
[06:10:13] <e^ipi> united
[06:10:14] <jbk> fu air? :)
[06:10:19] <jbk> ahh.. close
[06:10:41] <e^ipi> yeah, united seem pretty bad
[06:10:48] <e^ipi> almost as bad as air canada
[06:10:57] <jbk> i've never had the pleasure
[06:11:03] <jbk> only driven in/out of canada
[06:11:04] <e^ipi> avoid if possible
[06:11:07] <moazamraja> re
[06:11:20] <e^ipi> if you ever need to fly in canada, try and catch a westjet flight
[06:11:23] <e^ipi> they're pretty good
[06:11:35] <moazamraja> I got a survey type question for everyone...
[06:11:46] <jbk> i actually like southwest, but i know their 'no assigned seating' policy seems to offend some people
[06:12:05] <moazamraja> what features on Solaris are key for developers that they can't access on other OSes (linux/windows/osx, etc.)?
[06:12:18] <e^ipi> moazamraja: rbac
[06:12:34] <jbk> i think here soon zfs (at least for anyone doing kernel hacking)
[06:12:37] <moazamraja> features that would make developers apps shine on Solaris and be difficult on any other OS
[06:12:40] <sleepcat> rbac looks complicated
[06:12:43] <e^ipi> dtrace
[06:12:46] <moazamraja> non-kernel/OS-developers
[06:12:47] <e^ipi> sleepcat: it's really not
[06:12:47] <moazamraja> app developers
[06:13:10] <jbk> rbac, dtrace, smf are the biggest off the top of my head
[06:13:13] *** Megaf has quit IRC
[06:13:37] *** sahafeez has joined #opensolaris
[06:13:38] <jbk> maybe ptools + mdb as well
[06:13:40] <e^ipi> thread performance
[06:14:36] <e^ipi> STREAMS undoubtedly has some sort of advantage
[06:14:41] <moazamraja> well...for example, on Windows you get DirectX
[06:14:43] <e^ipi> SCO has that though
[06:14:52] <moazamraja> on OS X you get Quartz, Core Graphics, Core Data
[06:14:54] <jbk> if no one else gets to it sooner, i might eventually attempt to tackle source browsing support for mdb (note i said attempt, doesn't mean i'll succeed), but got enough other stuff ahead of that, might be a long time..
[06:15:04] *** wnorrix_ has quit IRC
[06:15:05] <moazamraja> which make developers apps a lot easier to write, using free functionality from the OS
[06:15:10] <moazamraja> that's the type of stuff i'm talking about
[06:15:28] <e^ipi> DISM
[06:15:54] <jbk> well most of those seems to be graphic oriented
[06:16:08] <jbk> that is i think a bit of a weak area (currently)
[06:16:09] <moazamraja> coredata is not
[06:17:08] <friendly12345> I'm currently trying to install Solaris Express under vmware and it's taking centuries to bring up the windowing system. Is this normal?
[10:26:02] *** echelog has joined #opensolaris
[10:30:05] <dsch04> Morning all
[10:58:28] *** echelog has joined #opensolaris
[11:00:56] <flyingparchment> Giaco: lrc is for legacy init scripts
[11:00:56] *** JWheeler has quit IRC
[11:01:05] *** Mazon has joined #opensolaris
[11:01:50] *** polk__ has quit IRC
[11:01:51] <Giaco> flyingparchment, how are started the lrc scripts ?
[11:01:59] <flyingparchment> what do you mean?
[11:02:17] *** linma has joined #opensolaris
[11:02:20] <Giaco> svcadm enable make services started at boot time
[11:02:33] *** NikolaVeber has joined #opensolaris
[11:02:35] <Giaco> and what about /etc/init.d/ scripts ?
[11:02:42] <flyingparchment> symlinks in /etc/rc?.d/
[11:02:57] *** polk__ has joined #opensolaris
[11:03:56] <Giaco> there is already a K16apache
[11:05:20] *** sioraiocht has quit IRC
[11:05:53] <quasi> that's for stopping
[11:08:43] <trochej> Giaco: You understand lrc scripts?
[11:10:37] *** renihs has quit IRC
[11:11:20] <Giaco> not much
[11:11:44] <Giaco> I am trying to install lightsquid
[11:12:57] *** renihs has joined #opensolaris
[11:13:04] <Giaco> but It seems that apache don't see perl script
[11:13:08] <Giaco> don't read
[11:16:57] *** sponix has joined #opensolaris
[11:16:59] <quasi> #apache might be worth a try
[11:17:27] *** calumb has joined #opensolaris
[11:29:50] *** JWheeler has joined #opensolaris
[11:35:36] *** simford has quit IRC
[11:35:56] <sponix> NV B74 Working good for everyone ?
[11:36:27] <trochej> YUp
[11:36:34] * trochej being everyone now :)
[11:37:25] <Giaco> ok with apache2 all works
[11:37:40] <Teltariat> I still have yet to get b74 installed, as it still has that stupid "Can't Find Install Media" bug
[11:38:07] <sponix> Just looking for a good build, wondering if I should wait for b75, or the next SXDE
[11:38:21] <JWheeler> I expect a lot of people will be holding out for b75
[11:38:38] <Teltariat> The SXDEs force you to install a full install, don't they?
[11:38:57] <Teltariat> All 8 gigs, right?
[11:39:02] <sponix> Teltariat: yep, SXCE lets you choose ?
[11:39:17] <Teltariat> Yes
[11:39:22] <Teltariat> For b69 at least
[11:39:26] <sponix> when are the Indiana builds scheduled to kick off ?
[11:39:34] <Teltariat> As I mentioned, I'm having issues with b74.
[11:39:45] <cmihai> Teltariat, it's not 8GB
[11:39:48] <Giaco> is a squid directive to compress logs after the -k rotate ?
[11:40:01] <sponix> I have SXDE b70 right now, few little things, but over all not a bad build
[11:40:19] <Teltariat> cmihai: well its still a lot.  the full install is definitely more than 6 gigs, I think.  Am I wrong?
[11:40:19] <cmihai> and you should do a Full Install + OEM regardless. It's the Developer edition "caiman" installer that does that, yes. You can still pick the old installer and don't install development tools
[11:40:39] <Teltariat> I'm still a fan of Reduced Core Network
[11:40:43] <cmihai> More like 5-6 GB
[11:40:53] <cmihai> Teltariat, that seems like a waste of time....
[11:41:01] <Teltariat> cmihai, not for my purposes.
[11:41:04] <cmihai> You save what, 1-2GB for hours of searching for depends? :-)
[11:41:07] *** wnorrix has quit IRC
[11:41:10] <iMax> btw does anyone know if the memory footprint of the installer is expected to change?
[11:41:15] <iMax> the new one I mean
[11:41:24] <cmihai> iMax, does it matter?
[11:41:30] <cmihai> Unused memory is wasted memory.
[11:41:37] <cmihai> You can still install in 256MB in text mode.
[11:41:38] <Teltariat> cmihai: CSW picks up most of the slack for me. :D  And my total disk space uses is still less than .9 gigs.
[11:41:45] <cmihai> anything less than 256MB won't really work anyway.
[11:41:50] <iMax> well, yes, have 512 MB assigned on my VM, caiman needed 750 or something
[11:42:01] <iMax> sounds a bit much for just an installer
[11:42:06] <Teltariat> agreed
[11:42:13] <cmihai> iMax, nobody forces you to use it.
[11:42:18] <iMax> ture
[11:42:19] <cmihai> Install in text mode with the old installer.
[11:42:21] <iMax> true
[11:42:26] <cmihai> Or JumpStart.
[11:42:28] <iMax> that is what I did...just asking
[11:42:48] <iMax> I did not say it is bad or anything, just sounds a bit much to me
[11:43:01] <Teltariat> It seems you can Jumpstart off of an OS other than Solaris.  Anyone done this before?
[11:43:08] <Teltariat> <-- Jumpstart newb
[11:43:31] <sponix> iMax: From what I've read, they are looking to trim the base install down, both size and memory wise, at least with the Project Indiana goals
[11:43:50] <quasi> Teltariat: not exactly jumpstart
[11:44:07] *** noyb has quit IRC
[11:44:14] <sponix> quasi: just using dhcp to cast the image out by mac, riiight ?
[11:44:24] *** nrubsig has quit IRC
[11:44:25] <iMax> sponix: ok thanks, because the installer requiring more memory than the OS with a GUI is a bit weird
[11:44:39] <quasi> sponix: something like that
[11:44:49] <Teltariat> hmm
[11:44:52] <Teltariat> worth a shot
[11:45:20] <cmihai> Teltariat, do it from Solaris or don't do it at all.
[11:45:21] <sponix> iMax: actually, that is close to the norm, Installs are more intense than most day to day operations, running several package copies and configurations
[11:45:36] <cmihai> Teltariat, do it in VMware if you must, but do it on Solaris. You can use JET too.
[11:45:50] <Teltariat> cmihai: Yes sir.
[11:46:15] <sponix> iMax: if you go look at reqs for bsd or linux you will see the same pattern, base required mem being less than fancy installer mem
[11:46:32] <iMax> sponix: yes, of course, also taking care of all package dependencies is probably not that easy
[11:46:46] <iMax> sponix: never noticed that :)
[11:46:48] <sponix> Know what kills me ?
[11:47:07] <quasi> Teltariat: http://www.samag.com/documents/s=9598/sam0504e/
[11:47:16] <kjetilho> cmihai: replacing the DHCP server is not an option for all of us ...
[11:47:27] *** LuckyLuke has joined #opensolaris
[11:47:28] <kjetilho> for a home network I guess it's no ordeal, though
[11:47:31] <cmihai> kjetilho, you don't have to.
[11:47:36] <cmihai> You ca use JumpStart with ANY DHCP server.
[11:47:39] <cmihai> Even ISC's.
[11:47:42] *** calumb has quit IRC
[11:47:51] <kjetilho> yes, but you're saying "do it on Solaris"
[11:47:52] <Teltariat> cmihai: Have just a FreeBSD laptop here, can't install Solaris on it, don't want to wipe out anything on it; Solaris _refuses_ to run in qemu.  So what can I do? The target machine cannot install b74 due to that "Cannot Find Install Media" issue.
[11:47:56] <kjetilho> what do you mean by that then?
[11:48:12] <quasi> kjetilho: even so, just setting up solaris dhcp to only answer specific mac addys, should be fine
[11:48:12] <cmihai> kjetilho, you can JumpStart on Solaris with the DHCP server on another machine.
[11:48:20] *** tfb has joined #opensolaris
[11:48:21] <Teltariat> Thank you, quasi
[11:48:26] <sponix> Take SXDE && Ubuntu-Desktop for example, both do a _full_ install no matter what (basically), yet they both set there resolving all the base package deps like every single package is ignorant of the next. Why not just extract a big tarBall of all the common base stuff, and then cp its package use/deps to match, forget all the CPU grinding for no reason
[11:48:39] <sponix> silly if you ask me
[11:49:09] <kjetilho> cmihai: as I mentioned yesterday, I do Jumpstart with no Solaris involved
[11:49:21] <kjetilho> Jumpstart only runs on the machine you're installing, really
[11:49:32] <sponix> Big Options during install, keyboard, timezone, video driver package... Why not just have a prefab base of _everything_ else, and then roll out that specific junk at the end
[11:49:35] <kjetilho> so it depends on what you mean by "running Jumpstart"
[11:49:53] *** The-spiki has quit IRC
[11:53:22] *** Cyrille has quit IRC
[11:53:47] *** KermitTheFragger has joined #opensolaris
[11:55:15] *** jossh has joined #opensolaris
[11:55:53] <iMax> sponix: agreed. but I have to say I never tried to program an installer, so there might be some downsides :)
[11:57:32] *** linma has quit IRC
[11:58:15] *** ichigo has joined #opensolaris
[11:59:33] *** bnitz has joined #opensolaris
[12:07:08] *** salamanders has joined #opensolaris
[12:07:47] *** mega has joined #opensolaris
[12:17:07] *** dsch04_ has joined #opensolaris
[12:19:39] *** FelipeBare has joined #opensolaris
[12:22:06] *** calumb has joined #opensolaris
[12:24:25] *** Giaco has quit IRC
[12:31:14] *** Giaco has joined #opensolaris
[12:32:12] *** victori_ has quit IRC
[12:32:13] *** __hsilva has joined #opensolaris
[12:32:47] *** victori_ has joined #opensolaris
[12:36:28] *** dsch04 has quit IRC
[12:45:25] *** _hsilva has quit IRC
[12:45:26] *** __hsilva is now known as _hsilva
[12:48:41] *** cydork has quit IRC
[12:49:19] *** cydork has joined #opensolaris
[12:50:27] *** bengtf has quit IRC
[12:50:52] *** sparc-kly has joined #opensolaris
[12:50:53] *** ChanServ sets mode: +o sparc-kly
[12:52:16] *** victori_ has quit IRC
[12:53:21] *** sponix has quit IRC
[12:55:13] *** victori_ has joined #opensolaris
[13:05:50] *** jossh has quit IRC
[13:10:19] *** renihs has quit IRC
[13:11:50] *** sparc-kly_ has quit IRC
[13:13:01] *** victori_ has quit IRC
[13:13:12] *** bengtf has joined #opensolaris
[13:13:14] *** rasputnik has joined #opensolaris
[13:16:52] *** victori_ has joined #opensolaris
[13:18:27] *** Giaco has quit IRC
[13:18:54] *** dsch04_ is now known as dsch04
[13:18:59] <dsch04> Hi guys
[13:19:29] <dsch04> Anyone know if the Promise SATA150 TX4 controller is supported in OpenSolaris?
[13:19:39] *** dpn` has joined #opensolaris
[13:21:55] <quasi> dsch04: the HCL probably knows
[13:22:04] <dsch04> Got a link
[13:22:05] <dsch04> ?
[13:22:08] <dsch04> Please :)
[13:22:36] <cmihai> google
[13:22:53] <cmihai> http://www.google.com/search?client=opera&rls=en&q=solaris+hcl&sourceid=opera&ie=utf-8&oe=utf-8
[13:22:54] <dsch04> Bah.
[13:22:59] <dsch04> It seems not: http://www.opensolaris.org/jive/thread.jspa?messageID=151864
[13:23:04] <cmihai> duh
[13:23:05] <quasi> dsch04: bigadmin.com/HCL
[13:23:07] <cmihai> Promise is fake raid
[13:23:16] <dsch04> Don't want raid
[13:23:19] <dsch04> jbod
[13:23:22] <dsch04> using zfs
[13:23:26] <quasi> ehrm, bigadmin.com/hcl
[13:23:31] <cmihai> still junk :P
[13:23:44] <dsch04> I've been trying to install with a SAT2-MV8
[13:24:30] <dsch04> After a couple of days of messing around, I've RMA'd it with a bad port 3 (which caused Solaris to hang when booting from the install DVD)
[13:24:51] <dsch04> So, I thought I'd throw in a couple of those Promise cards I had kicking around
[13:24:58] <dsch04> Typical that it's not supported!
[13:25:24] <dsch04> OK, next option is: replace SATA DVD drive with IDE and put all 6 drives on the ICH9 controller on the mobo!
[13:26:26] <dsch04> One might say I've not had a smooth ride with this! :)
[13:26:45] <quasi> dsch04: I've heard others having issues with sata dvd
[13:26:54] <jmcp> you're on the bleeding edge there :)
[13:29:02] *** SYS64738 has quit IRC
[13:30:53] *** delewis has quit IRC
[13:31:04] *** delewis has joined #opensolaris
[13:34:08] *** jcea has joined #opensolaris
[13:35:19] *** victori_ has quit IRC
[13:35:25] *** jcea has left #opensolaris
[13:35:34] *** jcea has joined #opensolaris
[13:36:05] *** victori_ has joined #opensolaris
[13:36:52] <dsch04> quasi, The DVD seems to work fine
[13:37:07] <dsch04> I'm only replacing it with IDE so I can use all 6 SATA ports for hard drives.
[13:39:06] *** halton has left #opensolaris
[13:46:05] *** rasputnik has quit IRC
[13:46:40] *** SirFunk has quit IRC
[13:51:37] *** jossh has joined #opensolaris
[13:52:03] *** renihs has joined #opensolaris
[13:53:01] *** Downix has joined #opensolaris
[13:53:45] *** dlg has quit IRC
[13:57:04] *** karrotx has quit IRC
[14:06:28] <kjetilho> hey, any native English speakers here who can tell me -- what does "daemon" *really* mean?
[14:07:37] <JWheeler> I have confess, it's history is before my time
[14:07:52] <jmcp> it's an archaic (ie ooooooold) spelling of Demon
[14:08:14] <jmcp> the traditional usage tends to ignore the evil connotations of "demon"
[14:08:43] <JWheeler> http://en.wikipedia.org/wiki/Daemon_(computer_software)#Terminology
[14:09:00] *** spoown has joined #opensolaris
[14:10:44] *** nostoi has joined #opensolaris
[14:11:09] <trochej> kjetilho: Greek speakers would be better :)
[14:12:26] *** jossh has quit IRC
[14:12:37] <kjetilho> trochej: hehe
[14:13:10] *** friendly12345 has left #opensolaris
[14:13:23] <kjetilho> JWheeler: someone claims that every 16yo will know that "daemon" is good, and "demon" is evil.  from Dante and Milton, you know
[14:13:47] <kjetilho> I think it's a bit optimistic on behalf of the teenagers of our time.
[14:14:16] <trochej> kjetilho: If they pay attention to their history of literature classes :)
[14:14:34] <jmcp> I've heard of people wearing the BSDaemon tshirt in the south of the US, and getting run out of town for promoting devil worship
[14:15:30] <quasi> that doesn't really surprise me
[14:15:31] <tfb> Daemon in computing may well come from CTSS (so the jargon file claims, anyway) and probably derives more closely from the classical Greek usage (some kind of supernatural being who was half-way between a god and human)
[14:15:36] <jteo> that's a problem with the south of US, rather than the t-shirt.
[14:17:00] <kjetilho> this British person thinks it is ludicruous that "daemon" (the Unix type) is translated into "nisse" in Norwegian.
[14:17:03] <tfb> Also (again from Jargon file) poss from Maxwell's demon
[14:17:42] <jmcp> jteo: that's my understanding too
[14:19:18] <jteo> but i'll steer far away from stereotyping; stupidity is an international disease.
[14:19:31] <trochej> that a stereotype
[14:19:34] <trochej> :)
[14:21:25] *** victori_ has quit IRC
[14:22:07] *** victori_ has joined #opensolaris
[14:23:40] *** spoown has quit IRC
[14:24:56] *** SYS64738 has joined #opensolaris
[14:33:52] <SYS64738> may I use logadm to compress squid logs ?
[14:38:06] *** alanbur has joined #opensolaris
[14:45:44] *** edvil has joined #opensolaris
[14:48:51] *** sommerfeld has joined #opensolaris
[14:48:51] *** ChanServ sets mode: +o sommerfeld
[14:49:02] <cmihai> SYS64738, I suggest you take a look at http://unixsadm.blogspot.com/2007/10/rzip-great-compression-rates-for-logs.html . It's rzip, you get better compress rates (even 10 times better than gzip) for log files and such. I have a nice review / benchmark there.
[14:50:11] <cmihai> SYS64738, you can use it with logadm too, logadm can rotate any log you point it at.
[14:50:27] <Stric> I tried on some http logs we have.. 340MB for gzip, 200MB for bzip2, 170MB for rzip..
[14:50:47] <Stric> but rzip has some disadvantages.. like memory usage, no stdin/out operations, no *grep etc
[14:50:48] <cmihai> :-)
[14:50:59] <cmihai> Stric, I've covered though.... they're not so bad for it's role.
[14:51:12] <cmihai> SYS64738, just edit /etc/logadm.conf and add a line to the log dir.
[14:51:21] <cmihai> Where are your logs stored
[14:51:46] <cmihai>  Add something like:
[14:52:28] <cmihai>  /var/sadm/log/squid -C 10 - P 'insert timestamp here' -a pkill -HUP squid
[14:52:42] <cmihai> Sometimes you need to send SIGHUP to the process.
[14:53:30] *** victori_ has quit IRC
[14:53:46] <cmihai> Stric, it's great if you have to store logs on write once media (UDO, DVD-R, Blu Ray BD-R and such).
[14:54:27] <cmihai> You don't need pipes or anything for that. And memory is quite cheap... a 500-900MB mem usage spike every log arhive isn't that big a deal.
[14:56:32] *** victori_ has joined #opensolaris
[14:57:38] *** Dar has quit IRC
[14:59:54] *** tsoome1_ has quit IRC
[15:00:43] <SYS64738> thanks cmihai
[15:01:45] *** cga has joined #opensolaris
[15:05:51] <SYS64738> cmihai, can I do a squid -k rotate with the -a option ?
[15:06:08] <cmihai> I guess...
[15:06:24] <cmihai> It's a post-rotate command
[15:06:33] <cmihai> once it executes the rotate, it runs that... but why?
[15:07:01] *** alanbur has left #opensolaris
[15:07:57] *** dennis has left #opensolaris
[15:07:59] <SYS64738> fot the term sign
[15:08:00] <SYS64738> for
[15:09:45] <cmihai> You can send squid SIGUSR1
[15:09:53] <cmihai> It closes and renames files
[15:10:00] <cmihai> Just use cron
[15:10:01] <SYS64738> I would like to something this:
[15:10:02] <SYS64738> squid -C 60 -a '/opt/local/squid/sbin/squid -k rotate; sleep 2; rm -f /logs/squi
[15:10:03] <SYS64738> d/*.log.0' -c -p 1d -t '$file.%Y%m%d' -z 0 '/logs/squid/{access,cache,store}.log
[15:11:56] *** victori_ has quit IRC
[15:12:17] <cmihai> Heh, funniest thing I've seen in UNIX
[15:12:22] <cmihai> rm -rf $HOME as root!
[15:13:07] <Teltariat> cmihai: No man, I've got one that takes the cake:
[15:13:15] <cmihai> Some idiot did it as root on machine... something like $HOME or $HOME/* :-). As root's home is / ....
[15:13:21] <Teltariat> It was a Perl script that, as part of its testing procedure, deleted root
[15:13:27] <cmihai> The dimwit deleted everything ;-\
[15:13:34] <cmihai> Teltariat, what did it test? Backup and recovery tapes?
[15:13:45] <Teltariat> cmihai: Let me find the link
[15:13:57] <jteo> maybe it tested sysadmin blood pressure.
[15:14:24] <Teltariat> cmihai: http://use.perl.org/~Alias/journal/34680
[15:14:48] <Teltariat> During CPAN installation
[15:15:01] <Teltariat> a newly installed module is built, and then tested, before it is installed
[15:15:08] <Teltariat> during this one's testing phase...
[15:15:16] <cmihai> Sweet
[15:16:28] <jteo> mmm. new n810 tablet.
[15:17:13] <Teltariat> "Hey, Junior SysAdmin, did you install the required Perl modules for this software package on our only most important sole production server that has no backups?"       Jr.: "Well....about that..... we need to talk.... "
[15:18:08] <kjetilho> never build and install software as root ...
[15:18:12] <cmihai> Man, that's stupidity to a whole new leve.
[15:18:23] <cmihai> You don't need root to build software or test it.
[15:18:36] <cmihai> Actually, guess what
[15:18:45] <cmihai> production systems and development system and testing systems DON'T mix.
[15:19:00] <cmihai> Don't let your devels run stuff on your production machine, or change things on the fly.'
[15:19:12] <jteo> given the existance of cheap virtualization like zones, why would you even test on your live machine?
[15:19:18] <cmihai> Even if you only have one machine, you could test in a Container or qemu
[15:19:23] <LeftWing> Or, in many cases, develop software at all.
[15:19:38] <LeftWing> Bloody web developers. :)
[15:19:50] <cmihai> Web devels?
[15:19:51] <jteo> define web-developers: l33t HTML code?
[15:19:52] <cmihai> No such thing
[15:19:54] <jteo> ;)
[15:19:57] <cmihai> 14 year old script kitties :P
[15:20:20] *** nostoi has quit IRC
[15:20:21] <Gekz> kitties?
[15:20:26] <LeftWing> Well web development done in a proper language is, as much as it makes me squirm, still coding at least.
[15:20:27] <jteo> i wonder who wrote the masterpiece known as the Storm worm -wistful admiration-
[15:20:32] <cmihai> Really, I see a person over 18 still doing web development / java script / AJAX or PHP for a living...
[15:20:41] <cmihai> I know he fucked up bad.
[15:20:42] <Gekz> www.icanhascheezburger.com
[15:21:10] <LeftWing> DO NOT WANT.
[15:22:07] *** Triskelios has joined #opensolaris
[15:22:32] <jteo> ?
[15:23:34] <Tempt> haha.
[15:23:53] <LeftWing> Nothing brings you out like a Macro. =P
[15:24:00] <Tempt> That's the one.
[15:24:07] <Tempt> Back in a sane place now.
[15:24:16] <LeftWing> Ah, can has Melbourne locale?
[15:24:21] <Berny_> hey Tempt
[15:24:31] <Berny_> how's the holiday been?
[15:24:45] <Tempt> What holiday?
[15:24:49] *** victori_ has joined #opensolaris
[15:25:01] <jteo> ...
[15:25:04] <Tempt> Working in Sydney != a holiday
[15:25:08] <Berny_> didn't you say something about a week off before new job starts?
[15:25:18] <jteo> depends on definition of work.
[15:25:21] <Tempt> Berny_: More like a couple of days in fact.
[15:25:32] <Berny_> oh well better than nothing ;-)
[15:25:39] * Berny_ has a day off next week
[15:26:10] <Berny_> and still 21 days of holiday to burn for this year :-\
[15:27:16] <jteo> Berny_: you say that like it is a bad thing.
[15:28:01] <Berny_> it is... cause i caanot take that many days off until end of year
[15:28:10] <Berny_> and i cannot trade them for money :-\
[15:28:25] *** bengtf has quit IRC
[15:28:37] <jteo> Berny_: tis a common situation.
[15:30:03] <Berny_> true... still annoying
[15:33:05] <jteo> it's sobering to realize it's cheaper for me to order a tonne of books from Amazon.com then to buy them locally.
[15:34:28] <Berny_> saves you a walk :-)
[15:36:20] <jteo> true. more of a drive really.
[15:36:30] <SYS64738> cmihai, could you make me an example for the timestamp to put after the -P (I would like to rotate every 24h at 0.00) ?
[15:36:31] *** nicoAMG has joined #OpenSolaris
[15:39:02] <cmihai> SYS64738, http://www.sun.com/bigadmin/content/submitted/solaris_logadm.html
[15:39:14] <SYS64738> thanks
[15:40:58] <SYS64738> I don't understand who has set the timestamp in the lines in logadm.conf
[15:41:36] *** victori_ has quit IRC
[15:42:24] *** victori_ has joined #opensolaris
[15:42:48] *** calumb is now known as calAFK
[15:43:19] <SYS64738> that page is useless
[15:45:48] <SYS64738> bye
[15:45:58] <cmihai> So read logadm(1M)
[15:46:32] <cmihai> It has examples...
[15:46:49] <Downix> has anyone ever had the screen refuse to restore after a day of no activity?  I can't get the local display to work, but I can access the machine remotely
[15:47:24] <cmihai> Restart cde-login
[15:47:35] <cmihai> pkill xscreensaver
[15:47:36] <cmihai> etc
[15:47:39] <cmihai> see what it's doing
[15:47:56] *** cub- has joined #opensolaris
[15:48:12] <Downix> that worked, thanks
[15:48:21] <Downix> (the xscreensaver bit)
[15:49:31] <cmihai> You're welcome :-)\
[15:50:27] <cub-> hmm...what's the command for sun update again?
[15:50:34] <cub-> patchupdate or something
[15:51:29] <Berny_> smpatch, updatemanager, patchadd
[15:51:52] <cub-> ahh, updatemanager
[15:52:23] <cub-> now they all require a contract services to work , right ?
[15:52:26] <quasi> smpatch
[15:52:39] *** bengtf has joined #opensolaris
[15:54:47] *** calAFK is now known as calumb
[15:55:46] *** Dar has joined #opensolaris
[15:55:58] *** csjp has quit IRC
[16:01:16] *** Gropi has joined #opensolaris
[16:07:43] *** victori_ has quit IRC
[16:08:15] *** victori_ has joined #opensolaris
[16:10:19] *** bowdengl has joined #opensolaris
[16:11:08] <cmihai>  zpool list
[16:11:08] <cmihai> NAME                    SIZE    USED   AVAIL    CAP  HEALTH     ALTROOT
[16:11:08] <cmihai> storage                    -       -       -      -  FAULTED    -
[16:11:11] <cmihai> Well, this is fun.
[16:11:20] <cub-> uh oh
[16:11:23] <cmihai> Yep.
[16:11:30] <cmihai> Management wouldn't buy more disks.
[16:11:35] <cmihai> They halved the number of disks I've oredered.
[16:11:41] <cmihai> They said they're "too expensive"
[16:11:53] <cub-> ???
[16:12:10] <Berny_> go for usb sticks :-P
[16:12:17] <cmihai> they got me half the number of disks I've ordered, so I couldn't do RAID-6.
[16:12:22] <cub-> shutdown production servers and take a vacation
[16:12:23] <cmihai> erm RAIDZ2
[16:12:30] <cmihai> cub-, yep.
[16:12:37] <cmihai> Ah well.
[16:12:41] <cmihai> Hooray for data loss.
[16:13:19] <Teltariat> cub-: that'll be a _PERMANENT_ vacation. :-|
[16:15:00] <cub-> doesn't matter anyway....since if data loss happens, he's fired anyway
[16:15:15] <jteo> unless he blames someone else
[16:15:31] <cub-> naw...he won't win the management
[16:16:00] <Teltariat> Its about time we link to that "Problem? Blame someone else" flowchart.
[16:16:46] <cub-> but for the long run, i forsee no bright future for you... where do they get the money for your raises/promotion if they can't afford the disks for servers.  I think it's time to update your resume
[16:17:15] *** mog has joined #opensolaris
[16:17:21] <cmihai> Not my fault :-)
[16:17:48] <cmihai> I didn't get the hardware I needed, I bitched about it every day for 6 months, shit happens, it's the data owner's fault :-).
[16:17:51] <cmihai> As in, management.
[16:17:54] <Teltariat> I swear, I don't have a heart strong enough for that.  If I were in cmihai's place, and saw 'FAULTED' when trying to list my data pools, I might have lost consciousness.....
[16:18:10] <cmihai> Teltariat, want to know the fun part?
[16:18:14] <Teltariat> uh oh
[16:18:17] <Teltariat> do tell.
[16:18:22] <cmihai> It was the VMware quorum device :-)
[16:18:31] <cmihai> Nothing like all your machines going down :D
[16:18:31] <Teltariat> Quorum?
[16:18:32] *** rcorreia has joined #opensolaris
[16:18:39] <cmihai> erm
[16:18:41] <cmihai> shared storage
[16:18:45] <Teltariat> oh shi
[16:18:53] <cmihai> Yep.
[16:19:07] <cmihai> So basically every node in the VMware infrastructure cluster is useles :-)
[16:19:20] <Teltariat> cmihai: time to hang a hammock from the rack doors?  Might as well, everythings fucked, right?
[16:19:43] <jteo> true.
[16:20:02] <cmihai> I'm mildly optimistic, I keep a few "all in one" bsd servers for emergencies :-).
[16:21:14] <Teltariat> And whats an all-in-one BSD server gonna do in this case?  You running a secret version of FreeBSD that comes with a 'magical-zpool-recovery-fairy' utility?
[16:21:15] *** laca has joined #opensolaris
[16:21:19] <cub-> managers are usually better in the finger pointing business ... the better ones at it hold higher management positions
[16:21:24] <Teltariat> Oh man, I'd die.
[16:21:26] <jteo> cub-: agreed
[16:21:52] <cmihai> cub-, here, I'm covered in signed papers :-)
[16:21:56] *** victori_ has quit IRC
[16:22:26] <cub-> keep all your email trails about your proposal for buying the disks...that's the best way to protect yourself
[16:22:47] <cub-> and make everything in writing/emails as well as CC a few other co-workers
[16:23:05] *** victori_ has joined #opensolaris
[16:23:15] <jteo> ah...black and white.
[16:23:16] <Teltariat> $ cat archives/email-evidence.1
[16:23:19] <Berny_> you gotta bcc the right people
[16:23:30] <Teltariat> "> YOU FUCKING IDIOTS!  YOU NEED TO GET US SOME MORE DRIVES!"
[16:23:46] <Teltariat> "...."
[16:23:50] <Teltariat> "Was I eloquent enough?"
[16:24:02] *** stukreit has quit IRC
[16:24:17] <cmihai> Teltariat, that's pretty much how I've put it :-)
[16:24:30] <Teltariat> cmihai: I don't doubt that :D
[16:24:44] <cmihai> I made sure to bring it up at every meeting too :-)
[16:24:47] <Berny_> oh well you still have a backup at hand
[16:25:14] * Teltariat snorts
[16:25:17] <cmihai> Yeah, sort of. But I can't bring the system up for... dunno, 5-6 days
[16:25:29] <Berny_> fair enough
[16:25:32] <cmihai> Make that 3 weeks... knowing how long it takes to buy drives around here
[16:25:35] <Berny_> that will teach them
[16:25:36] <cub-> yeah, some things dont' just work with a restore from backup
[16:27:01] * Berny_ had an incident few years ago... 5 disk softraid with samfs on top (been begging for hardware raid with hotspare for months) lost 2 disks in a power failure...
[16:27:08] <cmihai> Actually, I'm kind of sad I planned ahea
[16:27:09] <cmihai> d
[16:27:42] *** bowdengl has left #opensolaris
[16:27:44] <Berny_> had to restore the samfs stuff from the archiver logs... walk through like 60 tapes at the time
[16:27:51] <Berny_> that did take a while
[16:27:54] <cmihai> And only put non-critical services... like management consoles, SNMP collectors, log servers, graph servers, update servers and such. Nothing that's needed 24/7
[16:28:27] *** cygnusecks has quit IRC
[16:30:20] *** phimic has quit IRC
[16:30:35] *** stukreit has joined #opensolaris
[16:33:26] *** stukreit has quit IRC
[16:34:13] *** victori_ has quit IRC
[16:35:42] *** tfb has left #opensolaris
[16:36:30] *** tsoome has joined #opensolaris
[16:39:17] *** Pietro_S has joined #opensolaris
[16:44:40] *** cmihai has quit IRC
[16:45:36] *** blindfish has joined #opensolaris
[16:47:06] *** szt has quit IRC
[16:49:51] *** charlesnw has joined #opensolaris
[16:49:52] *** tsoome1 has joined #opensolaris
[16:49:55] *** victori_ has joined #opensolaris
[16:50:51] *** hsn_ has joined #opensolaris
[16:51:12] *** GmanAFK has quit IRC
[16:52:40] *** tsoome1_ has joined #opensolaris
[16:54:05] *** Gman has joined #opensolaris
[16:54:06] *** ChanServ sets mode: +o Gman
[16:54:07] *** Giahomo has joined #opensolaris
[16:56:08] *** Fullmoon has joined #opensolaris
[16:56:27] *** Fullmoon has quit IRC
[16:57:05] *** laca is now known as lacaAFK
[16:59:20] *** tsoome has quit IRC
[17:02:42] *** sommerfeld has quit IRC
[17:03:01] *** stukreit has joined #opensolaris
[17:03:38] *** gobbler has joined #opensolaris
[17:04:44] <Giahomo> what's the right syntax for logadm.conf to make rotate a file each day at midnight ?
[17:05:05] <Giahomo> I don't understand the right function for -P
[17:05:22] <gobbler> does shmget() reserve swap? and if yes, what happens when fork() is called - does the new process have to allocate swap for the shm region as well?
[17:05:52] *** glagasse has joined #opensolaris
[17:10:56] *** sommerfeld has joined #opensolaris
[17:10:57] *** ChanServ sets mode: +o sommerfeld
[17:11:17] *** tsoome1 has quit IRC
[17:14:03] *** legolas has joined #opensolaris
[17:14:38] <legolas> Hi
[17:14:51] <legolas> I installed SXDE and i have some problems with it.
[17:15:33] <legolas> when it boot up (Console based login prompt come) it took 5 minutes to boot into GUI
[17:17:19] *** Fullmoon has joined #opensolaris
[17:17:21] <legolas> Another question is about boot loader, I tried to use fdisk to find which partition is my windows to add the boot loader changes and fdisk did not work
[17:17:35] *** victori_ has quit IRC
[17:17:54] <Pietro_S> legolas: that's normal, cause at first start OpenSolaris read all smf profiles and initialize them
[17:19:07] <legolas> I tried and add  rootnoverify (hd1,0) 	chainloader +1 to make it load my windows and it did not.
[17:20:25] <legolas> Another problem is about Network, I want to change IP address and Gateway and DNS of my connection myself but when i run Network application from start menu it said that some automatic thingy is running and It can not configure the network.
[17:20:47] <Pietro_S> (hd1, 0) is second hardisk
[17:22:30] <legolas> How does it count hard disks? Does the disk that Solaris itself is installed is hd0 and the other one is hd1  and so on or it is based on BIOS?
[17:24:21] *** victori_ has joined #opensolaris
[17:28:11] <Triskelios> legolas: svcadm disable nwam if you want manual config
[17:28:27] *** calumb has quit IRC
[17:28:56] <Triskelios> legolas: for grub, it's based on the BIOS (there is no other source of information)
[17:32:16] *** dlynes_laptop has joined #opensolaris
[17:33:02] *** sommerfeld has quit IRC
[17:34:35] <legolas> Thank you all for your comments.
[17:34:45] *** peteh has quit IRC
[17:35:30] *** Tempt770 has joined #opensolaris
[17:35:41] <WickedWicky> it's Tempt770!
[17:35:58] <Tempt770> on my nokia n770
[17:36:03] <WickedWicky> w00t
[17:36:17] <timeless> um, s/n770/770/
[17:36:23] <Tempt770> nifty toy
[17:36:24] <timeless> the n was added for the n800 and n810
[17:36:29] <timeless> it's just nokia 770 :(
[17:36:47] <Tempt770> oh, details, details
[17:36:51] <WickedWicky> lol
[17:36:56] <WickedWicky> it's nifty
[17:36:59] <WickedWicky> with capital N
[17:37:08] <timeless> which os are you running?
[17:37:20] <Tempt770> the 800x480 screen is great
[17:37:41] <timeless> http://browser.garage.maemo.org/news/8/
[17:37:49] <trygvis> opensolaris? :)
[17:37:56] <Tempt770> still running tthe shipped os
[17:37:58] <timeless> trygvis: doesn't support arm
[17:38:03] *** sarahj has joined #opensolaris
[17:38:06] <trygvis> file a bug!
[17:38:20] <timeless> did anyone get bugzilla accounts for osol?
[17:38:21] *** tenex has joined #opensolaris
[17:38:26] <timeless> i was waiting for mine
[17:38:37] <legolas> Is it the correct way to use rootnoverify (hd1,0) 	chainloader +1 in order to add windows to solaris boot loader?
[17:38:39] <timeless> tempt770: i think they change the firmware over time
[17:38:51] <timeless> you could be running 3.2006.49-2 or something older
[17:38:54] <timeless> (see url)
[17:39:04] <Tempt770> fair enough ...
[17:39:46] *** joevandyk has joined #opensolaris
[17:40:00] <joevandyk> anyone got any tips on tuning opensolaris for high performance web serving?  i'm hitting the open file maximum.
[17:40:03] <Tempt770> anyway, enough of this for tonight.
[17:40:15] <Tempt770> sleep calls
[17:40:16] <joevandyk> (when stress testing)
[17:40:38] *** Tempt770 has quit IRC
[17:40:41] <WickedWicky> night night!
[17:42:52] *** Giaco has joined #opensolaris
[17:42:52] *** SYS64738 has quit IRC
[17:43:37] *** victori_ has quit IRC
[17:45:52] *** salamanders has quit IRC
[17:50:22] *** dlynes_laptop has quit IRC
[17:50:38] *** axisys has quit IRC
[17:51:12] *** mega has quit IRC
[17:52:33] *** pjlv has joined #opensolaris
[17:54:12] *** Gman has quit IRC
[17:54:24] *** gobbler has quit IRC
[17:55:39] *** sioraiocht has joined #opensolaris
[17:55:45] *** RaD|Tz has joined #opensolaris
[17:57:58] *** johnniez has joined #opensolaris
[18:00:58] *** Cyrille has joined #opensolaris
[18:02:24] *** legolas has quit IRC
[18:02:38] *** victori_ has joined #opensolaris
[18:07:29] *** catena has joined #opensolaris
[18:07:42] *** nostoi has joined #opensolaris
[18:08:06] *** victori_ has quit IRC
[18:09:20] *** plocher has joined #opensolaris
[18:10:25] *** stevel has joined #opensolaris
[18:10:25] *** ChanServ sets mode: +o stevel
[18:10:58] *** Snake007uk has quit IRC
[18:11:01] *** cmihai has joined #OpenSolaris
[18:11:52] *** sahafeez has joined #opensolaris
[18:13:26] *** cub- has quit IRC
[18:15:14] *** MattMan has quit IRC
[18:16:30] *** virgee has joined #opensolaris
[18:19:18] *** catena has quit IRC
[18:20:23] *** sioraiocht has quit IRC
[18:21:44] *** vmlemon has joined #opensolaris
[18:21:50] *** dnilsson has joined #opensolaris
[18:22:10] *** nachox has joined #opensolaris
[18:23:06] * JoergB is away: for the night
[18:24:17] *** victori_ has joined #opensolaris
[18:24:39] *** Gman has joined #opensolaris
[18:24:39] *** ChanServ sets mode: +o Gman
[18:25:58] *** _Megaf has joined #OpenSolaris
[18:30:32] *** Cyrille has quit IRC
[18:31:22] <claws_bugs> what is that thing that sees if my hardware is solaris friednly?
[18:32:46] *** dmesg has joined #opensolaris
[18:32:53] <dmesg> hi dudes and dudettes too
[18:33:08] <cmihai> claws_bugs, HCL
[18:33:21] <dmesg> is posible to make a sun ray thinclient work with another OS than solaris?
[18:33:23] <cmihai> claws_bugs, http://www.sun.com/bigadmin/hcl/
[18:33:51] <dmesg> like linux but with out the srss
[18:33:53] <cmihai> claws_bugs, http://www.sun.com/bigadmin/hcl/hcts/device_detect.html is automagic.
[18:34:20] *** sommerfeld has joined #opensolaris
[18:34:20] *** ChanServ sets mode: +o sommerfeld
[18:34:27] *** sriramnrn__ has joined #opensolaris
[18:34:30] *** sriramnrn__ is now known as sriramnrn
[18:36:34] *** victori_ has quit IRC
[18:36:57] *** tek-ops has joined #opensolaris
[18:37:04] <tek-ops> I have a solaris 8 routing quest
[18:37:09] <tek-ops> s/quest/question
[18:37:19] *** sriramnrn_ has quit IRC
[18:37:32] *** victori_ has joined #opensolaris
[18:37:37] <tek-ops> I know that I place my default gateway in /etc/defaultrouter
[18:38:05] <tek-ops> however, I want to add static routing entries
[18:38:22] <tek-ops> I believe these go in /etc/gateways, but I'm not sure if that holds true for solaris 8
[18:38:27] <Giahomo> cmihai, I am still playing with logadm, but I don't understand how to tell him to make a job at specific time
[18:38:40] <cmihai> Giahomo, cron mate
[18:38:41] <cmihai> crontab
[18:38:42] <cmihai> :-)
[18:39:36] <tek-ops> I've successfully added  the static routes with "route add" but I need them to be retained after reboots
[18:41:13] *** timsf has joined #opensolaris
[18:41:25] <timsf> hi folks
[18:42:26] <Pietro_S> is it HAL allowed to shutdown/restart computer?
[18:42:27] <tek-ops> is this not the place to ask solaris 8 questions
[18:43:05] *** _Megaf has quit IRC
[18:43:13] <tenex> tek-ops: #solaris...
[18:43:40] <tek-ops> thank tenex
[18:47:05] *** stukreit_ has joined #opensolaris
[18:47:25] *** cygnusecks has joined #opensolaris
[18:47:56] *** catena has joined #opensolaris
[18:49:28] <Pietro_S> hmm, looks like HAL doesn't have such permisions
[18:50:31] <tek-ops> I'm nervous by the responses I'm getting in #solaris
[18:50:38] <Pietro_S> doesn't anyone know how the shutdown button was done by JDS? I even can't find that source
[18:50:43] *** estibi has joined #opensolaris
[18:50:44] <tek-ops> has anyone here dealt with the /etc/gateways file
[18:50:49] <flyingparchment> i finally found a reason ubuntu is much awesomer than solaris: it has progress bars that go backwards
[18:51:12] *** jhawk has joined #opensolaris
[18:51:17] <Giahomo> ahhhhhh logadm without -w
[18:51:28] <Giahomo> e ditelo cazzo
[18:51:35] *** mikefut has quit IRC
[18:54:28] *** Rivelli has joined #opensolaris
[18:54:35] <stukreit_> dmesg:  Sunray units (called "DTU"'s) require srss, which is currently available for Linux. We are also making an Xorg/Solaris/sparc,x86 version now.
[18:55:03] *** jhawk has quit IRC
[18:56:51] <RElling> tek-ops: I dealt with an /etc/gateways file once... once
[18:57:28] <tek-ops> relling is it intended for storing your static routes, and I have presumed?
[18:57:32] <tek-ops> I'm missing the man page for it
[18:58:20] *** stukreit_ has left #opensolaris
[18:58:26] <claws_bugs> does init 6 hurt stuff?
[18:58:35] <claws_bugs> is it same as shutdown -r now
[18:58:37] <claws_bugs> ???
[18:58:44] <claws_bugs> someone told me you can lose data with init
[18:59:20] <RElling> tek-ops: yes, it has been around since BSD 4.2 or so.  You should find the man page online
[18:59:33] <tek-ops> ok great, thank you
[18:59:38] *** Giahomo has quit IRC
[19:01:07] *** vmlemon_ has joined #opensolaris
[19:01:08] *** szt has joined #opensolaris
[19:02:12] *** stukreit has quit IRC
[19:02:25] <Pietro_S> claws_bugs: using init 6 is safe in ON concept
[19:02:49] *** vmlemon has quit IRC
[19:02:50] <Pietro_S> if application is wrongly written than you can lose some data
[19:02:55] *** vmlemon_ is now known as vmlemon
[19:03:14] <Pietro_S> but init won't make mess in your disks ..
[19:03:36] *** yippi has joined #opensolaris
[19:03:53] *** JBeck has joined #opensolaris
[19:03:54] *** ChanServ sets mode: +o JBeck
[19:05:32] <flyingparchment> root@vandale:/var/svc/manifest/application# svccfg import jira.xml
[19:05:33] <flyingparchment> svccfg: couldn't parse document
[19:05:36] <flyingparchment> this error sucks ---^
[19:07:42] *** NikolaVeber_ has joined #opensolaris
[19:07:43] <RElling>  XML error?
[19:07:52] <trygvis> try svccfg validate
[19:08:02] <flyingparchment> i found the problem, but the error message still sucks :)
[19:08:19] *** tek-ops has quit IRC
[19:08:30] <jteo> true.
[19:08:31] <RElling> probably comes from libxml... or something similar
[19:10:46] <WickedWicky> vandale? your server is our national dictionary, or?
[19:10:49] *** theRealBallchalk has quit IRC
[19:11:02] <flyingparchment> WickedWicky: our servers are named after encyclopedists and similar
[19:11:09] <WickedWicky> Ahhhhhhhh
[19:11:20] <WickedWicky> figures :D
[19:11:23] <jteo> that's an interesting naming scheme.
[19:11:30] <flyingparchment> jteo: we run an encyclopedia
[19:11:45] <jteo> flyingparchment: ahhhhh.
[19:11:48] <Downix> hello
[19:12:26] <flyingparchment> that particular server is in holland, so it got a dutch name ;)
[19:12:39] <WickedWicky> nice :D
[19:13:39] * WickedWicky just received an -EGF_NEEDS_ATTENTION
[19:13:40] <WickedWicky> so
[19:13:46] <WickedWicky> I better solve that error
[19:14:07] *** _Megaf has joined #OpenSolaris
[19:14:09] <WickedWicky> which means I'll nod, smile and hear about her day at work and all :P
[19:14:21] * WickedWicky runs of
[19:14:53] <jteo> wb Downix.
[19:14:56] <dmesg> stukreit_ yes, but is very expensive :/ the srss
[19:15:14] <Downix> It appears in my solaris install I messed up the partitioning
[19:15:22] <Downix> went with my usual Linux partition setup
[19:16:46] *** catena has quit IRC
[19:17:29] <Downix> any method to fix or is it a reinstall?
[19:19:17] <flyingparchment> Downix: how exactly did you mess it up?
[19:19:35] *** deather_ has joined #opensolaris
[19:19:47] *** Timsen has quit IRC
[19:20:06] <Downix> flyingparchment, did not realize that solaris utilizes the /opt directory so heavily, so used a small / with larger /usr and /var
[19:20:16] <flyingparchment> Downix: got any spare space?
[19:21:15] *** Timsen has joined #opensolaris
[19:21:43] <Downix> flyingparchment, On /? Not really, installed the webserver and *poof* there it went
[19:21:44] *** lasseoe has quit IRC
[19:21:46] *** lasseoe has joined #opensolaris
[19:22:25] *** sriramnrn__ has joined #opensolaris
[19:22:48] <flyingparchment> Downix: no, anywhere
[19:23:00] <Downix> plenty under /var
[19:23:06] <Downix> it's a whole dedicated 160GB HD
[19:23:17] *** sahafeez has quit IRC
[19:23:32] <Downix> which I was considering just migrating everything to 1 hard drive and RAID'ing them
[19:25:45] <sickness> I'm back
[19:27:05] *** NikolaVeber has quit IRC
[19:27:12] *** loky has joined #opensolaris
[19:27:23] <flyingparchment> Downix: hmm.. /var isn't so good
[19:27:28] <flyingparchment> Downix: you really need a spare, unused partition
[19:27:45] <flyingparchment> Downix: then you can use liveupgrade to copy your OS into that partition, and (if you want) repartition the rest and copy it back
[19:28:02] *** aka_druid has quit IRC
[19:28:43] *** aka_druid has joined #opensolaris
[19:30:43] <Downix> let me see
[19:33:29] <Downix> hmm, /export/home is pretty empty too
[19:33:33] *** sriramnrn___ has joined #opensolaris
[19:33:40] *** sriramnrn has quit IRC
[19:33:45] *** sriramnrn___ is now known as sriramnrn
[19:33:52] <elektronkind> would anyone here be using the PHP DTrace provider
[19:33:53] <elektronkind> ?
[19:34:03] <flyingparchment> elektronkind: i used it when it was at like v0.1 :)
[19:34:15] <Downix> flyingparchment, How do I live upgrade?
[19:34:21] <elektronkind> flyingparchment: but not currently I take it? :)
[19:34:34] <flyingparchment> Downix: check the manual at docs.sun.com.  basically, you create a new boot environment.  that copies your existing OS to a new slice.
[19:34:48] <flyingparchment> Downix: then you just leave that as it is, without upgrading.  (despite the name, it doesn't force an actual upgrade)
[19:34:54] *** sarahj has quit IRC
[19:35:01] <flyingparchment> elektronkind: nope
[19:35:19] <flyingparchment> elektronkind: maybe i will though.. just set up a php server on solaris, would be interesting to instrument our app
[19:35:42] <Downix> ok
[19:35:46] <Downix> I'll be doing this for a bit
[19:36:08] * flyingparchment `pecl install dtrace`
[19:36:51] *** deather has quit IRC
[19:38:51] <elektronkind> I'm having big problems although the dtrace pecl module is installed and phpinfo() says its there... none of the probes seem to be firing though :/
[19:38:56] <joevandyk> is the refresh SMF option automatically defined to send a -HUP signal to the process?
[19:39:05] <joevandyk> or do I need to explicitly put that in the SMF?
[19:39:26] <Downix> you know flyingparchment, I don't have anything on here yet other than the webserver (no sites), it likely would be easier to reinstall
[19:39:45] *** Mdx4 has joined #opensolaris
[19:40:28] <flyingparchment> Downix: yes, quite possibly
[19:40:33] *** sahafeez has joined #opensolaris
[19:40:46] <flyingparchment> Downix: in general, create a completely separate partition for your data - that way you can reinstall, upgrade, etc. the OS without affecting your apps/data/services
[19:40:52] <Downix> *nods*
[19:41:04] <Downix> Under my linux install, all data wound up under /var
[19:41:09] <flyingparchment> elektronkind: hmm, i just installed it, added 'extension=dtrace.so' and it's working great
[19:41:16] <Downix> but here, it seems to be more... complicated
[19:41:26] <flyingparchment> well, you can put it all under /var in solaris if you like
[19:41:29] <flyingparchment> but it sucks ;)
[19:41:42] <Downix> never made sence to me
[19:41:55] <Downix> but then if I reinstall I can set up the two drives to RAID each other
[19:42:04] <Downix> providing I can figure out how to do it
[19:42:43] <flyingparchment> is using /copyinstr(arg0) == "foo"/ as a dtrace condition a bad idea?
[19:42:49] *** axisys_ has joined #opensolaris
[19:42:52] *** sahafeez has quit IRC
[19:43:00] <joevandyk> What does an etry of *LK* mean in /etc/shadow?
[19:43:09] <seanmcg> lock
[19:43:10] <flyingparchment> joevandyk: account is locked and cannot be logged into
[19:43:13] <joevandyk> sweet
[19:43:14] <joevandyk> thanks
[19:43:22] <flyingparchment> joevandyk: as opposed to NP, which means no password, but can login using other authentication methods
[19:43:26] <flyingparchment> (like ssh public keys or kerberos)
[19:44:54] <flyingparchment> hm: dtrace: error on enabled probe ID 2 (ID 5779: php13801:dtrace.so:php_dtrace_execute:function-entry): invalid address (0xff) in predicate at DIF offset 28
[19:46:33] *** axisys has joined #opensolaris
[19:47:49] *** sriramnrn__ has quit IRC
[19:49:15] *** sriramnrn___ has joined #opensolaris
[19:49:53] *** sriramnrn___ has quit IRC
[19:50:26] *** polk__ has quit IRC
[19:51:21] *** dynamicproxy has joined #opensolaris
[19:51:39] *** polk__ has joined #opensolaris
[19:53:27] *** axisys_ has quit IRC
[19:58:05] *** nahamu has left #opensolaris
[19:58:13] *** eclix_ is now known as eclix
[19:58:29] <Downix> now, to figure out how the heck to set up a RAID in this machine
[19:58:37] <flyingparchment> what controller?
[19:58:39] <dmesg> how can i update gcc with out pkg-get?
[19:58:59] *** lacaAFK is now known as laca
[19:59:29] <Downix> the onboard one for a Dell PRecision 690
[19:59:47] <flyingparchment> yuck
[19:59:56] <flyingparchment> dell raid controllers -> pain
[19:59:58] <dmesg> :S
[20:00:26] <Downix> that's what I heared
[20:01:36] *** nicoAMG has quit IRC
[20:01:41] *** wms_ has joined #opensolaris
[20:01:47] <Downix> any suggestions on how to utilize two 160GB drives in a solaris setup then?
[20:01:56] <flyingparchment> SVM
[20:03:00] <Downix> ko
[20:03:21] <Triskelios> or just make a mirrored zpool
[20:03:30] *** yippi has quit IRC
[20:03:33] *** Dar has quit IRC
[20:03:44] *** sriramnrn has quit IRC
[20:04:26] *** TrogL has joined #opensolaris
[20:04:49] *** natefoo has joined #opensolaris
[20:05:37] <TrogL> I'm having a wierd problem with swap on Open Solaris.  Basically it won't swap.  I'm testing ORACLE, hence there's been a certain amoujnt of kernel tweaking, but my own code won't swap either.
[20:05:38] <natefoo> anyone know what the proper location is for site-specific smf methods is?
[20:06:56] <TrogL> I'm only allowed to consume about 3861 Mb of memory.  There's plenty more on the box.
[20:06:59] <TrogL> the code:
[20:07:03] *** ZeleniZub__ has joined #opensolaris
[20:07:14] <PerterB> TrogL: 32bit app by any chance?
[20:07:24] <TrogL> hmmm
[20:07:37] <TrogL> compiled using Sun Studio, doens't it default to 64bit?
[20:07:53] <PerterB> nope
[20:08:41] <PerterB> but running file(1) against your executable will give you the definitive answer
[20:09:11] <TrogL> I'm doing a loop of 10 int *ptr=malloc(100000 * 1000 * sizeof (long)); that works
[20:09:16] *** claws_bugs has left #opensolaris
[20:09:18] <TrogL> 11 loops fails
[20:09:31] <TrogL> ELF 32bit, you're right
[20:09:40] *** Gropi has quit IRC
[20:10:20] <natefoo> err, s/is//
[20:10:54] *** cydork has quit IRC
[20:10:55] *** RaD|Tz has quit IRC
[20:10:55] <TrogL> More hints.  I'm using ZFS.  I'm running top, it's saying 8185M phys, 230M free, 8193 free swap.
[20:11:00] <TrogL> no mattter what i do, I never use swap
[20:11:13] *** hile_ has quit IRC
[20:11:17] <TrogL> ORACLE bombs saying out of memory, but it never uses swap either
[20:11:36] <sommerfeld> so, swap space is being reserved but not actually allocated until you touch the memory.
[20:12:09] <sommerfeld> if you follow up the malloc() with a memset(ptr, 0x42, length)
[20:12:29] <TrogL> I'm not sure it's beign reserved.  top says 8193M total, 8193M free.  but I'll try that
[20:12:34] <sommerfeld> what does swap -l say?
[20:12:56] <TrogL> /dev/md/dsk/d3      85,3      16 16780208 16780208
[20:13:14] <TrogL> ie everytrhing free
[20:13:19] <PerterB> I think he means swap -s
[20:13:29] <sommerfeld> that too
[20:13:44] <TrogL> total: 4633664k bytes allocated + 386160k reserved = 5019824k used, 6917312k available
[20:14:43] <cmihai> TrogL, try this
[20:15:21] <cmihai> TrogL, http://unixsadm.blogspot.com/2007/08/getting-detailed-system-information-on.html the ::memstat line.
[20:15:44] <cmihai> pastebin the result on paste.lisp.org
[20:16:08] <cmihai> Don't forget
[20:16:09] <cmihai> ::kmastat
[20:16:14] <TrogL> k
[20:17:56] <Pietro_S> lacaAFK: how the JDS shutdown button is implemented? Though HAL? I want to make same thing for Xfce - but xfce with HAL doesn't work at all ..
[20:18:07] *** hile_ has joined #opensolaris
[20:19:39] <TrogL> cmihai: how long for ::memstat to return?  it's been awhile
[20:19:52] <cmihai> It has?
[20:20:02] <cmihai> Shouldn't take more than a minute or so
[20:20:05] <cmihai> Give it 5 minutes
[20:20:09] <TrogL> k
[20:20:19] <cmihai> As you can see, the output is pretty detailed though.
[20:20:33] <BadKarma> any idea what PSU power may dual xeon mobo need?
[20:20:36] <cmihai> It might shed some light on your troubles. Or just waste our time. Either way is good :-)
[20:21:08] <TrogL> I'm watching it on top.  it sure changes CPUs a lot
[20:21:09] <cmihai> BadKarma, CPU wouldn't even matter much. Dual Xeon, consider the worst at 100-150W, and you're still fine with a nice 500W
[20:21:18] <cmihai> TrogL, don't use top.
[20:21:21] <cmihai> TrogL, use prstat.
[20:21:30] <cmihai> top has... issues :-)
[20:21:47] <TrogL> taht's what I've been trying to tell them.  errr... waht issues?
[20:21:56] <myrkraverk> and the least one is it's homicidal maniacal tendencies
[20:22:04] *** rcorreia_ has joined #opensolaris
[20:22:23] <TrogL> we're back
[20:22:33] <cmihai> I don't want to rant here, but it's inexact and consumes a lot of resources itself. You'd see if you'd dtrace them both :P. Anyway, it's not part of the Solaris base system, so just use prstat :P
[20:22:55] <BadKarma> cmihai: I've got a nice 500W and a tyan S2723 + 2x Xeon 2.66GHz for some reason its not starting.. I mean no video / beep but the cpus are getting power also if I plug a net cable to one of onboard nics led flashes
[20:23:06] *** Kush- has quit IRC
[20:23:07] *** jmcp has quit IRC
[20:23:08] <BadKarma> tired with one cpu / no ram
[20:23:20] <cmihai> Remove disks / PCI cards / stuff
[20:23:21] <BadKarma> /s/tired/tried/
[20:23:21] *** mick_work has joined #opensolaris
[20:23:24] <cmihai> does it start?
[20:23:28] *** jmcp has joined #opensolaris
[20:23:35] <BadKarma> did that its just a board + cpus now
[20:23:43] <nachox> does anyone have a solaris 8 installed in vmware?
[20:23:44] <BadKarma> even removed ram
[20:23:48] <cmihai> Hell, I've got dual Xeon servers here and they just have hot swapable 500W PSUs
[20:24:00] <cmihai> BadKarma, I think there's other issues there mate.
[20:24:15] <BadKarma> also tried different cpus
[20:24:23] <cmihai> nachox, nope, just 9 in qemu :P
[20:24:26] <BadKarma> which are confirmed to work
[20:25:01] <cmihai> It's either bad cabling or broken... does it.. erm.. beep/
[20:25:14] <cmihai> Any beeping singnals would warn you of BAD memory and such
[20:25:17] <BadKarma> nothing beeps
[20:25:19] <nachox> crap, i just needed to know what was the network driver it used
[20:25:25] <cmihai> BadKarma, try another PSU?
[20:25:27] <mick_work> hi everyone - i'm trying to throw some perl together to make a container/zone and it *looks like* it works - but it actually doesn't - if someone has an idea, i'd love to hear what i'm doing wrong: http://erxz.com/pb/4825
[20:25:28] <laca> Pietro_S: take a look at SUNWgnome-sys-suspend
[20:25:29] <BadKarma> yeah I have an instructin manual
[20:25:36] <laca> or man gnome-sys-suspend
[20:25:46] <BadKarma> tried a different one 450W
[20:25:46] <mick_work> it shows a terminal session that does exactly what i want
[20:25:52] <BadKarma> still nothing
[20:25:58] <mick_work> but when i verify it - it doesn't actually create it
[20:26:11] <cmihai> BadKarma, could be a bad MB
[20:26:42] <BadKarma> it was taken out of a working box
[20:26:50] *** bnitz has left #opensolaris
[20:26:55] <cmihai> Put it back in
[20:27:02] <BadKarma> don't have that anymore
[20:27:03] *** NikolaVeber_ has quit IRC
[20:27:04] <cmihai> With most of the components, one at a time.
[20:27:26] <BadKarma> hmm but it gives me an idea
[20:27:29] <BadKarma> brb
[20:28:09] *** bondolo has quit IRC
[20:28:58] *** Drone has quit IRC
[20:29:01] <Pietro_S> laca: thanks
[20:30:12] *** JBeck has quit IRC
[20:30:20] <BadKarma> cmihai: call me a WANKER... the speaker was broken
[20:30:40] <BadKarma> looks like it cries about memory anyway
[20:30:41] <cmihai> BadKarma, you wanker!
[20:30:42] <cmihai> :-)
[20:30:50] <BadKarma> thanks bro, thats appreciated
[20:31:00] <cmihai> You're welcome :-]
[20:31:01] <BadKarma> ok ... lemme see if I have any other memory available
[20:31:55] <Pietro_S> laca: hmm, it looks like it needs SUID root, so I probably rather find code which shutdown in xfce and instead of sudo I will allow using RBAC (pfexec)
[20:33:10] <Gman> http://today.slac.stanford.edu/images/2007/blackbox-openhouse.jpg
[20:33:11] <Gman> ^ whitebox :)
[20:34:31] *** JBeck has joined #opensolaris
[20:34:32] *** ChanServ sets mode: +o JBeck
[20:35:01] <stevel> all hail the ON overlord
[20:35:06] <stevel> :-p
[20:35:33] <Downix> flyingparchment, Ok, I'm at the partitioning stage... don't see an svm setup.
[20:35:42] <flyingparchment> Downix: you set up mirroring after you installed
[20:35:47] <Downix> ah, ok
[20:36:02] <Downix> so, should I skip anything on the second HD, just format it and put it somewhere?
[20:36:11] <JBeck> stevel: :-P yourself
[20:36:14] <TrogL> cmihai : http://www.pastebin.ca/741450
[20:36:43] <Downix> I could just put it as /docs for the websites
[20:36:52] <Downix> or under /export/home
[20:37:06] <Downix> Ok, /export/home
[20:37:31] <cmihai> TrogL, oh my, that took a while.
[20:37:33] *** rcorreia has quit IRC
[20:37:43] *** Drone has joined #opensolaris
[20:37:49] <TrogL> cmihai: yeah, copy-paste hell
[20:38:07] <Downix> that way should something bad happen to the OS, I still have all of the data
[20:38:24] <TrogL> cmihai: lott zio_*_bufs
[20:38:28] <BadKarma> cmihai: 9 beeps which would indicate (according to manual) only one mem stick instred
[20:38:32] <BadKarma> there are 2 however
[20:38:37] <TrogL> cmihai: lots of  zio_*_bufs
[20:38:50] <BadKarma> looks like this board doesnt like diff kinds of sticks
[20:40:21] <Downix> I'll still need a redundancy option
[20:40:22] <Downix> hmm
[20:40:33] <Downix> anyone make a good tape backup able to hold ~160GB?
[20:40:34] <Downix> 8)
[20:40:53] <sommerfeld> Downix: it will cost you
[20:41:08] <TrogL> Downix: DLT is your friend.  expensive friend
[20:41:17] <Downix> sommerfeld, I wouldn't expect otherwise
[20:41:21] <cmihai> TrogL, bloody hell man
[20:41:30] <Pietro_S> laca: I heard that you are starting building SFE to IPS ... I was about to write tool which makes IPS from spec files ... do you already have some tool or do you convert it manually?
[20:41:31] * TrogL panics
[20:41:52] <cmihai> What do you run on the thing, Oracle? :-)
[20:41:58] <sommerfeld> LTO, DLT, etc
[20:42:14] <TrogL> cmihai:  aye, ORACLE.
[20:42:27] <laca> Pietro_S: no, i'm still building svr4
[20:42:38] <laca> if you want to work on such a tool, that would be great
[20:42:46] <laca> you can post your ideas to desktop-discuss
[20:43:01] <sommerfeld> not to mention the storagetek T10k's
[20:44:06] <TrogL> cmihai: ORACLE asked for certain things in /etc/system, which ended up in an /etc/projects entry.  I'm wondering if some were a Bad Idea
[20:44:44] <cmihai> TrogL, well, some no longer apply for Oracle 10g
[20:44:45] <TrogL> cmihai: but ORACLE wouldn't run at all without them
[20:44:49] <cmihai> So you might want to tune those parameters.
[20:44:59] <cmihai> TrogL, eye, I know.
[20:45:23] <cmihai> Oracle requires heavy kernel tuning and tuning itself.
[20:46:31] <TrogL> cmihai: they wanted (in /etc/system) set noexec_user_stack=1
[20:46:58] <TrogL> cmihai: and I gave a project entry:
[20:47:01] <TrogL> user.oracle:100:Oracle:::process.max-sem-nsems=(priv,256,deny);project.max-sem-ids=(priv,100,deny);project.max-shm-ids=(priv,100,deny);project.max-shm-memory=(priv,4294967295,deny)
[20:47:39] <TrogL> cmihai: however, I can't get non-oracle-user test code to swap
[20:47:43] *** Kush- has joined #opensolaris
[20:48:02] <cmihai> Hm...
[20:48:20] <cmihai> I don't have an Oracle devel box handy to test.
[20:48:30] <cmihai> I'll see if I can figure something out tomorrow.
[20:48:50] <TrogL> OK, I'll be here.  I'll hack away in the meantime
[20:49:11] <cmihai> Yeah, let me know what you figure out :-).
[20:49:32] <axisys> I am getting this for sol 10 u4
[20:49:34] <axisys> add_install_client: /jumpstart/sol10u4x86/Solaris_10/Tools/Boot is not a valid Solaris install boot image
[20:49:54] <axisys> what am I doing wrong?
[20:50:39] <TrogL> axisys: not enough parameters I think you need -s at a minimum
[20:52:49] <axisys> ./add_install_client -d -e "00:e0:81:5d:cb:9f" -s 2.7.0.25:/jumpstart/sol10u4x86
[20:53:41] <TrogL> axisys: that work?
[20:53:48] <axisys> TrogL: nah!
[20:54:06] *** yippi has joined #opensolaris
[20:54:13] <axisys> add_install_client: /jumpstart/sol10u4x86/Solaris_10/Tools/Boot is not a valid Solaris install boot image
[20:54:24] <holcomb> is it?
[20:54:48] <axisys> now worked
[20:54:55] <axisys> i forgot to put platform group
[20:55:15] *** cygnusecks has quit IRC
[20:56:04] *** cygnusecks has joined #opensolaris
[20:56:18] <TrogL> how do I quit out of mdb?
[20:56:23] <holcomb> ::quit
[20:56:40] <TrogL> thansk
[20:57:30] <dsch04> <sigh> I don't want to give up yet, but I'm close.
[20:58:19] <dsch04> A recap: I'm trying to get OpenSolaris (b74) installed on an Asus P5K WS board, with 6xSATA disks
[20:58:34] <e^ipi> opensolaris is some code
[20:58:39] <e^ipi> I assume you mean solaris express
[20:58:44] <dsch04> Sorry, yes.
[20:58:59] <dsch04> I originally had a SAT2-MV8 controller, but RMA'd that today as I thought it was faulty
[20:59:30] <dsch04> I've now got an IDE DVD drive installed and all 6 HDDs on the onboard ICH9 controller
[20:59:47] <dsch04> The install booted OK, and completed.
[20:59:57] <dsch04> But, on reboot, it's hanging
[21:00:22] *** rcorreia__ has joined #opensolaris
[21:00:30] <dsch04> The last thing I see is: sda2 at: blah blah
[21:00:36] *** rcorreia__ is now known as rcorreia
[21:00:37] <dsch04> sd2 is: blah blah
[21:00:44] <dsch04> s/sda2/sd2/
[21:00:50] *** Downix has quit IRC
[21:01:08] <dsch04> It detects the first two drives fine but hangs on the third
[21:01:08] <e^ipi> boot it with the -v option, may give you more info
[21:01:24] <dsch04> That is with the -vk options
[21:01:54] *** IvanR_ has quit IRC
[21:02:44] <seanmcg> did you break into kmdb and see where its hanging at ?
[21:02:53] <dsch04> No, how do I do that?
[21:03:06] <Pietro_S> laca: where is right place to take look at source of pkgbuild? As /opt/jdsbld/lib/pkgbuild-1.2.0/pkgbuild.pl tells that it was generated by configure ...
[21:03:09] <seanmcg> F1-a  (the F1 key with the letter 'a' key)
[21:03:14] <dsch04> ok
[21:03:18] <seanmcg> thats from the console.
[21:03:26] <BadKarma> huh
[21:03:51] <laca> Pietro_S: the source is in CVS at sourceforge
[21:03:56] <dsch04> F1-a doesn't do anything
[21:04:02] <laca> you should be able to check it out as anonymous
[21:05:34] <laca> Pietro_S: it's better to work from CVS anyway, so you can generate patches easily
[21:05:46] *** jcea has left #opensolaris
[21:06:13] <seanmcg> dsch04 you're not on a remote console or anything ?
[21:06:28] <dsch04> USB keyboard
[21:06:42] <dsch04> plugged into server
[21:07:01] <dsch04> (actually, via a KVM, but same thing)
[21:07:14] <seanmcg> F1-A should work.. and have the -k flag.  Odd.
[21:07:33] *** dmesg has quit IRC
[21:08:08] *** dynamicproxy has quit IRC
[21:08:26] <Pietro_S> laca: oki
[21:08:59] <dsch04> <shrug>
[21:09:11] <dsch04> Not having a good time of it!
[21:11:08] <Pietro_S> laca: it's few years ago when I used cvs so I'm a bit slow ...
[21:12:00] *** virgee has quit IRC
[21:12:56] *** blindfish has quit IRC
[21:14:02] *** rcorreia_ has quit IRC
[21:15:37] <laca> Pietro_S: instructions here: http://sourceforge.net/cvs/?group_id=125443
[21:16:16] <Pietro_S> laca: thanks, already got it
[21:16:55] *** nostoi has quit IRC
[21:19:07] *** KermitTheFragger has quit IRC
[21:22:22] <Pietro_S> laca: I got it I will take look deeply on it tomorrow and also I will post some ideas on desktop-discuss ... but right now I really need bath
[21:23:40] *** styX-Xyts has joined #opensolaris
[21:24:03] <TrogL> heresy
[21:25:59] *** styX-Xyts has left #opensolaris
[21:26:18] <e^ipi> anyone happen to know offhand if the boot procedure for EFI is well documented
[21:26:25] <e^ipi> that is, file formats, entry points, etc
[21:28:10] *** l1s has joined #opensolaris
[21:28:31] *** styX-Xyts has joined #opensolaris
[21:31:01] *** victori_ has quit IRC
[21:39:08] *** sahafeez has joined #opensolaris
[21:48:04] *** bengtf_ has joined #opensolaris
[21:48:08] *** locy has joined #opensolaris
[21:56:54] *** blindfish has joined #opensolaris
[21:59:47] *** Fish has quit IRC
[21:59:48] *** phips has quit IRC
[22:03:30] *** phips has joined #opensolaris
[22:04:11] <dsch04> aaarrgggh
[22:04:24] *** phips has quit IRC
[22:04:37] *** loky has quit IRC
[22:04:41] <dsch04> I've now got a "working" system - it booted into Solaris Express and all 6 drives are visible
[22:04:43] <dsch04> Yay!
[22:04:51] <dsch04> But ... (there's always a but...)
[22:04:59] <styX-Xyts> bonne nuit
[22:05:00] <styX-Xyts> oups
[22:05:02] <styX-Xyts> sorry
[22:05:07] <dsch04> The network isn't working
[22:05:11] <styX-Xyts> it is not the fr chan
[22:05:18] *** FelipeBare has quit IRC
[22:05:27] <hsn_> anybody here installed sx 9/07 on machine with 256MB?
[22:05:28] <styX-Xyts> good night
[22:05:35] <dsch04> It all appears to be installed/configured
[22:05:37] *** styX-Xyts has left #opensolaris
[22:06:32] <dsch04> I removed the SK98sol pkg and installed the YUKONXsol driver
[22:06:43] <dsch04> I've configured it, set the default route, set the nameserver
[22:06:52] <dsch04> It appears all to be up:
[22:08:16] <jteo> and?
[22:09:35] <dsch04> Damn, forgot I can't paste from it 'cos the network isn't working yet!
[22:09:58] <dsch04> I can't ping anything
[22:10:08] <jteo> i presume you plumbed it?
[22:10:20] * dsch04 slaps head
[22:10:25] * dsch04 new to Solaris
[22:10:39] * dsch04 had niggle in back of mind that I'd forgotten something...
[22:10:53] <jteo> don't worry. i'm a noob too. :)
[22:12:31] *** nachox has quit IRC
[22:12:50] <dsch04> Hmm, it said SIOCSLIFNAME for ip yukon0: already exists
[22:13:18] <jteo> ifconfig yukon0 ?
[22:13:27] <jteo> have you set your ip?
[22:13:48] <dsch04> Yes
[22:13:55] <jteo> <UP,BROADCAST,RUNNING,MULTICAST,DHCP,IPv4,CoS>
[22:14:07] <jteo> you should have that (less the DHCP since you don't use it)
[22:14:26] *** phips has joined #opensolaris
[22:14:41] <dsch04> No CoS
[22:14:54] <dsch04> I have NOTRAILERS
[22:15:07] <jteo> i'm outta ideas.
[22:15:10] * dsch04 too
[22:15:21] <jteo> got a spare NIC? :P
[22:15:40] <dsch04> both onboard.
[22:16:26] <jteo> ugh.
[22:16:33] * dsch04 bangs head on desk
[22:16:37] <jteo> :(
[22:16:38] <jteo> be patient.
[22:17:06] <dsch04> No, at my stupidity
[22:17:13] <jteo> hmm?
[22:17:21] <jteo> you didn't plug in your cable?
[22:17:50] <dsch04> For some bizarre, uneartly unexplainable reason I had the cable in the 2nd NIC port (yukonx1)
[22:18:19] <dsch04> I can now do this:
[22:18:20] <dsch04> # ifconfig yukonx0
[22:18:20] <dsch04> yukonx0: flags=1000863<UP,BROADCAST,NOTRAILERS,RUNNING,MULTICAST,IPv4> mtu 1500 index 4
[22:18:20] <dsch04>         inet 192.168.1.10 netmask ffffff00 broadcast 192.168.1.255
[22:18:20] <dsch04>         ether 0:1b:fc:b3:73:b5
[22:18:31] <jteo> looks good.
[22:18:36] <jteo> :)
[22:18:41] <dsch04> Working!
[22:18:45] <jteo> kudos.
[22:18:51] <jteo> now backup/write down what you did.
[22:19:03] <dsch04> I don't know what I did!
[22:19:05] <jteo> ...
[22:19:09] <dsch04> Which bit do you mean?
[22:19:18] <dsch04> I've written down what I did to configure the network
[22:19:29] <dsch04> But I have no idea why the disks have started working
[22:19:42] <jteo> intriguing.
[22:20:50] <flyingparchment> heh notrailers
[22:21:00] <flyingparchment> really useful, so you can enable trailers on the 2 ethernets still using them
[22:21:16] <jteo> what does notrailers mean?
[22:21:19] * dsch04 has no idea what it means
[22:21:20] <jteo> (noob0
[22:21:28] <flyingparchment> trailers = putting the header at the end of the frame
[22:21:33] *** sahafeez has quit IRC
[22:21:37] <flyingparchment> sometimes done a very long time ago because it was a bit faster
[22:22:05] <jteo> ah.
[22:22:13] <dsch04> You know how to enable jumbo frames?
[22:22:50] <flyingparchment> ifconfig yukonx0 mtu 9000 # i think
[22:23:04] <dsch04> Ok, will experiment
[22:23:07] *** timsf has quit IRC
[22:23:12] <flyingparchment> write 'mtu 9000' after the host in /etc/hostname.yukonx0, if that works, to do it on boot
[22:23:36] <quasi> http://blogs.sun.com/taylor22/entry/configuring_jumbo_frames_on_the - but ymmv on other nics
[22:23:45] <dsch04> thx
[22:23:58] <dsch04> You know, I've about 4 days of pain just for zfs!
[22:24:05] <dsch04> (I've had...)
[22:24:25] <dsch04> All so I can type:
[22:24:26] <jteo> dsch04: suffering is good for the soul. and you had IRC with us to keep you company!
[22:24:31] <dsch04> bash-3.2# zpool create tank c1t1d0 c1t2d0 c1t3d0 c1t4d0 c1t5d0
[22:24:31] <dsch04> bash-3.2# zfs create -o shareiscsi=on tank/iscsi_luns
[22:24:31] <dsch04> bash-3.2# zfs create -s -V 25g tank/iscsi_luns/vol001
[22:24:34] * bda got hosts zfs booting via Jumpstart this morning.
[22:24:57] <quasi> bda: nice
[22:25:02] <dsch04> Jumpstart?
[22:25:06] <bda> Yeah, it was easy. I'm pretty pleased.
[22:25:10] <quasi> bda: will you be blogging a profile?
[22:25:13] <dsch04> Anyting like Redhat kickstart?
[22:25:25] <cmihai> JumpStart.
[22:25:30] <bda> dsch04: Where do you think they stole Kickstart from?
[22:25:34] <dsch04> ;)
[22:25:38] <cmihai> (and the name)
[22:25:56] <jteo> ,et/
[22:26:00] <bda> quasi: I could, but it's really just as easy as the README says.
[22:26:01] <dsch04> It's a conspiracy, you know. :)
[22:26:05] <cmihai> At least HP came up with an original name.. IgniteUX
[22:26:13] <jteo> let's not throw stones. Redhat's installer was lightyears ahead of the solaris one.
[22:26:33] <dsch04> s/was/is
[22:26:36] * dsch04 ducks
[22:26:42] <cmihai> yeah/right/whatever
[22:26:47] <quasi> bda: ah, so all I need to do is dig out the README ;)
[22:26:47] <dsch04> heh
[22:27:16] <dsch04> Is JumpStart suitable for a 1/2 host install?
[22:27:24] <bda> quasi: http://www.opensolaris.org/os/community/zfs/boot/netinstall/ # :)
[22:27:33] *** mikefut has joined #opensolaris
[22:27:45] <dsch04> So I can do the install a few times in a repeatable way?
[22:27:53] <quasi> bda: thanks
[22:27:57] <bda> dsch04: Yes.
[22:28:04] <bda> dsch04: http://jet.maui.co.uk/wiki/index.php/JETPrimer
[22:28:07] <dsch04> Are there any good docs?
[22:28:08] <bda> JET makes JumpStart not a pain.
[22:28:13] <dsch04> Ooops. Thanks.
[22:28:18] * quasi has a bit of time to wait before sxce b75
[22:28:40] <cmihai> Can't you ask bigadmin or docs.sun.com before? Most of the good stuff is there or on blogs :P
[22:28:53] <jteo> oh yes JET.
[22:28:54] <dsch04> Is that directed at me?
[22:29:09] <bda> jteo: Some things about it kind of confuse me, but eh.
[22:29:17] <jteo> bda: lesser of 2 evils.
[22:29:30] <bda> jteo: Like it doesn't seem to have a concept of "classes", wherein the same set of scripts are applied to every host without having to exist in the <client> dir...
[22:29:36] *** cga has quit IRC
[22:29:44] <bda> I can hack around it, but it's a weird lack.
[22:29:58] <bda> (Or I'm totally wrong, and will be happy to be so.)
[22:30:16] <quasi> dsch04: http://soulfood.dk/archives/2006/09/20/T23_43_37/index.html gives you a quick idea of how to jumpstart a box
[22:30:25] <dsch04> Cool
[22:30:44] <jteo> jet is a bandaid around jumpstart, which is arcane to begin with.
[22:30:51] <bda> For sure.
[22:31:07] <dsch04> Would have been useful over the past 4 days, given the number of installs I've done!
[22:31:38] <jteo> dsch04: you could have asked. ;)
[22:31:45] <quasi> jteo: perhaps a bandaid, but then at least one of the nice ones with disney characters on it
[22:32:01] <jteo> quasi: still hurts when it doesn't work.
[22:32:07] <dsch04> I'm pretty handy with Redhat kickstart configs
[22:32:45] <dsch04> I do all my installs with them - I can specify just what packages get installed, and set the boxes up in the post-install scripts
[22:32:46] <quasi> jteo: jet has worked for me when plain jumpstart was driving me crazy
[22:32:53] <dsch04> I'd like to do the same with SE
[22:33:04] <bda> It makes me miss FAI, a lot.
[22:33:09] <jteo> quasi: true. i just dislike the cryptic error messages inherent with the whole thing.
[22:33:25] <jteo> quasi: things like Jumpstart should be *easy*
[22:33:31] <quasi> jteo: especially it is very nice not to have to mess with the dhcp server and all the extensions
[22:33:36] <bda> Yes.
[22:33:39] <quasi> jteo: agreed
[22:33:41] *** sioraiocht has joined #opensolaris
[22:33:44] <bda> The DHCP crap was why I tried it in the first place. :P
[22:33:47] <quasi> jteo: that's why I blogged
[22:33:55] <quasi> about it
[22:34:00] <jteo> on the bright sides, it keeps solaris SAs in employment.
[22:34:04] <jteo> quasi: ahh.. :)
[22:34:52] <quasi> jteo: but still - it is all there, just a small pile of scripts to look into
[22:35:23] <quasi> jteo: and the "support" response on the yahoo jet group is pretty darn good
[22:36:04] *** sioraiocht has quit IRC
[22:37:51] <tomww> dsch04: you've gotten a regular guest here in #opensolaris :-) ?!
[22:38:02] <dsch04> Indeed!
[22:39:03] <quasi> tomww: just wait till he sees the bill ;)
[22:39:33] <tomww> quasi: psssst! don't tell!
[22:40:31] <quasi> tomww: I figure we've got enough already to clean him out and repo his machine so no sense in going on ;)
[22:42:22] <dsch04> Heh
[22:42:25] *** blindfish has quit IRC
[22:42:41] <dsch04> I've got into all this just because I want to use zfs
[22:42:55] <joevandyk> How do I set the ulimit (for increasing max open files) in SMF?
[22:43:09] <dsch04> I've just created a new zpool across 5 disks
[22:43:12] <dsch04> raidz2
[22:43:22] <elektronkind> yay
[22:43:28] <elektronkind> oh hey dsch04, you got that card working?
[22:43:34] <dsch04>  zpool create tank raidz2 c1t1d0 c1t2d0 c1t3d0 c1t4d0 c1t5d0
[22:43:39] <dsch04> No
[22:43:47] <elektronkind> everyone likes the pool name of tank it seems
[22:43:48] <dsch04> I RMA'd it today
[22:43:48] <seanmcg> no hot spares :)
[22:44:01] <dsch04> Although, I'm not convinced it was faulty
[22:44:16] <elektronkind> I seems that I'm the only one that likes the pool name of "kittens"
[22:44:23] <dsch04> Heh
[22:44:24] <dsch04> Anyways
[22:44:46] <dsch04> I would expect to "lose" the capacity of 2 drives with a 5-disk raidz2
[22:44:56] <e^ipi> the name "tank" is because all the old documentation used that as the metasyntactic variable
[22:44:57] <Triskelios> joevandyk: ulimit is a user environment thing
[22:45:15] <e^ipi> i name my pools after greek philosophers *shrug*
[22:45:17] <dsch04> I'm using 5x250GB drives, so I expected to have a pool of about 750GB
[22:45:20] <elektronkind> metasyntactic is the secret word of the day!
[22:45:26] <joevandyk> Triskelios, so i have a SMF for nginx.  How can I start it so that it can open more than 256 files?
[22:45:32] * elektronkind queues the Pee Wee mayhem
[22:45:53] <dsch04> However:
[22:45:56] <dsch04> # zpool list tank
[22:45:56] <dsch04> NAME                    SIZE    USED   AVAIL    CAP  HEALTH     ALTROOT
[22:45:56] <dsch04> tank                   1.14T    429K   1.14T     0%  ONLINE     -
[22:46:16] <quasi> dsch04: looks like it striped
[22:46:18] <elektronkind> joevandyk: you'd put the ulimit command in the method script for ngix
[22:46:21] <Triskelios> joevandyk: that is lower than the default, so the program is either changing it or it's running into some other limit
[22:46:34] <dsch04> It's showing as raidz2:
[22:46:35] <elektronkind> joevandyk: ie in /lib/svc/method/ngix or whatever the script is called
[22:47:00] <dsch04> http://opensolaris.pastebin.ca/741597
[22:47:02] <joevandyk> elektronkind, it's being exected directly in the SMF.  No method scripts.
[22:47:35] <joevandyk> Triskelios, 256 isn't the default?
[22:48:19] *** wms_ has quit IRC
[22:48:37] <Triskelios> joevandyk: there's no limit by default, and the hard limit in the kernel is probably 4k
[22:49:55] *** Chihan has quit IRC
[22:50:46] <Triskelios> no, it's actually 64k..
[22:52:00] *** emergo has quit IRC
[22:52:02] <Triskelios> sure you're not running into the fopen() limit?
[22:52:28] <dsch04> So, anyone any idea why I'm seeing a 1.14T zpool with 5x250GB drives in a raidz2 arrangement?
[22:53:15] *** yippi has quit IRC
[22:53:31] <Triskelios> joevandyk: http://developers.sun.com/solaris/articles/stdio_256.html - solution is either to build a 64-bit app or LD_PRELOAD the workaround (or switch to regular open())
[22:53:49] <holcomb> i think zpool list lies
[22:54:16] <quasi> dsch04: try zpool status -v
[22:54:20] <dsch04> k
[22:54:36] <quasi> dsch04: I'm guessing it didn't raidz2
[22:55:00] <dsch04> http://opensolaris.pastebin.ca/741597
[22:55:03] <dsch04> It says it did
[22:56:04] <quasi> dsch04: odd
[22:56:16] <dsch04> v. odd
[22:56:20] <dsch04> bug?
[22:57:07] <quasi> maybe you stuck 400G drives in there instead? ;)
[22:57:11] <dsch04> Looks like it's just done raidz1
[22:57:13] <dsch04> Heh
[22:57:34] <dsch04> I just destroyed it, and did the same command with raidz1 and get the same capacity
[22:57:57] <quasi> raidz1 wouldn't give you 1.14T out of 5 250G drives
[22:58:00] *** mikefut has quit IRC
[22:58:05] <dsch04> It does
[22:58:17] <dsch04> Or rather, it does on my system
[22:58:24] *** cmihai has quit IRC
[22:58:32] *** _Megaf has quit IRC
[22:58:40] <quasi> well, only by striping - or it just lies about size
[22:59:01] <dsch04> It must be lieing
[22:59:06] <dsch04> lying?
[22:59:12] *** _Megaf has joined #OpenSolaris
[22:59:27] <dsch04> A stripe has capacity 1.13T
[22:59:45] <e^ipi> haha, magical ZFS
[22:59:49] <quasi> close
[22:59:55] <e^ipi> take that... laws of thermodynamics
[23:00:05] <dsch04> Wow - perpetual motion!
[23:00:35] <dsch04> aha, df shows a sane view
[23:01:09] <dsch04> Filesystem             size   used  avail capacity  Mounted on
[23:01:09] <dsch04> pool00                 685G    32K   685G     1%    /pool00
[23:01:11] <RElling> zpool list shows different accounting than df
[23:01:28] <dsch04> That's v. confusing
[23:01:47] <dsch04> zpool list tells me that I have 1.14TB available
[23:01:51] <dsch04> THat's wrong
[23:02:01] <quasi> RElling: is that new? my snv_56 gets it right for a mirror
[23:02:18] <dsch04> See above ^^^
[23:02:41] <RElling> zfs is dynamic... unlike raid-5, raidz may or may not consume (N-1)/N space
[23:03:05] <RElling> mirrors should be right
[23:03:09] <quasi> ah
[23:03:14] <quasi> good point
[23:03:55] <RElling> I sometimes call it Enron accounting, but only amongst friends :-)
[23:04:08] <dsch04> Yes, but a raidz2 pool with 5xN Gb drives should be somewhere around 3xN Gb in size, not 5x NGb
[23:04:52] <RElling> from ZFS's view, there is 5xN GBytes of space available
[23:05:04] <dsch04> It doesn't matter if it's dynamic or not - it's reporting that I have 1.14TB of space when I only have 685GB
[23:05:13] *** victori_ has joined #opensolaris
[23:05:18] <RElling> but when you consume, you don't consume 1:1
[23:05:26] <dsch04> Ah, but is zfs running the "zpool list" command?
[23:05:35] <dsch04> No - the sysadmin is
[23:05:59] <dsch04> And from the sysadmin's view, he only has 685GB of space available, not 1.14TB
[23:06:47] <RElling> ... except he may or may not have 685GB of space available... not quite as confusing as tmpfs though
[23:07:02] <dsch04> Hmmm, zfs list does as I would expect
[23:07:48] <dsch04> OK, I am not confused. I am not sure what zpool list is telling me
[23:07:58] <dsch04> OK, I am now confused. I am not sure what zpool list is telling me
[23:08:27] <g4lt-mordant> zpool list is literally the exact space that the entire pool physically has
[23:08:40] *** szt has left #opensolaris
[23:08:55] <dsch04> i.e. the sum total of all devices in the pool?
[23:09:11] <g4lt-mordant> yup
[23:09:26] <RElling> or, as the man page says, This command reports actual physical space available  to the  storage  pool.
[23:09:34] <dsch04> The wording is cofusing
[23:09:37] <dsch04> confusing
[23:09:50] <RElling> The physical space can be different from the  total  amount  of  space  that  any  contained datasets can actually use.
[23:10:12] <dsch04> having seen zfs list, I can understand that
[23:10:31] <RElling> The amount of space used in a raidz configuration depends on  the  characteristics  of the  data  being written.
[23:10:42] <dsch04> But, what does "Size 1.14T" vs "AVAIL 1.14T" mean?
[23:11:07] <dsch04> i.e. what's the difference?
[23:11:32] <dsch04> Isn't the total space available to the pool going to remain constant?
[23:11:42] <dsch04> Or is that to cater for failed devices?
[23:11:44] <RElling> nothing, if it is empty, but as you consume you will see used go up faster than the logical size of the data
[23:11:59] *** _Megaf has quit IRC
[23:13:45] <dsch04> So, in round numbers..
[23:14:40] <dsch04> If I have a raidz2 pool with 1TB available (zpool list) which is showing 700GB of usable space (zfs list)
[23:14:52] <dsch04> I then write 200GB to it
[23:15:05] <dsch04> zfs list will show 500GB usable
[23:15:45] <dsch04> zpool list will show 1TB - ( 200 * F )
[23:16:20] <dsch04> Where F is approx 1 + 2/N = 1.4 (for N=5)
[23:17:03] <RElling> I believe you are on the right track
[23:17:04] <dsch04> So zpool list will show 280GB used and 720GB available
[23:17:16] <dsch04> In round numbers, of course
[23:17:19] <dsch04> OK
[23:17:38] <dsch04> I think the bottom line is - use zfs list
[23:17:38] *** jhawk has joined #opensolaris
[23:17:39] <dsch04> :)
[23:17:55] <RElling> when you get your mind wrapped around that, you can tackle "REFER" :-)
[23:18:07] <dsch04> Where's that?
[23:18:13] <RElling> zfs list
[23:18:23] <sommerfeld> or just enable compression; the inside of your pool will be bigger than the outside.
[23:18:26] *** jhawk has quit IRC
[23:18:52] *** laca is now known as lacaAFK
[23:19:13] <dsch04> Or create sparse volumes
[23:19:21] <dsch04> 150TB anyone?
[23:19:29] *** jcea has joined #opensolaris
[23:20:03] <dsch04> Ok, I give up, what's REFER
[23:20:03] <dsch04> ?
[23:20:59] <RElling> you can have multiple file systems (and snapshots) in a pool, so we want to know how much data is being referred to by a dataset
[23:21:40] <dsch04> Are the guys that designed this gibbering wrecks?
[23:21:51] <bda> No, but they are pretty goofy dudes.
[23:22:02] <bda> (Their talks are pretty highly entertaining.)
[23:22:25] <RElling> give up mortal, you will never know how much space you really have in your puny little universe :-)
[23:22:26] <dsch04> I've only found one major issue with zfs so far
[23:22:29] <bda> dsch04: http://video.google.com/videoplay?docid=8164805665567899891
[23:22:40] <dsch04> bda, thx, will check it late
[23:22:43] <dsch04> r
[23:23:07] <dsch04> Where "major issue" = "something I'd really like to be able to do but is not supported"
[23:23:35] <dsch04> I'd like to be able to add drives to a raidz2 array
[23:24:24] *** locy has quit IRC
[23:24:54] <RElling> in what way?  make a raidz[2] set wider, or just add more protected storage to the pool?
[23:24:55] *** glagasse has quit IRC
[23:26:02] <sommerfeld> dsch04: i find myself using mirrors rather than raidz on servers with small numbers of disks.
[23:26:08] <sommerfeld> and you can then add in pairs.
[23:26:44] <holcomb> isn't the 'add more disks to raidz' feature under construction?
[23:27:21] <dsch04> RElling, For example, create a raidz2 array with 6 disks, then add another one, then another, etc.
[23:28:11] <RElling> alas, the dynamic nature and COW make such enlargement non-trivial, so it hasn't been implemented yet
[23:28:24] <dsch04> I understand that.
[23:28:35] <dsch04> I still want to do it though :)
[23:30:03] <RElling> http://baarf.com
[23:32:26] *** yippi has joined #opensolaris
[23:40:06] <dsch04> bda, are the slides from that presentation online anywhere?
[23:40:19] <bda> Dunno.
[23:40:22] *** dduvall has quit IRC
[23:41:52] *** Bahra has joined #opensolaris
[23:41:55] <dsch04> sommerfeld, I'm looking at 8 disk NAS servers for home use
[23:42:06] <dsch04> Well, home/business use
[23:42:24] *** Bahra is now known as samy
[23:42:58] *** samy is now known as Bahra
[23:43:29] <dsch04> (I'm a System Consultant and use xen/XenSource a lot, so I want a NAS exporting volumes as iSCSI to use for xen guests on a separate Xen server)
[23:43:46] <dsch04> I also have a large collection of flac files. :)
[23:44:54] <Kitty> erm
[23:45:03] <Kitty> iSCSI would be SAN more than NAS, surely
[23:45:28] <sommerfeld> depends on how you define SAN and NAS.
[23:46:04] <sommerfeld> if you define "SAN" as "fiberchannel", it's NAS.  if you define "SAN" as "disks over a network", it's SAN.
[23:47:00] <dsch04> Sorry, I was typing via a mirror.
[23:47:01] <holcomb> san = blocks, nas = files?
[23:47:19] <dsch04> I believe the definitions are rather fluid
[23:47:35] <tomww> ask wikipedia - it never lies
[23:48:04] <dsch04> Well, the server holding the flac files will definitely be a NAS
[23:48:16] <dsch04> Accessible via NFS and Samba
[23:48:37] <dsch04> The other server will just export iSCSI LUNS
[23:49:00] <RElling> I'd go for something with an NForce 680a or similar (12 SATA ports on the mobo, nv_sata driver in SXCE)
[23:50:59] <sickness> RElling: omg, wich mobo has really all of that? =)
[23:52:44] <sommerfeld> dsch04: like "switch", it's a term that has been captured by marketing and is therefore not useful.
[23:52:44] *** natefoo has left #opensolaris
[23:53:23] *** yippi has quit IRC
[23:53:40] <dsch04> RElling, can you recommend a board?
[23:55:12] *** estibi_ has joined #opensolaris
[23:56:13] <RElling> I'll let NVidia do the references, http://www.nvidia.com/page/nforce_600a.html
[23:56:30] <RElling> or tigerdirect, http://www.tigerdirect.com/applications/searchtools/item-Details.asp?EdpNo=2853609&sku=A455-2323
[23:56:45] *** dduvall has joined #opensolaris
[23:57:38] <RElling> for the anti-clickers, Asus L1N64-SLI WS NVIDIA Dual Socket CEB Motherboard
[23:57:42] <dsch04> RElling, thx
[23:57:42] <sickness> heh dual socket :/
[23:57:58] <sickness> I'd prefer a 1 socket board, anyway tnx :)
[23:58:46] *** estibi has quit IRC
[23:59:36] <RElling> I guess you have to be in the "multimedia enthusiast" market segment to really enjoy it :-)





top