[00:00:21] <CIA-26> sangeeta: 6601771 ipfil_sendpkt causes panic with punchin
[00:00:58] *** peemus has quit IRC
[00:04:33] *** tombhad has joined #opensolaris
[00:04:43] *** nostoi has quit IRC
[00:10:34] *** tinman2k has quit IRC
[00:10:45] *** sarahj has quit IRC
[00:11:36] *** bengtf has quit IRC
[00:12:13] *** baumer11 has joined #opensolaris
[00:12:40] <baumer11> i'm working on getting my OS box to connect to a windows domain
[00:13:03] <baumer11> I get an error that the DNS domain of my OS server doesn't match my windows domain
[00:13:17] <baumer11> where do I set the domain of my OS box?
[00:15:44] <RElling> baumer11: there is a field in Launch -> Places -> Connect to server...
[00:15:51] *** mikefut has quit IRC
[00:16:32] *** jlc has joined #opensolaris
[00:19:27] <Teltariat> Quick question: does anyone know whether or not Solaris 10 8/07 has the bits to support booting from ZFS?
[00:19:34] <RElling> no
[00:19:40] <Teltariat> owsh
[00:19:46] *** tombhad-AC has quit IRC
[00:19:46] <RElling> NV only for now
[00:19:50] <Teltariat> :(
[00:20:14] *** vmlemon has quit IRC
[00:20:31] *** Cas1 has joined #opensolaris
[00:24:55] *** bondolo has quit IRC
[00:25:58] *** Mdx4 has quit IRC
[00:28:53] *** bondolo has joined #opensolaris
[00:28:59] *** jlc has quit IRC
[00:32:13] *** Cas1 has left #opensolaris
[00:33:20] *** Mdx4 has joined #opensolaris
[00:34:14] *** bengtf has joined #opensolaris
[00:36:14] *** baijiutong has quit IRC
[00:41:20] *** Mdx4 has quit IRC
[00:42:14] *** sarahj has joined #opensolaris
[00:42:46] *** Somethingelse has joined #opensolaris
[00:43:51] *** Dar_HOME has quit IRC
[00:44:11] *** Mdx4 has joined #opensolaris
[00:44:52] *** cmihai has quit IRC
[00:48:34] *** cmihai has joined #OpenSolaris
[00:48:39] *** mog has joined #opensolaris
[00:50:37] <mog> morning
[00:51:35] <mog> where does nexenta fit in with project Indiana?
[00:52:09] <wesolows> the way a filing cabinet fits into a norfolk pine
[00:52:54] <mog> really, it was my understanding at the very least indiana would be a working distro by itself, i mean i wouldnt make same comparison between ubuntu and debian
[00:53:18] *** Mdx4 has quit IRC
[00:53:27] *** NikolaVeber has quit IRC
[00:53:40] <wesolows> nexenta is a product of some company whose name I forget (might be nexenta)
[00:54:03] <wesolows> indiana is a name being used by Sun marketing to cover a large number of opensolaris projects
[00:54:08] *** noyb has quit IRC
[00:54:19] <wesolows> some people at Sun would like to see indiana become a second Sun distribution as well
[00:54:40] <elektronkind> SunSoft Solaris Desktop and Server editions!
[00:55:09] <kjetilho> Solaris ME!
[00:55:23] <mog> from listening to ian talk about indiana it sounded like a core distro, much like debian
[00:55:36] <jbk> Javis :)
[00:56:00] <RElling> cool.  Adaptec storage manager agent uses SMF
[00:56:05] <Gman> mog, that's the idea
[00:56:06] <Triskelios> it's mostly assembling the output of a bunch of projects at this stage
[00:56:19] <Gman> right now it's prototype quality
[00:56:34] <mog> is there anyway to build indiana yet or is nexenta closest thing to it till initial release?
[00:56:57] <Gman> mog, nothing yet, later this month
[00:57:13] <mog> great way to start an free software project.....
[00:57:20] <Triskelios> mog: I imagine it'll be more closely related to SXCE than Nexenta
[00:57:42] <Gman> mog, hrm?
[00:58:10] <mog> that sucks as i find the solaris userland pales in comparison to the gnu one
[00:58:23] <mog> generally people release code then announce
[00:58:32] <mog> i feel like i have been hearing about indiana for some time now
[00:58:36] *** myrkr has joined #opensolaris
[00:58:47] <alanc> but then you're stuck with their decisions, and don't get a chance to shape it
[00:58:50] <Gman> mog, more or less all the code is out there already, someone just has to do the work to pull it together
[00:59:04] <Triskelios> mog: the SX userland is much nicer than S10's was
[01:00:30] <CIA-26> dougm: 6546230 sharemgr: "nosuid=false" caused extra comma in share command's output, 6570719 share command cores dump when /etc/dfs/sharetab and /etc/dfs/dfstab are out of sync, 6586857 sharemgr: libshare double-frees xml node on teardown
[01:00:37] *** baumer11 has quit IRC
[01:00:42] <jbk> i think the two biggest things are: arrow key support + default $PATH so that stuff isn't hidden -- make those defaults and that'd go a long way towards addressing gripes
[01:01:58] <mog> darn, i wanted to reinstall something on my lappy today
[01:02:07] <mog> guess ill go with debian
[01:02:28] <jbk> why not sxce?
[01:02:53] <cypromis_> lesbian linux ?
[01:04:42] <mog> heh
[01:05:12] <mog> jbk im to comfortable with the gnu userland, gnu tar, gnu make, gnu ls
[01:05:12] <mog> etc
[01:05:23] <mog> i was tempted to try nexenta
[01:05:28] <mog> but feel its gonna be left behind
[01:05:31] <wesolows> what's wrong with Solaris?
[01:05:34] <mog> nothing
[01:05:39] <Triskelios> mog: all of those ship with SXCE (but with g prefix)
[01:05:42] <mog> its my preference of the gnu userland
[01:05:45] <mog> not a complaint
[01:05:49] <wesolows> You said you prefer the GNU userland; are there bugs or RFEs that are bothering you?
[01:06:06] <mog> ?
[01:06:07] <Triskelios> if you're really twisted you could run Debian in a zone
[01:06:16] <wesolows> BTW, in recent versions of SX, you can add /usr/gnu/bin to your path ahead of /usr/bin and get many of the GNU utilities as your defaults.
[01:06:27] <mog> that is very attractive to me Triskelios  if the zones are as fast as people say
[01:06:35] <Triskelios> wesolows: a lot of that is still in /usr/sfw
[01:06:48] <wesolows> Triskelios: Then he should file bugs to have it moved, no?
[01:07:16] <wesolows> As I said, if you can say "I'm not using Solaris because of bugs/RFEs X, Y, and Z" that's useful for Sun and perhaps for the rest of the OpenSolaris community as well.
[01:07:24] *** Somethingelse has quit IRC
[01:07:50] <wesolows> Just saying that the userland "pales in comparison to the GNU userland" makes you sound like a dogmatic pisspot full of unfounded bias.
[01:08:02] <mog> no im totally being that way
[01:08:03] <Triskelios> wesolows: well, either way it's just a subdir, the real solution is fixing PATH
[01:08:13] <mog> there is no reason for not liking their userland
[01:08:18] <mog> im just comfortable
[01:08:23] <wesolows> Triskelios: Not true.  /usr/gnu/foo is generally equivalent to /usr/sfw/gfoo.
[01:08:30] <mog> so i should get CE version?
[01:08:34] <Triskelios> mog: zones have almost no overhead
[01:08:34] <mog> and try moving paths
[01:08:36] <wesolows> Triskelios: /usr/gnu has a very specific purpose, very different from what /usr/sfw is/was.
[01:08:53] <Triskelios> wesolows: ah
[01:09:17] <mog> also is blastwave the best place to get free software from ?
[01:09:20] <Triskelios> wesolows: sorry, I misunderstood then
[01:09:27] <wesolows> Zones are not hardware virtualisation nor paravirtualisation.  They are (though the comparison does Zones an injustice) more similar in performance terms to jails; there is basically zero overhead.
[01:09:40] <jbk> i have been contemplating proposing a simply utility to control 'personalities' (essentially preconfigured user defaults for $PATH, etc.)
[01:09:50] <mog> well i thought linux emulation would bare some penality wesolows
[01:10:09] <jbk> so you could select from a list the desired behavior (gnu userland, posix, traditional solaris, bsd, etc.)
[01:10:16] <Triskelios> mog: I have been preferring spec-files-extra for a while, but there are no automated builds yet (so you would have to generate the packages)
[01:10:20] <wesolows> mog: BrandZ has some performance hit, but it's quite small.  Check with the BrandZ community group to see if they've published anything on it.
[01:10:25] <RElling> jbk: lots of prior art there, including my LISA paper from 1992
[01:10:26] <mog> is there any difference between cd vs dvd version? i mean can i upgrade those things?
[01:10:27] <cmihai> You mean like "Linux Kernel Personality"? Heh :-)
[01:10:28] <Triskelios> mog: as a repository of Free software, I mean
[01:10:44] <wesolows> The figure I was hearing before it was integrated was 10%, but it may be better/worse in practice now.
[01:10:46] <jbk> RElling: i figured i wasn't the first to think of it :)
[01:10:59] <cmihai> mog, you only need the DVD :-)
[01:11:08] <cmihai> It's more... convenient.
[01:11:15] <RElling> of course, I was extending John Furlani's work from LISA in 1991
[01:11:24] <cmihai> Be it JumpStart or LiveUpgrades from mounted iso images...
[01:11:47] <mog> oh its 6 cds....
[01:11:50] <mog> i thought it was one
[01:11:56] <cmihai> no..
[01:12:01] <wesolows> which is why the DVD is more convenient :-)
[01:12:02] <cmihai> Just get the DVD, then liveupgrade
[01:12:03] <RElling> GUIs pretty much obviate the need for *most* folks
[01:12:05] <mog> and you could do something similar to apt-get
[01:12:11] <mog> mmhm
[01:12:14] <cmihai> mog, read this: http://unixsadm.blogspot.com/2007/08/liveupgrade-sxce.html
[01:12:24] <wesolows> That would be a great feature for someone to implement in their distribution.
[01:12:31] <cmihai> Initial install can be done via CD, DVD or JumpStart.
[01:12:36] <wesolows> I was encouraging someone here to do that yesterday, I believe.
[01:12:46] <Triskelios> mog: there isn't a full repository yet, that is something indiana is supposed to address
[01:12:49] *** Capt` has joined #opensolaris
[01:12:55] *** Capt` has left #opensolaris
[01:13:04] <mog> mmhm
[01:13:09] *** sleepcat has joined #opensolaris
[01:13:16] <mog> i wish i had heard about this stuff a few months from now
[01:13:23] * mog is very impatient
[01:13:34] <Triskelios> I'm pretty happy with SXCE on my laptop
[01:13:49] <Triskelios> I also have Debian etch in a zone
[01:13:57] <mog> nice
[01:14:00] <wesolows> Impatience with open source without a willingness to contribute effort toward achieving the object of that impatience is not really constructive.
[01:14:03] <mog> that is what i would need
[01:14:14] <mog> if i ever had a frustration i could jump there
[01:14:24] <mog> its not very helpful for opensource project to not be released...
[01:14:28] <Triskelios> I barely ever boot the zone except to try to get skype working, though...
[01:14:33] <wesolows> Blame Sun marketing for that.
[01:14:36] <jbk> i did see that ipkg is able to do a successful install now
[01:14:58] <mog> heh, i think i will be very happy with sxce and debian zone
[01:15:02] <mog> if its not too slow
[01:15:07] <Triskelios> jbk: URL?
[01:15:18] 
[01:15:20] <jmcp> mog: why would it be slow?
[01:15:22] <bda> Triskelios: Docs to installing the Debian zone?
[01:15:33] *** pitty has left #opensolaris
[01:15:35] <Triskelios> bda: google "etch-zone"
[01:15:50] <bda> Triskelios: You are defeating my laziness with your google.
[01:15:53] <mog> debian zone in emulation
[01:15:57] <jbk> Triskelios: it's on the pkg-discuss list
[01:16:12] <bda> Triskelios: Sweet.
[01:16:31] <Triskelios> jbk: ah... I should probably get on that
[01:16:34] <jbk> though i suspect that doesn't include all the installer bits -- i'm guessing there was some manual lifting to get the system to the point where pkg could run
[01:16:39] <wesolows> To see that no one has done it and become annoyed with some marketing team's vague announcement rather than decide to go do it yourself is unhelpful.
[01:16:40] <mog> thank you that is awesome
[01:16:43] <wesolows> jbk: Yep.
[01:17:08] <wesolows> jbk: I figure it's about a week's work.  I can't say how I know this, but I'm 100% certain it can be done in that time.
[01:17:09] <sleepcat> how stable is SXCE 73 compared to lets say solaris 10 u4?
[01:17:10] <mog> well i intend to help now that i can
[01:17:47] <wesolows> sleepcat: Stability is per-interface, so it's hard to say.  Certainly it's more likely for incompatible changes to be made from one build to the next in SX, since it's not a "release" yet.
[01:18:00] <Triskelios> wesolows: guess he meant panic-wise
[01:18:18] <wesolows> That's not stability, it's availability.
[01:18:50] <sleepcat> wesolows: how available is SXCE 73 compared with u4?
[01:18:58] <mog> at the very least ill add the packages i have to build from source to blastwave for others
[01:19:06] <wesolows> There are occasionally bad SX builds that will panic/fail on certain hardware or in certain configurations.
[01:19:30] <wesolows> If you don't have such a build, or don't have that configuration, you'll probably find SX to be at least as good as S10.
[01:19:41] <sleepcat> can I just download the first CD and see if it works?
[01:20:00] <wesolows> In short, YMMV, but since bugs are fixed in SX before they're fixed in patches (and well before those patches are rolled up into the next update), SX should in theory be better.
[01:20:03] <Triskelios> mog: consider contributing to SFE, their packages integrate a lot of better into the OS
[01:20:11] <Triskelios> mog: and are easy to create
[01:20:17] <mog> sfe?
[01:20:18] <wesolows> sure, you can do whatever you want.  Why not just d/l the 4th CD and see how well that works?
[01:20:20] <elektronkind> the problem with SX is... no patches ;)
[01:20:31] <stevel> i'd say the beauty of SX is... no patches
[01:20:32] <wesolows> elektronkind: The great thing about SX is... no patches. :-)
[01:20:53] <wesolows> patches are where the sausage factory aspect of Solaris comes in
[01:20:54] <Triskelios> mog: spec-files-extra, it's an offshoot of the JDS (gnome) packaging effort
[01:20:59] <jbk> heh
[01:21:02] <mog> heh , acidently started downloading sparc build dont think that would work well on my thinkpad
[01:21:08] <wesolows> if you see how they're made (and tested), you'll never consider using them again
[01:21:19] <sleepcat> wesolows: would it be a good idea to have the solaris CD 1 install the OS plus the JDS and not much else?
[01:21:27] <stevel> we need a "The Jungle" for RPE and patch generation
[01:21:32] <mog> Triskelios, is this going to work with the new packaging system?
[01:21:39] <mog> or yet another one
[01:21:40] <wesolows> sleepcat: Sure, all you'd need is a way to fit 3GB on a CD.
[01:21:49] <wesolows> sleepcat: Thankfully, we already have that; it's called a DVD.
[01:21:50] *** ichigo has quit IRC
[01:21:54] <stevel> mog: SFE is a build system for generating SVRv4 packages
[01:21:57] <mog> ahh
[01:22:02] <Triskelios> mog: it produces standard SVR4 packages, so I imagine it will work with anything in the future
[01:22:09] <mog> ooh rpm like
[01:22:12] <Triskelios> mog: it uses the same spec file format as RPMs
[01:22:16] <stevel> it's like rpmbuild & spec files
[01:22:17] <mog> oh well could be worse
[01:22:24] <stevel> huh. all 3 of us basically just said the same thing
[01:22:29] <sleepcat> does anyone have screenshots of SXCE 73?
[01:22:31] <Triskelios> lol
[01:22:32] <mog> i wonder why ian didnt fight for .dpkg
[01:22:33] <elektronkind> wesolows: so what to do when there's a nice feature or bug fix in, say, the e1000g driver? Reinstall or BFU a production server in order to get it?
[01:22:36] <wesolows> there's no fundamental difference between rpm and pkg*
[01:22:36] <Triskelios> sleepcat: I can take one
[01:22:38] <mog> i mean the one he made..
[01:22:40] <wesolows> they're all just packages
[01:22:44] <mog> true
[01:22:51] <wesolows> elektronkind: You could LU
[01:22:54] *** sahafeez has quit IRC
[01:23:00] <wesolows> elektronkind: You could cluster
[01:23:23] <Triskelios> sleepcat: http://trisk.acm.jhu.edu/snv_73.png
[01:23:26] <elektronkind> LU plausible. clustering just for that? silly.
[01:23:29] <wesolows> elektronkind: And if you don't have 24x7 availability requirements, you could use a maintenance window which you'll need anyway...
[01:23:49] <wesolows> just for that, silly, I agree...but if it's production you might be clustering already...aren't you? :-)
[01:24:08] <elektronkind> in a horizontal sense, yes
[01:24:44] <wesolows> if the bloody installer didn't take 2 hours to do its job, reinstalling would be pretty attractive actually
[01:24:45] <Triskelios> mog: dpkg is actually kind of fragile because the manifest is a flat file
[01:24:57] <sleepcat> the icons are a lot better
[01:24:58] <wesolows> reinstallation is, after all, the simplest and most reliable way to get a box into a known state
[01:25:01] <mog> Triskelios, is there a repo to contribe sfes to?
[01:25:04] <elektronkind> wesolows: that's how I'm going from pre-s10u4 to u4
[01:25:22] *** WickedWicky has quit IRC
[01:25:25] <Triskelios> mog: yeah, SFE is actually a subversion repo at the moment
[01:25:30] <elektronkind> in fact, my main research NFS server is being jumped to s10u4 now
[01:25:35] <mog> spiffy
[01:25:48] <elektronkind> need iscsitgt goodness on it
[01:26:00] <elektronkind> don't want to hunt and peck with pkgadd
[01:26:01] <Triskelios> I've migrated all the programs I usually build from source to SFE
[01:26:05] <mog> lots of em
[01:26:32] <Triskelios> (there's some games I still have backlogged)
[01:27:34] <mog> is there any way to have suspend just work? currently i close my laptop and it suspends to ram, i saw some scripts that will take down your nic etc and then allow you to sleep, and then run said scripts when you wake backup, but is there a way to have that automagically?
[01:27:54] <Pietro_S> Triskelios: which games, cause I'm working on few of them?
[01:27:57] <sleepcat> can I upgrade from u4 to SXCE 73 easily?  or is it risky?
[01:28:00] <Triskelios> suspend no workie on SX yet =\
[01:28:08] <Triskelios> tesla-dev is working on it
[01:28:09] <mog> ahh no worries
[01:28:28] <mog> i usually only suspend when i travel
[01:28:48] <elektronkind>         SUNWiscsitgtu....................done.  1755.41 Mbytes remaining.
[01:28:50] <elektronkind> yays
[01:29:59] <Triskelios> Pietro_S: ones I can think of at the moment are tuxracer, frozen-bubble (haha penguin themed), and whatever other popular quake3 derivatives there are
[01:30:09] <Pietro_S> mog: so far tesla-devs made good job - as watt consuption was smallest compared aginst linux and windows
[01:30:23] <Pietro_S> Triskelios: extreme tux racer?
[01:30:26] <mog> spiffy
[01:30:39] <mog> my laptop gets good suspend time
[01:30:42] <mog> i think about a day
[01:30:45] <mog> from full bat
[01:30:53] <Triskelios> Pietro_S: never heard of that
[01:31:28] <Triskelios> ah, a new fork
[01:31:49] <Pietro_S> Triskelios: that's some fork from tux racer, if I wouldn't have cold last dayes it would be already in SFE ;-)
[01:32:51] <Pietro_S> good night
[01:33:18] *** Pietro_S has quit IRC
[01:33:43] * Triskelios runs to dinner
[01:35:43] <mog> im reading this doc on setting up zone for debian, both examples show binding to a real nic, you cant create tun tap like device to route with? or just not there?
[01:40:37] *** cypromis_ has quit IRC
[01:40:45] <sleepcat> i have ssh enabled sometimes.  Is there a way to block a login via ssh after multiple login failures over the net?
[01:40:46] <Triskelios> mog: well, I wrote one of them, and honestly I've never tried, but the opensolaris qemu site has some info for tun/tap, and there might be other instructions for zones in general
[01:40:51] *** eryc has joined #opensolaris
[01:41:02] <Triskelios> mog: wrote one of the docs, I mean
[01:41:24] <mog> uh hu, it would just be nice to have networking just work as i move between work and home
[01:43:36] <mog> you realize i was asking about having debian running inside of a zone on opensolaris with virtual nic to connect out my real interfaces, much like i would run an os inside of qemu right?
[01:44:04] *** sleepcat has quit IRC
[01:48:30] *** RealWickedWicky has joined #opensolaris
[01:50:21] *** comay_ has joined #opensolaris
[01:50:47] *** kadath_ has joined #opensolaris
[01:50:55] <delewis> heh, the direct IO & ZFS thread on zfs-discuss reminds me of the discussions regarding the advantages/disadvantages of using raw devices for tablespaces.
[01:51:36] <delewis> it seems with every leap in filesystem technology, there's always some particular aspect of a particular workload that's left behind, which causes a huge amount of discussion.
[01:51:52] <RElling> if Oracle had decent builtin-RAS we wouldn't be having these discussions ;-)
[01:52:16] <delewis> this is true, and I've been wondering myself whether or not Oracle would implement some aspects of ZFS in ASM.
[01:52:24] <delewis> namely, checksums and transactions (if they don't have that)
[01:52:34] * delewis isn't a fan of ASM
[01:52:48] <delewis> it's always struck me as logical volume management for the DBA.
[01:52:54] <RElling> they do checksums, but people tend to turn that off first, especially performance benchmarkers
[01:53:05] <delewis> RElling: really? is that a 10g feature?
[01:53:13] <delewis> I'm surprised I hadn't heard anything about this.
[01:53:17] <palowoda> I'm sorry Oracle working with Sun on a collabrate level?
[01:53:36] <delewis> palowoda: no, Oracle keeping ASM competitive. :-)
[01:53:54] <delewis> if ZFS offers as easy managability as ASM, that limits the advantages of ASM.
[01:53:55] <palowoda> Ah an Oracle only problem.
[01:54:22] <RElling> isn't Oracle behind btrfs?
[01:54:36] <delewis> well, the ZFS/Direct IO discussion on zfs-discuss is applicable to most RBDMSs.
[01:54:59] * delewis googles btrfs
[01:55:12] <RElling> true, but ARC is very different than the (UFS) page cache, and ZFS doesn't have the single writer lock problem to begin with
[01:55:27] <delewis> http://oss.oracle.com/projects/btrfs/
[01:55:28] <delewis> interesting.
[01:55:38] <delewis> RElling: right
[01:56:04] <RElling> so it isn't really clear the the caching "problem" solved by UFS direct I/O is also a ZFS problem
[01:56:16] <delewis> RElling: judging by the 'features' of btrfs from that link, it sounds quite ZFS-ish.
[01:56:17] <elektronkind> anyhow, Solaris 10, per Oracle, is their "preferred" 64bit platform
[01:56:18] <RElling> in the long run, though, it may be easier to JFDI
[01:57:02] <delewis> elektronkind: right, I always find it somewhat entertaining when someone has the impression that Oracle is pushing Linux as its primary platform.
[01:57:16] <elektronkind> yeah, OracleOS
[01:57:31] <RElling> Larux? :-)
[01:57:40] <delewis> as an Oracle employee explained to me, Linux is a buzzword and for most people it has a much lower adoption cost when a customer is looking at Oracle. Of course, that's now irrelevant given the 'free-ness' of Solaris.
[01:58:26] <jbk> oracle does do checksums, but i believe they are stored with the block (unlike zfs)
[01:58:58] <jbk> i just find some of the outdated 'tribal knowledge' frustrating
[01:59:00] <delewis> does Oracle do checksums for all types of tablespaces? or is this an ASM-only feature?
[01:59:28] <jbk> dbas will still (for example) turn off async i/o, or refuse to use large SGAs on large ram systems
[01:59:37] * delewis sighs
[01:59:40] <jbk> because 'there were issues with oracle 8'
[01:59:46] <jbk> 'well are we running oracle 8?'
[01:59:47] <jbk> 'no'
[01:59:52] <jbk> 'how is that relevant?'
[01:59:53] <delewis> that's a problem of DBAs applying pre-Oracle 8i methodologies.
[02:00:05] <RElling> would love to continue the chat, but gotta go feed the horses... l8r
[02:00:06] <jbk> but it's still very common
[02:00:21] <delewis> later, RElling.
[02:00:38] <jbk> i'm still trying to figure out at work what group is driving all of this linux adoption
[02:00:43] <jbk> people still feel it's cheaper
[02:00:54] <jbk> which of course it's not (as they're always referring to redhat w/ a support contract)
[02:01:01] * delewis tries to come up with an effective benchmark that'd illustrate whether or not the advantages of direct IO apply to ZFS
[02:01:43] <delewis> obviously, you'd want something with a lot of COMMITs, and I guess, somewhat lengthy transactions.
[02:02:20] <jbk> the last place i worked
[02:02:25] <jbk> they ran all their dbs on vxfs
[02:02:27] <delewis> and something that'd utilize a large SGA to see if ZFS's or more generally Solaris's ability to free up the ARC is efficient.
[02:02:40] <jbk> odm
[02:03:44] <jbk> it finally took showing how in just one instance, they were looking at having to buy over $1mil in extra hardware to buy enough cpu horsepower to get equivalent performance
[02:04:10] <jbk> but then ASM came, and now all the DBAs were dead set on using nothing but ASM
[02:04:16] <delewis> yeah
[02:04:33] <delewis> like I said, ASM strikes me as volume management for the DBA.
[02:04:56] <delewis> and that raises the question of whether or not the DBA should even be doing volume management (ASM gives that responsibility to them)
[02:05:27] <delewis> doing SELECTs for volume status/statistics is just wrong on so many levels. :-)
[02:05:31] <elektronkind> couldn't you toss them zvols ?
[02:05:38] <elektronkind> sick ASM on that
[02:05:47] <delewis> yeah, that'd work.
[02:06:14] <delewis> doubt they'd get all the features of ASM (just as using LUNs from a SAN and ZFS), though.
[02:06:21] <delewis> let them sort that out :-)
[02:06:30] <elektronkind> "touch ANYTHING under /dev/rdsk and I kill you"
[02:06:35] <jbk> well it does provide for better management
[02:06:51] <delewis> ASM does have some advantages that DBAs find attractive, like the tight integration between table statistics, etc. and tablespace volumes.
[02:06:58] <jbk> since kinda like how zfs means you no longer have to play 'filesystem size' genie, they don't have to play table/index/etc. size genie
[02:07:25] <jbk> and like zfs, makes physical layout less of a concern
[02:07:30] <delewis> jbk: right.
[02:07:42] <delewis> and that makes those table statistics, etc. that ASM gives you irrelevant in most cases.
[02:08:26] <delewis> like I said, in every leap of filesystem technology, there's always this kind of discussion. With raw tablespaces, it was filesystems being efficient with their physical layout, and nowadays with ZFS, the physical layout is irrelevant.
[02:09:07] *** uebayasi has joined #opensolaris
[02:09:30] <elektronkind> oh for cyring out loud
[02:09:51] <elektronkind> I mistakingly reboot the one old server in my farm which requires a keyboard in order to get past BIOS
[02:10:03] * elektronkind puts that box at the top of the hardware refresh list
[02:10:03] <delewis> elektronkind: :-)
[02:10:06] *** sarahj has quit IRC
[02:10:18] * delewis would put that on the top of the 'shoot-on-sight' list
[02:10:37] <elektronkind> it's an old Dell 1600 series "server"
[02:10:38] <sommerfeld> "keyboard failure.  press f1 to continue"
[02:10:45] <elektronkind> don't ya love it
[02:10:51] <delewis> elektronkind: basically, a peecee sold under a 'server' brand, right?
[02:10:56] <delewis> hence your quotes, I presume. :-)
[02:10:57] <elektronkind> pretty much
[02:11:05] <elektronkind> its getting a surplus v20z
[02:12:36] * delewis sighs
[02:12:41] <delewis> looks like Update Manager has frozen.
[02:12:59] <delewis> every now and then I just have an itch to see if Update Manager/smpatch/etc. can redeem themselves.
[02:13:22] <elektronkind> pca++
[02:13:52] <delewis> sure, but I do like the ability to see the SunSolve page for the patch to see what it fixes, as I try not to install patches blindly.
[02:13:53] *** medar has quit IRC
[02:14:05] <delewis> that's probably about the only feature of Update Manager that I appreciate.
[02:14:06] <holcomb> pca -r
[02:14:11] <delewis> (yes, I know about pca -r)
[02:14:26] <elektronkind> I just feed it a list of self-approved patches
[02:14:41] <elektronkind> better than smpatch roulette at least
[02:14:58] <delewis> elektronkind: so you just get the patch list, look up each on SunSolve, and feed it the 'approved' ones? (that's what I used to do)
[02:14:59] <elektronkind> although I admit it's been a while since I've tried smpatch.
[02:15:04] <holcomb> smpatch was nice in its first iteration
[02:15:13] <delewis> holcomb: when it was PatchPro?
[02:15:17] <elektronkind> delewis: essentially that's the end result, yes
[02:15:32] <delewis> or the generation after that, but before it started using the CACAO bullshit? :-)
[02:15:36] <holcomb> yes
[02:15:47] <holcomb> cacaoadm disable; cacaoadm stop
[02:16:07] <delewis> does that let Update Manager and smpatch still work?
[02:16:16] <delewis> I thought they depended on CACAO, nowadays.
[02:17:10] <delewis> what's a safe way to kill Update Manager? pkill? or have faith that 'Cancel' will actually kill it?
[02:17:56] * delewis wonders what it is with Java applications randomly freezing
[02:18:10] <delewis> I ran into a similar issue with the DB2 v8.1 installer and S10u4.
[02:18:14] *** JBeck has quit IRC
[02:23:52] *** sleepcat has joined #opensolaris
[02:25:33] *** rubymonk has quit IRC
[02:28:15] <palowoda> Actually you can remove SUNWcacaort safely.  It's not needed.
[02:28:43] <delewis> is cacao the reason why smpatch freezes, though?
[02:29:21] *** kloczek has quit IRC
[02:29:54] <palowoda> You don't need smpatch either.
[02:30:02] <delewis> well, sure. :-)
[02:30:14] <delewis> I usually just install the recommended+security patchset.
[02:30:59] <palowoda> Oh that old stuff.
[02:31:19] <delewis> it works, and usually patches all the security holes.
[02:31:27] <palowoda> Sometimes.
[02:31:34] <delewis> I'll install other patches under the advice of a vendor.
[02:31:51] <palowoda> I haven't seen a vendor here.
[02:32:07] <delewis> occasionally, like on my workstation, I do like to install other patches, namely patches that are X11- or JDS-related.
[02:32:27] <palowoda> On old Gnome 2.6 based stuff?
[02:33:10] <palowoda> Are they really support that?
[02:33:21] *** mog has left #opensolaris
[02:33:44] *** mog has joined #opensolaris
[02:34:00] <delewis> whatever version Solaris 10 shipped with, yes.
[02:34:24] <palowoda> Must be some old weird tree.
[02:34:37] *** Atomdrac1e has joined #opensolaris
[02:35:02] *** mog has left #opensolaris
[02:35:37] *** mog has joined #opensolaris
[02:36:39] *** Atomdrache has quit IRC
[02:36:47] *** Atomdrac1e is now known as Atomdrache
[02:36:55] *** mog has left #opensolaris
[02:37:08] *** palowoda has quit IRC
[02:38:36] *** jcsmith has quit IRC
[02:38:49] * elektronkind drums his fingers as he waits for a x4100 ilom to upgrade from 1.0.0 to 1.1.8
[02:39:10] *** palowoda has joined #opensolaris
[02:39:45] <sleepcat> i downloaded cups from blastwave and now i want to add my usb printer.  I need to do a lpadmin -p myprinter -v xxxxxx.  What do I put in xxxxx for my usb printer?
[02:40:17] <delewis> elektronkind: least you aren't an x2100 or x2200 user that's having to patch the LOM because of security reasons. :-)
[02:40:28] <elektronkind> heh, I saw that alert
[02:40:55] *** mog has joined #opensolaris
[02:40:57] <elektronkind> never knew there was an ipmi with smtp relay RFC ;)
[02:41:42] <palowoda> If opensolaris is dependent on patching the hardware than don't recommend Sun hardware.
[02:42:03] <elektronkind> opensolaris isn't dependent on it
[02:42:24] <elektronkind> it's just the ilom, which is actually ppc linux
[02:42:25] <palowoda> Can you remove patch and use opensolaris to update the lom?
[02:42:35] *** comay_ has quit IRC
[02:42:41] <palowoda> Ah ilom.
[02:42:50] <palowoda> ppc linux?
[02:42:53] <elektronkind> yup
[02:43:04] <elektronkind> that what ilom be, under the hood
[02:43:44] <palowoda> Missed that, was off line for a sec.
[02:43:53] <elektronkind> [(flash)root@SUNSP0003BACD6A3E:~]# uname -a
[02:43:54] <elektronkind> Linux SUNSP0003BACD6A3E 2.4.22 #1 Sat Apr 21 11:28:21 PDT 2007 ppc unknown
[02:44:47] <elektronkind> it lets me reassure my crazy CIO that, yes, we run linux
[02:45:30] <palowoda> Tell him your proud.
[02:46:14] <elektronkind> I'll be sure to tell him my proud
[02:47:40] *** sleepcat has left #opensolaris
[02:49:38] *** charlesnw has quit IRC
[02:50:11] *** cmihai has quit IRC
[02:53:19] <stevel> elektronkind: i laughed, fwiw
[02:57:25] *** Megaf has joined #OpenSolaris
[02:57:32] *** gavagai_ is now known as _gavagai
[02:57:39] *** tombhad has quit IRC
[02:57:55] *** tombhadAC has joined #opensolaris
[02:58:36] *** yarihm has quit IRC
[03:00:35] *** sleepcat has joined #opensolaris
[03:01:16] <sleepcat> how can I turn my mouse cursor pink in solaris 10 u4?
[03:02:39] <palowoda> You mean in Gnome Terminal to Edit-Current Profile and turn off the blink?
[03:04:15] <sleepcat> no, i'd like to change the colour to pink
[03:04:24] <Bartman007> sleepcat: rose colored glasses.
[03:04:36] <palowoda> Ah come on pink?
[03:04:55] <palowoda> You want color ls too?
[03:05:22] *** baijiutong has joined #opensolaris
[03:05:23] <sleepcat> palowoda: yes,
[03:05:30] <stevel> sleepcat: you can't in s10u4
[03:05:35] <stevel> you need snv_73 or higher
[03:05:38] <stevel> i think
[03:05:40] <stevel> maybe 72
[03:05:53] <sleepcat> can I print if I have a usb printer then?
[03:05:55] <stevel> see xwin-discuss
[03:06:02] <stevel> usb printers should be supported just fine, yeah
[03:06:15] *** stevel has quit IRC
[03:06:29] <sleepcat> how do I tell cups what my printer is in the -v variable?
[03:08:15] *** mog has left #opensolaris
[03:10:19] *** sleepcat has quit IRC
[03:14:40] *** victori_ has quit IRC
[03:15:33] *** victori_ has joined #opensolaris
[03:17:10] *** comay_ has joined #opensolaris
[03:19:25] <Triskelios> wait, snv_73 finally has libXcursor? why didn't anyone tell me?
[03:20:18] *** _mw46_ has quit IRC
[03:20:25] <palowoda> alanc announce this kind os stuff on xwindow-discuss all the time.
[03:24:20] *** medar has joined #opensolaris
[03:25:44] *** comay_ has quit IRC
[03:29:13] *** master_o1_master has quit IRC
[03:30:31] <Triskelios> oh, it's snv_74..
[03:30:58] *** baijiutong has quit IRC
[03:31:07] *** baijiutong has joined #opensolaris
[03:35:28] *** masta has joined #opensolaris
[03:36:01] <moazamraja> re
[03:36:25] <moazamraja> anyone got good advice on reducing (without reboot) the amount of CLOSE_WAIT connections I have running?
[03:37:35] <jamesd> unplug the network cable.. if no one is requesting opens .. there are no tasks   waiting to close ;-)
[03:38:32] <jamesd> better yet... and fool proof... pull the power plug(s) ...  all close waits will die..
[03:40:06] <sommerfeld> moazamraja: CLOSE_WAIT means that the peer closed the connection but a local process hasn't closed its end yet
[03:40:14] <moazamraja> yeah
[03:40:23] <sommerfeld> finding the local process which has it open and killing it is the usual way.
[03:40:30] <moazamraja> but I got a ton of em (190+), and i'd like to chop em
[03:40:40] <moazamraja> unfortunately, i believe that is Oracle :(
[03:40:41] <sommerfeld> pfiles should show who's the offender
[03:40:52] <sommerfeld> so stop & restart oracle
[03:40:58] * sommerfeld has no experience with oracle
[03:41:02] <myrkr> hmm, how do I get thread safe std::cout with studio 12?
[03:41:16] <myrkr> I'm confused
[03:41:19] <elektronkind> well, he'd want to stop/start tnslistener
[03:41:56] *** sleepcat has joined #opensolaris
[03:41:58] *** master_of_master has joined #opensolaris
[03:42:28] <elektronkind> people have long complained about tnslistener keep stuff in close_wait
[03:42:46] <elektronkind> moazamraja
[03:43:06] <elektronkind> there is a timeout on close_waits where the OS should reap them, iirc
[03:43:16] <sommerfeld> elektronkind: Uh, no.
[03:43:25] <sommerfeld> close_wait is a potentially long-term stable half-open state
[03:43:52] <sommerfeld> there are other tcp states which are transient (TIME_WAIT)
[03:43:55] <elektronkind> I thought it was that close_wait implied a FIN was received
[03:44:04] <elektronkind> ah, maybe I'm thinking of time_wait
[03:44:24] <sommerfeld> close_wait means we've received the FIN but haven't sent the FIN for the other direction
[03:44:33] <elektronkind> I see
[03:44:42] <sleepcat> i ran cfgadm on my machine and it says I have my printer on usb0/2.  How does that translate to /dev/ ?
[03:44:47] <elektronkind> does SO_REUSE have bearking on this at all?
[03:44:50] <elektronkind> er bearing
[03:46:43] <e^ipi> sleepcat: /dev/printers/0
[03:46:48] <e^ipi> at least on my box
[03:46:59] <e^ipi> or is it "printer" singular?
[03:47:01] <e^ipi> one of them anyways
[03:47:15] <sleepcat> e^ipi: thanks, i'll go check it out.
[03:47:48] <moazamraja> elektronkind: i'm trying to lower that timeout...but the close_waits are still there
[03:48:22] <elektronkind> are the close_waits causing a real problem, or are they just an annoyance?
[03:48:33] <moazamraja> we think they're causing a problem
[03:48:34] *** chrisso has joined #opensolaris
[03:48:43] <elektronkind> blink tns listener if you can
[03:48:44] <moazamraja> i.e., stopping other machines from being able to connect to the db
[03:49:14] <elektronkind> lsnrctl stop/start (as oracle)
[03:49:36] <elektronkind> unless you use SMF for that
[03:49:43] <moazamraja> we dont
[03:50:26] <sleepcat> so does this look correct?  ./lpadmin -p HP_DeskJet -v usb:/dev/printers/0
[03:51:38] <sleepcat> I'm making a FAQ for CUPS on solaris so newbies like myself won't ask so much...
[03:52:09] <elektronkind> CUPS also has that nifty UI thingy
[03:52:17] *** mschenck has quit IRC
[03:52:19] <e^ipi> s/nifty/frustrating & useless
[03:52:34] <sleepcat> elektronkind: it asks for my username and password repeatedly
[03:52:40] <e^ipi> sometimes I think they add broken features to things just to taunt you
[03:52:56] <e^ipi> GUI crap sucks generally anyways
[03:53:02] <sleepcat> the command line is my only hope. :-)
[03:53:03] <Triskelios> sleepcat: the web interface is very reliable, actually
[03:53:18] <e^ipi> clicking around until things start working is a retarded way of setting things up & is very prone to error
[03:53:41] <sleepcat> Triskelios: i can't get past the point where it keeps asking for my root password, I give the correct answer and it keeps asking on the web interface
[03:54:09] <sleepcat> I did the whole lppasswd thing
[03:55:35] <sleepcat> i'll probably stop here, because I've been trying to get printing to work for about 5 hours..
[03:55:40] <Triskelios> sleepcat: the authentication is configurable in the config file
[03:55:51] *** ferret_0567 has joined #opensolaris
[03:56:05] <sleepcat> Triskelios: i set it to None, and then cupsd freaked out when I did a restart
[03:56:50] <sleepcat> the only thing holding me back is the -v variable where I specify usb:/blablabla
[03:57:00] <sleepcat> on the command line
[03:57:14] <moazamraja> very odd
[03:57:21] <moazamraja> we did a tnslistner stop
[03:57:30] <moazamraja> and yet we still have 165 CLOSE_WAITs on 1521
[03:58:07] <Triskelios> I generally stick with the Solaris print system (presto helps)
[03:58:40] <sleepcat> Triskelios: I tried that first, i couldn't figure it out
[03:58:54] <sleepcat> i'm beginning to think that i'm too stupid to use solaris
[03:59:54] <e^ipi> nah, probably just too used to some other system
[04:00:01] <Triskelios> sleepcat: well, printmgr is not pretty but it seems pretty straightforward
[04:00:01] <e^ipi> that goes away eventually
[04:00:24] <Triskelios> I've only had to add one printer "manually" so far, though, thanks to presto
[04:00:35] <sleepcat> i have a usb printer so i think that adds to the complexity
[04:01:47] <Triskelios> all of my local printers are usb, but presto found them so I didn't have to
[04:02:56] <sleepcat> what is presto?
[04:03:00] <e^ipi> mine wasn't supported, I had to get the driver & PPD
[04:03:14] <sleepcat> e^ipi: do you have a usb printer?
[04:03:21] <e^ipi> yeah
[04:03:31] <Triskelios> sleepcat: the automatic printing setup tools
[04:03:50] <sleepcat> Triskelios: is that on solaris 10 u4 also?
[04:03:57] <Triskelios> don't know if they work in S10
[04:04:35] <sleepcat> e^ipi: do you use CUPS?
[04:04:41] *** jcsmith has joined #opensolaris
[04:04:50] <e^ipi> sure do
[04:05:29] *** jamesd has quit IRC
[04:05:30] <sleepcat> to add a printer one does lpadmin -p myprinter -v xxxx -P myprinter.ppd
[04:05:39] <sleepcat> what did you put under the -v ?
[04:06:17] <Triskelios> also, curiously, I just looked in my printers.conf and all the printers have an ipp URL?
[04:06:34] <e^ipi> IIRC i just put /dev/printers/0
[04:06:38] * elektronkind shakes his fist at mpxio
[04:06:39] <e^ipi> or printer
[04:06:44] <e^ipi> don't remember which
[04:06:49] <sleepcat> ok so i think i did this right
[04:07:06] <sleepcat> now did you specify usb:/dev/printers/0 ?
[04:07:19] <sleepcat> or was it file:/dev/printers/0 ?
[04:09:17] <e^ipi> no, just /dev/printers/0 i think
[04:10:02] <jmcp> e^ipi: what's your beef with mpxio?
[04:10:03] *** alobbs has quit IRC
[04:10:25] *** alobbs has joined #opensolaris
[04:12:02] <sommerfeld> my beef with mpxio is the human-unfriendly device names.
[04:12:06] <sommerfeld> starts, and ends there.
[04:12:45] *** alanc is now known as alanc-away
[04:13:11] <jmcp> sommerfeld: you mean to say that you can't remember a 64bit or 128bit hexadecimal target number?
[04:13:14] * jmcp faints from shock
[04:13:16] <jmcp> :-)
[04:14:40] <elektronkind> I can get over the dev names, it's mpathadm's argument structure that bugs me
[04:14:49] <e^ipi> jcsmith: huh?
[04:14:56] * elektronkind gets his xraids re-mpxioized
[04:14:58] <e^ipi> I don't have a beef with mpxio
[04:16:27] *** lloy0076 has joined #opensolaris
[04:16:30] <elektronkind> although, I am seeing something odd
[04:16:41] <elektronkind> short dev names for stuff that mpxio says is under its control
[04:17:12] <elektronkind> example: http://opensolaris.pastebin.ca/725055
[04:17:39] <elektronkind> the device names in the first two zfs mirror sets are to be expected, the second are not
[04:17:50] <elektronkind> but mpathadm shows the shorties as being mpath'd
[04:19:01] <jbk> evening
[04:19:40] <jmcp> elektronkind: the "shorties" are using a different sort of GUID
[04:20:26] *** masta has quit IRC
[04:20:27] <jmcp> elektronkind: http://www.jmcp.homeunix.com/~jmcp/WhatIsAGuide.pdf
[04:22:12] <Triskelios> jmcp: extra "e" =P
[04:22:17] <jmcp> dammit, added that again
[04:22:23] * jmcp coughs, goes to make a symlink
[04:22:37] <jmcp> done
[04:22:57] <jmcp> let's try that again::     http://www.jmcp.homeunix.com/~jmcp/WhatIsAGuid.pdf
[04:23:08] <elektronkind> got it :) thanks
[04:25:41] <Triskelios> nice cover photo
[04:26:00] *** jamesd has joined #opensolaris
[04:26:00] *** ChanServ sets mode: +o jamesd
[04:26:01] <jmcp> front or back?
[04:26:47] <moazamraja> g'damnit
[04:26:54] <elektronkind> this is an excellent preso
[04:27:01] <jmcp> thankyou :)
[04:27:03] <moazamraja> i even shut down the connecting clients and the server is still holding on to the CLOSE_WAITs
[04:27:22] <jmcp> Triskelios: I took the one at the back, the one at the front is stock from the corporate template
[04:27:33] <jbk> did the work to separate out the closed bits from the non-encumbered bits of mpxio ever get finished?
[04:27:39] <jmcp> yeah
[04:27:43] <jmcp> I need to update the presentation
[04:28:57] *** Shiv_1 has joined #opensolaris
[04:29:36] <jmcp> Triskelios: the photo at the end of the presentation is http://www.jmcp.homeunix.com/gallery2/main.php/v/jmcp/US_Trip_2006_2007/SantaCruz/_MG_2129.jpg.html
[04:30:21] <Triskelios> heh, ok
[04:30:23] <jmcp> sorry, this one - http://www.jmcp.homeunix.com/gallery2/main.php/v/jmcp/US_Trip_2006_2007/SantaCruz/_MG_2131.jpg.html
[04:32:36] <jbk> nice
[04:32:39] <jmcp> ta
[04:33:15] *** Fullmoon has joined #opensolaris
[04:35:11] * lloy0076 wonders if that is James surfing at the end
[04:35:30] <jmcp> nope, I just took the photo
[04:35:33] <jmcp> I can't surf
[04:36:56] *** sommerfeld has quit IRC
[04:37:32] <Bartman007> jmcp: a Fire25K should make a great surfboard.
[04:37:44] <Bartman007> E25K*
[04:37:53] <jbk> haha
[04:37:58] <jbk> kinda bulky i think
[04:38:01] <jbk> maybe the door
[04:38:08] <jbk> and heavy
[04:38:55] *** RainDoctor has joined #opensolaris
[04:44:37] *** ferret_0567 is now known as ferret_0567_linu
[04:44:50] *** ferret_0567_linu is now known as ferret_0567-linu
[04:45:12] *** jamesd has quit IRC
[04:54:13] *** sioraiocht has quit IRC
[04:54:25] <sleepcat> http://www.jmcp.homeunix.com/gallery2/main.php/v/jmcp/US_Trip_2006_2007/SantaCruz/_MG_2065.jpg.html
[04:54:32] <sleepcat> who is that hot little number?
[04:55:33] *** _gavagai has quit IRC
[04:55:55] <Gman> looks like a couple of birds to me ;)
[04:56:07] <Gman> [nice to see the location for next week's opensolaris summit]
[04:57:09] <jbk> sc is nice from what i recall (been about 7 years since i was there)
[05:01:59] <e^ipi> except for the san francisco airport, i've never been to california
[05:03:36] <Bartman007> e^ipi: SFO isn't that special :-P
[05:04:04] <jbk> it's nice, just unless you're an executive, or happen to get some pre-ipo options in a company that ends up going public (and can cash them out), it's pretty much impossible to live there and actually get ahead
[05:04:07] *** myrkr has quit IRC
[05:04:12] <moazamraja> sleepcat: the hottie with the glasses? yeah, that's jmcp.
[05:04:15] <flyingparchment> hmm, sshd seems to have some extraneous debug information left it..
[05:05:38] <flyingparchment> http://rafb.net/p/xuG8uo10.html < should i file a bug?
[05:05:58] *** gavagai_ has joined #opensolaris
[05:06:07] <moazamraja> phuckin eh. lsof isn't that useful in zones :(
[05:06:39] <delewis> I would say I'm surprised, but given how many private interfaces lsof digs in...
[05:10:59] *** jamesd has joined #opensolaris
[05:10:59] *** ChanServ sets mode: +o jamesd
[05:20:32] *** oninoshiko has joined #opensolaris
[05:23:21] *** RainDoctor has quit IRC
[05:24:15] *** polk__ has quit IRC
[05:25:20] *** polk__ has joined #opensolaris
[05:29:45] *** simford has joined #opensolaris
[05:36:55] <delewis> ah, installing the kernel patches has gotten to be so much fun.
[05:37:09] <jbk> heh
[05:37:22] <jbk> are the to the point where they also patch /bin/true? :)
[05:37:23] <delewis> it's almost to the point where doing a Live Upgrade is more sane than just installing a kernel patch if that patch falls within a given releaes (as do the latest patches).
[05:37:39] <delewis> jbk: http://sunsolve.sun.com/search/advsearch.do?collection=PATCH&type=collections&max=50&language=en&queryKey5=120011&toDocument=yes
[05:37:47] <delewis> look at the 'Special Notes' section for a laugh.
[05:38:15] <delewis> I've never seen that many notes for a single patch.
[05:38:37] <delewis> it makes me wonder if the previous patches were as complex and Sun just didn't tell you :-)
[05:38:59] <Bartman007> that's a lot of bugs fixed.
[05:39:04] <delewis> yep.
[05:39:18] <delewis> it's basically a huge backport from Nevada.
[05:39:28] <delewis> looks like most of the things are from snv_57 and onwards.
[05:39:42] <jbk> it'd be interesting to compare that to the list of files installed by whatever the smallest install cluster is (can't remember off the top of my head)
[05:39:44] <delewis> I see a few things from ~ snv_40
[05:39:56] <jbk> see which one is larger :)
[05:40:01] *** Megaf is now known as _Megaf
[05:40:05] *** _Megaf is now known as _Megaf_
[05:40:11] <delewis> jbk: recommended+security for Solaris 10/SPARC is close to 300MB, nowadays.
[05:40:16] <delewis> that kernel patch is almost 70MB.
[05:40:32] *** _Megaf_ is now known as _Megaf
[05:40:34] <sleepcat> http://opensolaris.pastebin.ca/725123
[05:40:39] <delewis> I'm not sure what just recommended is, as I always install recommended+security.
[05:42:21] <delewis> wow, the Solaris 10 patch cluster is much, much larger than the Solaris 8 patch cluster.
[05:42:21] *** _Megaf has quit IRC
[05:42:40] *** _Megaf has joined #OpenSolaris
[05:42:42] <delewis> I guess that would make sense, given Solaris 10 is a much larger release than Solaris 8 is, but that's also scary, given how long 8 has been supported vs. 10.
[05:42:59] * delewis thinks it's about time for a new minor release
[05:43:15] <sleepcat> maybe solaris 8 had better engineering?
[05:43:21] <sleepcat> less patches needed
[05:43:46] <sleepcat> you can only lay off so many before quality suffers
[05:44:13] <delewis> I wouldn't say 'better engineered'. It's just that Sun has quite a few customers (especially, large customers) that are still demanding Solaris 8 support, like the government.
[05:44:31] <delewis> and that's why I find it ironic that the Solaris 10 patch cluster is much larger than the Solaris 8 one.
[05:44:55] <delewis> that Solaris 10 kernel patch I referenced earlier is about 70MB, which would make up almost half of the Solaris 8 patch cluster.
[05:44:56] *** mschenck has joined #opensolaris
[05:45:17] <Bartman007> Open Source leads to bloat?
[05:45:22] * Bartman007 ducks.
[05:45:56] <delewis> if that were true, then how do you explain Oracle? :-)
[05:46:00] <sleepcat> solaris 8 was released at the hight of SUN?
[05:46:33] <oninoshiko> maybe solaris 8 was less feature rich
[05:46:45] <delewis> Solaris 8 wasn't really that good of a release, though.
[05:46:56] <oninoshiko> more features = more code = more stuff to fix
[05:46:57] <delewis> there were a few features that were introduced in it, but they weren't fulfilled completely until Solaris 9.
[05:47:03] <delewis> like resource controls and RBAC.
[05:47:08] <sleepcat> oninoshiko: dtrace , zfs and jds are the only new things aren't they?
[05:47:25] <delewis> sleepcat: there's a lot of 'smaller stuff' that the end-user isn't really aware of.
[05:47:36] <oninoshiko> comparing to 8, not in the least
[05:47:57] <Bartman007> sleepcat: brandz/containers, x86_64
[05:48:15] <sleepcat> x86 was supported in 8 wasn't it?
[05:48:21] <delewis> sure.
[05:48:28] <delewis> not x86_64, though.
[05:48:35] <oninoshiko> as delewis said, resource controls, RBAC
[05:48:40] <delewis> Solaris 9 (9/04?) supports x86_64.
[05:48:58] <delewis> or whatever the 2005 release of Solaris 9 was.
[05:48:59] <sleepcat> so dtrace, zfs, zones, RBAC (how much code can RBAC really take though)..
[05:49:05] <e^ipi> network stack rewrites
[05:49:10] <delewis> sleepcat: you'd be surprised!
[05:49:19] <delewis> RBAC isn't a simple mechanism.
[05:49:20] <e^ipi> sleepcat: RBAC takes a lot of code
[05:49:29] <delewis> and yes, Fire Engine was a Solaris 10 development.
[05:49:35] <delewis> which was a major, major rewrite of the network stack.
[05:49:38] <oninoshiko> the new services management
[05:49:49] <Bartman007> oh yeah... smf... :-)
[05:49:50] <delewis> (Fire Engine replaced the existing STREAMS implementation -- /dev/ip)
[05:50:14] <sleepcat> how good is fire engine?
[05:50:15] * oninoshiko just loves that... oninoshiko hugs smf
[05:50:24] <jmcp> sleepcat: it's outstanding
[05:50:39] <sleepcat> i never really experienced life without fire engine since i'm a new convert
[05:50:45] <delewis> sleepcat: it scales much better than the older STREAMS-based implementation, which was geared for fewer long-lived connections.
[05:51:02] <delewis> of course the workload of network applications has changed nowadays to something like more short-lived connections.
[05:51:05] <oninoshiko> did 8 support sctp?
[05:51:15] <sleepcat> does anyone really use RBAC?
[05:51:21] <delewis> with the STREAMS implementation it was extremely costly by comparison to FireEngine to build up a connection.
[05:51:24] <e^ipi> I use it all the time
[05:51:33] <e^ipi> I haven't logged in as root in over a month
[05:51:34] <sleepcat> i probably should read how to use it.
[05:51:41] <e^ipi> RBAC is like sudo on steroids
[05:51:42] <sleepcat> i log in as root all the time
[05:51:43] <delewis> sleepcat: it's basically a more featureful sudo.
[05:51:57] <delewis> sleepcat: with RBAC, root literally means nothing on Solaris.
[05:52:08] <delewis> instead of applications simply doing uid == 0, they look at the privileges that role has.
[05:52:14] <jmcp> oninoshiko: not that I recall
[05:52:17] *** Shiv_1 has quit IRC
[05:53:03] <delewis> but as I said earlier, there have been many, many changes from Solaris 8 and 9 to 10 that the end-user isn't going to see.
[05:53:16] <e^ipi> RBAC is a fantastic and incredibly under-rated feature
[05:53:21] <delewis> sure, you hear about zones, DTrace, etc. in the press, but there's a lot of other interesting things that deserve recognition.
[05:53:30] <jmcp> all of which add up to "wtf would you want to install s8 or s9 when you should install s10?"
[05:53:33] <delewis> RBAC, FireEngine, etc.
[05:53:51] <delewis> there's also been some resource control improvements.
[05:54:01] <jbk> heh
[05:54:06] <delewis> and MPO was improved somewhat from its introduction in Solaris 9.
[05:54:06] <jbk> coincidentially
[05:54:06] <e^ipi> I'm not a good writer, otherwise I'd be tempted to write an "idiot's guide to RBAC"
[05:54:17] <sleepcat> e^ipi: you should
[05:54:22] <jbk> there was an email chain at work about some solaris 10 boxes where they needed to up the shared memory max value
[05:54:29] <jbk> and thought they needed to reboot
[05:54:32] *** asarch has joined #opensolaris
[05:54:35] <jmcp> hehe
[05:54:40] <delewis> jbk: oh, dear. :-)
[05:54:45] <sleepcat> jbk: i have to reboot right now.
[05:54:48] <oninoshiko> i would like to experiment with it more... seems to get impressive network improvements (SCTP) (when you can get clients/servers using it).
[05:54:52] <jmcp> removing those sysv ipc limits .... a *godsend*
[05:55:03] <sleepcat> i have a kernel patch that wants a good reboot
[05:55:04] <jbk> i responded back 'this is solaris 10, why do you feel that is necessary when the prctl command will do it online?'
[05:55:10] <delewis> yes
[05:55:13] <dlg_> bah, sctp
[05:55:14] <jbk> and still no one's responded
[05:55:17] <asarch> Have you noted that there is an overheat using SXCE on laptops?
[05:55:18] <delewis> no more convoluted, customized /etc/system files.
[05:55:28] <delewis> it seems like *everyone* had their own /etc/system incantations.
[05:55:29] <jbk> so i'm guessing they're gonna just modify /etc/system and reboot
[05:55:36] <jbk> heh
[05:55:44] <jbk> storage vendors piss me off the most over /etc/system
[05:55:51] <sleepcat> I want to use RBAC to beat CUPS into submission.  How would I go about configuring cups as a non-root user using RBAC?
[05:56:01] <jmcp> delewis: which they copied from solaris 2.6....
[05:56:02] <jbk> EMC insists you need to set sd:sd_max_throttle when connecting to an EMC array using leadville
[05:56:09] <delewis> jbk: Oracle pisses me off about /etc/system, but they've at least started to jump on the dynamic tunables introduced in 10.
[05:56:24] <jbk> delewis: a bit, though crs still doesn't call setproject()
[05:56:36] * delewis avoids crs like the plague
[05:56:36] <jmcp> jbk: I think that might still be correct - but only due to their raid engine implementation
[05:56:44] <jbk> jmcp: with leadville?
[05:56:54] <jmcp> I think so
[05:57:02] <jbk> why would ssd devices be using paramters for sd devices?
[05:57:11] <jmcp> it's the same driver underneath the hood
[05:57:20] <e^ipi> sleepcat: you should only need to monkey with exec_attr and user_attr I think
[05:57:30] <delewis> that's what I thought. I was about to ask what's the difference between sd and ssd. :-)
[05:57:35] <jbk> but two different binaries -- i would think there should be an equivalent -- ssd_max_throttle
[05:57:37] <jmcp> "s" :)
[05:57:40] <jmcp> nope
[05:58:20] <sleepcat> 3000 trapped in mine is south africa
[05:58:33] <sleepcat> http://apnews.myway.com/article/20071004/D8S24LV00.html
[05:58:34] <jbk> though the more annoying one is setting fcp_auto_config(?) on s10 -- i'm pretty sure that is the default (granted it's not on 8 or 9)
[05:58:38] <delewis> ugh, I see pca is still "somewhat" broken.
[05:58:53] <jmcp> jbk: I don't know of that variable
[05:58:57] <delewis> I'm iterating with pca to make sure all the patches install, as some HTTP requests still fail. :-)
[05:58:59] *** sleepcat has quit IRC
[05:59:06] <jmcp> jbk: what's it supposed to work around?
[05:59:57] <jbk> as far as i can tell, it does nothing
[06:00:30] <jmcp> jbk: we've had automatic enumeration in leadville for *aaaaages*
[06:01:10] <jbk> jmcp: at work they still don't use leadville, instead still use emulex's driver (which means editing sd.conf & doing manual wwn to target binding)
[06:01:15] <jmcp> gah
[06:01:20] <jmcp> sigh
[06:01:29] <jmcp> "we must use the same driver across all platforms"
[06:01:34] <jmcp> their loss
[06:01:51] <jbk> naah
[06:01:56] <jbk> i think it's more a matter of ignorance
[06:01:57] <delewis> well, they still have to support multiple Solaris releases.
[06:02:04] <delewis> leadville is only considered mature in Solaris 10, right?
[06:02:12] <jbk> works fine on 8 & 9
[06:02:17] <jbk> just a separate install
[06:02:19] <jmcp> agreed
[06:02:26] <delewis> hmm
[06:02:29] <jmcp> in s10 it's integrated, so easier to manage
[06:02:41] <jbk> yeah, that's the biggest thing
[06:03:05] <delewis> can you update the firmware on an Emulex HBA using Leadville or does that require you to use the whole Emulex driver stack?
[06:03:14] <delewis> Its been awhile since I've done it.
[06:03:31] <jmcp> I believe that new fw is downloaded on boot, if required
[06:03:39] <delewis> IIRC, the Emulex management interface doesn't work with Leadville.
[06:04:04] <jmcp> that's what I recall too
[06:06:58] <Teltariat> I'm going to ask the newbie question that I know is asked most often here: I have a CD-ROM device, iostat -E says its at "sd0".  I can't find sd0 anywhere in /dev.  How do I mount it?  I'm running Reduced Core with what seems to be no vold
[06:07:01] *** smtms has quit IRC
[06:07:15] <flyingparchment> Teltariat: iostat -En tells the device name
[06:07:20] <flyingparchment> Teltariat: to mount it, you most likely want the p0 device
[06:07:22] <Teltariat> Thanks parchment
[06:07:55] <Teltariat> flyingparchment: yea, I know about p0, I just figured out that my raw devices are to be found in /dev/rdsk, and its only there I can fdisk the p0's.
[06:09:11] *** simford has quit IRC
[06:09:49] <jmcp> mount -F hsfs -o ro /dev/dsk/.....s2
[06:10:01] <flyingparchment> does an iso9660 cd-rom have an s2?
[06:10:04] * flyingparchment forgets
[06:10:09] <jmcp> most of them do
[06:10:14] <jmcp> it's a reasonably safe assumption
[06:10:19] <jmcp> if that doesn't work, try an s0
[06:15:31] <Teltariat> p0 worked fine for me
[06:15:46] <flyingparchment> the terms of use page on the new sunsolve is incredibly annoying
[06:15:49] <flyingparchment> it seems to appear on every second click
[06:16:17] <RealWickedWicky> good morning happy people
[06:16:30] <jmcp> Teltariat: did the cd mount?
[06:16:41] <Teltariat> Yes sir, without any issues
[06:16:46] <jmcp> cool
[06:16:53] <Teltariat> Again, thank you Mr. flyingparchment.
[06:17:00] <jmcp> to be pedantic, you should have use the s2 rather than p0
[06:17:04] <flyingparchment> jmcp: do you know if 6446083 will make it into S10?
[06:17:05] <Teltariat> Why?
[06:17:12] <Teltariat> Its a DVD
[06:17:17] <Teltariat> Why s2?
[06:17:30] <jmcp> since the pX numbers are primary partitions, but cd/dvd don't - according to the standard, as I understand it - have that concept
[06:17:30] * RealWickedWicky wonders if jmcp can write a added_extra_hours_of_sleep.diff for me
[06:17:37] <jmcp> RealWickedWicky: sorry :(
[06:18:03] <Teltariat> Does the CD/DVD device driver know to automatically equate p0 with s2, that made the mount work?
[06:18:11] <Teltariat> s/that/to
[06:18:13] <RealWickedWicky> that's okay, I can always try ;-) so coffee it is then, brb
[06:18:20] <jmcp> Teltariat: kinda-sorta
[06:18:37] <jmcp> RealWickedWicky: 6446083 is slated for u5
[06:18:52] <RealWickedWicky> I think that was for flyingparchment
[06:18:58] <jmcp> sorry, yeah
[06:18:59] <flyingparchment> i think so too.. thanks :)
[06:19:00] <Teltariat> Well whatever. :) I'm happy that it mounted, so that I can add a bit of comfort to my bare as the Sahara Reduced Core installation.
[06:19:16] <flyingparchment> i see raidctl does understand mpt raid arrays, so that's the only reason i still need itmpt
[06:19:17] * Teltariat resumes pkgadding away....
[06:19:38] *** Chihan has joined #OpenSolaris
[06:20:14] *** laca has quit IRC
[06:20:39] * Teltariat wonders how people get along on a shell without command history
[06:20:45] <Teltariat> I personally find it maddening without
[06:20:55] <jmcp> Teltariat: carefully, that's how
[06:20:58] <Teltariat> I'd have to claw my eyes out
[06:21:07] <flyingparchment> Teltariat: ksh, csh, /usr/xpg4/bin/sh etc. all support command history.. just not /bin/sh
[06:21:29] <Teltariat> And no tab completion? Divine have mercy on your soul.....
[06:21:45] <Teltariat> xpg4-sh uses the fc utility?
[06:21:50] <flyingparchment> yes
[06:21:55] <flyingparchment> fc is defined in xpg iirc
[06:22:04] <Teltariat> Oh, I see
[06:22:33] <flyingparchment> no, i think i'm wrong there..
[06:22:49] <flyingparchment> but anyway, i think it supports it.. and as it's really ksh, it supports set -o [vi|emacs] too
[06:23:23] <flyingparchment> man, openssl really has a ridiculous number of security issues.  someone needs to do a proper audit :(
[06:24:00] <Teltariat> Hmm.  This machine is now running S10_8/07, but it was the one that couldn't install Nevada 72, and 73.  I wanted to try to use this machine as a Jumpstart point to install the latest Nevada onto another machine, but I've never done Jumpstart before.  I've read up on it, and "clusters" seem to be the key to maintaining sanity
[06:24:07] *** AtomicPunk has quit IRC
[06:24:11] <Teltariat> but I don't see any SUNWC* on the nevada disks
[06:24:21] <Teltariat> do we define SUNWC* cluster files ourselves?
[06:24:26] <Teltariat> Or am I talking out of my ass?
[06:24:37] <flyingparchment> i don't think those are real packages
[06:24:48] * oninoshiko motions that flyingparchment be elected to the position "do i have a second?"
[06:25:02] <Teltariat> Don't clusters server as manifests of the packages you want installed
[06:25:09] <Teltariat> ?
[06:25:18] <Teltariat> s/server/serve
[06:25:29] <oninoshiko> the are described in a manifest file on the cd
[06:25:40] <Teltariat> ah, alright
[06:25:43] <oninoshiko> i would have to look up the name
[06:25:47] <Teltariat> So any pointers on Jumpstarting?
[06:25:52] <Teltariat> <sheepish grin>
[06:26:10] * oninoshiko hasnt dont it, but should be looking into it
[06:26:10] <flyingparchment> Teltariat: there's a how-to in the install guide that covers add_install_server, profiles and such
[06:26:33] <flyingparchment> Teltariat: you might also look at JES, i'm told it's good but never used it myself
[06:26:39] <flyingparchment> no, JET, not JES
[06:26:42] <Teltariat> JES?  Wazzat?
[06:27:06] <oninoshiko> is there a changelist for each svn somewhere?
[06:27:16] <flyingparchment> Teltariat: jumpstart enterprise toolkit
[06:27:25] <Teltariat> I see
[06:28:28] <Teltariat> flyingparchment: the install guide you mentioned is on opensolaris.org, or BigAdmin?
[06:28:39] *** victori_ has quit IRC
[06:28:57] <flyingparchment> docs.sun.com
[06:29:30] <flyingparchment> there's probably something on bigadmin too though
[06:29:53] <oninoshiko> might
[06:29:55] *** victori_ has joined #opensolaris
[06:29:56] <oninoshiko> night*
[06:29:59] *** oninoshiko has quit IRC
[06:31:24] *** Gman has quit IRC
[06:32:06] <Teltariat> thanks parchment
[06:33:08] *** tsoome has quit IRC
[06:33:15] <Teltariat> ISC DHCPd has not been playing nice though.  I try to tell it to ignore the Jumpstart client's DHCP requests so that they can be picked up by the not-yet-setup Jumpstart server, and it ignores me happily, continuing to merrily service any and all DHCP requests.
[06:34:03] <Teltariat> I figure I'll have to take to beating it down with a stick to curb its enthusiasm.
[06:34:32] <jbk> i wonder if the isc dhcpd code is anything like the bind9 code..
[06:35:47] <Teltariat> I remember trying to look through DHCPd code.  I'm not an expert C coder (or even novice for that matter), but I know something opaque when I see it.
[06:36:04] <RealWickedWicky> I'm going to the office, see you all from the bus, good luck and have fun at work/school/college/uni wherever you  are
[06:36:17] <Teltariat> Wait, lol, my last statement didn't make any sense, did it. :P
[06:36:30] <Teltariat> Enjoy your day, Wicky
[06:36:35] <Teltariat> Night, whatever
[06:37:14] *** RealWickedWicky has quit IRC
[06:38:04] *** asarch has quit IRC
[06:38:36] <trochej> Fok
[06:39:23] <trochej> my new uberadmin laughed at my opensolaris t-shirt and now it seems he's a good friend with Polish most laughed at "hacker"
[06:45:23] *** movement has quit IRC
[06:49:25] *** movement has joined #opensolaris
[06:51:36] <flyingparchment> when a disk in a raid5 array breaks, is it normal for it to store bad data and corrupt the array without the controller failing it?
[06:55:02] <RElling> flyingparchment: it certainly is possible, but depends on the array software - reason #423 why ZFS does end-to-end checksumming
[07:03:58] *** Triskelios has quit IRC
[07:04:31] *** RealWickedWicky has joined #opensolaris
[07:18:21] *** karrotx has joined #opensolaris
[07:19:51] *** sleepcat has joined #opensolaris
[07:20:20] <sleepcat> i hate cups
[07:20:48] <sleepcat> i hate usb printers
[07:21:40] <e^ipi> PEBKAC?
[07:21:53] <e^ipi> I managed to get cups working just fine with my usb printer
[07:22:03] <sleepcat> e^ipi: it is PEBKAC
[07:22:09] <e^ipi> and I got it accepting jobs from the network too
[07:22:49] <sleepcat> i don't know what I'm doing wrong
[07:24:20] <sleepcat> ah well, i've been at it all day.  9 hours of trying to print
[07:25:51] <sleepcat> who wants root to my sun box?
[07:26:45] <sleepcat> you can format my hard drive if you want
[07:27:40] <e^ipi> because you haven't figgured out how to monkey with printing yet?
[07:28:11] <sleepcat> e^ipi: i'll install solaris express
[07:28:16] <e^ipi> jesus, here: http://www.sun.com/bigadmin/content/submitted/duplex_printing.html
[07:28:30] <e^ipi> SX might be a better choice though, there's presto there
[07:28:33] <e^ipi> plug & give'r
[07:28:56] <e^ipi> unless you buy a ghetto cheap printer, in which case you have to compile & install the driver first
[07:29:02] <e^ipi> like I had to
[07:29:07] <e^ipi> but after that it worked fine
[07:29:13] *** jimgris has joined #opensolaris
[07:29:20] <e^ipi> hey jimgris
[07:29:28] <jimgris> hey
[07:29:40] <e^ipi> how's it going?
[07:30:18] <jimgris> going good. reading os.o threads and working on slides :)
[07:30:41] <e^ipi> the second is good, the first is a form of self abuse
[07:31:08] <jimgris> Indeed. I'm getting ready for a IEEE conf here in Japan
[07:31:12] <jimgris> should be cool
[07:31:15] <jimgris> my first IEEE
[07:31:16] <e^ipi> you'll never find a more vile cesspool of scum and villany on the internets than the OS.o forums
[07:31:35] <jimgris> well, that's going a tad far, but I get your point. :)
[07:32:04] <e^ipi> what're you doing at the IEEE conf? just attending?
[07:32:36] <sleepcat> what's os.o threads?
[07:33:04] *** baijiutong has quit IRC
[07:33:07] <e^ipi> sleepcat: the forums at opensolaris.org
[07:33:14] <e^ipi> I'd avoid them if you value your sanity
[07:33:21] <jmcp> sleepcat: there's a fairly heated one on ogb-discuss at the moment
[07:33:34] <e^ipi> when is there not a heated one on ogb-discuss?
[07:33:44] <jimgris> e^ipi: I'll just be doing an OpenSolaris talk here: http://www.u-aizu.ac.jp/conference/cit07/
[07:34:23] *** amdx has joined #opensolaris
[07:34:25] <jimgris> That university in Japan is rather unique in that 50% of the professors teach in English
[07:34:42] <e^ipi> neat
[07:35:08] <jimgris> tomorrow there is an Open Source conf in Tokyo that some of the Sun JP team will be presenting OpenSolarsi and Indiana
[07:35:14] <jimgris> I'll take pics
[07:35:19] *** amdx has left #opensolaris
[07:36:38] <e^ipi> how is solaris-as-hobby penetration in japan? as i understand it's quite good in india and pretty poor in america ( I'm the only one I know that's not a windows or linux user )
[07:37:51] <jimgris> There is a *lot* of Solaris here. However, the OpenSolaris community is tiny. It needs to be built from scratch.
[07:38:28] <jimgris> There are some very cool communities in Tokyo (Linux, Ruby, Postgress, etc) that I'm trying to meet.
[07:41:00] *** nostoi has joined #opensolaris
[07:43:07] *** RainDoctor has joined #opensolaris
[07:46:36] <e^ipi> as much as it's annoying that we're largely ignored by the larger open-source ecosystem, it's kinna neat that opensolaris is young & small enough that most of the community knows a little about each other
[07:46:58] *** _Megaf has quit IRC
[07:47:54] <jimgris> why do you say we are ignored?
[07:48:53] <jimgris> But yes, we are small. :)
[07:49:30] <sleepcat> e^ipi: hey i'm in the usa too
[07:49:35] <sleepcat> that's 2 solaris users
[07:49:42] <e^ipi> good chunks of open-source software doesn't compile, when people talk about open source operating systems they make reference to "linux and the BSD's" and so forth
[07:50:34] <e^ipi> though it is kinna cool that some of the linux higherups are noticing & insulting opensolaris
[07:50:59] <jimgris> Indeed. :) That has been interesting lately.
[07:52:45] <moazamraja> re
[07:53:02] <moazamraja> anyone have a Griffin Powermate (the lil knob thingie) working in Solaris?
[07:54:02] *** cmihai has joined #OpenSolaris
[07:54:08] *** Kernel86 has joined #OpenSolaris
[07:54:18] *** sleepcat has quit IRC
[07:56:25] <RealWickedWicky> GOD
[07:56:40] <e^ipi> ZEUS
[07:56:45] <RealWickedWicky> why cant scripts that work with bash2 just work with bash3
[07:56:46] <RealWickedWicky> fucking eh
[07:56:53] <e^ipi> TEAPOT ORBITING THE SUN
[07:57:12] <e^ipi> oh, it's bash... just write scripts against POSIX and to hell with GNU
[07:57:18] <RealWickedWicky> yes yes
[07:57:20] <jmcp> RealWickedWicky: because it's bash, and back-asswards compatibility is for those insane people who insist on Stability rather than 133tness
[07:57:23] <RealWickedWicky> tell that to the RHEL lovahs here
[07:57:33] <RealWickedWicky> ah yes
[07:57:47] <RealWickedWicky> I knew there was a reason I wrote this migration plan from RHEL to Solaris
[07:58:06] <RealWickedWicky> I think this is one of those "I told you so" moments
[07:59:02] <RealWickedWicky> haha
[07:59:13] <RealWickedWicky> even /bin/sh is /bin/bash on RHEL4.3
[07:59:17] <RealWickedWicky> oh my ffing god
[07:59:29] <RealWickedWicky> well, somebody is gonna rewrite scripts today, and its not gonna be me
[07:59:45] *** derchris^ has quit IRC
[07:59:54] *** derchris^ has joined #opensolaris
[08:00:55] *** gavagai_ has quit IRC
[08:00:59] *** WickyZZZ has joined #opensolaris
[08:01:08] <WickyZZZ> ugh
[08:01:13] *** WickyZZZ is now known as WickedWicky
[08:02:21] <WickedWicky> hooray for engineering
[08:03:30] <WickedWicky> "wow patrick, you're really into this solaris thing arent you, do they pay you or something?" <-- stupid retarded questions from coworkers
[08:03:39] <cmihai> Gee
[08:03:47] <WickedWicky> no mofos, things just work, all my scripts I wrote on solaris 8 work on the solaris 10 boxes
[08:04:28] <cmihai> Well, duh, use ksh88 and simple portable shell basic techniques.
[08:04:29] <WickedWicky> I wouldnt mind sun paying me, just to keep things fair
[08:04:40] <WickedWicky> but they dont :P
[08:04:49] <cmihai> And for the love of GOD, DON"T use csh!
[08:05:23] <WickedWicky> it are not that many scripts this guy has to rewrite, that makes it a kinda good thing
[08:05:24] <WickedWicky> just.. 20.
[08:07:47] *** RealWickedWicky has quit IRC
[08:08:46] <trochej> Coffeeeeeeeeee....
[08:08:51] <cmihai> You guys heard of rzip? :-)
[08:09:16] <WickedWicky> I think Tempt mentioned it to me
[08:09:20] <cmihai> 2-3x better compression rations than bzip for say binary logs :-].
[08:09:21] <WickedWicky> and not to use it on pebbles
[08:09:32] *** tsoome has joined #opensolaris
[08:09:33] <WickedWicky> plus that amount in CPU cycles
[08:09:36] <WickedWicky> or?
[08:09:43] <cmihai> Hm.. no, this: http://unixsadm.blogspot.com/2007/10/rzip-great-compression-rates-for-logs.html
[08:09:56] <cmihai> It's actually faster then bzip... though it uses more memory.
[08:11:59] <WickedWicky> I'll read that link in a bit
[08:12:02] <WickedWicky> I arrive at the office now
[08:12:29] <cmihai> Nice if you want to dump your logs to DVD-R or something.
[08:12:32] *** jimgris has left #opensolaris
[08:16:13] *** slowhog has left #opensolaris
[08:19:08] <tsoome> i'd prefer some standard methos for storing logs, so they could be accessible even some years are passing.... .P
[08:19:35] <cmihai> tsoome, actualy, some security standards require you to store them on write once media
[08:19:58] <cmihai> And keep them for 1 or more years...
[08:20:24] <tsoome> yes, write once does not mean never read;)
[08:20:53] <palowoda> sarbane oxley
[08:20:57] <palowoda> sp?
[08:21:00] <tsoome> every moron can write, just try to read...
[08:21:18] <cmihai> Well, if you're talking about reliable media...
[08:21:21] *** RainDoctor has quit IRC
[08:21:23] <tsoome> no
[08:21:35] <cmihai> but?
[08:21:45] *** estibi has quit IRC
[08:21:54] <tsoome> im talking about yes another new compression program
[08:22:04] <tsoome> s/yes/yet/
[08:22:20] <cmihai> Oh, you're worried people won't have rzip to decompress it or what
[08:22:43] <cmihai> If that were the case, we'd all be using compress...
[08:22:44] <tsoome> anyone remembers those gazillion compress/archive programs for dos?
[08:23:10] <palowoda> lawsuits
[08:23:13] <cmihai> Well, consider it security through obcurity.
[08:23:19] <cmihai> obscurity
[08:23:22] <tsoome> yea, and compress still is very widely in use;)
[08:23:27] <cmihai> How often do you need to read your logs? :-
[08:23:40] <cmihai> The ones that have already been rotated and are off the log server for example
[08:23:41] <tsoome> depends on environment
[08:23:52] <tsoome> just now i almost read none
[08:24:02] <palowoda> The mail logs and intrustion detection frequently
[08:24:05] <cmihai> Even so, rzip is quite portable (works on Solaris anyway).
[08:24:09] <tsoome> when i did work in uni, it did happen quite often
[08:24:31] <cmihai> palowoda, ah, but once you offload them from secondary storage to tape or other media?
[08:24:40] <tsoome> portability is not wuestion - will it be alive and available after 5 years?
[08:24:47] <cmihai> Yes.
[08:24:48] <palowoda> Ugh never read anything from tape freq.
[08:24:52] <cmihai> You have the source code.
[08:24:57] <cmihai> It compiles.
[08:25:05] <cmihai> Doesn't meen it needs to be _updated_.
[08:25:09] <kjetilho> cmihai: I'd be interested to see numbers for plain gzip and bzip2 without -9.  in particular gzip doesn't gain much extra, but uses a lot of extra CPU when you use -9.
[08:25:10] <cmihai> Compress hasn't changed for 20 years ffs
[08:25:17] <cmihai> kjetilho, what do you mean?
[08:25:26] <tsoome> will you have manpower to deal with yet another ancient app after 5 years?
[08:25:32] <kjetilho> ok, but not for gzip
[08:25:56] <cmihai> tsoome, In 5 years, sure, Solaris binary compatibility will handle that :-).
[08:26:13] <cmihai> Besides, this is from the Samba blokes mate, it's gonna be around for a while. It's not all that new either.
[08:27:21] <e^ipi> cmihai: assuming murdock et. al. don't throw that baby out with the bathwater
[08:27:57] <palowoda> Oh no the murdock syndrome.
[08:29:47] <palowoda> so is log retention just getting more popular because of euro requirements of some system details or is this just personal preferences?
[08:30:00] <cmihai> Requirements, usually.
[08:30:23] <cmihai> You're supposed to keep them for specified periods of time, on write once media.
[08:30:40] <palowoda> Friggen requirements are so different between US and EU.
[08:30:52] <cmihai> And trust me, when you have say 100GB of logs a day, things change.
[08:30:53] <kjetilho> heh, in Norway it's the opposite, you're not allowed to keep logs with private data longer than necessary.
[08:31:10] <cmihai> kjetilho, even in .gov?
[08:31:12] <e^ipi> I like norway...
[08:31:32] <tsoome> well, it depends what you call to be the private data:D
[08:31:33] <kjetilho> cmihai: sure?  financial transactions is different, of course
[08:31:34] <e^ipi> i've never been, but it sounds like a pretty sensible place, encroachment on rights wise
[08:32:22] *** MattMan has joined #opensolaris
[08:32:33] *** test1 has joined #opensolaris
[08:34:16] <cmihai> Well, in the US there's a bunch of bull about hearsay evidence and such...
[08:34:21] <palowoda> So if log retention requires RO is that optical and tape excluded?
[08:35:05] <cmihai> For example logs, when they aren't use to prove statements but as they are, aren't considered such... hence they're a valuable resource in legal procedings.
[08:35:48] *** Dar has joined #opensolaris
[08:35:49] <cmihai> palowoda, not exactly.
[08:35:51] <cmihai> HP has UDO
[08:36:05] <cmihai> They're magneto-optical disks
[08:36:15] <cmihai> 30GB stuff
[08:36:23] <palowoda> Isn't that old technology?
[08:36:26] <cmihai> Though seeing how a Blue Ray is 27.3 times cheaper...
[08:36:32] <cmihai> palowoda, yeah, pretty much.
[08:36:43] <cmihai> They've tried magneto-optical floppies years ago...
[08:36:50] <palowoda> Yeah I kind of remember that when working at Plexus.
[08:36:50] *** RealWickedWicky has joined #opensolaris
[08:36:51] <cmihai> you could write 32MB on regular floppies :P
[08:36:55] <kjetilho> BluRay isn't Write-Once, is it?
[08:37:12] <tsoome> there are many alternatives, most storage vendors willbe happy to sell you some kind of write once technology
[08:37:37] <cmihai> kjetilho, BD-R
[08:37:48] *** WickedWicky has quit IRC
[08:38:08] <e^ipi> log > /dev/null
[08:38:15] <cmihai> BD-R is write once, BD-RE is rewritable
[08:39:42] <palowoda> e^ipi Homeland security did that stupid link from /dev/null to a file.
[08:40:25] *** RealWickedWicky is now known as WickedWicky
[08:40:30] <palowoda> I hate that when it happens.
[08:49:37] *** Mdx4 has joined #opensolaris
[08:56:40] *** mikefut has joined #opensolaris
[08:56:58] *** chrisso has quit IRC
[09:03:19] *** bengtf_ has quit IRC
[09:06:28] *** Mdx4 has quit IRC
[09:08:36] *** Giaco has joined #opensolaris
[09:08:37] *** SYS64738 has quit IRC
[09:14:58] <test1> hi, i am new to opensolaris, from where i can download its complete CDs (or DVDs) ?
[09:15:33] <dlg_> get solaris express
[09:15:38] *** dlg_ is now known as dlg
[09:15:40] <palowoda> http://www.opensolaris.org/os/downloads/
[09:17:33] *** cydork has joined #opensolaris
[09:18:44] <e^ipi> opensolaris is a pile of code, not an OS
[09:19:21] <e^ipi> solaris express is the most widely used distribution at the moment
[09:20:09] <palowoda> It would be interesting if you could get opensolaris code via CD's or DVD's.
[09:20:24] *** Cyrille has joined #opensolaris
[09:20:44] <e^ipi> the ON source is on the starter kit
[09:20:52] <e^ipi> given, an old version of ON, but it's there
[09:20:59] <palowoda> Which version?
[09:21:11] <e^ipi> whichever version is on the starter kit
[09:21:19] <e^ipi> on mine it's build 52 or something
[09:21:27] <palowoda> Ah waste of time I guess.
[09:21:47] <e^ipi> why would it be a waste of time?
[09:21:50] <palowoda> Unless you have a modem.
[09:22:17] <palowoda> Your in Canada, your bandwidth is cheap.
[09:22:34] <e^ipi> helluva cheap
[09:22:46] <palowoda> Must be the taxes.
[09:23:32] <e^ipi> unless I start cranking it up to 25Mbps downstream
[09:23:40] <e^ipi> in which case it goes up to about $90/month
[09:23:42] <palowoda> Or buying CD's :)
[09:24:00] *** renihs has joined #opensolaris
[09:24:24] <palowoda> Damn I pay 90 for DSL curcuit. ASDL too boot.
[09:25:06] <e^ipi> i pay $50 for 10Mbps down/512k up
[09:25:35] <palowoda> 90.00 a month for 6M/768
[09:25:54] <e^ipi> oh canada...
[09:26:03] <palowoda> oh taxes
[09:26:12] <e^ipi> has nothing to do with taxes
[09:26:25] <e^ipi> government doesn't own the backbone, the phone companies do
[09:26:33] <palowoda> Yeah you get it free because in canada your better looking right?
[09:27:04] <kjetilho> e^ipi: are you saying the Canadian government has no backbone?
[09:27:05] <palowoda> Winter is coming drink more coffee
[09:27:14] <e^ipi> kjetilho: this much should be apparant
[09:29:34] <palowoda> e^ipi:  I'm shipping you 10,000 free blank dvd's.
[09:31:34] <e^ipi> well, that's a random comment
[09:32:02] <palowoda> Hey we get them for 12 cents a piece here.  But yes a random comment.
[09:33:20] <e^ipi> umm... okay
[09:33:24] <sniffy> $50 for 100/10 (ye mbps)
[09:33:41] *** Gekz[sleep] is now known as Gekz
[09:33:52] <palowoda> sniffy: location?
[09:34:08] <sniffy> sweden
[09:34:14] <sniffy> naturally
[09:34:19] <palowoda> figures.
[09:34:39] <sniffy> free umts at full blow is like $20 a month
[09:34:58] <sniffy> so pleanty of options for getting online quick
[09:36:19] <palowoda> sniffy: How much do you pay for say a 500Gig disk drive in sweden?  SATA
[09:37:17] <sniffy> uhm dunno around 800 sek for a wd usb one - so less i think - which is ~$100 or so
[09:37:52] <e^ipi> there's a point attempting to be proved somewhere in here, i'm sure
[09:38:05] <sniffy> but hardware is a one time fee per thing - subscriptions ar monthly or what not
[09:38:12] <trygvis> $178 for a Samsung SpinPoint T166 500GB SATA2 16MB 7200RPM
[09:38:16] <trygvis> (in norway=
[09:38:16] <sniffy> that includes sales tax
[09:38:29] <palowoda> Just wondering about the differences in pricing of hard media verses bandwidth fee's.
[09:38:46] <e^ipi> enough bandwith, and who gives a damn about hard media fees
[09:38:47] <e^ipi> ;)
[09:39:16] <palowoda> Here in the US it's about 90 to 100 USD for 500Gig drives.  plus 5 percent tax where needed.
[09:39:59] <palowoda> For some reason we a flooded with cheap optical media.
[09:40:51] <sniffy> well sweden seems to match that then
[09:40:56] <e^ipi> DVD media is cheap here
[09:40:59] <palowoda> e^ipi: I thought you had to pay extra taxes on blank optical media in Canada?
[09:41:17] <e^ipi> $30 for a 100 spindle, so about .30 a piece
[09:41:35] <e^ipi> and that's normalized for a much lower canadian dollar, so the price'll adjust in a couple months
[09:42:11] <palowoda> So no extra tax surcharge on dvd or cd's in Canada?
[09:42:26] <e^ipi> we pay a bit of an extra tax on blank media, but i think it's only on blank CD's, not dvd's
[09:42:38] <e^ipi> and then it's only a couple cents per.
[09:42:49] <palowoda> Well there is a brain dead tax policy.
[09:43:11] <e^ipi> I don't mind my taxes if they're spent well
[09:44:36] <e^ipi> government waste is a seperate issue than high taxes. the first implies the second, but not neccessarily the other way around
[09:45:16] <lloy0076> e^ipi: I love paying the music industry for the music I DON'T PIRATE.
[09:45:23] <lloy0076> I may as well pirate the stuff.
[09:45:26] *** Erwann has joined #opensolaris
[09:45:31] <kjetilho> in Norway we had (have) a tax on blank cassettes, to compensate for music copying.  this makes sense, since it is *legal* to make copies and give them away to friends.
[09:45:45] <e^ipi> I did the calculation based on the tax rate in canada, the population, and the national healthcare budget, and the average canadian pays about $100 per year for health insurance
[09:45:50] <kjetilho> for some reason, the music industry didn't think it was a good idea to legitimize copying of CDs
[09:46:01] <palowoda> At .30 cents a CD I wonder how much the authors of music really get.
[09:46:09] <quasi> kjetilho: we have the tax, but it still isn't legal to copy
[09:46:09] <lloy0076> palowoda: A pittance.
[09:46:10] <palowoda> or a percentage of that.
[09:46:16] <WickedWicky> $100 per year?!
[09:46:17] <kjetilho> quasi: that's just absurd
[09:46:24] <WickedWicky> I pay 99 Euro a month
[09:46:31] <e^ipi> palowoda: do you actually think the artist sees that much from CD sales in the first place?
[09:46:48] <quasi> kjetilho: yeah, just another way of making money
[09:47:12] <e^ipi> the recording industry is a bunch of snakes
[09:47:27] <palowoda> e^ipi: With our layers in the US you would be suprised.
[09:47:42] <palowoda> lawyers.
[09:47:44] <e^ipi> but honestly, i'll take the extra $.10 per blank CD instead of the DMCA any day of the week
[09:49:45] <palowoda> It was 21 cents in 2003 did they lower the musician's pay?
[09:50:17] *** phimic has joined #opensolaris
[09:50:58] <e^ipi> maybe
[09:51:07] <e^ipi> maybe it's just blank cdr's
[09:51:18] <e^ipi> those are still on the pricy side
[09:51:44] <palowoda> Maybe they should take it down to say 2 cents a cd.  Sound reasonable?
[09:51:52] *** sioraiocht has joined #opensolaris
[09:52:11] <palowoda> reasonable
[09:52:41] <e^ipi> I honestly don't use blank media that much, so i'm a poor judge of what's reasonable or not
[09:53:29] <palowoda> But the CA gov is a good judge of what the artists should be payed right?
[09:54:27] <paul> People in the USA think they pay less than tax than those in the 'socialist' Europe..
[09:54:38] <paul> that's not true though.
[09:54:44] <e^ipi> paul: all my taxes are consolidated
[09:54:51] <e^ipi> i'm not nickle and dimed at every corner
[09:55:05] <paul> AFAICT, tax in states like Ireland and UK is /lower/ than in USA (compared to Georgia anyway)
[09:55:38] <paul> plus over here you /don't/ need to pay for emergency medical care.
[09:56:02] <e^ipi> I pay 15% federal tax and 5% provincial tax
[09:56:03] <paul> so if you're youngish and healthy you can skip medical insurance.
[09:56:08] <e^ipi> most of which I write off
[09:56:25] <e^ipi> and that's it
[09:56:40] <paul> e^ipi: that's income tax or ..?
[09:56:44] <e^ipi> no road taxes, or municipal water taxes, or anything
[09:56:49] <e^ipi> paul: yeah, that's income tax
[09:57:27] <palowoda> e^ipi: Your a student.
[09:57:42] <e^ipi> palowoda: yes, but that just covers the writeoff
[09:58:01] <palowoda> Wait till your earning 400K.
[09:58:03] <e^ipi> if i quit school i'd still only be paying 20% income taxes
[09:58:21] <renihs> <paul> so if you're youngish and healthy you can skip medical insurance < omg? :p
[09:58:22] 
[09:58:49] <e^ipi> if i earned 400k p/a i'd pay 29% federal and 14% provincial
[09:59:01] <paul> for that I get excellent emergency medical care for free.
[09:59:07] <e^ipi> and that's only on the amount over 100k or so
[09:59:08] *** _WildPikachu_ has joined #opensolaris
[09:59:16] <palowoda> In the US you write that down to 10percent.
[09:59:19] <paul> and free emergency medical care all over EU
[09:59:28] <_WildPikachu_> I'm trying to understand what the requirements are for me to burn OpenSolaris discs for my clients
[09:59:43] <e^ipi> plus, if you're making over 100k/year you get a ton of places to write off most of that anyways
[09:59:44] <paul> _WildPikachu_: which discs?
[09:59:59] <renihs> paul, not just emergency
[10:00:00] *** nostoi has quit IRC
[10:00:08] *** dmarker has quit IRC
[10:00:27] *** dmarker has joined #opensolaris
[10:00:35] <paul> renihs: the situation is more complex for beyond emergency..
[10:00:49] <renihs> paul, it is but you are not being dumped on street to die
[10:00:56] <paul> (it depends on the state, it may be free but not very adequate in terms of waiting time, etc)
[10:01:22] <paul> renihs: right, all time-critical medical care is essentially free in the EU
[10:01:25] <renihs> thats not being argued, if you can put a few 100k euros on the table you will get the new heart quickly
[10:01:36] <renihs> else you will wait long, but thats just corruption
[10:02:02] <paul> e^ipi: what would taxes be on a more typical income? (ie an average industrial income)
[10:02:09] <_WildPikachu_> paul, this one   => "Solaris Express Community Edition"
[10:02:28] <palowoda> _WildPikachu_: CD or DVD?
[10:02:33] <_WildPikachu_> dvd
[10:02:48] <palowoda> You going to burn it in Windows?
[10:03:05] <_WildPikachu_> nope
[10:03:14] <palowoda> What OS are you going to burn in?
[10:03:19] <e^ipi> 22% on money between 40k and 75k ( the first 40k are taxed at the lower rate) federal, and 8% provincial
[10:03:20] <_WildPikachu_> it doesn't matter how i burn it ... i'm curious as to the distribution requirements
[10:03:29] <paul> _WildPikachu_: which requirements? do you mean whether you legally may distribute burnt discs to clients?
[10:03:39] <_WildPikachu_> paul, yep
[10:03:44] <_WildPikachu_> sorry if i was unclear
[10:03:56] <palowoda> You can't redistribute.
[10:04:22] <palowoda> At least Solaris Express.  Maybe Nexenta or Belenix versions.
[10:04:51] <paul> _WildPikachu_: afaik, technically you may not redistribute copies of Solaris - even though it's a free download.
[10:05:16] <_WildPikachu_> ok
[10:05:23] <e^ipi> and again, there's piles of writeoffs all over the place
[10:05:43] <e^ipi> children, investments, job related expenses ( tools, uniforms, etc )
[10:05:52] <paul> _WildPikachu_: if you acted wholly as an agent of your client, you might not redistribute such images. Whether you can act as agent 'in bulk' is something you'd need to ask a lawyer.
[10:06:05] <trochej> Any word if sxce 74 will be out this friday? Maybe I download 73 unnecesomething
[10:06:10] <palowoda> e^ipi: I guess you can worry about that after you graduate.
[10:06:12] <trochej> Drat, cant espell that word
[10:06:17] <_WildPikachu_> thanks paul
[10:06:20] *** peteh has joined #opensolaris
[10:06:24] <paul> e^ipi: so your tax isn't much better than over here ;)
[10:07:07] <palowoda> Heck in the US we play games and buy from states that don't have tax.
[10:07:49] <paul> i guess it depends on whether you believe your tax should be spent on subsidising medical care and education, or whether it should go to the military-industrial complex.. ;)
[10:08:19] <palowoda> We also seem to blow up everyting in site.
[10:09:25] <palowoda> And ask questions later.  Not that I agree with that.
[10:09:27] <paul> (College tends to be free, or else very cheap over here)
[10:09:57] <paul> (if you live in a country where college is not completely free, just go study in one the countries where it is)
[10:10:48] *** lloy0076 has quit IRC
[10:10:49] <palowoda> Heh don't come here it's three times the cost for any edu instituion.
[10:11:32] <test1> is ther any mirror for http://www.opensolaris.org/os/downloads/ for downloding opensolaris ?
[10:11:39] <palowoda> no
[10:13:41] <palowoda> test1: Just out of curiosity where are you located and why are you concerened with bandwidth in downloading?
[10:16:12] <palowoda> I've always wonder if project Indiana was just invented to resolve such issues.
[10:16:51] *** agony__ has joined #opensolaris
[10:17:08] <test1> palowoda, my bandwidth is low, and i want to give its address to one of my friends that in its organization has some restrictions in connection to some sites, thus i want to give him a list  of mirrors so that he can download if from at least one of them.
[10:17:46] *** dpn` has joined #opensolaris
[10:18:07] <palowoda> test1: Do you mean your friends are using modems or have less than 200K download?
[10:18:39] <trochej> HMmm, I got here 10 Mbit internet pipe
[10:18:42] <trochej> naaaaajjjsss
[10:19:18] <test1> palowoda, no his speed is good(512 Kbps)
[10:19:41] <test1> is there any mirror for http://www.opensolaris.org/os/downloads/ ?
[10:20:07] *** dnilsson has joined #opensolaris
[10:20:55] *** dnilsson has quit IRC
[10:21:48] <palowoda> test1: I said no there is no mirrors.  That is not how Sun downloads work.   The Sun network has internal routing for differenet geographical areas but for some reason you want external sources as a mirrior for some reason.
[10:22:11] *** razrX has quit IRC
[10:22:17] *** andyshack has joined #opensolaris
[10:22:26] <paul> test1: some of the opensolaris distributions have mirrors, Nexenta definitely does.
[10:22:35] <paul> test1: Solaris Express does not
[10:22:42] <palowoda> Nexenta is not Sun.
[10:23:15] *** razrX has joined #opensolaris
[10:23:16] <palowoda> Two different companies
[10:23:17] <trochej> palowoda: SXCE contains things that cannot be freely redistributable
[10:23:34] <trochej> Nexenta seems to contain ON and typical GNU shajt
[10:23:52] <palowoda> trochej: Like I didn't know that.
[10:24:13] <trochej> palowoda: I know that you know, I was saying to paul :)
[10:24:19] <palowoda> Ah
[10:24:58] <trochej> but it seems I should have said to to test1
[10:24:58] <trochej> :)
[10:25:01] <palowoda> Paul is learning all the different versions of Opensolaris binary distributions.
[10:25:34] <paul> hmm, I don't think I said Nexenta was Sun..
[10:25:41] *** agony_ has quit IRC
[10:25:42] <trochej> palowoda: You didn't :)
[10:25:45] <trochej> Drat
[10:25:50] <trochej> paul: You didn't :)
[10:26:33] <paul> anyway
[10:26:34] <paul> breakfast!
[10:26:55] <palowoda> Wait I did fire up my smoker for some good smoked salmon the past few days.
[10:27:22] <flyingparchment> how friendly is veritas with zones?
[10:27:56] <trochej> flyingparchment: Very. They get horizontal every chance they get. :)
[10:28:08] <test1> at last, from where can i get opensolaris :-((
[10:28:45] <trochej> test1: Sources?
[10:28:46] <palowoda> test1: I gave you the url.  You didn't check it out?
[10:28:46] <e^ipi> opensolaris is just code
[10:28:47] *** alanc_away has quit IRC
[10:29:23] <palowoda> Ok votes for test1 for troll or bozo?
[10:29:39] <trochej> I don't
[10:29:46] <trochej> One can be a little bit confused at start
[10:29:55] <trochej> test1: OS so far is a source code.
[10:30:03] <e^ipi> so far?
[10:30:10] <trochej> You can get working os as: Solaris Express Community Edition
[10:30:12] <palowoda> Come on test1 did you look at the opensolaris.org website?
[10:30:19] <trochej> That one doesn't have mirrors, sot o speak
[10:30:52] <trochej> test1: Nexenta, Belenix, SChilinx and such, linked from opensolaris.org, are freely redistributable and probably have mirrors
[10:30:59] <trochej> for that matter, belenix has a tortrent file
[10:31:04] <trochej> torrent
[10:31:36] <palowoda> Say what is the perfered torrent client on Solaris?
[10:31:48] <cmihai> Azureus since it's Jaba
[10:31:53] <cmihai> ctorrent, rtorrent might also work
[10:31:59] <trochej> palowoda: I don't know, I don't use torrent
[10:32:20] <palowoda> Azureus is the vote.
[10:32:25] <e^ipi> i use azureus because it reports ratios properly & has a console UI
[10:32:52] <Teltariat> um
[10:32:55] <Teltariat> quick question
[10:32:56] <flyingparchment> i hear rtorrent is popular (a command-line client)
[10:33:01] <flyingparchment> never used it though
[10:33:11] <Teltariat> Does Solaris 10 x86 automatically monitor HDD SMART data?
[10:33:42] <palowoda> Wait how can you reccomend something you never used.
[10:33:45] <Teltariat> Azureus + HTMLui plugin is what you need.  You can run Azureus headless, and then manage it and run torrents through a browser
[10:34:04] <Teltariat> ./azureus --ui=console
[10:34:15] <e^ipi> and why the hell would I want to do that when ssh plus gnu screen work perfectly well?
[10:34:18] <Teltariat> Then you go to <ip_of_az_box>:6886/
[10:34:21] <Teltariat> in your browser
[10:34:24] <andyshack> Teltariat is right.
[10:34:37] <e^ipi> monkey browser UI annoys me more than stupid GUI's
[10:35:23] <palowoda> Seems like azureus should be one of those apps included in Solaris Express.
[10:35:30] <e^ipi> yeah, why not
[10:35:33] <Teltariat> e^ipi: I used screen + Python Bittornado, and tired of it after a while because with the web ui, I could easily manage the entire host of torrents I had running in a simple tab next to all the other tabs in my browser.
[10:35:34] <e^ipi> it already takes up a DVD
[10:35:35] <palowoda> license?
[10:36:04] <Teltariat> palowodo: open source or something or other.  go see: http://azureus.sf.net
[10:36:20] <Teltariat> Download the classic azureus, or else you'll get the abomination that is Azureus 3
[10:36:38] <Teltariat> which is called "Vuze".  (Lord have mercy.)
[10:39:13] *** Mdx4 has joined #opensolaris
[10:39:16] <palowoda> azureus seems like gpl just like java so no big deal
[10:40:33] <palowoda> But than again Sun isn't into bittorrent so through that out the window.
[10:44:59] <trochej> HM
[10:47:15] <WickedWicky> when /opt is a diff filesystem from / , and I wanna use live upgrade, I'll have to make an /altroot and an /altopt if I understand correctly, am I understanding it correctly?
[10:48:31] <palowoda> You might be confusing filesystems and slices.
[10:48:36] <WickedWicky> slices
[10:48:41] <WickedWicky> potato potatoe
[10:48:56] <palowoda> did you ever use prtconf?
[10:49:03] <WickedWicky> yesh
[10:49:22] <palowoda> you alt is on which slice?
[10:49:32] <palowoda> your opt is on which slice?
[10:49:39] <WickedWicky> I am like, setting up the system as we speak out
[10:49:53] <WickedWicky> I am laying out filesystems, and yes, the installer calls it laying out filesystems
[10:50:17] <palowoda> so how much and which slice did you configure for lu?
[10:50:36] <WickedWicky> ok, lemme make the question short and simple, do I need an /altopt when I have an /opt
[10:50:51] <PerterB> yes
[10:50:56] <WickedWicky> 15000MB for /altroot, since / is 15000MB
[10:51:05] <WickedWicky> /opt is 30000MB
[10:51:13] <WickedWicky> haleluja, thanks PerterB
[10:51:58] <palowoda> One would think you could link and mount a common /opt or make a slice for it.  Which ever is useful.
[10:52:18] <WickedWicky> /opt is slice 4
[10:52:22] <WickedWicky> root is slice 0
[10:52:25] <WickedWicky> if that's what you mean
[10:52:35] <flyingparchment> WickedWicky: LU won't try to upgrade /opt because there's no OS there
[10:52:48] <PerterB> You could think that, but you would be wrong... If you have a common /opt it'll all go horribly wrong when LU upgrades packages with components in /opt
[10:52:57] <WickedWicky> amm
[10:53:04] <flyingparchment> PerterB: hm.. it does that?
[10:53:07] <WickedWicky> doesnt SXCE place packages in /opt by default?
[10:53:14] <palowoda> the compilers
[10:53:26] <quasi> PerterB: if you actually have anything covered by the base os in /opt
[10:53:26] <flyingparchment> i don't think LU will try to upgrade studio
[10:53:31] <palowoda> although they can be installed seprate.
[10:53:48] <quasi> flyingparchment: LU upgrades what you ask it to upgrade
[10:53:52] <PerterB> flyingparchment: studio no, but things like star office are in the base OS
[10:53:55] <palowoda> Just don't use the silly DX install.
[10:54:02] <flyingparchment> PerterB: but that's in /usr/staroffice isn't it?
[10:54:12] <WickedWicky> SUNWmlib, staroffice8, SUNWjavadb
[10:54:15] <WickedWicky> they're not upgraded?
[10:54:28] <flyingparchment> i only have S10 here, maybe it moved in SX
[10:54:29] <palowoda> Not in /opt
[10:55:01] <PerterB> flyingparchment: no, it's in /opt with the other components WickedWicky mentions
[10:55:48] *** NikolaVeber has joined #opensolaris
[10:56:08] <WickedWicky> PerterB and these components in /opt are not upgraded then when using LU?
[10:56:45] *** Hystrix has joined #opensolaris
[10:56:51] <PerterB> WickedWicky: they *are* upgraded by LU, which is why your new boot environment needs its own /opt, which is why you need a /altopt
[10:57:04] <WickedWicky> ok, thank you
[10:57:58] <palowoda> Heh than by virtual /opt isn't opt via LU.
[10:58:16] <PerterB> and I am late for work again... shower time :)
[10:58:28] <palowoda> virtue
[10:58:55] <palowoda> But than again LU isn't open source solaris either.
[10:59:19] <WickedWicky> but LU is offered as upgrade solution by opensolaris
[10:59:30] <palowoda> In source form?
[10:59:32] <WickedWicky> no
[10:59:36] <WickedWicky> as functionality
[10:59:38] *** Mdx4 has quit IRC
[10:59:47] <WickedWicky> which is why I ask about its usage and i got what I read correctly
[10:59:59] <WickedWicky> you dont have to answer questions you dont want to answer of course
[11:00:07] <WickedWicky> but just dont then.
[11:00:18] <palowoda> Opensolaris questions?
[11:00:20] * WickedWicky goes for a coffee
[11:00:31] <WickedWicky> dude
[11:00:42] <WickedWicky> I am gonna try to LU upgrade SXCE72 to SXCE73
[11:00:47] <WickedWicky> how opensolaris do you want it to get?
[11:01:03] <WickedWicky> but I am not gonna argue with you, coffee is way more valueable
[11:01:10] <palowoda> Just goof'en.
[11:02:02] <palowoda> Hell I'm suprised you have a 30G /opt.
[11:02:31] <trygvis> blastwave! :D
[11:02:32] *** test1 has left #opensolaris
[11:02:59] <palowoda> I simlink blastwave to a zfs partion from opt
[11:03:33] <palowoda> wouldn't affect lu anyways.
[11:04:31] <razrX> palowoda: /opt/csw is zfs for me, not a symlink tho
[11:04:32] * quasi stores blastwave in /dev/null
[11:04:37] <razrX> quasi: lol
[11:04:40] <trochej> :)
[11:05:05] <quasi> plenty of room for the insane dependency tree
[11:05:38] <quasi> svn was something more than 30 packages to get a commandline tool
[11:05:41] <palowoda> Say is the /opt/SUNWappserver or SUNWjavadb upgraded via live upgrade or nevada builds in general?
[11:06:23] <trochej> VFAT is stupid
[11:06:24] <trochej> It can't accept files with : in name
[11:06:51] <palowoda> Wait those packages are not required but /opt/SUNWmlib might be important.
[11:09:23] *** timsf has joined #opensolaris
[11:09:32] <timsf> hi all
[11:09:48] <trochej> hi
[11:11:57] <palowoda> Does Sunstudio load redundent libmlib files in /opt?
[11:13:11] <kjetilho> remember Sun Studio can't assume Solaris 10
[11:15:14] <paul> software today has lots of dependencies shocker..
[11:15:27] <paul> In-depth special report at 11
[11:15:38] <palowoda> so /opt/SUNWmlib is kind of useless with opensolais delivered in solaris express.
[11:16:50] *** RainDoctor has joined #opensolaris
[11:18:32] <palowoda> PeterB: So which packages where required /altopt when using LU?
[11:19:54] *** migi has joined #opensolaris
[11:20:53] *** Gman has joined #opensolaris
[11:21:10] <PerterB> varies from solaris release to solaris release, there's less in /opt in nevada than there was in solaris 8, but at the very least SUNWstaroffice*
[11:21:46] <PerterB> but more to the point, once you reboot onto your new boot environment, if you want /opt to still be a seperate filesystem then you have to have allocated some space for it
[11:22:33] <palowoda> I was under the impression that the lu upgrades where not going to be tied to nevada build upgrades. That is all.
[11:22:36] <PerterB> (you can, of course, "flatten" filesystems into the same slice during LU, but you can't undo that until the next LU)
[11:23:03] <PerterB> what do you mean by "tied to nevada build upgrades"?
[11:23:28] <palowoda> I mean optional software should be tied to upgrades.
[11:23:33] *** derchris^ has quit IRC
[11:23:50] <quasi> PerterB: couldn't you "undo" the flattening with lumake?
[11:24:01] <PerterB> quasi: uh, maybe :)
[11:24:51] <quasi> PerterB: once you're happy with your new BE, you just lumake to populate the old BE from the new env
[11:24:53] <PerterB> palowoda: once you have created a new boot environment, you can do more than just upgrade the OS with luupgrade, you *can* upgrade optional software too, but currently you'd have to do that by hand
[11:26:09] <palowoda> It all boils down to LU being part of OpenSolaris or not.  Which it's not right now.
[11:26:34] <PerterB> quasi: coo, I never even realised that existed... all my systems tend to be mirrored to I rarely need more than lucreate/ludestroy
[11:26:41] *** Griffous has joined #opensolaris
[11:27:08] <PerterB> palowoda: you mean because you can't enhance it to upgrade other components easily?
[11:27:21] <quasi> PerterB: it is very nice if you don't want to break your mirrors or if you want to change your disk layout
[11:27:32] <PerterB> yeah, I can see that :)
[11:28:27] * quasi is a bit paranoid about breaking mirrors ;)
[11:29:02] *** catena has joined #opensolaris
[11:29:46] <PerterB> 7 years bad luck ;)
[11:29:52] <quasi> heh
[11:29:55] <palowoda> PeterB: Was that the intent of OpenSolaris and LU.  I think not.  LU has some incumbered technology tied to it.  Why advertise it's features as a Opensolaris feature?
[11:30:19] <palowoda> s/a/an/
[11:30:20] <PerterB> uh, because it's a useful feature and it's available to open solaris users?
[11:30:21] *** JWheeler has quit IRC
[11:30:25] <catena> i don't know what happened to my system: it takes a lot time to resolve hosts
[11:30:29] <quasi> palowoda: because upgrading in osol woul be a pain without lu?
[11:30:52] <palowoda> PeterB: so are the studio comilers useful but they aren't open source.
[11:31:07] <catena> can anyone help?
[11:31:26] <PerterB> absolutely they're useful... lots of software that isn't open source is useful
[11:31:38] *** bnitz has joined #opensolaris
[11:32:17] <PerterB> catena: check whether the name servers listed in /etc/resolv.conf are still working, maybe the first one you have listed has become unavailable
[11:33:03] <palowoda> useful and dependent by design might have different meanings.
[11:34:36] <catena> traceroute to the primary server prints as : http://pastebin.com/d7d5b5691
[11:35:52] <palowoda> wow you have to go through a lot of hops to your primary dns server.
[11:36:03] <PerterB> that doesn't prove much...use "nslookup - 202.9.145.6" to test it
[11:36:21] <PerterB> anyway, gotta run (sorry, palowoda, we can pick this up some other time)
[11:36:38] <palowoda> sound good
[11:37:04] <catena> traceroute took approx. 15s
[11:37:53] <palowoda> catena: I see 37ms in your traceroute.
[11:38:09] <catena> it's making me confused!
[11:38:32] <palowoda> me too, but I'm a pro at being confused. :)
[11:38:53] <catena> :)
[11:40:31] <catena> online speed tests show: 591kbps download && 364kbps upload,,, nslookup tells: http://pastebin.com/d2cb10e1e
[11:42:02] <palowoda> So many primary dns server ips so little time.
[11:42:31] *** phips|mb has joined #opensolaris
[11:42:36] <catena> didn't get that
[11:43:13] <palowoda> Hey I didn't get the different pastebins either.  But like I said I get confused easy.
[11:46:58] <catena> i got it, the primary i specified is down: 202.9.145.6, the rest are working
[11:48:09] <palowoda> Don't you hate it when you don't control your own dns server. :)
[11:50:55] <catena> it's the second time i'm experiencing this rare problem, first at windows, now at solaris
[11:51:30] <palowoda> Who is your DNS daddy?
[11:51:38] <catena> VSNL
[11:51:55] <palowoda> Tell them they are all screwed up.
[11:53:26] *** Griffous is now known as JWheeler
[11:53:57] <catena> Of coz, they're all screwed up, when i ring them up, all i hear is a lady speaking nuts && bolts
[11:54:35] *** Gekz is now known as Gekz[zzz]
[11:58:16] <palowoda> I don't know if a female was talking about nuts and bolts to me on the other end of the line it might get my attention.
[12:01:29] <palowoda> Than again I'd probably forget what the problem was.
[12:02:03] <catena> so you forget everything in 15mins? :)
[12:02:27] <palowoda> I thought it only toke 3 mins.
[12:05:12] <catena> something like cyclic ___ syndrome?
[12:06:33] <palowoda> Could be.  But I might be either too old for that or just bored. :)
[12:07:14] *** Gekz[zzz] is now known as Gekz
[12:16:40] *** andyshack has quit IRC
[12:19:24] *** Fish has joined #opensolaris
[12:19:45] <Fish> hello
[12:28:53] *** RainDoctor has quit IRC
[12:35:31] *** ghatak has left #opensolaris
[12:44:49] *** timsf has quit IRC
[12:46:28] *** Hystrix has quit IRC
[12:50:51] *** Gman has quit IRC
[12:52:45] *** timsf has joined #opensolaris
[12:53:44] <trygvis> I wonder how noisy a box like this would be: http://www.supermicro.com/Aplus/system/Tower/4021/AS-4021M-32R.cfm
[12:54:16] *** cmihai has quit IRC
[12:57:33] *** Chihan has quit IRC
[12:58:46] *** Gekz is now known as Gekz[zzz]
[12:59:12] *** catena has left #opensolaris
[12:59:58] <trochej> trygvis: A little
[13:00:28] <trochej> :)
[13:01:17] * JWheeler attempts to wave down delewis
[13:01:44] * JWheeler flails limbs
[13:02:15] *** kszwed has joined #opensolaris
[13:04:49] * trochej looks up
[13:04:50] *** debuggerboy has joined #opensolaris
[13:04:53] <trochej> Can't see him
[13:05:26] <jamesd> trygvis, all depends on the quality of fans and harddrives...   if everything is new and quiet the box is quiet as the drives and fans age... bearings begin to go and it will get noiserier
[13:07:01] *** Murmuria has joined #opensolaris
[13:16:21] *** uebayasi has quit IRC
[13:17:10] <ofu> trygvis: i use this tower with 4 disks.. it is not ultra-quiet...
[13:17:54] <ofu> i wouldnt want to work next to one of these towers with 8 disks for 8 hours a day
[13:18:08] <trygvis> hmm
[13:18:27] <trygvis> I wondering if I should get one to use as a workstation in my living room :)
[13:19:49] <PerterB> not if you also like peace and quiet there :)
[13:20:02] <PerterB> but then, I find the fan on my laptop annoying at times
[13:25:43] <trygvis> try a macbook :)
[13:25:54] <trygvis> I always play music so it's not that important
[13:29:38] <razrX> trygvis: ditto, couldn't work without mah music
[13:30:39] *** FelipeBare has joined #opensolaris
[13:32:07] <PerterB> trygvis: it *is* a macbook (pro) ... under CPU load they get pretty loud, and I like to appreciate the quieter passages in music without that
[13:32:30] <trygvis> bummer
[13:33:39] <PerterB> meh, on the list of things that annoy me, it's not in the top 10 for sure
[13:36:14] *** kFuQ has quit IRC
[13:38:37] *** dpn` has quit IRC
[13:38:50] * trochej eats crap food
[13:38:52] <trochej> Mhhhmmm
[13:39:05] <trochej> KFC
[13:39:12] <trochej> Your most tasty syntetic chicken ever
[13:41:05] <trochej> I used to work next to a small room with four servers and one 12 drive matrix
[13:41:18] <trochej> the door was open whole day
[13:41:28] <trochej> I can take your 4 drive tower with no pain :)
[13:43:37] *** kloczek has joined #opensolaris
[13:44:28] *** nadir has joined #opensolaris
[13:44:33] <nadir> mount -F pcfs /dev/dsk/c3t0d0p0:c /mnt/usb/
[13:44:36] <nadir> mount: /dev/dsk/c3t0d0p0:c is not a DOS filesystem.
[13:44:57] <nadir> :o, I thought it was fat, but perhaps htfs, any idea how to mount?
[13:45:04] <kjetilho> any of you know how far back VIS 2 support has been in SPARC processors?
[13:48:28] <PerterB> nadir: fstyp? also (and I'm never really clear on this) I think unless you have a DOS extended partition (which is unlikely on a USB drive), you drop the :c
[13:49:30] <kjetilho> okay, VIS 1.0 is UltraSPARC I or II, VIS 2.0 is UltraSPARC III.
[13:49:37] <kjetilho> I wonder how SPARC64 fits into this
[13:50:04] <nadir> PerterB: damn your right
[13:50:11] <nadir> worked when I dropped the c
[13:50:17] <nadir> thanks
[13:51:33] *** RainDoctor has joined #opensolaris
[13:57:00] *** jonkri has joined #opensolaris
[14:05:26] *** Mdx4 has joined #opensolaris
[14:11:51] <delewis> kjetilho: given VIS is UltraSPARC-specific, I very much doubt SPARC64 has it.
[14:11:55] *** peteh has quit IRC
[14:12:20] <delewis> from my understanding Fujitsu took the same path that IBM did with pre-POWER6 processors and SIMD -- don't.
[14:12:23] *** peteh has joined #opensolaris
[14:12:55] <kjetilho> delewis: I found a reference which says UltraSPARC V has VIS 1.0, and VI has 2.0
[14:13:09] <delewis> huh?
[14:13:18] <kjetilho> ehh, SPARC64!
[14:13:58] <kjetilho> SPARC64 V came out in 2003, so it's pretty safe to ship software compiled with VIS 1.0.
[14:14:22] *** jonkri has quit IRC
[14:14:40] <delewis> no, it's not.
[14:15:08] <flyingparchment> 2003 doesn't sound very safe
[14:15:09] <delewis> VIS or any other SIMD instruction set is not specificied in the SPARCv9 specification, therefore, it's completely optional for an implementation to have such an instruction set.
[14:16:06] <delewis> if you want a binary to have the greatest amount of portability, then it is wise to strictly comply with SPARCv9 specification.
[14:16:18] <delewis> Solaris does provide VIS emulation, btw, as it does quad-precision arithmetic in software.
[14:16:48] <WickedWicky> but it's noticably slower, right?
[14:17:05] <delewis> if you execute a VIS instruction on a non-VIS system, it'll generate a trap that you can catch, and use a software routine.
[14:17:12] <delewis> WickedWicky: much.
[14:17:15] <delewis> but it does let your binaries "work"
[14:17:19] <delewis> anyway, off to class.
[14:17:23] <WickedWicky> good luck :)
[14:17:43] <kjetilho> depends on your target public, of course
[14:18:01] <kjetilho> Blastwave would be insane to enable VIS with all those SS20 still around :)
[14:18:24] <kjetilho> but stuff like Oracle could do it.
[14:21:45] <JWheeler> oh, delewis, you're here!
[14:21:55] <JWheeler> oh no... wait. he's not
[14:22:25] *** master_o1_master has joined #opensolaris
[14:22:54] *** timsf has quit IRC
[14:23:10] <WickedWicky> I dunno why
[14:23:15] <WickedWicky> but I am looking forward to xmas
[14:24:30] *** gavagai_ has joined #opensolaris
[14:25:29] *** Murmuria has left #opensolaris
[14:27:15] *** gavagai__ has joined #opensolaris
[14:31:28] <flyingparchment> apparently, i have two patches installed that deliver 1 fix each - the same fix
[14:31:29] <flyingparchment> i love solaris patching ;)
[14:33:36] <jmcp> away dleeping
[14:34:56] <dlg> n8
[14:36:07] *** master_of_master has quit IRC
[14:41:39] *** RainDoctor has quit IRC
[14:45:21] *** gavagai_ has quit IRC
[14:47:36] *** timsf has joined #opensolaris
[14:47:53] *** alanbur has joined #opensolaris
[14:49:05] <kjetilho> I'm disappointed to read that Dynamic Resource Allocation by poold isn't supported in a zone setting in S10u3 -- has this been implemented in OpenSolaris?
[14:54:52] *** Gekz[zzz] is now known as Gekz[PDA]
[15:03:08] *** Nishaway has joined #opensolaris
[15:04:28] *** sarahj has joined #opensolaris
[15:14:18] *** loke_ has joined #opensolaris
[15:17:15] *** WickedWicky has quit IRC
[15:23:33] *** karrotx has quit IRC
[15:27:25] *** loke has quit IRC
[15:32:41] *** alanbur has left #opensolaris
[15:33:05] *** Nishaway has quit IRC
[15:36:16] *** RaD|Tz has joined #opensolaris
[15:36:22] *** debuggerboy has left #opensolaris
[15:42:58] *** gavagai__ has quit IRC
[15:49:29] <quasi> kjetilho: have you checked U4?
[15:50:08] <kjetilho> quasi: well, reading the updated docs.  for my purposes, the FSS is probably a better solution anyway :)
[15:51:02] <quasi> kjetilho: fss for zones has been available all along s10
[15:51:18] <quasi> kjetilho: but U4 got memory capping as well
[15:52:30] *** nostoi has joined #opensolaris
[15:52:43] *** laca has joined #opensolaris
[15:55:33] <kjetilho> that's great, I'll have to read on :-)
[15:59:34] *** dnilsson has joined #opensolaris
[16:06:47] *** cherry_ has joined #opensolaris
[16:06:50] <cherry_> hello
[16:08:12] <cherry_> i installed the most recent sxce and it wont start Xorg, as there was no config file, i created one with Xorg -configure , and then it couldnt find a correct mode for my video card which is weird as it worked in the current sxde version, as well as working in linux
[16:08:40] *** tinman2k has joined #opensolaris
[16:08:51] <cherry_> when i switched the driver in the config file from intel to vesa
[16:09:06] <cherry_> i get a black screen
[16:13:22] <kjetilho> quasi: I can't seem to find the right syntax for allocating cpu-shares to the zone, though?
[16:14:29] <tinman2k> kjetiho: using zonecfg?
[16:14:44] <kjetilho> tinman2k: no, that's definitely an u4 addition
[16:15:03] <kjetilho> AFAICT
[16:15:17] <kjetilho> trying to use prctl (non-persistent, but that's fixable)
[16:15:34] <kjetilho> but it wants a specific PID, and init isn't acceptable
[16:16:01] <tinman2k> I am pretty sure you can do cpu sharing in u3
[16:16:06] <tinman2k> let me check real quick
[16:17:53] <quasi> kjetilho: there's a link from the zones community to brendang's site iirc - but it got easier with U4 iirc
[16:18:05] <kjetilho> yes, u4 looks *very* nice.
[16:19:21] <cherry_> anybody?
[16:20:01] <kjetilho> cherry_: perhaps you could copy your config file from Linux or current sxde?
[16:20:09] <kjetilho> worth a shot, anyway
[16:20:10] <quasi> cherry_: not apart from looking at the errors in the x log
[16:20:11] *** alanc_away has joined #opensolaris
[16:20:20] <tinman2k> sorry cherry, I may not have got anything you said for the driver switch. are you using sparc or x86?
[16:20:26] <tinman2k> x86 duh. intel
[16:20:29] <tinman2k> nm
[16:20:30] <cherry_> i coppied from linux and it had loads of issues
[16:20:38] <cherry_> i dont have the sxde config
[16:20:44] <tinman2k> are you using xsun or x11?
[16:20:45] <cherry_> but im suprised vesa fails
[16:20:52] <cherry_> i believe its running Xorg
[16:20:57] <cherry_> i start it from gdm
[16:21:26] <tinman2k> ok. I had problems trying that with xsun. xorg didn't give me any problems so I don't think I will be much help.
[16:23:43] <tinman2k> kjetiho, using zonecfg to allocate cpu shares has been availible for awhile. such as zonecfg -z zonename > add rctl > set name=zone.cpu-shares > set value=(priv=privleged,limit=30,action=none) end
[16:23:46] *** kaiwai has joined #opensolaris
[16:23:57] <tinman2k> of course that number (30) is just the %
[16:24:18] <tinman2k> but I run Trusted Extensions so that may have something to do with it
[16:24:27] <tinman2k> that is from my notes i took awhile back
[16:24:47] <kjetilho> tinman2k: ah, right
[16:25:11] <kaiwai> damn, someones really ballsed up the java website
[16:25:46] <tinman2k> the prctl is used for Global Zone I believe
[16:25:50] <tinman2k> FSS
[16:26:57] <kaiwai> http://swsdlcuat.central.sun.com/ECom/docs/Welcome.jsp?StoreId=22%20&%20PartDetailId=jdk-6u3-oth-JPR%20&%20SiteId=JSC%20&%20TransactionId=noreg%20& <-- am I the only one who see's somethingn wrong with that address?
[16:26:58] <kjetilho> tinman2k: it's percent?  I thought it would just add up the shares and use the proportional value.
[16:27:30] <tinman2k> you can, that example was if I wanted to allocate 30% to that zone.
[16:28:09] *** Gekz[PDA] is now known as Gekz[sleep]
[16:31:53] <kaiwai> hmm, dinner
[16:33:02] *** calumb has joined #opensolaris
[16:34:20] <kaiwai> 'allo calumb
[16:35:19] *** moo_cow has joined #opensolaris
[16:36:07] <quasi> kjetilho: fss is a proportional value
[16:36:27] *** moo_cow has quit IRC
[16:37:13] <calumb> hi kaiwai
[16:37:47] *** _Megaf has joined #OpenSolaris
[16:39:19] *** Joerg has quit IRC
[16:40:09] *** calumb is now known as calAFK
[16:41:06] <kaiwai> anyone running Vermillion B75?
[16:41:30] *** Pietro_S has joined #opensolaris
[16:42:26] *** jonkri has joined #opensolaris
[16:44:41] *** Fullmoon has quit IRC
[16:47:17] *** calAFK is now known as calumb
[16:48:02] <madhatter> Can somebody tell me, if the onboard sata contoller chipset VIA VT8237S is working with OpenSolaris?
[16:48:16] <madhatter> In the hcl is only the vt8235
[16:49:44] <tinman2k> The HCL, unless it is verified by sun is full of user submitted hardware that they tested.  It may be one of those things you just have to try.
[16:50:22] <madhatter> tinman2k: I tried a lot already. I am getting short on money ;)
[16:50:49] <Berny_> get a sponsor... or use the 14 days money back ;-)
[16:51:25] <tinman2k> Let me see if anyone here remembers the onboard SATA controller we tested last.
[16:52:03] <flyingparchment> wow that's annoying.. the IDR 'do you really wish to install this patch' prompt is displayed for every zone
[16:52:08] <quasi> madhatter: http://www.opensolaris.org/jive/thread.jspa?messageID=90750 if VT8237S is the same as VT8237
[16:52:33] <quasi> nv sata is a pretty safe bet
[16:52:57] <tinman2k> flyingparchment: is that 'fixed' in U4?
[16:53:33] <flyingparchment> tinman2k: i don't know, i'm still using U3
[16:54:01] <tinman2k> We are testing u4, I will try it out and see
[16:56:41] <madhatter> quasi: Thanks, but it'
[16:56:44] <madhatter> hups
[16:57:00] <madhatter> quasi: Thanks, but it's not 100% safe though...
[16:57:27] <tinman2k> There was a setting on one of ours we had to change to get it to boot
[16:57:46] <tinman2k> Top of my head I don't remember.
[16:58:39] *** sommerfeld has joined #opensolaris
[16:58:40] *** ChanServ sets mode: +o sommerfeld
[17:01:09] *** flyingparchment has quit IRC
[17:01:27] <quasi> yay, zfs crypto alpha
[17:01:28] *** cherry_ has quit IRC
[17:02:02] <tinman2k> quasi: really?
[17:02:24] <quasi> tinman2k: http://opensolaris.org/os/project/zfs-crypto/phase1/alpha/
[17:02:35] <tinman2k> Damn, and I thought I was going to be productive at work today...
[17:05:05] *** Vanuatoo has left #opensolaris
[17:06:30] *** wms has joined #opensolaris
[17:07:03] *** Vanuatoo has joined #opensolaris
[17:08:25] <logic_> quasi: strange the non debug file doesn't exist :(
[17:09:27] <logic_> quasi: ow i see, the link is wrong, if you go to : http://dlc.sun.com/osol/zfscrypto/downloads/alpha/ you see the 2 files there
[17:09:37] <tinman2k> http://dlc.sun.com/osol/zfscrypto/downloads/alpha/zfscrypto-nondebug-20071004.i386.tar.bz2
[17:09:41] <tinman2k> lol
[17:09:46] <tinman2k> thats what I did
[17:10:13] *** tsoome has quit IRC
[17:10:23] <logic_> tinman2k: hehe ok , are you gonna try it? i am wondering how it works before i try it ;)
[17:11:13] *** ichigo has joined #opensolaris
[17:12:10] <tinman2k> I have a few machines around here. I will through it on one around lunch. See what I can break
[17:12:51] <tinman2k> see how well it plays with Trusted Extensions
[17:13:56] <logic_> tinman2k: hehe thats handy, i dont want to break my server
[17:15:28] *** blindfish has joined #opensolaris
[17:17:16] <tinman2k> I just go back into the warehouse and snag a machine.  Since we have to migrate around 2000 installs of tsol 8 to SOL10 TX in about a year and a half, they let us play with anything we want while testing
[17:18:26] *** flyingparchment has joined #opensolaris
[17:23:19] <logic_> tinman2k: wish i could do that :)
[17:23:20] <PerterB> that's a lot of tsol 8....
[17:24:24] <quasi> in a setup like that I would almost expect tinman2k to be connecting from a .mil ;)
[17:24:43] <PerterB> nah, I would expect him to be hiding that part :)
[17:24:45] <tinman2k> Yeah, but fortunately we just have to go around to all the locations and make a few master drives and have the local admins dupe them.  Same hardware at all locations
[17:27:05] <tinman2k> quasi, perterb: :-X
[17:27:32] <PerterB> that looks like duct tape over someone's mouth :)
[17:28:08] *** jonkri has quit IRC
[17:34:03] <elektronkind> Sun will announce the new T2 based servers as well as preview the new AMD quad core systems on October 9th.
[17:34:10] <elektronkind> * Sun SPARC Enterprise T5120 and T5220 servers  - T2 servers
[17:34:12] <elektronkind> * Sun Blade T6320 server - T2 blade server
[17:34:20] <elektronkind> * Next-generation Galaxy systems based on the AMD Opteron processor (preview) - Barcelona
[17:34:22] *** linux_user400354 has joined #opensolaris
[17:34:25] <elektronkind> yay
[17:34:47] <kszwed> i had a 5120 two months ago.
[17:34:52] <kszwed> very nice units
[17:34:54] <flyingparchment> veritas is hard :( it won't see my nice array
[17:35:24] <the-decider> hmm, the big question as always is -- wuzzitcost?
[17:35:25] <elektronkind> kszwed: the 5120 and 5220 1u/2u analogs ?
[17:35:29] <Vanuatoo> Hopefully when Phenom is out, Ultra 2x line will also be updated
[17:35:31] <kszwed> elektronkind, yes
[17:36:03] <kszwed> elektronkind, 5120 has lots of little sas disks, dvd and redundant psus so they're little rockets.
[17:36:11] <elektronkind> vera nice
[17:36:21] * elektronkind does a borat thumbs up
[17:36:25] <the-decider> kinda like the intels, eh?
[17:36:31] <the-decider> good move on sun's part.
[17:36:40] <elektronkind> I wonder if they use the same system chassis
[17:36:54] * the-decider thinks its funny watching the CIO play "AV club"
[17:37:23] <elektronkind> hopefully they incorporate micro channel archetecture buss
[17:37:51] <the-decider> microchannel!  Yay!
[17:38:02] <ofu> does anybody know if the new barcelona box uses all 4 ht links?
[17:38:03] <the-decider> Then I'll have a use for all of those tokenring cards.
[17:38:23] <ofu> 5120 is a very nice system, yes
[17:38:42] *** jonkri has joined #opensolaris
[17:39:03] * the-decider laughs
[17:39:30] <wesolows> all our systems from here on out will be MCA-based, yes
[17:39:37] <wesolows> though ISA will be available as an option
[17:40:05] <ofu> as long as they are shipped with isa configuration floppies
[17:40:08] <wesolows> PCIe just wasn't meeting our customers' performance needs
[17:40:24] <wesolows> no, you're thinking of ISAPNP
[17:40:34] <wesolows> only jumper cards are supported
[17:40:44] <wesolows> but we do make sure that IRQ 5 and 7 are both free
[17:40:46] <the-decider> and there were just TOO many options available.  Its easier to support the limited MCA cards available.
[17:40:54] <tomww> hom many pepole needed to confiugre just one card? 5 or so?
[17:41:01] <the-decider> you did need those floppies that had the card-specific configs for MCA stuff.
[17:41:20] * the-decider misses the PS/2 ;)
[17:41:26] <wesolows> you'll have a choice of 360k or 720k floppy drives
[17:41:34] <PerterB> luxury
[17:41:38] <tomww> high density
[17:41:42] <wesolows> 1.2M "coming soon"
[17:41:49] *** kszwed has quit IRC
[17:42:28] <sommerfeld> mmm, isa prug and pray
[17:43:01] * wesolows got better results by skipping the prayer and showing the hardware a 30-pound sledgehammer before booting
[17:43:21] *** dlg has quit IRC
[17:43:24] *** bnitz has left #opensolaris
[17:43:33] *** comay has quit IRC
[17:43:44] <tomww> ah, now i realize what "my computer is hammered" tells. *g*
[17:44:19] <wesolows> I especially liked the cards that had jumpers which made it possible to jumper +12 directly to GND
[17:44:47] *** thepumpkin_w has joined #opensolaris
[17:44:49] <tomww> sales opporunities to be jumpered
[17:45:28] <wesolows> Sometimes the computer was still usable; the card, never
[17:45:54] <the-decider> ah, those were the good-ole days
[17:46:23] <sommerfeld> Commandment 0x6 - Thou shalt not connect PWR to GND.
[17:47:37] <wesolows> Much like the prohibition on adultery, the commandment contains much wisdom but at the end of the day it's still kind of a killjoy.
[17:48:51] <sommerfeld> (from http://hacks.mit.edu/Hacks/by_year/1992/cathedral_7/father_tool_description.html)
[17:49:08] *** tsoome has joined #opensolaris
[17:50:07] <wesolows> indeed; the real killjoy there is commandment 0x5
[17:50:40] <paul> wesolows: ofu might have been thinking of EISA..
[17:50:58] <wesolows> Could be, yes
[17:51:00] <delewis> OK, this is odd. Looks like /var/tmp isn't created in a zone on S10u4.
[17:51:03] <wesolows> now there was a real abortion
[17:51:06] *** kaiwai has quit IRC
[17:51:13] <PerterB> <homer>young lady, in _this_ house we obey the laws of thermodynamics</homer>
[17:51:14] <delewis> anyone seen this problem in Nevada?
[17:51:21] *** hile_ has quit IRC
[17:51:37] <sommerfeld> yah, isapnp seemed to involve backporting some of eisa to isa
[17:52:11] <the-decider> Does solaris support ESDI drives? ;)
[17:52:28] <sickness> should...
[17:52:52] *** thepumpkin_w has left #opensolaris
[17:53:21] <elektronkind> oh, which brings me to a question I voiced to the-decider yesterday. What were those huge drives that were often found attached to sun3 systems called?
[17:53:28] <elektronkind> ISTR a 3 letter acronym
[17:53:54] <flyingparchment> does veritas not support iscsi disks or something?  it just shows my lun as "error"
[17:53:54] <tinman2k> BFD
[17:53:55] <elektronkind> they had two ribbon cable connectors
[17:54:04] <wesolows> I something I think
[17:55:25] <elektronkind> ah. SMD
[17:55:32] <the-decider> yes!  that was it.
[17:55:37] <wesolows> yep, IPI
[17:55:42] <wesolows> oh, different interface then
[17:55:53] <the-decider> it showed up as /dev/hernia
[17:56:04] <elektronkind> rofl
[17:56:27] <the-decider> there wasa reason they were always at the bottom of the racks.
[17:56:48] <wesolows> those big old eagle drives with their own AC power connector?
[17:57:37] <elektronkind> complete with cooling fans
[17:57:40] <the-decider> the ones that used lead platters ;)
[17:57:54] *** jonkri has quit IRC
[17:57:56] *** _Megaf has quit IRC
[17:58:20] *** RealWickedWicky has joined #opensolaris
[17:58:23] <RealWickedWicky> howedy
[17:58:26] *** RealWickedWicky is now known as WickedWicky
[18:01:15] *** calumb is now known as calMTG
[18:02:17] *** TrogL has joined #opensolaris
[18:03:29] *** bubbva has joined #opensolaris
[18:04:31] *** KermitTheFragger has joined #opensolaris
[18:04:52] *** stevel has joined #opensolaris
[18:04:53] *** ChanServ sets mode: +o stevel
[18:06:11] *** victori_ has quit IRC
[18:07:02] *** victori_ has joined #opensolaris
[18:08:21] *** mikefut_ has joined #opensolaris
[18:08:21] *** mikefut has quit IRC
[18:10:59] <elektronkind> hey sommerfeld, this stuart dempster album is nice. it was on itms
[18:12:44] *** cmihai has joined #OpenSolaris
[18:16:19] <TrogL> if you do prtcl -n project.something -i project user.root, does that affect all processes, even those run on other accounts?
[18:16:58] <TrogL> s/prtcl/prctl/
[18:19:28] *** stevel has quit IRC
[18:19:42] *** stevel has joined #opensolaris
[18:19:43] *** ChanServ sets mode: +o stevel
[18:20:26] *** stevel has quit IRC
[18:20:37] *** stevel has joined #opensolaris
[18:20:38] *** ChanServ sets mode: +o stevel
[18:21:25] <elektronkind> hey stevel
[18:21:33] <elektronkind> still considering that colfax box?
[18:21:36] <stevel> heya
[18:21:47] * elektronkind is in the same predicament as you
[18:21:49] <stevel> yeah, it looks like a good fit for what i need, i've heard good things about them as a vendor, and they're local
[18:22:09] <elektronkind> I'm just leary about the impi aspect
[18:22:13] <flyingparchment> okay, so i added /dev/dsk/c2t6d0 to veritas with 'vxddladm addforeign', but it doesn't appear in 'vxdisk list'.  why not?
[18:22:14] <elektronkind> er ipmi
[18:22:20] <stevel> plus, apparently, the co-op i colo with might get an 8% discount there
[18:22:27] *** Cyrille has quit IRC
[18:22:31] <elektronkind> flyingparchment: vxdctl enable ?
[18:22:37] <flyingparchment> elektronkind: done many times ;)
[18:22:43] <stevel> yeah i don't know what their ipmi interface is like
[18:23:02] <elektronkind> probably requires ipmitool and has no web ui
[18:23:11] <elektronkind> ilom has spoiled me
[18:23:26] <stevel> i can deal, i don't need to get to the serial console that often fortunately
[18:23:40] * the-decider was pretty impressed with the Dell 850 -- the current model of that is the 860
[18:24:03] <delewis> the 860 is OK if you're on an extreme budget.
[18:24:20] <flyingparchment> ime, anything requiring ipmitool should be avoided.  i never made it work properly.
[18:24:21] <delewis> I've set one up for a client that functions as an offsite replication/backup solution (ZFS+rsync)
[18:24:26] <elektronkind> flyingparchment: does it show in a 'vxdisk list' ?
[18:24:41] * stevel wants a machine with 4 SATA bays.
[18:24:45] <stevel> i think the 860 only has 2
[18:24:45] <the-decider> well, if you're looking for a 1u server to load up with a couple phat SATA drives to house your personal email & data backup, is perfect ;)
[18:24:45] <elektronkind> http://www.colfax-intl.com/jlrid/SpotLight_more.asp?L=120&S=42&B=1500
[18:24:51] <delewis> stevel: yep.
[18:24:52] <elektronkind> I take it you're looking at that one eh
[18:24:55] <flyingparchment> elektronkind: no.  i did try adding the directory instead (blockdir=...), then an entry for every disk appears, but they're all 'error' except for the root disk
[18:25:19] <stevel> elektronkind: http://www.colfax-intl.com/jlrid/SpotLight_more.asp?L=74&S=2&B=1390
[18:25:20] <flyingparchment> i did label and newfs the disk, so i know it's working.
[18:25:26] <stevel> the quad core intel one...
[18:25:46] <the-decider> hmm, looks nice
[18:26:00] <flyingparchment> stevel: saw sun's new 1RU server with 8 sas bays? ;)
[18:26:21] <stevel> flyingparchment: yeah, but i need cheap storage. and i'd rather have 4 3.5" SATA bays with 750GB drives than 8 2.5" SAS drives
[18:26:31] * delewis is thinking about purchasing one of those or an x4100 once he has a few paychecks on his new job
[18:27:18] <stevel> if i could afford a bunch of SAS drives, then i'd love to have the x4150; but i'll have to find something cheaper for now :)
[18:27:57] *** slowhog has joined #opensolaris
[18:28:02] <flyingparchment> elektronkind: i'm wondering if something about this array is confusing veritas (it's an infortrend iscsi array..)
[18:28:41] <elektronkind> flyingparchment: well veritas should at least be able to see the vx disk lable on it
[18:28:53] <elektronkind> slice 4 iirc
[18:29:02] <elektronkind> it's been a while since I've had to fool with vxvm
[18:29:08] <flyingparchment> there's no vx label yet - i can't do anything in vxdiskadm, it just says 'invalid disk'
[18:29:31] <elektronkind> so are you adding a disk in, or trying to use one that was previously under vx control?
[18:29:33] <the-decider> veritas has detedted your purchas of affordable storage.
[18:29:36] * elektronkind missed that part
[18:29:57] <the-decider> and has decided that since you paid big bucks for your veritas license, you should be able to afford better.
[18:30:10] <flyingparchment> i'm using the free veritas :P
[18:30:30] <the-decider> there's a free veritas?
[18:30:31] <elektronkind> flyingparchment: any luck trying to initialize it with vxdiskadm ?
[18:30:43] <flyingparchment> the-decider: storage foundation basic, only supports 4 volumes and 4 filesystems
[18:31:03] <flyingparchment> elektronkind: nope.  Select disk devices to add: [<pattern-list>,all,list,q,?] c2t6d0    No matching disks found.
[18:31:12] <the-decider> perhaps they have some limitation on the types of devices they "support"
[18:31:15] <flyingparchment> it is in the list, but as 'error'
[18:31:20] <elektronkind> yar
[18:31:25] <elektronkind> I had this problem, once
[18:31:34] <flyingparchment> the-decider: maybe, but i set it up on a similar crappy infortrend (SCSI instead of iSCSI) and it worked there
[18:31:38] <elektronkind> but years ago. I'm talking vxvm 3.2 at least ;)
[18:32:29] *** estibi has joined #opensolaris
[18:32:31] <the-decider> Veritias could be checking your actual device path, and saying "hey, this isn't scsi/fc, I dunna wanna play"
[18:33:10] <elektronkind> flyingparchment: one thing you can try is use format to go in and zero out the slices and then 'vxdctl enable', and then vxdiskadm it
[18:33:47] <elektronkind> adding drives was something veritas never got 100% reliable/foolproof
[18:34:12] <elektronkind> also, disable DMP if that's on
[18:34:19] <delewis> nor when a path completely failed.
[18:34:33] * delewis has nightmares about vgdisabled states
[18:34:45] <flyingparchment> maybe i need to reboot after the initial config stuff?  i assumed not, since it started all the daemons..
[18:35:12] <elektronkind> doing that probably won't hurt at least
[18:35:40] * flyingparchment tries
[18:36:32] *** flyingparchment has quit IRC
[18:38:32] *** Erwann has quit IRC
[18:41:54] *** victori_ has quit IRC
[18:44:16] *** Mdx4 has quit IRC
[18:46:05] *** flyingparchment has joined #opensolaris
[18:46:15] <flyingparchment> nope, still error :/
[18:48:16] <madhatter> re
[18:49:29] *** cmihai_ has joined #OpenSolaris
[18:50:44] *** cmihai_ has quit IRC
[18:54:24] *** peteh has quit IRC
[18:57:46] *** victori_ has joined #opensolaris
[18:59:59] <madhatter> quasi: I looked up those chipsets of via, seems that vt8251 is the last version of vt8237S and differs only in the 2x Sata-Lite ports instead of two more full sata II ports
[19:04:52] *** RaD|Tz has quit IRC
[19:06:13] *** hyegeek has joined #opensolaris
[19:07:32] <Teltariat> Can you create zpools on slices?
[19:07:38] <elektronkind> yes
[19:07:56] <Teltariat> k, thanks
[19:08:14] *** cmihai has quit IRC
[19:09:19] *** vega| has joined #opensolaris
[19:09:28] <vega|> Hi
[19:09:50] *** tombhadAC has quit IRC
[19:09:52] <vega|> Any1 tried 8/07 edition of Solaris 10 for x86 architecture ?
[19:10:20] <elektronkind> I'm running it now on a x4100
[19:10:29] <vega|> Wow
[19:10:36] *** migi has quit IRC
[19:10:39] <vega|> It's Intel powered server or AMD ?
[19:10:43] <elektronkind> amd
[19:10:48] <vega|> Kewl.
[19:10:50] *** JBeck has joined #opensolaris
[19:10:51] *** ChanServ sets mode: +o JBeck
[19:10:52] <elektronkind> quad opteron
[19:10:54] <vega|> What are additions ?
[19:10:55] <wesolows> most likely it's powered by natural gas or coal, actually
[19:10:58] <vega|> Linux support ?
[19:11:07] <the-decider> quad laser!
[19:11:07] <vega|> I heard
[19:11:12] <vega|> This edition has
[19:11:22] <vega|> some inbuilt linux compatible tools ...
[19:11:29] <elektronkind> BrandZ
[19:12:01] <vega|> I tried to install Samba on Solaris 10 11/06 edition for my SPARC
[19:12:05] <vega|> I failed.
[19:12:16] <vega|> I removed the samba package
[19:12:26] <vega|> They made the process complicated.
[19:12:54] <vega|> By default there is no SWAT location..1 must create manually and edit sample file provided by Sun
[19:14:12] <wesolows> how is this different from any other samba installation?
[19:15:08] <vega|> I have to bang my keyboard three times to install samba on Solaris 10...Hehe
[19:15:16] <vega|> type >>>> pkginstall
[19:15:22] <vega|> sigh ~
[19:16:12] <wesolows> But Samba is installed already if you select the right package metacluster.
[19:16:18] <vega|> SWAT is not compatible with the given pkg
[19:16:20] <wesolows> All you should have to do is configure it.
[19:16:37] <vega|> While installing Solaris ?
[19:16:49] <vega|> Just like I do in Slackware ?
[19:16:52] <wesolows> Yes; I seem to have smbd, smbpasswd, et al on my system
[19:17:10] <RElling> In Nevada, Samba is as easy as a mouse click... Java Cup -> Administration -> Shared Folders...
[19:17:19] <vega|> I missed it .
[19:17:20] <wesolows> I don't know about swat; istr it doesn't install any binaries?
[19:17:27] *** WickedWicky has quit IRC
[19:17:35] <vega|> I am going to install latest edition on my ThinkPad
[19:17:40] <wesolows> well, swat(1M) is there and seems helpful enough
[19:17:41] <vega|> Hehe
[19:17:46] <wesolows> This should be a no-brainer.
[19:17:50] <vega|> Wow...GUI ..Java GUI ?
[19:18:10] <RElling> I don't know or care if it is a Java GUI... it just works
[19:18:16] <vega|> Baaaaaaa
[19:18:21] *** deather_ has joined #opensolaris
[19:18:32] *** catena has joined #opensolaris
[19:18:34] <vega|> I'll try it.
[19:18:38] <RElling> its almost as easy as a Mac
[19:18:41] <wesolows> If you care more about ease of use than about availability or support, SX (Nevada) will likely be better than S10.
[19:18:42] <vega|> On my laptop ...
[19:18:49] <vega|> Mac..We have Mac G5
[19:18:59] <vega|> It eats up lot of power
[19:19:31] <RElling> thats why Apple gave up on IBM and went with Intel (IBM roadmap was dead for Power laptops)
[19:19:32] <vega|> SMPS of G5 consumes too much electricity
[19:20:10] <RElling> at least one reason, but an important one seeing as how more laptops are sold than desktops now
[19:20:11] <vega|> Apple lake for support in other countries than US.
[19:20:38] <vega|> I live in India. They are offering 3 years warranty just for additional 28 US$
[19:20:56] <vega|> Yeah right. Every1 looking for mobility .
[19:21:13] <wesolows> They obviously don't expect to service any of those laptops under that warranty then.
[19:21:30] <vega|> Not just laptops but desktops too
[19:21:52] <vega|> Ofcourse yes..And beleive me they are offering on site warranty
[19:21:55] <wesolows> $28 won't cover the first 30 minutes of diagnosis, much less parts or labour to actually fix it if broken, or shipping to the nearest depot, etc.
[19:21:56] *** theRealBallchalk has joined #opensolaris
[19:22:13] <vega|> Because Apple want to make some market in India..Intel still rules..But I really like AMD
[19:22:19] <wesolows> So they either really really believe there won't be any defects, don't expect anyone to present a machine for service, or plan to dishonour the warranty.
[19:22:31] <vega|> You know what last month SMPS of 1 Power G5 blown up
[19:22:32] <wesolows> Or, I suppose there's a fourth alternative: they intend to lose a boatload of money.
[19:22:48] <vega|> They fixed it for free...Original price of that SMPS is more than 300 US$ ...
[19:22:56] <vega|> Nope !
[19:23:09] <vega|> They want business ! This is foundation !
[19:23:30] <wesolows> You're essentially saying their plan is option 4: lose a billion dollars on warranty to buy market share.
[19:23:40] *** stevel has quit IRC
[19:23:41] <wesolows> Might be a good strategy, might not.
[19:24:01] <vega|> True Apple user knows that Apple PC rarely gets screwed up..
[19:24:17] <RElling> actually, Apple has one of the lowest warranty reserves in the industry, ~2%, I'd love to get Sun to that level
[19:24:46] <wesolows> I don't know; this office has a number of extremely experienced engineers who have over a decade of Apple user experience, and I know at least one has had his Macbook in for service at least twice in the past year.
[19:24:52] <vega|> HP , IBM offering 5 years warranty even for entry level servers these days in India
[19:25:03] <delewis> a true Apple user also doesn't appreciate the observability tools that he lacks to tell if his system is 'screwed up' or not.
[19:25:12] <wesolows> delewis++
[19:25:16] <vega|> Must be "Made in China"..Hehe
[19:25:21] *** MattMan is now known as MattAFC
[19:25:22] *** tcuji`e has quit IRC
[19:26:03] <vega|> Yes...because those true users are nothing but bunch of graphic designers....ADOBE is there world...Hehe...They enjoy the apple in 1 hand
[19:26:08] <RElling> ignorance is bliss sometimes... look at the trouble we had when we spewed copious DRAM error messages to syslog
[19:26:22] *** RealWickedWicky has joined #opensolaris
[19:26:24] <vega|> hmm
[19:26:28] <RElling> remove the messages, remove the psychological impact
[19:26:41] *** smesjz has joined #opensolaris
[19:26:48] <delewis> let Fault Manager sort it out. :-)
[19:26:52] <vega|> Cisco says syslog is the most good way to store logs...Haha
[19:27:25] <vega|> Any1 ever tried to open up Cisco PIX and fix normal SD Ram ?
[19:27:39] <wesolows> cisco's management tools are in the 1970s, completele useless for the most part
[19:27:42] <delewis> the memory isn't the only thing that's normal about a PIX.
[19:27:46] <delewis> it's commodity hardware.
[19:27:47] <delewis> all of it.
[19:27:49] <delewis> 100%.
[19:27:54] * elektronkind reads zfs-discuss and his head spins
[19:28:04] <elektronkind> why must everyone talk in terms of 1000 disk servers
[19:28:22] <delewis> elektronkind: I noticed that to in the latter part of the Direct IO thread.
[19:28:28] <Teltariat> How can I ask Solaris to give me a listing of all the filesystems that it knows how to work with?
[19:28:30] <RElling> do elektrons spin to the left or right? :-)
[19:28:31] <elektronkind> that's what I just read
[19:28:52] <delewis> I guess they're wanting to maximize saturation visibility.
[19:28:55] <timsf> ls /kernel/fs
[19:29:00] <elektronkind> RElling: hard to say. some days I dont even know what shell I'm in :/
[19:29:04] <Teltariat> timsf: Thank you kindly
[19:29:52] <timsf> (unless someone's gone & installed filesystem drivers elsewhere...)
[19:30:02] *** alfism has joined #opensolaris
[19:30:03] <vega|> I like BASH any time
[19:30:15] <vega|> When I telnet i never use kron shell
[19:30:20] <vega|> Disguessting...
[19:30:21] * elektronkind was referring to electron enery levels
[19:30:26] <wesolows> tell us what you think of ksh93
[19:30:29] <delewis> you still use telnet?
[19:30:35] <vega|> Haha
[19:30:39] <wesolows> elektronkind: yes, I got it and was about to make a ksh93 physics joke, but thought the better of it
[19:30:43] * delewis checks the calendar to make sure this isn't 1995.
[19:30:50] <vega|> We dont have cluster environment
[19:31:01] <vega|> Just a small server room
[19:31:02] <wesolows> what's clustering got to do with telnet?
[19:31:04] <delewis> what does that have to do with anything?
[19:31:06] <wesolows> no one uses telnet any more
[19:31:08] <vega|> Containing 7 Sun boxes
[19:31:09] <delewis> use SSH. :-P
[19:31:09] <wesolows> use ssh instead
[19:31:12] 
[19:31:13] <vega|> I dont no ...
[19:31:13] *** Mdx4 has joined #opensolaris
[19:31:20] <vega|> SSH ...Hmmm ....
[19:31:21] <elektronkind> how'd that ? get in there
[19:31:21] <the-decider> elektronkind: right on!
[19:31:24] <delewis> elektronkind: :-)
[19:31:46] *** bengtf has quit IRC
[19:31:47] <vega|> I am not a solaris expert :D
[19:31:51] <elektronkind> my keyboard is crapping out
[19:31:52] <vega|> I am curious to know
[19:32:03] <wesolows> instead of typing telnet <host>, type ssh <host>.
[19:32:03] <delewis> vega|: SSH isn't Solaris-specific.
[19:32:04] <vega|> Why top command is not included by default in Solaris 10
[19:32:12] <vega|> I know i know
[19:32:13] <wesolows> because you should use prstat instead
[19:32:14] <delewis> vega|: because top isn't part of any standard.
[19:32:20] <vega|> I use it on OpenSuSe, Slackware, Debian
[19:32:24] <vega|> Hmm
[19:32:27] <vega|> In Sol 10
[19:32:30] *** calMTG is now known as calumb
[19:32:31] <vega|> I have to type
[19:32:33] <vega|> >>>> ./top
[19:32:41] <wesolows> Use prstat instead
[19:32:42] <delewis> any POSIX implementation is free to implement whatever interactive, real-time process observability tool it likes.
[19:32:47] <wesolows> top is garbage
[19:32:50] <delewis> and besides, top is terribly inefficient.
[19:32:53] <vega|> Cool
[19:33:01] <the-decider> delewis: but so pretty
[19:33:12] <delewis> the-decider: prstat -Z is pretty, too.
[19:33:26] <vega|> Hehe
[19:33:40] <vega|> Easy to find above pkg ?
[19:33:44] <wesolows> it's conversations like this that will someday find me standing on the rook of the transamerica building waving a bazooka and laughing maniacally
[19:33:49] <wesolows> *roof
[19:34:07] <vega|> hahahahahahahahahahaha
[19:34:12] <the-decider> delewis: NICE.  Thanks for pointing that out.
[19:34:14] <wesolows> prstat is installed by default
[19:34:19] <delewis> the-decider: sure.
[19:34:22] <vega|> I am a part time Solaris admin...LoL
[19:34:24] <wesolows> you don't need a separate package
[19:34:26] <delewis> it's much  nicer when you have a few zones :-)
[19:34:27] <delewis> anyway
[19:34:28] <delewis> off to lunch
[19:34:33] <vega|> What ? Oh....
[19:34:36] <RealWickedWicky> eat well
[19:34:37] *** calumb has quit IRC
[19:34:39] <sommerfeld> .. or -J if you use projects
[19:34:52] *** cydork has quit IRC
[19:34:55] <vega|> where pstat is located @ ?
[19:35:00] <wesolows> /usr/bin
[19:35:06] <vega|> thanx
[19:35:27] <vega|> I tried MySQL on Solaris 9 last month
[19:35:33] *** deather has quit IRC
[19:35:34] <vega|> It was a unsuccessful try
[19:35:38] <vega|> mysqld failed
[19:35:45] <vega|> Even I given privilage and everything
[19:35:51] <vega|> My SQL Daemon failed too
[19:35:59] <wesolows> In Nevada at least mysql is also installed with the system, in /usr/sfw
[19:36:06] <vega|> I went through doc before installing
[19:36:08] <Mdx4> vega|: get a real RDBMS use postgres :)
[19:36:17] <RealWickedWicky> Oracle ftw
[19:36:21] <vega|> Developers wanted MySQL only
[19:36:39] <vega|> Otherwise we already have many instances Of Oracle 8i, 9i , 10g
[19:36:45] <Mdx4> vega|: and this is the reason why they aren't called PROGRAMMERS.
[19:36:46] <vega|> Waiting for 11 release
[19:36:52] <vega|> Yes
[19:36:57] <vega|> They are crap
[19:36:58] <vega|> They dont no anything
[19:37:10] <vega|> Not even version of MySQL which they want
[19:37:26] <vega|> They just want to sit and start work without much efforts
[19:37:45] <vega|> They are too lazy ...They want SAMBA mapping
[19:37:48] <vega|> duh ~~
[19:38:03] <Mdx4> vega|: you have looked if sun has done its package for mysql ?
[19:38:22] <vega|> I got not much info
[19:38:32] <vega|> I posted my problem @ opensolaris forum
[19:38:46] <vega|> But replies I found didnt work
[19:38:51] <vega|> I left that thing already.
[19:38:54] <Mdx4> get pkg-get and install it via pkg-get .
[19:38:55] *** stevel has joined #opensolaris
[19:38:55] *** ChanServ sets mode: +o stevel
[19:39:04] <vega|> But still not happy because problem remained as it is
[19:39:26] <vega|> How quick installation of NFS is ?
[19:39:46] <vega|> If I do on Linux box ...Solaris box will support ?
[19:39:47] <wesolows> it's part of the OS
[19:39:50] <wesolows> no installation needed
[19:39:57] <vega|> Hmmm
[19:40:09] <vega|> I want single user ID for all linux/Solaris boxes
[19:40:20] <vega|> For every user
[19:40:28] <wesolows> GNU/Linux's NFSv4 support is broken, according to some reports.  If you use GNU/Linux clients with Solaris servers, you should consider disabling v4 on the clients.
[19:40:34] <vega|> Single ID will work on all RedHat/Solaris server
[19:40:35] <wesolows> That's NIS or LDAP.
[19:40:48] <wesolows> NIS server and client are part of Solaris by default.
[19:40:50] <vega|> Sorry ..Yes NIS
[19:40:52] <wesolows> LDAP client as well.
[19:40:54] *** timsf has quit IRC
[19:40:55] <vega|> LDAP
[19:40:57] <Mdx4> vega|: http://www.mysql.org/downloads/mysql/5.1.html#solaris-pgadd
[19:41:08] *** sahafeez has joined #opensolaris
[19:41:15] <vega|> I have to try it out
[19:41:20] <wesolows> If you want an LDAP server, you need to get JES/SunDS or OpenLDAP (on the companion CD) or another server of your choice.
[19:41:29] <vega|> I m looking at the link...thanx
[19:42:14] <vega|> You all from United States ?
[19:42:54] <wesolows> Many people here are from Australia, New Zealand, and Canada as well as elsewhere.  There are a few from India too.
[19:43:05] <vega|> Includes me...
[19:43:06] <wesolows> People who are awake now tend to be North Americans. :-)
[19:43:21] <RealWickedWicky> and europe
[19:43:25] <wesolows> yep
[19:43:25] * RealWickedWicky raises hand
[19:43:33] *** RealWickedWicky is now known as WickedWicky
[19:43:38] * Mdx4 raises hand too
[19:43:47] <WickedWicky> ciao
[19:43:48] <WickedWicky> :)
[19:43:59] <vega|> End user rarely realized backbone of most bigger computer systems are Sun boxes
[19:44:13] <wesolows> It's about as global as it gets for an English-speaking channel.
[19:44:31] <vega|> I got to know about Solaris OS hardly 1 year back
[19:44:34] <vega|> Ha .
[19:44:46] *** gdamore has joined #opensolaris
[19:45:32] <vega|> Sun started throwing out free Solaris 10 DVD's....it was a good step
[19:45:41] <vega|> UBUNTU is a big joke !!!
[19:45:58] *** smesjz has quit IRC
[19:46:04] <Mdx4> Ubuntu is a good product for a desktop.
[19:46:32] *** catena has quit IRC
[19:46:44] <sahafeez> sorry but as far as linux goes ubuntu is very well done with a great installer and is starting to reach the it just works stage. as a desktop in the unix world is better then every out there expect os x.
[19:47:01] <vega|> I dont no why they broke KDE and GNOME in to UBUNTU & KUBUNTU edition
[19:47:21] <vega|> I still prefer OpenSuSe
[19:47:41] <WickedWicky> as of this morning I hate Linux, at least for today and the comming week
[19:47:49] <Pietro_S> vega|: because they wanted to have *1* installed CD
[19:47:52] <vega|> ba hahahaha
[19:48:06] <Mdx4> SUN has loose a lot of time with CDE.
[19:48:12] <vega|> Hmm ....
[19:48:16] <vega|> Yepzzz...CDE
[19:48:23] <WickedWicky> whoever made /bin/sh be /bin/bash and make bash v3 at some points incompatible with bash v2 should really suffer
[19:48:33] <vega|> When I run CDE in Windows using Hummingbird
[19:48:38] <vega|> It runs too slow
[19:48:41] <WickedWicky> so should the 'engineer' who did stupid programing though
[19:48:45] <vega|> Better to use PuTTy or etc
[19:48:46] <Mdx4> and now they are gettin' too much complicated Solaris..
[19:48:56] <WickedWicky> I loved Exceed
[19:48:59] <Mdx4> beginning from the update manager..
[19:49:07] <Mdx4> with a lot of dependencies in java.
[19:49:22] <vega|> Sure ...
[19:49:31] <vega|> .NET seems to have cloudy days..
[19:49:45] *** yarihm has joined #OpenSolaris
[19:49:45] <Mdx4> that has shown to be defective to detect the hardware where is working on.
[19:50:00] *** QwaZ has joined #opensolaris
[19:51:03] <vega|> 1 disappointing thing in Sun hardware is PCI/ PCI express card compatible with SPARC server is over expensive. Price never looks suitable. Nobody buys a graphic card for Sun server but...It doesnt mean that they should charge 600 US$ just for the card..
[19:51:42] <wesolows> SPARC, expensive? No...
[19:51:44] <wesolows> say it ain't so!
[19:51:57] <Mdx4> now is no more expensive..
[19:52:29] <Mdx4> but when i've seen Microsoft logo on sun.com i've thinked: "ALL YOUR BASE ARE BELONG TO US"
[19:52:34] <WickedWicky> I'm hungry, going for a taco
[19:52:36] <QwaZ> maybe not the server it self -- but graphics cards are
[19:52:44] <vega|> haha
[19:53:02] <vega|> Agreed QwaZ
[19:53:13] <Mdx4> about smc ...
[19:53:44] <vega|> SMC ?
[19:54:21] <Mdx4> too much complicated to customize and no helper programs.
[19:54:45] <QwaZ> it's a pitty you cannot use PC's PCI cards -- I guess because of the firmware
[19:55:08] <wesolows> You can, in some cases.
[19:55:14] <Mdx4> QwaZ: the same thing that Apple has done in the past.
[19:55:25] <vega|> I affraid if the server gets blown up ...
[19:55:26] <wesolows> Just can't boot from them, and if they rely on BIOS boot code to work, that obviously won't happen.
[19:55:51] <vega|> Sun takes no responsibility ...
[19:56:16] <vega|> Otherwise I would go for ATi radeon graphic card :D
[19:56:27] <QwaZ> ...I tried to use video card in my sun blade -- I even compiled XOrg for the Sun -- but it didn't work
[19:56:32] <vega|> I think there are VESA drivers included in Solaris 10
[19:56:48] <vega|> whoooops ...!! Interesting ...
[19:56:50] *** MattAFC has quit IRC
[19:57:10] <wesolows> VGA cards generally rely on BIOS ROMs to be executed in order to work.
[19:57:22] <QwaZ> I guess that was the problem
[19:57:26] <wesolows> So they don't function in non-OC machines
[19:57:28] <wesolows> PC
[19:57:41] <wesolows> NICs, HBAs, etc, those generally work fine
[19:57:46] <wesolows> though you won't be able to boot from them
[19:58:20] *** Shiv_1 has joined #opensolaris
[19:58:32] <vega|> They dont give fibre channel card right ?
[19:58:42] <vega|> Any1 using Fibre channel card ?
[19:58:48] <vega|> For ur Sun server ?
[19:58:55] <wesolows> We sell several
[19:58:58] <Mdx4> vega|: i but not on SPARC machines.
[19:59:11] <wesolows> Some of which no doubt are supported on SPARC.
[19:59:27] <vega|> We r having HP stoarage library connected to a XTREME gig switch ...We bought separatr Fibre chanel card
[19:59:36] <Mdx4> i've normal Qlogic card on AMD ones.
[19:59:53] <vega|> Cool
[20:00:11] <vega|> Fibre channel cables are still too expensive
[20:00:15] *** bengtf has joined #opensolaris
[20:00:22] <vega|> Just OK within server room
[20:00:33] <vega|> Not beyond that
[20:00:43] <Mdx4> surelly isn't that the problem; costs a lot more the storage ;^)
[20:01:01] <vega|> How Google works...I always wonder
[20:01:15] <vega|> Do they have big Sun black box shown 2 months back ?
[20:01:26] <vega|> Just like a big shipping container
[20:01:36] <wesolows> it *is* a shipping container, actually
[20:01:45] <vega|> :)
[20:01:58] *** bondolo has quit IRC
[20:03:07] *** phips|mb has quit IRC
[20:03:08] <vega|> Any1 ever saw Sun Microsystems manufactured laptop ? Hehe
[20:03:27] *** gdamore has quit IRC
[20:03:36] <Mdx4> vega|: there are some SPARC laptops.
[20:03:45] <vega|> wooo.>Really ?
[20:03:52] <vega|> Wow...I never heard
[20:03:56] <vega|> May be only in US
[20:04:13] <Mdx4> SPARCLE
[20:04:46] <vega|> Many things remains within United States..never gets out ...Like Mac ...Or Juniper in networking.....Big players
[20:04:49] <Mdx4> look at http://www.tadpolecomputer.com/
[20:05:04] <Pietro_S> tadpole made some of them
[20:06:10] <vega|> Looking neat
[20:06:30] <vega|> Thin Client was a cocept introduced by Sun and theft by others ?
[20:07:16] * the-decider thinks a tadpole with a T1 would be crazy fun
[20:07:20] <Mdx4> a complete failure of the 90ies. like the choice to abandon solaris x86
[20:09:34] *** phips|mb has joined #opensolaris
[20:09:35] <vega|> i wonder if I am the only person here running Windows right now ....
[20:09:41] <vega|> :P
[20:10:03] <Mdx4> i'm running OSX
[20:10:12] *** Dar is now known as Dar_HOME
[20:10:13] <delewis> nope, I'm rdesktop'd into a Windows 2003 server at the moment.
[20:10:15] <vega|> So no worries about viruses
[20:10:27] <vega|> I am waiting for Windows 2008
[20:10:34] <the-decider> ooh -- macrumors says leopard on 10/26
[20:10:35] <vega|> They added NAC ...pretty late
[20:10:58] <Mdx4> the-decider: i wait for a merge with Solaris ;)
[20:10:58] <vega|> Leopard had so many rumours from underground world I guess
[20:11:01] <Mdx4> dtrace and zfs are there..
[20:11:15] <the-decider> solaris kernel, osx userland.  it'd be wonderful.
[20:11:37] <moazamraja> it's not a solaris kernel
[20:11:55] *** hile_ has joined #opensolaris
[20:11:57] <the-decider> I know -- but what I'm suggesting would be interesting.
[20:12:17] <vega|> Mac OS is easier than Solaris
[20:12:20] <vega|> :)
[20:12:36] <vega|> from the end user prospective ..
[20:12:40] <holcomb> depends on what you want to do
[20:12:47] * holcomb quickly re-exits the conversation
[20:12:50] <wesolows> Respectfully disagree, but it's always a matter of personal opinion
[20:13:01] <axisys> is there a way to increase the limit of searc list in resolv.conf? according to man resolv.conf it is six
[20:13:16] * wesolows . o O ( one mouse button?  WTF am I supposed to do with that? )
[20:13:16] <vega|> I know 1 company who own a big mail server of Apple.....How weird ...
[20:13:40] <vega|> hahahaha
[20:13:42] <bda> wesolows: Mac mice have two buttons. And you could always get your own damn mouse.
[20:13:46] <vega|> Throw the mouse
[20:13:50] <wesolows> bda: yeah yeah
[20:13:58] <bda> And you've always had context menus.
[20:13:59] <vega|> Use the tiny mechanical keyboard provided by Sun Microsystems
[20:14:06] *** _mw46_ has joined #opensolaris
[20:14:13] <bda> Which is not to say that Apple mice have not previously been massively jacktarded pieces of idiocy.
[20:14:25] <wesolows> vega|: eh?  The only Sun-branded keyboards I've seen are normal size.  Which product was this?
[20:14:28] <bda> Or that OS X is in any way a pinnacle of software engineering.
[20:14:29] <vega|> hahahahahahaaha
[20:14:38] <vega|> USB
[20:14:45] <vega|> Sun keyboard
[20:14:58] <wesolows> I'm using a USB Sun-brnaded keyboard at this moment; it's normal.
[20:14:58] <vega|> We got with Sun V230 I guess
[20:15:04] <vega|> Very old folk in our company
[20:15:23] <wesolows> Type6 and type7 are normal-sized keyboards...maybe there's a differently-localised one sold in India that's tiny.
[20:15:34] <vega|> We got Sun compliers for free on that server
[20:15:37] <vega|> For free
[20:15:46] <the-decider> ...they are free.
[20:15:47] <vega|> Hehe
[20:15:55] <vega|> They aint ??????
[20:15:56] <wesolows> Sun distributes Studio 11 at no cost to anyone
[20:16:00] <vega|> Where can I get them ?
[20:16:06] <vega|> Oh
[20:16:08] <Pietro_S> same with studio 12
[20:16:15] *** peemus has joined #opensolaris
[20:16:21] <vega|> I had talk with a linux developer today evening
[20:16:34] <wesolows> Pietro_S: Keep forgetting that it's out (since it doesn't actually work for compiling anything I care about :))
[20:16:54] <vega|> I told him Solaris 10  8/07 edition available with me...He said what about compilers ? He want compilers...
[20:16:59] *** mikefut_ has quit IRC
[20:17:08] <wesolows> Solaris comes with gcc in /usr/sfw/bin by default.
[20:17:11] <wesolows> Works great.
[20:17:16] <wesolows> If you want Studio, no cost.
[20:17:28] <wesolows> And it's in SXDE now as well, on the same media.
[20:17:35] <vega|> So compilers are included in Sun Studio ... ?
[20:17:49] <wesolows> The compilers *are* studio, unless I'm horribly confused.
[20:18:10] <wesolows> if there's other stuff in there I have no idea what it is; we don't deliver TeamWare any more, sadly.
[20:18:10] <the-decider> Sun Studio is more... so much more.  More than you can imagine ;)
[20:18:10] <vega|> I never used it.
[20:18:22] <wesolows> It has lint, and I think dbx.
[20:18:24] <vega|> Hmm....
[20:18:35] <vega|> I prefer BigAdmin newsletter rather...
[20:18:36] <wesolows> and a few C++ libraries you need to build.
[20:18:52] <vega|> yepz...Thats what he was asking me for
[20:18:59] <vega|> C ++ libs for Sol 10
[20:19:14] <vega|> pearl etc
[20:19:21] <wesolows> well, libCrun comes with Solaris 10 by default.
[20:19:25] <wesolows> You are confusing me.
[20:19:27] <wesolows> I go now.
[20:19:41] <Shiv_1> wesolows: SS bundle has cscope, tcov, gvim, xemacs, ....
[20:19:43] <vega|> Oh
[20:19:53] <vega|> Get some waffers and Coke
[20:19:55] <vega|> Hehe
[20:20:04] <wesolows> yes, coke, that's what I need
[20:20:24] <vega|> Waffers which has trade mark of Sun Microsystems
[20:20:29] <vega|> ba hahahahahaahahahahaha
[20:20:38] <wesolows> perhaps a speedball
[20:20:57] <the-decider> or a spaceball
[20:21:18] <vega|> You wanna go for NASA's next expidition ?
[20:21:24] <vega|> :P
[20:22:42] *** Gropi has joined #opensolaris
[20:23:35] <vega|> Its 11:53 PM in India
[20:23:56] <vega|> I m on 10 kbps GPRS link....EDGE support ..Thank god ...
[20:24:00] <Mdx4> vega|: for the spatial expedition try the LEON Sparc processor.
[20:24:15] <vega|> :)
[20:25:06] <vega|> Does this also covers my insurance ... ? Haha
[20:25:20] <vega|> Life insurance ...
[20:27:19] <Mdx4> i'm the only or the sun.com site has problems ?
[20:27:40] <vega|> No...I m browsing without any prob
[20:27:47] <vega|> even opensolaris.org
[20:28:12] <vega|> on 10 kbps connection ...
[20:28:29] <vega|> Ur SCSA ?
[20:28:34] <Mdx4> sun has problems with a link.
[20:28:44] <vega|> which link ?
[20:29:03] *** hyegeek has left #opensolaris
[20:29:20] <Mdx4> i arrive correctly to sanfrancisco then my packets goes to /dev/null
[20:29:34] <wesolows> yes, I'm watching them right now
[20:29:50] <wesolows> they come in over the bay bridge and flow stright into a giant sinkhole near embarcadero station
[20:29:52] <vega|> Spy !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! hehehehehehe
[20:29:55] <wesolows> there are a LOT of them!
[20:30:03] <vega|> Where is my camera ... ?
[20:30:05] <wesolows> and the sinkhole is filling up rapidly
[20:30:26] <Mdx4> wesolows: i'll saturate the BW of /dev/null :)
[20:30:27] <peemus> just redirect them to the bay
[20:30:46] <_mw46_> do we have a libsocket expert here?
[20:30:51] <vega|> Call init 5
[20:30:54] <wesolows> peemus: they're overflowing into the bay now; the SFFD is trying to set up a hazardous material containment zone
[20:30:58] <wesolows> but it looks hopeless
[20:31:06] <wesolows> I'm leaving for higher ground now
[20:31:32] <_mw46_> I've the following stack trace from a postfix smtpd : http://pastebin.ca/725790
[20:31:54] <vega|> any1 have spair sun server ?
[20:32:05] <vega|> Where I can do some practice ?
[20:32:07] <vega|> Hehe
[20:32:20] <vega|> All servers in my company are production servers
[20:32:24] <vega|> Cant play on it
[20:32:27] <_mw46_> it looks like getnameinfo() has a problem in cases when it does not find a hostname.
[20:33:10] <wesolows> this reminds me of a bugfix I just saw go by
[20:33:46] *** QwaZ has quit IRC
[20:34:11] <vega|> ZFS users pls clap ur hands ...  :)
[20:34:20] <vega|> I have some questions
[20:34:23] <vega|> About zones
[20:34:55] <wesolows> maybe 6609869?
[20:35:30] <vega|> waz tat ?
[20:35:35] <_mw46_> wesolows: are you talking to me?
[20:35:39] <wesolows> _mw46_: yes
[20:36:02] <_mw46_> do you have a ling to your bug DB handy?
[20:36:34] <_mw46_> I think I know where the code is kind of questionable.
[20:37:05] <_mw46_> http://cvs.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/lib/libsocket/inet/getnameinfo.c around line 150.
[20:37:53] <wesolows> bogs.opensolaris.org
[20:37:56] <wesolows> err bugs
[20:38:01] <wesolows> same place as always
[20:38:27] <wesolows> the fix went into usr/src/lib/nsswitch/dns/common/dns_common.c though
[20:38:32] <wesolows> it may not be the same bug
[20:39:57] *** gdamore has joined #opensolaris
[20:42:18] <_mw46_> it looks a little bit different.
[20:43:01] <_mw46_> But I don't know (yet?) what the smtpd stumpled over. But it looks like the strlen() was called on NULL:-(
[20:43:25] <vega|> Back
[20:43:34] <vega|> O am falling asleep
[20:43:36] <_mw46_> I'll go back to my postfix log files and see if I can find one.
[20:43:41] <vega|> 12:00 AM
[20:43:43] *** millhouse has joined #opensolaris
[20:43:44] <vega|> See ya
[20:43:49] <millhouse> hey
[20:43:58] <vega|> ciao
[20:46:39] *** vega| has quit IRC
[20:50:52] <elektronkind> hehe oh joy
[20:51:06] <elektronkind> I'm using CAM to manage a IBM DS4200 ;)
[20:52:53] *** blindfish has quit IRC
[20:52:54] *** millhouse has quit IRC
[20:53:17] *** mikefut has joined #opensolaris
[20:56:31] *** dmesg has joined #opensolaris
[20:56:45] <dmesg> hi dudes,
[20:57:25] <dmesg> does srss 3.x can run on a x86 solris 10?
[20:57:42] *** phips|mb has quit IRC
[20:58:10] <agliodbs> feh
[20:58:12] *** phips|mb has joined #opensolaris
[20:58:23] <agliodbs> not going to be able to go to the OSOL summit after all
[20:59:19] <dmesg> so it wont work?
[21:00:17] *** mikefut has quit IRC
[21:00:25] <agliodbs> dmesg: srss?
[21:00:34] <_mw46_> wesolows: the IP will the problem is :
[21:00:39] <_mw46_> 8.44.159.67.in-addr.arpa        name = .
[21:00:52] <dmesg> agliodbs sun ray server software
[21:01:10] <_mw46_> that apparently kills the code in getnameinfo.
[21:01:13] <agliodbs> oh, beats me
[21:01:18] *** cmihai has joined #OpenSolaris
[21:01:19] <agliodbs> Gman == Glynn?
[21:01:39] <_mw46_> so how do I file a bug against opensolaris today?
[21:02:07] *** hile_ has quit IRC
[21:02:40] <RElling> http://bugs.opensolaris.org/
[21:02:46] <e^ipi> _mw46_: bugs.opensolaris.org
[21:03:21] <_mw46_> I can look at bugs there, but how do I file one?
[21:03:31] <palowoda> http://www.opensolaris.org/bug/report.jspa
[21:03:32] <e^ipi> unless it's a security bug, in which case fulldisclosure at insecurity dot org
[21:03:39] <the-decider> 175 * 4
[21:03:44] <the-decider> oops
[21:03:48] <e^ipi> it gets fixed much faster if you send security bugs there
[21:03:59] *** sleepcat has joined #opensolaris
[21:04:38] <cmihai> e^ipi, and even faster if you send them to me.
[21:04:46] <CIA-26> ky115808: 6612154 PCI slot NAC names should be corrected in fmtopo -V output
[21:05:14] <cmihai> Don't bother unless it's 0day / remote system access :P
[21:07:56] <e^ipi> when b.o.o stops dissapearing (in the KGB sense) security bugs, I'll stop suggesting that people route them to fulldisclosure
[21:10:17] <stevel> agliodbs: yeah Gman == Glynn
[21:11:16] <WickedWicky> e^ipi: there are no bugs, there are no people againt the regime, and if they were, they "changed their mind"
[21:11:29] <WickedWicky> I thought the KGB stopped existing long ago btw :P
[21:11:40] <stevel> they morphed into the FSB i thought
[21:12:32] <e^ipi> WickedWicky: putin seems to think otherwise it seems
[21:12:41] <agliodbs> wish the summit wasn't on a weekend
[21:13:13] <e^ipi> agliodbs: because then people wouldn't be able to come?
[21:13:36] <WickedWicky> ok, OT question.. what would a T2000 generaly be used for? as in, where would you deploy one?
[21:13:52] <e^ipi> WickedWicky: database, web
[21:13:52] <cmihai> Iraq
[21:13:55] <e^ipi> transaction processing
[21:14:00] <WickedWicky> something tells me it's a bit over kill for an apache webserver
[21:14:01] <cmihai> It makes a poor db server methinks
[21:14:13] <e^ipi> depends on the workload I'd assume
[21:14:22] <sleepcat> a T2000 is used to generate revenue for SUN Microsystems Inc.
[21:14:34] <sleepcat> ticker symbol JAVA.
[21:14:46] <WickedWicky> should i see these coolthreads as more or less the hyperthreading technology in Xeons? or is it something totaly different?
[21:14:47] <e^ipi> then it does a piss poor job of it because they're selling the hell out of amd64 machines
[21:15:18] <sleepcat> e^ipi: I'd get the amd64 machines, once the quad cores come out
[21:15:31] <WickedWicky> they have quad core xeons now
[21:15:33] <WickedWicky> they being sun
[21:16:01] <sleepcat> i'll wait for amd, they consume less power
[21:16:16] <cmihai> Meh
[21:16:29] <cmihai> What's this whole think green bullshit about in terms of servers?
[21:16:51] <sleepcat> think green as in save money on my electric bill?
[21:16:55] <cmihai> I really doubt there's THAT big a difference in TCO / costs when it comes to CPU power usage for your servers.
[21:17:15] <elektronkind> you wanna bet?
[21:17:26] <elektronkind> we
[21:17:35] <elektronkind> we're looking at two competing systems
[21:18:03] <elektronkind> system 1: a IBM p 570 with 4x 2 core POWER6 CPUs
[21:18:20] <elektronkind> system 2: a Sun T2000 (or 5220)
[21:18:26] <cmihai> I think the new quad core Xeons are 60-80W
[21:18:37] <cmihai> elektronkind, competing?
[21:18:40] <cmihai> Please
[21:18:47] *** Shiv_1 has quit IRC
[21:18:49] <elektronkind> system 1's power draw is 1400W
[21:18:54] <cmihai> Sun T2000 is way out of IBM's Power 6's league
[21:19:12] *** virgee has joined #opensolaris
[21:19:16] <elektronkind> system 2: ~640W
[21:19:19] <Mdx4> elektronkind: sorry but the Power6 processors aren't partitionables :) and almost dead :)
[21:19:54] <cmihai> Dead?
[21:19:56] <elektronkind> Mdx4: my CIO thinks wants us to look at it, so we are. We know what our answer will be as I'm sure you would as well, but we're humoring him.
[21:20:02] <cmihai> You mean the 4.7 Ghz Power 6 released a few months ago?
[21:20:41] <RElling> elektronkind: can you wait until next week?
[21:21:13] <elektronkind> for many data centers, including ours, AC capacity is near full. Gone are the days when we can toss a fridge-sized, dual 208VAC monster in there without thinking about BTU
[21:21:14] <Mdx4> cmihai: they 'll reach 6Ghz and then ?
[21:21:20] <the-decider> elektronkind: you forgot PA/RISC!  Shouldn't we hae eerything on the table?
[21:21:36] <cmihai> elektronkind, I'm willing to bet a single CPU Power 6 system will outperform T2000 as say.. an Oracle DB server
[21:21:37] <elektronkind> RElling: I'd love nothing more than to get on the Niagara2 bandwagon
[21:21:57] <Mdx4> cmihai: false for OLTP :)
[21:22:12] <elektronkind> cmihai: yes, but not enough to justify its consumption requirements
[21:22:21] <the-decider> cmihai: outperform, as in what?  Cooking capacity?
[21:22:24] <the-decider> ;)
[21:22:40] <Mdx4> the-decider: he is right for the DSS workload.
[21:23:01] <Mdx4> a JOIN is CPU intensive and a higher clock does the job quicker.
[21:23:02] <elektronkind> RElling: but it'll take some intense engagement from our Sun sales team to convinve the CIO to go that route.
[21:23:11] <the-decider> mhz has its place.
[21:23:14] <elektronkind> er convince
[21:23:19] *** virgee has quit IRC
[21:23:52] <Mdx4> the-decider: but is also true that on real DDS begin to take place the PARALLEL in many queries.
[21:24:00] <elektronkind> Intel realized the clock speed game was bunk and existed with it put the last nail in to the Netburst coffin. IBM with its POWER chips is still stubbornly towing that line.
[21:24:05] <Mdx4> on the OLTP more thread is better.
[21:24:15] * the-decider is pro-sun for this usage...
[21:24:53] <e^ipi> thank god intel killed off that abortion, by the way...
[21:25:23] <RElling> T2000 already beats Power6 on several benchmarks, N2 will make P6 look silly
[21:25:32] <cmihai> Yeah, right.
[21:25:48] * the-decider puts on his CIO's hat...
[21:25:53] <sleepcat> Mdx4: OLTP more thread is better IF the RDBMS uses threads...postgresql is process based...
[21:26:00] <elektronkind> RElling: honestly, what we need is for a person like you and BMSeer to come over and sit down with our CIO
[21:26:00] <the-decider> "but, Sun is going out of business..."
[21:26:04] *** deather_ is now known as deather
[21:26:37] <Mdx4> sleepcat: is the same, i talk about "CPU Threads" :) in postgresql you can have more processes running at the same time.
[21:26:43] <elektronkind> unfortunately "coming over" means Baltimore, and I don't blame you if you want to avoid baltimoe ;)
[21:26:49] <RElling> elektronkind: I'd love to, but speeds-and-feeds rarely work with CIOs, so the conversation would be very different than here
[21:27:21] <RElling> I've been to Baltimore several times, it has its charms :-)
[21:28:00] <Mdx4> sleepcat: oracle also runs basically on processes and not thread.
[21:28:03] <the-decider> RElling: he tries to be rather technical, you see.
[21:28:09] *** yarihm has quit IRC
[21:28:31] <Mdx4> sleepcat: don't talk me about MTS because is only a multiplexer.
[21:28:39] <the-decider> RElling: he wanted a command by command comparison of "how" AIX was harder to administer than Solaris
[21:29:19] <elektronkind> IBM sauntered in and poo-poo'd the T2000 of course. This is for a Oracle/Peoplesoft server. P6 vs. T2000. They harped on fact that US-T1 has a shared FPU. So our CIO says to us after that meeting "But the T2000 has bad floating point performance!" and we ask him if he thinks that's relevant to Oracle.
[21:29:29] <Mdx4> the-decider: AIX is a unix of third choice (Solaris, HP-UX, AIX).
[21:29:30] *** Gman has joined #opensolaris
[21:29:42] <RElling> I'd sick John Meyer, http://blogs.sun.com/jmeyer on him, John lives near there
[21:30:08] <Mdx4> elektronkind: for ORACLE FPU is almost irrilevant if you don't use the internal JVM and/or SPATIAL :)
[21:30:20] <the-decider> the-decider: AIX isn't unix, it's MVS in a unix costume, under an unrecognizable pile of IBM poop
[21:30:44] <the-decider> and PeopleSoft doesn't ;)
[21:30:47] <Mdx4> the-decider: MVS isn't bad as AIX :)
[21:30:50] <Stric> it has some good stuff and it has some odd stuff..
[21:30:52] <elektronkind> Mdx4: well yeah, that's a big "duh" to you and me. The CIO took IBM's word for it though. This is what the-decider and I are fighting against ;)
[21:30:53] <sleepcat> elektronkind: how good is it with NUMERIC datatypes ?
[21:31:00] <RElling> Tuesdays are the preferred new product announcement days at Sun :-)
[21:31:16] * the-decider will get him drunk and yell at him on Monday
[21:31:20] <cmihai> Funny, same day Microsoft announces bug fixes.
[21:31:39] <Mdx4> sleepcat: in oracle aren't translated to Floating point (floating point looses decimals).
[21:31:54] <elektronkind> it's fixed point
[21:32:23] <cmihai> Yeah, you want to have some fun with that, try something like 2.0 -1.1 in Java :-)
[21:32:27] <sleepcat> Mdx4: i understand, does the T2000 have good performance with NUMERIC datatypes as found in postgresql?
[21:32:30] <cmihai> 0.8999999999999999 :-\
[21:32:31] <Mdx4> elektronkind: in oracle you can do a LIKE over a NUMBER field try to imagine how are managed.
[21:32:55] <elektronkind> Mdx4: I leave that up to the DBAs to worry about ;)
[21:32:56] *** alanc-away is now known as alanc
[21:33:12] <Mdx4> sleepcat: halt, in postgresql there is a FLOAT field that is a treated as FLOAT.
[21:33:32] <Mdx4> sleepcat: but you know if you are doing scientific calculus over your machine.
[21:33:53] <sleepcat> float is good for statistical OLAP, but a lot of times we use NUMERIC.
[21:34:20] <sleepcat> float is very bad if you are doing anything with money.....
[21:34:28] <Mdx4> sleepcat: indeeed :)
[21:34:41] <the-decider> the db backend is for peoplesoft.  It makes the app-servers do all the lifting anyhow
[21:34:47] <Mdx4> sleepcat: but for a discrimination you haven't to involve the FPU :)
[21:34:55] *** bengtf_ has joined #opensolaris
[21:35:09] *** slowhog has quit IRC
[21:35:38] <Mdx4> sleepcat: the meaning is if you use an RDBMS as a DATABASE ok, if you have choosen an RDBMS like a sobstitute for Mathematica you know what you get.
[21:35:47] <Mdx4> s/sob/sub/g
[21:36:31] <sleepcat> i put all the business logic in the database using stored procedures and functions
[21:36:42] *** sebroy has joined #opensolaris
[21:36:44] <e^ipi> you can evidently impliment lambda calculus using nothing but C++ template matching rules, so you can make it turing complete...
[21:36:49] <Mdx4> sleepcat: so better a higher clock CPU :)
[21:36:57] <e^ipi> sometimes you should think about using the correct tool for the job though
[21:37:20] <the-decider> I still haven't figured out if sendmailrewrite rules are turing complete ;)
[21:37:22] *** millhouse has joined #opensolaris
[21:37:34] <sleepcat> e^ipi: Tom Kyte from oracle and all the others recommend doing it that way.
[21:37:54] <e^ipi> sleepcat: what? implementing things with C++ templates?
[21:37:59] <sleepcat> no
[21:38:11] <e^ipi> good, that'd be a damnedfool idea
[21:38:12] <the-decider> implemeting c++ template with sendmail rules ;)
[21:38:12] <sleepcat> herb sutter abuse of C++
[21:38:31] <Mdx4> please let us sendmail rest in peace :)
[21:38:37] <Mdx4> had done its epoch :)
[21:38:41] <the-decider> it lives man
[21:38:48] <sleepcat> i still have to get CUPS to work....
[21:38:49] <the-decider> has achieved sentience
[21:38:52] <Stric> no, it's a zombie that's haunting people :)
[21:38:58] <Mdx4> the-decider: better say SURVIVES.
[21:39:06] <the-decider> or solving their problems ;)
[21:39:15] <e^ipi> sleepcat: rewrite it in the C++ template version of sendmail
[21:39:25] <elektronkind> I walked in to the-decider's office one day while he was working on a new sendmail rule and I asked him why one of his terminal windows was full of comic-style cuss words
[21:39:44] <WickedWicky> lol
[21:39:52] <WickedWicky> *&#&(*&#@)(*$#(*&$#^&#@*&#@)($#@)(*@)(*$#
[21:39:53] *** KermitTheFragger has quit IRC
[21:39:53] <Mdx4> elektronkind: a file full of cat's walk on the keyboard :^)..
[21:40:20] <elektronkind> "what rob, catting binaries again?"
[21:40:31] *** sleepcat has left #opensolaris
[21:41:06] <elektronkind> in fact, I bet you could cat a binary file and cut+paste the resulting garbase on your screen into sendmail.cf, and it would work
[21:41:12] <elektronkind> er garbage
[21:41:25] *** alobbs has quit IRC
[21:41:49] <Stric> elektronkind: then you can play the game, "where will my mail end up today?"
[21:41:49] <Mdx4> the-decider: without any dubt sendmail does more than deliver mail, UUCP etc, but almost nowdays we have to deliver mails over the net.
[21:42:17] <Mdx4> m4 is for wimps ;)
[21:42:40] *** karrotx has joined #opensolaris
[21:43:55] <Mdx4> Stric: the problem is the sendmail's attitude to be heavvy, its binaries "size" and the difficult to setup it correctly :)
[21:44:52] <Mdx4> and the attitude of the mbox to corrupt easy :)
[21:44:53] <Stric> but then you have sendmail-x or whatever it's called, whose design looks quite similar to postfix
[21:45:35] <Mdx4> i can use sendmail today because is "well-tested", if i have today postfix why choose a postfix like ?
[21:45:52] <Stric> that.. I don't kow.
[21:45:53] <Stric> know.
[21:46:06] <Stric> because it has the "sendmail" name? :)
[21:46:30] * Stric replaced sendmail with postfix many many years ago
[21:46:34] *** Fullmoon has joined #opensolaris
[21:46:54] <Mdx4> i can't understand why solaris doesn't let the user to choose if use postfix or sendmail :)
[21:47:12] <elektronkind> because postfix isn't part of it
[21:47:20] <Mdx4> neither sendmail.
[21:47:24] <e^ipi> sendmail is
[21:47:31] <e^ipi> svcs -a | grep sendmail
[21:47:36] <Mdx4> sendmail is a opensource like postfix
[21:47:38] <elektronkind> sendmail's there. SUNWsndm(r|u)
[21:47:51] <elektronkind> Mdx4: svcs -a | grep smtp
[21:48:03] <elektronkind> or you don't have it installed.
[21:48:19] * Mdx4 hates svcs :)
[21:48:23] <elektronkind> pkginfo -l SUNWsndmr
[21:48:26] <elektronkind> to verify
[21:49:05] <Mdx4> elektronkind: what i meant is that solaris isn't a software designed by SUN :)
[21:49:27] <elektronkind> you mean sendmail isn't software designed by Sun
[21:49:38] <Mdx4> sure..
[21:49:43] <elektronkind> that's right, but Sun people contribute to it
[21:49:55] <elektronkind> just like any other open source project out there
[21:50:08] <e^ipi> there's lots of software in solaris not made by sun
[21:51:04] *** devel1 has joined #opensolaris
[21:51:08] <devel1> Hello
[21:51:24] <elektronkind> welcome to my channel. I kiss you!
[21:51:47] <devel1> :)
[21:51:58] <Mdx4> e^ipi: they could let us choice between another MTA, i'm not asking too much, i prefer choice what MTA use than which desktop use.
[21:52:02] <elektronkind> I guess everyone has forgotten Mahir
[21:52:28] <elektronkind> Mdx4: nothing prevents you from installing Postfix and using that
[21:52:37] <elektronkind> no one has to "let" you do that
[21:52:55] <Mdx4> elektronkind: i've to fight with "mail" and a lot of other binaries :)
[21:53:43] <elektronkind> I think what you're looking for is that in Solaris, both Sendmail and Postfix are installed as a part of the OS, and the admin can choose which one to use
[21:54:20] <Mdx4> elektronkind: perfect :)
[21:54:29] <elektronkind> right now only Sendmail is included, and if you want something else like Postfix, then you have to just install that yourself
[21:54:53] <Mdx4> but i get two desktops :^)
[21:55:00] <Mdx4> <grin>
[21:55:01] <e^ipi> not for long
[21:55:18] <e^ipi> CDE's going to be EOSL'd soon iirc
[21:55:29] <elektronkind> it's already SOL
[21:57:58] *** phips|mb has quit IRC
[21:58:54] <Mdx4> anyway i agree on the choice to use bash as default shell for users.
[21:59:11] <Mdx4> and is a step ahead.
[22:01:46] <e^ipi> what?
[22:02:00] <e^ipi> you haven't used ksh93 lately, have you?
[22:02:14] <e^ipi> people don't want bash per se, they want tab complete, and ksh93 has that
[22:02:28] <e^ipi> it is also standards compliant, whereas bash can't even comply with itself
[22:03:20] <RElling> Mdx4: IMHO, no MTA on a desktop is the best
[22:03:20] <sommerfeld> zsh!
[22:04:11] <e^ipi> linux distros use bash because that's the way it's always been... we have a choice though, and so we can choose what's actually better
[22:05:59] <alanc> linux distros use bash because neither bourne nor korn shells were open source when the first distros came out, so bash was the only choice for a bourne/korn-like shell
[22:06:02] *** dmesg has quit IRC
[22:06:04] <Mdx4> RElling: i meant, i can choose the DESKTOP to use, for a server like solaris i would have the possibility to choice about the MTA instead.
[22:07:15] *** phips|mb has joined #opensolaris
[22:07:15] <Mdx4> e^ipi: i use HP-UX daily i know ksh :)
[22:08:00] <Mdx4> e^ipi: the old default shell on solaris was the plain SH :) the most SHish shell with frills is bash.
[22:08:43] *** slowhog has joined #opensolaris
[22:08:51] <e^ipi> say that in front of roland
[22:09:08] <Mdx4> mcdonalds ? ;^^)
[22:09:22] <e^ipi> roland, not ronald
[22:09:27] <e^ipi> roland mainz
[22:09:43] *** karrotx has quit IRC
[22:09:44] <RElling> Mdx4: I think I agree... traditionally Sun has gone to great lengths to hide the dozen or so desktops they ship on a Solaris distro
[22:10:26] *** sleepcat has joined #opensolaris
[22:12:11] *** Murmuria has joined #opensolaris
[22:12:49] <moazamraja> is there a dtrace enabled memcached for solaris ?
[22:12:55] <moazamraja> i.e., with some extra dtrace hooks in it
[22:23:15] <quasi> moazamraja: I didn't see any dtrace mentioned when it was discussed as part of the webstack
[22:23:58] *** tinman2k has quit IRC
[22:25:04] *** Drone has quit IRC
[22:26:40] <alanc> they didn't mention dtrace in memcached in their ARC review
[22:28:59] *** millhouse has quit IRC
[22:29:18] <moazamraja> webstack?
[22:29:27] <moazamraja> must've missed out on that discussion
[22:29:29] <quasi> moazamraja: webstack-discuss
[22:31:08] <moazamraja> oh
[22:31:13] * moazamraja subscribes
[22:31:33] *** sleepcat has quit IRC
[22:33:31] *** noyb has joined #opensolaris
[22:34:48] <quasi> it's been very lively lately
[22:34:53] *** Triskelios has joined #opensolaris
[22:35:48] *** Murmuria has quit IRC
[22:37:39] *** Fish has quit IRC
[22:42:51] *** nostoi has quit IRC
[22:43:53] <Triskelios> which release of snv has gcc in /usr/gnu?
[22:45:34] *** theRealBallchalk has quit IRC
[22:45:41] <RElling> dunno, b72 has /usr/sfw/bin/gcc
[22:47:30] <peemus> saweet! my boss just handed me an iPhone - now how can I get solaris on it?
[22:47:32] <peemus> jk
[22:47:42] <the-decider> I CAN HAS SOLARIS?
[22:48:06] *** millhouse has joined #opensolaris
[22:48:09] <millhouse> hey
[22:50:06] <alanc> no gcc in /usr/gnu/bin on nv_74, guess it hasn't moved yet
[22:50:31] <wesolows> is it supposed to be there as cc?
[22:51:00] <wesolows> should be /usr/bin/gcc, /usr/gnu/bin/cc, no?
[22:51:04] <alanc> right, it would be /usr/bin/gcc, /usr/gnu/bin/cc
[22:51:23] <wesolows> sort of like /usr/ucb/cc, except it works :-)
[22:51:35] *** victori has quit IRC
[22:51:40] <alanc> but either way, my nv_74 still has it in /usr/sfw/bin/gcc
[22:51:43] <wesolows> sigh
[22:52:18] <alanc> there have been a lot of /usr/sfw -> /usr/gnu cases coming through ARC the last couple of weeks, so there's work going on there
[22:52:23] <wesolows> there are several other bugs in the way we build and deliver gcc too
[22:52:25] *** victori has joined #opensolaris
[22:52:35] <wesolows> hopefully whoever moves it will fix them too
[22:52:53] <RElling> just point out who did /usr/ucb/cc and I'll fetch my baseball bat..
[22:53:26] *** hile_ has joined #opensolaris
[22:53:27] *** phips|mb has quit IRC
[22:53:28] <alanc> hope so - would love to get rid of the workarounds I have in my makefiles for gcc to find it's MMX intrinsic header files, which aren't in it's default include path or anywhere logical
[22:54:02] *** Berny__ has joined #opensolaris
[22:54:10] <wesolows> and the library path stuff is broken I believe, for 64-bit
[22:54:46] <millhouse> how do I use GDM and not the default Solaris login?
[22:54:56] *** Berny_ has quit IRC
[22:55:10] *** Murmuria has joined #opensolaris
[22:55:21] *** Murmuria has left #opensolaris
[22:55:35] <alanc> svcadm disable cde-login ; svcadm enable gdm
[22:56:09] <alanc> it will kill all running desktop sessions instantly though
[22:56:42] <Triskelios> the services really should conflict with each other to prevent the reverse order from starting two X servers
[22:57:15] <alanc> does SMF allow you to define conflicts?
[22:57:29] <millhouse> alanc:  Thanks..  As it turns out, I found out how to do it right after i posted the question.  but thanks anyway :)
[22:57:37] *** FelipeBare has quit IRC
[22:57:58] <alanc> though I think in theory you could run both with different X servers, like gdm on console, dtlogin for your Sun Rays
[22:58:10] <millhouse> yeah, i accidentally enabled gdm2 login w/o first disabling cde login...  the login screen had cde on top and gdm behind it.  it was quite amusing, but not very useful :-)
[22:58:21] <wesolows> yes, SMF allows "exclude" dependencies
[22:58:46] <Triskelios> wesolows: thanks, couldn't remember what the directive was
[22:58:53] <wesolows> exclude_all I believe
[22:59:42] <Triskelios> millhouse: yeah, I think that happened to me after an upgrade
[23:00:16] <CIA-26> gd78059: 6317553 Wrong fix implemented in 4877168 for dmfe Rx buffer unavailable messages., 6380155 panic assertion failed: (dmfe_ring_get32(desc...sun4u/io/dmfe/dmfe_main.c line 1024, 6555168 do not deliver dmfe header files, 6592772 'ndd -get /dev/dmfe0 autoneg_cap' prints a value greater than the max allowed in dmfe's nd_params, 6603848 dmfe does not restart link when changing advertised parameters with ndd, 6603852 dmfe should support x86, 6603953 r
[23:00:54] *** sebroy has quit IRC
[23:05:53] <the-decider> what the hell
[23:05:55] *** tombhadAC has joined #opensolaris
[23:06:29] <the-decider> "catallaxy paradigm for decentralized operation of dynamic application networks"
[23:06:36] <the-decider> (www.catnets.org)
[23:06:44] <the-decider> that is a mouthful
[23:17:05] <millhouse> I've got vncserver running on a solaris system, but when i log into it, I get the basic twm desktop...  how can I get vnc to give me the gnome desktop instead?
[23:19:16] *** yarihm has joined #OpenSolaris
[23:20:51] <bengtf_> there is a line commented out in your created config file  millhouse
[23:21:02] *** Triskelios has quit IRC
[23:21:22] *** yarihm has quit IRC
[23:23:16] *** Gropi has quit IRC
[23:24:17] <bengtf_> millhouse: ~/.vnc/xstart file edit it
[23:28:32] <millhouse> bengtf_:  ah, thanks..
[23:28:49] *** cmihai has quit IRC
[23:30:25] <HCoyote> I keep seeing references to "txg" when dealing with ZFS.  What exactly is txg?
[23:31:14] <HCoyote> ahha.  nevermind.  google to the rescue.
[23:32:13] *** sleepcat has joined #opensolaris
[23:32:39] <sleepcat> which has better print support , SXDE or SXCE?
[23:35:18] <bengtf_> dont really know, but I got very surprised when my hp3200m worked perfectly ;) in sxce
[23:35:46] <sleepcat> how is sxce?
[23:35:51] <sleepcat> 73?
[23:35:52] *** dlg has joined #Opensolaris
[23:36:08] <e^ipi> sxde is just an older version of sxce
[23:36:40] <bengtf_> it works from b63 atleast so b73 should work
[23:37:03] <bengtf_> I run b63 as print server
[23:37:11] <sleepcat> i'm thinking of downloading sxce 73 in hopes of being able to print from my usb HP deskjet printer
[23:37:24] <sleepcat> otherwise I have to do this in windoze
[23:37:42] <bengtf_> which one
[23:37:51] <sleepcat> 3940
[23:38:27] <sleepcat> i spent two days trying to get it to work with CUPS
[23:39:45] <bengtf_> it doesnt recognize it in b63
[23:40:47] <sleepcat> ah typo 3930
[23:40:52] *** bigjohnto has joined #opensolaris
[23:40:53] <bigjohnto> ok, solaris workstation hangs, and does nothing, no mouse movement etc... i kept a tail on the messages log on another machine but nothing showed any last messages, and nothing showed on the monitor connected to the workstation.  What in the world could be causing the workstation to freeze?
[23:41:08] <bigjohnto> what tool can I use to figure out what the problem is?
[23:41:09] <sleepcat> bigjohnto: nvidia driver
[23:41:17] <bigjohnto> its a sparc
[23:41:53] <bengtf_> what version and what workstation
[23:42:12] <bigjohnto> hehe
[23:42:18] <bigjohnto> 8 and ultra5
[23:42:37] <alanc> if it's a SPARC and the mouse isn't moving, then your kernel is most likely hung, since mouse updates are done in the kernel
[23:42:44] <the-decider> can you stop-a it?
[23:43:14] <bigjohnto> well i restarted it, but next time i will try to stop-A it,
[23:43:15] *** Giaco is now known as SYS64738
[23:43:31] <bigjohnto> or do you want me to stopA it now?
[23:43:38] <the-decider> nah, when its frozen.
[23:43:43] <the-decider> by restarting, you power cycled?
[23:43:53] <bigjohnto> yes
[23:43:58] <e^ipi> well that's a silly place to put mouse updates
[23:44:36] <the-decider> if you can't stop-a it and get an ok> prompt when its hung like that, there's a greater chance of it being hardware
[23:45:05] <alanc> e^ipi: saved context switches when those were costly - the mouse input comes in from the device to the kernel driver, the update goes out to the frame buffer kernel driver - why go out to the X server and back in between the two if you don't have to
[23:45:27] <alanc> so a streams module is pushed onto the mouse device that sends updates directly to the fb driver
[23:46:31] *** Fullmoon has quit IRC
[23:46:35] <sleepcat> what is this stop-A stuff all about?
[23:46:43] <the-decider> its things like that which made "real" UNIX workstations (such as SGIs & Suns) feel like they had better interactive response then similar hung (speed-wise) linux boxen back in the day.
[23:46:46] <_mw46_> wesolows: So I filed a bug, the web site said it was filed. But I did not get an email and I can't find it when I try to search for it?
[23:47:06] <alanc> that stream module is the one bit of kernel code in the X consolidation - been wishing for years it would move to ON....wonder if I could ask on request-sponsor for someone in ON to move it?
[23:47:07] <the-decider> sleepcat: on sparc, pauses the OS and drops you to prom.
[23:47:28] <sleepcat> you can "pause" an os?  What can you do in a prom?
[23:47:37] <e^ipi> alanc: is this true of xorg on sparc as well?
[23:47:38] <the-decider> write forth code. ;)
[23:47:39] <bigjohnto> ok, I will try that, so far I have replaced the Sparc Processer and the memory
[23:47:45] <bigjohnto> also the hard drive
[23:47:46] <e^ipi> (given the very limited subsets of cards that run in such a manner
[23:48:05] <alanc> e^ipi: porting that code to Xorg is on the todo list still
[23:48:06] <the-decider> panic it, type "go" and jump back to the OS, all sorts of fun
[23:48:27] <e^ipi> seems an unstable way of doing things
[23:48:42] <the-decider> bigjohnto: there's not much left ;)
[23:48:47] <bigjohnto> but lets say i do have the same issue, i am assuming i will have to download a hardware tester from sun and just run some diags eh
[23:48:52] <e^ipi> the more crap you cram in to the kernel, the more there is to go wrong with it
[23:49:00] <bigjohnto> well i assuming it could just be the SOB mobo
[23:50:20] <bengtf_> bigjohnto: which u5 the early or late one ?
[23:50:31] <alanc> it was around the same time similar hacks were being done to put telnet & rlogin processing into streams modules in the kernel to save context switches
[23:51:23] <dlg> cos telnet has to perform
[23:52:07] <wesolows> _mw46_: that's normal; it should show up tomorrow (it's a horrid bridge + bugs by mail hackery) - I can get your bug id if you want
[23:52:28] <_mw46_> yes, I would need that.
[23:52:46] <wesolows> ordinarily it'll be mailed to you by one of the Tonic people
[23:52:52] <_mw46_> I have to do something for our ops people so that they can close their ticket.
[23:52:55] <bigjohnto> bengtf_, I think the early
[23:52:57] <wesolows> but that sometimes takes longer than it should, and it's inefficient as hell
[23:53:08] <wesolows> You want 6613211
[23:53:19] <_mw46_> thanks.
[23:54:30] <sleepcat> bengtf_: does the HP DeskJet 3930 work in SXCE?
[23:55:09] <bengtf_> dont know the only one online is b63 and it doesnt recognise it
[23:55:33] <bengtf_> get a little larger one ;)
[23:56:26] <sleepcat> My dad probably has a larger laser printer
[23:56:41] <sleepcat> that I could steal...errr borrow
[23:57:06] <e^ipi> sleepcat: http://openprinting.org/show_printer.cgi?recnum=HP-DeskJet_3920
[23:57:35] <e^ipi> you may need to install the hpijs driver, get the PPD from there and edit it to reflect the correct path of hpijs
[23:58:17] <sleepcat> i'm looking for something more click n' drool.
[23:58:32] <sleepcat> i spent 2 days trying to get cups to work
[23:58:34] <e^ipi> then buy a supported printer
[23:58:45] <sleepcat> i probably will
[23:59:18] *** peemus has quit IRC
[23:59:21] <e^ipi> meanwhile, the link I gave you is a matter of download, ./configure; gmake; gmake install, then using the included ppd





top