[00:05:05] *** the_unmaker has quit IRC
[00:06:18] *** cypromis_ has joined #opensolaris
[00:06:23] *** hannesd has quit IRC
[00:10:13] *** cypromis_ has quit IRC
[00:13:53] *** cypromis has quit IRC
[00:15:34] *** stux|work is now known as stux
[00:18:16] *** saite-f00f has quit IRC
[00:19:18] *** lolmac has quit IRC
[00:22:30] *** not-me-guv has joined #opensolaris
[00:27:21] *** ninjaslim has quit IRC
[00:27:28] *** lesterc has quit IRC
[00:27:39] *** ninjaslim has joined #opensolaris
[00:29:53] *** Wez has quit IRC
[00:31:23] *** capaz has quit IRC
[00:31:50] *** NeZetiC has quit IRC
[00:34:19] *** pumpkin_ has joined #opensolaris
[00:35:47] *** netj has quit IRC
[00:36:57] *** netj has joined #opensolaris
[00:37:30] *** netj has quit IRC
[00:38:31] *** netj has joined #opensolaris
[00:40:06] *** netj has quit IRC
[00:50:57] <ZOP> my download speed at work usually depends on the other end
[00:51:08] <ZOP> our smallest inbound pipe is like 50mbit.
[00:51:33] *** Cass has joined #opensolaris
[00:53:38] <jbk> is it me, or are the default font sizes getting smaller and smaller on sucessive sxce builds?
[00:58:14] <alanc> they reduced the default gnome desktop font in 99 or 100 I think
[00:58:14] *** duri has quit IRC
[00:58:18] *** duri has joined #opensolaris
[00:58:23] *** NeZetiC has joined #opensolaris
[00:59:05] *** ninjaslim has quit IRC
[01:05:14] *** NeZetiC has quit IRC
[01:06:58] *** ruse39[home] has joined #opensolaris
[01:07:29] <piwi> anybody knows when ips is synced to 99? hopefully i'll get new hardware tomorrow, and i wonder if i should start downloading indiana 98 or wait for 99 (if it comes before the weekend)
[01:08:14] <alanc> when they finish fixing the bugs they found in building the isos & repo updates for 99
[01:08:51] <alanc> were supposed to be trying to build again today with the latest fixes and start testing them, not sure how long it will take to test and decide if it's ready
[01:09:18] <piwi> i know the blocker bug for the final 2008.11 release, are there any other resources to get info?
[01:09:49] <alanc> pkg-discuss & indiana-discuss mailing lists
[01:09:59] <piwi> ok, thank you
[01:10:19] <alanc> don't remember how much I saw there vs. internal mail though
[01:11:02] <piwi> nothing on indiana discussion about 99
[01:11:19] <piwi> i'll check pkg
[01:11:43] <ZOP> http://www.newwest.net/topic/article/fifteen_hundred_boise_workers_to_lose_micron_jobs/C559/L559/
[01:11:48] <ZOP> microns stopping their NAND plant
[01:11:52] <ZOP> which...is bad
[01:12:01] <ZOP> because in 6 to 12 months demand will pick up again
[01:12:25] *** the_unmaker has joined #opensolaris
[01:12:53] <the_unmaker> wel
[01:12:58] <the_unmaker> opensolaris workin nicely
[01:13:00] <the_unmaker> heh
[01:13:08] *** lesterc has joined #opensolaris
[01:29:55] <the_unmaker> not too shabby
[01:31:18] *** alanc has quit IRC
[01:31:18] *** ZOP has quit IRC
[01:31:41] *** alanc has joined #opensolaris
[01:31:42] *** ChanServ sets mode: +o alanc
[01:38:58] *** pumpkin- has joined #opensolaris
[01:39:31] *** alanc has quit IRC
[01:40:49] *** sartek has quit IRC
[01:44:30] *** ZOP has joined #opensolaris
[01:49:24] <the_unmaker> ah the subtle differences from linux
[01:49:37] <the_unmaker> like for example where do i see loadavg?
[01:49:44] *** ninjaslim has joined #opensolaris
[01:49:45] *** jay-away has quit IRC
[01:50:25] <the_unmaker> prstat at bootom ah ok
[01:51:05] <the_unmaker> wow firefox takes RSS 140M?
[01:51:08] <the_unmaker> what a piggy
[01:53:41] *** sah-work has joined #opensolaris
[01:55:06] *** pumpkin_ has quit IRC
[01:56:39] *** spiki has quit IRC
[01:57:42] *** syamajala2_ has joined #opensolaris
[01:59:28] <CIA-25> Alexandre Chartre <Alexandre.Chartre at Sun dot COM>: 6640564 vds should not serialize disk image IO, 6684721 file backed virtual i/o should be synchronous, 6742587 vds can ACK a request twice
[02:00:11] *** ruse39[home] has quit IRC
[02:04:17] <the_unmaker> uh termianl wont let me do a colon :
[02:04:18] <the_unmaker> ?
[02:05:52] *** ruse39[home] has joined #opensolaris
[02:11:58] *** houst0n- has quit IRC
[02:12:46] *** syamajala2 has quit IRC
[02:13:22] *** houst0n- has joined #opensolaris
[02:16:30] *** crichardso has quit IRC
[02:17:23] <kimc> i remember when micron began making memory in Idaho ~15 years ago
[02:17:51] <kimc> at the time it was remarkable they could do it in Idaho of all places
[02:18:04] *** syamajala2_ has quit IRC
[02:18:09] <kimc> not Singapore not..
[02:18:36] <kimc> potato farmers with some money to invest
[02:18:57] <kimc> and the technology to create the product
[02:19:09] *** syamajala2 has joined #opensolaris
[02:19:43] <kimc> it was great -still is
[02:20:50] *** alanc-away has joined #opensolaris
[02:21:46] *** alanc-away is now known as alanc
[02:22:59] *** faber has joined #opensolaris
[02:23:19] <faber> hai
[02:23:24] <faber> i want to ask about solaris
[02:23:35] <faber> any body can answer
[02:23:44] <e^ipi> metaquestions are annoying
[02:23:49] <e^ipi> don't ask if you can ask
[02:23:51] <e^ipi> just ask
[02:24:12] <faber> hmm
[02:24:20] <alanc> there are too many things you could ask about solaris to know if we can answer until you ask - no one knows everything, everyone knows something
[02:24:31] *** bubbva has quit IRC
[02:24:43] *** ChanServ sets mode: +o alanc
[02:25:25] <faber> i want to transform my operating system to solaris from fedora core to solaris
[02:25:32] <faber> what do you think
[02:25:58] <e^ipi> if it would make you happy, do it
[02:26:40] <faber> i am thinking about how solaris can support my laptop toshiba satellite A50
[02:26:55] <piwi> try the live cd
[02:27:52] <faber> what the best things from solaris
[02:28:04] <faber> can you tell me about it
[02:28:26] <e^ipi> why doesn't marketing have a single page that we can point people to for that by now anyways?
[02:28:29] <e^ipi> Gman_: make it so.
[02:28:40] *** coffman has joined #opensolaris
[02:29:08] <piwi> shouldn't be osol.com this kind of page?
[02:29:09] *** coffman has quit IRC
[02:29:17] <e^ipi> yes, it should
[02:29:46] <faber> hmm
[02:30:18] <piwi> faber: what do you want hear?
[02:30:29] <Cass> faber, i think the question is what you need from solaris that you use on fedora
[02:30:30] <faber> hear about what?
[02:30:32] <e^ipi> faber: google for these terms: zfs, solaris zones, rbac, smf
[02:30:39] <piwi> using it as a desktop os? for development? as a server?
[02:30:40] <evocallaghan1> faber:Burn this to CD and boot it http://www.sun.com/bigadmin/hcl/hcts/install_check.jsp
[02:30:59] <faber> i want to learn about netbeans and mysql
[02:31:17] <faber> i think solaris is open source os from sun
[02:31:26] <e^ipi> yes, it is...
[02:31:27] <piwi> so far so right
[02:31:28] <faber> so solaris can support all java development kit
[02:31:32] <e^ipi> yes
[02:31:40] <faber> hmm
[02:31:45] <Gman_> e^ipi: i'm sure www.sun.com/software/solaris has something useful
[02:31:50] <Cass> faber:  then if thats all you want then providing the display/network drivers are available then you should have fun
[02:32:04] *** Odin- has quit IRC
[02:32:23] <faber> ok
[02:32:33] <faber> i'll transform my os to solaris
[02:32:56] <Cass> if you need the other stuff multimedia codecs etc then maybe its not so easy as fedora
[02:33:01] <faber> what do you advise to me
[02:33:03] <e^ipi> Gman_: well, there's this http://www.sun.com/software/solaris/features.jsp but it'd still be cool to have an opensolaris features cheat sheet
[02:33:19] <Gman_> e^ipi: yeah, i think there is something
[02:33:22] <Gman_> but it's pretty small
[02:33:34] <Cass> i would say solaris 100% of the time bur im biased and i have a mac for all the nicities
[02:33:52] <Gman_> http://www.opensolaris.com/opensolaris_datasheet.pdf
[02:33:57] <faber> solaris as well as mac ?
[02:33:58] <piwi> 10 things you would love to have in your os
[02:33:58] <Gman_> it's a bit short on information
[02:34:10] <Gman_> (nothing from stopping any of you guys from writing one...)
[02:34:20] <Cass> yes, im typing this in Nevada b96 in vmware on my mac
[02:34:37] <e^ipi> Gman_: :P fine, i'll write up a blog entry on it...
[02:34:51] <Gman_> :)
[02:35:30] <faber> hmm
[02:35:50] <faber> thx
[02:36:41] <faber> hai
[02:37:03] <faber> how do long you use solaris?
[02:37:10] <Cass> faber: i did use solaris / linux 100% of the time as a desktop OS before i went the mac route but i was tired of hacking around trying to get things to do as i liked ...
[02:37:32] <Cass> i use solaris for 8 years now .. im a solaris sys admin as a daytime job
[02:38:17] <faber> how command in solaris as same as linux
[02:38:21] <Cass> i dont want to come home and have to piss around with stuff just to edit photos or watch a film in whatever format, hence my mgration to mac
[02:38:29] <e^ipi> faber: there are some differences
[02:38:42] <e^ipi> the manual pages are good though
[02:38:46] *** stevel has quit IRC
[02:39:45] <faber> how much percent different?
[02:39:51] <piwi> you may find this useful: http://bhami.com/rosetta.html
[02:40:06] <piwi> but i think no one can measure it
[02:40:22] *** alanc changes topic to "SXCE 99, ON 100, IPS 98 || Step one: see if SAG answers your question: http://docs.sun.com/app/docs/coll/47.16 || The answer to "I do $foo on *ux, how do I do it on Solaris" is http://bhami.com/rosetta.html || Clipboard: http://opensolaris.pastebin.ca"
[02:40:46] <e^ipi> whatever percentage you would assign to the difference between a SysV operating system and a GNU operating system
[02:41:08] <faber> hmm
[02:41:13] <faber> just to know
[02:41:23] <e^ipi> it's noticable
[02:41:32] <faber> how can i transform my knowledge to learn a new os
[02:41:35] *** syamajala2 has quit IRC
[02:41:38] <Cass> faber:  just install it, play and see what you think yourself... its the true test of %
[02:41:40] <e^ipi> it's not impossible or even hard to get the difference
[02:42:11] <faber> ok
[02:42:26] <e^ipi> i used linux for almost 10 years before i started using solaris, it took me a couple days to feel comfortable in the new environment
[02:42:30] <faber> solaris support dual boot?
[02:42:37] <e^ipi> of course
[02:42:38] <Cass> yes
[02:42:48] <faber> great
[02:42:56] <faber> solaris support novell?
[02:43:05] <e^ipi> novell what?
[02:43:17] <Cass> its grub as a bootloader so you can boot what you like
[02:43:24] <faber> novell = like software for networking
[02:43:33] <Cass> netware ?
[02:43:46] <faber> yes
[02:43:47] <faber> hehehehe
[02:43:48] <faber> sorry
[02:44:01] <faber> like that
[02:44:06] <e^ipi> is that thing still alive?
[02:44:36] <Cass> they use tcp/ip these days do they not ..
[02:44:55] <Cass> so id say yes ... im sure there are nds hooks from novell for solaris if thats your need
[02:45:29] <faber> how about oracle on solaris
[02:45:44] <faber> that's work?
[02:45:44] <Cass> well supported
[02:45:49] <faber> great
[02:45:55] <faber> i'll like it
[02:46:21] <Cass> you may have an issue getting oracle 9 on x86 but 10 and 11 are available im sure
[02:46:47] <faber> how to troubleshoot in solaris
[02:46:55] *** uebayasi has quit IRC
[02:46:57] <Cass> same as linux
[02:47:06] <Cass> depends on the problem
[02:47:14] <faber> hmm
[02:47:16] <faber> Cass
[02:47:22] <Cass> with solaris you have truss/dtrace for deep issues
[02:47:24] <faber> what do you choose?
[02:47:29] <Cass> for ?
[02:47:30] <faber> networking or database?
[02:47:41] <Cass> i dont understand the question
[02:47:50] <Cass> i hate databases so none ...
[02:47:54] <faber> hehehehe
[02:47:59] <Cass> but networking im not sure what you mean
[02:48:03] <faber> u like networking of course
[02:48:06] <Cass> yes
[02:48:41] <faber> r u learn about networking
[02:48:45] *** lolmac has joined #opensolaris
[02:48:57] <Cass> i know network troubleshooting .. yes ...
[02:49:18] <Cass> ping, traceroute, snoop/tcpdump .. cisco stuff
[02:49:29] <faber> yeah
[02:49:35] <faber> i learn about cisco
[02:49:39] <faber> get the certify
[02:49:41] <Cass> yeah it comes up every day at work
[02:49:49] <Cass> i did a ccna years ago
[02:50:01] <faber> u get ccnp
[02:50:31] <Cass> nope, didnt value the cert, ccna was useless enough in my opinion, too much money for no value add as solaris SA
[02:52:36] *** sah-work has quit IRC
[02:52:54] <faber> ohh
[02:53:09] <faber> r u software developer
[02:53:25] <Cass> no, im a systems administrator, i manage solaris systems
[02:53:59] <Cass> cisco certs are really only good if you configure cisco equipment on a daily basis, which i dont
[02:54:41] <Cass> i know enough to log in and troubleshoot to a certain degree to help in my job but for conectvity issues i have no control over i pass to out outsourced network team
[02:55:08] <Cass> i dont install or architect networks so cisco certs are useless to me
[02:57:42] <faber> Is solaris good to learn data center?
[02:58:03] <Cass> what you looking for ?
[02:58:23] <faber> make me know
[02:58:27] <faber> what is solaris for?
[02:58:34] <Cass> solaris is the best server os in the world ... if you want to learn unix use it if not dont ..
[02:58:41] <faber> so i can use it for long time?
[02:58:49] <e^ipi> most of us do
[02:58:55] <Cass> are you trolling ?  looking for homework answers ? whats all the questions leading up to ?
[02:58:56] <faber> yeah
[02:59:05] <faber> no
[02:59:22] <faber> all questions to make me ensure
[02:59:27] <CIA-25> Mark Johnson <Mark.Johnson at Sun dot COM>: 6756798 xpvtap doesn't delete as_callback if IO not performed
[02:59:34] <faber> so i can commit to solaris
[02:59:38] <e^ipi> as for servers... dclarke has some old solaris 8 machines with years and years of uptime IIRC
[02:59:48] <Cass> whats the worry ... you dont like it after install remove it and continue about your business
[03:00:09] <faber> too much worry
[03:00:17] <Cass> why ?
[03:00:18] <faber> like compability to my laptop
[03:00:26] <Cass> google laptop make and solaris
[03:00:28] <faber> yeah
[03:00:43] <faber> ok cas
[03:00:48] <e^ipi> or just try it...
[03:00:49] <faber> u r great
[03:00:52] <Cass> get vmware and install via that to see whats up
[03:00:56] <e^ipi> if it doesn't work, sux, file a bug and move on
[03:00:59] <faber> i'll try live cd
[03:01:08] <faber> ok
[03:02:44] *** `Matir has joined #opensolaris
[03:02:56] <faber> ok
[03:03:17] <faber> solaris dekstop like kde or gnome?
[03:03:24] <e^ipi> gnome
[03:03:54] <faber> if you choose one
[03:04:01] <faber> kde or gnome
[03:04:11] <faber> because my kde always crash
[03:04:35] <Cass> gnome
[03:05:33] <faber> hmm
[03:05:38] <faber> like ubuntu
[03:05:44] <faber> i like
[03:05:56] <faber> i'll like to change os
[03:06:01] <faber> so complicated
[03:06:10] <e^ipi> not really
[03:06:10] <faber> i need tentor
[03:06:17] <Cass> Nexenta, is nevada but with gnu tools, like ubuntu
[03:07:56] <faber> Cass
[03:08:26] <Cass> yes
[03:08:50] <faber> i need friend
[03:08:56] <Cass> easy tiger
[03:09:08] <faber> can u support me
[03:09:22] <Cass> this is what this channel is all about
[03:09:31] <Cass> im not into personal support
[03:09:40] <Cass> you have a problem, come here ..
[03:10:09] <Cass> not trying to be harsh but im way too busy to be on call for internet support
[03:10:47] *** mukiex has joined #opensolaris
[03:11:56] <e^ipi> faber: if you ask a specific, well formulated question, people are happy to answer.
[03:12:22] <e^ipi> vague calls for nonspecific help will result either in being ignored or snide remarks, depending what sort of mood i'm in
[03:13:30] <mukiex> I want to make a fileserver that may also act as a printserver, webserver, svn server, and maybe, just maybe, a MythTV *back*end box. It'll be an 8-drive Raid6 (Raid-Z2 is fine too) array. I kind of want to use ZFS. Is OpenSolaris right for me, or should I look into ZSF-Fuze for Linux?
[03:14:38] <e^ipi> you'd trust zfs-fuse?
[03:14:45] <e^ipi> that seems kinna foolish don't you think?
[03:14:47] <mukiex> See, that's wherein the problem lies.
[03:14:56] <mukiex> I know very little about its current stability...
[03:15:12] <e^ipi> any reason why you'd think solaris couldn't do print/web/svn/whatever ?
[03:15:19] <e^ipi> because that's pretty much what my fileserver does
[03:15:22] <Cass> if mythtv is on solaris then solaris is the guy
[03:15:42] <mukiex> MythTV is the only major worry.
[03:16:19] <mukiex> I mean, print should be fine, hell, all my printers are HP, which are supposed to be BSD/Linux friendly, so I figure it's probably fine on Solaris as well...
[03:16:43] <mukiex> How good is RAID-Z2?
[03:16:58] <e^ipi> in what sense?
[03:17:12] <piwi> easy answer: it's better :)
[03:17:18] <e^ipi> it lets 2 disks die
[03:17:21] <e^ipi> so, hooray
[03:17:31] <mukiex> I mean, compared to RAID6
[03:17:38] <e^ipi> they are the same thing
[03:17:48] <e^ipi> w/ raidz2 you get all the ZFS goodies though
[03:17:56] <mukiex> Is there a scrubbing function? (This may be the dumbest question you hear all day, but I know next-to-zilch about OS)
[03:17:56] <piwi> no "write hole" as they call it
[03:18:00] <e^ipi> ( checksums and associated reliability, snapshots, etc )
[03:18:04] <piwi> yes
[03:18:32] <e^ipi> it's called 'scrub' as a matter of fact
[03:18:48] <mukiex> Noice =3
[03:18:52] <Cass> http://opensolaris.org/jive/thread.jspa?messageID=288006&tstart=0
[03:19:04] <Cass> might be worth a look for a clue on mythtv for opensolaris
[03:19:09] <Cass> others have asked
[03:19:18] *** lolmac has quit IRC
[03:19:28] <mukiex> See, letting 2 disks die isn't as important to me as "let's an extra parity stripe fix things if one disk dies and things aren't perfect when rebuilding that disk"
[03:19:29] <piwi> raid-z from the author: http://blogs.sun.com/bonwick/entry/raid_z
[03:19:58] <tnelson> Hey, so, I've given up trying to get native Solaris LDAP talking to my Windows Server 2008 AD and moved onto NIS instead.
[03:20:20] *** fr4g has quit IRC
[03:20:29] <tnelson> The *only* way NIS works is if I launch ypbind w/ -broadcast.
[03:20:31] <e^ipi> mukiex: that's the same thing as two disks dying
[03:20:54] <tnelson> If I rely on /var/yp/binding/domain/ypservers (10.0.0.1), it just flat out doesn't work.
[03:20:56] <Cass> tnelson: ypinit -c and hardcode the nis servers does not work for you ?
[03:21:03] <Cass> weird
[03:21:16] <tnelson> I've traced the packets as well via wireshark, absolutely nothing is sent to 10.0.0.1.
[03:21:49] *** syamajala2 has joined #opensolaris
[03:22:06] *** syamajala2 has quit IRC
[03:22:09] <tnelson> Bunch of other OSes on this network don't have any issue; AIX 6, FreeBSD, DragonFlyBSD, Linux, etc.
[03:22:22] <tnelson> (Don't have any issue when 10.0.0.1 is explicitly set, that is.)
[03:22:41] <tnelson> I've even patchadd'd up to the latest everything.
[03:22:43] <tnelson> Still no dice.
[03:23:03] <mukiex> Wow, it almost looks like RaidZ2 doesn't even *need* scrubbin' =3
[03:23:17] <the_unmaker> ok for web under heavy load glassfish tomcat alone ot sun webserver 7?
[03:23:20] <tnelson> Given that I've been messing around with this crap for nearly a week -- can I fool the svc nis/client service?
[03:23:23] <the_unmaker> which can handle tons fo hits?
[03:23:24] <e^ipi> mukiex: well, it does...
[03:23:36] <mukiex> Oh I know it does, but the self-healing looks neat =3
[03:23:42] <e^ipi> mukiex: zfs does checksums and can self heal
[03:23:45] <tnelson> Like, change /var/yp/binding/domain/ypservers to 0.0.0.0 or something.
[03:23:51] <e^ipi> but if it never accesses a block, it never checks the checksums
[03:24:08] <e^ipi> scrubbing though can fix problems instead of leaving you with corrupted blocks
[03:24:18] * evocallaghan1 is porting MythTV to solaris :D
[03:24:38] <evocallaghan1> Anyone like to stepup and help with it is welcome
[03:25:08] <Cass> tnelson: your trying to bind the server to itself or a another client to the nis server ?
[03:25:10] <tnelson> Ah, nevermind, just removing ../domain/ypservers did the trick.
[03:25:26] <tnelson> (It now ypbind -broadcast's, and everything finally freakin' works.)
[03:25:44] <tnelson> The Solaris 10 box that I'm on is just a NIS client.
[03:25:57] <tnelson> 10.0.0.1 is the Active Directory serving NIS.
[03:26:12] <Cass> yeah, never seen that problem before, hardcoding the server always has worked for me
[03:26:21] <tnelson> rm'ing ypservers did the trick, so I'm just going to leave it like that.
[03:26:35] <tnelson> Trying to get LDAP working nearly brought me to tears.
[03:26:47] <Cass> that would annoy me to be honest :-)
[03:27:05] <the_unmaker> ok why does my terminal not allow me to do ":"
[03:27:06] <the_unmaker> ?
[03:27:12] <the_unmaker> can't scp stuff
[03:27:15] <the_unmaker> what the heck?
[03:27:35] <Cass> wrong keyboard map  ?  what do you get instead of : ?
[03:27:46] <the_unmaker> oh weird now it work
[03:27:51] <the_unmaker> ?!
[03:28:06] <Cass> you didnt press the button did you
[03:28:08] <Cass> lol
[03:28:09] <the_unmaker> Cass: you are good
[03:28:16] <the_unmaker> no t was in some weird mode
[03:28:18] <mukiex> e^ipi : How hard is it to move from, say, Ubuntu to Opensolaris, for a server? In terms of feel, I mean; I SSH to my Ubi server a lot.
[03:28:20] <the_unmaker> and i backspaced
[03:28:28] <the_unmaker> then retyped last leter of server anem and added :
[03:28:32] <the_unmaker> and ti let me
[03:28:40] <the_unmaker> ferginfargin!
[03:28:47] <Cass> haha
[03:29:50] <e^ipi> mukiex: there are differences
[03:30:05] <e^ipi> but it's not bad, you'll get a feel for it in a week or two
[03:30:49] <mukiex> I might try some RAID-Z2 with some usb drives or something first. =3
[03:30:58] <e^ipi> yeah, go for it
[03:31:07] <e^ipi> you can also use on-disk files
[03:31:12] <e^ipi> ( loopback mounting )
[03:32:28] <Cass> anyone here zfs root on a sparc box ?
[03:32:42] <Cass> just noticed today its now possible
[03:32:48] <e^ipi> yeah, it is
[03:32:52] <mukiex> Might be a bit of a divergence, but what do you think would make a good proc for an 8-drive RAID-Z2 array? =3 I'm probably gonna build the fileserver from scratch, haven't dealt with the performance hit on striping, let alone the parity-writing...
[03:32:54] <Cass> need obp update ?
[03:33:01] <e^ipi> nope
[03:33:16] <e^ipi> mukiex: as cheap as you can find, but 64 bit is a must
[03:33:25] <piwi> buy ram
[03:33:25] <mukiex> Quad? Dual?
[03:33:28] <Cass> cool ... figured it did given some of the boot commands
[03:33:29] <piwi> lots of it
[03:33:33] <e^ipi> single, dual,whatever's cheap
[03:33:37] <mukiex> 4G's enough?
[03:33:40] <e^ipi> but yes, as much ram as you can afford
[03:33:49] <mukiex> Caching reasons?
[03:34:03] <e^ipi> if it's a difference between 4 cores x 4 gigs and 2 cores x 8 gigs, buy the dual core and spend the rest on ram
[03:34:10] <mukiex> wow
[03:34:13] <e^ipi> yeah, ZFS caches very aggressively
[03:34:18] *** XYOX has joined #opensolaris
[03:34:19] <mukiex> Ah.
[03:34:21] *** XYOX has left #opensolaris
[03:34:22] <mukiex> Wow.
[03:34:24] <piwi> zfs uses every free bit as cache
[03:34:35] <e^ipi> it doesn't interfere with your apps, but it likes a lot of ram
[03:34:40] *** faber has quit IRC
[03:34:48] <mukiex> Well, given that the hard drives are probably going to cost the most, I'll make sure to look for a mobo with enough ram slots for 8 gigs.
[03:35:00] <e^ipi> these days that's 4 slots
[03:35:05] <e^ipi> that's standard
[03:35:21] <mukiex> That makes sense, I've got 2 2GB dims on my desktop =3
[03:35:49] <mukiex> (too bad it seems like no OS I've tried on it knows how to use 4 gigs. Even fucken Vista 64 only sees 3.25 AFAICT)
[03:36:41] <e^ipi> on this machine:
[03:36:42] <e^ipi> Memory size: 8191 Megabytes
[03:36:51] <Cass> yeah the rest is kernel memory
[03:37:00] *** kmays has joined #opensolaris
[03:38:26] <piwi> e^ipi: consumer mobo, oder something more workstation like?
[03:38:50] <e^ipi> nah, cheapass asus athlon64 kit
[03:39:30] <piwi> bought a m2a-vm some days ago, hope it runs with 8 gig
[03:39:38] <the_unmaker> hm I see some apps are 64 bit on my new opensolaris
[03:39:47] <e^ipi> some are, some aren't
[03:39:48] <the_unmaker> but arck -k gives me i386
[03:39:54] <the_unmaker> ..
[03:40:02] <e^ipi> isainfo
[03:40:06] <the_unmaker> did I make a mistake and get i386
[03:40:19] <e^ipi> you can't "get" i386
[03:40:23] <e^ipi> there's only one iso
[03:40:23] <the_unmaker> oh that gives amd64 i386
[03:40:27] <the_unmaker> ok
[03:40:30] <e^ipi> solaris chooses the correct kernel to boot
[03:40:34] <the_unmaker> ah ha!
[03:40:50] <the_unmaker> uname -a SunOS opensolaris 5.11 snv_95 i86pc i386 i86pc Solaris
[03:40:59] <the_unmaker> shouldn't this say amd64?
[03:41:07] <e^ipi> no
[03:41:09] <the_unmaker> ok
[03:41:22] <e^ipi> the architecture is still the same
[03:41:30] <the_unmaker> isainfo gives amd64 adn i386
[03:41:39] <the_unmaker> thats interesting
[03:41:44] <e^ipi> why?
[03:41:49] <e^ipi> amd64 is just a superset of i386
[03:41:55] <the_unmaker> so is it hybrid? some apps 32 bit?
[03:41:57] <the_unmaker> oh
[03:42:05] <e^ipi> yes, most apps are 32 bit
[03:42:13] *** sartek has joined #opensolaris
[03:42:15] <the_unmaker> whys that
[03:42:18] <e^ipi> because having a 64-bit 'ls' makes no sense at all
[03:42:22] <the_unmaker> ah ok
[03:42:26] <the_unmaker> ok
[03:42:32] <the_unmaker> so solaris builders are smart
[03:42:39] <the_unmaker> only use 64bit when it counts
[03:42:45] <e^ipi> yes
[03:42:56] <mukiex> Nice
[03:43:01] <e^ipi> sun's been doing 64 bit for decades
[03:43:12] <e^ipi> they know how to do it correctly by now
[03:44:28] <mukiex> You do make a good point.
[03:47:24] <mukiex> Plus, you can't buy laptops capable of bludgeoning someone to death for any other OS =3
[03:47:32] <e^ipi> and isaexec will choose the correct binary
[03:47:54] <e^ipi> assuming there's a difference
[03:52:26] *** jklyekai has joined #opensolaris
[03:54:48] *** xtrondo has quit IRC
[03:55:04] *** Plazma has joined #opensolaris
[03:55:56] <maxote> e^ipi, no, sun did nothing for decades
[03:56:10] <e^ipi> pardon?
[03:56:49] <maxote> remember the slowaris story
[03:56:57] <e^ipi> yeah, good times
[03:57:10] <e^ipi> but that hasn't been true for quite a while
[03:58:50] <e^ipi> for that matter on multicore machines, you'd be hard pressed to find something that beats it
[03:59:28] <CIA-25> Jeff Bonwick <Jeff.Bonwick at Sun dot COM>: 6755897 recursive mutex enter in spa_vdev_enter when running test suite
[04:01:19] <Cass> slowaris, brings back fond memories, fetchmail, sendmail, cde dtmail app, dial up modems ...  them were the days
[04:01:32] *** jfisc has quit IRC
[04:02:28] <Plazma> xmodem :D
[04:02:32] <Plazma> that takes me back
[04:02:50] <snakesqzns> kermit
[04:02:52] <snakesqzns> :p
[04:03:07] <Plazma> i still use kermit to this day
[04:03:10] <Plazma> by far the best IMHO for what it does
[04:03:13] <Plazma> i detest minicom
[04:03:23] <the_unmaker> hey how do I switch to another teminal screen while im in x to confuse anyone who wants to fuck with my box now hat im going home?
[04:03:26] <Plazma> but i also like tip/cu :D
[04:03:27] <the_unmaker> i haev ad ownlaod runnig
[04:03:36] <the_unmaker> oh wait maybe gnome can lock me screen...
[04:03:44] <e^ipi> it can
[04:03:45] <the_unmaker> ooow
[04:03:49] <the_unmaker> how very convenient
[04:03:52] <snakesqzns> i used to dialup my local library. it dropped you into lynx, but you could drop to a shell are run something (slurp i think) to get a tcp stack for running mosaic
[04:04:33] <Plazma> come to think of it, i hate xmodem too
[04:04:36] <Cass> tip was my dialer of choice :-)
[04:04:36] <snakesqzns> The-spiki, control-alt-L
[04:04:48] <Plazma> i <3 kermit with a passion.. it just "works"
[04:04:50] <snakesqzns> er, the_unmaker
[04:04:53] <Plazma> and its got an ingenious design
[04:06:30] <snakesqzns> i think i always used zmodem for transfers
[04:06:54] *** Wil has left #opensolaris
[04:07:51] <jbk> well b99 seems to be better than b98
[04:08:21] <e^ipi> jbk: can you create a zone such that you get a new zfs dataset out of it?
[04:08:31] <jbk> hmm haven't tried
[04:09:08] <e^ipi> ( that is, zfs create mypool/zones; zonecfg -z newzone 'set zonepath=/zones/newzone'; zoneadm install newzone )
[04:09:19] <e^ipi> i'm on b100 and i can't
[04:09:22] <e^ipi> and it's annoying
[04:09:55] <snakesqzns> i believe zoneadm takes care of creating the zfs filesystem
[04:10:03] <snakesqzns> if /zones is zfs already
[04:10:19] <e^ipi> snakesqzns: yes, that is the behavior i'm expecting.
[04:10:33] <e^ipi> not the behavior i'm getting
[04:11:08] <snakesqzns> i was upset to discover zones on zfs isn't supported on production solaris
[04:11:29] <e^ipi> supported as in buying support, or supported as in can do it at all
[04:11:43] <snakesqzns> hold on, trying to find the link
[04:11:53] <jbk> ferrari% pfexec /usr/sbin/zoneadm -z test install
[04:11:53] <jbk> A ZFS file system has been created for this zone.
[04:12:08] <e^ipi> indiana or nevada?
[04:12:15] <jbk> sxce b99
[04:12:22] <e^ipi> i guess it's a b100 bug
[04:12:28] <jbk> and it did create the zone
[04:12:38] <e^ipi> that sucks :-/
[04:13:03] *** nachox has joined #opensolaris
[04:13:45] *** kmays has quit IRC
[04:13:50] <jbk> now as my laptop becomes choppy as it's poor hd churns :)
[04:14:46] <snakesqzns> e^ipi, http://docs.sun.com/app/docs/doc/817-2271/gayov?l=en&q=zfs+zones&a=view
[04:15:03] <snakesqzns> e^ipi, oops that's not it
[04:15:44] *** bhall has quit IRC
[04:16:40] <evocallaghan1> e^ipi:What's the problem with 100 ? Should I skip it if I use zones/zfs ?
[04:17:55] <snakesqzns> e^ipi, oh wait that was it!
[04:17:56] <jbk> hmm.. i wonder if nfs or cifs will be modified to use volo
[04:18:21] <snakesqzns> this one worries me "Do not use a ZFS file system for a global zone root path or a non-global zone root path in the Solaris 10 releases. You can use ZFS as a zone root path in the Solaris Express releases, but keep in mind that patching or upgrading these zones is not supported."
[04:18:23] <e^ipi> snakesqzns: i'm well aware of how it works
[04:18:34] <snakesqzns> e^ipi, well excuse me.
[04:19:20] <jbk> i wonder with s10u6 if the 'no upgrade zones on zfs' limitation goes away
[04:19:53] <e^ipi> one would hope so
[04:19:54] <evocallaghan1> jbk:I am wondering if u6 is going to have zfs root support :p ?
[04:20:04] <jbk> it's supposed to
[04:20:11] <jbk> hence why i'm wondering
[04:20:23] <jbk> since I _really_ want to upgrade all the boxes at work to u6 w/ zfs root once it's out
[04:20:37] <jbk> now if there was only a  way to get them off oracle's ebusiness suite :)
[04:21:51] <evocallaghan1> :p
[04:22:10] <jbk> you can't even cluster it too well unless you use rac *shudder*
[04:22:18] *** uebayasi has joined #opensolaris
[04:23:15] <evocallaghan1> Well I had a panic yesterday .. So been spending my time sorting that out
[04:23:53] <maxote> this = a waste of time
[04:24:00] *** Wil has joined #opensolaris
[04:25:22] <evocallaghan1> ay?
[04:26:16] <nachox> evocallaghan1, that nick is seriously annoying for those of us using xchat in a small window, it's too long
[04:26:28] <e^ipi> hmm?
[04:26:37] <e^ipi> ( re maxote )
[04:26:49] <evocallaghan1> speaking of which
[04:26:55] *** evocallaghan1 is now known as evocallaghan
[04:27:07] <nachox> still long
[04:27:25] <nachox> actually anything longer than 8 chars is long
[04:27:39] <evocallaghan> nachox:Get reall. This is not 1984
[04:27:46] <maxote> what's the norm?
[04:27:47] <evocallaghan> s/reall/real/
[04:28:20] <evocallaghan> A normal size monitor with correct res/dpi settings maybe
[04:28:29] <maxote> max. 80 - 15 (nick) = 65 columns?
[04:28:34] *** e^ipi has quit IRC
[04:28:42] * evocallaghan *shrugs*
[04:28:46] <ninjaslim> hi guys, what's the roadmap for SXCE, is it going to get IPS or will it continue to have SVR4 packages?
[04:29:15] <nachox> i'm guessing it will eventually be replaced by indiana
[04:29:19] *** syamajala2 has joined #opensolaris
[04:29:29] <evocallaghan> ninjaslim:prob conitue with SVR4 with some hope
[04:29:52] <ninjaslim> evocallaghan: why do you with some hope?
[04:30:15] <evocallaghan> Many *many* problems with IPS
[04:30:25] <ninjaslim> nachox: why do you think that, that means all the folks who want real Unix will have to go for standard Solaris
[04:30:32] <evocallaghan> beyond a lab project, its useless in production
[04:30:33] <ninjaslim> evocallaghan: yes i've noticed those too
[04:30:52] <nachox> ninjaslim, what do you exactly mean by "real unix"?
[04:31:15] <evocallaghan> I just moved back to sxce after a good 9m? of indiana madness
[04:31:18] <ninjaslim> indiana is GNU/OpenSolaris
[04:31:28] <evocallaghan> POSIX/SUS3
[04:31:55] <evocallaghan> http://www.unix.org/
[04:32:11] <evocallaghan> and a correct $PATH
[04:32:25] <ninjaslim> wait what
[04:33:10] <nachox> ninjaslim, indiana includes the regular solaris tools, you just need to correct the PATH, it IS a real unix, and i'm sure sun will certify it just like it did with solaris 10 when the time comes
[04:33:43] <evocallaghan> http://rafb.net/p/nmYSQk30.html
[04:34:14] <evocallaghan> ninjaslim:Correct PATH ^
[04:34:15] <nachox> and the default PATH for solaris when it is finally indiana based is not set in stone, nothing was said about that yet
[04:34:32] <ninjaslim> i see
[04:34:49] <ninjaslim> i really couldn't use OpenSolaris much, SXCE is very nice though
[04:35:00] <evocallaghan> +1
[04:35:49] *** `Matir has quit IRC
[04:36:46] <ninjaslim> i'm tempted to get a box and just set up SXCE on there
[04:39:33] *** freakazoid0223 has joined #opensolaris
[04:40:24] <evocallaghan> Why can't I access my bug on bugster ? http://bugs.opensolaris.org/view_bug.do?bug_id=6757777
[04:40:48] <evocallaghan> ninjaslim:Try to go for 100 if you can
[04:41:56] <evocallaghan> CR 6757777
[04:42:25] <ninjaslim> build 100?
[04:42:43] *** e^ipi has joined #opensolaris
[04:42:43] *** ChanServ sets mode: +o e^ipi
[04:43:04] <e^ipi> that's much worse than not creating a zfs filesystem
[04:43:10] <e^ipi> it panicked
[04:43:42] <nachox> ?
[04:44:15] <e^ipi> stupid crap nvidia sata drivers i think
[04:44:37] <evocallaghan> ninjaslim:yes, snv_100+
[04:44:59] <evocallaghan> e^ipi:Got a problem; "This is a high priority CR and requires your immediate attention.
[04:44:59] 
[04:45:03] <jbk> e^ipi: btw, i looked at the diffs for zoneadm between b98 & current, and can't see anything that'd explain what you're experiencing
[04:45:12] <evocallaghan> But I can't find it in the db yet ?
[04:45:29] <e^ipi> no, it has to go through people first
[04:45:38] <evocallaghan> ok
[04:45:49] <e^ipi> you'll get an email in the morning most likely
[04:45:52] <evocallaghan> e^ipi:But its expecting me to provide info..
[04:45:58] <alanc> bugs.opensolaris.org only gets updates from the internal bug db once a day
[04:46:05] <evocallaghan> e^ipi:I got a heap of emails already
[04:46:35] <evocallaghan> alanc:thanks for the heaps up
[04:46:50] <evocallaghan> Two of the emails have said they need my immediate attention..
[04:47:27] <evocallaghan> I contacted Garrett on his blog about the panic http://gdamore.blogspot.com/2008/10/new-stuff-in-opensolaris-200811.html
[04:47:32] <alanc> I think that's intended for the Sun engineers working on it, and they forgot to remove that from the script that generates the internal e-mail
[04:47:41] <evocallaghan> He got back to me in 40min which was great !
[04:47:54] <evocallaghan> alanc:OK thanks
[04:48:21] <alanc> the internal comments suggest it may be a dup of either 6750056 (fixed in nv_100) or 6756588 (fix being worked on now)
[04:48:28] <evocallaghan> I stuck on my blog a workaround for the problem
[04:48:44] <evocallaghan> alanc:That is what I see here to
[04:49:08] <evocallaghan> 6756588 (a better fix for 6750056)
[04:49:38] <sstallion> evening all
[04:49:59] <e^ipi> oh for christ sakes...
[04:50:07] <e^ipi> the panic didn't leave me a core file
[04:50:20] <evocallaghan> e^ipi:That's no good
[04:50:24] <sstallion> e^ipi: it wouldnt have. it should have dumped to your dump device
[04:50:46] <evocallaghan> We had a camera phone yesterday so I could acturly see the panic :p
[04:50:56] <e^ipi> sstallion: s'what i meant
[04:51:30] <jbk> e^ipi: what are you getting?
[04:51:37] <alanc> ninjaslim: the roadmap is SXCE & Indiana eventually merge into one, with IPS as the package system
[04:52:02] <e^ipi> jbk: flashes by too fast to see, but something along the call chain that involves the nv sata driver
[04:52:12] <alanc> though that's probably a year out or so
[04:52:29] <evocallaghan> Garrett seems to be doing some really fantastic work ! I'm glad he got back to me so quick and its going to be sorted so soon :D
[04:52:56] <sstallion> e^ipi: boot with -d set
[04:53:00] <evocallaghan> alanc:Only a year O_o
[04:53:02] <jbk> hmm wonder if that is what i was hitting
[04:53:07] <jbk> are you using the latest ON build?
[04:53:10] <sstallion> cause the panic, and then issue ::msgbuf, and of course $c
[04:53:17] <jbk> or -kv
[04:54:45] <e^ipi> yeah, that's what I 'm doing now
[04:55:42] *** ludc has joined #opensolaris
[04:56:09] *** ludc has left #opensolaris
[04:56:25] *** ludc has joined #opensolaris
[04:57:29] *** ludc has left #opensolaris
[05:00:24] *** freakazoid0223 has quit IRC
[05:05:58] <e^ipi> damn you nv_sgp_check_set_cmn !!
[05:06:27] <sstallion> e^ipi: ?
[05:07:01] <e^ipi> that's where this stack trace ends and the panic calls begin
[05:07:16] *** donour has quit IRC
[05:07:18] <sstallion> can you rafb the entire backtrace ?
[05:07:20] <e^ipi> seems it pukes on writing
[05:07:33] <sstallion> also post the ::msgbuf output
[05:14:08] *** nachox has quit IRC
[05:16:43] <e^ipi> http://rafb.net/p/0zZekv68.html
[05:19:41] *** uebayasi has quit IRC
[05:21:20] *** uebayasi has joined #opensolaris
[05:21:46] *** ormandj has quit IRC
[05:22:54] *** erast has quit IRC
[05:23:15] *** erast has joined #opensolaris
[05:26:30] *** ninjaslim has quit IRC
[05:28:02] <evocallaghan> e^ipi:I think you could be hitting http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6747163
[05:28:24] <e^ipi> no, i'm not
[05:28:31] <e^ipi> for one, that was fixed
[05:28:43] <e^ipi> and for two, it's not on drive_connect / boot
[05:29:22] <evocallaghan> ah yes. Just had a closer look
[05:29:40] <evocallaghan> I did say *think* :p
[05:33:34] <sstallion> e^ipi: I'm not going to be much help on this one... jmcp might be though
[05:35:01] <e^ipi> he's vanished off the face of the earth
[05:36:52] <alanc> isn't australia the other end from the face?
[05:37:33] <e^ipi> heh
[05:38:16] <evocallaghan> :p
[05:38:28] <alanc> his calendar is marked off for vacation it hs weekthough
[05:38:49] <alanc> wow, I type bad when there's big lag in seeing the output
[05:39:16] <e^ipi> typing over SWAN?
[05:39:38] <alanc> no, typing while an app was starting and hosing my system i/o
[05:40:41] <evocallaghan> alanc:that's prob the best way to explain my intel card problems. Its like typing like that while the system is doing nothing but running firefox
[05:42:10] <alanc> i did just upgrade my sb2500 from nv_80 to nv_100 today - still getting used to the changes in gnome 2.24, firefox 3.0, etc.
[05:42:22] <alanc> am happy to have /usr/bin/xchat come as part of the install though
[05:43:00] <lesterc> guys - my test box core dumped when I tried to upgrade from nv94 to nv99 - can someone point me to the right docs to troubleshoot it further?
[05:43:06] <evocallaghan> That's a big rev jump alanc :D
[05:43:18] <evocallaghan> Are you using zfs root now on that ?
[05:43:24] <jbk> i do like the ssh key integration they did in b99
[05:43:31] <alanc> yeah, I had to clean off disk space because previous upgrade attempts ran out of space
[05:43:41] <alanc> haven't migrated to zfs root on here
[05:43:42] <jbk> and b99 doesn't seem to be a horrendously low a b98 in gnome
[05:43:49] <jbk> (not sure where the problem was with that)
[05:44:08] <alanc> tried that on my opteron box, but it failed to boot afterwards - need to try again (it's at least on nv_92 though)
[05:44:20] <lesterc> alanc:  is nv100 out? :)
[05:44:31] <alanc> and of course, all the fonts look different from nv80 8-)
[05:44:54] <alanc> lesterc: nope, it's only in right now (inside Sun)
[05:45:40] <alanc> I think they may be respinning nv100 before releasing it due to a compiler bug they found after the ON switch to Studio 12
[05:45:44] <evocallaghan> ON 100 is out though
[05:45:49] * sstallion falls over
[05:45:51] <sstallion> Oct  9 22:45:38 devel last message repeated 97598 times
[05:46:01] <alanc> oh yeah, GNOME spams syslog
[05:46:12] <alanc> that's supposed to be fixed in nv_101
[05:46:16] <sstallion> alanc: actually it was an interrupt handler i'm debugging :)
[05:46:27] <alanc> heh
[05:46:28] <sstallion> for some reason the chipset is going nuts
[05:46:30] <lesterc> alanc: right.
[05:46:31] <jbk> uugh compiler bug? that mean more patches to ss12?
[05:46:40] <evocallaghan> alanc:What's this problem in ss12 /
[05:46:45] <evocallaghan> s///?/
[05:47:47] *** sponix has joined #opensolaris
[05:49:03] <alanc> problem with the way arguments are put on the stack on x86 - breaks debugging with mdb
[05:49:18] <techqbert> Anybody in here run kernel samba server in a multiuser environment?  I can't find a meaningful tutorial or help in isolating user space.
[05:49:49] <evocallaghan> ah great ok
[05:49:53] <e^ipi> "isolating user space" ?
[05:49:54] <alanc> bug id is 6753543, but since it's a compiler bug, it won't show up on bugs.opensolaris.org
[05:50:27] <evocallaghan> I thought I could hit it in kde/mythtv/ardour work I am doing
[05:50:47] *** MrBIOS_ has joined #OpenSolaris
[05:50:49] <evocallaghan> although not much time for kde at the moment :p
[05:51:08] <evocallaghan> Just porting Jack to solaris today
[05:52:48] <_mary_kate_> techqbert: the kernel-mode CIFS server isn't samba
[05:53:14] <techqbert> mary_kate_: ah.  should I get rid of kernel server and just configure samba
[05:53:34] <_mary_kate_> er.. i don't think that's what i said.
[05:54:15] <techqbert> @e^ipi: by "isolating user space," I mean giving rw access to one pool for one user and no access to another pool and so on
[05:54:19] <sponix> saw someone post in #solaris that you can add devices to a zpool on the fly now, is that true... just zpool add <newdevice> is what they had listed, does that work ?
[05:54:35] <_mary_kate_> sponix: that's been possible forever
[05:55:11] <techqbert> _mary_kate_: certainly.  what would you recommend for a more versatile approach that gives certain users rw and others no access?
[05:55:17] <sponix> _really_ I knew you could replace a device on the fly, but didn't know about resizing the zpools, does zpool del <device> to shrink work as well ?
[05:56:03] <e^ipi> techqbert: so you mean ACL's
[05:56:12] <e^ipi> for that i would reccomend you use ACL's
[05:57:02] *** alanc is now known as alanc-away
[05:57:24] <_mary_kate_> sponix: no
[06:00:47] *** nprice has joined #opensolaris
[06:03:08] <e^ipi> damn, solaris doesn't like it when you try to migrate the root disk to another controller
[06:04:42] *** axisys has joined #opensolaris
[06:05:57] *** snakesqzns has quit IRC
[06:08:54] *** ShadowHntr has quit IRC
[06:13:22] *** erast has quit IRC
[06:16:02] *** ormandj has joined #opensolaris
[06:18:57] *** comay has joined #opensolaris
[06:19:12] *** ChanServ sets mode: +o comay
[06:19:22] *** iceD[R] has joined #opensolaris
[06:20:21] <iceD[R]> someone speak portuguese?
[06:20:22] *** ormandj has quit IRC
[06:20:57] *** iceD[R] has quit IRC
[06:21:36] *** ClayB_WFCS has joined #opensolaris
[06:23:03] *** not-me-guv has quit IRC
[06:28:21] *** RElling1 has joined #opensolaris
[06:30:42] *** plavcik has joined #opensolaris
[06:35:43] *** RElling has quit IRC
[06:38:26] *** ormandj has joined #opensolaris
[06:40:16] <_mary_kate_> e^ipi: need to edit bootenv.rc
[06:41:57] *** ZOP has quit IRC
[06:42:04] *** ZOP has joined #opensolaris
[06:46:24] <plavcik> I used zfs to export some datasets via nfs and cifs (smb), which docuemnt describe, how to configure and enable nfs and cifs daemons?
[06:47:16] <e^ipi> google://solaris+cifs
[06:48:13] <mukiex> How does Solaris compare to Linux in terms of CPU scheduling, e^ipi? (note : for desktop use, Linux' cpu scheduler SUCKS >_<)
[06:50:22] *** Triskelios has joined #opensolaris
[06:51:11] *** perlmongo has quit IRC
[06:51:37] <e^ipi> there's a half dozen schedulers that you can assign your projects to at run time
[06:53:56] <e^ipi> but i dunno, it's fine i guess
[06:54:53] <ClayB_WFCS> It takes a lot to starve a Solaris box, but I don't know much about either's behavior (or actual code) at the scheduler-level to speak to that.
[06:55:44] <_mary_kate_> solaris has the IA scheduler, not sure linux does that..
[06:55:54] <plavcik> how to display setting for status of some (all) serverices (someting like svcadm status nfs/server)
[06:55:59] <ClayB_WFCS> The real time scheduler is pretty unique
[06:57:09] *** ormandj has quit IRC
[06:57:27] <e^ipi> plavcik: svcs
[06:57:32] <e^ipi> read the man page
[06:57:44] *** freakazoid0223 has joined #opensolaris
[06:57:47] *** ormandj has joined #opensolaris
[06:58:33] <plavcik> thx
[06:59:29] <CIA-25> Vikram Hegde <Vikram.Hegde at Sun dot COM>: 6757930 GRUB cap file version number must be bumped up to track zpool changes
[07:16:50] *** jgracin has joined #opensolaris
[07:17:13] *** erast has joined #opensolaris
[07:20:40] *** piwi has quit IRC
[07:21:37] *** anilg has joined #opensolaris
[07:24:45] *** plavcik has quit IRC
[07:32:15] *** tombhadAC has quit IRC
[07:32:52] *** tsoome has quit IRC
[07:34:56] *** dclarke has quit IRC
[07:36:57] *** ClayB_WFCS has quit IRC
[07:42:24] *** r_nd_ has joined #opensolaris
[07:46:01] *** r_nd has quit IRC
[07:46:42] *** mukiex has left #opensolaris
[07:47:24] *** dunc has quit IRC
[07:51:05] *** ClayB_WFCS has joined #opensolaris
[07:52:43] *** sponix has quit IRC
[07:54:24] *** mikea has joined #opensolaris
[07:54:53] *** moazamraja has joined #opensolaris
[07:55:09] <moazamraja> anyone have a Lenovo IdeaPad Y510 here?
[07:55:23] <moazamraja> installed opensolaris, doesn't seem to detect the broadcom ethernet
[07:56:30] <ClayB_WFCS> Moaza May need the BCME driver from Broadcom
[07:57:02] <ClayB_WFCS> It's free and a Google search away, otherwise, you may want to check bugs.opensolaris.org to see if it's a newer one under development and not yet released.
[07:57:14] <ClayB_WFCS> (Sadly I don't think BCME is open source.)
[07:59:53] *** RElling has joined #opensolaris
[08:03:56] <e^ipi> auto-snap would be less annoying if it didn't try to snapshot my swap device...
[08:03:57] <CosmicDJ> moooo'ning
[08:04:01] <CosmicDJ> wc
[08:04:05] *** jbasse has joined #opensolaris
[08:04:16] <e^ipi> i don't really need a snapshot of swap
[08:04:19] <e^ipi> it's not very helpful
[08:05:28] <Triskelios> I don't recall it selecting any filesystems by default
[08:05:47] <e^ipi> i've got a whole lot of spew that contradicts that
[08:07:38] *** RElling1 has quit IRC
[08:10:45] <Triskelios> moazamraja: the builtin bge driver seems to refer to it, maybe you just have a different PCI ID?
[08:11:09] <Triskelios> if this is a 5906/5906M according to the reviews of that laptop
[08:13:37] *** yongsun has joined #opensolaris
[08:15:38] <Triskelios> moazamraja: ah, see http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6603710
[08:15:58] *** sophokles has joined #opensolaris
[08:16:57] <_mary_kate_> how do i change the mode of the .zfs directory?
[08:19:08] *** the_unmaker1 has joined #opensolaris
[08:19:17] <the_unmaker1> uh
[08:19:35] <the_unmaker1> is 2008.11 opensolaris the one by ian?
[08:19:58] <_mary_kate_> there is no "one by ian"
[08:20:12] <_mary_kate_> since starting at sun, he has had very little to do with (open)solaris
[08:20:20] <ClayB_WFCS> It involves a couple more hundreds or thousand people than that
[08:20:36] *** div8 has joined #opensolaris
[08:21:28] *** twisti is now known as twisti_home
[08:22:30] <CosmicDJ> _mary_kate_: man zfs, search for snapdir
[08:22:45] <_mary_kate_> CosmicDJ: that doesn't seem to change the mode, it just makes it visible or not in 'ls'
[08:22:59] <CosmicDJ> ah hm...
[08:27:11] <_mary_kate_> hmm, nicstat seems confused
[08:27:21] <_mary_kate_> it reports 21MB/s read, 33MB/sec write as 44% utilisation of a GE nic
[08:27:37] *** mikl has joined #opensolaris
[08:28:22] <CosmicDJ> so you think your gig nic can do 1000MB/s?
[08:28:57] <_mary_kate_> it can do 125MB/sec in each direction, so 250MB/sec.  21+33=54MB/sec.  seems more like 20%
[08:29:35] <moazamraja> phuckin card is a Broadcom 5906M
[08:31:11] <_mary_kate_> i guess it's adding the send+recv and comparing that against 125MB/sec total
[08:31:48] <_mary_kate_> (which yields a figure of around 44%)
[08:32:38] *** mikea has quit IRC
[08:35:02] *** comay has quit IRC
[08:38:42] <CosmicDJ> _mary_kate_: btw, did you ever play with jumbo packets?
[08:39:30] <CosmicDJ> s/packets/frames/
[08:42:19] <_mary_kate_> no, but on my todo list is seeing if this iscsi array can do jumbo frames, in which case i will
[08:42:19] <the_unmaker1> I love shemale porn
[08:42:48] *** sophokles has quit IRC
[08:42:59] *** gerard13 has joined #opensolaris
[08:43:16] * _mary_kate_ mailed the nicstat author.. also suggested integrating it, it's way more useful than half the stuff that's been integrated recently
[08:44:26] *** sophokles has joined #opensolaris
[08:44:43] *** sophokles has left #opensolaris
[08:45:25] *** perlmongo has joined #opensolaris
[08:45:35] *** carl- has joined #opensolaris
[08:45:44] <_mary_kate_> hm, it has 'jumbo frames [enable/disable]'
[08:45:50] <_mary_kate_> i assume 'enable' must mean 9k frames
[08:46:22] *** jbasse has quit IRC
[08:46:47] *** StayTuned has quit IRC
[08:49:30] <CosmicDJ> rKB/s = 45305.4 and wKB/s = 116290  ==  100 %Util on a e1000g...
[08:50:17] <_mary_kate_> i believe it's limiting the result at 100% (i looked at the source, it does that)
[08:50:37] <_mary_kate_> using the logic nicstat uses, you would have 126% util otherwise
[08:52:48] *** the_unmaker1 has quit IRC
[08:53:25] <trochej> Coffee
[08:55:28] *** mikefut has joined #opensolaris
[08:58:00] <evocallaghan> Oh yes please !
[08:58:08] *** juriskr has joined #opensolaris
[09:01:24] *** axisys has quit IRC
[09:08:39] *** tsoome has joined #opensolaris
[09:10:10] *** DTEIT has joined #opensolaris
[09:10:36] <DTEIT> morning
[09:12:03] *** yongsun has quit IRC
[09:12:33] <moazamraja> greeeeat.
[09:12:40] *** ormandj has quit IRC
[09:13:14] <moazamraja> right after I rip a b99 DVD, I find out that b100 will have support for the BCM5906 NIC in my machine :P
[09:13:31] <moazamraja> (i'm assuming snv_100 is the same as b99)
[09:13:54] <CosmicDJ> http://blogs.sun.com/chriszhu/entry/find_and_try_faster_ruby
[09:14:21] <ClayB_WFCS> Moazamraja NV100 == B100
[09:15:17] <moazamraja> ok
[09:15:19] <moazamraja> soooo
[09:15:28] <moazamraja> any idea when b100 is going to be released?
[09:15:41] <moazamraja> or am I going to have to go with b99 and do some sort of manual bfu update to get the driver?
[09:15:41] <_mary_kate_> two weeks after 99
[09:15:45] <ClayB_WFCS> Moazamraja, builds come out every two weeks
[09:15:48] <moazamraja> ah
[09:15:50] <moazamraja> ok
[09:15:53] <ClayB_WFCS> Soon :)
[09:16:40] <moazamraja> b99 was released when ?
[09:17:12] <trochej> Wow, on 100?
[09:18:00] <_mary_kate_> http://opensolaris.org/os/community/on/schedule/
[09:18:22] <_mary_kate_> add 10 days to 'delivery to WOS' to get the public release date
[09:18:29] <moazamraja> _mary_kate_: perfect, thx
[09:18:37] <_mary_kate_> hm, actually, you need to add another two weeks to that for QA, i think
[09:18:42] *** ludc has joined #opensolaris
[09:18:56] <moazamraja> _mary_kate_: so installing b99 + bfu to onnv100 would work though...right?
[09:19:06] <_mary_kate_> yes, but then you can never upgrade again
[09:19:16] <ludc> hi...exists any mp3 player in repository of opensolaris?
[09:19:19] <moazamraja> i can bfu again tho, can't I?
[09:19:37] <_mary_kate_> yes, but BFU only upgrades ON, not the rest of the system.  so it's not ideal for everyone
[09:19:59] <ClayB_WFCS> Ludc Not in a Sun.COM repo due to licensing. Blastwave may. Fluendo used to have a free "purchase" to get the Gstreamer MP3 plugin
[09:20:10] <_mary_kate_> used to?  did they remove it?
[09:20:26] <moazamraja> _mary_kate_: ok
[09:20:33] <ClayB_WFCS> Mary I just haven't been there for a while, so I can't attest to how new it is.
[09:20:40] <ClayB_WFCS> Or if it moved.
[09:20:48] <moazamraja> worse case, I'll have to wait 2-3 weeks to get this laptop on OpenSolaris
[09:21:57] <ClayB_WFCS> Moazamraja Build 100 is going to be very, very sweet for all sorts of stuff. Suspend to RAM, universal AudioHD, many, many driver fixes, a better FireFox 3 (though that may be 99). It'll be cool!
[09:22:30] *** pumpkin- has quit IRC
[09:24:11] <moazamraja> ClayB_WFCS: sweet. my spare Lenovo IdeaPad Y510 awaits it
[09:24:28] <moazamraja> now I wonder if I can hack this machine to accept more than 3-4GB of RAM, that'd be nice
[09:24:49] <ClayB_WFCS> Yeah, I'm trying to wait patiently. I'm on the caiman team so I have to wait for the full OpenSolaris build opposed to the SXCE build to image-update to it.
[09:25:24] <_mary_kate_> moazamraja: you might be able to take the driver from 100 and install it on 99
[09:25:29] <trochej> ClayB_WFCS: Does caiman have it's irc # ?
[09:25:35] <_mary_kate_> moazamraja: if the driver doesn't depend on new interfaces or bug fixes introduced in 100, it should work
[09:26:23] <ClayB_WFCS> Yeah #caiman-discuss, we're trying to hang out there more, but it's been pretty quiet.
[09:27:08] <moazamraja> _mary_kate_: i'll have to read up on how to do that, I'm not sure where the b100 driver files will be for me the transplant them
[09:27:13] *** LuckyLuke has joined #opensolaris
[09:27:42] <_mary_kate_> moazamraja: it'll be in /kernel/drv/{amd64,}/<name>
[09:27:55] <_mary_kate_> moazamraja: copy it to the same location on your system (use amd64 for 64-bit systems) and run add_drv
[09:28:02] <_mary_kate_> (or if you're replacing an existing driver, just reboot)
[09:30:10] *** xRaich[o]2x has joined #opensolaris
[09:34:02] *** kokoko1 has joined #opensolaris
[09:34:23] *** the_unmaker1 has joined #opensolaris
[09:34:35] <kokoko1> hi guys, do we have a native package of apache2 ?
[09:34:44] <_mary_kate_> /usr/apache2/2.2/
[09:35:20] <moazamraja> _mary_kate_: cool! I'm installing b99 now. Will try manually adding the b100 driver in the morning
[09:35:47] <kokoko1> _mary_kate_, how did you install it imean what's the package name?
[09:38:52] *** jklyekai is now known as likaijun
[09:40:14] <moazamraja> broadcom drivers are bnx? not bge?
[09:40:27] <CosmicDJ> kokoko1: http://blogs.warwick.ac.uk/chrismay/entry/opensolaris_adventure_part_1_2/
[09:40:28] <ClayB_WFCS> KoKoKo1 pkg:/SUNWapch22
[09:40:47] <kokoko1> thanks guys
[09:41:58] *** ludc has left #opensolaris
[09:42:11] <kokoko1> heh i was search without -r :s
[09:42:18] <kokoko1> pkg search apache
[09:42:33] <ClayB_WFCS> :)
[09:42:50] <kokoko1> gtg, see ya
[09:42:59] * kokoko1 disappears
[09:45:52] *** minidev has joined #opensolaris
[09:49:05] *** hanthana has joined #opensolaris
[09:50:58] *** Triskelios has quit IRC
[09:51:26] *** lesterc has quit IRC
[09:51:37] *** luc^ has joined #opensolaris
[09:51:39] *** ClayB_WFCS has quit IRC
[10:01:59] *** hanthana has quit IRC
[10:12:35] <_mary_kate_> i'm confused about how to enable jumbo frames for bge(7d).  it says to set default_mtu=9000 in bge.conf, but what if i want bge0 to use 1500 MTU, and bge1 to use 9K?  can i reduce the mtu in /etc/hostname.bge0?
[10:15:41] <CosmicDJ> _mary_kate_: the ifconfig manpage lists an mtu option...
[10:16:22] *** loquitus_of_bor1 has joined #opensolaris
[10:16:34] <loquitus_of_bor1> Hello?
[10:17:37] <loquitus_of_bor1> Can somebody tell me what gcc compiler version they have?
[10:18:04] <CosmicDJ> 3.4.3
[10:18:25] *** MattMan has joined #opensolaris
[10:18:30] <evocallaghan> gcc (GCC) 3.4.3
[10:18:34] <evocallaghan> and 4.2
[10:20:03] <_mary_kate_> CosmicDJ: yes, but i can't set it to >1500.  ifconfig: setifmtu: SIOCSLIFMTU: bge0: Invalid argument
[10:20:15] <_mary_kate_> CosmicDJ: i assume bge(7d) describes using bge.conf to enable it because that's the only way..
[10:22:40] <mui> does frequent snapshotting cause performance issues in say when sharing volume over iscsi
[10:22:45] *** smtms has quit IRC
[10:22:54] <mui> eg. does it matter if I snapshot every 5sec vs. every 120sec
[10:24:39] <trochej> mui: I'd say test it, but I
[10:24:46] *** not-me-guv has joined #opensolaris
[10:24:47] <trochej> mui: I'd say test it, but I think it wouldn't be aproblem
[10:25:06] <mui> hm
[10:25:12] <CosmicDJ> _mary_kate_: and your bge supports jumbo frames? (the bge manpages lists them)
[10:25:33] <CosmicDJ> mui: you'll just run out of diskspace sometime ;)
[10:25:47] <trochej> mui: What *may* be a problem, is sheer number of shapshots you'll have to manage
[10:25:51] <trochej> And the diskspace
[10:26:09] <mui> actually I'm just doing snapshot, sending it incrementally to remote location, removing old one and starting from beginning
[10:26:30] <trochej> Oh, then every 5 seconds may be a problem. :)
[10:26:52] <Stric> mui: Try it and see for yourself if the performance hit is a problem.
[10:26:52] <trochej> Depends on the amount of changes your snapshot reference
[10:26:54] <mui> depends, If I do it in fast pace theres less data to be transferred
[10:26:59] *** jhfisc has quit IRC
[10:27:18] <Stric> mui: a snapshot etc requires a transaction to be committed, so it might disturb the data flow
[10:27:28] <trochej> mui: The more busy fs, the bigger snapshots and highet i/o rate and al
[10:27:34] *** xRaich[o]2x has quit IRC
[10:27:39] <Stric> Best way is to find out how it affects _your_ setup
[10:27:39] <mui> yeah
[10:27:46] <mui> I was just wondering
[10:27:47] *** Cass has quit IRC
[10:28:15] <mui> got pretty cool setup anyway, too solaris boxes sharing iscsi to xenserver hosts
[10:28:37] <mui> and planning to snapshot one to another
[10:28:49] <mui> so in case of disaster I can just shareiscsi=on from another solaris
[10:28:53] <CosmicDJ> mui: you can remote mirroring/replicat. with sun's availability suite (AVS)
[10:28:56] <CosmicDJ> mui: http://www.c0t0d0s0.org/archives/4351-Less-known-Solaris-Features-Point-in-time-copy-with-AVS.html
[10:29:17] <_mary_kate_> CosmicDJ: yes, it's a V20z, BCM5704C
[10:29:28] <trochej> mui: I think your solution was tested by a friend of mine and he wasn't satisfied.
[10:29:34] <trochej> Can't remember why, now
[10:30:14] <CosmicDJ> _mary_kate_: hm maybe you have to do some ndd fun like in http://prefetch.net/blog/index.php/2005/10/16/enabling-jumbo-frames-with-sun-solaris-and-cisco-catos/
[10:30:46] <_mary_kate_> CosmicDJ: well, i'm assuming the manual page describes the correct method
[10:30:50] <mui> iscsi works very well with xenserver
[10:30:55] <_mary_kate_> CosmicDJ: i'm just not sure how to apply it to just a single interface
[10:31:00] <mui> and snapshots and rollbacking works like they should
[10:31:53] <mui> solaris just announces broken scsiid to xenhost, but when you understand the problem you can pretty much live with it
[10:32:11] <mui> afaik, its fixed in opensolaris builds already
[10:32:32] <_mary_kate_> nge.conf has an example of how to apply it to just one card, but i'm not sure that's a general method, or something specific to nge
[10:35:31] *** Rarok has joined #opensolaris
[10:35:54] <the_unmaker1> standard ml
[10:36:56] *** anilg has quit IRC
[10:39:41] *** twisti_work is now known as twisti
[10:40:20] *** maxote has quit IRC
[10:43:41] *** Raroko-chan has joined #opensolaris
[10:44:26] *** Rarok has quit IRC
[10:53:25] *** maxote has joined #opensolaris
[10:55:12] *** the_unmaker1 has quit IRC
[10:55:23] *** KOHJU is now known as kohju
[10:58:35] *** MrBIOS_ has quit IRC
[11:03:04] *** Erwann has joined #opensolaris
[11:06:41] <trochej> Coffee
[11:08:23] *** coffman has joined #opensolaris
[11:12:40] *** timsf has joined #opensolaris
[11:13:40] <trochej> timsf: Coffee?
[11:14:07] *** ormandj has joined #opensolaris
[11:15:14] <timsf> Good idea - definitely need it
[11:16:14] *** kohju is now known as KOHJU
[11:16:18] *** KOHJU is now known as kohju
[11:16:48] *** dunc has joined #opensolaris
[11:17:10] <timsf> how're the plans for the conference going trochej?
[11:17:15] <timsf> Got much attendance?
[11:18:02] <_mary_kate_> CosmicDJ: btw, i won't be able to reboot this system until monday, but i'
[11:18:07] <moazamraja> guess i'll have to wait for b100
[11:18:07] <_mary_kate_> ll let you know how it goes with jumbo frames
[11:18:25] *** e^ipi_ has joined #opensolaris
[11:18:35] <moazamraja> i slapped the bnx driver from onnv100 /kernel/drv/bnx into my b99 build, no go :/
[11:18:47] <_mary_kate_> moazamraja: what happened?
[11:19:00] <moazamraja> on boot,    bnx0: Failed to memory map device
[11:19:02] *** xRaich[o]2x has joined #opensolaris
[11:19:35] *** vmlemon_ has joined #opensolaris
[11:19:45] *** Dar has joined #opensolaris
[11:20:12] *** e^ipi_ has quit IRC
[11:20:31] *** e^ipi has quit IRC
[11:20:55] *** e^ipi has joined #opensolaris
[11:20:56] *** ChanServ sets mode: +o e^ipi
[11:21:44] <moazamraja> i also noticed that the first time i run 'ifconfig -a' i get a nice fat core
[11:22:30] <moazamraja> altho, the core is from gnome-terminal :/
[11:23:09] <moazamraja>  ifconfig bnx plumb
[11:23:09] <moazamraja> ifconfig: cannot open link "bnx": invalid DLPI linkname
[11:24:13] *** RElling1 has joined #opensolaris
[11:25:47] <CosmicDJ> moazamraja: you cp'ed a b100 kernel module to a b99 system?
[11:30:03] <moazamraja> yeah :)
[11:30:16] <moazamraja> at 2am, why not.
[11:31:42] *** RElling has quit IRC
[11:32:38] <CosmicDJ> cause there's only backward binary compatibility; and not foward :)
[11:32:44] <CosmicDJ> +r
[11:32:45] *** kohju is now known as KOHJU
[11:33:24] <e^ipi> =quit
[11:33:28] *** e^ipi has quit IRC
[11:33:40] *** [1]kimc has joined #opensolaris
[11:34:12] *** likaijun has quit IRC
[11:34:18] *** likaijun has joined #opensolaris
[11:35:04] *** e^ipi has joined #opensolaris
[11:35:04] *** ChanServ sets mode: +o e^ipi
[11:36:02] *** dunc has quit IRC
[11:40:22] *** bnitz has joined #opensolaris
[11:40:38] *** e^ipi has quit IRC
[11:41:35] *** e^ipi has joined #opensolaris
[11:41:35] *** ChanServ sets mode: +o e^ipi
[11:47:55] <TomJ> Does Azureus still need messing about with to work on SXCE, due to Motif?
[11:48:36] <oxygene> it still needs messing around to get SWT to work, I think. ask ibm when they finally really fix swt builds on solaris (esp. solaris/x86, solaris/sparc is somewhat supported)
[11:48:55] <TomJ> ok thanks
[11:49:32] *** kimc has quit IRC
[11:49:32] *** [1]kimc is now known as kimc
[11:50:20] *** not-me-guv has quit IRC
[11:55:30] *** derchris^ has quit IRC
[11:55:34] *** derchris has joined #opensolaris
[11:56:39] *** mbz has joined #opensolaris
[12:00:04] *** not-me-guv has joined #opensolaris
[12:00:11] *** smtms has joined #opensolaris
[12:02:16] *** bnitz1 has joined #opensolaris
[12:10:35] *** bara|noid has joined #opensolaris
[12:11:30] *** bnitz has quit IRC
[12:12:12] *** MeP3aBeu has joined #opensolaris
[12:13:39] *** tomj_ has joined #opensolaris
[12:16:26] *** xtrondo has joined #opensolaris
[12:20:04] *** tomj__ has joined #opensolaris
[12:20:51] *** mbz has quit IRC
[12:20:56] *** TomJ has quit IRC
[12:23:59] *** Raroko-chan has quit IRC
[12:25:12] *** stux is now known as stux|work
[12:25:14] *** tomj__ has quit IRC
[12:25:29] *** tomj__ has joined #opensolaris
[12:27:20] *** Dar has quit IRC
[12:30:36] *** Dar has joined #opensolaris
[12:31:35] *** calumb has joined #opensolaris
[12:31:52] *** mikl has quit IRC
[12:34:52] *** cypromis has joined #opensolaris
[12:36:16] *** mikl has joined #opensolaris
[12:36:23] *** anilg has joined #opensolaris
[12:37:37] *** tomj_ has quit IRC
[12:37:52] *** tomj_ has joined #opensolaris
[12:39:10] *** zapata has joined #opensolaris
[12:41:57] *** _teo_ has joined #opensolaris
[12:44:40] *** tomj__ has quit IRC
[12:45:13] *** tomj__ has joined #opensolaris
[12:45:30] *** tomj__ has quit IRC
[12:45:44] *** TomJ has joined #opensolaris
[12:51:31] <TomJ> Got a networking issue.. SXCEb99 is my LAN router/NAT/firewall using PPPoE.  It runs dns/server.  LAN client (Linux) has private IP of SXCE as only entry in /etc/resolv.conf.   Looking up a hostname from the SXCE box itself, e.g.:  host www.blah.com  - takes <0.5seconds.  Running host blah.com from the Linux client takes many seconds, and occasionally times out completely
[12:51:59] <TomJ> oh wait
[12:52:05] <TomJ> SXCE host isn't using itself in resolv.conf
[12:54:48] *** h3sp4wn has joined #opensolaris
[12:57:56] *** tomj_ has quit IRC
[12:58:22] *** KOHJU is now known as kohju
[12:58:41] <codestr0m> good afternoon.. if any porters are up.. line 231: too many struct/union initializers  http://rafb.net/p/sLwYgu27.html Looks like something lint would catch, but sun cc doesn't like it
[13:00:35] *** m0zzzy has quit IRC
[13:01:19] *** LordIllpalazo has joined #opensolaris
[13:04:27] *** calumb has quit IRC
[13:09:07] *** loquitus_of_bor1 has quit IRC
[13:09:25] *** kohju is now known as KOHJU
[13:18:33] <glance> anyone know why u6 was moved from mid-october to end-of-october?
[13:24:49] <tijo007> please, how do i create new pooladm.conf if file is missing?
[13:25:36] *** sophokles has joined #opensolaris
[13:26:09] *** sophokles has left #opensolaris
[13:29:21] <CosmicDJ> tijo007: man pooladm (I think you want the -c option)
[13:30:10] <tijo007> Dominic: yep, done. i forgot to clear the maintenance state and run 'pooladm -s /etc/pooladm.conf'
[13:30:29] <CosmicDJ> glance: some folks even say they'll release u6 not until november...
[13:31:21] *** LordIllpalazoThe has joined #opensolaris
[13:31:35] *** anilg has quit IRC
[13:31:36] <glance> okay? do you know what the blocker is?
[13:32:40] *** LordIllpalazo has quit IRC
[13:32:55] *** dunc_ has joined #opensolaris
[13:33:34] <CosmicDJ> glance: no idea
[13:37:07] *** jgracin_ has joined #opensolaris
[13:37:14] *** heB_z0rLDree has joined #opensolaris
[13:37:48] *** LordIllpalazoThe has quit IRC
[13:46:51] *** jgracin has quit IRC
[13:50:07] *** Odin- has joined #opensolaris
[13:50:29] *** sipior has joined #opensolaris
[13:50:32] *** TomJ has quit IRC
[13:56:32] <CosmicDJ> wow, openbsd 4.4 runs on us-iii, us-iv, t1, t2, sparc64-v/vi/vii; nice work
[13:56:40] <CosmicDJ> http://www.de.openbsd.org/sparc64.html
[13:57:06] <kokoko1> heh, pkg image-update work cool
[13:59:06] <kokoko1> it would be nice if they add option to pkg so that it start from where it was last disconnected instead of starting all over again when downloading hundreds of packages
[13:59:08] <kokoko1> :s
[14:00:01] <kokoko1> uname -a
[14:00:10] <kokoko1> SunOS opensolaris 5.11 snv_86 i86pc i386 i86pc
[14:00:53] <th> i hot-plugged a sata disk on supermicro marvel. how would i get my scsi device back after plugging in?
[14:01:00] <th> something like volcheck for scsi devices?
[14:01:24] <th> its not listed in `format`
[14:01:59] <evocallaghan> kokoko1:It does now.. update to 98
[14:02:26] <evocallaghan> kokoko1:first update pkg though
[14:02:34] <kokoko1> and when doing pkg image-update it again showing me 551 packages?
[14:02:49] <kokoko1> evocallaghan, it mean one more upgrade is required?
[14:03:21] <evocallaghan> kokoko1:follow my blog or the release notes for 98
[14:03:29] <kokoko1> link?
[14:03:45] <CosmicDJ> th: man cfgadm
[14:04:03] <th> CosmicDJ: thanks
[14:05:17] <evocallaghan> kokoko1: Still applyshttp://ultravioletos.blogspot.com/2008/06/upgrading-opensolaris200805-to-snv91.html
[14:05:29] <kokoko1> thx
[14:05:35] <CosmicDJ> th: smth like cfgadm -c configure c0::dsk/c0t0d0
[14:05:39] <evocallaghan> np, change SUNWipkg at 0 dot 5.11,5.11-0.90 to SUNWipkg at 0 dot 5.11,5.11-0.98
[14:06:34] <kokoko1> evocallaghan, by change you mean upgrade using pkg image-update right?
[14:06:45] <evocallaghan> Just read the blog
[14:06:53] <evocallaghan> It should be clear whats going on
[14:07:04] <kokoko1> right, one more thing
[14:07:22] <kokoko1> does the pkg image-updateg delete the old packages
[14:07:23] <kokoko1> ?
[14:07:46] <kokoko1> coz my virtual harddrive showing 2 gb increase after the image-uprade
[14:08:17] <timsf> Nope, doesn't delete the older boot environment,
[14:08:23] <_mary_kate_> th: cfgadm -al; cfgadm -c sataX/Y configure
[14:08:23] <timsf> however it is a space-efficient upgrade,
[14:08:35] <th> _mary_kate_: yes - thanks. i just figured.
[14:08:40] <timsf> in that it uses zfs clones to reference the older blocks on disk if they haven't changed.
[14:09:20] <timsf> I've done image-updates which only added a few hundred mb of new stuff.
[14:09:22] <th> _mary_kate_: but it's -c configure sataX/Y    actually
[14:11:58] <evocallaghan> Any good with configure.ac and SS12. - http://rafb.net/p/X8PEoJ15.html ; I need to wrap this around with a statment that tests for gcc and then test for SS with the two defines fixed with the correct flags. I don't you what you do instead of -Wall ?
[14:13:35] <kokoko1> hmm
[14:14:27] <CosmicDJ> evocallaghan: you can drop -Wall if you're not interested in fixing compiler warnings...
[14:15:14] <kokoko1> damn FAT32 ihave to move .vdi (harddisk image of os) to ext3 after virtuabox refuses to start guest when it exceeded the fat32 filesize limit (4gb)
[14:16:00] <CosmicDJ> evocallaghan: -O is -xO3 on Solaris IIRC; I'd drop "-ffast-math -funroll-loops -fprefetch-loop-arrays" as well
[14:16:28] <kokoko1> tomorrow i'll update to _98, coz i do not have fast internet at home only (512kbps) and at work i join fat pipe
[14:16:28] <CosmicDJ> s/Solaris/SS12/
[14:16:55] <kokoko1> s/join/enjoy
[14:20:16] *** jgracin_ has quit IRC
[14:22:24] <evocallaghan> CosmicDJ:thanks man
[14:29:04] *** sponix has joined #opensolaris
[14:31:32] <CosmicDJ> glance: btw, there's a "What's new in u6" show in secondlife in about 4 hrs AFAIK; why don't you ask there? ;)
[14:32:23] *** kokoko1 has quit IRC
[14:33:54] <glance> secondlife? i don't touch that thing, even with a long stick =)
[14:35:12] <CosmicDJ> same here :)
[14:37:21] <oxygene> secondlife.. and how are you supposed to login there as all-sun customer?
[14:37:44] <evocallaghan> haha wtf
[14:38:05] <evocallaghan> sun promotes serial killers ? :p
[14:40:50] * evocallaghan head hits desk
[14:40:58] <evocallaghan> Night lads
[14:42:07] <CosmicDJ> http://blogs.sun.com/solariscampus/entry/what_s_new_in_solaris
[14:45:43] *** letz_ has joined #opensolaris
[14:47:23] <sponix> evocallaghan:  you decent with Windows and Solaris Integration ? I'm just looking for a simple backup solution, to take XP client laptops, and backup their "My Documents" and PST (outlook email) to a Solaris Box
[14:48:53] *** Odin- has quit IRC
[14:49:20] <CosmicDJ> sponix: configure samba for domain logons, all there Documents and Settings files will be on the solaris box then
[14:49:30] <CosmicDJ> s/there/their/
[14:51:09] <codestr0m> CosmicDJ: 2ndlife doesn't run on solaris does it? (someone would have to get mono to compile)
[14:53:35] <CosmicDJ> codestr0m: no idea because I wouldn't touch it, too
[14:53:58] <codestr0m> mono or 2ndlife?
[14:54:43] <evocallaghan> sponix:yea, Its what we do here
[14:54:54] <CosmicDJ> 2ndlife
[14:55:09] <evocallaghan> sponix:Just make a few zfs FS's and share them with CIFS
[14:55:51] <evocallaghan> sponix:Then setup the age old windows backup tool to backup to them over the network. That is the easiest solution
[14:56:09] *** not-me-guv has quit IRC
[14:56:30] <evocallaghan> sponix:I put some notes on my blog already but your need 820-2429.pdf from docs.sun.com
[14:57:13] <sponix> evocallaghan:  OK, I can tell Solaris that easy, but it is the whole picture I'm having trouble with.. Can you give me a quick step by step... domain is called NASE that I am working with
[14:58:26] <sponix> But I have little to no understanding of Windows AD/domains/permissions, so unattended backups are out of my scope at this point, even from one windows machine to another
[14:59:50] <evocallaghan> Sorry, i'm about to go to bed
[15:00:23] <evocallaghan> sponix:Read _though_ that pdf I recommend from docs.sun.com
[15:01:09] <sponix> Thanks for the Documentation Link, I'll have a look, any knowledge is better than none ;)
[15:01:36] <sponix> Windows World is killing me... Sucking my will to live
[15:01:51] * evocallaghan falls over
[15:02:04] * evocallaghan &
[15:02:46] <CosmicDJ> sponix: you already have an AD controller?
[15:03:05] <CosmicDJ> in your network?
[15:03:18] <sponix> Yes, there is a Windows Server 200* AD Server in place already
[15:03:48] <CosmicDJ> sponix: good, then join it and tell the AD admin to place all home directories on your sun box :)
[15:05:05] <sponix> If only life was that simple ;)
[15:05:40] *** MattMan is now known as MattAFC
[15:05:43] <CosmicDJ> you wanted a quick step by step :p
[15:06:23] <sponix> that exact setup with Roaming profiles for users would be slick
[15:06:33] *** CIA-25 has quit IRC
[15:08:10] <sponix> Strange question...
[15:09:43] *** dunc_ has quit IRC
[15:09:48] <sponix> Will a setup like this work.... 147 networked Windows XP box gets users data pushed to it, and a cross over cable on a separate 192 local NIC goes to a standalone Solaris box to pickup the files for storage ?
[15:11:34] *** heB_z0rLDree has quit IRC
[15:11:42] <sponix> Basically running the Solaris box like a Local NAS drive to the XP box
[15:13:53] <CosmicDJ> sponix: just one box? you know that you'll have 147 angry users with forks and torches in front of your door when that box breaks...
[15:14:55] <sponix> CosmicDJ:  1 Backup box is better than 0
[15:15:19] *** rodney_h has joined #opensolaris
[15:15:42] *** zapata has quit IRC
[15:15:46] <CosmicDJ> sponix: ah this is the backup box? and the user directories should remain on the ad server?
[15:17:15] <sponix> CosmicDJ:  I'm not up to par on MS stuff at all, but I _think_ the auth is coming from AD, and the users still have their profiles/directories right on their laptops at this point
[15:17:34] <h3sp4wn> sponix: Why do you not just join the solaris box to the active directory ? with smbadm and install the ad integration for unix
[15:17:47] <sponix> I'm just looking to backup their local work: "My Documents", and "PST" file(s) to an outside system in case their laptops fail
[15:17:57] <rodney_h> Hello, I need a little help! I recently ordered the free opensolaris cd from sun microsystems by mail and recieved it yesterday but when I try to boot my machine from the cd all it will do is load a lot of text lines upto a certain point and stop when it stops it says Console login service(s) cannot run.?
[15:19:38] <sponix> h3sp4wn:  2 reasons: 1) I don't know what the hell I'm doing (yet), and 2) the reason I am looking to Run the Solaris box on a local 192 hooked by cross over cable to the windows machine is attempting to get a Solaris box accredited for this network is a nightmare from what I'm told
[15:19:39] <CosmicDJ> rodney_h: try the 2008.11 development build from http://www.opensolaris.org/os/downloads/
[15:21:31] <rodney_h> CosmicDJ...I'm not sure what I got by mail, but figured it would be latest version?
[15:22:32] *** Yorlik has joined #opensolaris
[15:22:48] <CosmicDJ> rodney_h: it *is* the latest "stable" and somewhat supported version
[15:23:23] *** yarihm has joined #opensolaris
[15:23:28] <Yorlik> Do I see right, that mcrypt for php is not standard in SXCE98 ?
[15:24:02] <_mary_kate_> probably not, as that would need for mcrypt to be integrated as well
[15:24:25] <oxygene> uh, is php standard in sxce?
[15:24:39] <Yorlik> its included, yes
[15:25:16] <Yorlik> Seems i need to compile from new as well ...
[15:25:39] *** CIA-25 has joined #opensolaris
[15:25:47] <rodney_h> CosmicDJ: I am trying to run this simultaneous on my Windows Vista Machine with a separate partition for each, a few of my systems specs are Gateway Intel Core 2 Quad model GM5478.
[15:26:06] <oxygene> given how that's the most likely answer to anything php related, that php package is probably a checkbox item for linux immigrants ("does it have php? check" "do I have to recompile it? uh, didn't ask that")
[15:26:41] <rodney_h> CosmicDJ: I'm pretty new to other OS's!
[15:26:43] <CosmicDJ> rodney_h: fine, but you said it doesn't work; thats why I said: try the development version
[15:28:45] <rodney_h> CosmicDJ: Well, yes, that is what I said but not exactly true...what I meant was it will not complete startup on initial boot trying to run as livecd!
[15:29:18] *** yarihm has quit IRC
[15:31:09] <rodney_h> CosmicDJ: It is an original disc from Sun MicroSystems! Question...is it something wrong with disc, my machine, or perhaps something I'm doing wrong?
[15:31:48] <CosmicDJ> rodney_h: I bet some of your HW is causing trouble
[15:32:03] <rodney_h> CosmicDJ:
[15:32:23] <rodney_h> CosmicDJ: Yes this is what I thought also, but?
[15:34:00] <rodney_h> I am running Gateway model: GM5478 Intel Core 2 Quad 2.4 GHz. 2 GB System RAM 500 GB SATA2 HDD...
[15:35:20] <rodney_h> CosmicDJ: It should be able to handle running the OS?
[15:35:44] <CosmicDJ> rodney_h: look, opensolaris is bleeding edge code, insufficient testing on the majority of hw configurations out there and only a small set of supported hardware
[15:36:34] <seanmcg> CosmicDJ, it does get tested on a _large_ set of supported hardware ! :)
[15:37:03] <CosmicDJ> seanmcg: yeah, all branded with a sun logo? ;)
[15:37:33] <_mary_kate_> sun actually supports a reasonable amount of non-sun stuff now, at least in the server market.. consumer desktop PCs probably not so much
[15:37:55] <rodney_h> CosmicDJ: I'm not trying to argue with you guy, I'm just trying to figure out what the problem is, I really would like to use the original cd I got, I am trying to learn a different OS other than MicroShaft WinBlows....
[15:38:35] *** luc^ has quit IRC
[15:38:37] *** Rarok has joined #opensolaris
[15:39:30] *** calumb has joined #opensolaris
[15:40:20] <sponix> _mary_kate_:  yeah, Military Buys a Ton of Sun Boxen and has them come with Windows Server pre-installed, have to say it makes me sick
[15:40:33] <_mary_kate_> sponix: i meant supporting solaris on non-sun systems
[15:40:34] <seanmcg> CosmicDJ, :)
[15:40:45] <sponix> I am of course glad to see Sun is still getting business either way though ;)
[15:40:49] <h3sp4wn> rodney_h: Even if you did install it from the original disk if you ran the updates then you would download over 600MB anyway
[15:40:50] *** KOHJU is now known as kohju
[15:40:53] <timsf> rodney_h: have you tried just running it in VirtualBox to begin with?
[15:41:03] <seanmcg> sponix, remember Sun is a hardware company too :)
[15:41:05] <sponix> Oh, like Dell doing Solaris Installs now
[15:41:19] <CosmicDJ> seanmcg: I thought sun is a service company ;)
[15:41:20] <timsf> that'd give you a good feel for the OS, and a chance to learn it, before trying to deal with the hassle of potential hardware issues
[15:41:24] <_mary_kate_> no, like you buying dell hardware and installing solaris on it and getting support for it from sun
[15:41:24] *** loquitus_of_bor1 has joined #opensolaris
[15:41:34] *** Rarok has quit IRC
[15:41:34] <timsf> Sun's a systems company.
[15:41:44] <rodney_h> timsf: I believe that is what I'm trying to do? Run from LiveCD!
[15:41:54] <CosmicDJ> timsf: no sun is a storage company :p
[15:42:01] <sponix> seanmcg:  I know, kills me how they can stand in front of me and justify Sun hardware "because it is the most reliable", and then 1) Have no redundancy on systems 2) Run Windows
[15:42:03] <CosmicDJ> I think we have the 4 s' now..
[15:42:14] <timsf> Nope, the LiveCD runs it on your physical hardware, running it in VirtualBox would be running it on virtual hardware...
[15:42:22] <sponix> seanmcg:  Probably just my emotions, but it seems like a Mixed Bag to me ;)
[15:42:57] <rodney_h> timsf: Okay, now I'm totally confused, maybe you can explain what you mean?
[15:43:02] <timsf> Sure
[15:43:36] <timsf>  - the live CD runs on your actual machine, sees physical hardware and depends on software support for that hardware.
[15:43:39] <CosmicDJ> timsf: well at least he knows then that his cd is not to blame (when it runs inside vbox)
[15:43:55] *** anomaly has quit IRC
[15:44:09] <timsf> VirtualBox presents the illusion of a physical computer,
[15:44:15] <sponix> CosmicDJ:  only if he has ran an md5sum on the ISO image ;)
[15:44:26] <timsf> but it's all emulated (for want of a better, more accurate word)
[15:44:29] *** anomaly has joined #opensolaris
[15:44:33] <timsf> and we know OpenSolaris runs on VirtualBox
[15:44:36] <timsf>  - you boot windows,
[15:44:38] <CosmicDJ> sponix: he got his opensolaris cd via snailmail from sun
[15:44:46] <timsf> install and download VirtualBox,
[15:44:50] <timsf> then start a virtual machine,
[15:44:54] <sponix> Have to say, VirtualBox has been kicking some ass since Sun acquired it. Tons of improvement
[15:45:02] <timsf> and run the LiveCD inside the virtual machine
[15:45:17] *** stux|work has left #opensolaris
[15:45:24] <timsf>  - sure it won't be as fast as running it on the physical computer (called "bare metal" , in the trade)
[15:45:30] <sponix> CosmicDJ:  I'd suck it in from the Drive to an ISO and md5sum that then ;)
[15:45:48] <timsf> but at least you can explore the OS, get used to it, then worry about whether you've got hardware support for it.
[15:46:31] <rodney_h> timsf: Ok, so if I understand what you're telling me I need to go to Sun's website and download an application called VirtualBox and use it to run the OS from the cd?
[15:46:42] <timsf> Yep, exactly
[15:46:58] <sponix> timsf:  I like the USB support now, I can hookup a USB hard drive and access it within virtualbox
[15:47:07] <timsf>  - with more experience, you could start trying to work out why the OS isn't running on your physical hardware,
[15:47:18] <timsf> but that's proabably not the best solution for a beginner
[15:47:49] <h3sp4wn> Is usb support working for Solaris hosts now ?
[15:47:59] <rodney_h> timsf: Okay, next question? Is this App VBOX free to download or am I going to have to pay for it? :D
[15:48:01] <sponix> rodney_h:  virtualbox.org has information about what a "virtual machine" is. Basically running an OS within another OS
[15:48:05] <timsf> Nope, it's free.
[15:49:04] <rodney_h> timsf: cool!
[15:49:15] <rodney_h> Thanks guys for all your help!
[15:49:32] <sponix> Example: I run Ubuntu Linux 8.04.1 as my Main OS (Host OS), from there I now have Solaris 10, OpenSolaris 200805-1, SXCE, and Windows XP running at the same time in separate windows inside my Linux Desktop Screen
[15:49:44] *** __teo__ has joined #opensolaris
[15:49:46] *** sophokles has joined #opensolaris
[15:49:52] <sponix> rodney_h:  It lets me test, work with, and play on all the OS's at the same time
[15:50:04] * rodney_h going to go check this baby out!
[15:50:09] <sponix> Virtual Computers inside a Computer
[15:50:24] <sponix> If one say Crashed, I could just restart that Virtual Computer with my Main one still going
[15:50:27] *** sophokles has left #opensolaris
[15:51:02] *** rodney_h has quit IRC
[15:51:29] <h3sp4wn> Speaking of virtualisation has anyone managed to get the paravirtualised disk driver working properly for windows on xvm
[15:51:40] *** Rarok has joined #opensolaris
[15:52:26] <sponix> h3sp4wn:  Can't say I have, but I am very curious to exactly what that is ;)
[15:52:26] *** _teo_ has quit IRC
[15:53:21] <oxygene> h3sp4wn: uh, windows and paravirtualized? how does that fit together?
[15:53:33] <_mary_kate_> oxygene: he's probably using hvm with a pv driver
[15:53:40] <_mary_kate_> improves performance without needing a pv-aware OS
[15:54:08] *** syamajala2 has quit IRC
[15:54:14] <h3sp4wn> https://cds.sun.com/is-bin/INTERSHOP.enfinity/WFS/CDS-CDS_SMI-Site/en_US/-/USD/ViewProductDetail-Start?ProductRef=sun-xVM-guest_add-G-F@CDS-CDS_SMI
[15:54:47] <oxygene> _mary_kate_: ie. hvm for the cpu/mmu, pv stuff for disk and network? interesting mix
[15:54:58] <_mary_kate_> oxygene: yes
[15:55:15] <_mary_kate_> oxygene: iirc, that's how sun wants to virtualise S10 in the short term as well
[15:57:22] * timsf nods
[15:58:10] *** letz_ has quit IRC
[15:59:03] <h3sp4wn> sponix: what _mary_kate_ said
[16:00:27] <sponix> _mary_kate_:  umm, So, they want a base platform Host to run S10 as a guest OS ?
[16:00:33] <sponix> <-- more lost
[16:01:00] <carl-> i run snmpwalk from a linuxhost against a solarishost .. and it gets all kinds of stuff but it allways ends in "Timeout: No Response from ipaddress"
[16:01:08] <h3sp4wn> sponix: Its the equivalent of the windows guest additions but for xvm server (xen) instead of virtualbox
[16:01:14] <oxygene> sponix: that setup is the "easy way" to run solaris 10 inside xen. no need to change the kernel
[16:01:24] <_mary_kate_> sponix: i don't understand your question
[16:01:25] <sponix> awww
[16:01:28] <oxygene> sponix: xen is a virtualization technology (dubbed xvm by sun, due to trademark reasons)
[16:01:56] <sponix> Yes, and how is that compared to VirtualBox ?
[16:02:00] <_mary_kate_> sponix: normally, virtualisation for non-Xen-aware OSs emulates a network card and disk driver.  PV drivers allow the host to provide the native xen network and disk drivers (which is much faster) without requiring the guest OS to understand Xen
[16:02:03] <sponix> think I'll have to try that out
[16:04:25] <sponix> OK, Latest x86 Solaris 10, in JDS as root, needing to create a User, hit Launch, where on this freaking menu did they put the Users and Groups tool ?
[16:04:33] <sponix> Know it is there somewhere
[16:05:38] <sponix> found it in SXCE right off
[16:06:42] <calumb> sponix: pretty sure we didn't ship it with S10... it's part of gnome-system-tools, which we added later...
[16:07:48] <sponix> I'm just spoiled, that is what you are saying? ... So off to CDE, or running useradd/adduser from the CLI aye ?
[16:08:15] <h3sp4wn> You could just run the smc
[16:08:23] <sponix> been a minute since I've ran Solaris 10
[16:08:40] <sponix> h3sp4wn:  true
[16:09:28] <sponix> Got 2G dedicated to this Solaris 10 session and it is still running sluggish compared to SXCE... Might be because I don't have the Guest Additions package installed yet though
[16:09:29] <h3sp4wn> Now I wonder if I can switch this domU to boot off iscsi (as the network driver is working perfectly)
[16:10:13] <oxygene> sponix: my sol10 in vbox is still sluggish, despite guest additions. a bit less than before, but still
[16:11:18] <sponix> oxygene:  I badly need to back this box up and just put Solaris on it native, but I'm in limbo about what flavor to run with
[16:11:18] *** letz_ has joined #opensolaris
[16:11:35] *** Openfree has joined #opensolaris
[16:11:40] *** letz_ has quit IRC
[16:14:47] <sponix> h3sp4wn:  So, performance wise Xen kicks the crap out of Virtualbox ?
[16:14:59] <oxygene> as usual, "depends" ;)
[16:15:38] <sponix> Basically if I reload this box I'll need the Power of Solaris, to my wife it will have to be transparent
[16:16:19] <sponix> She is use to Ubuntu8.04.1 and doesn't like change, so a VM of it will be on Top, and needs to be near native speed, with full Video and Audio support
[16:16:38] *** ruse39[home] has quit IRC
[16:16:44] <sponix> Have yet to see a VM I could play a movie smoothly in though ;)
[16:17:26] <h3sp4wn> I can play a movie over the network smoothly with VirtualGL
[16:18:25] *** timsf has quit IRC
[16:18:29] <sponix> Man, I am really _really_ starting to miss my research time, I am very outdated
[16:18:42] <sponix> VirtualGL, I take it a Virtual OpenGL flavor ?
[16:20:09] <h3sp4wn> Yeah its part of SXCE - I have a noisy Sun Blade 1000 - with an XVR-1000 and you can use the local video card to accelerate and then use turbo vnc its faster than anything I have seen yet (including citrix)
[16:20:32] <h3sp4wn> Apparantly you can use it with exceed on Windows instead also
[16:22:37] *** stux has joined #opensolaris
[16:23:57] *** tsoome has quit IRC
[16:24:15] *** stux is now known as stux|work
[16:31:35] *** ejray has quit IRC
[16:32:06] *** timsf has joined #opensolaris
[16:32:09] <carl-> and someone here know how i can easily access data such as load , processes in different states such as blocking ( from running vmstat etc )
[16:34:02] *** calumb is now known as calAFK
[16:37:03] <h3sp4wn> prstat ? ps ?
[16:40:05] *** jamesd has quit IRC
[16:40:22] <oxygene> hmm.. seems like solaris 10 still can't do efficient clones of zones on zfs (via zfs clones), contrary to what the documnetation says
[16:43:52] *** calAFK is now known as calumb
[16:43:56] <sponix> Any Belenix or Nexenta users in here ?
[16:44:38] *** vmlemon_ has quit IRC
[16:45:40] *** vmlemon_ has joined #opensolaris
[16:52:14] <sponix> what is the revisions with os200805-1.iso ? I didn't see a release log or changelog, does it have patches to os200805 or just a repack of the ISO ?
[16:54:23] *** minidev has quit IRC
[16:55:08] *** tsoome has joined #opensolaris
[16:56:43] *** jamesd has joined #opensolaris
[16:56:43] *** ChanServ sets mode: +o jamesd
[16:56:57] *** nibbler has quit IRC
[17:00:09] *** not-me-guv has joined #opensolaris
[17:10:25] *** twisti is now known as twisti_work
[17:13:04] *** juriskr has quit IRC
[17:15:27] *** perlmongo has quit IRC
[17:16:42] *** gerard13 has quit IRC
[17:17:05] *** capaz has joined #opensolaris
[17:23:51] *** MeP3aBeu has quit IRC
[17:28:49] *** carl- has quit IRC
[17:33:16] *** stukag has joined #opensolaris
[17:35:01] *** norman has quit IRC
[17:41:04] *** kohju is now known as KOHJU
[17:42:23] *** Triskelios has joined #opensolaris
[17:43:48] *** MattAFC is now known as MattMan
[17:44:38] <jbk> do ufsdump or *tar properly handle backing up/restoring hard links?
[17:47:00] *** anomaly has quit IRC
[17:47:25] *** anomaly has joined #opensolaris
[17:51:49] *** dunc_ has joined #opensolaris
[17:52:54] <CosmicDJ> jbk: trying it out with tar should be no prob, don't you think? :)
[17:56:34] *** jfndi has joined #opensolaris
[17:57:50] *** dellastreet has quit IRC
[18:00:27] *** MrBIOS_ has joined #OpenSolaris
[18:04:35] *** jay-away has joined #opensolaris
[18:04:39] *** MrBIOS_ has quit IRC
[18:07:39] *** stevel has joined #opensolaris
[18:07:40] *** ChanServ sets mode: +o stevel
[18:11:16] *** sartek has quit IRC
[18:12:12] *** wms has joined #opensolaris
[18:12:33] *** mikefut has quit IRC
[18:19:33] *** SYS64738 has quit IRC
[18:22:24] <loquitus_of_bor1> Has anybody build gcc on OpenSolaris? I am having some trouble.
[18:23:17] *** bondolo has joined #opensolaris
[18:23:39] <CosmicDJ> loquitus_of_bor1: you are not the first one, IIRC there was lenghty thread on comp.unix.solaris
[18:23:48] *** dunc_ has quit IRC
[18:24:53] <loquitus_of_bor1> CosmicDJ: can you direct me to where I can find the directions to do this? the compiler that came with OpenSolaris 2008.05 is ANCIENT!
[18:25:29] *** Fish- has joined #opensolaris
[18:26:14] *** minunea_konsole has joined #opensolaris
[18:27:16] <e^ipi> loquitus_of_bor1: use ss-dev
[18:27:24] <e^ipi> it's faster, stronger... better
[18:27:51] *** DTEIT has quit IRC
[18:28:03] <CosmicDJ> loquitus_of_bor1: or http://cooltools.sunsource.net/gcc/
[18:28:12] <loquitus_of_bor1> e^ipi: can you please elaborate? what is ss-dev? do you mean to say I should use it to build gcc, or use it INSTEAD of gcc?
[18:28:20] <e^ipi> instead of gcc
[18:28:30] <e^ipi> it generates much better code
[18:31:05] <cypromis> ss-dev is sunstudio
[18:31:09] <cypromis> gcc is a desaster
[18:31:10] <loquitus_of_bor1> e^ipi: the reason I am updating gcc is because the one I have on 2008.05 is ancient, and I am suspecting that might explain why I am getting alot of annoying and ridiculous errors in my C++ code that never happen with the gcc compiler I have on Linux for the same codebase....
[18:32:11] <Triskelios> loquitus_of_bor1: by "annonying and ridiculous" you probably mean "standards-complaint". although the studio default template impl. is not great, try the stlport option
[18:33:05] <e^ipi> stefan just got his apache libstdcxx arc case approved, so that'll be an even nicer & more standards compliant C++ standard lib
[18:33:23] *** logic has joined #opensolaris
[18:34:09] <loquitus_of_bor1> I have sun studio express. is that different from ss-dev? so do I need to install ss-dev as WELL now?
[18:34:17] <e^ipi> same thing
[18:34:21] <loquitus_of_bor1> stlport option? can you give an example?
[18:34:22] *** minunea_konsole has left #opensolaris
[18:34:26] <Triskelios> e^ipi: would be nice if there were a binary compatible solution to just ditch old roguewave stl implementation
[18:34:39] <Triskelios> (probably technically unfeasible)
[18:34:44] <loquitus_of_bor1> yes... these are stl issues I have...
[18:34:56] <Triskelios> loquitus_of_bor1: try enabling stlport
[18:35:26] <Triskelios> the default stl implementation is really limited for compatibility issues
[18:36:21] <loquitus_of_bor1> Triskelios: can you give me a real quick example... is stlport a directive for the gcc compiler or sun studio express
[18:36:38] <e^ipi> studio
[18:36:45] <e^ipi> you'd use -library=stlport5 IIRC
[18:36:59] <e^ipi> ( or it could be stlport4, i don't remember)
[18:37:00] <CosmicDJ> stlport4
[18:37:07] <e^ipi> there you go
[18:37:39] <loquitus_of_bor1> so like "CC -library=stlport4 foobar.c -o foobar" ?
[18:37:55] <e^ipi> ya
[18:44:12] *** stukag has quit IRC
[18:44:12] *** hudnix has quit IRC
[18:44:37] *** logic855 has quit IRC
[18:44:37] *** logic is now known as logic855
[18:47:46] *** mega has joined #opensolaris
[18:49:48] *** Fish- has quit IRC
[18:49:51] *** sipior has left #opensolaris
[18:52:24] <loquitus_of_bor1> Triskelios: so is that in: "CC -library=stlport4 -staticlib=stlport4 foobar.c -o foobar"
[18:53:29] <Triskelios> loquitus_of_bor1: that should work
[18:54:01] <CosmicDJ> s/foobar.c/foobar.cc/
[18:55:52] <e^ipi> i usually give my c++ files the .cpp suffix
[18:55:55] <e^ipi> *shrug*
[18:56:24] *** SYS64738 has joined #opensolaris
[18:56:30] <loquitus_of_bor1> CosmicDJ: us the .cc ending required?
[18:56:40] *** div8 has quit IRC
[18:56:51] <e^ipi> loquitus_of_bor1: no, he's pointing out that you wouldn't be compiling C files like that, only C++
[18:57:01] <loquitus_of_bor1> gocha
[18:57:10] <e^ipi> ( C typically doesn't use the C++ STL )
[18:57:18] <loquitus_of_bor1> yep. typo
[18:57:35] *** anomaly has quit IRC
[18:57:39] <e^ipi> yeah, i assumed you knew what you were talking about
[18:59:06] *** bnitz has joined #opensolaris
[18:59:31] *** Fish- has joined #opensolaris
[19:00:25] *** m0zzzy has joined #opensolaris
[19:01:16] *** sartek has joined #opensolaris
[19:03:13] <loquitus_of_bor1> Are these compiler (CXXFLAGS) directives?
[19:03:23] <e^ipi> yeah
[19:04:35] <loquitus_of_bor1> so just to double check then... I would typically put in BOTH -library=stlport4 -staticlib=stlport4 if I do NOT intend to distribute studio libs with the software I am building?
[19:04:56] <e^ipi> i believe that is the case, yes
[19:05:01] <Triskelios> correct; I believe the latter flag implies the former but it doesn't hurt to put both
[19:05:02] <loquitus_of_bor1> and -library=stlport4 if I am willing to distribute the dynamic libs with it.
[19:05:27] <loquitus_of_bor1> ah ok. it is curious... these sound like LDFLAG directives, yet they affect the compiler.
[19:05:52] <loquitus_of_bor1> is stlport4 some magic tag that tells it to somehow "be nicer to me" in terms of my STL code?
[19:06:11] *** bnitz1 has quit IRC
[19:06:13] <e^ipi> there's a whole different set of linker flags & preprocessor flags , in addition to a compiler flag that tells it not to link to the default
[19:06:26] <e^ipi> so that's abstracted away with a single compiler flag
[19:06:35] *** bhall has joined #opensolaris
[19:06:44] <Triskelios> it uses a completely different stl implementation. different linker flags are chosen by the compiler automatically based on this choice as well..
[19:08:06] <loquitus_of_bor1> Triskelios: so technically if I am linking as a seperate command than the compiling step, I would neeed to provide these two directives on both? (ie: Both LDFLAGS AND CXXFLAGS)?
[19:09:54] <Triskelios> loquitus_of_bor1: correct. typically the compiler can also be used to link so the same options apply
[19:10:04] *** vmlemon_ has quit IRC
[19:12:53] <fraggeln> no sxce_100 yet? :)
[19:13:16] <e^ipi> the ON gate just became available
[19:13:17] <e^ipi> chill.
[19:13:46] *** jfndi has quit IRC
[19:14:28] *** bubbva has joined #opensolaris
[19:15:52] <loquitus_of_bor1> As soon as I added the two new directives to my LD flags to generate "export LDFLAGS="-L/opt/csw/lib -library=stlport4 -staticlib=stlport4"", the configure script failed saying I can't build executables.
[19:16:12] *** vmlemon_ has joined #opensolaris
[19:16:21] <e^ipi> no, just give them to the compiler
[19:16:23] <loquitus_of_bor1> Mind you it did this trying to builder conftest.c
[19:16:29] <loquitus_of_bor1> ah ok. so not ldflags.
[19:17:17] <fraggeln> i must say, i realy like what you guys are dooing, keep up the good work.
[19:18:28] *** mikefut has joined #opensolaris
[19:21:57] *** kimc has quit IRC
[19:22:51] <sstallion_work> e^ipi: any chance you know how to check irq assignments on sparc /
[19:28:05] *** vmlemon_ has quit IRC
[19:28:24] <loquitus_of_bor1> Triskellios, e&ipi, CosmicDJ: thanks. it seems to have compiled. Now I am just resolving some boost library issue that is most likely not related
[19:29:36] *** ludc has joined #opensolaris
[19:31:46] <CosmicDJ> loquitus_of_bor1: link issues?
[19:32:34] *** vmlemon_ has joined #opensolaris
[19:32:36] <Triskelios> loquitus_of_bor1: boost is normally built with stlport as well, but keep in mind that other C++ libraries may be using the older STL. they are unfortunately not ABI compatible
[19:32:41] *** crichardso has joined #opensolaris
[19:33:48] *** hannesd has joined #opensolaris
[19:34:20] <Triskelios> (also libraries created with g++ are similarly not compatible, but you were probably aware)
[19:35:00] *** TomJ has joined #opensolaris
[19:35:25] <loquitus_of_bor1> CosmicDJ: I don't have boost's filesystem libraries installed.
[19:35:48] <loquitus_of_bor1> Triskelios: so if I am installing boost's libraries on here, I'd need to build those the same way right?
[19:36:53] <CosmicDJ> loquitus_of_bor1: http://blogs.sun.com/sga/
[19:36:55] <Triskelios> loquitus_of_bor1: it's the only way to build them with studio, afaik. SFE has a package recipe for boost
[19:38:29] <loquitus_of_bor1> CosmicDJ: that's terrific! I just need to find bjam now :)
[19:40:11] <Triskelios> it's part of the boost package; the first thing built before boost itself
[19:42:38] <TomJ> My SXCE b99 box is hanging regularly.  It's a server I've not used for a long time, so could well be hardware.   It just freezes, no panic, no messages on console, no messages in adm/messages or syslog.  Can I enable anything that might hlep track what is causing the death?   It seems to be happening on disk access (e.g. rsync), but zfs scrubs are not triggering it (and show no errors.)
[19:43:01] <CosmicDJ> TomJ: http://prefetch.net/blog/index.php/2007/02/11/recovering-from-solaris-hangs-with-the-deadman-timer/
[19:43:33] *** pumpkin_ has joined #opensolaris
[19:43:45] <TomJ> ok thanks
[19:44:06] <loquitus_of_bor1> Triskelios: it is built before boost, but which directory do I go into do run the build script?
[19:44:45] <loquitus_of_bor1> found it
[19:44:46] *** MeP3aBeu has joined #opensolaris
[19:45:54] *** pjfloyd has joined #opensolaris
[19:46:13] <Triskelios> loquitus_of_bor1: tools/jam/src. I made one of the boost packages in SFE, which is why I'm suggesting that you try them
[19:47:59] *** not-me-guv has quit IRC
[19:48:09] *** hannesd has quit IRC
[19:49:19] <Triskelios> bbl
[19:49:26] <loquitus_of_bor1> Triskelios: SFE?
[19:51:33] <Triskelios> spec-files-extra. a packaging project
[19:51:40] *** Triskelios has quit IRC
[19:52:34] <loquitus_of_bor1> Damn. what does he mean by that?
[19:53:07] <Berny> http://pkgtool.sf.net/ look for SFE there
[19:53:26] <Berny> aehm pkgbuild.sf.net even
[19:53:35] <loquitus_of_bor1> that link goes not work. but is sfe another "package manager" like pkg-get ?
[19:54:56] <Berny> no it's a project to build packages using spec files
[19:55:15] <Berny> aehm pkgbuild.sf.net does work
[19:56:30] <Berny> you grab the specfile for a program you want, fire of a pkgtool build --download specfile.spec
[19:56:38] <Berny> wait a bit and there you are
[19:57:02] <tomww> ...happy
[19:57:50] *** Wez has joined #opensolaris
[19:57:57] *** twisti_home is now known as twisti
[19:58:27] <Berny> hopefully :-)
[19:59:01] * Berny is off to wife and kid
[19:59:03] <Berny> laters
[19:59:13] <loquitus_of_bor1> thanks
[19:59:14] <loquitus_of_bor1> !
[20:00:58] *** calumb has quit IRC
[20:05:04] *** kimc has joined #opensolaris
[20:10:04] *** xtrondo has quit IRC
[20:11:11] *** tarbo_ has joined #opensolaris
[20:12:27] *** jgracin has joined #opensolaris
[20:14:19] *** pjfloyd has left #opensolaris
[20:15:47] *** chubs has joined #opensolaris
[20:18:56] <TomJ> My dump device is rpool/dump - how can I access this from teh filesystem?
[20:20:12] <e^ipi> you can't, it's a dump device
[20:20:17] *** [1]kimc has joined #opensolaris
[20:20:38] <TomJ> ok, so how do I access the dumps generated by a panic?  I use dumpadm to specify they're saved to a filesystem?
[20:20:50] <e^ipi> /var/crash/<hostname>/
[20:21:01] <TomJ> hmm, so it will happen by default?
[20:21:03] <TomJ> it's empty
[20:21:11] <TomJ> sxce is hanging constantly
[20:21:19] <e^ipi> hang != panic
[20:21:29] <TomJ> I did:  set snooping=1 in /etc/system as advicsed by  http://prefetch.net/blog/index.php/2007/02/11/recovering-from-solaris-hangs-with-the-deadman-timer/
[20:21:38] <TomJ> which is meant to cause a panic to trigger even if the system has hung
[20:21:48] *** tarbo has quit IRC
[20:21:49] <TomJ> but I guess this is a 'deeper' hang and that deadman timer isn't even working
[20:22:15] <TomJ> not realy sure how I'm supposed to work out what is causing the hang
[20:23:37] *** myrkraverk has quit IRC
[20:24:28] *** stux|work is now known as stux16777216Away
[20:25:42] *** sponix has quit IRC
[20:26:08] *** LordIllpalazo has joined #opensolaris
[20:27:13] *** AxeZ has joined #opensolaris
[20:33:32] <CosmicDJ> TomJ: did you wait 500 secs?
[20:33:52] <TomJ> I set snoop_interval to 90 seconds
[20:33:57] <CosmicDJ> ah ok
[20:33:59] *** kimc has quit IRC
[20:33:59] *** [1]kimc is now known as kimc
[20:34:06] <TomJ> set snoop_interval=9000000
[20:34:23] <TomJ> actually after adding those system settings and rebooting, the box hung almost immediately, even before I had a chance to login
[20:34:30] <TomJ> then I rebooted again and it lasted 5 mins or so
[20:34:31] <CosmicDJ> no activity after that? i.e. blinking drive leds indiating that a crashdump is written to swap?
[20:34:36] <TomJ> this may just be coincidental
[20:34:51] <TomJ> but before it was only crashing on activit,y like rsync, and (oddly) installing Sun Studio
[20:35:08] <CosmicDJ> TomJ: is this a x86/64 box?
[20:35:18] <TomJ> yeah dual Opteron 264, amd64
[20:35:20] <TomJ> *246
[20:35:29] <CosmicDJ> TomJ: try memtest86 :)
[20:35:36] <TomJ> yeah not a bad idea
[20:35:38] <CosmicDJ> or Sun's VTS
[20:35:42] <CosmicDJ> or both :)
[20:35:52] <TomJ> it was definitely having memory issues,  I had hangs related to memory a couple of days ago - fmadm faulty was showing bad DIMMs
[20:36:11] <CosmicDJ> and you didn't replace them?
[20:36:11] <TomJ> but I pulled some dimms and that went away, but now this has happened and fmadm faulty is not reporting anything
[20:36:28] *** tarbo has joined #opensolaris
[20:36:38] <TomJ> well I couldnt actually work out which dimms it meant for sure. so I pulled some I'd added recently, and the crashes stopped for a day or two.. then this happened, and unlike before, no fmadm faulty output this time
[20:36:46] <TomJ> but yes I suppose it could still be related
[20:36:59] <TomJ> I'll try memtest and/or cutting it down the bare minimum memory it'll run on
[20:37:23] <TomJ> when it was crashing before,  fmadm would print to console then it'd hang.  this time it just hangs, no output of any kind
[20:37:31] <CosmicDJ> test one stick at a time; this should help finding out which one(s) are bad
[20:37:42] <TomJ> unfortautneyl they have to be in pairs, but yeah I'll cut down to one pair
[20:38:34] *** NeZetiC has joined #opensolaris
[20:38:41] <CosmicDJ> we also had two hba's failing today, and you thought you're safe because they were redundant...
[20:39:07] <TomJ> hehe ouch
[20:39:42] <CosmicDJ> I heard this killed their exchange server...
[20:40:49] <CosmicDJ> and same stuff have their calendars in exchange; they complained they have no idea when/which/what appointment they have today...
[20:41:23] <CosmicDJ> damn we are so depended on those machines nowaday...
[20:41:58] *** stukag has joined #opensolaris
[20:42:08] <CosmicDJ> +s
[20:42:28] <CosmicDJ> I want my typewrite back ;)
[20:42:30] <CosmicDJ> +r
[20:44:26] *** timsf has quit IRC
[20:45:55] <TomJ> by far the most annoying thing I've had recently was last weekend.. setting up this box in fact.  it has 16 x 400GB SATA drives that had previously been used in linux, now I was puttnig them into solaris on 2 x supermicro MV8 SATA2 controllers.  8 worked perfectly,  the other 8 would cause sxce b99 to simply hang on install, or on boot once it had been insatlled to one of the good drives.
[20:46:22] <TomJ> all 8 of those drives were on one controller, so must be the controller.. no, wasn't.. all 8 were on one side of the chassis, maybe bad power to that side? nope. they all had black cables (other 8 had red), so must be cables?  nope.  dead drives? hmm, could be, beacuse the drive also failed when booted on an adaptec controller (with read errors)
[20:46:25] <TomJ> but all 8 drives dying in one go?
[20:46:58] <TomJ> turned out that they were all still part of a 3WARE hardware RAID array.. and for some reason, being part of a hardware raid array had changed the drive contents such as to make them show read errors in adaptec,  and cause complet ehangs on supermicro
[20:47:02] <TomJ> took me 12 hours to diagnose that
[20:47:14] <CosmicDJ> hehe
[20:47:27] *** tarbo_ has quit IRC
[20:47:42] *** vmlemon_ has quit IRC
[20:47:54] *** jerivard has left #opensolaris
[20:47:54] <CosmicDJ> I also had some fun trying to install solaris 10 on a EFI labeled harddisk...
[20:48:24] <CosmicDJ> as you might guess, it didn't work and the installer dropped out for no reason
[20:48:37] *** capaz has quit IRC
[20:51:05] *** [1]kimc has joined #opensolaris
[20:53:28] <e^ipi> hey, bitchin'
[20:53:34] <e^ipi> transmission-daemon is coming along for the ride
[20:54:18] *** mega has quit IRC
[20:55:14] <e^ipi> no SMF service though
[20:56:01] *** capaz has joined #opensolaris
[20:56:32] *** kimc has quit IRC
[20:56:32] *** [1]kimc is now known as kimc
[20:57:25] *** mega_ has joined #opensolaris
[20:58:52] <TomJ> ok looks like it was probably ram then.. cut down to 2GB and now got Studio installed
[20:59:03] <TomJ> damn this box is falling apart.. is meant to have 8gb and now I'm down to 2
[20:59:25] <CIA-25> Richard Bean <Richard.Bean at Sun dot COM>: 6560842 sparc drivers should not use SCCS keywords in user-visible strings
[20:59:26] <CIA-25> Ramesh Kumar Katla <Ramesh.K at Sun dot COM>: 6712497 gratuitous arp not sent sometimes during ipmp failover under heavy network load
[20:59:27] <TomJ> hope it's the chips not the slots
[21:02:41] *** rhqq has joined #opensolaris
[21:02:47] *** stukag has quit IRC
[21:03:26] *** stukag has joined #opensolaris
[21:04:04] <rhqq> hi. is there any support for ati/amd graphic cards in opensolaris?
[21:04:21] *** xRaich[o]2x has left #opensolaris
[21:05:19] *** storycrafter has joined #opensolaris
[21:05:33] <CosmicDJ> rhqq: 2D, sure :)
[21:06:40] <rhqq> well... i want to use blender on os
[21:06:58] <rhqq> :/
[21:07:27] <rhqq> so better choice would be a linux based system, right?
[21:08:41] *** bnitz has quit IRC
[21:08:53] <e^ipi> or windows
[21:09:00] *** Erwann has quit IRC
[21:09:03] <CosmicDJ> or irix *cought*
[21:09:07] <e^ipi> +1
[21:09:14] <e^ipi> or use a supported card like nvidia or intel
[21:09:15] <rhqq> ehh, thanks
[21:09:27] <rhqq> i have 4870...
[21:10:07] <e^ipi> mind you, does blender even use the 3d hardware, or does it do everything via CPU?
[21:10:17] <rhqq> rendering is via cpu
[21:10:27] <e^ipi> okay, so there shouldn't be an issue
[21:10:34] <rhqq> all projecting is based on gpu/ g-ram and ram
[21:11:15] <e^ipi> not sure what that means
[21:11:31] <rhqq> all textures and vertex are in gpu
[21:11:40] <rhqq> while projecting
[21:12:01] <rhqq> cpu does only rendering
[21:12:35] <rhqq> well. i try
[21:12:59] <storycrafter> rhqq, you should be fine
[21:13:58] <rhqq> thx, i tell you results ;) now im gonna install it ;)
[21:14:10] *** chubs has quit IRC
[21:14:11] *** ahe has joined #opensolaris
[21:15:55] *** gehmehgeh has joined #opensolaris
[21:16:09] *** myrkraverk has joined #opensolaris
[21:17:08] *** wms has quit IRC
[21:20:09] *** gehmehgeh has quit IRC
[21:23:57] *** jwit has quit IRC
[21:24:16] <ludc> the directory /var/spool/pkg can be the my local repository ?
[21:28:09] *** plavcik has joined #opensolaris
[21:29:40] *** coffman has quit IRC
[21:30:32] *** snakesqzns has joined #opensolaris
[21:34:34] <plavcik> I exported some dirs via ZFS command, I see it with sharemgr show -vp zfs, I can'nt mount it from other machine, what I have to adjust, please?
[21:36:57] *** snakesqzns has left #opensolaris
[21:41:01] *** Dar has quit IRC
[21:41:56] *** jwit has joined #opensolaris
[21:42:54] *** mattc_ has joined #opensolaris
[21:46:03] <plavcik> where are is log from nfs server for unsuccessfull attempts for mount?
[21:48:39] *** MattMan has quit IRC
[21:53:09] <seanmcg> plavcik, from the other machine can you do a showmount -e $nfs-server ?
[21:57:35] *** freakazoid0223 has quit IRC
[21:58:07] <plavcik> other machine is linux, can I try something similar
[21:59:29] <CIA-25> Santwona Behera <Santwona.Behera at Sun dot COM>: 6723808 ifconfig plumb fails on PlatinumII: 2 * 10G fiber ports and 2 * 1G copper ports
[21:59:49] <seanmcg> what command did you use to share ?  I do: zfs set sharenfs=ro tank/some-zfs
[22:00:43] *** ludc has left #opensolaris
[22:00:44] <plavcik> zfs set sharenfs=rw data/images
[22:01:37] *** CoolMa has joined #opensolaris
[22:02:01] <plavcik> mount.nfs: access denied by server while mounting 192.168.1.250:/srv/data/images
[22:02:25] <plavcik> something with user ids? shall I put ip range somewhere like with /etc/exports?
[22:03:35] <plavcik> showmount -e 192.168.1.250 show /export/images (everyone)
[22:04:01] <plavcik> I have two network cards on server, can this be related?
[22:05:07] <plavcik> directory on server have 777 mode, I did not touched ZFS ACL
[22:05:41] <seanmcg> plavcik, you're trying to mount 192.168.1.250:/srv/data/images, but the showmount shows the /export/images directory shared, not /srv/data/images
[22:07:13] <plavcik> you right, unfortunatellymount.nfs: access denied by server while mounting 192.168.1.250:/export/images
[22:07:44] *** ivan has quit IRC
[22:07:44] *** djinni has quit IRC
[22:07:52] <plavcik> I used firstly command via sudo, then fstab (which have previously old line)
[22:08:48] *** djinni has joined #opensolaris
[22:09:02] <seanmcg> on the server is the nfs service running ?  svcs nfs/server
[22:09:35] <seanmcg> and on the server can you cd /net/localhost/export/images ?
[22:09:52] <plavcik> online          6:42:02 svc:/network/nfs/server:default
[22:10:20] *** Rarok has joined #opensolaris
[22:10:42] <plavcik> as root yes
[22:10:52] <plavcik> exist sudo user? to try it as user
[22:11:36] <seanmcg> since you can cd /net/localhost/export/images, it doesn't appear to be a problem on the nfs server side.
[22:11:57] <plavcik> ok, I will try it from OpenBSD then
[22:13:38] <plavcik> mount_nfs: ct access /export/images: Permission denied :(
[22:14:01] <CosmicDJ> plavcik: show us the exact mount command
[22:14:04] <blahee> yeap. one doesn't get root access to mount while doing something like 'zfs set sharenfs=rw data/images'
[22:14:28] <plavcik> on server, I not have /etc/exports
[22:14:51] <CosmicDJ> plavcik: you don't need /etc/exports on solaris
[22:14:54] *** Mdx4 has joined #opensolaris
[22:14:56] <plavcik> as user mount /elrond/images
[22:15:00] <blahee> something like 'zfs set sharenfs=sec=sys,rw,root=.you.domain.tld' does work given that there is DNS for domain
[22:15:03] *** TomJ has quit IRC
[22:15:19] <plavcik> 192.168.1.250:/export/images /elrond/images nfs rw,nodev,nosuid,noexec,noauto,user 0 0
[22:15:25] <plavcik> in fstab
[22:16:12] *** ivan has joined #opensolaris
[22:16:50] <plavcik> I dont neet a root acces, shall I change owner of that dir on server?
[22:19:38] <CoolMa> hi there all
[22:20:29] <blahee> i am no expert in all the fine details of nfs, but usually i do mount from client _as_ root and therefor all the users has access in their level (uid macthed in both ends). efectively this "root access" is that the mounting uid from client side is root (as in uid=0)
[22:21:15] <blahee> one could alter the uid mounting the share from client side, if needed
[22:21:21] *** myrkraverk has quit IRC
[22:22:26] <plavcik> data/images  sharenfs  sec=sys,rw,root=.mydomain.tld  local
[22:22:54] <plavcik> as root then mount 192.168.1.250:/export/images /elrond/images
[22:22:59] <plavcik> mount.nfs: access denied by server while mounting 192.168.1.250:/export/images
[22:23:04] <blahee> plavcik: i'd assume that you don't really have DNS for .mydomain.tld?
[22:23:23] <plavcik> no
[22:23:23] <blahee> plavcik: if you don't have proper DNS setup, use /etc/hosts and hostnames
[22:23:53] <plavcik> 127.0.0.1 full.domain.name is fine?
[22:24:05] <plavcik> or on static IP adresses
[22:24:07] <plavcik> ?
[22:24:33] <blahee> i think the "root=aa.b.cc.dd" should work too
[22:24:56] *** mikefut has quit IRC
[22:25:06] <plavcik> where aa.b.cc.dd is client IP ?
[22:25:14] <blahee> yeap
[22:25:41] <blahee> i have had proper DNS setup  at home so long that i can't even remember when i didn't have it
[22:26:08] <blahee> ie. all the hosts does have PTR records too
[22:26:10] *** derchris^ has joined #opensolaris
[22:26:18] *** derchris has quit IRC
[22:26:27] <plavcik> data/images  sharenfs  sec=sys,rw,root=192.168.1.13  local
[22:26:55] <plavcik> mount.nfs: access denied by server while mounting 192.168.1.250:/export/images
[22:28:02] <plavcik> I remember, then once on OpenBSD I have to set IP address for nfsd, because I hava two network cards
[22:28:07] <blahee> where does this /export/images come from? I see this as /data/images?
[22:28:25] <plavcik> ZFS data/images
[22:28:52] <plavcik> I can access that via ssh
[22:29:15] <plavcik> I can access via ssh /export/images
[22:29:36] <blahee> what does command exportfs list it as exported?
[22:30:41] <plavcik> sharemgr show -vp zfs     zfs/data/images nfs=() nfs:sys=(rw="*" root="192.168.1.13")
[22:31:31] <blahee> i don't know. i'd put this 192.168.1.13 to /etc/hosts as line for hostname and would use this for export permissions. Dunno if it even work with IP
[22:33:05] <plavcik> yes!! that is , adding line with IP and full domain name of client solved this
[22:33:17] <plavcik> thanks for all help, I learned a lot hier!
[22:33:24] <blahee> the reason being reverse resolving
[22:33:33] <CoolMa> booting of opensolaris livecd 200805 stops at "configuring devices." what could be the problem?
[22:33:49] <blahee> the /etc/hosts entry does resolve IP<->hostname both ways while pure IP doesn't
[22:34:00] <plavcik> I see
[22:34:25] <plavcik> where I can learn how to imporve my 8 years old dhcp server config to do it there?
[22:34:49] <plavcik> or is matter of bind (tinydns) only?
[22:35:14] <blahee> bind (as in DNS)
[22:35:30] <blahee> dhcp doesn't do DNS
[22:35:39] <plavcik> ok, thx again
[22:36:17] *** Odin- has joined #opensolaris
[22:36:19] *** mattc_ has quit IRC
[22:36:42] *** freakazoid0223 has joined #opensolaris
[22:39:56] *** tombhadAC has joined #opensolaris
[22:42:06] *** not-me-guv has joined #opensolaris
[22:42:24] *** jwit has quit IRC
[22:42:35] *** pjfloyd has joined #opensolaris
[22:45:54] *** jwit has joined #opensolaris
[22:46:48] *** Atomdrache has joined #opensolaris
[22:46:53] *** pjfloyd has left #opensolaris
[22:48:03] *** bondolo has quit IRC
[22:52:24] *** xtrondo has joined #opensolaris
[22:59:00] *** brendang has quit IRC
[22:59:24] <CIA-25> Ramesh Kumar Katla <Ramesh.K at Sun dot COM>: 6653407 Multicast 01:80:c2:00:00:02 is enabled after aggregation is deleted, 6656421 aggr: aggr_send_port_disable can be called without aggr grp write lock
[22:59:25] <CIA-25> gt29601@anthrax: 6739879 Oracle database keep crashing while using direct NFS feature
[22:59:38] *** brendang has joined #opensolaris
[23:00:52] <jbk> heh
[23:01:07] <jbk> i have to wonder if direct NFS wasn't a response to the quality of NFS on linux
[23:01:31] <jbk> i mean.. writing your own userland NFS client inside of a database?
[23:02:18] <sstallion_work> jbk: ?
[23:03:14] *** Rarok has quit IRC
[23:03:25] <jbk> a new feature in oracle
[23:03:28] <jbk> 'direct nfs'
[23:03:40] <jbk> which is basically an NFS client built into the database
[23:03:47] <jbk> for cases where you're running your database over NAS
[23:04:01] <RElling1> Larry wants the whole software stack :-)
[23:04:01] <sstallion_work> yech
[23:04:07] <sstallion_work> sounds like a silly idea
[23:04:15] <sstallion_work> better to use iscsi
[23:04:30] <jbk> well they already have ASM, OCFS, their own (horrible) cluster software (CRS)
[23:04:33] <sstallion_work> reminds me of the direct disk stuff that came along a few years ago
[23:04:47] <jbk> god i hate CRS
[23:04:59] <jbk> i woudln't trust it to cluster an addressbook, let alone a database
[23:05:40] *** Gman_ has quit IRC
[23:05:43] <jbk> iirc, solaris 9 (w/ certain patches) and later can support windows sizes > 32k
[23:05:54] <jbk> i'd be suprised if netapp doesn't support that as wekk
[23:05:56] <jbk> err well
[23:06:22] <jbk> yet, oracle _requires_ any NFS mounted fileystems w/ databases have 'rsize=32768,wsize=32768' as part of the mount options
[23:06:22] *** stukag has quit IRC
[23:06:55] *** medar has quit IRC
[23:07:11] *** jgracin has quit IRC
[23:09:46] *** medar has joined #opensolaris
[23:10:38] *** norman has joined #opensolaris
[23:11:20] <sstallion_work> yeesh
[23:11:22] <sstallion_work> thats retarded
[23:11:38] <sstallion_work> Although, I really cringe at the idea of keeping a database on NFS
[23:11:55] <sstallion_work> SANs exist for a reason
[23:12:24] <jbk> it works well enough
[23:14:06] <jbk> i really dislike the whole 'you can only use X,Y,Z HBAs and switches A,B,C with our storage array, even though all of this is theoritically standardized and in theory should all interoperate and in theory should work just like ethernetcards'
[23:14:12] *** AxeZ has quit IRC
[23:14:24] *** TomJ has joined #opensolaris
[23:15:42] <jbk> the major storage vendors still do that today
[23:16:01] <jbk> which to me makes the whole fc standard a farce
[23:16:18] *** CoolMa has left #opensolaris
[23:17:38] <loquitus_of_bor1> hey people. How do I tell the suns studio compiler to use the static version of a library to link against?
[23:17:55] <loquitus_of_bor1> for example, I am doing -lfoobar, but I want to link against foobar.a, not foobar.so
[23:18:11] <oxygene> loquitus_of_bor1: refer to /foo/bar/libfoobar.a, instead of -lfoobar
[23:18:19] <jbk> don't you just include the library in the link line?
[23:18:21] <oxygene> ie. refer to it by name
[23:18:32] <loquitus_of_bor1> oxygene: I do not want to have to specify the path to it, obviously.
[23:18:34] <oxygene> but be careful, file name order is important, then
[23:19:51] <loquitus_of_bor1> Isn't there some directive that lets me specify the library name alone but does a static link for that one library?
[23:19:57] <TomJ> How do I get Solaris DHCP to give out IPs within a subnet?  I've done:  dhcpconfig -D -r SUNWfiles -p /var/dhcp ;  dhcpconfig -N 192.168.0.0 -t 192.168.0.10    #  but now I don't know how to tell it to give out IPs in the range 192.168.0.0/24
[23:20:12] <loquitus_of_bor1> i thought that -Bstatic preceding it would work, but it doesn't
[23:20:16] <CosmicDJ> loquitus_of_bor1: http://blogs.sun.com/rie/entry/static_linking_where_did_it
[23:20:44] <CosmicDJ> loquitus_of_bor1: in short; no static bin's anymore on sol10+
[23:21:18] <loquitus_of_bor1> CosmicDJ: does that apply to me as well, where I want some of my libraries to be linked in statically but not all?
[23:21:53] <loquitus_of_bor1> I mean I would hate to have to bundle boost libraries whenever I package my software up, even as a dependency. I wanted to just shove the boost code into the software
[23:22:45] <jbk> no static linking of _system_ libraries
[23:22:45] <oxygene> loquitus_of_bor1: then reference the full path
[23:22:51] <sstallion_work> loquitus_of_bor1: no. unless you want to support a separate boot lib package, you are better off either sandboxing your install with the boot .so's, or just linking statically
[23:22:54] <jbk> libc, etc.
[23:22:56] <oxygene> loquitus_of_bor1: you need it anyway, to figure out the -L parameter
[23:23:04] <sstallion_work>  s/boot/boost/
[23:23:15] <loquitus_of_bor1> sstallion_work: I thought I was just told I cannot link statically?
[23:23:19] *** capaz has left #opensolaris
[23:23:20] <sstallion_work> no
[23:23:34] <sstallion_work> solaris just does not use static links for cmd anymore
[23:23:41] <jbk> loquitus_of_bor1: see above comment
[23:23:44] <jbk> (16:23) < jbk> no static linking of _system_ libraries
[23:23:56] <jbk> if you have your own libfoo.a, you can statically link against that
[23:23:59] <loquitus_of_bor1> is boost considered a system library?
[23:24:02] <jbk> no
[23:24:13] <sstallion_work> jbk: did they do away with libc.a ?
[23:24:36] <jbk> yes
[23:24:56] <loquitus_of_bor1> so let me get this straight then. the -l parameter only works for shared libs. if I want a static lib, I have to put in the full path to the static lib?
[23:25:09] <sstallion_work> ahh interesting. its been so long since I've had to statically link anything I had never checked
[23:25:10] <loquitus_of_bor1> (yes, not system libraries)
[23:25:25] <jbk> sstallion_work: starting with solaris 10
[23:25:30] <oxygene> loquitus_of_bor1: entirely different handling, so yes
[23:25:33] <sstallion_work> gotcha
[23:25:54] <loquitus_of_bor1> ok. another question I have is... where do I specify in OpenSolaris an additional directory to search for .so files in?
[23:26:02] <loquitus_of_bor1> On Linux it was ldconfig, I believe.
[23:26:07] <loquitus_of_bor1> but on there, there is not one
[23:26:16] <jbk> link it correctly :)
[23:26:28] <jbk>  /etc/ld.so.conf is an abomination
[23:26:35] <jbk> use -R/path
[23:26:47] <jbk> when linking to append a runtime search path into the binary
[23:26:50] <loquitus_of_bor1> jbk: it is. boost got installed and put into /usr/local/lib but /usr/local/lib is not in the library search path
[23:27:07] <jbk> add -R /usr/local/lib
[23:27:14] <jbk> or -R '$ORIGIN/../lib'
[23:27:18] <jbk> or both
[23:27:22] <jbk> to the link line
[23:27:26] <loquitus_of_bor1> jbk: -R/usr/local/lib would be added to the link stage of the software I am linking against boost?
[23:27:30] <e^ipi> loquitus_of_bor1, post-facto, you can use elfedit to correct the problem without fucking with the system
[23:27:47] <jbk> yes, that will cause the binary to search that path (in addition to the system defaults) when looking for libraries
[23:28:03] <e^ipi> but yeah, you should just build it with '-R $ORIGIN/../lib'
[23:28:13] <jbk> $ORIGIN/../lib will do it relative to the location of the executable
[23:28:48] <loquitus_of_bor1> e^ipi: how will $ORIGIN translate to /usr/local?
[23:28:49] *** mattc_ has joined #opensolaris
[23:29:07] <jbk> $ORIGIN is the location of the binary when it is run
[23:29:16] <e^ipi> what he said
[23:29:29] <loquitus_of_bor1> right. so that would only work if the binary was put into /usr/local/bin right?
[23:29:34] *** Fish- has quit IRC
[23:29:45] <e^ipi> is that not where you're putting it?
[23:29:57] <jbk> it means 'whever the binary is, search ../lib from where it's at for libraries'
[23:30:07] <jbk> so if it's in /usr/local/bin, it'll searhc /usr/local/lib
[23:30:14] <loquitus_of_bor1> maybe. maybe not. I don't want to assume. I might put it into /usr/bin/ the fact that boost is not there is kind of lame.
[23:30:18] <jbk> if it's in /foo/bar/bin, it will search /foo/bar/lib
[23:30:20] <loquitus_of_bor1> yah
[23:30:38] <e^ipi> loquitus_of_bor1, if you put it in /usr/bin you're likely to drop the library in to /usr/lib
[23:30:46] <storycrafter> gang, have a great weekend.
[23:30:49] <e^ipi> so $ORIGIN/../ is still correct
[23:31:02] *** storycrafter has left #opensolaris
[23:31:33] <jbk> the $ORIGIN syntax allows you to move things around, and still have it work as long as the relative directory structure is intact
[23:32:07] <loquitus_of_bor1> e^ipi: I realize that. but do understand that my binary is using boost libraries that got installed in a whole separate and complicated process that did not put boost into /usr/lib. boost got into /usr/local/lib. so now what you are indirectly saying is if I used ORIGIN, my binary would be tied to where boost got put.
[23:32:28] <jbk> you could use both
[23:32:39] <loquitus_of_bor1> true
[23:32:41] <loquitus_of_bor1> I could.
[23:33:45] <e^ipi> isn't most of boost just headers and inlines anyways?
[23:34:49] <oxygene> no, a couple of libraries, too
[23:34:59] <oxygene> but if you stick to the right subset of boost, you'll get by with headers and templates
[23:38:20] *** stukag has joined #opensolaris
[23:39:15] <loquitus_of_bor1> Not really OpenSolaris necessarily, but anyone know what I might be missing in terms of linked libraries: symbol __1cDstdM_String_base4Ccn0AJallocator4Cc___V_M_throw_out_of_range6kM_v_: referenced symbol not found]
[23:39:21] *** cypromis has quit IRC
[23:40:03] *** mattc_ has quit IRC
[23:40:24] *** mattc_ has joined #opensolaris
[23:40:31] <tsoome> dem command will translate it;)
[23:40:37] <e^ipi> void std::_String_base<char,std::allocator<char> >::_M_throw_out_of_range()const
[23:40:40] *** mattc_ has quit IRC
[23:41:10] <e^ipi> ( and yeah, dem(1) is how you translate that
[23:41:34] <loquitus_of_bor1> "dem"? some command line util?
[23:41:41] <e^ipi> yeah
[23:41:48] <e^ipi> short for C++ demangle
[23:42:26] <loquitus_of_bor1> damn. i don't even have that installed. ok. another thing to fix. what do you suppose that is inside of, in terms of library? I am surprised
[23:42:44] <e^ipi> some string library or another from the looks of it
[23:42:51] <e^ipi> exception handler most likely
[23:43:05] <loquitus_of_bor1> any idea which lib?
[23:43:23] *** stukag has quit IRC
[23:43:54] <loquitus_of_bor1> sounds like some stl library. wierd.
[23:43:56] *** freakazoid0223 has quit IRC
[23:45:00] *** lolmac has joined #opensolaris
[23:45:15] <oxygene> loquitus_of_bor1: there are at least two STL implementations in sun studio
[23:45:31] <oxygene> loquitus_of_bor1: and I think the standard one still doesn't support boost (because of compatibility requirements)
[23:45:55] <e^ipi> yeah, should be an internal representation of something. either whoever wrote that code is naughty, or you built against 2 different c++ standard libs
[23:46:25] <oxygene> try -library=stlport4
[23:46:48] <loquitus_of_bor1> oxygene: I am using that -library=stlport4 -staticlib=stlport4 directive
[23:47:05] <loquitus_of_bor1> but what i am running into here is a runtime link error.
[23:47:07] <oxygene> loquitus_of_bor1: and you did that for the entire build?
[23:47:15] <loquitus_of_bor1> for compiling. not linking.
[23:47:27] <loquitus_of_bor1> I am not referring to anything like that in the link stage
[23:48:07] *** Gman_ has joined #opensolaris
[23:50:11] <loquitus_of_bor1> yes, the entire build. everything was built using that. I set CXXFLAGS to use that
[23:50:56] <loquitus_of_bor1> Umm... one thing. I have no idea what boost was built using. is this possibly because boost is not built using the right libs?
[23:51:16] *** Gman_ is now known as Gman
[23:51:32] *** ahe has quit IRC
[23:51:38] *** ChanServ sets mode: +o Gman
[23:52:16] *** TomJ has quit IRC
[23:52:21] <e^ipi> loquitus_of_bor1, yeah, that'd be my guess
[23:53:56] <e^ipi> luckily libCstd and STLport4 are getting marked obsolete in favour of the actually good libstdcxx
[23:54:56] <loquitus_of_bor1> e^phi: actually it WAS built with those directives.
[23:55:02] <loquitus_of_bor1> boost, that is.
[23:55:06] *** Wez has quit IRC
[23:55:25] <oxygene> loquitus_of_bor1: maybe some more, too?
[23:55:51] <loquitus_of_bor1> oxygene: who knows... I mean I link against alot of stuff... anything could be broken right?
[23:56:04] <loquitus_of_bor1> libgtk. libpng. where is the limit?
[23:56:15] <oxygene> loquitus_of_bor1: those aren't c++
[23:56:17] <e^ipi> C libraries aren't an issue
[23:56:22] <e^ipi> only the C++ stuff collides
[23:56:23] <loquitus_of_bor1> is there anyway for me to know what it is that is brreaking this somehow?
[23:56:28] <loquitus_of_bor1> yeah, understood. c++ stuff
[23:56:29] *** weredragon has joined #opensolaris
[23:56:34] <oxygene> loquitus_of_bor1: but if you're so annoyed: how about rebuilding boost, and all c++ libs and your tool with gcc?
[23:56:46] *** weredragon has left #opensolaris
[23:56:56] <loquitus_of_bor1> oxygene: it is not that I am annoyed. I am just trying to get this to run. I have rebuilt boost already
[23:57:18] <e^ipi> yeah, because gcc /NEVER/ breaks binary compatibility *snirk*
[23:57:25] <loquitus_of_bor1> as for c++ libs, I do not know how. are you referring to me rebuilding all the sun studio c++ libraries?
[23:57:34] <loquitus_of_bor1> I am not using gcc
[23:57:51] <e^ipi> loquitus_of_bor1, that was directed at oxygene mostly :)
[23:57:59] <oxygene> e^ipi: they pretty much sorted it out recently. and just having 3 libraries (+ various compiler tweaks that might change names) in parallel isn't too helpful either
[23:58:36] <loquitus_of_bor1> so you folks think this is a matter now of me trolling through all the C++ dependencies I have to figure out which one is broken?
[23:59:07] <oxygene> loquitus_of_bor1: use nm on the various libraries and look for which library refers to that missing symbol
[23:59:22] <e^ipi> what are you trying to build, anyways?





top