[00:00:05] *** dj2 has quit IRC
[00:00:44] *** gm152 has quit IRC
[00:03:03] *** gm152 has joined #opensolaris
[00:07:45] *** laca has quit IRC
[00:09:42] *** laca has joined #opensolaris
[00:13:57] *** ShadowHntr has quit IRC
[00:28:52] *** delewis has joined #opensolaris
[00:31:29] *** salmandr has quit IRC
[00:42:19] *** elops has joined #opensolaris
[00:42:38] <elops> stupid question, but is is possible to force a commit of an older file to a cvs repository.  I have a couple of developers that are really screwing up their project and I just can't figure out how?
[00:42:48] <elops> e.g. they have checked out v 1.1 of a file, but repository is now at v 1.5.  can you force a commit of the v1.1 file over the v1.5 file
[00:49:12] <boyd> I have no idea, but worst case couldn't they get a diff of their changes then apply that to a checked out 1.5 version?
[00:49:58] *** jteo has quit IRC
[00:49:59] *** mdj has quit IRC
[00:50:48] *** mdj has joined #opensolaris
[00:51:07] *** mdj has quit IRC
[00:51:34] *** mdj has joined #opensolaris
[00:51:41] <Stric> elops: make a copy of that "1.1 file", checkout latest version, copy over your 1.1 and cvs ci
[00:52:30] *** uebayasi has joined #opensolaris
[00:53:02] *** mdj has quit IRC
[00:53:08] *** mdj has joined #opensolaris
[00:53:32] *** Cass has quit IRC
[00:53:51] *** Cass has joined #opensolaris
[01:04:48] *** mritun_ has joined #opensolaris
[01:05:07] <mritun_> yarr folks
[01:05:54] *** mdj has quit IRC
[01:06:22] *** mdj has joined #opensolaris
[01:08:03] <mritun_> nobody alive ?
[01:08:13] <esaxe> i'm here
[01:08:21] <mritun_> Hi esaxe
[01:08:24] <esaxe> how's things?
[01:08:29] <mritun_> fine fine...
[01:09:10] <mritun_> there has been a war on the discussion forum :-/
[01:09:18] <mritun_> the "newbie..." thread
[01:09:54] <mritun_> now if someone who can wrap his head around the discussion and can update the wikipedia entry on opensolaris
[01:10:13] <mritun_> then this bloodshed might be avoidable in future
[01:10:42] *** mdj has quit IRC
[01:12:18] <esaxe> wow, long thread.
[01:12:39] *** mdj has joined #opensolaris
[01:12:48] <delewis> yes, and its mostly a regurgitation of various statements that have been repreated for the last year.
[01:12:55] <delewis> repeated, rather.
[01:13:27] <mritun_> but not recorded
[01:13:36] <mritun_> hence the flamefest Part Deux
[01:14:18] <mritun_> I'm sill wraping my head around it
[01:14:34] <mritun_> and finding the most politically neutral language to put it in
[01:14:52] <richlowe> at least if the same argument repeats over and over it's easy to spot which threads to delete :)
[01:14:54] <mritun_> or there would be another repeat on wikipedia :-/
[01:14:59] <mritun_> :D
[01:15:00] <delewis> Solaris is Sun's Opensolaris-based distribution -- that's pretty straight-forward.
[01:15:03] <mritun_> hehe
[01:15:12] <mritun_> naw
[01:15:19] <mritun_> there are dsagreements
[01:15:24] <mritun_> disagreements
[01:15:40] <delewis> mritun_, disagreements are irrelevant. That's the truth of the matter.
[01:15:44] <mritun_> like Solaris == OpenSolaris - some pieces + some closed bits
[01:15:53] * mritun_ ducks
[01:15:58] <delewis> mritun_, that does not contradict what I said.
[01:16:01] <mritun_> not another one here
[01:16:03] *** edwardocallaghan has joined #opensolaris
[01:16:14] <mritun_> delewis: its ok
[01:16:16] <delewis> you're applying a completely different mindset towards the relationship of OpenSolaris and Solaris than you would say, Red Hat and the Linux kernel.
[01:16:29] <mritun_> just that an agreeable paragraph needs to be written
[01:16:44] *** mdj has quit IRC
[01:16:49] <mritun_> that both guides the newbies, and doesn't step on toes of those involved
[01:17:11] *** mdj has joined #opensolaris
[01:17:25] <mritun_> delewis: I'm not ;) thankfully I'm following all this for quite sometime
[01:17:42] <edwardocallaghan> ext3 is poo, when you lose power !
[01:17:43] <mritun_> so I have idea of what it is, and what it's not
[01:18:10] <mritun_> but writing it down neutrally is what I'm looking for
[01:18:18] *** elops has quit IRC
[01:18:26] <edwardocallaghan> Sorry, good evening everyone :)
[01:18:36] <mritun_> good evening edwardocallaghan
[01:18:44] <edwardocallaghan> ;)
[01:19:26] <edwardocallaghan> Just work myself up a little today about files systems going mad on me when Linux crashs
[01:20:22] <sch0> I don't think that genunix.org's wiki requires the NPOV efforts that Wikipedia might.  A paragraph would be helpful.
[01:21:06] *** yarihm has quit IRC
[01:21:35] <boyd> edwardocallaghan: But it's Journalled! :)
[01:22:11] <mritun_> in my experience, pretty much everything goes tits up when you lose power
[01:22:27] <mritun_> nothing is engineered with hard failures in mind
[01:22:36] <edwardocallaghan> Same
[01:22:53] <edwardocallaghan> Is ZFS good in this way?
[01:22:54] <dvorak> I had a v40z eat itself on tuesday actually, when it lost power
[01:22:55] <boyd> Yes, I was being sarcastic
[01:23:12] <mritun_> some people might say X has feature Y which lets it survive hard failures (power loss etc..)
[01:23:34] <mritun_> but the fact is, when you put all pieces together, the system doesn't work on the whole
[01:23:57] <mritun_> if your FS is reliable, your disks might not obey cache flush ad nauseum
[01:24:00] <edwardocallaghan> I hope ZFS does well
[01:24:24] <mritun_> edwardocallaghan: even ZFS can do zilch faced with disks that lie
[01:24:34] <boyd> edwardocallaghan: ZFS is good in that way. It's not really graceful in the case of a non-reduvdant pool disk failure, and I think there may be some issued with not giving up trying to write to a dead disk, but they are not systemic problems. The design is for always consistent on-disk data
[01:24:35] <edwardocallaghan> Maybe just bad disks
[01:24:38] <edwardocallaghan> yes sure
[01:24:41] <jengelh> or a bad day
[01:24:52] *** axisys has joined #opensolaris
[01:25:28] <edwardocallaghan> Every day is a bad day for me on Computers
[01:25:44] <edwardocallaghan> Anything that can go wrong will go wrong
[01:25:44] <mritun_> when I read forums... I see problems like that
[01:25:53] <mritun_> edwardocallaghan: precisely
[01:26:06] <edwardocallaghan> lol good to know I'm not the only one
[01:26:15] <mritun_> even if ONE component in the system is bad... the whole system goes down the toilet
[01:26:30] <edwardocallaghan> But I have never had a problem with any SPARC station!
[01:26:35] <mritun_> classic case on forum today
[01:26:44] <mritun_> someone had issue with failover
[01:26:54] <mritun_> system takes 5mins to fail over a yanked FC
[01:26:56] <boyd> Failover of what?
[01:26:57] *** Burana has joined #opensolaris
[01:27:12] <mritun_> boyd: FC multipathing
[01:27:22] * boyd nods
[01:27:25] <mritun_> he yanks one, system takes 5min to fail over to other
[01:27:25] <edwardocallaghan> FC (sorry?) Fedora Core?
[01:27:31] <twincest> fibre channel
[01:27:32] <mritun_> fiber channel SAN
[01:27:34] <Auralis> fibre channel
[01:27:38] <edwardocallaghan> Oh sorry :p
[01:28:04] <mritun_> turns out the sd driver was merrily doing 3-5 retries with 60s timeouts
[01:28:07] <edwardocallaghan> Oh I don't like SAN if it's the SAN I am thinking of
[01:28:19] <dvorak> ...
[01:28:19] <edwardocallaghan> Oh dear !
[01:28:30] <mritun_> on second note
[01:28:35] <edwardocallaghan> Sorry before I forget
[01:28:43] <mritun_> I'm QUITE amazed at the SUN engineers
[01:28:50] <edwardocallaghan> Can I ask a stupid question please
[01:28:56] <mritun_> issues like that pin pointed in minutes....
[01:29:05] <edwardocallaghan> Same, see the new Ultra 40M2 it rocks !
[01:29:06] <mritun_> just not possible in real world!
[01:29:30] <edwardocallaghan> My stupid question... I have not used containers yet, can it be used like VMPlayer?
[01:29:51] <mritun_> you mean zones ?
[01:30:00] <axisys> some group got a 6140 and need help to setup. I setup a 3510 .. is it much different?
[01:30:01] <mritun_> nope
[01:30:04] <edwardocallaghan> Do I ok...
[01:30:11] <jmcp> axisys: yes, actually, it is
[01:30:21] <jmcp> axisys: the basic principles are the same, but the interface is different
[01:30:26] <boyd> containers is a marketing term for (zones + resource controls)
[01:30:35] <edwardocallaghan> OH
[01:30:38] <mritun_> nope as in - you can not run arbitrary OS in them
[01:30:39] <boyd> morning jmcp
[01:30:42] <jmcp> hi boyd
[01:30:56] <mritun_> there is a nice page that sums it up
[01:30:59] <mritun_> lemme fish it out
[01:31:00] <edwardocallaghan> Is it a hypervisor then?
[01:31:10] <boyd> edwardocallaghan: Zones are a way of partitioning solaris processes from each other at the process, net and FS level
[01:31:11] <syndrome71> edward: Not in the least.  What you would really be looking for is Xen
[01:31:12] <axisys> jmcp: so there are different type of fiber chnl outside
[01:31:24] <boyd> hey syndrome71
[01:31:33] * syndrome71 lurks...
[01:31:34] <Tpenta> hello nathan
[01:31:37] <syndrome71> :)
[01:31:42] <boyd> hey Tpenta
[01:31:50] <edwardocallaghan> What do you guys think of Xen?
[01:31:59] <boyd> edwardocallaghan: There is only one kernel, and the procs are normal solaris procs.
[01:32:04] <edwardocallaghan> I have had alttile play with it
[01:32:08] <mritun_> http://uadmin.blogspot.com/
[01:32:17] <Tpenta> sufficiently well that tere is a project to get it integrated *very* shortly
[01:32:20] <axisys> jmcp: if i have 16 disks w/ two raid controllers.. i can create 16 luns per controllers and then create a multipath on sol 10
[01:32:33] <mritun_> there is an entry on differences in zones/Xen/Vmware/UML/blah there
[01:32:36] <axisys> jmcp: sorry that was a question
[01:33:19] <mritun_> http://uadmin.blogspot.com/2005/06/zones-vs-jail.html
[01:33:27] <boyd> Tpenta: Do you know something I dont
[01:33:28] <edwardocallaghan> Doesn't Xen need to be complied into the Kernel?
[01:33:29] <boyd> ?
[01:33:32] <boyd> edwardocallaghan: Yes
[01:33:44] <Tpenta> boyd: maybe
[01:33:44] <mritun_> edwardocallaghan: partially correct
[01:33:53] <edwardocallaghan> Is it in Solaris?
[01:33:54] <boyd> edwardocallaghan: It's, in effect, a port to a different (but v similar) architecture
[01:33:55] <edwardocallaghan> *yet
[01:34:08] <richlowe> edwardocallaghan: not yet, but there are project bits up on opensolaris.org
[01:34:09] <boyd> edwardocallaghan:  not yet
[01:34:10] <mritun_> if there is virtualization support in CPU (most recent opterons, Core2 Duo, etc)
[01:34:14] <Tpenta> not yet, real soon
[01:34:21] <mritun_> then guest OS can run unchanged
[01:34:21] <edwardocallaghan> Ah Jails yes FreeBSD
[01:34:30] *** mdj has quit IRC
[01:34:42] <jmcp> axisys: .... sorry, was doing something else for a bit
[01:34:55] <jmcp> axisys: you can create as many luns as the controller will allow.
[01:35:01] <edwardocallaghan> Yes its doing full virtualization with a hypervisor
[01:35:12] <edwardocallaghan> I have been looking at some code
[01:35:14] <jmcp> axisys: with the 3510, the general idea is that one controller is the alternate path.
[01:35:15] <axisys> axisys: hey i am getting free consulation :-)
[01:35:21] <jmcp> axisys: you should rtfm, of course
[01:35:28] <axisys> jmcp: yep
[01:35:31] <boyd> So xen == hypervisor, zones ~ jails + features
[01:35:35] <jmcp> axisys: and make darned sure you have the latest SAN patches installed
[01:35:45] <axisys> jmcp: just like to get a warm up thru someone who knows about it
[01:36:00] <jmcp> ha :)
[01:36:22] *** mdj has joined #opensolaris
[01:36:43] <mritun_> anyone knows what's happening in the bluetooth project ?
[01:36:51] <jmcp> axisys: here's what I would do:    create a massive raid5 vol out of all your disks. then create logical volumes (that's the 3510 term) and present those to the host through both primary and secondary (alternate) controllers
[01:36:56] <axisys> jmcp: so assuming (i have not seen it yet.. on vacation) i have two raid contrlr.. i could put 16 luns per controller and make a multipath ?
[01:36:58] <jmcp> then turn on mpxio
[01:37:05] <jmcp> yes, generally
[01:37:14] <boyd> mritun_: Not much yet.. .It's just been founded
[01:37:40] <mritun_> boyd: any idea which code base might be the base ?
[01:37:47] <axisys> jmcp: cool.. so i will use zfs for the raidz in my case..
[01:37:52] <mritun_> NetBSD one was in talks
[01:37:53] <boyd> Anyone got any ideas where a zsh-on-solaris project would fit?
[01:38:01] <jmcp> axisys: actually .... you create a total of 16 luns in the unit, and then present them through the primary and secondary controllers.
[01:38:06] <richlowe> boyd: in a giant argument with Roland.
[01:38:06] <boyd> mritun_: I don't think that's decided yet
[01:38:10] <richlowe> boyd: or is that not what you meant? :)
[01:38:11] <axisys> yes
[01:38:18] <boyd> richlowe: lol
[01:38:22] <jmcp> axisys: and then with the mpxio devices you can use raidz
[01:38:29] <axisys> jmcp: yep 16 luns present to OS and zfs and go from there
[01:38:32] <mritun_> boyd: zsh !
[01:38:33] <axisys> jmcp: exactly..
[01:38:39] <Auralis> zsh on solaris? zsh is already delivered with the other free shells
[01:38:44] <jmcp> axisys: so you'll have software (zfs raidz) as well as hardware (3510) protection. which is The Way To Do It (tm)
[01:38:49] <mritun_> isn't it already available ?
[01:38:49] <axisys> jmcp: very cool.. i guess i have some basic ideas then ;-)
[01:39:00] * jmcp nods
[01:39:03] <mritun_> what would the project address ?
[01:39:05] <jmcp> you're on the right track
[01:39:12] <jmcp> axisys: just remember the SAN patches
[01:39:17] <boyd> I'm thinking of solaris specific features... like completion functions for zfs, zones etc... maybe some direct support for RBAC
[01:39:18] <axisys> jmcp: ok
[01:39:29] <jmcp> the fix for 5077933 is directly applicable to se3510 .... and I created it :)
[01:39:35] <edwardocallaghan> You guys are fast, I got too many IRC's open on gaim hehe
[01:39:36] <axisys> googling for the latest san patches for sol 10
[01:39:44] <mritun_> boyd: cool...
[01:39:47] <Auralis> boyd: ahh, well, i'm all for that since zsh is my default shell :)
[01:39:47] <axisys> jmcp: alrighty !
[01:40:04] <mritun_> but wouldn't it be good as something that can be contributed back to zsh team ?
[01:40:07] <axisys> brb
[01:40:08] <jmcp> axisys: it didn't make it into s10u2 though, ...
[01:40:11] <boyd> Auralis: Me too...
[01:40:14] *** stevel has quit IRC
[01:40:14] <mritun_> may be part of main distribution ?
[01:40:23] <boyd> I think there are many of us.. I just want to concentrate the effort
[01:40:44] <mritun_> bash has its completion project hosted on sf... dunno why it's not in main release
[01:40:46] <Auralis> ZSH. The Last Word in shells :)
[01:40:47] <boyd> mritun_: Sure, we can do that...
[01:40:57] <boyd> Auralis: Except for zzzzzzzsh
[01:41:08] <Auralis> no, thats the snoozer shell fork :)
[01:41:13] <boyd> Heh
[01:41:25] <mritun_> zzzzzsh
[01:41:39] <mritun_> we do sleep(1) best !! (TM)
[01:41:41] <mritun_> :D
[01:41:48] <boyd> What I'm thinking of is just some kind of organization for solaris zsh users. I've heerd of at least 3 people working on zones completion, e.g.
[01:41:49] <edwardocallaghan> Right, one reason form each please, why is zsh better then bash?
[01:41:51] <boyd> independently
[01:42:02] *** hile_ has quit IRC
[01:42:09] <boyd> edwardocallaghan: I have to pick *one* reason?
[01:42:14] *** hile_ has joined #opensolaris
[01:42:22] <mritun_> edwardocallaghan: echo $(( 3.5 / 2.0 ))
[01:42:25] <jmcp> boyd: would that be a secret zsh society?
[01:42:32] *** hile_ has quit IRC
[01:42:33] <edwardocallaghan> OK go on
[01:42:45] *** hile__ has joined #opensolaris
[01:42:46] <mritun_> zftp
[01:42:49] *** hile_ has joined #opensolaris
[01:43:00] <mritun_> modules
[01:43:03] <boyd> tab-complete nearly *anything*
[01:43:05] <mritun_> 3 for me already :-/
[01:43:08] <twincest> does bash do **?
[01:43:14] <boyd> (with arrow selection in menus)
[01:43:26] <boyd> twincest: Don't know.. didn't last time I looked
[01:43:37] <twincest> also set -J
[01:43:40] <boyd> Generally *very* enhanced globbing.
[01:43:49] <boyd> rm *~*.keep
[01:43:53] <edwardocallaghan> OK looks like I got more to learn
[01:44:00] <edwardocallaghan> Thanks guys
[01:44:09] <hile_> boyd: talking of why zsh rocks?
[01:44:17] <boyd> yeah
[01:44:24] <boyd> hile_:  edwardocallaghan asked
[01:44:27] *** Burana has quit IRC
[01:44:29] *** Burana has joined #opensolaris
[01:44:30] <mritun_> well
[01:44:49] <mritun_> now if iPython and scsh guys were in channel
[01:44:54] <edwardocallaghan> Any tricks people can share please?
[01:44:55] <mritun_> we zsh folks would be toast
[01:45:00] <mritun_> crispy toast :D
[01:45:05] <edwardocallaghan> lol
[01:45:11] <twincest> zsh doesn't have a bug that breaks the 'pwd' command with certain zone configurations
[01:45:27] <edwardocallaghan> Ah good
[01:45:31] <edwardocallaghan> I hate that one
[01:45:39] <edwardocallaghan> Thanks, next...
[01:45:56] <boyd> There's a bug with pwd?
[01:46:01] <twincest> no, bash
[01:46:06] <boyd> sure
[01:46:08] <twincest> i mean, pwd in bash
[01:46:13] <twincest> http://bugs.opensolaris.org/view_bug.do?bug_id=6231865
[01:46:14] <boyd> yeah. I get it
[01:46:15] <edwardocallaghan> You guys are great, I learn so much here thanks
[01:46:49] <edwardocallaghan> Yes if you are in the dir tree of pwd then bash makes you bash your head on a wall
[01:47:02] <twincest> that one was really annoying my users a while ago.  had to fix it manually and recompile bash
[01:47:04] <boyd> edwardocallaghan: print -l /etc/**/a*(.)      == list all plain files starting with a in the /etc/ dir recursively
[01:47:20] <twincest> (for some reason sun still hasn't fixed it yet)
[01:47:28] <boyd> twincest: Yuck
[01:48:22] <elektronkind> quick question: drive with a solaris ufs slice from a sparc box. Need to mount it on x86 linux. will it not work due to the endianess of ufs since it's coming from a sparc box?
[01:48:35] <elektronkind> everything I've seen says no
[01:48:38] <boyd> elektronkind: It will not work
[01:48:45] <axisys> back
[01:48:46] * mritun_ doesn't think so too
[01:48:50] <boyd> (I can't work out all the negatives in your question)
[01:48:54] <elektronkind> thought so.
[01:48:55] <twincest> elek: use zfs instead
[01:48:56] <axisys> jmcp: did u mean SAN patch?
[01:49:01] <twincest> elek: it still won't work but it'll be much nicer
[01:49:01] <mritun_> endian issues are bang into the ditch
[01:49:29] <boyd> elektronkind: You need to backup/restore. Does anyone know if ufsdump/restore works across endiannesses
[01:49:38] <richlowe> Are we *sure* that x86 linux doesn't flip UFS as necessary?
[01:49:46] <twincest> richlowe: for some reason i actually thought it did
[01:49:51] <twincest> but then everyone else said 'no'
[01:49:54] <richlowe> twincest: me too, that's why I'm asking.
[01:49:59] <mritun_> richlowe: I'd bet it won't
[01:50:00] <elektronkind> twincest: no no, some dude here is frantic. his ultra 10 died. It ran solaris 8 anyway. He wants to get his data back by mounting the drive in one of his linux boxes.
[01:50:15] <mritun_> linux doesn't have single tree for multiple archs
[01:50:22] <hile_> shit, restore it from backups would be quicker
[01:50:39] <edwardocallaghan> x86 is Irish ... Backwards ahhahha
[01:50:42] <Tpenta> boot a solaris install disk and mount the thing
[01:50:53] <edwardocallaghan> Note I am half Irish so I can say that
[01:50:57] <mritun_> elektronkind: just a random thought
[01:51:00] <twincest> google suggests linux will read both endians
[01:51:07] <mritun_> elektronkind: get a nexenta live CD
[01:51:14] <mritun_> it'd mount it just fine
[01:51:26] <mritun_> then ufsdump | ncftpput somewhere
[01:51:35] <elektronkind> hahah. this man we're talking about was too sure of himself to bother with backups. I'm trying to see how long I can have him worry before I hand him a spare ultra 5 I have sitting in my cabinet.
[01:51:46] <mritun_> :D
[01:51:53] <mritun_> BOFH
[01:52:09] <twincest> from the source:
[01:52:18] <twincest>  *  Big-endian to little-endian byte-swapping/bitmaps by
[01:52:19] <twincest>  *        David S. Miller (davem at caip dot rutgers.edu), 1995
[01:52:22] <elektronkind> it's mean, but I don't get a chance to be mean. And I'm really curious about my original question anyway.
[01:52:24] <twincest> but that might be applying to minix fs rather than ufs
[01:52:29] <jmcp> axisys: yes
[01:52:38] 
[01:52:41] <elektronkind> david miller is mr. sparc/linux
[01:52:42] <boyd> Oh, x86 LINUX... sorry I didn't read. No idea
[01:52:53] <edwardocallaghan> Just a idea
[01:52:54] <jengelh> who would want an ultra 10 nowadays
[01:52:59] <mritun_> ME !!!!
[01:53:09] <twincest> it does include the byte swapping routines tho
[01:53:11] <boyd> I have one
[01:53:16] <twincest> so i'm putting $5 on it working
[01:53:18] <jengelh> get something new already :)
[01:53:21] *** bobbyz has quit IRC
[01:53:22] <mritun_> the shipping would kill the deal for me though :(
[01:53:46] <edwardocallaghan> Make good webservers
[01:53:59] <boyd> Terrible, terrible IDE controller in them
[01:54:27] 
[01:54:31] *** darkcmd has joined #opensolaris
[01:54:44] <mritun_> I have sweet memories of them
[01:54:49] <edwardocallaghan> True but how much is a IO card these days
[01:55:00] <boyd> Not much..
[01:55:02] <mritun_> a U10 was main student + faculty server
[01:55:07] <mritun_> + mail + web + nntp
[01:55:13] * boyd wonders if he could put a SATA controller in one
[01:55:20] <mritun_> even with 10 matlab sessions wouldn't break a sweat
[01:55:23] <boyd> or in a SB100 for that matter
[01:55:30] <jmcp> boyd: yeah, get one with a silicon image chip
[01:55:40] <darkcmd> is it possible to compile JDS for NetBSD?
[01:55:40] <boyd> mritun_: Jeez... nowadays they struggle with JDS :)
[01:55:43] <dvorak> a 10 is still stupid slow
[01:55:51] <boyd> jmcp: Cool.... Any recommendations locally?
[01:55:58] <mritun_> boyd: good ol' days :D
[01:56:03] <boyd> darkcmd: It's gnome. Why?
[01:56:04] <alanc> JDS is just GNOME with patches, and GNOME runs on NetBSD...
[01:56:13] <mritun_> that was a dual proc btw
[01:56:29] <boyd> Dual proc U10? I think not.
[01:56:32] <syndrome71> boyd: I bought one off ebay about 3 months ago. Came from China - 4 port and all...
[01:56:35] <mritun_> some 266 MHz x 2
[01:56:45] <boyd> U60 Maybe
[01:56:50] <mritun_> US II if I'm correct
[01:56:55] <alanc> trying to actually use the JDS spec files might be fun, unless someone's ported rpm to NetBSD
[01:56:58] <elektronkind> boyd: http://www.siig.com/product.asp?pid=1018
[01:57:10] <dvorak> I'd be surprised if rpm isn't ported to netbsd
[01:57:11] <syndrome71> boyd: that being said - I'm using it in x86, not sparc. ;)
[01:57:15] <dvorak> it's been ported to most everything else
[01:57:20] <darkcmd> boyd, so you can?
[01:57:21] <elektronkind> boyd: there's a PCI version of that, of course.
[01:57:44] <edwardocallaghan> People, how does the top UltraSPARC box (Ultra45) compare to the top AMD box (Ultra40M2)
[01:58:09] <mritun_> edwardocallaghan: ditch the U45... get a T1K
[01:58:25] <edwardocallaghan> T1 chip?
[01:58:27] <elektronkind> erm, not for a desktop...
[01:58:38] <mritun_> would make a nice desktop :D
[01:58:55] <elektronkind> it would make a loud desktop
[01:59:02] <richlowe> elektronkind: stick SRSS on it, and get a ray :)
[01:59:20] <mritun_> 4 core T1K + 2GB ram
[01:59:29] <mritun_> and Ray terminals all over home
[01:59:33] <mritun_> perfect setup
[01:59:34] <boyd> syndrome71: Yeah, I''l ask you about that more later
[01:59:36] <edwardocallaghan> The question... please stick to the question
[01:59:56] <darkcmd> are there many people running opensolaris on x86?
[02:00:00] <elektronkind> the t1000 won my computer room's recent Howler Of The Year away, narrowly beating out a x2100.
[02:00:01] *** Cass has quit IRC
[02:00:02] *** peteh has quit IRC
[02:00:08] <elektronkind> s/away/award
[02:00:09] <edwardocallaghan> I know but I am interested in comparing them
[02:00:15] <boyd> darkcmd: Most of the kernel devs
[02:00:22] *** Cass has joined #opensolaris
[02:00:33] *** Burana has quit IRC
[02:00:36] <darkcmd> boyd, so most of the people developing solaris are on x86?
[02:00:37] <boyd> edwardocallaghan: Ask me in about a month. I'll have 13 of each to compare
[02:00:50] <boyd> darkcmd: I'd guess, but not 100% sure
[02:00:59] <mritun_> darkcmd: most would be an exagerration
[02:01:05] *** peteh has joined #opensolaris
[02:01:12] <mritun_> but yes, most people would have x86 boxen
[02:01:20] <edwardocallaghan> Really, what's performance like?
[02:01:23] <mritun_> though a large number would have a sparc on disposal too
[02:01:27] <darkcmd> because most of the people I meet in Solaris channels say that x86 suck
[02:01:37] <mritun_> darkcmd: it does :D
[02:01:47] <darkcmd> mritun_, why?
[02:01:49] <mritun_> but the reality is- it's the market leader
[02:01:52] * boyd works
[02:02:08] <mritun_> so even if you don't like it, you'd have to live with it
[02:02:29] <darkcmd> why does it suck?
[02:02:38] <mritun_> the instruction set et al
[02:02:42] <jmcp> darkcmd: clearly you're hanging out in the wrong channels
[02:02:51] <darkcmd> jmcp, what do you mean?
[02:02:53] <mritun_> compilers have to jump through hoops
[02:03:16] <mritun_> sparc has nice register windows
[02:03:29] <mritun_> and small instruction set
[02:03:32] <jmcp> darkcmd: people who say that (open)solaris on x86 or x64 suck are failing to state their assumptions, or their experiences in the past which have prevented them from considering the possibility that things have most definitely improved from the bad old s8 days
[02:03:39] <mritun_> but yes, it's wrong channel for that ;)
[02:04:00] <mritun_> jmcp: erm
[02:04:01] <syndrome71> jmcp - I think you missed something. I think there were refering to the ISA, not to solaris running on it...
[02:04:01] <darkcmd> jmcp, no I meant, x86 as an architecture sucking
[02:04:03] <jmcp> mritun_: I wholeheartedly dislike the x86 and x64 ISAs, but they're a fact of life.
[02:04:08] <mritun_> the statement was x86 sucks
[02:04:16] <jmcp> syndrome71: ah ... bad assumption
[02:04:19] <mritun_> (in comparison to sparc or ....)
[02:04:29] <mritun_> not that solaris ON x86 scks
[02:04:30] <jmcp> fortunately for us, we can run Solaris on x86 and x64 :)
[02:04:31] <darkcmd> although, I have an Ultra 2 workstation
[02:04:47] <jmcp> mritun_: that's the problem with flipping between desktops :)
[02:04:50] * jmcp shrugs
[02:04:55] <mritun_> jmcp: yeps... agree
[02:05:00] <darkcmd> does anyone know if it's still possible to download 4.3 BSD for VAX?
[02:05:08] <mritun_> ewwww
[02:05:23] <dlg> im sure source is around
[02:05:26] <mritun_> you really belong to dark ages don't you ?
[02:05:29] <dlg> binaries might be harder to find
[02:06:01] <jmcp> darkcmd: you into computer forensics?
[02:06:05] <mritun_> :D
[02:06:07] <mritun_> lol
[02:06:08] <darkcmd> jmcp, no why?
[02:06:25] <darkcmd> I have this VAX and am interested in messing around with it
[02:06:26] <jmcp> 4.3BSD? for Vax? ....
[02:06:32] * jmcp sighs
[02:06:35] *** |tsoome| has quit IRC
[02:06:36] <richlowe> TUHS has most of the BSD sources.
[02:06:40] <darkcmd> TUHS?
[02:06:44] <richlowe> I think they have *some* tape images, but I'm not sure for which releases.
[02:06:51] <mritun_> netbsd folks might have one lying around
[02:06:52] <darkcmd> ah, found it
[02:07:12] <mritun_> !!!
[02:07:27] <edwardocallaghan> A VAX
[02:07:29] * syndrome71 has many-a-vax in the cupboard... Uses Vax emulator on his PC as it's *way* faster...
[02:07:31] <dlg> darkcmd: there are more modern operating systems running on the fax
[02:07:53] * syndrome71 prefers VMS on Vax / Alpha anyways... ;)
[02:07:54] * mritun_ 's phone too
[02:07:56] <darkcmd> dlg, I have netbsd, and vms on it
[02:07:57] *** mdj has quit IRC
[02:08:01] <twincest> there are 4.3 binaries for VAX on the internets.  i forget where though (maybe tuhs has them)
[02:08:06] <edwardocallaghan> Does that hold your paper work down?
[02:08:11] <dlg> darkcmd: netbsd works on it?
[02:08:14] <dlg> thats amazing
[02:08:28] *** mdj has joined #opensolaris
[02:08:33] <mritun_> netbsd runs on anything with a MMU
[02:08:41] <dlg> hahaha
[02:08:42] <darkcmd> yeah
[02:08:45] <dlg> ok
[02:09:22] <mritun_> the amazing thing is... it manages it all from the same code base
[02:09:23] <jmcp> syndrome71: I prefer to block out the disgusting evil memories of having to admin vms on vax
[02:09:59] <mritun_> I want to see one someday
[02:10:28] <mritun_> and a PDP11 too
[02:10:35] <jbk> heh.. i seem to recall someone telling me about 'protecting the fuse at all costs' when dealing with a vax
[02:10:46] <mritun_> ew
[02:10:57] <mritun_> at all costs 8-|
[02:11:00] <darkcmd> the AMD opterons don't see so bad
[02:11:32] <jbk> of course, just today even, someone was talking about how vms and dg-ux did clustering 'right' and no one else has come even close..
[02:11:32] <mritun_> nice procs
[02:11:47] <jbk> i'd love to know how their clustering worked and why everything else out today suppsoedly falls short
[02:11:54] <mritun_> well... QNX did it well for one
[02:11:59] <mritun_> and Plan9 does it too
[02:12:11] <darkcmd> mritun_, opterons are nice processors?
[02:12:36] <mritun_> darkcmd: I like them...
[02:13:00] * Tpenta logs 6502219 If SMI label exists, hal does not probe logical disks
[02:13:00] <edwardocallaghan> So how do the opteron Sun Ultra 40 M2's compare to the UltraSPARC Sun Ultra 45's?
[02:13:03] <mritun_> good FPU, low power, 64 bit, MP support....
[02:13:07] *** mdj has quit IRC
[02:13:50] <mritun_> edwardocallaghan: the U40 is 2 socket right ?
[02:14:09] <edwardocallaghan> Yes they both are two socket
[02:14:20] <mritun_> then go with U40 M2
[02:14:24] <edwardocallaghan> photos.sun.com to see inside
[02:14:41] <mritun_> H2 2007 quad core opterons would be out
[02:14:46] <syndrome71> and so can take 2 X dual core CPU's - and 4 core ones if AMD release the quad in the AM2 socket...
[02:14:49] <mritun_> and you'd be able to pop in two of them
[02:14:51] <edwardocallaghan> I know
[02:14:52] <mritun_> 8 core box ;)
[02:15:08] <syndrome71> the U45 is 2 X USIIIi+
[02:15:10] <edwardocallaghan> But its not all about the CPU
[02:15:21] <darkcmd> what else is it about?
[02:15:27] *** sch0 has left #opensolaris
[02:15:30] <syndrome71> Clock for clock (running at the same speed), the sparc and opteron chips are reasonably close for many workloads.
[02:15:31] <edwardocallaghan> What about the bus and though put
[02:15:42] <syndrome71> as the opteron is clocked faster, it'll feel faster...
[02:15:42] *** mdj has joined #opensolaris
[02:16:09] <edwardocallaghan> Well it depends on how many things it does in them clocks
[02:16:34] <mritun_> edwardocallaghan: the opteron woud have an edge
[02:16:44] <mritun_> USIIIi is barely 1.5GHz
[02:16:46] <edwardocallaghan> If the FSB is too slow then all them clocks are bit of a wast of power
[02:16:55] <mritun_> and FSB is not great either
[02:17:04] <darkcmd> I like these new aluminum sun cases
[02:17:08] <edwardocallaghan> No clocks are not everything
[02:17:23] <mritun_> the opteron with HT, integrated memory controller would be a screamer
[02:17:58] <edwardocallaghan> Oh for sure the Ultra 40 M2 is a screamer !
[02:18:10] <mritun_> edwardocallaghan: if you're IO bound, then anyways you won't be getting anything better then PCIe
[02:18:23] <edwardocallaghan> But how do does the Ultra 45 compare ?
[02:18:52] <mritun_> well you said the cpu isn't everything
[02:19:07] <mritun_> and on IO side, the things are bit brighter for AMD
[02:19:12] <mritun_> so I guess there you go
[02:19:19] <edwardocallaghan> True but it has a different mother broad as well you know !
[02:19:39] <mritun_> and why would you care ?
[02:19:45] <edwardocallaghan> OK
[02:20:22] <edwardocallaghan> I am very interested in electronics so ARK of the system interests me
[02:21:13] <edwardocallaghan> So are Sun going to upgrade there SPARC workstation line or are they going to let the SPARC die on the Workstation front ?
[02:21:34] <Tpenta> ummm isnt that what the u45 was?
[02:21:54] <syndrome71> Tpenta: Only an evolutionary upgrade though...
[02:22:07] <edwardocallaghan> I would not like to see that as they are the only ones I know that don't make x86 Desktops
[02:22:08] <darkcmd> I think they're letting SPARC die
[02:22:38] <syndrome71> I'd like to think that Niagara2 might make it into something desktoppy... But that's just me dreaming. :)
[02:22:39] <edwardocallaghan> That makes me want to cry, call me sad but it does
[02:23:19] <darkcmd> what makes you want to cry?
[02:23:30] <edwardocallaghan> That means 100% of the Desktop market will be x86 because of f**king M$
[02:23:41] <darkcmd> yeah
[02:23:49] <darkcmd> but you don't have to run M$ based operating systems
[02:24:11] <edwardocallaghan> And me and my kids have to grow up with no RISC systems
[02:24:24] <edwardocallaghan> I know I don't
[02:24:29] <syndrome71> Sun's not the only people that make SPARC...
[02:25:02] <edwardocallaghan> HP and Fuji right?
[02:25:10] * mritun_ thinks the only way is to work your ass off, and get a T2K
[02:25:19] <edwardocallaghan> I love Sun products though
[02:25:31] <mritun_> I have little trust in HP
[02:25:53] <edwardocallaghan> I am very little trust in HP
[02:25:55] <mritun_> but given that HP and Dell, and IBM are up against SUN
[02:26:04] <mritun_> I dunno how'd things look in long term
[02:26:12] <mritun_> SUN *has* to make profits
[02:26:17] <edwardocallaghan> Dell=Hell
[02:26:24] <mritun_> and deliver exceptional user experience
[02:26:35] <mritun_> Dell == Sells
[02:26:40] *** karrotx_ has joined #opensolaris
[02:26:42] <edwardocallaghan> They always have for me
[02:26:47] <darkcmd> how about homebrew x86 machines?
[02:27:24] <edwardocallaghan> Sun need to make more advertisement in the UK for there Desktops !!! I hope Sun are hearing this !!!
[02:27:41] <edwardocallaghan> No, I use to build my own from age 8
[02:27:56] <darkcmd> Sun's getting a lot of good publicity with OpenSolaris
[02:28:04] <darkcmd> edwardocallaghan, you don't like homebrew x86 machines?
[02:28:04] <edwardocallaghan> But Sun make better systems and there BIOS is not so full of bugs
[02:28:31] <darkcmd> SPARC machines don't use a BIOS though
[02:28:42] <darkcmd> they use a different sort of boot firmware
[02:28:43] <mritun_> SUN's marketing, sales and more especially the evangelism department SUCKS
[02:28:48] <edwardocallaghan> Dell sells because its advertisement in the UK is every where!
[02:28:54] <mritun_> sucks as in it *reallly* sucks
[02:29:03] <dlg> edwardocallaghan: dell bioses have less issues than the sun one
[02:29:13] <dlg> since sun doesnt really have one
[02:29:27] *** jgilje has quit IRC
[02:29:33] <edwardocallaghan> Sun use the 'open BIOS'
[02:29:36] <jmcp> dlg: on the ultra20 and ultra40? ever looked at the actual motherboard?
[02:29:43] <edwardocallaghan> Yes
[02:29:44] <Plaidrab> openboot
[02:29:47] <jmcp> on the sparc boxes they use OpenBoot / ieee1275
[02:29:48] <edwardocallaghan> I have
[02:29:57] <dlg> sparc boxen are beautiful
[02:30:08] <dlg> jmcp: the u20 feels like a pc with a flimsy sun wrapper on it
[02:30:11] <edwardocallaghan> photos.sun.com just look how good they are built !
[02:30:29] <edwardocallaghan> take a look at the 40's
[02:30:40] <edwardocallaghan> 20 mean to be cheap
[02:30:45] <Plaidrab> 1000s, 1500s, etc. Those were sexy
[02:30:53] <jmcp> dlg: correct. It's got a tyan motherboard. Ditto with the u40
[02:31:01] <dlg> exactly
[02:31:06] <darkcmd> I wish there was an openboot solution for x86
[02:31:09] <dlg> sun doesnt do the bios
[02:31:18] * jmcp sees some split hairs
[02:31:36] * dlg feel half asleep
[02:31:41] <edwardocallaghan> http://www.openbios.info/Welcome_to_OpenBIOS
[02:31:41] <dlg> sorry if i seem slow atm
[02:31:42] <Plaidrab> I saw a quote today that said Sun's moving 100 mil in T1 boxen a quarter.
[02:31:50] <darkcmd> edwardocallaghan, does it work?
[02:32:06] <edwardocallaghan> Yes
[02:32:20] <Tpenta> plaidrab, that has been the case for teh last 2 quarters
[02:32:21] <darkcmd> how can I use it?
[02:33:03] <edwardocallaghan> SUN released OpenBOOT source code
[02:33:04] <mritun_> yeah
[02:33:15] <mritun_> but they need to make profits
[02:33:19] <darkcmd> is that what openbios is based on?
[02:33:37] * jmcp concalls
[02:33:39] <edwardocallaghan> Don't know I can't think right now
[02:33:49] <edwardocallaghan> You need to read the docs
[02:33:56] <Plaidrab> dats a lotta t1s
[02:35:05] <edwardocallaghan> That's great
[02:35:22] <edwardocallaghan> I think I may buy a T1 for a Desktop
[02:35:33] <boyd> They're noisy
[02:35:41] <edwardocallaghan> But graphics would be nice
[02:35:44] <Tpenta> you might want to have a listen to the fans first, esp if you are considering a t1000
[02:35:55] <Tpenta> it's made to go into a datacentre
[02:35:57] <edwardocallaghan> Don't care about Noise
[02:36:17] <edwardocallaghan> I know what they sound like
[02:36:22] <Tpenta> the T1000 sounds like an industrial vacuum cleaner (you know like the ones you see folks wearing on their backs in the movies)
[02:36:26] <edwardocallaghan> Seen one, look fantastic !
[02:36:39] <edwardocallaghan> lol
[02:37:43] * boyd has an image of someone wearing a t1000 on their back
[02:37:53] <edwardocallaghan> I hope with the advent of OpenSPARC that it splits into a server version and a Desktop version
[02:38:21] <edwardocallaghan> If no one is going to bother then I will make my own company !
[02:38:33] <edwardocallaghan> And I'm not joking !
[02:38:52] <boyd> I'll buy a share
[02:39:01] <boyd> Just one, most likely :)
[02:39:06] <Tpenta> the problem with pushing openboot onto commodity x86 hardware is, what is joe sixpack going to do when he comes up to his screen to see it saying "ok " ?
[02:39:33] <boyd> That need not happen.... Apple had it on the PPC laptops and most people never saw it
[02:39:34] *** karlo has joined #opensolaris
[02:39:40] <Plaidrab> Mac hide it quite nicely without getting rid of it
[02:39:52] <Tpenta> true
[02:40:01] <Plaidrab> hide/hid
[02:40:23] <darkcmd> so OpenSPARC is going to be a completely open architecture?
[02:40:31] <boyd> I think EFI addresses many of the limitations of conventioal BIOS. It seems like that may be the way things go
[02:40:38] <Tpenta> what do you meen going to be?
[02:40:50] <darkcmd> not sure
[02:40:53] <boyd> darkcmd: It is now. GPL
[02:41:04] <darkcmd> wow
[02:41:20] <Tpenta> T1 is GPL, the documentation for SPARC was always far more available than for i86
[02:41:35] <Tpenta> just ask amd how open IA is
[02:41:38] <boyd> funny that you don't hear the FSF saying "We must not use the proprietary  x86 arch. Must go to t1"
[02:41:44] <Plaidrab> All you had to do, IIRC?, was join the trade group.
[02:41:56] <darkcmd> edwardocallaghan, would it be possible to make a PCI card that boots openbios part way through the BIOS and use OpenBIOS?
[02:42:06] <boyd> Plaidrab: Maybe... I think it was downloadable by anyone... is now, anyway
[02:42:26] <darkcmd> Tpenta, so IA isn't very open at all?
[02:42:39] <boyd> darkcmd: I think you would loose much of the benefit... e.g. boot time reduction
[02:42:39] <Plaidrab> I remember looking at it at one time out of curiousity and thinking it was incredibly easy, if you know what to do with the info. :)
[02:42:43] <boyd> s/loose/lose
[02:42:53] <darkcmd> boyd, yeah, I'm just wondering if it's possible
[02:43:10] <Tpenta> darkcmd, try not at all
[02:43:19] <boyd> darkcmd: Do you see the source for current intel chips anywhere?
[02:43:26] <darkcmd> no
[02:43:52] <boyd> There you are then
[02:43:57] <edwardocallaghan> darkcmd: it's a idea ??? Worth having a think
[02:44:10] <mritun_> well I can speak for the T1
[02:44:21] <mritun_> my team simulated it
[02:44:29] <boyd> It seems that booting to openbios on a flash drive would achieve much of the same thing wouldn't it?
[02:44:32] <mritun_> the code works
[02:45:09] <boyd> Hmm... boot to openbios rather than GRUB.
[02:45:23] <Tpenta> boyd: sounds like DCA to me
[02:45:25] <boyd> Of course, then it's not a BIOS as such...
[02:45:28] <darkcmd> you can boot openbios instead off of flash?
[02:45:29] *** dj2 has joined #opensolaris
[02:45:38] <boyd> Tpenta: Yes, that's where I was going :)
[02:45:50] <boyd> darkcmd: I have no idea
[02:46:05] <edwardocallaghan> GRUB is a bootloader by the way, sure you knew that
[02:46:07] <Tpenta> boyd: which to build required an obsolete microsoft compiler and an NT box
[02:46:19] <mritun_> isn't boot time more a function of getting services up than the base OS alone ?
[02:46:24] <boyd> Well, we need not carry all that cruft
[02:46:49] <Plaidrab> The most important part of OpenBoot is POST anyway
[02:46:51] <mritun_> so then what's the benefit of using OB
[02:46:52] <boyd> mritun_: Well, sure... but I think that linuxbios poweron->kernel running in 3 seconds has to help
[02:47:02] <mritun_> (except getting a nerdy FORTH shell )
[02:47:15] <boyd> mmmmm.... nerdy forth.....
[02:47:37] <dlg> mritun_: personally i like the support for serial stuff
[02:47:38] <mritun_> boyd: sure... but then Oracle RAC takes bloody ages to startup that saving of 10 secs is moot
[02:47:47] <boyd> No doubt
[02:48:24] <mritun_> dlg: agreed. serial console is a killer feature
[02:48:52] <dlg> a lom on top of that is very easy
[02:49:10] * boyd reads about zulu... good... lucreate really needed more command-line features.... NOT!
[02:49:47] <richlowe> it's one extra colon.
[02:50:14] <boyd> Trouble is if you only offer serial console then you rule out windows
[02:50:46] <edwardocallaghan> I think that drivers that have firmware to load should be in the 'BOIS'
[02:50:54] <edwardocallaghan> *'BIOS'
[02:50:59] <boyd> richlowe: True, but there are already so many colons/commas, etc, "detach,attach,preserve"
[02:51:07] * mritun_ thinks windows was never meant for servers... it was retro-fitted
[02:51:07] *** yongsun has joined #opensolaris
[02:51:32] <boyd> mritun_: Of course. The graphics console is a dead giveaway
[02:51:33] <edwardocallaghan> Why use Windows when there are doors?
[02:51:58] <mritun_> edwardocallaghan: if the customers pay, you do it ;)
[02:52:06] <Plaidrab> No one plays door games anymore?
[02:52:15] <boyd> door games?
[02:52:22] <mritun_> @-)
[02:52:43] <mritun_> mmmm games
[02:52:57] <mritun_> I've seen paddle ball being played on IRC
[02:53:07] <mritun_> (clip is on bash.org)
[02:53:20] <mritun_> (no kidding)
[02:53:37] <boyd> |.      |
[02:53:40] <Auralis> | o
[02:53:43] <Plaidrab> BBS terminology. There were a couple of semistandard ways to "spawn" pluginable games. Through "Doors"
[02:54:01] <boyd> Oh yeah... sorry, that info was paged out
[02:54:02] <edwardocallaghan> Oh dear, I did not start that ^ Sir...
[02:54:12] <mritun_> | .                                        |
[02:54:23] <boyd> |                 .                 |
[02:54:25] <Plaidrab> : chuckles at boyd.
[02:54:50] <edwardocallaghan> Pongo running on Vista now at full speed
[02:55:38] <mritun_> Btw, has anyone see the "Plea for help... Thumper w ZFS/NFS" thread ?
[02:56:03] <mritun_> I have a strange feeling that the OP is from Joyent
[02:56:19] <jmcp> mritun_: yes, Benr is from Joyent
[02:56:23] <mritun_> and this is the same thumper they lugged out in a car boot from SUN HQ
[02:56:28] <jmcp> he's been there for a few months now
[02:56:31] <mritun_> aha
[02:56:43] <mritun_> two things ave it away
[02:56:46] <mritun_> 1. Thumper
[02:56:54] <mritun_> 2. The pool name "joyous"
[02:57:14] <mritun_> zfs is getting some bad press now a days
[02:57:42] <boyd> mritun_: Most people n the community know that ben is there... It's no secret
[02:57:50] <boyd> It's getting some good too...
[02:57:54] <mritun_> inspite of being very simple... people are using it wrongly and blaming it to point out the problems
[02:58:06] <mritun_> boyd: I didn't
[02:58:23] <boyd> Well... I did say "most" not "all"....
[02:58:29] <mritun_> I twitch when I see people shooting the messenger
[02:58:50] <edwardocallaghan> OK after about a year of thought on this
[02:58:56] <mritun_> zfs says you've got a problem and then people start blaming it for "eating their data"
[02:59:13] <edwardocallaghan> I think stupid people who just can't use a computer should not be on a computer
[02:59:14] *** hile__ has quit IRC
[02:59:17] <mritun_> (cf.  The zfs ate my pool thread)
[02:59:33] <mritun_> edwardocallaghan: worse. They're managing TB arrays
[02:59:40] <edwardocallaghan> Like a car, you need a license and if you run over people that's bad
[03:00:09] <mritun_> I think it's rather a problem of mediocrity
[03:00:10] <edwardocallaghan> lol
[03:00:19] <mritun_> people accept mediocre systems
[03:00:40] <edwardocallaghan> People should have a license to use a computer I think
[03:00:44] <mritun_> and when a superior system points out where they've crapped, they blame it
[03:00:51] <mritun_> edwardocallaghan: precisely
[03:00:56] <edwardocallaghan> And it should be part of school with a test at the end
[03:01:24] <edwardocallaghan> Thanks I am glad you agree!
[03:01:29] <mritun_> I think it should be something like HAM radio process
[03:01:34] <edwardocallaghan> In fact I am very glad
[03:01:42] <edwardocallaghan> Yes indeed !
[03:01:46] <mritun_> you write exams and do some practical training and get Levels of licenses
[03:01:57] <edwardocallaghan> Yes yes !
[03:01:58] <mritun_> 1. Luser
[03:02:03] <edwardocallaghan> lol
[03:02:05] <mritun_> 2. power user
[03:02:06] *** halton has joined #opensolaris
[03:02:07] <mritun_> 3. geek
[03:02:10] <mritun_> 4. nerd
[03:02:13] <mritun_> 5. ubergeek
[03:02:18] <mritun_> likewise
[03:02:48] <edwardocallaghan> 6. Ultra amazing cool Unix man
[03:02:54] <mritun_> naah
[03:03:02] <mritun_> 6. Solaris admin :D
[03:03:12] <edwardocallaghan> same thing ;)
[03:03:13] <Error_404> 7. guru
[03:03:48] <mritun_> 8. Lord of the blings
[03:04:04] <mritun_> (the blinken-lights server room ref.)
[03:04:19] <edwardocallaghan> 10^100 Can configure sendmail
[03:04:24] <mritun_> LOL
[03:04:25] <mritun_> :D
[03:04:32] <edwardocallaghan> ;)
[03:04:41] * mritun_ shakes his head
[03:04:52] <mritun_> did you read the guy's interview ?
[03:04:55] <edwardocallaghan> Well it did it once when I was 14
[03:05:11] <mritun_> he who invented internet mail alias sendmail-author ?
[03:05:11] <edwardocallaghan> Sorry what guy?
[03:05:20] <edwardocallaghan> No?
[03:05:31] <edwardocallaghan> Have you got a URL to it please?
[03:05:44] <mritun_> he's sorry :D
[03:05:54] <mritun_> and thinks the effort "wasn't worth it" ;)
[03:06:04] *** jbalint has joined #opensolaris
[03:06:11] <mritun_> http://www.theregister.co.uk/2006/12/01/email_troublesome_inventor_says/
[03:07:05] <mritun_> apparantly he was not paid for it
[03:07:16] <mritun_> and the returns weren't great either
[03:07:20] <mritun_> and he's sad
[03:07:34] <mritun_> and apparantly it shows in the sendmail.cf :P
[03:11:41] *** karlo has left #opensolaris
[03:13:43] <edwardocallaghan> Thanks made a interesting read.
[03:13:53] <mritun_> np
[03:14:04] <edwardocallaghan> How hard is a SCSA ?
[03:14:05] <mritun_> he went to school with Bill Joy !!
[03:14:21] <edwardocallaghan> Yes that did surprise me
[03:14:24] <dlg> edwardocallaghan: what are you asking?
[03:14:32] <edwardocallaghan> That was the main point I was looking at
[03:14:34] <Tpenta> i like the comment about people following bill around in th hope of picking up somethign they could turn into a PhD
[03:14:39] <mritun_> yeps
[03:14:41] <mritun_> :D
[03:14:49] <mritun_> amazing
[03:15:00] <mritun_> and how the timing is so important in startups
[03:15:05] <edwardocallaghan> Sun Certified System Administration
[03:15:16] <edwardocallaghan> SCSA
[03:15:17] <mritun_> that guys didn't hot millions because email lagged behind
[03:15:59] <mritun_> It'd morning now... I should go get some tea
[03:16:33] <edwardocallaghan> lol
[03:16:55] <edwardocallaghan> Cool that's very like me mritun_
[03:18:00] <mritun_> where are you located ?
[03:18:16] <dlg> not the scsi midlayer thing in solaris
[03:19:05] <edwardocallaghan> London
[03:19:14] <edwardocallaghan> Although I am moving to Australia
[03:19:25] <edwardocallaghan> To study and try to live there
[03:19:32] * boyd wonders how edwardocallaghan will go moving from London to Canberra
[03:19:49] <edwardocallaghan> Oh you remember
[03:19:57] <mritun_> wow
[03:20:03] <mritun_> quite a distance
[03:20:04] <mritun_> :d
[03:20:16] <edwardocallaghan> for a 19year old I guess
[03:20:17] <mritun_> I'm in India
[03:20:20] <boyd> edwardocallaghan: I remember.... there's not that many of us in Aus
[03:20:38] <edwardocallaghan> Oh sorry are you in Australia?
[03:20:52] <Tpenta> boyd, jmcp & I are in Oz
[03:21:07] <edwardocallaghan> Oh great, good on you guys
[03:21:16] <mritun_> haha
[03:21:36] <mritun_> should've guessed.... not many from west likely at this time
[03:21:44] <edwardocallaghan> Looking for anyone at my age to sponsor?
[03:22:54] <edwardocallaghan> Well, us type of people can stay up at all sorts of time;)
[03:24:14] <mritun_> yeps
[03:24:27] <mritun_> but next day of work limits the timings
[03:24:34] <edwardocallaghan> hehe
[03:24:35] <mritun_> people have to show up at work
[03:24:45] <edwardocallaghan> Do they hehe
[03:24:52] <mritun_> mostly :D
[03:24:53] <edwardocallaghan> Sure sure
[03:24:58] <edwardocallaghan> :D
[03:25:38] <edwardocallaghan> So anyone in Oz want to sponsor me?
[03:26:15] <edwardocallaghan> I can speak English unlike most people in London...
[03:26:18] <edwardocallaghan> :p
[03:26:58] <mritun_> Lol
[03:26:58] <edwardocallaghan> The British education system, well it's not a system. It's junk !
[03:27:13] <mritun_> why would you say so ?
[03:27:25] *** stevel has joined #opensolaris
[03:27:26] *** ChanServ sets mode: +o stevel
[03:27:41] <Tpenta> hi steve
[03:27:42] <Plaidrab> The SCSA is somewhat hard. The 10 is harder than the 9
[03:27:44] <mritun_> The Indian education system is mostly based on british model... it's bad, but even then not totally junk
[03:28:05] <stevel> hey alan
[03:28:20] <Plaidrab> It expects you to know x86 and Sparc
[03:28:53] <edwardocallaghan> Well people need a calculator to do logs !!! How stupid, logs are so you don't need a calculator
[03:28:58] <Yamazaki-kun> mritun_: Did you see the recent NYT article on Indian higher education?
[03:29:07] <mritun_> Yamazaki-kun: yes I did
[03:29:14] <Yamazaki-kun> I hope that was the worst 5%.
[03:29:37] <mritun_> Yamazaki-kun: Usual xenophopic stuff that NYT gets paid to write, I'd say
[03:30:01] <Yamazaki-kun> I'm not sure I'd go for xenophobic... but maybe NYC-centric.
[03:30:02] <edwardocallaghan> Well by the age of 20 I will have a LPIC1 and be working on a SCSA in my own time with my CCNA
[03:30:32] <mritun_> NYT has to cater to a large segment of people
[03:30:58] <mritun_> so it usually publishes what is the dominant political group at the moment
[03:31:07] <edwardocallaghan> Do you guys in Oz think I can get sponsor with that kind of thing or do I need to work harder?
[03:31:26] <mritun_> you remember, not too long back they published the equation of IIT = Stanford + MIT + Harvard
[03:31:29] <edwardocallaghan> p.s. Sorry for being off top
[03:31:32] <Tpenta> its not something that I've realy looked in to
[03:31:39] <mritun_> quite embarrasing
[03:35:35] <mritun_> Yamazaki-san, brb
[03:35:51] *** mritun_ is now known as mritun^breakfast
[03:46:33] <edwardocallaghan> Oh I really want to get my site up and running
[04:01:47] <onbot> commit by xz162242:  6501277 "dladm reset-linkprop <link>" causes a segmentation fault
[04:02:58] *** freakazoid0223 has joined #opensolaris
[04:08:42] *** peteh has quit IRC
[04:09:22] *** peteh has joined #opensolaris
[04:10:42] *** sol1 has joined #opensolaris
[04:10:58] <sol1> tomww: hey.. it still doesn`t work
[04:11:36] <sol1> you got to help me..
[04:12:21] *** sol1 has quit IRC
[04:12:36] <edwardocallaghan> Good night guys all the best, Thanks again
[04:12:39] *** edwardocallaghan has left #opensolaris
[04:12:44] <stevel> and you had approximately 45 seconds in which to help him
[04:13:11] <Tpenta> my b54 bins are nearly done
[04:13:34] <Tpenta> well the x86 ones are nearly done, sparc wil be a bit longer yet
[04:14:02] *** stevel changes topic to "Latest builds: SXCR: 53 | ON build: 54"
[04:14:10] <stevel> (sorta, not really)
[04:14:14] <stevel> the 54 bundle is done
[04:14:23] <stevel> the 54 delivery is churning now (for me too)
[04:14:24] <Tpenta> is it up yet?
[04:14:29] <stevel> the bundle is
[04:14:52] <Tpenta> my x86 one is into te final non-debug build, the sparc one has not finished the first debug build
[04:14:55] <stevel> my 54 didn't work because the finally updated the EK to be 11 instead of 10.1
[04:15:02] <stevel> s/the/they/
[04:15:03] <Tpenta> EK?
[04:15:05] <stevel> so i had to rerun it
[04:15:08] <stevel> encryption kit
[04:15:22] <alanc> isn't that dead yet?
[04:15:25] <Tpenta> ok, any changes to bindrop i need to know abt?
[04:15:33] <stevel> yeah, s/10\.1/11/
[04:15:40] <stevel> i hope that's all :)
[04:15:43] <stevel> alanc: is it dead already?
[04:15:51] <stevel> i thought it was planned, but not done yet
[04:16:06] <richlowe> I didn't think it was done.
[04:16:07] <Tpenta> not done yet, and no-one has spoken with danek about it
[04:16:10] <alanc> not sure - I know they were working on killing it
[04:16:36] *** Yamazaki-kun has quit IRC
[04:16:51] <Tpenta> which reminds me that I shoudl ping arren about it
[04:17:38] <Tpenta> OK, I hacked /opt/onbld/bin/bindrop rather than fixing and redelivering onbld to my inprogress build boxen :)
[04:18:08] <dduvall> bubbva keeps talking about the crypto packages going away, but it's all vaporware as far as I'm concerned.  :)
[04:18:38] <Tpenta> oh hi danek, ears burning? ;)
[04:18:47] <dduvall> beeping, more like it.  :)
[04:18:50] <Tpenta> stevel: did you see the hal bug i logged?
[04:18:58] <stevel> nope. should i have?
[04:19:15] <Tpenta> sorted out what the problem is with my external disk not mounting stuff automagically now
[04:19:36] <Tpenta> looks like if any partition has an SMI label on it, hal doesnt go and check the other logical partitions
[04:19:42] <stevel> damnit. klinsmann turned down the US job. we're getting Bradley
[04:20:05] <alanc> darrenm did file a PSARC fasttrack to kill it, 2006/610, but that's the last I heard
[04:20:16] <Tpenta> that's how i know abt it
[04:21:28] <Tpenta> anyone know how to tell teamware that I no longer want a particular file to be in the list of modified files?
[04:21:37] <Tpenta> say for example usr/src/Makefile ;)
[04:21:44] <stevel> are you using wx?
[04:21:47] <Tpenta> yea
[04:21:50] <Tpenta> wx unedit?
[04:21:54] <alanc> though given the gate limitations for nv_54 & 55, it's not surprising it hasn't gone farther yet
[04:22:06] <stevel> do you want to revert it?  or just remove it from the active list?
[04:22:20] <Tpenta> well the fix that it has in it is now in the gate
[04:22:51] <stevel> wx reedit, and then wx unedit
[04:23:26] <stevel> is how i revert files, and then bringover to grab the change from the gate
[04:23:43] <Tpenta> what does reedit do?
[04:24:04] <stevel> edits the file with the same delta (i.e.: doesn't give it a new delta)
[04:24:19] <Tpenta> ahhh i tend to do a redelget anyway
[04:24:25] <Tpenta> so theer is only one delta
[04:24:39] <stevel> right, so what you are doing is putting the file into 'edit mode' on that one new delta
[04:24:48] <stevel> and then 'unedit'ing it to undo that delta
[04:25:02] <Tpenta> oh i get it
[04:25:15] <Tpenta> ok, I'll do athta AFTER i do my b53 zfs snapshot before the b54 bringover
[04:25:23] <Tpenta> a/athta/after
[04:25:32] <Tpenta> gawd my typiong sucks today
[04:26:09] <richlowe> wx reset?
[04:26:36] <richlowe> wx reset <file>, rather, I guess.
[04:26:57] <richlowe> though now I see the comment in the help, perhaps not.
[04:27:54] <whaq> Is anyone here handling the storage subsystem (raidz, sata)? I'm having problem w/ intermittent 'inactivities' of a raidz set . Whenever I do long operations (copy big files, scrub raidz) things will just stall for no apparent reason every now and then. Has anyone experienced this or may know what is wrong with it?
[04:31:38] <Tpenta> actually richlowe, reset looks like what i want
[04:31:57] <Tpenta> but I'll try both, the wonder of snapshot/rollback
[04:32:20] <richlowe> Tpenta: yeah, the "if something goes wrong, restore from your last backup" warning caused me concern, however. :)
[04:34:21] <dwc-> backups are good
[04:34:42] <Tpenta> that's what snapshot rollback is for ;)
[04:35:05] <richlowe> dwc-: the implication that problems are more expected than usual isn't though. :)
[04:36:40] <dwc-> yea, well...
[04:36:59] *** karrotx_ has quit IRC
[04:37:43] *** karrotx_ has joined #opensolaris
[04:40:41] *** mritun^breakfast has left #opensolaris
[04:44:18] *** vmhobbes- is now known as vmhobbes
[04:51:48] *** laca has quit IRC
[04:55:12] *** ShadowHntr has joined #opensolaris
[04:55:49] *** jamesd has joined #opensolaris
[04:57:35] *** unixconsole has joined #opensolaris
[05:02:33] *** bobbyz has joined #opensolaris
[05:05:46] <unixconsole> quiet in here
[05:05:49] <bobbyz> I don't mean to be a pain since I'm sure you guys have already been getting this question, but is Sol 10 11/06 still planned to be released?  :)
[05:06:00] <boyd> yes
[05:06:13] <jmcp> bobbyz: real soon now
[05:06:17] <jmcp> no idea just when though
[05:06:22] <unixconsole> a lot of stuff is suppose to go into 11/06..
[05:06:36] <bobbyz> awesome!  :)  I was just debating about whether or not I should download SXCR: 53, but if 11/06 is still coming out I'll wait
[05:07:04] <bobbyz> sounds good...I'd rather the release be stable than be rushed out the door
[05:07:07] <boyd> I'd guess before christmas. But how long before.....
[05:07:34] <unixconsole> well 11/06 is still sol10.. sxcr:53 is nevada.
[05:07:47] <unixconsole> however, 11/06 will have some nice upgrades:)
[05:08:31] <boyd> certainly b53 will have way more new things that 11/06.... and 11/06 will have some things that are not in nevada
[05:08:42] *** Gman has joined #opensolaris
[05:08:57] <stevel> 'evening glynn
[05:09:06] <Gman> hi steve
[05:09:08] <stevel> isn't it an ungodly hour in ireland for you?
[05:09:13] <Gman> yeah, 4am
[05:09:16] <Gman> jetlag, fun!
[05:09:27] <stevel> heh
[05:09:38] <Gman> [also sick, but mostly jetlag]
[05:09:45] <jmcp> Gman: hi
[05:09:50] *** uebayasi has quit IRC
[05:09:51] <unixconsole> hi
[05:10:07] <unixconsole> yeah, 11/06 will have support for some new products:)
[05:10:19] <bobbyz> How about sata changes?  I should probably just buy a better sata controller, but I'm being stubborn  :)  I've got a sii3114 that runs ridiculously slow and a promise tx4 that isn't supported at all.  So I was going to try one of the new releases out and see if I can get things working better
[05:10:19] <Gman> jmcp, unixconsole: howdy
[05:11:37] <bobbyz> For the meantime I'm using linux since it supports my tx4, but it drives me nuts using md software raid5 and lvm2 when zfs is sitting out there
[05:11:50] <unixconsole> don't know about sata changes, I don't recall seeing much in the update3 beta.
[05:11:55] <bobbyz> gotcha
[05:12:44] <unixconsole> definitely a lot nevada backports though.
[05:13:35] <bobbyz> well I'm still excited regardless of whether either of my lame sata cards works better  :)
[05:15:52] <unixconsole> anyone here with an ultra20?
[05:17:20] <boyd> unixconsole: I think jmcp has one (M2)
[05:18:47] <unixconsole> jmcp: do you like the m2? I was thinking about buying one. but wanted to ask around about it first.
[05:19:28] <Plaidrab> I need to get a SATA card too. I forgot I only had two spots on my board. :)
[05:19:53] <Plaidrab> But all I find are combo cards or Raid. I just want a couple of extra ports. heh
[05:19:56] <jmcp> unixconsole: yes, I've got the u20m2 ... I like a lot
[05:20:06] <jmcp> unixconsole: it's based on a tyan s2865 mobo iirc
[05:20:09] <jmcp> dual-core
[05:20:18] <Plaidrab> drooly...
[05:20:39] <jamesd> i have a u20
[05:20:50] <unixconsole> jpmc: yeah, I figured out what mobo it's based on. I found that if I tried to build my own, I wouldn't save much money.
[05:21:59] <boyd> You won't get support like sun's on a self-build either :)
[05:22:15] <unixconsole> this is true.. and I'd be missing a logo and a kewl case:)
[05:22:34] <unixconsole> have you added anything to it?
[05:22:47] <unixconsole> what video to you recommend?
[05:27:10] <jmcp> unixconsole: I have the fx560 card and 2 320gb seagate sata disks which I mirror with svm (root, /usr, /var, /opt) and the rest is done with zfs
[05:27:21] <jmcp> I got a good deal on the whole config
[05:28:54] <jamesd> jmcp, why put /opt as   svm? why not zfs
[05:30:34] <jmcp> jamesd: because I couldn't be bothered, basically. Well, to be correct, I have the basic /opt on svm, and I mount zfs filesystems on /opt/{local|csw|SUNWspro|...}
[05:31:05] <jamesd> oh
[05:32:47] <unixconsole> I never break /usr or /var from /, the use of quotas and logadm prevent that need. but I use zfs a lot for my jet environment (/opt/SUNWjet and /export/install. I also use it for my home dirs, zones, backups, etc.
[05:33:50] <jmcp> unixconsole: I haven't broken them out from the actual / slice
[05:34:18] *** alanc is now known as alanc-away
[05:34:35] <unixconsole> jmcp: oh okay
[05:35:20] <jmcp> I just have this habit of specifying them all separately otherwise people make assumptions :)
[05:36:17] <boyd> You use quotas!?!
[05:37:34] <Plaidrab> mmm. quotas
[05:37:44] <unixconsole> not personally.. but for a couple of shops quotas are a good way to keep ppl from filling up disk space. the advent of monitoring software for things like disk space has reduced the need in most shops.
[05:38:47] <unixconsole> the main concern for most companies seems to be /var because of logs. and that's when it's time to introduce ppl to logadm.
[05:39:19] <boyd> Yeah, logadm is pretty nice
[05:39:48] <unixconsole> not a bad replacement for stuff like logrotate
[05:40:26] <unixconsole> but quotas are very handy in web hosting companies that pile tons of customers on each node.
[05:44:07] *** freakazoid0223 has quit IRC
[05:45:56] <whaq> bobbyz, what's wrong with your sil3114?
[05:47:06] *** jamesd_ has joined #opensolaris
[05:48:37] <bobbyz> whaq: It recognizes all of the attached SATA drives, but data written to or read from the drives hits a limit of about 5-6MB/s, even when being written through raid5 or a zfs raidz pool.  These same drives on the same computer perform above 20MB/s on the promise tx4 controller with linux, and I'm quite certain linux isn't nearly as efficient or speedy as solaris
[05:48:42] <unixconsole> seems like the ultra20m2 is a variant of this mobo http://www.tyan.com/products/html/tomcatn3400b.html
[05:49:40] <jmcp> bobbyz: what's the value of maxphys in your kernel?
[05:49:48] <bobbyz> I tried changing maxphys to a higher value as per suggested by jcmp, but that darn sii3114 card just doesn't want to perform better
[05:49:51] <jmcp> :(
[05:49:53] <jmcp> bugger
[05:49:56] <bobbyz> yeah  :)
[05:50:01] <whaq> bobbyz, ah .. I'm sort of in the same boat. Having a sil3132 based controller. Each drive can do >20MB, but they stutter.. sort of. Cannot sustain high throughput. What's a good sata card then?
[05:50:28] <bobbyz> whaq: no clue.  I love my promise tx4, but it's not supported on solaris
[05:50:42] <whaq> marvell based ones perhaps?
[05:51:08] <unixconsole> probably have to buy from from lsi logic
[05:51:18] <bobbyz> which stinks because I'd much rather use solaris than linux, but I don't have the money to purchase a good high-end card for personal use  :)
[05:51:48] <unixconsole> perhaps something in the hcl?
[05:51:49] <whaq> why would you get a RAID card if zfs\raidz won't be using its facilities though?
[05:53:02] <bobbyz> whaq: I personally would just by a better sata controller and not use any raid functions on it if I were to buy again, but that's just me
[05:54:06] <bobbyz> of course you could still use zfs and with a hardware raid 5 and just not make a raidz pool
[05:54:43] <whaq> and be exposed to the raid5 write-hole?
[05:54:58] <boyd> I'm with you bobbyz (on the non-raid controller)
[05:54:58] <jamesd_> bobbyz, and live with the write hole.. and have no redunctancy for checksum errors
[05:55:40] <bobbyz> hey, I said I'd rather use a non-raid controller, but even people doing hardware raid could still benefit from the cool features of zfs
[05:55:40] <boyd> He did present it as the less favoured option
[05:55:41] <whaq> sonnet has a marvell based 4 & 8 port sata controller.. but it's pci 4x
[05:55:49] <whaq> true
[05:56:43] <whaq> i'm dying to use this http://www.raidon.com.tw/web/pro-st6600.htm but the Sil3132 is flaky on my machine for some reason.
[05:56:51] <Triskelios> anyone know about serial ports on a netra?
[05:57:06] <unixconsole> triskelios: sure
[05:57:55] <Triskelios> unixconsole, someone  (not me) wants to know the pinout
[05:58:04] <bobbyz> whaq: looks pretty nice
[05:58:56] <unixconsole> what kinda of netra?
[05:59:04] <boyd> Grr ", Zulu doesn't provide support for upgrading zones whose zonepath is on a ZFS filesystem" Damn.. that's the nicest way to make them
[05:59:09] <whaq> bobbyz, only uses a single esata cable to connect to the controller. But as of b50, solaris needs to be rebooted 3-4 times for it to recognize all the ports :(
[05:59:16] <Triskelios> t1 105
[05:59:56] <unixconsole> ah.. hold on a sec and I'll find the pinout info. off the top of my head.. it's the same pinout as a cisco box
[06:00:11] <Triskelios> ahh
[06:00:20] <boyd> living up to your nick, hey unixconsole ?
[06:00:23] <Triskelios> ok, this guy has one of those already
[06:00:40] <Triskelios> so he just needs a null-modem converter
[06:01:17] <Triskelios> boyd, hehe
[06:01:19] <unixconsole> Pin 1  RTS
[06:01:19] 
[06:01:19] 
[06:01:19] 
[06:01:19] 
[06:01:20] 
[06:01:21] * boyd just noticed the links to "Sunsolve 6.0 beta" and clicks with some trepidation
[06:01:22] 
[06:01:24] 
[06:02:44] <boyd> Wow... So far I can't see any difference. This does not seem worth a major version number jump
[06:03:20] <unixconsole> here we go.. I was trying to remember where this was..
[06:03:23] <unixconsole> http://www.sunhelp.org/unix-serial-port-resources/serial-pinouts/#netrat1105.link
[06:03:50] <Triskelios> thanks
[06:04:32] <unixconsole> no prob
[06:06:49] <unixconsole> boyd: yeah, after having to make my own serial cable from a i386 to an ipx back in the 90's, came up with this nickname.
[06:06:54] * dunc_ just finished epic 650 mile journey
[06:06:59] <boyd> unixconsole: lol
[06:07:00] <dunc_> phew
[06:07:21] <boyd> dunc_: You should get, like, a car or something :)
[06:07:33] <unixconsole> or take a plane
[06:07:36] <dunc_> lol, i did have a car
[06:07:51] <boyd> unixconsole: I remember those days. is seemed like every single port in the world was different to every other
[06:07:58] <boyd> s/is/it
[06:10:17] <Triskelios> at least the mini din-8s the IPXes have were fairly common..
[06:10:22] <unixconsole> boyd: agreed. I'm glad it's a little easier these days.
[06:10:42] <unixconsole> the ipxes were fun.
[06:11:30] <unixconsole> I recently had to deal with a 15k in our lab that had lost one of the mini-din -> db25 cables. luckily someone found it;)
[06:12:17] * boyd was at DEC... we had mostly MMJ, oh, and RJ11, RJ12, RJ45, DB9, DB25, male, female, null modem, HW flow control or none and some others that I've no doubt forgotten
[06:12:18] <unixconsole> at one point I had an elc.. a really funky one that had a clear case for some reason.
[06:12:21] <Triskelios> I use the IPX at home as a console for an SGI machine since I don't have an adapter for the mini din there
[06:13:04] <unixconsole> yeah, the DEC stuff was pretty interesting:)
[06:13:20] <unixconsole> I miss the lk401? keyboard sometimes.
[06:13:37] <unixconsole> I did like srm.. good stuff.
[06:13:42] <jamesd_> Triskelios, the ipx and sgi most likely use the same cable.. my ipx and  indy and indigo2  all use the same cable
[06:14:21] *** jamesd has quit IRC
[06:14:22] <boyd> unixconsole: yeah srm was good
[06:14:28] <Triskelios> jamesd_, which is why I use a regular mini din cable between them, yeah
[06:14:54] <boyd> unixconsole: I don't miss the lack of an escape key on some of their kbs
[06:15:05] <unixconsole> boyd: too bad things didn't work out with the alpha in the long run. digital unix/tru64 was an interesting trip.
[06:15:42] <boyd> Sure was.. the chip was ahead of it's time. Could have done well... 64bit unix in ~1992
[06:15:53] <unixconsole> boyd: yeah, the missing esc key and some of the funky symbols on the keys definitley confused ppl.
[06:16:17] <boyd> But DEC couldn'd market to save themselves and neither CQ or HP knew what to do with the stuff
[06:16:41] <boyd> Trucluster was quote nice too
[06:16:45] <boyd> quite
[06:16:52] <unixconsole> boyd: sadly I spent time trying to help the slackware guys port to the alpha.. then the compaq/hp thing kinda sunk all the interest.
[06:17:09] <unixconsole> very true.. dec had good stuff.. especially on the clustering side.
[06:17:19] * boyd nods... He still has an alpha in a cupboard somewhere
[06:17:29] <unixconsole> but they didn't know how to market their stuff well.
[06:17:42] * boyd sighs
[06:18:05] <unixconsole> I was forced to sell a lot of odds and ends.. included a nice xp1000 workstation and a lx64 clone
[06:18:51] <unixconsole> I remember when amd was suppose to release a mobo with the ev5 socket that would take a k7 chip and an alpha chip. never materialized:(
[06:18:53] * boyd feels more at home here. On a quick visit to #vim a few weeks ago he mentioned that he'd learned vi on a serial terminal and they treated him like Encino man
[06:20:44] <unixconsole> well after working at Sun, I've been on the solaris bandwagon. definitely quieter at home without a bunch of dec's, pc's, and old ipx's. makes the wife happy;)
[06:20:56] <boyd> DEC has a thing.... umm Centris XL? That had the pentium on a daughterboard and you could put an alpha daughterboard in instead
[06:21:18] * Triskelios eyes the DEC storageworks array whining in the corner...
[06:21:28] <boyd> unixconsole: I know what you mean. I had a neighbour complain about the noise from my Alpha's disk once
[06:21:30] <unixconsole> but I gotta get rid of my netra x1 and probably the ultra60. that's why I'm thinking the ultra20m2 would be a good replacement.
[06:21:31] <Triskelios> (doing ZFS of course)
[06:22:02] <boyd> Heh... and I'm just making way for a SB100 and a couple of e250s
[06:23:10] <unixconsole> I was thinking of getting sb1000, but considering it uses more power than my ultra60 and the storage is expensive.. the ultra20m2 seems like a better deal. personally.. I'd be happy with a t1000.
[06:23:21] <unixconsole> but that would cross the noise barrier:)
[06:23:27] <boyd> Hehe
[06:23:28] *** yongsun has quit IRC
[06:23:33] <unixconsole> and the $$ barrier:)
[06:23:43] <boyd> BTW I really did mean a sb100, not 1000 :)
[06:24:25] *** gm152 has quit IRC
[06:24:37] <unixconsole> the sb100's are okay, but in some ways I rather the ultra10 for a USIIi.
[06:25:21] <boyd> Yeah, I agree... but I can get 2GB in this one and it's a slight faster CPU than the U10 I have now
[06:25:22] *** yongsun has joined #opensolaris
[06:25:31] <unixconsole> this is true.
[06:25:31] *** yongsun has quit IRC
[06:25:32] <boyd> And I think the IDE controller is less awful
[06:25:44] <richlowe> I think it's only marginally less awful.
[06:25:47] <unixconsole> true as well.. but I do like the u10 case.
[06:26:08] <boyd> Sure, but every step up from the pit of hell is still a step up :)
[06:26:22] <unixconsole> I'm just happy my u60 has a noise reduction kit on it.. definitely makes a difference.
[06:26:36] <boyd> Plus I have some sunswifts here so I may go the SCSI option
[06:27:20] <boyd> unixconsole: What exactly is in the noise reduction kit? A padded box you put over the top?
[06:27:24] <unixconsole> I just need to bum a sun employee's discount for an ultra20m2. but I gotta sell the stuff I have now.
[06:27:59] <unixconsole> no.. it's acutally a temp sensor with a voltage regulator that you connect between the power supply and the fans.
[06:28:21] <unixconsole> sun made it as an option and included it in the last revs of the u60.
[06:28:53] * boyd nods
[06:29:11] <unixconsole> definitely makes things nicer;)
[06:29:41] <unixconsole> I also added usb2.0 to it.. good for external usb hdd for backups:)
[06:30:19] <boyd> How generic a USB card will it take?
[06:30:20] <unixconsole> would be nice to add sata to it.. haven't tried that.
[06:30:29] <ShadowHntr> i'd like a gigaswift for my u5
[06:30:33] <boyd> I'm wondering about that too
[06:30:40] <unixconsole> I bought a belkin pci card at best buy
[06:30:41] <boyd> ShadowHntr: Where are you?
[06:30:48] <ShadowHntr> US/Tennessee
[06:30:52] <ShadowHntr> Nashville/Chattanooga
[06:31:14] <boyd> I could ship one but the shipping would be more that the cost of buying one I think
[06:31:25] <ShadowHntr> well
[06:31:32] <ShadowHntr> i've heard something around 400 for it
[06:31:52] <boyd> Yikes! That seems... extreme
[06:32:03] <unixconsole> work's like a charm.. I even use a usb iogear kvm to switch between my u60 and my laptop.. use a sun type6 kb and a logitech trackball mouse.
[06:32:19] <boyd> Nice... I may have to look into that...
[06:33:01] <unixconsole> but I keep a type 5 around since the obp doesn't know anything about usb kb
[06:33:22] <ShadowHntr> i'd love to have gigabit on copper on my Ultra 5
[06:33:26] <ShadowHntr> but i doubt it would happen.
[06:33:29] <boyd> Ah... yeah I thikn you need OBP4.x for that. It works on the blade 100 anyway
[06:33:36] *** Sieghard has joined #opensolaris
[06:33:58] <unixconsole> true.. but the USII 450's are better cpus..
[06:34:05] <boyd> ShadowHntr: Oh, sorry I saw gigaswift and read sunswift... I only have them (hme)
[06:35:10] *** Sieghard has quit IRC
[06:35:22] <ShadowHntr> don't worry bout it
[06:35:26] <ShadowHntr> it works fine in 100mbit mode
[06:35:43] <unixconsole> still a second hme is not bad for ipmp;)
[06:37:57] *** vmhobbes has quit IRC
[06:38:32] *** calvin_ has joined #opensolaris
[06:41:42] *** nexrafa has joined #opensolaris
[06:43:47] <boyd> Sheesh...  370000 acres burned in bushfires here. And it's only the start of summer
[06:45:18] *** vmhobbes has joined #OpenSolaris
[06:47:04] <ShadowHntr> anyone here have a 400MHz processor module for an Ultra 5?
[06:47:10] <ShadowHntr> or will i have to go on ebay to get it :P
[06:47:24] <ShadowHntr> boyd: sucks.
[06:47:27] <ShadowHntr> hope nobody's hurt
[06:48:02] <boyd> ShadowHntr: Not yet AFAIK
[06:48:13] <ShadowHntr> boyd: where you based, canada?
[06:48:22] <boyd> Victoria, Australia
[06:48:26] <ShadowHntr> oy vey
[06:48:34] <ShadowHntr> i saw the on.net and guessed
[06:48:38] <boyd> I don't think it's the start of summer in Canada :)
[06:48:39] <ShadowHntr> yeah that would be expensive.
[06:48:51] <boyd> ShadowHntr: Ah, Ontario :)
[06:48:51] <ShadowHntr> boyd: i wasn't paying attention to who i was talking to. heh.
[06:51:00] <nexrafa> i hate making storage configurations at 3:50am
[06:51:13] <nexrafa> im not seeing anything clear at this point
[06:51:13] <nexrafa> :(
[06:51:21] <nexrafa> to much numbers, names, volumes disks for me
[06:51:22] <nexrafa> hehe
[06:52:12] <ShadowHntr> lol
[06:52:21] <ShadowHntr> i'm worried that i'll get lost when i try to learn SANs
[06:52:28] <ShadowHntr> i barely know anything about network attached storage either
[06:52:37] *** halton has quit IRC
[06:52:40] <ShadowHntr> the advantages and disadvantages of both.
[06:52:48] <nexrafa> ShadowHntr: hehe
[06:53:04] <nexrafa> I've done it lots of times and i get lost also ;)
[06:53:05] <nexrafa> hehehe
[06:53:20] <nexrafa> im making an spreedsheat with 300 lines of volumes, disks
[06:53:26] <nexrafa> wwns, storages, luns
[06:53:27] <ShadowHntr> oy.
[06:53:32] <nexrafa> virtual disks, domain profiles
[06:53:33] <nexrafa> bleh bleh
[06:53:36] <ShadowHntr> what's lwps stand for?
[06:53:38] <ShadowHntr> in prstat
[06:53:47] <nexrafa> light weight process
[06:53:48] <ShadowHntr> throwing off questions while i have my mind on solaris
[06:53:49] <ShadowHntr> :)
[06:53:53] <ShadowHntr> cool thanks
[06:53:57] <nexrafa> :)
[06:54:02] <nexrafa> its a small part of the parent process
[06:54:07] <nexrafa> or a "thread"
[06:54:08] <nexrafa> hehe
[06:54:08] <ShadowHntr> i like prstat
[06:54:13] <ShadowHntr> i'm familiar with threads
[06:54:16] <nexrafa> are u using solaris 10 ?
[06:54:18] <ShadowHntr> i'm not a systems programmer
[06:54:31] <ShadowHntr> but i'm familiar with most concepts. :)
[06:54:33] <ShadowHntr> yeap
[06:54:35] <ShadowHntr> 6/06
[06:54:37] <ShadowHntr> SPARC
[06:54:38] <ShadowHntr> on an Ultra 5
[06:54:45] <nexrafa> ps -ef -o pid,ppid,zone,zoneid
[06:54:54] <nexrafa> or try
[06:54:56] <nexrafa> prstat -Z
[06:54:57] <nexrafa> then
[06:54:58] <nexrafa> prstat -J
[06:55:04] <ShadowHntr> i don't use zones
[06:55:05] <nexrafa> you will get info for zones
[06:55:08] <nexrafa> and for projects
[06:55:16] <nexrafa> and all these stuff, really cool ;)
[06:55:56] <unixconsole> the proc tools in general are very kewl (pfiles, pargs, pcred, pldd, etc.).
[06:56:11] <ShadowHntr> pretty cool
[06:56:17] <ShadowHntr> i use pgrep and pkill all the time
[06:56:18] <ShadowHntr> :)
[06:56:20] <nexrafa> pstack ;)
[06:56:31] <nexrafa> try to make a pstack over a jvm :).. you are going to see
[06:56:38] <nexrafa> all java process and lwp
[06:56:46] <ShadowHntr> i wish i knew about pgrep and pkill when i was in computer science class at university
[06:56:47] <nexrafa> and you can alalyze only one thread of jvm
[06:56:49] <ShadowHntr> i wrote a program
[06:56:54] <nexrafa> doing pstack pid/nlwp
[06:57:02] <ShadowHntr> while (1) { fork(); }
[06:57:07] <ShadowHntr> and forgot the sleep statement
[06:57:08] <nexrafa> hihi
[06:57:17] <nexrafa> ShadowHntr: solaris 10 saves you from that
[06:57:22] <nexrafa> using resource manager ;)
[06:57:23] <ShadowHntr> on HP-UX
[06:57:28] <nexrafa> eargh
[06:57:31] <ShadowHntr> had to reboot an HP 9000
[06:57:36] <nexrafa> hihi
[06:57:41] <ShadowHntr> our university mainframe is an HP 9000
[06:57:48] <unixconsole> yuck..
[06:57:49] <nexrafa> out
[06:57:50] <ShadowHntr> running HP-UX 11
[06:57:51] <nexrafa> outchh
[06:57:52] <nexrafa> hehe
[06:57:53] <ShadowHntr> i'm not happy with it
[06:57:58] <ShadowHntr> but i'm familiar with it
[06:58:05] <ShadowHntr> my next project is learning IRIX on an SGI Indy
[06:58:12] *** nprice has left #opensolaris
[06:58:15] <ShadowHntr> i got at PhreakNIC in Nashville in October for free
[06:58:21] <nexrafa> Based on all these above:
[06:58:22] <nexrafa>
[06:58:22] <nexrafa> 			storage (domain, profiles, luns, virtual disks)
[06:58:22] <nexrafa> 			solaris (volumes, disk groups, disks, sub-disks)
[06:58:22] <nexrafa>
[06:58:22] <nexrafa> Based on SUN's best practices, some notes about the configuration was made:
[06:58:23] <nexrafa> hehehe
[06:58:29] <nexrafa> my doc is a "txt" file now
[06:58:47] <nexrafa> i dont wanna use line-wrap for my email, orelse nobody is going to understand nothing
[06:58:49] <nexrafa> hehehe
[06:59:09] *** stevel has quit IRC
[06:59:57] *** yongsun has joined #opensolaris
[07:05:06] *** sol1 has joined #opensolaris
[07:05:13] <sol1> hey...
[07:05:26] <sol1> elektronkind: hey
[07:05:29] <sol1> tomww: hey
[07:06:28] <sol1> can anyone help me configure my lan card. I just have one small problem left
[07:06:37] *** yongsun has quit IRC
[07:06:47] <sol1> the output of netstat -r is as follows
[07:07:09] <sol1> 192.168.1.0        192.168.1.2 dnet0
[07:07:30] <sol1> 244.0.0.0        localhost lo0
[07:07:41] <sol1> default         192.168.1.1
[07:07:59] <sol1> The default also is supposed to have the interface dnet0
[07:08:20] <sol1> thats the only problem how do i get dnet0 configured as default ?
[07:08:43] <sol1> please someone help
[07:09:12] *** bougie has joined #opensolaris
[07:09:14] <sol1> I have been strugling with this from yesterday..i need internet working in Solaris..please
[07:09:19] <Auralis> i think you are misconcepting here
[07:09:19] <sol1> help
[07:09:32] <Auralis> what is your gateways ip?
[07:09:33] <sol1> ah.. tell me ..
[07:09:38] <sol1> 192.168.1.1
[07:09:47] <Auralis> thend efaukt is correct
[07:09:59] <Auralis> what is the ip of your system? .1.2?
[07:10:00] <sol1> ok..
[07:10:03] <sol1> yes
[07:10:18] <sol1> then why can`t i surf the web
[07:10:18] <Auralis> ok, now what exactly is not working?
[07:10:39] <sol1> i even do this ifconfig dnet0 up
[07:10:43] <Auralis> do you have a /etc/resolv.conf?
[07:10:55] <Tpenta> actually, try this:
[07:10:58] <jmcp> sol1: and please define "cannot surf the web" more precisely
[07:10:58] <sol1> yes
[07:11:00] <Tpenta> tracerout e-n 1.1.1.1
[07:11:08] <Tpenta> traceroute -n 1.1.1.1
[07:11:13] <Tpenta> do you get past your router?
[07:11:28] *** meow_ has joined #opensolaris
[07:11:33] <sol1> i am using a live cd now.. I only have solaris installed so i have to reboot to do this
[07:11:47] *** ddunham_ has joined #opensolaris
[07:12:04] <sol1> Tpenta: i tried this ping 192.168.1.1 it did not work
[07:12:09] <Tpenta> that traceroute will attempt to follow defaults for quite some way without doing name lookups
[07:12:14] <Tpenta> ah
[07:12:19] *** Gman has quit IRC
[07:12:45] *** ddunham_ has quit IRC
[07:12:51] <sol1> jmcp: i tried using a browser...
[07:13:09] *** DataStream has left #opensolaris
[07:13:26] <dlg> yay, hba arrived
[07:13:28] <sol1> Tpenta: i have edited the following files with the foll data just tell me if its right ..
[07:13:31] <dlg> oh, wrong channel
[07:13:51] *** DataStream has joined #opensolaris
[07:14:17] <sol1> /etc/hostname.dnet0 contains 192.168.1.2
[07:14:38] <sol1> /etc/defaultrouter contains 192.168.1.1
[07:14:51] <sol1> /etc/resolv.conf contains nameserver 192.168.1.1
[07:15:10] <Tpenta> do you have other machines on your net?
[07:15:11] <sol1> and /etc/netmasks contains 192.168.1.255 255.255.255.0
[07:15:13] *** you_nix has joined #opensolaris
[07:15:22] <sol1> no this is my only system
[07:15:25] *** you_nix has quit IRC
[07:15:37] <Tpenta> hmmmmm
[07:15:38] *** you_nix has joined #opensolaris
[07:15:41] *** you_nix has left #opensolaris
[07:15:42] <sol1> Tpenta: are those files correct ?
[07:15:45] <Tpenta> it might be interesting to run a snoop while you are trying the pings
[07:15:56] <Auralis> check if you have the line hosts: files dns  in /etc/nsswitch.conf
[07:15:57] <jmcp> sol1: why do you have an IP address in /etc/hostname.dnet0 rather than an actual name?
[07:16:04] <Tpenta> you shouldnt need netmasks. that's a proper C net
[07:16:10] <Tpenta> that wont be your problem tho
[07:16:25] <sol1> yesterday  i was told to do that by elektronkind
[07:16:39] <Tpenta> auralis, he can't ping the ip address of the router
[07:16:48] <Tpenta> what error does it get? "no answer" ?
[07:16:51] <boyd> Is dnet0 up?
[07:16:54] <sol1> Tpenta: so i remove the /etc/netmasks ?
[07:16:55] <Auralis> oh, hrm
[07:17:10] <Tpenta> it's just something yu didnt need to add, leave it it should nto hurt
[07:17:29] <sol1> boyd: yes when  i do ifconfig dnet0 up i don`t get any errors
[07:17:38] <sol1> Tpenta: ok..
[07:17:39] *** saturnine has quit IRC
[07:18:11] *** meow_ is now known as saturnine
[07:18:22] <boyd> sol1: I'd suggest you paste the output of ifconfig -a and netstat -rn if you can
[07:18:32] <sol1> ok..
[07:18:47] *** dj2 has quit IRC
[07:18:50] <sol1> i already have written it down in a paper
[07:18:57] <sol1> ifconfig -a
[07:19:12] <sol1> dnet0
[07:19:35] <sol1> inet 192.168.2 netmask fffffff0
[07:19:55] <sol1> ether 0:40:f6:2c:26:e1
[07:19:58] <Auralis> if that ip is correct written then it is wrong
[07:20:05] <boyd> There's a line of flags missing there, and the ip is not right
[07:20:07] <Tpenta> ummm should not the netmask be ffffff00 ?
[07:20:14] <Tpenta> pr are you subnetting?
[07:20:21] <boyd> good spot, Tpenta
[07:20:23] <richlowe> Tpenta: with an address like that, who knows?
[07:20:23] <sol1> broadcast 192.168.1.255
[07:20:27] <sol1> Tpenta: no
[07:20:38] <sol1> sorry it is ffffff00
[07:20:46] <Tpenta> ok
[07:20:52] <boyd> Is the IP address really as you typed above?
[07:21:06] <sol1> 192.168.1.2 sorry another typo
[07:21:18] <sol1> i am a bit tensed
[07:21:23] <boyd> :)
[07:21:46] *** calvin_ has quit IRC
[07:21:49] <sol1> i just installed Solaris yesterday..
[07:21:49] <boyd> and the flags= part includes "UP" ?
[07:22:04] <sol1> i didn`t note down that part
[07:22:11] <sol1> but i think it does
[07:22:22] <boyd> it should after ifconfig dnet0 up anyway
[07:22:30] <boyd> Is there anything else on the net?
[07:22:35] <boyd> (apart from the router?
[07:23:03] <sol1> boyd: i don`t understand the question..
[07:23:19] <boyd> Is anything else apart from the router plugged into the network?
[07:23:48] <sol1> i have a lan card and that is connected to an adsl modem thats is provided by the isp
[07:24:01] <boyd> So no other machines?
[07:24:03] <sol1> netstat -r
[07:24:05] <sol1> no
[07:24:08] <boyd> Ok
[07:24:28] <Auralis> ah! does the adsl modem delivers normal ethernet or do you need pppoe?
[07:24:44] <sol1> 192.168.1.0        192.168.1.2 dnet0
[07:24:45] <boyd> And you're sure that the router is configured for 192.168.1.1?
[07:25:05] <sol1> 244.0.0.0 localhost lo0
[07:25:14] <sol1> default 192.168.1.1
[07:25:24] <sol1> localhost localhost lo0
[07:25:36] <sol1> Auralis: normal i guess
[07:26:02] <Auralis> sol1: check, i think that is your problem, you are not running pppoe on the sun box
[07:26:17] <sol1> Auralis: i don`t know what that is ..
[07:26:26] <sol1> pppoe ?
[07:26:35] <Auralis> you said you are on a live cd?
[07:26:42] <sol1> how do i do that ? commands please
[07:26:45] <boyd> Auralis: I don't have anything handy with that... would if show on snoop output?
[07:26:53] <sol1> Auralis: yes linux live cd
[07:27:07] <Auralis> boyd: should
[07:27:23] <Auralis> sol1: ok, how did you configured the network for the session?
[07:27:44] <sol1> I enter the ip address for the device eth0
[07:27:49] <sol1> and press activate
[07:27:51] *** vmhobbes has quit IRC
[07:27:52] <sol1> thats it
[07:28:21] <Auralis> hrm ok, sounds not like pppoe
[07:28:27] <sol1> sometime it gets autoconfigured via dhcp though
[07:28:38] <Auralis> hoes the ifconfig looking on the linux?
[07:28:53] <sol1> wait...
[07:29:15] *** sol1 has quit IRC
[07:29:17] <dwc-> why don't you just dhcp it under solaris
[07:31:51] *** jamesd_ has quit IRC
[07:32:36] *** sol1 has joined #opensolaris
[07:32:50] <sol1> this is the ifconfig in linux
[07:32:59] <sol1> eth0      Link encap:Ethernet  HWaddr 00:40:F6:2C:26:E1
[07:33:00] <sol1>           inet addr:192.168.1.2  Bcast:192.168.1.255  Mask:255.255.255.0
[07:33:01] <sol1>           inet6 addr: fe80::240:f6ff:fe2c:26e1/64 Scope:Link
[07:33:01] <sol1>           UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
[07:33:01] <sol1>           RX packets:283 errors:0 dropped:0 overruns:0 frame:0
[07:33:02] <sol1>           TX packets:291 errors:2 dropped:0 overruns:0 carrier:2
[07:33:04] <sol1>           collisions:0 txqueuelen:1000
[07:33:07] <sol1>           RX bytes:55282 (53.9 KiB)  TX bytes:23435 (22.8 KiB)
[07:33:11] <sol1>           Interrupt:18 Base address:0xb800
[07:33:12] <sol1> lo        Link encap:Local Loopback
[07:33:15] <sol1>           inet addr:127.0.0.1  Mask:255.255.255.255
[07:33:17] <sol1>           inet6 addr: ::1/128 Scope:Host
[07:33:20] <sol1>           UP LOOPBACK RUNNING  MTU:16436  Metric:1
[07:33:23] <sol1>           RX packets:0 errors:0 dropped:0 overruns:0 frame:0
[07:33:24] <sol1>           TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
[07:33:28] <sol1>           collisions:0 txqueuelen:0
[07:33:29] <sol1>           RX bytes:0 (0.0 b)  TX bytes:0 (0.0 b)
[07:33:33] <sol1> sit0      Link encap:IPv6-in-IPv4
[07:33:35] <sol1>           NOARP  MTU:1480  Metric:1
[07:33:38] <sol1>           RX packets:0 errors:0 dropped:0 overruns:0 frame:0
[07:33:39] <dwc-> why don't you just dhcp it under solaris
[07:33:40] <sol1>           TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
[07:33:43] <sol1>           collisions:0 txqueuelen:0
[07:33:45] <sol1>           RX bytes:0 (0.0 b)  TX bytes:0 (0.0 b)
[07:33:48] <sol1> oops i should have used a pastebin osrry
[07:33:51] <dwc-> and you really don't need to paste all of that ...
[07:33:54] <sol1> how doi da that ?
[07:34:21] <dwc-> http://www.google.com/search?q=solaris+dhcp
[07:34:24] <sol1> ifconfig autodhcp ?
[07:34:27] <dwc-> first link
[07:34:39] *** dunc has joined #opensolaris
[07:34:51] <dwc-> or there's probably some from sun.com too
[07:35:12] <sol1> Auralis: hers is the outpuy
[07:35:15] <sol1> Auralis: hers is the output
[07:35:18] *** vmhobbes has joined #OpenSolaris
[07:35:51] <whaq> elektronkind, u around?
[07:37:11] <sol1> dwc-: it saus i have to create an le0
[07:37:22] <sol1> in my case is it dnet0 ?
[07:38:25] <sol1> hey ?
[07:38:26] <dwc-> yes
[07:38:47] <sol1> ok..do i have to remove the other files ?
[07:38:55] <dwc-> what other files?
[07:39:02] <sol1> /etc/netmasks , /etc/defaultrouter ?
[07:39:10] <dwc-> also, if it's still not working, try traceroute -n 4.2.2.1
[07:39:15] <dwc-> and save the output
[07:39:26] <sol1> bok..
[07:39:28] <sol1> ok
[07:39:39] <dwc-> yea, you should be able to remove those
[07:39:40] <sol1> i will try and get back. pls be here itself
[07:39:57] <sol1> should i remove them ?
[07:40:07] <sol1> ok.. i will try and get back
[07:40:08] <dwc-> remove whatever you added
[07:40:09] *** sol1 has quit IRC
[07:40:49] <dwc-> why does every irc network think of themself as an airline ...
[07:41:39] <LeftWing> Err.. what?
[07:43:04] *** bobbyz has quit IRC
[07:43:35] <boyd> Yeah, what? You mean they serve bad food and run late without apology?
[07:44:01] * jmcp sods off
[07:44:46] *** DrAk0 has quit IRC
[07:44:55] <LeftWing> boyd: I suspect that's not quite where he's going. ;P
[07:45:13] *** jmcp has quit IRC
[07:45:23] <galt> dwc-, would you prefer tehy thought of themselves as a brothel?
[07:45:25] *** darkcmd has quit IRC
[07:48:01] <boyd> http://www.flickr.com/photos/megganallison/311383545
[07:51:31] <dlg> WHY DOESNT THE U20 HAVE A SERIAL PORT?
[07:52:38] *** yongsun has joined #opensolaris
[07:53:15] <dwc-> brothel would be a better metaphor
[07:53:22] <boyd> Part of the general no-legacy philosophy I guess... but it sucks
[07:53:23] <dwc-> lots of people coming through
[07:53:27] <dwc-> no respect for anyone
[07:53:41] *** slowhog has left #opensolaris
[07:53:42] <dwc-> random zealots attacking it
[07:54:04] <dwc-> dlg: because they can stick a usb port in with the same board real estate
[07:54:16] <dwc-> and more people use usb these days than serial
[07:54:21] <razrX> morning all
[07:54:42] <dwc-> oh... and people are always getting fucked
[07:54:53] <dwc-> sounds like irc to me!
[07:57:25] *** vmhobbes has quit IRC
[07:58:19] <LeftWing> dwc-: So many communities and work/social centres are like that, I hardly think the human condition is limited to IRC.
[08:00:58] <dlg> dwc-: real estate on what?
[08:01:02] <dlg> the mobo? or on the backplace?
[08:01:05] <dlg> backplate?
[08:01:17] <dlg> cos the mobo always has tracks for teh serial port
[08:01:22] <dlg> and the backplate has space
[08:01:30] <dlg> but they skip on a part
[08:01:39] <dwc-> both on the mobo and on the case
[08:02:39] *** sol1 has joined #opensolaris
[08:02:45] <sol1> dwc-: hey
[08:02:50] <sol1> dhcp failed
[08:02:55] <dlg> the mobo has a header for it
[08:03:29] <sol1> then i again created those files for a static ip and then did traceroute -n 4.2.2.1
[08:03:38] <sol1> dwc-: are u ther ?
[08:03:44] <dwc-> yea
[08:03:53] <sol1> ok.thanks god.. here is the output
[08:04:00] <sol1> 1.* * *
[08:04:03] <sol1> 2.* * *
[08:04:08] <sol1> 3.* * *
[08:04:18] <sol1> 4.* 202.56.193.25
[08:04:19] <dwc-> okay, anything other than * * * ?
[08:04:32] <sol1> 5.59.145.1.73
[08:04:34] <dwc-> interesting
[08:04:37] <dwc-> okay, you can stop
[08:04:52] <sol1> oh.. i have it till 10
[08:04:53] <dwc-> that means your network is up
[08:04:59] <sol1> the i tried this..
[08:05:05] <dwc-> and you can get packets outside of your router
[08:05:11] <dwc-> and onto the internet
[08:05:13] <sol1> ping 192.168.1.1
[08:05:24] <dwc-> it's odd that nothing answered for the first 3 hops
[08:05:24] <sol1> 192.198.1.1 is alive
[08:05:32] <sol1> ping 192.168.1.2
[08:05:39] <sol1> 192.168.1.2 is alive
[08:05:46] <sol1> ping 192.168.1.255
[08:05:51] <sol1> 192.168.1.255 is alive
[08:05:54] <sol1> host yahoo.com
[08:06:10] <sol1> not found / not reachable
[08:06:19] <Tpenta> !seen stevel
[08:06:20] <Drone> stevel (stevel!n=stevel at 192 dot 18.43.225) was last seen in #opensolaris on Fri 08 Dec 2006 04:13 GMT, saying 'heh'.
[08:06:30] <dwc-> so it sounds like just your dns is broken then
[08:07:01] <sol1> oh... this is the contents of /etc/resolv.conf
[08:07:08] <sol1> nameserver 192.168.1.1
[08:07:11] <sol1> is that right ?
[08:07:16] <dwc-> most likely, yes
[08:07:27] <sol1> Then how is it broken ?
[08:07:36] <dwc-> I can't say for certain that your dsl router runs a dns server
[08:07:45] <Auralis> cp /etc/nsswitch.dns /etc/nsswitch.conf
[08:07:56] <sol1> its working now.. on the livecd
[08:08:06] <sol1> Auralis: do i have to do that ?
[08:08:26] <dwc-> then it should work on solaris as well.
[08:08:28] <Auralis> you need to tell the system that you want to use dns for hostnames
[08:08:45] <sol1> Auralis: tell me all my options please.. its really painfull i have been rebooting to live cd and solaris from 10pm last night
[08:08:49] <Auralis> soalris can use different name resoultion methods simulatanously
[08:09:01] <sol1> its now 12:30 pm the next day !
[08:09:10] <Auralis> just make sure /etc/nsswitc,conf holdes the line
[08:09:13] <Auralis> hosts: files dns
[08:09:16] <sol1> Still no internet
[08:09:21] *** vmhobbes has joined #OpenSolaris
[08:09:24] <sol1> ok..
[08:10:02] *** bobbyz has joined #opensolaris
[08:10:23] <sol1> Auralis: shall i try that ?
[08:10:27] <ShadowHntr> sol1: check to see if your /etc/defaultrouter is set
[08:10:32] <sol1> Auralis: anything else ?
[08:10:35] <sol1> yes it is set
[08:10:40] <sol1> 192.168.1.1
[08:10:59] <Auralis> sol1: since you have a resolv.conf that should be all to activate dns resolution
[08:11:15] <sol1> Auralis: ok..
[08:11:30] <sol1> then why is it not working ?
[08:11:41] <dwc-> you do "have internet"
[08:11:49] <dwc-> otherwise the traceroute wouldn't have made it out of your ISP
[08:11:58] *** gustav3d has quit IRC
[08:12:08] <sol1> dwc-: but i am not able to surf and even host yahoo.com fails
[08:12:15] <Auralis> err, you still need the nsswitch.conf corect s i said
[08:12:34] <sol1> Auralis: ok.. i will check that out and come back..
[08:12:43] <sol1> hope it works this time atleast
[08:12:58] <Auralis> host fails because it uses the system configured name resolution, and with nsswitch.conf only pointing to files if fails because dns is not used
[08:13:06] <sol1> how long will  u guys be here ?
[08:13:17] <sol1> Auralis: ok.. will try that
[08:13:20] <sol1> Thanks
[08:13:40] *** sol1 has left #opensolaris
[08:16:10] *** triplah has joined #opensolaris
[08:28:37] *** Mazon is now known as mazon
[08:29:50] *** sol1 has joined #opensolaris
[08:30:13] <sol1> hey Auralis IT worked !
[08:30:24] <Auralis> :)
[08:30:30] <sol1> i chatting from solaris 10 !!!!
[08:30:39] <sol1> Thanks a lot guys
[08:30:47] <sol1> Ok.. next problem sound card
[08:30:54] <sol1> any idea on that i
[08:31:58] <Auralis> this is x86 hardware voodoo, i have no clue about that
[08:32:03] <sol1> i do this  /usr/X11/bin/scanpci
[08:32:22] <sol1> pci bus 0x0000 cardnum 0x06 function 0x00: vendor 0x10de device 0x01b1
[08:32:23] <sol1>  nVidia Corporation nForce Audio
[08:32:25] <sol1> i get this
[08:32:52] <dlg> i think you should be more interested in prtconf -D output
[08:33:07] <Triskelios> install the oss drivers from opensound
[08:33:10] <sol1> Auralis: even for the card elektronkind asked me to paste this info and using the numbes he enables my lan card
[08:33:30] <sol1> enabled
[08:33:39] <dlg> nv audio stuff should be supported by a generic ac97 driver
[08:34:02] *** mazon is now known as Mazon
[08:34:03] <sol1> dlg: ok how do i do it ?
[08:34:39] <dlg> if you can match the prtconf -D output to the output you just pasted from scanpci, you should be able to tell if there is already a driver hooked up to it
[08:34:58] <dlg> since the u20 uses nvidia nforce audio id be surprised if it isnt already working
[08:35:49] <sol1> dlg: which is the audio in prtconf -D
[08:36:02] <sol1> prtconf -D
[08:36:03] <sol1> System Configuration:  Sun Microsystems  i86pc
[08:36:05] <sol1> Memory size: 224 Megabytes
[08:36:06] <sol1> System Peripherals (Software Nodes):
[08:36:08] <sol1> i86pc (driver name: rootnex)
[08:36:09] <sol1>     scsi_vhci, instance #0 (driver name: scsi_vhci)
[08:36:11] <sol1>     +boot
[08:36:12] <sol1>         memory
[08:36:14] <sol1>     aliases
[08:36:15] <sol1>     chosen
[08:36:17] <sol1>     i86pc-memory
[08:36:18] <sol1>     i86pc-mmu
[08:36:20] <sol1>     openprom
[08:36:21] <sol1>     options, instance #0 (driver name: options)
[08:36:23] <sol1>     packages
[08:36:24] <sol1>     delayed-writes
[08:36:26] <Triskelios> sol1: argh, don't flood!
[08:36:27] *** jmcp has joined #opensolaris
[08:36:27] <sol1>     itu-props
[08:36:28] <sol1>     isa, instance #0 (driver name: isa)
[08:36:30] <sol1>         motherboard
[08:36:33] <sol1>         INT0800
[08:36:34] <sol1>         lp (driver name: ecpp)
[08:36:35] <sol1>         joyst
[08:36:37] <sol1>         i8042, instance #0 (driver name: i8042)
[08:36:39] <sol1>             keyboard, instance #0 (driver name: kb8042)
[08:36:41] <sol1>         asy, instance #0 (driver name: asy)
[08:36:44] <sol1>         asy, instance #1 (driver name: asy)
[08:36:45] <sol1>     pci, instance #0 (driver name: pci)
[08:36:47] <sol1>         pci10de,1a4
[08:36:49] <sol1>         pci10de,c11
[08:36:51] <sol1>         pci10de,c11
[08:36:52] <triplah> someone kick him
[08:36:53] <sol1>         pci10de,c11
[08:36:55] <sol1>         pci10de,c11
[08:36:57] <sol1>         pci10de,c11
[08:36:59] <sol1>         pci10de,c11, instance #0 (driver name: ohci)
[08:37:02] <sol1>             mouse, instance #1 (driver name: hid)
[08:37:03] <sol1>         pci10de,c11, instance #1 (driver name: ohci)
[08:37:05] <sol1>         pci1043,c11
[08:37:07] <sol1>         pci1043,8384
[08:37:09] <sol1>         pci10de,1b8, instance #0 (driver name: pci_pci)
[08:37:12] <sol1>             pci8e2e,5100, instance #0 (driver name: dnet)
[08:37:13] <sol1>         pci-ide, instance #0 (driver name: pci-ide)
[08:37:15] <Triskelios> jmcp: ping?
[08:37:16] <sol1>             ide, instance #0 (driver name: ata)
[08:37:17] <sol1>                 cmdk, instance #0 (driver name: cmdk)
[08:37:19] <sol1>             ide, instance #1 (driver name: ata)
[08:37:21] <sol1>                 sd, instance #0 (driver name: sd)
[08:37:23] <sol1>                 sd, instance #1 (driver name: sd)
[08:37:25] <sol1>                 st, instance #0 (driver name: st)
[08:37:27] <sol1>                 st, instance #1 (driver name: st)
[08:37:30] <sol1>         pci10de,1b7, instance #1 (driver name: pci_pci)
[08:37:30] * jengelh sends a fireball towards sol1
[08:37:31] <sol1>             display, instance #0 (driver name: vgatext)
[08:37:33] <sol1>     used-resources
[08:37:35] <sol1>     pseudo, instance #0 (driver name: pseudo)
[08:37:37] <sol1>     xsvc, instance #0 (driver name: xsvc)
[08:37:39] <sol1>     objmgr, instance #0 (driver name: objmgr)
[08:37:42] <sol1>     cpus
[08:37:43] <sol1>         cpu, instance #0
[08:37:45] <sol1> sorry
[08:37:47] <sol1> Triskelios: how do i install new software ?
[08:37:49] <sol1> triplah: sorry..
[08:37:52] * sol1 gets hurt
[08:38:23] <sol1> dlg: any idea ?
[08:38:34] <dlg> several
[08:38:41] <dlg> but im trying to get some work done
[08:38:46] <sol1> ok.. tell me
[08:38:55] <sol1> Oh...
[08:39:17] <jmcp> Triskelios: hi
[08:39:51] <sol1> Auralis: you have any idea ? whats the soundcard device file name ?
[08:40:03] <sol1> like lan card is dnet0
[08:40:26] <sol1> Auralis: for the lan card i did
[08:40:29] <Triskelios> jmcp: nevermind, thought you had ops
[08:40:37] <sol1> dnet0"pci1011,19"
[08:41:00] <Triskelios> sol1: that's not what you did
[08:41:43] <sol1> elektronkind: asked me do that .. only after that ifconfig -a showed dnet0 till then onl lo0 was shown
[08:41:45] <Triskelios> you added the PCI ID to the dnet driver
[08:41:58] <Triskelios> using update_drv or similar
[08:42:11] <sol1> thats the command i typed
[08:42:26] <Triskelios> sol1: that's not a command
[08:42:35] <sol1> oh..
[08:42:43] <jmcp> Triskelios: I wish ...
[08:42:51] <Triskelios> sol1: please be accurate
[08:42:59] <jmcp> sol1: and *precise*
[08:43:00] <sol1> ok leave that..
[08:43:19] <sol1> now my audi card has the foll numbers
[08:44:01] <sol1> bus 0x0000 cardnum 0x06 function is 0x00 vendor 0x10de device 0x01b1
[08:44:15] <sol1> so what device would it be ?
[08:45:03] <andersmo> 0x10de - I think that's nvidia.
[08:45:07] <sol1> yes
[08:45:08] <jmcp> yup
[08:45:27] <sol1> t0x01b1 whats that ?
[08:45:40] <sol1> 0x01b1 sorry
[08:45:41] <Triskelios> sol1: the device ID...
[08:45:55] <jmcp> sol1: pci10de,1b1 is the "nForce Audio" interface
[08:46:04] <sol1> yes
[08:46:13] <sol1> how do get it working
[08:46:22] <sol1> so the card is supported by solaris right ?
[08:46:25] <andersmo> sol1: what jmcp said. http://pci-ids.ucw.cz/iii/ is useful. =)
[08:46:26] <Triskelios> that nforce chip is intel ich audio compatible
[08:46:28] <jmcp> sol1: your best bet for support of that chip - if the audio810 driver doesn't attach to it - is to grab Juergen Keil's Solaris audio driver
[08:46:50] <jmcp> sol1: http://www.tools.de/solaris/audio
[08:47:18] <jmcp> sol1: are you running s10 or nevada/sx ?
[08:47:23] <sol1> Triskelios: yes Linux detects it as intel only
[08:47:24] <jmcp> if you are, grab the v1.9 beta.
[08:47:26] <sol1> Solaris 10
[08:47:43] <sol1> jmcp: i am at the page what do i download ?
[08:47:56] <sol1> x86 package right ?
[08:48:13] <jmcp> sol1: see that bit where it say        Also available: Beta Version 1.9 of the audio drivers.
[08:48:15] <jmcp> grab that version
[08:48:26] <dlg> do x86s have mmus?
[08:48:34] <dlg> wait
[08:48:37] <Triskelios> dlg: lol!
[08:48:38] <dlg> thats a dumb question
[08:48:39] <dlg> iommus
[08:48:41] <sol1> i downloaded the second one..
[08:48:46] <jmcp> dlg: nope
[08:49:10] * dlg nod
[08:49:30] *** simford has joined #opensolaris
[08:49:40] <sol1> it extracted a .pkg file.. how do i install that ?
[08:49:57] <jmcp> ffs
[08:50:02] <jmcp> sol1: "man pkgadd"
[08:51:20] <sol1> so i do pkgadd -d /audio-1.8-i86pc.pkg ?
[08:51:31] <sol1> i extracted the file in /
[08:52:06] * jmcp shudders
[08:52:15] <jmcp> sol1: just how long have you been using a unix variant for?
[08:52:35] <dlg> jmcp: there is a danger he uses the one with the crappy doco, so he doesnt try manpages
[08:52:44] <sol1> 1 year and more
[08:52:45] <jmcp> dlg: that's what I'm thinking
[08:52:56] <jmcp> sol1: great. So you're not doing everything as root, are you
[08:53:02] <sol1> yes
[08:53:03] <dlg> i know i forget about them when i go back to a sane unix
[08:53:10] <jmcp> sol1: oh, good
[08:53:12] <Triskelios> maybe he should read a user guide
[08:53:19] <sol1> i know root is bad..
[08:53:32] <sol1> i have created another user..
[08:53:36] <sol1> already
[08:53:38] <jmcp> sol1: now for your next challenge, make sure the perms on / are such that you as an ordinary user cannot dump stuff there
[08:53:51] <sol1> ok..
[08:54:02] <sol1> so first let me install the driver
[08:54:02] <jmcp> sol1: so where should you have saved the pkg file?
[08:54:15] <sol1>  /tmp
[08:54:17] <jmcp> yes, your cmd line looks fine
[08:54:24] <jmcp> good
[08:54:44] <sol1> ok..so shall i  install it
[08:54:53] <jmcp> ... yes .....
[08:55:09] <sol1> pkgadd -d /tmp ?
[08:55:33] <jmcp> you had it correct the first time
[08:55:49] <sol1> ok..
[08:55:50] *** bobbyz has quit IRC
[08:57:03] <sol1> so i reboot noe ?
[08:57:07] <sol1> now
[08:57:22] <jmcp> what messages did you see?
[08:57:33] <sol1> that i need to reboot
[08:57:48] <jmcp> what other messages did you see?
[08:57:57] <sol1> many
[08:58:07] <sol1> I pressed y for all the packages
[08:58:38] <sol1> ok.. i will reboot now..
[08:58:44] <jmcp> sol1: here's a hint. Use the mouse to select all the text that resulted from your pkgadd command, and then use http://rafb.net/paste to provide a copy of it without flooding this channel
[08:58:57] <sol1> i closed the terminal
[08:59:22] <sol1> is there a log that i can copy from and paste ?
[08:59:30] <jmcp> probably not
[08:59:35] <jmcp> are you always this hasty?
[08:59:42] * jmcp feels like an Ent some times
[08:59:45] <sol1>  i am a bit tense now though
[08:59:52] <sol1> ok..
[08:59:57] <sol1> i will reboot
[09:00:02] *** sol1 has left #opensolaris
[09:00:09] * jmcp groans
[09:03:21] *** sol1 has joined #opensolaris
[09:03:57] <sol1> jmcp: i still see the mixer disabled
[09:04:12] <sol1> now i am as normal user not root
[09:05:04] <sol1> jmcp: are u there ?
[09:06:23] <sol1> hello !
[09:07:17] <sol1> how do i enable the driver ?
[09:07:52] <Triskelios> sol1: prtconf -D and check if you have the driver attached (should be audio810 or audioi810)
[09:08:01] <sol1> ok
[09:08:59] <sol1> no..
[09:09:17] <sol1> no audioi810
[09:10:58] <sol1> Triskelios: shall i install the driver again
[09:11:24] <sol1> which is the pastebin ?
[09:11:28] <Triskelios> sol1: no, assume it doesn't have your ID, and add it with update_drv
[09:11:37] <Triskelios> rafb.net/paste is a pastebin
[09:12:06] <sol1> Triskelios: how do add it using updata_drv ?
[09:12:27] <sol1> command please
[09:13:00] <sol1> update_drv audioi810
[09:13:01] <sol1> devfsadm: driver failed to attach: audioi810
[09:13:03] <sol1> Warning: Driver (audioi810) successfully added to system but failed to attach
[09:13:34] <Triskelios> update_drv -i '"pci10de,1b1"' audioi810
[09:14:14] <sol1> nope
[09:14:21] <sol1> NOTE: at least one of m/i/P/p must be specified with -a and -d.
[09:14:30] <Triskelios> oops, add -a in front
[09:14:52] <sol1>  update_drv -ai "pci10de,1b1" audioi810 Like this ?
[09:15:03] <Triskelios> no
[09:15:21] <Triskelios> type exactly what I said
[09:15:36] <Triskelios> except with -a before -i
[09:15:48] <sol1> done..
[09:15:51] <sol1> now what ?
[09:16:02] <Triskelios> you used single quotes right?
[09:16:03] <dlg> prtconf -D might show the driver attached somewhere
[09:16:10] <sol1> double quotes
[09:16:25] <Triskelios> you need both, duh
[09:16:39] <sol1>  pci1043,8384, instance #0 (driver name: audioi810)
[09:16:47] <sol1> anyway .. i think it worked
[09:17:02] <sol1> shall do again with both the single and double quotes ?
[09:17:36] <sol1> ok. samething
[09:17:49] <sol1> now what next ?
[09:18:17] <Triskelios> just test with something that plays sound
[09:18:40] <sol1> do i have to restart ?
[09:18:53] <sol1> mixer is still disabled
[09:19:40] <Triskelios> the gnome mixer in jds?
[09:19:44] <Triskelios> did you log back in?
[09:19:51] <sol1> no wait
[09:20:08] *** sol1 has quit IRC
[09:21:41] *** sol1 has joined #opensolaris
[09:22:11] <sol1>  i have only mp3`s
[09:22:26] <sol1> the java player doesn`t play that i guess
[09:22:39] <Triskelios> use rhythmbox or totem
[09:22:44] <sol1> any nice audio player small download as well
[09:22:55] <sol1> i have to download as .pkg ?
[09:23:17] <Triskelios> should be preinstalled... maybe only in nevada though
[09:23:29] <sol1> will beep media player work ?
[09:23:32] <Triskelios> you don't have any other players?
[09:23:33] <sol1> I use Solaris 10
[09:23:34] <Triskelios> yeah
[09:23:37] <sol1> what is neveda ?
[09:23:42] <Triskelios> solaris 11
[09:23:48] <sol1> no other players
[09:23:57] <sol1> i can`t download solaris 11 ?
[09:24:08] <sol1> its also free right ?
[09:24:25] <sol1> hey.. totem is there
[09:24:45] <sol1> There is no element present to handle the stream's mime type audio/mpeg.
[09:24:53] <sol1> no mp3 plugin huh ?
[09:25:04] <Triskelios> yup
[09:25:10] <sol1> hmmm..
[09:25:10] <Triskelios> don't know why it's missing
[09:25:26] <Triskelios> guess they added that later, too
[09:25:54] <Triskelios> if you have beep-media-player installed that should work
[09:25:57] <sol1> oh..
[09:26:07] <sol1> wait i wil download beep
[09:26:14] <sol1> .pkg right ?
[09:26:25] <Triskelios> yes
[09:26:54] <sol1> any site where i can find solaris aps ?
[09:27:14] <sol1> http://bmp.beep-media-player.org/index.php/Packages
[09:27:16] <Triskelios> uh... the companion CD site on sun.com and blastwave.org
[09:27:39] <sol1> ok..
[09:27:50] <sol1> wiill the archlinux .pkg work ?
[09:28:17] <Triskelios> no that's an unrelated format
[09:28:30] <Triskelios> and the binaries are for linux anyway
[09:28:43] <sol1> oh..ok
[09:29:27] <sol1> there is no blastwave.org
[09:29:35] <sol1> its there.. sorry
[09:30:23] <Triskelios> blastwave adds a pkg-get tool to download packages automatically, but it also uses a special prefix (so most things you download from there will add duplicate copies of software you already have installed)
[09:30:48] <sol1> http://www.blastwave.org/packages.php/xmms
[09:31:53] <sol1> will  realplayer work ?
[09:32:19] <Triskelios> yes realplayer will work
[09:33:50] <sol1> thanks a ton..
[09:33:52] <sol1> bye al
[09:34:29] *** alobbs has joined #opensolaris
[09:34:41] <sol1> Triskelios: any ebook that i can use to learn solaris 10 ?
[09:34:54] <andersmo> sol1: lots of them at docs.sun.com =)
[09:35:01] <sol1> ebooks ?
[09:35:10] <sol1> ok.. Thanks
[09:35:20] <andersmo> well, you can get many of the documentation "books" in PDF format at least.
[09:35:26] <andersmo> Though I prefer HTML myself.
[09:35:29] <sol1> thanks
[09:35:30] <sol1> heh
[09:37:59] *** |tsoome| has joined #opensolaris
[09:39:34] <sol1> http://forms.helixcommunity.org/helix/builds/?category=realplay-stable
[09:39:43] <sol1> does thiis one play mp3 `s ?
[09:40:43] <sol1> bye
[09:40:46] *** sol1 has left #opensolaris
[09:48:17] *** qdk has quit IRC
[09:48:46] *** cmihai has joined #OpenSolaris
[09:52:16] *** Burana has joined #opensolaris
[09:53:03] <trygvis> anyone have any experience with running oracle on a thumper?
[10:10:05] *** alanc_away has quit IRC
[10:10:30] *** alanc_away has joined #opensolaris
[10:12:08] <sickness> http://javashoplm.sun.com/ECom/docs/Welcome.jsp?StoreId=7&PartDetailId=Sol-Express_b52-DVD-x86-SP-G-B&TransactionId=try <- hey update this link to 53 :P
[10:12:31] *** qdk has joined #opensolaris
[10:12:38] *** dunc has quit IRC
[10:21:17] <onbot> commit by stevel:  Tagged the wrong rev as onnv_54
[10:21:18] <onbot> commit by zw161486:  6482744 i945 chipsets should have AGP support; 6482755 drm should support i945 chipsets
[10:22:55] <richlowe> ...well that makes the onnv-gate restrictions so much easier to understand, or something.
[10:23:54] <cmihai> Well, the new sdm is pretty sweet :)
[10:25:55] *** bobbyz has joined #opensolaris
[10:29:12] *** linma` has quit IRC
[10:38:34] *** Burana has quit IRC
[10:38:35] *** Burana has joined #opensolaris
[10:40:40] *** simford has quit IRC
[10:44:25] <raph_ael> hello
[10:55:14] *** Burana has quit IRC
[10:57:07] <whaq> hello
[10:57:30] <bobbyz> hello :)
[10:57:59] <whaq> Hey Bobbyz, how goes?
[10:58:10] *** bougie has quit IRC
[10:58:27] <bobbyz> oh just fine...bout to call it a night though...too much putsing around with afs for one day  :)
[11:00:18] *** bunker has joined #opensolaris
[11:00:28] <whaq> Cool, how many nodes do you run it on?
[11:02:35] <bobbyz> right now I only run it on 1, my mini test lab at home, but it has quickly become my favorite network fs
[11:03:29] <ShadowHntr> i wanna try zfs once i get a good sized disk array set up
[11:03:30] <ShadowHntr> :)
[11:03:34] <whaq> never used it. what's good about it?
[11:03:45] <bobbyz> support for it on the evil empire's os is what clenches my vote...otherwise it would be nfs4
[11:04:04] *** bougie has joined #opensolaris
[11:04:11] <ShadowHntr> from wikipedia
[11:04:12] <ShadowHntr> ZFS, is a free, open-source file system produced by Sun Microsystems for its Solaris Operating System. It is notable for its high capacity, integration of the concepts of filesystem and volume management, novel on-disk structure, lightweight filesystems, and easy storage pool management.
[11:04:22] <ShadowHntr> http://en.wikipedia.org/wiki/ZFS
[11:05:46] <whaq> i meant afs
[11:05:51] <whaq> :)
[11:05:52] <ShadowHntr> sorry
[11:05:53] <ShadowHntr> :)
[11:06:08] <ShadowHntr> i wanna get a netra disk array or a storedge multipack
[11:06:17] <bobbyz> whaq: afs is awesome!  It's got snapshot support, replication across multiple file servers (although at the moment, only one copy can read a read-write copy), excellent performance, kerberos 5 integration (can even get it to integrate into AD)
[11:06:43] <bobbyz> a single global namespace for files, excellence support on all unices and windows
[11:06:48] <whaq> nice!
[11:07:03] <whaq> what about OS support?
[11:07:04] *** vmhobbes has quit IRC
[11:07:26] <ShadowHntr> AFS ?
[11:07:29] <ShadowHntr> can i get a link somewhere?
[11:07:37] <bobbyz> ShadowHntr: www.openafs.org
[11:07:58] <ShadowHntr> danke
[11:07:59] *** bengtf2 has quit IRC
[11:08:12] <bobbyz> you can also run an nfs translation service and access it through nfs clients
[11:08:12] *** peteh has quit IRC
[11:08:32] *** peteh has joined #opensolaris
[11:09:22] <ShadowHntr> bobbyz: thanks mate
[11:10:16] <bobbyz> ShadowHntr: np.  It's not exactly the easiest thing to setup and deploy (especially if you wish to use kerberos 5 instead of the built-in authentication which is kerberos 4 based), but it's a terrific network FS
[11:10:32] <ShadowHntr> *nod*
[11:10:41] <bobbyz> </rant>
[11:10:48] <ShadowHntr> it's integratable with active directory (if i have to) ?
[11:11:11] <bobbyz> ShadowHntr: I'm currently using my win 2003 server to handle authentication for it
[11:11:18] <ShadowHntr> cool
[11:11:19] <ShadowHntr> :)
[11:11:36] <bobbyz> though it's easier to setup if you use mit kerberos instead
[11:12:16] *** fgd has quit IRC
[11:13:03] <bobbyz> and if you want to build a version on solaris that uses seam instead of mit kerberos, you need some headers and a tiny patch, but if you actually decide to use openafs and deploy it, youcould msg me and I could help with that
[11:13:24] <bobbyz> anyway, to bead with me...g'night
[11:13:34] <whaq> have you guys ever encountered this strange behaviour in solaris: disk access on affected controller is interrupted w/ intermittent freezes.
[11:13:56] <bobbyz> and good luck with your sii card, whaq  :)
[11:14:00] <whaq> I'm baffled because it happens on a Sil3132 controller (supposed to be supported) and the same system is fine w/ windows.
[11:14:13] <whaq> bobbyz, thanks mate :)
[11:14:22] <whaq> have a good nite
[11:14:36] * bobbyz bobbyzzz
[11:17:43] *** fgd has joined #opensolaris
[11:19:06] *** klocze1 has quit IRC
[11:20:07] *** kloczek has joined #opensolaris
[11:26:32] *** Snake007uk has joined #opensolaris
[11:27:03] <Snake007uk> morning all
[11:28:02] *** ShadowHntr has quit IRC
[11:41:05] <whaq> mooin
[11:44:51] *** yongsun has quit IRC
[11:49:41] *** klepplap has quit IRC
[11:50:11] *** klepplap has joined #opensolaris
[11:50:11] *** jengelh has quit IRC
[11:50:36] *** jengelh has joined #opensolaris
[11:58:44] <Griffous> what is the current best solution for flash under opensolaris?
[12:07:43] <timeless> last i checked solaris shipped w/ flash
[12:07:48] <timeless> why do you eed a solution?
[12:07:48] *** elopaa_ has joined #opensolaris
[12:07:54] <timeless> it's the *only* platform i have with working flash
[12:08:33] <timeless> oh goody, sxcr53 is up
[12:09:02] * timeless wonders if there's some magic way to ask a java instance to dump its threads
[12:09:07] <timeless> (stacks)
[12:11:15] *** bougie has quit IRC
[12:11:24] <andersmo> timeless: it is, at least in the recent jvms.
[12:11:27] *** jteo has joined #opensolaris
[12:11:57] <andersmo> timeless: does "jstack" do what you want?
[12:13:55] <timeless> um,
[12:14:00] <timeless> ControlPanel  java  java_vm  javaws  keytool  kinit  klist  ktab  orbd  pack200  policytool  rmid  rmiregistry  servertool  tnameserv  unpack200
[12:14:08] <timeless> which of those is jstack?
[12:14:22] <Griffous> timeless, it plays some flash, not all though :/
[12:15:11] <timeless> griffous: brandz lx centos?
[12:15:32] <Griffous> yeah, I had read about that option - I was hoping there might be an easier option
[12:15:42] <Griffous> I'm having a crack at compiling gnash atm
[12:15:43] <timeless> buy a mac? :)
[12:16:46] <timeless> oh wow, ctrl-\ worked (Eventually)
[12:18:31] <movement> timeless: pstack?
[12:19:19] <timeless> hrm, pstack is um, intesting
[12:20:03] <andersmo> timeless: jstack is a java 1.5 thing.
[12:20:35] <timeless> ok, so,ctrl-\ worked, is tehre a way to get a list of the lsat couple of exceptions, or a count of exceptions, or to start logging for the jvm or..
[12:21:19] <timeless> the jvm was told to use 3.5g and it isn't, and the app hasn't been doing much work for a while, so i'm trying to figure out if it got stuck and how
[12:22:37] <timeless> hrm, it's jvm 1.5.0_07-b03 (running on centos [lx brandz])
[12:25:56] *** calumb has joined #opensolaris
[12:28:51] *** bougie has joined #opensolaris
[12:38:29] *** jteo has quit IRC
[12:45:04] <richlowe> timeless: hey, you deal with bugzilla, right?
[12:50:11] <bougie> hello :)
[12:52:26] <bunker> hello, i'm configuring ipf, but it's always in "maintenance" mode, how can i resolve? I've tried with svcadm clear ipfilter & co but it remains always in maintenance mode
[12:52:43] <bunker> i've also restarted the server
[12:53:30] <bunker> i'm relative new to opensolaris, so if it's a stupid question, sorry :D
[12:56:54] <bunker> mhm, i'm very stupid... i forget to enable the interface in pfil.ap
[12:58:53] <bunker> right :)
[12:59:27] *** deedaw has joined #opensolaris
[13:02:25] *** Tpenta has quit IRC
[13:07:24] *** Burana has joined #opensolaris
[13:15:58] *** Xh4 has joined #opensolaris
[13:18:37] *** vmhobbes has joined #OpenSolaris
[13:18:58] *** vandenoever has joined #opensolaris
[13:19:10] <vandenoever> hello, are there vmware player images of opensolaris?
[13:19:28] <Xh4> :O
[13:19:31] <Xh4> You said the V word.
[13:19:44] *** calumb is now known as calLNCH
[13:21:18] *** uebayasi has joined #opensolaris
[13:21:51] <bunker> lol
[13:22:12] <vandenoever> eh, is this a faq?
[13:24:56] <lasseoe> no an irc channel
[13:27:29] <movement> vandenoever: no. there is a vmware image of nexenta, and one of Solaris 10
[13:27:54] <vandenoever> movement: does one of those have the forte compiler?
[13:28:05] <movement> the solaris 10 one does
[13:28:15] <movement> http://javashoplm.sun.com/ECom/docs/Welcome.jsp?StoreId=8&PartDetailId=SES-VM-1.0-G-F&TransactionId=try
[13:28:16] <vandenoever> movement: do you have a link to it?
[13:28:21] <vandenoever> :-) thanks
[13:28:21] <movement> it's an older version of solaris 10 alas
[13:30:06] <vandenoever> movement: it's for amd64 and i've to register...
[13:30:24] <movement> vandenoever: it works on 32-bit too, and yes you do
[13:30:37] <vandenoever> ok, thanks
[13:30:54] <dlg> half the fun of an os is figuring out how to install it
[13:31:20] <movement> it should work on 32-bit rather, not positive about that wrt vmware
[13:32:15] *** vmhobbes has quit IRC
[13:42:19] <richlowe> movement: I'd swear you're just out to make life difficult (tools-discuss) :)
[13:44:04] *** yoss has joined #opensolaris
[13:44:24] <yoss> hi all
[13:44:31] <Berny> just out of interest: how does live-upgrade handle zones? are they updated upon next reboot?
[13:45:02] <richlowe> LU handles zones in snv_53 and above.
[13:45:23] <Berny> not in plain sol10?
[13:46:17] <yoss> how can i change the defaultrouter within a zone without change the defaultrouter of the global zone?
[13:47:37] *** vmhobbes has joined #OpenSolaris
[13:47:47] *** elopaa_ has left #opensolaris
[13:48:12] *** Xh4 has quit IRC
[13:48:27] <Berny> vi /etc/defaultrouter (just my naive idea)
[13:49:09] <yoss> benny: doesn't work
[13:49:57] <yoss> he seems to ignore th file...
[13:50:07] *** gm152 has joined #opensolaris
[13:51:06] *** nexrafa has quit IRC
[13:52:26] *** polk__ has quit IRC
[13:56:16] <Berny> so to make sure i cannot use LU to update from sol10 03/05 to s10u3 (when it becomes available) if i have non-global zones?
[13:57:00] <richlowe> I had thought zone upgrade/live upgrade was going into u3
[13:57:05] *** polk__ has joined #opensolaris
[13:57:08] <richlowe> but obviously I'm not at all sure of that.
[13:57:17] <Berny> who is to ask? ;-)
[13:58:42] <richlowe> Hm, all Dave Miner's blog entry says on the subject is "A future Solaris 10 update release"
[13:59:45] <Berny> bugger
[14:00:02] <Berny> so i would have to run a "normal" update on the box with zones?
[14:00:05] *** calLNCH has quit IRC
[14:02:09] *** Burana has quit IRC
[14:02:12] *** Burana has joined #opensolaris
[14:04:49] <movement> richlowe: it's all so easy when it's not you doing the work! ;)
[14:05:32] <richlowe> movement: I sent you mail off-list regarding it all.
[14:05:35] <richlowe> movement: it's a laugh riot.
[14:06:14] * richlowe is still trying to sanitize some of the more... interesting choices of wording out of other notes on the subject, so they can go to tools-discuss.
[14:08:20] *** Somethingelse has joined #opensolaris
[14:09:52] *** movement has quit IRC
[14:14:17] *** yoss has quit IRC
[14:14:52] *** vmhobbes has quit IRC
[14:18:31] *** vmhobbes has joined #OpenSolaris
[14:22:50] *** Andrew___ has quit IRC
[14:23:26] *** Andrew___ has joined #opensolaris
[14:24:34] *** movement has joined #opensolaris
[14:26:31] *** darrenm has joined #opensolaris
[14:43:01] *** Burana has quit IRC
[14:46:18] *** bozi has joined #opensolaris
[14:47:31] *** laca has joined #opensolaris
[14:48:24] <bozi> how to i pass more than 1 -B parameter to the kernel? with 1 "-B" or with multiple "-B"?
[14:49:53] <movement> multiple I believe
[14:51:17] <bozi> *trying*
[14:51:42] *** lloy0076 has joined #opensolaris
[14:52:22] <lloy0076> Well, I now have nVidia drivers working nicely :)
[14:57:06] *** lloy0076 has left #opensolaris
[14:57:29] *** triplah has quit IRC
[15:01:37] *** nexrafa has joined #opensolaris
[15:03:36] <bozi> damn is doesnt work
[15:04:22] <bozi> neither 1 nor multiple "-B"
[15:04:53] *** gustav3d has joined #opensolaris
[15:05:53] <bozi> get anyone running the belenix live cd in qemu?
[15:08:20] <bozi> scanning local part./cdrom
[15:08:27] <bozi> pagefault
[15:08:35] <bozi> damn
[15:20:39] *** laca is now known as lacaAFK
[15:28:54] *** vandenoever has left #opensolaris
[15:43:07] *** dj2 has joined #opensolaris
[15:44:49] *** bozi has left #opensolaris
[15:46:18] *** mwaterl has joined #opensolaris
[15:47:53] *** mwaterl has quit IRC
[15:53:38] *** darrenm has quit IRC
[16:00:57] *** Alic1 has joined #opensolaris
[16:00:59] *** qdk has quit IRC
[16:01:17] *** Edward has joined #opensolaris
[16:01:36] <Edward> Hi
[16:02:28] *** Edward has quit IRC
[16:03:02] *** Edward has joined #opensolaris
[16:03:14] <Edward> Anyone around?
[16:04:17] <movement> nope
[16:04:20] *** Edward has quit IRC
[16:04:20] <Berny> no
[16:04:58] *** Edwar1 has joined #opensolaris
[16:04:59] <Berny> bugger
[16:05:25] <Edwar1> Hi well I got a question
[16:06:11] <Edwar1> Solaris build 52 does not seem to start if there is no network connection, how can I make it time out?
[16:07:14] *** Mazon is now known as mazon
[16:08:14] *** Alic1 has quit IRC
[16:09:09] *** Alic1 has joined #opensolaris
[16:12:50] *** Edwar1 has left #opensolaris
[16:13:47] *** mjh has joined #opensolaris
[16:14:09] *** TBCOOL has quit IRC
[16:16:12] *** edwardocallaghan has joined #opensolaris
[16:16:36] *** qdk has joined #opensolaris
[16:17:04] *** lacaAFK is now known as laca
[16:19:20] <elektronkind> hrm. s.o.o seems to be down
[16:19:34] <Berny> hi laca
[16:19:53] <laca> hey Berny
[16:20:23] <Berny> laca: do you know if anyone successfully build fspot yet (and got it working?)
[16:21:08] <laca> well, an older version worked okay(ish) for me
[16:21:14] <laca> haven't tried the latest yet
[16:22:09] <Berny> oh well i'll check it again some time...
[16:22:55] <Berny> another question: if i add a drawer to the panel and put some stuff inside the drawer is there a way to have it display a text with the icon?
[16:22:58] <laca> i was playing with dvdstyler last weekend, it's quite nice
[16:23:17] * Berny might give that a try
[16:23:42] <laca> hmm... i don't use drawers, let me see
[16:30:09] *** alanc_away is now known as alanc
[16:30:41] *** gm152 has quit IRC
[16:30:42] <laca> Berny: so, it looks like drawers are just icon containers, but what you can do is
[16:30:55] <laca> create a custom submenu and add it to your panel
[16:32:16] <edwardocallaghan> *stupid question: how do you add a shutdown button to Gnome on the Sun Java Desktop?
[16:35:48] <laca> edwardocallaghan: well, can you create a custom launcher for any command you want to run
[16:37:18] <edwardocallaghan> so init 0 right?
[16:37:34] <laca> the problem is, you need to do that as root
[16:37:47] <laca> so you will need to fiddle with profiles and pfexec a bit
[16:37:51] <unixconsole> wacky the download link on the opensolaris.org site brings you to b52 and not b53..
[16:38:17] <richlowe> if you have the snv_53 JDS bits, gksu init 5 may do it, I guess.
[16:38:59] <laca> yeah, that will ask for the root password
[16:39:19] <richlowe> unixconsole: the link is nearly always slightly out of date.
[16:39:26] <edwardocallaghan> Is it not a bit stange how you don't have a shutdown?
[16:39:31] * richlowe doesn't actually know who updates it, either.
[16:39:38] <edwardocallaghan> I sort of understand the reasoning though
[16:39:43] <richlowe> edwardocallaghan: having a shutdown that doesn't need a password would be Bad.
[16:39:47] <richlowe> (think sun-rays)
[16:39:59] <richlowe> So, doing it how, say, Ubuntu does it just wouldn't fly.
[16:40:25] <|tsoome|> there is config utility for gdm
[16:40:42] <edwardocallaghan> Yea I said I understand the reasoning
[16:40:43] <|tsoome|> but dtlogin is different beast...
[16:40:45] <laca> richlowe: what we used to have in jds/linux is, if you were logged in on the console, you could shut down on logout
[16:41:01] <richlowe> laca: are the bits behind gksu aware of RBAC?
[16:41:16] <laca> probably not
[16:41:18] <edwardocallaghan> But what if a home user want to use Solaris ?
[16:41:18] <richlowe> though I'm not entirely certain they could be.
[16:41:33] <edwardocallaghan> "Now how do I shut it down..."
[16:42:12] <edwardocallaghan> I though run level 5 is the run level with X running?
[16:42:19] <edwardocallaghan> Or is that Linux only?
[16:42:44] <richlowe> Not so much linux-only, as much as it varying in general.
[16:42:52] <galt> edwardocallaghan, what about the solaris home user?  I've been using solaris at home as my primary desktop for years.   still no issues
[16:42:56] <richlowe> init(1M) tells you which level does what.
[16:43:09] <richlowe> in this case 5 shuts you down so it's safe to remove power (and attempts to power down itself)
[16:43:17] <richlowe> whereas init 0 shuts you down to firmware.
[16:44:08] * laca uses "poweroff"
[16:44:14] <laca> less geeky, i know... (;
[16:44:15] <edwardocallaghan> ACPI right?
[16:44:22] * galt doesn't turn off his computer
[16:44:52] <edwardocallaghan> When I mean home user I mean the people who should not be on a computer anyway... ;)
[16:47:24] *** |tsoome| has quit IRC
[17:00:50] *** edwardocallaghan has left #opensolaris
[17:06:19] *** Kronuz has joined #OpenSolaris
[17:06:25] <Kronuz> hello
[17:07:20] <Kronuz> hey, I need some advice, has anyone used IBM's SAN systems? (like the DS4700 or DS4800)
[17:08:04] <Error_404> Kronuz: you were the guy building the cluster, right?
[17:08:15] <Kronuz> hehe
[17:08:22] <Kronuz> Error_404: I guess so :)
[17:08:28] <Error_404> i found a link for you a couple weeks ago, haven't seen you since
[17:08:30] <Error_404> http://ha-jdbc.sourceforge.net/
[17:08:43] <Kronuz> hmm
[17:08:51] <Error_404> high-availiability JDBC driver
[17:08:55] <Kronuz> thanks Error_404 :)
[17:09:06] <Error_404> np
[17:09:25] <Kronuz> sounds interesting, I'll have to take a close look at it
[17:09:28] <Kronuz> thanks
[17:09:49] <Kronuz> hey, Error_404, have you used IBM SANs?
[17:10:04] <Kronuz> Error_404: http://www-03.ibm.com/servers/storage/disk/ds4000/ds4700/specs.html
[17:10:04] <Error_404> that, I can't help you with
[17:10:15] <Kronuz> any SAN?
[17:10:23] <Error_404> nope, sorry
[17:10:30] <Kronuz> 'cause the thing is it says it has 8 ports
[17:10:30] <Kronuz> :(
[17:10:53] <Kronuz> well, I ask the question and hope someone will know...
[17:11:10] <elektronkind> ds4700 == sun 6140
[17:11:20] <elektronkind> both are based on the same Engenio chassis
[17:11:36] <Kronuz> elektronkind: yeah, the specs look similar
[17:11:43] <Kronuz> are they both the exact same thing?
[17:11:48] <elektronkind> they're essentiall the same box
[17:11:52] <Kronuz> oh
[17:12:04] <Kronuz> anyway, the IBM's DS4700 and DS4800 say they have 8 host ports
[17:12:09] <Kronuz> and I'm having only four servers
[17:12:25] <Kronuz> I was thinking to connect the servers directly without switches
[17:12:38] <Kronuz> four servers, two connections each = 8 ports needed
[17:12:52] <elektronkind> yeah...
[17:13:16] <Kronuz> so it shouldn't be a problem, but the IBM's sales person is telling me I need the switches 'cause half of the ports are for adding more stuff and can't be used to connect servers
[17:13:33] <Kronuz> it's just that I can't believe it... so I wanted to ask around
[17:14:09] <Kronuz> (I was hoping not to buy the expensive switches at the moment)
[17:14:10] <elektronkind> ah, maybe some of the ports are AL-only and are meant for daisy-chaining arrays
[17:14:25] <Kronuz> oh
[17:14:29] <elektronkind> qlogic sanbox 5200 is a fine, cheap switch.
[17:14:48] <elektronkind> apple sells them at the lowest price I've seen
[17:15:06] <Kronuz> ...so he might be saying the thruth then
[17:15:28] <elektronkind> he might... I'd actually search for the manual for those ibm boxes you're interested in and see what it says
[17:15:38] <Kronuz> elektronkind: how could I know for certain if that's the case?
[17:15:45] <Kronuz> what would I be looking in the manual?
[17:15:54] <elektronkind> read the boxe's manual :)
[17:16:51] <Kronuz> elektronkind: I don't yet have the manual ... and I haven't been able to even find detailed specs of anything on the IBM's website :(
[17:17:14] <Kronuz> hey, google!! :)
[17:17:23] <Kronuz> I think I found it
[17:20:53] *** estibi has joined #opensolaris
[17:21:48] <Kronuz> elektronkind: could you tell anything from their brochure?: ftp://ftp.software.ibm.com/common/ssi/rep_sp/n/TSD01828USEN/TSD01828USEN.PDF
[17:23:16] <elektronkind> I'd just ask your sales rep for a copy
[17:23:29] <elektronkind> to make the sale they'll do something like that
[17:24:23] <Kronuz> I guess I'll do that... I hope they "can" give it to me
[17:24:38] <Kronuz> ^_^
[17:25:26] <Kronuz> ...'cause the link to the manual I found seems dead :(
[17:29:49] <Kronuz> elektronkind: found the manual :)
[17:30:00] <whaq> 'evening all
[17:30:11] <Kronuz> elektronkind: what did you say I should be looking for? ('caus it will say it has 8 host ports)
[17:30:21] <Kronuz> hi whaq
[17:30:29] <whaq> how goes, Kronuz?
[17:31:19] <whaq> elektronkind - how's your NAS project?
[17:31:26] <Kronuz> whaq: great!, we got accepted to the Sun's Startup Essentials Program ;)
[17:31:42] *** jteo has joined #opensolaris
[17:32:03] <Kronuz> whaq: now I'm just trying to figure some stuff about the SAN and comparing IBM/Sun/Dell
[17:32:31] <whaq> Kronuz - congrats :)
[17:32:36] <Kronuz> :)
[17:33:12] <Kronuz> prices look amazing in that program
[17:33:41] <whaq> coolness, what kind of 'amazing' are we talking about?
[17:33:58] <Kronuz> 50-60% discounts!
[17:34:08] *** uebayasi has quit IRC
[17:34:09] <Kronuz> and now that intel released it's quad cores :S
[17:34:28] <Error_404> yeah, i'd take advantage of that program if it were outside the US
[17:34:52] <Error_404> or if i were inside i suppose as well
[17:34:57] *** |tsoome| has joined #opensolaris
[17:35:07] <Kronuz> Error_404: you have to be a US company
[17:35:16] <Error_404> Kronuz: yeah, that's what I mean
[17:35:39] <Kronuz> IBM baldes are pretty nice too tho' (IBM's hardware is pretty cool)
[17:36:04] <Kronuz> specially I liked their blades with intel quad cores
[17:36:04] <unixconsole> anyone know when sxcr 54 will be released? wondering if I should bother with b53.
[17:37:26] <whaq> nice.. too bad we're not eligible for that
[17:37:45] <Kronuz> now it's a tough decision... IBM quad core based blades or Sun's dual core Opteron rack model (half the price)
[17:38:26] *** stevel has joined #opensolaris
[17:38:26] <Kronuz> I guess we'll go for the price, but I'm not yet sure.... IBM could give us some good prices too
[17:38:26] *** ChanServ sets mode: +o stevel
[17:38:29] <unixconsole> you'lll be happier with opteron over intel quad cores..
[17:38:39] <Kronuz> unixconsole: really?
[17:38:50] <Kronuz> that's what I first thought (about intel vs amd)
[17:38:51] *** mjh has quit IRC
[17:38:59] <Kronuz> but now qith the quad cores... I'm not so sure
[17:39:05] <Kronuz> benchmarks look pretty amazing
[17:39:08] <unixconsole> yeah, performance wise the opteron stuff works better
[17:39:12] *** Somethingelse has quit IRC
[17:39:27] <unixconsole> benchmarks only show one view of performance
[17:39:34] *** steleman has quit IRC
[17:40:00] <unixconsole> besides.. if you buy the right opteron tech.. you can swap the dual-core chips out with quad-core's when they become available.
[17:40:05] <galt> the best way to acclerate wintel is 32'/sec/sec
[17:40:18] <Kronuz> supposedly the intel quad cores are about 1.5 times faster the fastest Opteron available
[17:40:43] <Kronuz> unixconsole: what would be the right technology?
[17:41:21] <unixconsole> if I remember it's the f or 1k sockets?
[17:41:23] <Kronuz> I was thinking about the SunFire X4200 M2 servers
[17:41:31] <unixconsole> those are the right ones:)
[17:41:49] <Kronuz> really? it's already known they'll work with the quad stuff?
[17:41:50] <unixconsole> the m2 boxes from sun have the right socket
[17:42:01] <Kronuz> cool
[17:42:15] <unixconsole> the socket was designed to support the quad's in the future
[17:42:26] <Kronuz> too bad Sun doesn't have "nice" blades like IBM's
[17:42:42] <Kronuz> and the blades it does have seem to be for the "old" Opterons
[17:42:55] <whaq> buy today's technology today..
[17:43:31] <Kronuz> hey, if you want to upgrade a CPU I can just get the processors anywhere, right? or do I have to get them from Sun, or IBM?
[17:43:48] <Kronuz> what about the heatsinks? for rack or blade servers they should be passive ones, don't they?
[17:43:54] <unixconsole> I think the 8000 frame blades are m2's as well.
[17:44:46] <Kronuz> unixconsole: I think I only saw Opteron 800 series there in the blades
[17:45:01] <unixconsole> the one thing I don't like about the ibm bladecenters is the need to ditch the frame when major cpu/mem upgrades happen.
[17:45:10] <unixconsole> let me look..
[17:45:44] <Kronuz> unixconsole: do you have IBM's BladeCenters?
[17:46:25] <unixconsole> I have some in our lab.. can't upgrade the blades to the latest.. I think we have p3's or early p4's in it.
[17:46:38] <Kronuz> now I'm thinking about getting single processor servers, and later upgrade to quad cores
[17:46:52] *** nexrafa has quit IRC
[17:46:53] <Kronuz> (from the X4200 M2)
[17:47:24] <Kronuz> If we get four of them, for instance, we can later put two of them with quad cores and two with dual cores
[17:48:34] <Kronuz> hence my question... can I just go to Best Buy or whatever and get the quad core Opterons and put them myself or I have to call Sun to make the upgrade?
[17:49:41] <Kronuz> I think we'll go fr Sun (prices will are great)
[17:50:12] <unixconsole> yeah, it looks like they are 800 series chips on the sun 8k blades.  it may be cheaper to go with the 1u boxes:)
[17:51:38] <Kronuz> yeah
[17:51:47] <unixconsole> from what I've seen sun does have some good prices on the cpu modules on the opteron side of things. I haven't been able to beat the price of the ultra20 by building something simliar. as for upgrading the cpu's yourself, I've done it on a v20z and it worked. really it just comes down to convience.
[17:51:54] <Kronuz> I didn't like the Sun's blades (they were out of the table very soon)
[17:52:14] <unixconsole> when you buy the modules from sun they include the heatsync/fan together.. so it's easy to install.
[17:52:36] <unixconsole> yeah.. sun's blade frames are still very telco oriented.
[17:53:03] <Kronuz> unixconsole: that's what I thought the heatsinks are different
[17:53:17] *** sol1 has joined #opensolaris
[17:53:25] <sol1>  i installed xmms it needs libgmodule..
[17:53:27] <Kronuz> I also like a lot the idea of having Solaris running :)
[17:53:30] <sol1> where i can get that ?
[17:54:08] <unixconsole> on the v20z's I've just pulled the heatsink from the old cpu and stuck it on the new one.. worked fine.
[17:54:09] <sol1> anyidea ?
[17:54:39] <sol1> dwc-:  i installed xmms it needs libgmodule..
[17:55:03] <unixconsole> with the sun gear you definitely want to use the sun parts for support and keeping things clean. not like you can do a lot of customization on the ibm blades or anyones blades for that matter.
[17:55:05] <sol1> where can i get that ?
[17:57:29] *** sol1 has quit IRC
[18:00:21] <Kronuz> unixconsole: aghrr! I forgot the differences between the X2000 and X4000 series
[18:02:04] <Kronuz> elektronkind:, how did you find out the IBM DS4700 and the StorageTek 6140 is the same one?
[18:02:27] <unixconsole> yeah, there are some differences.. the x21* and the x41* are the older opteron tech. the x22* and x42* are the next gen.
[18:02:43] <unixconsole> and the x4500 and x4600 are a bit larger;)
[18:04:32] <Kronuz> unixconsole: not entirely... there are also X4100 and X2100 with the new Opterons (X4100 M2 and X2100 M2)
[18:04:44] <Kronuz> so I'm still trying to figure out (again) the differences
[18:04:46] *** calumb has joined #opensolaris
[18:05:10] <unixconsole> kronuz: oh yeah.. forgot about that
[18:05:42] <Kronuz> should you buy a bunch of of boxes or just a couple with lots of memory and processing speed....
[18:05:51] <Kronuz> the neverending question ^_^
[18:06:08] <Kronuz> the biggest dilemma on earth!
[18:06:09] <unixconsole> depends on the application.
[18:06:17] <Kronuz> yeah :P
[18:06:27] <unixconsole> web, app, or database?
[18:06:33] <Kronuz> the only thing that would demand a big machine is the Pgsql database
[18:06:47] <unixconsole> or oracle:)
[18:06:48] <Kronuz> unixconsole: the other thing is a web app and mail server
[18:07:05] <Kronuz> unixconsole: that's totally out of the table it's way way too expensive
[18:07:12] <Kronuz> :P
[18:07:21] <unixconsole> hehe..
[18:07:56] <Kronuz> 40K for each core! and that wouldn't even solve the problem, I'd need the cluster stuff license which is even more expensive I presume (it says "call")
[18:08:07] <unixconsole> well for a db server something like the x4200 or a x4600 might be good. for webservers, you can go the x2200 path.
[18:09:13] <unixconsole> clustering depends on which solution. if you go with sun, you can do the jes pricing and it's not too bad. you could also go the vcs path.
[18:09:39] <Kronuz> is there a "free" solution ? :P
[18:09:44] <Kronuz> hehe
[18:10:15] <Kronuz> (free as in beer)
[18:10:18] <unixconsole> I think there is, but I can't remember the name of the project. at some point most of the jes stuff will be part of solaris.. including the clustering so..
[18:12:17] <unixconsole> for web serving a dual cpu or dual-core setup works well behind load balancers. depending on your web app( think java), you could even save $$ by using a t1000. those are really good on performance when you have lots of web transactions
[18:13:26] <Kronuz> it seems the power supply doesn't have a redundant option in the X2100
[18:13:45] <unixconsole> this is true of many 1u servers :(
[18:13:46] <Kronuz> but if I'm having a bunch of those, I wonder if that's really needed...
[18:14:15] <Kronuz> I guess one can still connect them to a SAN
[18:14:18] <unixconsole> not really.. that's the point of the 1u box.. buy many.. depend on the redundancy of the network.
[18:14:52] <Kronuz> what happens if the power dies on one of them?
[18:15:03] <Kronuz> it would crash the file system even if it's in a SAN?
[18:15:08] <Kronuz> I suppose so...
[18:15:12] *** ada has joined #opensolaris
[18:15:58] <Kronuz> I proly want a bunch of small servers and a couple big ones for the DB
[18:18:07] <Kronuz> tho' I'm not sure... maybe one can use some sort of middleware balancer/replicator like pgpool-II instead
[18:18:28] <Kronuz> (but I don't know how that would work, I've never used it)
[18:19:22] <unixconsole> do you need san on the x2100? are you placing your db there?
[18:19:30] <Kronuz> yep
[18:19:45] <Kronuz> I'm thinking of putting the DB in the SAN
[18:19:52] <Kronuz> (I hope it's a good way of doing it)
[18:20:10] <unixconsole> well if the power supply dies, you're down unless you have clustering.
[18:20:25] <unixconsole> definitely good to put the db on a san.
[18:20:27] <Kronuz> then I was thinking about using pgpool or something to load balance the database
[18:20:52] *** sommerfeld has joined #opensolaris
[18:20:52] *** ChanServ sets mode: +o sommerfeld
[18:21:08] <unixconsole> that's another option. not sure if pg works on NAS, but that might be an option for you.
[18:21:11] <Kronuz> but I'm not sure up to what point would be good to have many small servers vs. a big one using pgsql and pgpool-II
[18:22:11] <Kronuz> I'm sure small servers will reach a point where they can't handle pgsql even with pgpool-II
[18:22:12] <unixconsole> well I think the basic question is.. "how much uptime do you want?"
[18:22:30] <Kronuz> 99.9% :P
[18:22:30] <Kronuz> (or better)
[18:23:19] <Kronuz> but having small servers, pgsql will get to a point where it can't do the job, wouldn't it? (specially during writes) if the server is not big enough
[18:23:29] <unixconsole> then you should get bigger boxes.. make sure you have redundant networking and san paths. dual power supplies. do multipathing.
[18:25:07] <unixconsole> that is always a risk. that's why multipathing is important. you place one hba on one pci bus and the second on another pci bus slot. have your hba's go to different fc switches. same with networking.. you place your public drops on the onboard network and on a pci card. do ipmp and mpxio for multipathing.
[18:25:53] <unixconsole> same with power.. have one power supply going to one pdu/ups and the second going elsewhere.
[18:26:45] <unixconsole> sadly.. adding that level of redundancy basicly means buying twice the equipment needed to do the job. but you'll get your uptime.
[18:27:06] <unixconsole> it's also one the points of failure for a bladecenter.. if the backplane dies.. all of your servers in that frame are down.
[18:27:22] <postwait> rather buying exactly the equipment you need to do the job
[18:27:24] <unixconsole> so there are pluses to avoiding the blade route.
[18:27:29] <postwait> not buyingi t would be buying half the equipment you need.
[18:28:09] <unixconsole> blades are okay for web farms.. but not db's or app servers that need redundancy. even in a web farm, you'd want to blade frames.. so it's expensive either way.
[18:28:13] *** nprice has joined #opensolaris
[18:28:25] <unixconsole> if you look at it the right way, this is true..
[18:28:26] <delewis> greetings
[18:30:29] <unixconsole> two.. not to..
[18:31:25] *** Snake007uk has quit IRC
[18:31:57] <unixconsole> kronuz.. so it might be good to buy a couple of x2100m2's for your web stuff. but of for the db stuff, you probably want two x4200m2's with two hba's and a quad gige card.
[18:32:26] <Kronuz> unixconsole: that's exactly what I was thinking right now
[18:32:46] <unixconsole> just have to check if you can fit those three cards in the box.
[18:33:56] <Kronuz> the x4200 m2 come with quad gig-ethernet don't they?
[18:35:05] <|tsoome|> www.sun.com ;)
[18:43:53] <unixconsole> yeah, but for redundancy you should put half of your public network and cluster heartbeats on the onboard ports and the other half on a quad card. that way if something shorts, you are still up.
[18:46:46] *** alanc is now known as alanc_away
[18:47:59] <unixconsole> from looking at sunsolve that would work.
[18:48:36] *** yusufg has quit IRC
[18:48:37] <Kronuz> unixconsole: yeah, but only four gig-eth ports are needed, two for the hearbeats and two for the network
[18:49:02] <Kronuz> tho' I'm sure the HBA is going to have to be a dual port for the 1U... :(
[18:49:14] <Kronuz> there's probably not enough space to put two cards there
[18:49:51] <Kronuz> hey, what about scsi arrays instead of a SAN?
[18:50:02] <Kronuz> they must be cheaper, right?
[18:50:08] <Kronuz> (and probaby faster too)
[18:50:19] <|tsoome|> don't be so sure:P
[18:50:34] <unixconsole> well for a 1u.. you'd need a dual-port hba.
[18:50:52] <unixconsole> as for arrays.. fc is definitely faster.
[18:50:58] <Kronuz> yeah... and if it fries, there goes the redundancy :P
[18:51:09] <Kronuz> (the hba)
[18:51:22] <jengelh> then don't fry it
[18:51:23] <jengelh> bake it
[18:51:24] <unixconsole> and that's why you would not use a 1u for an HA solution.
[18:51:39] <unixconsole> not enough slots or power supplies
[18:51:49] <Kronuz> still if it dies some other server could take over
[18:52:13] <unixconsole> this is true. then it comes down to what cluster solution you'll use.
[18:52:16] <Kronuz> so I figure there's no much need for redundancy if it works like that
[18:52:35] <unixconsole> again.. it depends on the level of redunancy/uptime you need.
[18:52:50] <Kronuz> well, for pgsl I'll have to have two "big" servers
[18:52:51] *** dunc has joined #opensolaris
[18:53:21] <unixconsole> keeping the db up and consistent (without errors) is the most important.
[18:53:41] <Kronuz> I'm thinking of using single processors now for the DB
[18:53:44] <unixconsole> failovers are costly in time.
[18:53:51] <Kronuz> (and 16GB of ram)
[18:53:56] <unixconsole> unless you have the right solutions in place.
[18:54:08] <Kronuz> aghrr! this is all too complicated :P
[18:54:19] <Kronuz> seems like anything would work ^_^
[18:54:37] <unixconsole> well I would figure out the cluster part first.. that'll determine how much of a budget you have to work with.
[18:54:51] <sickness> unixconsole: what's the point in having double everything on one box, if you have multiple boxes and no box is a single point of failure?
[18:55:43] <Kronuz> that's what I said, there's no point (as unixconsole said, it comes to the cluster solution)
[18:55:46] <sickness> unixconsole: isn't cheaper to go anyway for 1 power supply per box, and distribute the plugs among the lines/ups?
[18:56:11] <sickness> I know that's a "poor man's" ha, but still... =P
[18:56:39] <unixconsole> depends on the application. if you don't want to go down because of networking, power, storage, etc. ontop of server failures.. you either need many servers.. or a few highly redundant components in a few servers.
[18:56:41] <Kronuz> sickness: well, don't be so sure, I'd say it's a smart move for some solutions
[18:57:55] <Kronuz> (that's how biology works, and it works, een when not everything has to be working at the same time)
[18:58:12] <sickness> yeah...
[18:58:19] <Kronuz> ... databases are not as easy... 'cause they don't cluster well (yet)
[18:58:24] <twincest> kronuz: 1 cpu sounds too little for 16GB, unless you have a very large working set
[18:58:29] <sickness> yeah, I know...
[18:58:35] <Kronuz> (orcale doesn't count, it's too expensive)
[18:59:18] <Kronuz> twincest: well, that's the thing, I'm not sure how to calculate the proportions
[18:59:30] <Kronuz> I just know the little formula: "disk > ram > cpu"
[18:59:35] <Kronuz> (for databases)
[18:59:40] <unixconsole> free db's have their cons.. when it comes to reliablity and scaling.. they fall short.
[19:00:13] *** dlynes_laptop has quit IRC
[19:00:16] <Kronuz> unixconsole: the thing is non-free are prohibitive
[19:00:19] <sickness> best I tried so far, without selling your soul and your house at least, was the mysql ndbcluster thing, it has a lot of limitations, but if you can get over it, you can build a fairly available cluster with 3+ boxes and no shared storage =)
[19:00:55] <unixconsole> the other aspect to think of is manageability.. it takes more effort to manage 100 1u servers than a dozen highly available servers.
[19:01:09] <Kronuz> sickness: your soul and your house *and* everything else (probably even your luck)
[19:01:18] <sickness> Kronuz: eheh
[19:01:29] <Kronuz> unixconsole: that's true
[19:01:34] <sickness> unixconsole: yeah, that's true, a few "big" boxes will end up consuming less energy too...
[19:01:41] <unixconsole> and space;)
[19:01:48] <sickness> heh, yeah, that too
[19:01:56] <Kronuz> sickness: but a hundred boxes look nicer too
[19:02:01] <sickness> eheheh
[19:02:06] <Kronuz> lol
[19:02:19] <sickness> more blinky lights to make a pic and post on your blog ;P
[19:02:20] <sickness> ghgh
[19:02:24] <twincest> 100 servers also costs more in power and rackspace
[19:02:32] <Kronuz> yeah
[19:02:35] <twincest> and infrastructure (networking e.g.)
[19:02:42] <Kronuz> too many is bad too
[19:03:02] <Kronuz> but there's a good balance
[19:03:02] <Kronuz> I'm trying to find it :)
[19:03:26] <twincest> we went with the '100 servers' approach and it works okay for us
[19:04:38] <twincest> there is something to be said for being able to buy + install 20 new servers with no changes, and have it work
[19:05:39] *** slowhog has joined #opensolaris
[19:05:59] <Kronuz> twincest: I figure I eventually have a bunch of boxes
[19:06:35] <Kronuz> twincest: but I want them to be enough with a couple to start with, so they should be big (I'll end up with bunch of "big" boxes)
[19:06:42] <Kronuz> (not so big, tho')
[19:06:59] <twincest> what are you actually doing, anyway?
[19:07:32] <Kronuz> twincest: magic
[19:07:48] <Kronuz> twincest: no, really, it's one of those new big new-bubble projects, but I 'can't tell much about it (they made me sign a whole bunch of papers) :(
[19:08:46] <Kronuz> I'm at the brink of getting fired too, 'cause it's taking me so long to get the servers :S
[19:08:52] *** mazon is now known as Mazon
[19:09:40] *** Somethingelse has joined #opensolaris
[19:10:44] <Kronuz> especially 'cause they think I know *everything* I need to know I told them I was a "wizard" at anything they asked me and I was able to fool them, but ... I I'm not... hehe
[19:11:03] <Kronuz> <.<
[19:11:03] <Kronuz> >.>
[19:13:05] <twincest> maybe you should stop worrying about details and just implement something that works.  you can always change later
[19:13:25] <jengelh> oh no
[19:13:33] <Kronuz> yeah, but it has to work fine, otherewise I'll have problemsm big problems :S
[19:13:46] <jengelh> if you implement it now, and "change later" you end up with cruft over cruft
[19:13:46] <unixconsole> it's easier to do it right the first time than to go back to the well after a major failure to fix something *you* put in place:)
[19:13:47] <Kronuz> it should work for the nex couple years without adding stuff
[19:14:23] <Kronuz> unixconsole: and that's right, if I get it wrong, I won't be the one going back :S
[19:14:53] <twincest> unix: not if you take so long to do it right that it never gets done
[19:16:12] <unixconsole> doing it right doesn't take that much time. it just takes money and planning.
[19:16:19] *** alanc-away is now known as alanc
[19:16:26] <Kronuz> twincest: yeah, it has taken me two months already and they're getting suspicious
[19:16:43] <Kronuz> dangerously suspicious :S
[19:17:12] <twincest> why did you apply for a job you can't do?
[19:17:29] <Kronuz> 'cause the payment
[19:17:30] <Kronuz> <.<
[19:17:35] <Kronuz> :P
[19:17:46] <Kronuz> and I can do it, I just need to learn more ^_^
[19:17:47] <unixconsole> brb
[19:18:32] <Kronuz> (I haven't selept well in weeks, reading and reading stuff... learning new things and shit... so, I think I'm getting there)
[19:19:25] <lasseoe> so you can't do it.
[19:19:44] <Kronuz> <.<
[19:19:44] <Kronuz> >.>
[19:20:03] <twincest> if you take so long to learn it that you get fired, you clearly can't do it :P
[19:20:26] <Kronuz> lasseoe: (but they don't know that... yet)
[19:20:47] <Kronuz> twincest: nah, I'm not really there yet
[19:20:56] <lasseoe> I mean, I could build a nuclear reactor on my own, I just need to learn more
[19:21:18] <Kronuz> they're still happy; suspicious but happy
[19:21:58] <lasseoe> I know this isn't very nice of me, at all, but I hope they fire you!
[19:22:11] <lasseoe> food.. bbl
[19:22:39] <Kronuz> O_O
[19:23:47] <sickness> lol
[19:24:03] <Kronuz> I mean, I mostly only lack experience, I know about the theory...
[19:24:56] <sickness> well, I used to do this at my previous job, the difference was that I always told *before* to my boss what I wasn't able to do, but he was nice and always let me figure it out :P
[19:25:03] <sickness> so I never had to actually lie to him
[19:25:36] <Kronuz> I mean you don't spend so many years in college and end up knowing nothing... plus I was top of my class :P (not that it makes any difference now that I'm realizing I know little about the real world
[19:26:08] <Kronuz> I can tell you anything about the theory... but that's it (well, now I know a bit more after a few months)
[19:26:10] <sickness> eheh
[19:26:38] <dj2> you can spend as many years in collage as you want, doesn't mean you'll know something about everything. depends on what you studied.
[19:26:39] <Kronuz> but it's hard to figure the size of the servers I'll need for instance
[19:27:40] <Kronuz> dj2: I studied computer science with a major in distributed systems
[19:28:06] <Kronuz> but without real world experience that means nothing
[19:29:04] <twincest> distributed systems?  should've got a job at a research institution or something
[19:29:10] <twincest> (probably a lot more interesting too)
[19:29:46] <Kronuz> yeah, but they're paying me here and giving stock for the new bubble
[19:30:15] <Kronuz> I just sell before it bursts ^_^
[19:30:34] <Kronuz> hehe "the new bubble"
[19:30:38] <jengelh> If it works, sell it. If it works well, sell it for more. If it does not work, quadruple the price and sell it as an antique.
[19:31:01] <Kronuz> "pretty bubbles inthe air"
[19:37:54] *** echelog has joined #opensolaris
[19:39:46] *** yippi has joined #opensolaris
[19:41:08] *** alobbs has quit IRC
[19:42:31] <Kronuz> is PCI-E better than PCI-X ?
[19:43:33] <delewis> yes
[19:44:02] <delewis> the latest revisions of PCI-E, if I'm not mistaken, can approach the throughput of XIO.
[19:44:51] <Kronuz> oh
[19:45:16] <delewis> PCI-X is just a synonym for 64-bit, 66MHz PCI, I think.
[19:45:45] <twincest> it's 64-bit 133MHz according to wp
[19:45:56] <delewis> hmm, maybe it's not a synonym then
[19:46:07] * delewis isn't always up on his peecee terminology
[19:47:10] <twincest> hm, and PCI-X can do 533MHz
[19:47:18] <twincest> PCC-X 2.0 that is
[19:48:20] <sommerfeld> you can plug a traditional PCI card into a PCI-X slot.  PCI-E has a different connector family.
[19:49:28] *** mazon is now known as Mazon
[20:07:34] *** echelog has joined #opensolaris
[20:11:23] *** deather_ has quit IRC
[20:14:04] *** deather__ is now known as deather
[20:15:10] *** Burana has joined #opensolaris
[20:28:49] *** cubb has joined #opensolaris
[20:29:21] <cubb> Does tcpdump exist on solaris 10?  Or is there something similar to tcpdump that comes with sol10 ?
[20:29:42] <twincest> snoop
[20:30:07] <twincest> tcpdump can be compiled
[20:30:08] <Kronuz> what do you think of the Sun Fire X4500 Server?
[20:30:25] <cubb> thanks twincest
[20:30:43] <Kronuz> I'm thinking of using it for the databse server
[20:31:58] <Kronuz> (and maybe even for common storage of the other servers instead of the SAN)
[20:34:01] <Kronuz> it's RAID is software based (ZFS)
[20:34:12] <Kronuz> so I'm not sure how that would affect too
[20:34:17] <Kronuz> (in the speed)
[20:34:18] <twincest> don't run databases on zfs
[20:34:34] <Kronuz> why?
[20:34:58] <twincest> not efficient for the workload
[20:35:00] <Kronuz> it sounded great for a database server... until software raid
[20:35:04] <twincest> (it's being worked on, i think)
[20:35:17] <Kronuz> interesting
[20:35:26] <Kronuz> so it might be a good solution in a few months
[20:35:38] <twincest> i'm not sure there's any reason the X4500 wouldn't work with SVM
[20:35:51] <Kronuz> SVM?
[20:35:55] <twincest> (or vxvm if you like that stuff)
[20:35:58] <twincest> solaris volume manager
[20:36:14] <Kronuz> which is?
[20:36:21] <Kronuz> for hardware RAIDs?
[20:36:24] <twincest> what solaris used for software raid before ZFS
[20:36:31] <Kronuz> oh
[20:36:34] <twincest> aka solstice disk suite before it was bundled
[20:36:37] <Kronuz> and that's fast?
[20:36:48] <twincest> that supports UFS
[20:36:51] <Kronuz> still the 48 drives should give pretty good performance
[20:36:52] <twincest> and UFS is faster than ZFS for databases
[20:36:53] <Kronuz> I think
[20:37:13] <Kronuz> (even SATA drives and software RAID)
[20:37:31] <Kronuz> but I'm not sure
[20:37:40] <Kronuz> what would you think of it as a database server?
[20:37:49] <Kronuz> the X4500
[20:38:24] <twincest> only 16GB RAM and 2 CPUs.  doesn't seem to be scaling much for databases
[20:39:03] <Kronuz> yeah, probably... but the disk access should be awesome
[20:39:07] <twincest> i suppose if you're mostly doing data warehousing that's not a problem
[20:39:10] <Kronuz> very fast, I suppose
[20:39:23] <twincest> kronuz: not as fast as cached data in ram
[20:39:29] *** yusufg has joined #opensolaris
[20:39:30] <Kronuz> nay,maybe it's not a good solution for me after all
[20:39:47] <Kronuz> so databases cache a lot to RAM?
[20:39:56] <Kronuz> pgsql, perticularly
[20:40:10] <twincest> postgres relies on the OS to cache, but the effect is the same
[20:40:25] <Kronuz> so I really want all the RAM I can get
[20:42:18] <twincest> depends what kind of workload
[20:42:46] <twincest> if your working set is too large to cache efficiently, more ram won't help as much as disk speed
[20:43:47] *** klepplap has quit IRC
[20:44:46] *** klepplap has joined #opensolaris
[20:49:32] *** Fish- has joined #opensolaris
[20:49:56] <Fish-> hello
[20:49:59] <Burana> I'm taking my first steps with databases on ZFS. (500MB instance). Somewhere you have to start :-)
[20:53:46] *** Fish- is now known as Fish
[20:58:30] *** _william_ has joined #opensolaris
[20:59:01] <_william_> hi all
[21:01:59] *** qbit has joined #opensolaris
[21:05:17] *** thedonvaughn has joined #opensolaris
[21:05:51] <thedonvaughn> Ok maybe I'm missing something.  I see the links for SXCR 53 are not up.  However everyone tells me to change 52 to 53 in the link, this doesn't work for me.  I switch to 53 and it still prompts me to download 52.
[21:06:25] <thedonvaughn> Now I am using firefox, and the link is extremely long.  I only changing the last part which is the file name. sol-nv-b53-x86-v1-iso.zip
[21:07:05] <thedonvaughn> any suggestions? :)
[21:10:23] <estibi> thedonvaughn: you can copy link and  change from 52 to 53
[21:10:23] *** kloczek has quit IRC
[21:10:39] *** Kronuz has quit IRC
[21:11:09] <estibi> thedonvaughn: it should works :)
[21:11:12] <thedonvaughn> estibi: that's what i mean, i'm trying that it's not working
[21:11:19] <thedonvaughn> it keeps making me download 52 even if i change to 53
[21:11:52] <thedonvaughn> http://192.18.108.228/ECom/EComTicketServlet/BEGINAD2440224CB0914C5123FCA502626549/116692119/1823323851/1/786050/785954/1823323851/2ts+/westCoastFSEND/Sol-Express_b52-x86-SP-G-B/Sol-Express_b52-x86-SP-G-B:10/sol-nv-b52-x86-v1-iso.zip  <- this is the link.  Do I change _ALL_ 52 to 53 or just the filename?
[21:12:08] <jengelh> the filename is not encoded in the name, but in the lots of hexdigits
[21:13:07] <thedonvaughn> jengelh: so if i change to 53, and it says 52 it really is 53?  Also I can type 55 60 any number and it'll still download 52
[21:13:27] <jengelh> like I said
[21:13:29] <thedonvaughn> ahh i see what jengelh is saying
[21:13:32] <thedonvaughn> i read that wrong
[21:13:47] <jengelh> sun download center is almost as half-assed as FilePlanet
[21:14:42] *** klocze1 has joined #opensolaris
[21:14:44] <estibi> thedonvaughn: did you try this: http://javashoplm.sun.com/ECom/docs/Welcome.jsp?StoreId=7&PartDetailId=Sol-Express_b53-x86-SP-G-B&TransactionId=try ?
[21:15:55] <estibi> it works for me :)
[21:16:14] <thedonvaughn> ah ok that works estibi thanks
[21:17:58] <estibi> you're welcome :)
[21:23:23] *** pikapika has joined #opensolaris
[21:24:02] <pikapika> hello
[21:30:23] *** salmandr has joined #opensolaris
[21:32:51] *** crib has quit IRC
[21:38:25] *** estibi has left #opensolaris
[21:38:26] *** estibi has joined #opensolaris
[21:39:39] *** dj2 has quit IRC
[21:42:45] <twincest> how can i turn "a.c b.cpp" into "a.o b.o" in a makefile?
[21:43:16] *** deedaw has quit IRC
[21:44:27] <dwc-> sed?
[21:44:42] <twincest> i was hoping for a make variable to do it :-(
[21:45:27] <springfield> twincest: just use '%.o : %.cpp' and '%.o : %.c'
[21:45:55] <|tsoome|> isn't that already predefined?
[21:46:00] <twincest> springfield: i have a rule which needs a list of sources (to generate a source distribution) and a rule that needs a list of objects (to link)
[21:46:06] <springfield> |tsoome|: depends on the make
[21:46:15] *** |tsoome| is now known as tsoome
[21:47:01] <springfield> twincest: what's your dev. platform? Do you have access to the GNU autotools?
[21:47:15] <twincest> if i was using automake i wouldn't have this problem ;-)
[21:47:55] <tsoome> ah, ok solaris make does have .C->.o
[21:48:09] <springfield> twincest: you need to read up on make then
[21:48:22] <tsoome> and  for .c of course
[21:48:51] <tsoome> twincest: check /usr/share/lib/make
[21:49:08] <twincest> tsoome: for what?
[21:49:16] <tsoome> for rules;)
[21:49:25] <twincest> i know how implicit rules work, that's not what i'm asking
[21:49:40] <springfield> twincest: it sure sounds like it
[21:49:46] <tsoome> :)
[21:50:39] <twincest> how will implicit rules give me a list of source and object files?
[21:51:29] <springfield> twincest: you have to define the source list yourself!
[21:51:52] <twincest> yes, i know that. i have a list of sources. (or i could have a list of objects, whichever)
[21:51:54] <springfield> twincest: you could use a glob if your make supports it I suppose
[21:52:05] <twincest> now i want to create a list of objects from it
[21:52:32] <springfield> wow you took the long way round asking that!
[21:52:48] <twincest> no i didn't
[21:52:53] <twincest> < twincest> how can i turn "a.c b.cpp" into "a.o b.o" in a makefile?
[21:53:12] <twincest> basically i want $(SRCS:.c=.o) except i have more than one source extension
[21:53:13] <springfield> yea, sounds like how can I make object files from source files?
[21:54:55] <Error_404> or "how does a compiler work?"
[21:55:05] *** swa_mobil has joined #opensolaris
[21:55:08] <tsoome> :)
[22:00:56] *** noyb has quit IRC
[22:05:04] *** triplah has joined #opensolaris
[22:08:57] *** laca has quit IRC
[22:09:26] *** laca has joined #opensolaris
[22:13:44] *** kb7sqi has quit IRC
[22:23:14] *** kb7sqi has joined #opensolaris
[22:25:55] *** thedonvaughn has quit IRC
[22:34:47] <twincest> hahah... apparently java acquires __malloc_lock itself
[22:34:57] <twincest> so much for stable interfaces :P
[22:35:28] *** regx has joined #opensolaris
[22:39:03] *** bunker has quit IRC
[22:45:15] <Error_404> I wonder if o'reilly sells a java pocket guide
[22:45:38] *** bobbyz has quit IRC
[22:46:04] <Error_404> or i could google it
[22:47:04] <twincest> does the CDDL let me remove the bit that says 'CDDL HEADER START'?  it's really ugly
[22:49:08] *** regx has quit IRC
[22:50:10] <Error_404> I'm relatively certain it does not
[23:02:13] * twincest wonders what primary_link_map is
[23:02:30] <twincest> (other than seemingly undocumented)
[23:07:42] <twincest> ah, it's set when TLS is initialised
[23:08:31] *** Yamazaki-kun has joined #opensolaris
[23:09:19] *** aFlag has joined #opensolaris
[23:09:51] <aFlag> hello, how is solaris for x86?
[23:10:30] <evad> "Good"
[23:12:08] <aFlag> should I expect more problems than a debian etch linux system?
[23:12:55] <evad> Depends on the hardware, of course
[23:13:04] <delewis> without knowing your problems, it is impossible to say.
[23:15:02] *** yippi has quit IRC
[23:16:34] <aFlag> I want to try solaris because my OS book have a few examples of how some stuff works on solaris, I specially want to try that role-based access control thing.
[23:16:54] <aFlag> is there something else which you find nice that I won't see on my linux machine?
[23:17:55] <evad> Lots of stuff
[23:17:58] <evad> Lots and lots and lots
[23:18:11] <evad> SMC, SMF, SVM, Zones, ZFS...
[23:18:34] <delewis> consistence?
[23:18:39] <delewis> consistency, rather.
[23:19:31] <movement> SMC is something nice?
[23:19:48] <delewis> I thought SMC was dead or dying
[23:19:54] *** hell__ has quit IRC
[23:20:02] <aFlag> zfs is the file system, right? I've heard people saying that's really nice, better than ext3 and reiserfs, but I don't know why. I'm not quite the filesystem wizard, is it faster? More reliable?
[23:20:37] <delewis> it guarantees data integrity, for one.
[23:20:48] <delewis> it also has built-in volume management
[23:21:00] <delewis> which is an after-thought for ext3 and most other filesystems
[23:22:02] <sommerfeld> it's a rethinking of filesystem design starting from a blank sheet of paper.
[23:22:10] <aFlag> what you mean by guaranteeing data integrity? If the power goes down it won't be inconsistent or something like that?
[23:22:34] <delewis> much more than that
[23:22:37] <sommerfeld> aFlag: if the disks lie to you, you'll know.
[23:22:48] <delewis> most people assume that the path between the disk and operating system is error free
[23:22:52] <delewis> this is false
[23:22:57] <delewis> and ZFS will reveal that to you
[23:23:43] <delewis> you'd be surprised what a small variation in your PSU can do to your data integrity from the disk to the controller to the operating system.
[23:23:56] <sommerfeld> zfs never overwrites a block that contains valid data -- it allocates a new block, writes the content, and then when that is successful, updates the parent block and frees the old block.
[23:24:17] <sommerfeld> all block pointers in zfs contain a checksum of the block they point at
[23:24:41] <sommerfeld> so when zfs gets a block back from disk it can tell whether it got the same block it put on disk.
[23:25:06] <sommerfeld> this turns out to enable some seemingly impossible robustness.
[23:25:29] <sommerfeld> if you have a redundant pool, you can dd /dev/random over random blocks of one of the disks and ZFS will just keep going.
[23:25:35] <aFlag> hum, what can you do with that besides knowing you lost data? I don't suppose you can recover using the checksum, cna you?
[23:25:38] <sommerfeld> applications won't notice anything
[23:25:42] <sommerfeld> other than slight slowdown
[23:25:55] <delewis> aFlag, you can recover if your volume is mirrored
[23:26:03] <delewis> and ZFS does that transparently
[23:26:08] <aFlag> oh
[23:26:11] <sommerfeld> aFlag: because ZFS handles the mirroring, it knows where the other copies are and can go find all of htem
[23:26:12] <aFlag> that's cool
[23:26:32] <sommerfeld> normally with mirroring, if the two sides of the mirror differ, you have no way to resolve the argument.
[23:26:37] <aFlag> that's really nice
[23:27:15] <aFlag> and reiserfs and ext3 doesn't do it for you, does it?
[23:27:19] <delewis> heh, of course not.
[23:27:26] <delewis> they have no concept of volume management, period.
[23:27:29] <Auralis> no other fs does
[23:27:33] <delewis> you use external utilities to do mirroring, striping, etc.
[23:27:44] <delewis> like I said, it's an after-thought.
[23:28:09] <aFlag> that's really cool stuff
[23:28:10] <delewis> and you can grow ZFS volumes dynamically
[23:28:17] <delewis> this is by no means new, but ZFS is free
[23:28:18] <cubb> Can someone please help...Why can't I resolve a name keyed inside the /etc/hosts  file ?
[23:28:27] <delewis> and traditionally volume managers that grow volumes dynamically have been pricey.
[23:28:33] <delewis> namely, VxVM.
[23:28:38] <cubb> it only goes to the external DNS server defined in resolv.conf
[23:28:41] <evad> movement: I liked SMC, it was at least a vague-ish attempt at something nice ;)
[23:29:02] <evad> I got the feeling the Solaris 11 dev's wish to work on it for 11.
[23:29:21] <delewis> I thought it was being EOL'd
[23:29:46] <Triskelios> cubb: what are you using to resolve the name? and does nsswitch.conf have 'files' set for ipnodes and hosts?
[23:29:46] <evad> bah :(
[23:29:57] <estibi> delewis: it grows already mounted partitions, like XFS ?
[23:29:59] <aFlag> and all those things are already ported to x86, right? The zfs, smf, and all that
[23:30:14] <delewis> estibi, XFS is just a filesystem.
[23:30:26] <delewis> and yes, it can grow them while they're mounted, hence dynamic.
[23:30:33] <delewis> whlie the filesystem is growing, the user will never notice.
[23:31:24] <delewis> VxVM and the AIX LVM have been doing this for years, but neither of those were free and neither of them guaranteed end-to-end data integrity.
[23:32:53] <movement> aFlag: SPARC and x86 Solaris are not separate products
[23:38:25] <twincest> hmm, i think i found a bug in libc malloc
[23:38:32] <twincest> it doesn't honour #define SEGMENTED as documented
[23:39:37] <cubb> Triskelios: yes , I have "files" for both ipnodes and hosts in nsswitch.conf
[23:40:10] <Auralis> cubb: hosts: files dns  or hosts: dns files ?
[23:40:14] <Auralis> order is importtant
[23:40:27] <cubb> Auralis:   hosts:  files dns
[23:40:54] <cubb> i can get there with the IP
[23:40:59] <Auralis> cubb: restart nscd
[23:41:21] <cubb> ohh...i'm doing  a sys-unconfig now
[23:41:24] <cubb> it's a pain
[23:42:37] *** Burana has quit IRC
[23:44:16] <sommerfeld> twincest: hmm.  I doubt the #define SEGMENTED code has been tested in the last decade.
[23:44:24] *** _william_ has quit IRC
[23:44:38] <twincest> sommerfield: well, i just tested it ;-D
[23:44:46] *** yippi has joined #opensolaris
[23:44:47] <sommerfeld> heh
[23:45:29] <sommerfeld> what are you attempting to do with it?
[23:45:30] <twincest> (i needed a malloc implementation that can work with this non-contiguous vm system)
[23:45:43] *** Tpenta has joined #opensolaris
[23:45:55] <twincest> sommer: malloc implementation on top of an L4 kernel
[23:46:13] <sommerfeld> try libumem instead
[23:46:15] <twincest> which likes to give you physical memory in non-contiguous chunks
[23:46:20] <sommerfeld> "L4" ?
[23:46:25] <twincest> hmm, well malloc works now i fixed the bug
[23:46:38] <twincest> http://en.wikipedia.org/wiki/L4_microkernel_family
[23:46:49] <sommerfeld> libumem should do much better in multithreaded programs
[23:49:06] *** bougie has quit IRC
[23:53:25] *** ShadowHntr has joined #opensolaris
[23:53:46] *** jamesd_ has joined #opensolaris
[23:59:42] <twincest> hmm, libumum looks similar to something i already have on top of malloc





top