[00:00:08] <richlowe> sahafeez: orange line goes down into fairfax county, blue/yellow line down through arlington toward springfield.
[00:00:22] <sahafeez> i know, my dad built it ;)
[00:00:23] *** nwf has quit IRC
[00:00:30] <sahafeez> thanks, tho.
[00:00:32] <delewis> Kronuz: IBM redbooks are helpful too
[00:00:37] <delewis> which is what Blueprints were modeled off of
[00:00:58] <hile_> a support contract for SunSolve and IBM tech access
[00:01:03] *** hell` has quit IRC
[00:01:06] <delewis> just ignore the IBM-specific stuff in the books (unless you're interested) and focus on the concepts and practices
[00:01:15] <hile_> as well as being good friends with your local SSE/ IBM CE, etc
[00:01:29] <Kronuz> this is a too big world I'm getting into... I'm beginning to think
[00:01:36] <sahafeez> i was generally bitching about the cost. seems to go up $100 per year..
[00:02:19] <hspaans> Kronuz: just start small
[00:03:17] <hspaans> sahafeez: how much is an appartment in that area?
[00:03:39] <sahafeez> no idea.
[00:03:54] <sahafeez> alot
[00:04:15] <hspaans> just wondering
[00:05:23] <richlowe> hspaans: a whole lot.
[00:05:33] <richlowe> unless it's further into eastern DC
[00:07:42] <Kronuz> okay, now where do I get SX and SX:CR ?
[00:08:19] <Error_404> http://javashoplm.sun.com/ECom/docs/Welcome.jsp?StoreId=7&PartDetailId=Sol-Express_b51-DVD-x86-SP-G-B&TransactionId=try
[00:08:22] <Error_404> that's the DVD versoin
[00:08:25] <Error_404> version
[00:08:35] <Error_404> http://javashoplm.sun.com/ECom/docs/Welcome.jsp?StoreId=7&PartDetailId=Sol-Express_b51-x86-SP-G-B&TransactionId=try
[00:08:38] <Error_404> for CD's
[00:08:57] <Kronuz> and you say it's free for commercial and non-commercial use?
[00:09:06] <Kronuz> or just for non-commercia use
[00:09:11] <Error_404> both
[00:09:14] <Kronuz> hmm
[00:09:27] *** kupfer has joined #opensolaris
[00:09:28] *** ChanServ sets mode: +o kupfer
[00:09:31] <Kronuz> why is that :Get Solaris Express for
[00:09:31] <Kronuz> Noncommercial Use
[00:09:46] <Error_404> *shrug*
[00:09:53] <Tpenta> good morning mike
[00:09:59] <kupfer> hi Alan
[00:10:10] <Error_404> probably because SX:CR doesn't have any support
[00:10:19] <Error_404> SX has some minor support, i'm told
[00:10:39] <twincest> SX requires a license for commercial use, doesn't it?
[00:10:52] <richlowe> kupfer: g'afternoon.
[00:11:05] <Kronuz> Solaris Express Subscription: $99
[00:11:10] <Error_404> twincest: i've never heard anything of the sort
[00:11:11] <Kronuz> (Commercial Use
[00:11:12] <Kronuz> )
[00:11:19] <kupfer> hi Rich.  thanks for forwarding me the tools list.
[00:11:22] <Error_404> oh... how bizzare
[00:11:29] <Kronuz> http://www.sun.com/software/solaris/solaris-express/get.jsp
[00:11:42] <hspaans> Kronuz: but why not just download solaris 10?
[00:12:09] <Kronuz> I was just wondering... plus I couldn't get it to work in my computer the last time I tried
[00:12:27] <Kronuz> (it freezed)
[00:13:06] <Kronuz> also, 'cause I had the idea that using LiveUpdate was a network upgrade
[00:13:14] <Kronuz> I suppose patches are better
[00:13:22] <delewis> much better
[00:13:25] <delewis> they can be rolled back
[00:13:27] <twincest> not better, it's a different thing
[00:13:30] *** Gman has quit IRC
[00:13:39] <twincest> you can't use patches to upgrade to a new release and you can't apply patches with live upgrade
[00:13:45] <Kronuz> hspaans: I come from the FreeBSD world, where you upgrade via the network
[00:14:12] *** Gman has joined #opensolaris
[00:14:15] <Error_404> i think the 'commercial use' option is just for the web based support deal
[00:14:26] <Error_404> and i don't think the same license applies to SX:CR
[00:14:30] <Error_404> but i could be wrong
[00:15:07] <Kronuz> do the releases of Solaris 10 change to include the new patches?
[00:15:24] <twincest> yes, there are update releases every so often that include the current patches
[00:15:29] <hspaans> yes those called updates and 11/06 is the next one
[00:15:32] <twincest> (e.g. s10 6/06, which is the current version)
[00:15:43] <Tpenta> as well as new h/w or feature support (we don't currently deliver new packages in patcehs)
[00:15:43] <Kronuz> as for instance if I got version 10 last year or a few months ago then it's differen from the one online right now
[00:15:55] <twincest> kronuz: depends how many months
[00:16:14] <Kronuz> (should be old)
[00:16:28] <Kronuz> I don't think there was a DVD version of it
[00:16:45] <Kronuz> (maybe there was... but I didn't get it so meyb enot)
[00:17:43] <twincest> Tpenta: does that mean SUNWzfs* were in 3/05 and just had no contents until the patch delivered the actual files?
[00:18:02] <Tpenta> zfs was a rather special exception :)
[00:18:37] <Kronuz> jmcp: hey, is there anything I have to chack on my servers so that I know they can run as a cluster alright?
[00:18:42] <twincest> i want more exceptions, upgrades are annoying :)
[00:18:45] <Kronuz> (before I buy them)
[00:18:57] <Tpenta> so use live upgrade
[00:19:09] <edp> does live upgrade work with non-global zones?
[00:19:10] <Kronuz> like the number of ethernet cards or something
[00:19:13] <jmcp> Kronuz: at least 2 cpus, at least 2 hbas, at least 4 ethernet interfaces
[00:19:17] <jmcp> edp: not yet
[00:19:18] <jmcp> :(
[00:19:26] <edp> not even in sx-cr?
[00:19:27] <Tpenta> edp: not yet, actually very shortly
[00:19:32] <delewis> nice :-)
[00:19:33] <Kronuz> jmcp: 4 ehternet per server ? O_O
[00:19:39] <delewis> I've been wanting Live Upgrade non-global zone support for awhile.
[00:19:56] <delewis> Kronuz: particuarly on separate PCI busses, if possible.
[00:20:03] <jmcp> Kronuz: heartbeats as well as IPMP for public interfaces
[00:20:04] <Kronuz> each of the four?
[00:20:11] <jmcp> Kronuz: no
[00:20:12] <Kronuz> hmm
[00:20:23] <jmcp> Kronuz: two interfaces for heartbeats (crossover to your other node)
[00:20:36] <twincest> most of sun's x86 stuff has four nics builtin these days
[00:20:38] <jmcp> and two interfaces configured with IPMP for delivery to the public lan
[00:20:45] <delewis> well, if you go hot-swapping, you don't want to take down all of your interfaces to remove one (provided your hardware supports hotswapping)
[00:21:00] <jmcp> Kronuz: see, it takes planning ... and if you want redundancy it costs
[00:21:09] <Kronuz> yep
[00:21:34] <Kronuz> well, my servers have two HBA cards and 2 processores, but only dual ethernet
[00:21:46] <jmcp> you'll need more nics
[00:21:59] <Kronuz> four sounded too much, but it's true (for the heartbeats)
[00:22:13] <Kronuz> could it be better to use something else for that?
[00:22:24] <Kronuz> or dfo they have to be gigabit too?
[00:22:34] *** pikapika has quit IRC
[00:22:37] <delewis> not unless you require it
[00:22:39] <jmcp> you can use fast ethernet
[00:22:54] <Kronuz> oh, 'cause if it's just for the hearbeat, I figured
[00:23:10] <delewis> all it does it basically ping the other system
[00:23:14] <delewis> nothing extravagant
[00:23:15] <Kronuz> still they'r eprobably about the same price nowdays... if I can still get fast ehternet
[00:23:20] <twincest> can you still buy fast ethernet cards? :)
[00:23:45] <Kronuz> twincest: yep, that's what I meant ;)
[00:23:54] <Gman> dudes
[00:23:59] <Gman> #opensolaris-i18n
[00:24:04] <Gman> for all your i18n needs ;0
[00:24:08] <Kronuz> lol
[00:24:11] <Stric> Sun still charges big bucks for gigE cards :/
[00:24:29] <delewis> Sun charges big bucks for everything.
[00:24:43] <delewis> Sun Store still tries to pawn off 36GB SCSI drives for $500
[00:24:43] <Tpenta> everything?
[00:24:47] <delewis> :-)
[00:24:48] <Kronuz> I'm still trying to find the right docs in Sun's BluePrints (about clusters)
[00:25:00] <Tpenta> have you done price comparisons of equivalant galaxy h/w recently?
[00:25:05] <Stric> delewis: and more for gigE cards
[00:25:18] <delewis> Tpenta: I'm talking about basic components, not systems.
[00:25:19] <Stric> (which are cheaper than those disks in real life)
[00:25:25] <delewis> Sun Store makes a killing off of that sort of thing.
[00:25:30] <Tpenta> ok, that's not "everything".
[00:25:39] <Kronuz> delewis: EMC was trying to sell me 146GB SCSI drives for $980
[00:25:52] <delewis> Kronuz: that's not as bad as Sun Store :-)
[00:26:00] <Kronuz> that's what I mean
[00:26:13] <Kronuz> and even at that I did't buy it ^_^
[00:26:15] <Error_404> speaking of the opteron systems, what SATA chipset do they use?
[00:26:40] <sahafeez> dude, i remember in 1994 working at IBM they wanted $900 for a 30gb ESDI drive
[00:26:44] <jamesd> Error_404, my u20 uses  nvidia,  thumper uses  maverick
[00:27:09] <twincest> a 30GB ESDI drive?
[00:27:15] <sahafeez> yes
[00:27:32] <Error_404> jamesd: what of the x2100/2200 ?
[00:27:35] <sickness> gisburn: PONG!
[00:27:40] <twincest> error: aren't those SAS?
[00:27:44] <sickness> gisburn: yeah, I was testing belenix usb so I rebooted =)
[00:27:45] <Error_404> are they?
[00:27:46] <jmcp> twincest: yes
[00:27:48] <jamesd> Error_404, not sure i don't own one of those boxes.
[00:27:53] <twincest> the X4200 is, at least
[00:27:56] <twincest> it uses an LSI SAS controller
[00:28:07] <sickness> gisburn: I'll reboot it to your install just now...
[00:28:28] <gisburn> sickness: thanks! :-)
[00:28:56] <sahafeez> twincest: http://en.wikipedia.org/wiki/Enhanced_Small_Disk_Interface
[00:30:20] *** sahafeez has quit IRC
[00:33:44] <Kronuz> hey, can one generally boot from a SAN
[00:34:02] <Kronuz> I mean, the internal drive for each server is no longer needed if you have a SAN, right?
[00:34:04] <delewis> depends on if the HBA supports it
[00:34:05] <jbk> on sparc at least, yes
[00:34:06] <Kronuz> (or it shouldn't)
[00:34:07] <delewis> generally, yes.
[00:34:09] <delewis> and on some x86 systems
[00:34:11] <jbk> with the right hba
[00:34:17] <Kronuz> oh
[00:34:28] <delewis> most Emulexes support it
[00:34:37] <jbk> and qlogic (if bought from sun at least)
[00:34:48] <Gman> stevel, there? and not stressing over a talk you're going to present soon?
[00:34:51] <jbk> though i'd suggest sticking with solaris 10 (or newer)
[00:34:54] <Kronuz> I'm getting a QLogic... but I'm not sure of the model
[00:34:56] <jbk> much easier to do the install
[00:35:06] <richlowe> mornin' Gman.
[00:35:13] <Gman> hey richlowe
[00:35:19] <Kronuz> how do you check for that? is there a "special" name for that booting capbility?
[00:35:52] <Kronuz> also, where exactly do one uses infiniband in the real world
[00:36:10] <elektronkind> Kronuz: are you dealing with a sparc or x86 box?
[00:36:17] <Kronuz> x86 most likely
[00:36:24] <Kronuz> AMD64
[00:36:43] <elektronkind> Kronuz: then the qlogic card has its own boot bios
[00:36:53] <elektronkind> Kronuz: just like a scsi adapter has
[00:37:10] <elektronkind> sparc is a different story though
[00:37:37] <elektronkind> on sparc, you'll need a sun-branded qlogic as it has the fcode for booting off the SAN from OpenBoot
[00:37:38] <delewis> OpenBoot is intelligent enough to read the Fcodes off the device
[00:37:47] <elektronkind> stock Qlogics don't have that
[00:37:47] <delewis> elektronkind: no, Emulex works, too
[00:38:00] <elektronkind> stock emulex or sun-branded emulex?
[00:38:13] <delewis> stock emulex works fine
[00:38:23] <delewis> the low-end models don't support it
[00:38:30] <delewis> but anything you'd actually buy for a production environment does
[00:38:34] <delewis> 9002ls
[00:38:35] <elektronkind> ah ok. I only deal in qlogics
[00:38:54] <Kronuz> what's the "TCP/IP Offload Engine-enabled (TOE)"  ?
[00:38:57] <delewis> and being OBP, I can take my Emulex HBAs and throw them in my pSeries and all works fine :-)
[00:39:04] <delewis> qLogics should be ok there, too
[00:39:07] <delewis> it's an OBP feature
[00:39:30] <Stric> Kronuz: make the card lower your cpu usage
[00:39:32] <elektronkind> the card itself needs the fcode though
[00:39:33] <Kronuz> damn... the more I read stuff the more new words and technologies I see :-(
[00:40:37] <elektronkind> if you buy a qlogic QLAxxxx from joeshouseofsan.com or the like, you don't get the fcode version... which I thought you can only get through Sun
[00:40:45] <Kronuz> look, this is one of the servers I was thinking about buying: http://www-03.ibm.com/systems/opteron/ls21/features.html
[00:40:49] <jmcp> Kronuz: a TOE allows the cpu proper to offload just about all TCP stack operations much like a gpu does for graphics
[00:41:14] <jmcp> Kronuz: does that match the price etc on stuff from Sun?
[00:41:28] <Kronuz> jmcp: what do you mean?
[00:41:50] <Kronuz> jmcp: I just thought about Sun... I'm still waiting for their call :P
[00:41:59] <jmcp> Kronuz: if your cpu (opteron / sparc / whatever) doesn't have to process TCP stuff like packet checksums or encryption, the cpu can do more stuff for you
[00:42:26] <jmcp> oh, vs ibm? well Sun seems to be damned competitive with Dell ... and I assume with IBM too
[00:42:51] <Kronuz> in pricing?
[00:42:53] <Kronuz> ;)
[00:43:19] <edp> in many cases, yes
[00:43:23] <Kronuz> O_O
[00:43:24] <Kronuz> hmm
[00:43:32] <jmcp> Kronuz: and power efficiency too
[00:43:34] <edp> x4100/x4200 are comparable in price to dell 1U and 2U servers
[00:44:04] <Kronuz> what do you think about blades?
[00:44:33] <Kronuz> 'cause that's what we were thinking of buying
[00:44:49] <Kronuz> the IBM BladeCenter H and LS21 servers
[00:45:07] <delewis> Blades work well if you use them properly.
[00:45:08] <Kronuz> (tho' they only can have up to two CPUs per blade)
[00:45:17] <Kronuz> (the LS21)
[00:45:26] <delewis> if you're going to boot off a SAN (as you do with most Blades), you need multiple-paths, etc.
[00:45:42] <Kronuz> yep, we've thought of that already
[00:45:49] <Kronuz> two FC two FC switches, etc.
[00:46:47] <Kronuz> we're spending about 60 grands in equipment and I'm to see what to buy :P
[00:47:18] <Kronuz> unfortunately I'm rather new to the servers world, as you may see... hehe
[00:47:25] <edp> what size SAN does that get you?
[00:47:36] <Kronuz> nah, it's not just for the SAN
[00:47:45] <Kronuz> we're planning on  getting a small SAN
[00:48:00] <Kronuz> about 500GB in five to eight drives
[00:48:07] <edp> ah ok
[00:48:08] <Kronuz> (after the RAID)
[00:48:24] <edp> i've been eyeing the thumper since its introduction
[00:49:43] * Kronuz checking sun's thumper
[00:50:13] <edp> thumper is fairly on par cost-wise with a dell solution of a server + 4 external SAS enclosures
[00:50:31] <edp> though thumper is a *much* more elegant and optimized solution
[00:50:59] <edp> thumper = x4500 server if you're not familiar with it
[00:51:33] <delewis> not to mention you can leverage ZFS
[00:54:04] <edp> is there a way to upgrade sxcr without live upgrade if i'm using zones?
[00:54:06] <Kronuz> what I liked about Sun solutions (in blades at least) is the power consumption as well as they can have four processors
[00:54:32] <delewis> edp: regular upgrade should work fine
[00:54:34] <Kronuz> what I didn't like was the size of the chassis and most likelt I won't like the price :P
[00:54:38] <Tpenta> I don't think the new live upgrade code is in yet (I could be wrong), but if it is not, then it's not far off
[00:54:40] <Kronuz> (as they look expensive)
[00:54:45] <edp> delewis, dumb question, but how do you do that?
[00:54:49] <Kronuz> it was the blade something 8000
[00:54:52] <delewis> edp: boot from the CD
[00:54:59] <edp> everything i read gives me the impression i can boot from the CD, but i get no option to upgrade
[00:55:07] <delewis> and when it comes around to select whether you want to do an initial install or an upgrade, choose upgrade.
[00:55:16] <edp> is that early on in the install?
[00:55:17] <delewis> then your system isn't upgradable :-(
[00:55:21] <edp> heh
[00:55:25] <delewis> edp: right before you slice up your disks
[00:55:29] <delewis> and after you configure the network
[00:55:41] <edp> what do i need for it to be upgradable?  this is a fresh install of b50
[00:55:52] <edp> maybe i gave up too soon.. let me run through it again
[00:57:03] <jmcp> Tpenta: it's not there yet, according to Sarah
[00:57:30] <Tpenta> it's gotta be close as the b51 sparc respin was a premature incorpration of some of that
[00:58:28] <Kronuz> jmcp: hey, seeing what we've talked about the cluster and stuff, do you think I could put the firewall software as part of the cluster or as a separate unit?
[00:58:46] <jmcp> the firewall should be outside the cluster to protect it
[00:59:17] <Kronuz> I understand... everything else I can put inside the cluster... or are there clusters of clusters?
[00:59:23] <Kronuz> O_O
[00:59:26] <jmcp> not really
[00:59:34] <Kronuz> :) good
[01:00:02] <alanc> don't you hate it when you search google for an error message and one of the top hits is your own post asking what it means?
[01:00:09] <Kronuz> lol
[01:00:15] <Tpenta> alanc: :-D
[01:00:16] <Kronuz> alanc: happens every time
[01:00:19] <jmcp> alanc: yeah, had that experience
[01:00:53] <alanc> though since it's an ld error, I could always just walk upstairs and ask Rod
[01:01:12] <edp> delewis, thanks, the upgrade option was there afterall
[01:01:17] *** lacaAFK is now known as laca
[01:02:25] <Gman> we need a RodEvensDrone for questions like that :)
[01:02:31] <twincest> hm, neat advantage of publicly documenting stuff: vendor support engineers already know how your system works before they meet you ;-)
[01:03:01] <alanc> hmm, looks like "R_386_GOTOFF: relocation must bind locally" means that gcc's hidden attribute doesn't play nicely with Solaris ld
[01:05:15] <Kronuz> jmcp: you remember you told me about 4 ethernet cards? would I need four ethernet switches too?
[01:05:22] <jmcp> no
[01:05:27] <Kronuz> :-)
[01:05:29] <delewis> Kronuz: in the ideal world, yes. :-)
[01:05:30] <jmcp> and I said nics, not cards
[01:05:30] <gisburn> alanc: traitor!
[01:05:33] * jmcp re-pedants
[01:05:35] <delewis> but you don't *need* it.
[01:05:36] <gisburn> alanc: wrong compiler!!!!!!
[01:05:39] * gisburn smacks alanc
[01:05:43] <Kronuz> that was worrying me :)
[01:05:46] <jmcp> delewis: not if two of them are for heartbeats and crossover cables
[01:06:06] <jmcp> Kronuz: if you've got more than two nodes in your cluster, you need redundant switches for heartbeats
[01:06:21] <delewis> jmcp: true, you could put one heartbeat and one crossover on and switch and the other pair on another
[01:06:28] <jmcp> true, true
[01:06:37] <Kronuz> O_O
[01:06:40] * gisburn wonders why alanc doens't kick back... ;-(
[01:06:42] <delewis> wouldn't really hurt if both the crossover and the heartbeat went out -- that path is dead, anyway :-)
[01:06:46] <jmcp> except that if you stick to the RAS principles you wouldn't do that
[01:06:57] <Kronuz> o_O
[01:09:19] <Kronuz> jmcp: okay say I have three nodes, each of which has 4 NICs then I'll need just two switches, right?
[01:09:28] <jmcp> no
[01:09:33] <Kronuz> oh
[01:09:52] <jmcp> three nodes requires three switch connections for heartbeats
[01:09:53] <Error_404> here's a question... an unsupported SATA chipset evidently can't hotswap?
[01:10:01] <jmcp> Error_404: correct
[01:10:03] <Kronuz> I can't connect to the same first switch both the first heartbeat and the first network connectin?
[01:10:14] <Kronuz> I'm getting confused :S
[01:10:15] <jmcp> *I* wouldn't do that
[01:10:16] <Error_404> so if a drive dies I'd have to take the machine down... lame
[01:10:18] <jmcp> clearly :|
[01:10:41] *** kupfer has quit IRC
[01:10:41] <Kronuz> I still can't find that darn documentation about clusters
[01:10:57] <Kronuz> I can't find anything at the BluePrints site :(
[01:11:03] <jmcp> Kronuz: there's also docs.sun.com
[01:11:04] <Error_404> what if i get a SAS adaptor & throw cheap sata drives on it?
[01:11:15] <alanc> oh that's where the beeping was coming from while I was off talking, gisburn was taunting me
[01:11:42] <Error_404> oh.. n/m... they cost ~$100
[01:11:46] <Error_404> err, $1000
[01:11:49] <alanc> gisburn: making sure things compile with both cc & gcc is not a bad thing
[01:11:54] <alanc> gives users a choice
[01:11:58] <Kronuz> jmcp: just please don't tell me five nodes require five switches ^_^
[01:12:05] <jmcp> Kronuz: http://www.sun.com/blueprints/0406/819-5783.pdf
[01:12:14] <Kronuz> hmm.. how did you find that?
[01:12:19] <jmcp> Kronuz: if you're seriously thinking about 5 nodes in a cluster, then you *really* must do things the right way
[01:12:20] *** uebayasi has joined #opensolaris
[01:12:24] <jmcp> http://www.sun.com/blueprints/browsedate.html
[01:12:35] <jmcp> then I used the nifty "search" feature that this upstart firefox browser has
[01:13:01] <hspaans> the old fashion search? ;-)
[01:13:08] <jmcp> hspaans: yup
[01:13:31] <jmcp> Kronuz: installation guide part I http://www.sun.com/blueprints/0403/817-2015.pdf and part II http://www.sun.com/blueprints/0503/817-2016.pdf,
[01:13:45] <Kronuz> you cheated, I was expecting "cluster" in the title...  :P
[01:13:46] <gisburn> alanc: I had my fun with gcc in the last days and I have a strong desire to throw some gcc authors into deep pits filled with komodo dragons, rattlesnakes and scorpions.
[01:13:56] <bondolo> i've just tried booting from an nv 51 DVD on x86 and am getting an "Error 28: Selected item cannot fit into memory". I haven't previously seen this error on other install attempts. The machine has 768MB. Did the requirements changed drastically recently or is this a known bug?
[01:14:13] <jmcp> http://www.sun.com/blueprints/1202/817-0905.pdf has "Modelling Sun Cluster Availability"
[01:14:20] <jmcp> Kronuz: you doing a case-sensitive search or something?
[01:14:31] <hspaans> jmcp: a cluster should be certified to get support from Sun if I'm not mistaken?
[01:14:31] <Kronuz> hmm
[01:14:38] <jmcp> hspaans: definitely
[01:15:25] <g4lt-U60> hspaans, that's okay, jmcp is certifiable ;P
[01:15:41] <jmcp> g4lt-U60: that's only on my good days
[01:15:42] * jmcp sobs
[01:15:43] <jmcp> :)
[01:15:44] <Tpenta> jmcp was certified a long time ago :-D
[01:15:48] <hspaans> g4lt-U60: so? my collega also ;-)
[01:15:51] <jmcp> in many ways
[01:16:15] <Stric> certified loony
[01:16:32] <jmcp> Kronuz: this book (http://www.sun.com/books/catalog/elling.xml, isbn 0-13-008458-1) is excellent. http://www.sun.com/books/blueprints.series.html
[01:16:38] * jmcp is biased though
[01:16:51] <jmcp> Kronuz: I was one of that book's pre-pub review team
[01:17:06] <Tpenta> you'll find jmcp's name in the front of it :)
[01:17:09] <hspaans> ieks biased
[01:17:18] <Kronuz> lol
[01:17:44] <jmcp> page xxix :)
[01:18:02] <Tpenta> what is sad is that h e*knows* that page number without having to look it up ;)
[01:18:20] <hspaans> jmcp: you're on a triple X-page? wow
[01:18:30] <jmcp> hspaans: yeah, I know
[01:18:31] <Tpenta> that's kinda scary
[01:18:34] <jmcp> my wife was impressed :)
[01:19:26] <hspaans> no details please ;-)
[01:19:35] <Kronuz> jmcp: with a five nodes cluster, how many nic cards per node and switches would I need (to do it right)... I'll try to ficure out why
[01:19:38] <Kronuz> figure*
[01:19:40] <jmcp> hspaans: I wouldn't tell you anyway :-)
[01:19:54] <hspaans> jmcp: that kinky? ;-)
[01:20:55] <Kronuz> I would guess it would be four switches and four nics per node
[01:21:04] <hspaans> Kronuz: 5 nics per server and 6 switches
[01:21:11] <jmcp> hspaans: nah, it's just none of your business :)
[01:21:26] <Kronuz> hspaans: -_-
[01:21:40] <jmcp> Kronuz: I'd prefer that you actually get a qualified person to help you with this
[01:21:43] <Kronuz> hspaans: you're just trying to get me more confused
[01:22:46] <hspaans> Kronuz: 2x IPMP production (2 switches), 2x IPMP heartbeat (2 switches), 1 backup (1 switch) , 1 RSC (1 switch)
[01:23:24] <twincest> how's the bug database resync coming?
[01:23:30] <Kronuz> jmcp: I'm just trying to get a little knowledge before asking at Sun or IBM guys, just to not ask really stupid questions and give them a way to sell me more than I need or at a worse price ;)
[01:23:58] <jmcp> you won't be asking stupid questions because you'll have read that book I just pointed you too, and thought about it
[01:24:18] <hspaans> Kronuz: is people sell you crap then thet can be legally responsible in most countries
[01:24:45] <Kronuz> jmcp: yep, I'll read that one :)
[01:24:47] *** Cyl has joined #opensolaris
[01:25:07] <Kronuz> hspaans: not that they sell crap, but they can sell more than I need or give me a worse price, I would guess
[01:25:11] <jmcp> Kronuz: it should be available from just about any online bookseller without too much hassle
[01:25:13] *** dunc_ has quit IRC
[01:25:22] <Cyl> Hi guys, has anyone ever configured jumbo frames for the nge network interface?
[01:25:27] <jmcp> Kronuz: be very clear when you talk to a vendor about what your requirements are
[01:25:44] <hspaans> Kronuz: that is what sales is for
[01:26:20] <Kronuz> yeah, I guess I'll read more now that I know where to read :)
[01:26:45] <Kronuz> and I'll be back with more questions afterwards... if I still have them ;)
[01:27:00] <hspaans> how fast do you need it and how is the weather in your place? ;-)
[01:27:14] <Kronuz> lol
[01:27:15] <Error_404> weather?
[01:27:51] *** lloy0076 has joined #opensolaris
[01:28:08] *** jcsmith has quit IRC
[01:28:17] <hspaans> its raining and 5 a 10 degrees Celcius over here
[01:28:26] *** jcsmith has joined #opensolaris
[01:29:06] <lloy0076> I've just accidentally discovered that you can run the "installer" script which is in the root directory of the ON (51) DVD iso I just built from downloads...it appears to be a "Live Upgrade".
[01:29:14] <Kronuz> jmcp: I really appreciate the help :)
[01:29:21] <jmcp> Kronuz: you're welcome.
[01:29:21] <lloy0076> Is that expected and what would it do if I actually did the Live Upgrade?
[01:29:34] <jmcp> Kronuz: the principle to keep in mind is "do it right the first time"
[01:29:55] <delewis> which isn't exactly the current hallmark of the IT world.
[01:30:06] <delewis> "do it quick, and enough to satisfy the management, and let the rest burn!"
[01:30:45] <Kronuz> also delewis, hspaans, Stric, Error_404, and Burana although he's not with us anymore :P
[01:30:48] <Kronuz> thank you :-)
[01:30:50] <hspaans> read as "we stay until the money runs out"
[01:31:22] <hspaans> Kronuz: np that is why there is a community
[01:31:25] <Kronuz> jmcp: yeah, that's why I'm doing all this reading and planning anf stuff
[01:31:31] *** bobbyz has quit IRC
[01:31:31] <jmcp> good
[01:31:41] <Kronuz> to get it right from the start
[01:31:43] <Kronuz> :)
[01:31:52] <Kronuz> or at least as right as possible
[01:32:06] <Kronuz> (given my ignorance and inexperience)
[01:32:10] <Kronuz> ^_^
[01:33:20] <hspaans> you have the balls to ask the questions and that is a good sign
[01:33:57] <delewis> much better than believing the hype.
[01:34:08] <Kronuz> yeah, well I just figured "those that ask look stupid but those that don't, stay that way"
[01:34:14] <delewis> which seems to be centered around "poor man's storage/clustering" solutions, nowadays
[01:34:42] <Kronuz> :P
[01:35:54] <Error_404> delewis: ugh, like those "whitebox thumpers" that started showing up about 2 weeks after sun debut theirs?
[01:36:00] <Kronuz> well, thanks again, you've given me a lot of light and a place to really start :)
[01:36:03] <delewis> Error_404: yeah.
[01:36:10] <delewis> "let's just cram a bunch of disks into a case and sell it"
[01:36:16] <Error_404> using lvm or some crap
[01:36:17] <delewis> regardless of any architectural concerns
[01:36:23] <delewis> like multiple controllers, etc.
[01:36:27] <delewis> multiple busses
[01:36:51] <Cyl> I wish I could afford a thumper
[01:36:58] <delewis> Cyl: don't we all :-)
[01:37:06] <Error_404> i would be happy w/ a tyan gt20
[01:37:07] <Cyl> I made the choice between a new car and a thumper
[01:37:11] <Cyl> and the new car won
[01:37:11] <delewis> though, an UltraSPARC-IV in a Thumper would've been a bit more cool, IMO. :-)
[01:37:14] <Cyl> :D
[01:37:33] <edp> i'm hoping for a 24tb thumper at work in the next month
[01:37:37] <Error_404> esp. since my current server is starting to die
[01:37:52] <delewis> well, I'll have to live with a Photon for the next few years :-)
[01:37:58] <delewis> 400GB should be enough for my needs
[01:38:09] <delewis> (I'm using a 36GB disk in my primary workstation)
[01:40:23] <Cyl> I'm jealous of education pricing Sun gives out
[01:40:52] <Cyl> the 24TB thumper is only 35k for EDU
[01:41:16] <edp> heh.. yeah i just got an email about the $35k offer :)
[01:41:32] *** lloy0076 has left #opensolaris
[01:41:33] * delewis would like to see Suns back at the computer labs of every campus
[01:41:40] <Cyl> edp: your in Edu?
[01:41:43] <edp> yeah
[01:41:46] <delewis> I imagine Sun does, too :-)
[01:41:49] <Cyl> college?
[01:41:52] <edp> educational discounts are sweet
[01:41:54] <edp> high school
[01:42:03] <Cyl> your high school needs a x4500 thumper?
[01:42:22] <edp> i'm planning on putting quite a bit on it..
[01:42:35] <edp> right now we have maybe 3 TB of data
[01:42:37] <Error_404> yeah, those highschools produce a remarkable amount of marketing data on the habits of teenagers
[01:42:39] <Cyl> hehe, but you know you really need two x4500 thumpers in a real production environment
[01:42:44] <Cyl> if one fails
[01:42:44] <jamesd> Cyl, of course...   3000 students even with 100MB each student adds up
[01:42:58] <edp> we're looking into some IP security cameras which take up quite a bit of space
[01:43:44] <edp> educational environments are odd when it comes to redundancy though
[01:44:10] <edp> businesses can make an easier case for how much loss there is for X hours of downtime
[01:44:13] <Cyl> jamesd: True, I worked in an univ. engineering comp lab and storage was always being upgraded
[01:45:02] <edp> i figure a bunch of zfs snapshots would add up to a decent amount of storage too
[01:45:34] <Cyl> edp: That's true, edu especially at a HS level could get away with downtime
[01:45:45] <jamesd> edp, snapshots them self don't use storage, only file deletions make  snapshots use space.
[01:45:48] <edp> 100 mb of changed data seems to equate to much more than 100 mb of extra disk space used
[01:46:47] <edp> Cyl, i've found most downtime comes from outside factors or human error anyways
[01:46:48] <Cyl> edp: Although, in most cases unless students are filling their storage space with mp3s and videos (which I'm assuming you'd police) they wouldn't use that much storage
[01:46:54] <_william_> gn
[01:46:55] *** _william_ has left #opensolaris
[01:47:19] <edp> most of the thumper would go towards security camera storage, but i'd like to have plenty of room to grow too
[01:47:27] <edp> you get tired of expanding storage after you do it enough
[01:47:37] <edp> it gives a lot of flexibility too when you have storage to spare
[01:47:50] <Cyl> edp: I know I've been there, ZFS makes it very easy though
[01:48:04] <edp> yep.. it definitely lives up to the hype
[01:48:13] *** d3vi1 has joined #opensolaris
[01:48:15] <Cyl> I'm hoping Coraid will work with solaris x86 soon
[01:48:18] <Stric> except for in various usage cases..
[01:48:27] <edp> coraid does work with x86
[01:48:34] <edp> not sure if it works with zfs though
[01:48:52] <Cyl> Coraid doesn't have a driver for Solaris x86 last time I checked only SPARC
[01:48:54] <Cyl> and that was in beta
[01:48:57] <edp> i tested it out a year ago or so i'd imagine they've made some progress since then
[01:49:30] <edp> i'd been waiting for snapshots in linux for forever but i've never gotten them to work decently
[01:49:39] *** B|nTaRa has quit IRC
[01:50:14] <Cyl> confirmed http://www.coraid.com/support/solaris/ no x86 driver only SPARC, unless you had a development driver
[01:50:28] <edp> let me check my email archives...
[01:50:34] *** B|nTaRa has joined #opensolaris
[01:50:35] *** eboutilier has joined #opensolaris
[01:50:44] *** d3vi1 has quit IRC
[01:50:55] <Cyl> edp: I was using evms to do snapshots in linux but I never tested if they worked right
[01:51:51] <edp> even when i kinda had them working they are geared more for a single snapshot for backup purposes
[01:52:26] <edp> i think i ended up compiling the AoE sparc driver for x86 and it worked
[01:53:04] <edp> i had tried version 1.1 of their driver around the b33 days
[01:53:45] <edp> my turnoff with coraid is that i think you only get 100 mbit to each disk
[01:54:16] *** Xach has joined #opensolaris
[01:54:29] <Cyl> edp: I thought they had GigE interfaces
[01:55:52] <edp> from their docs it seemed like you get 2 GigE to the enclosure, but then 100 mbit to each disk internally
[01:57:10] *** Xach has left #opensolaris
[01:57:22] *** bobbyz has joined #opensolaris
[01:57:29] <edp> they sell a single disk AoE card for developers that runs at 100 mbit so i may have just assumed they used that same thing internally in the enclosures
[01:58:46] <Cyl> edp: hmm I think you can get a gigE to a disk using the enclosures, I talked to some people who only have 2 disks in it and they are able to get 80mb/s or so
[02:00:04] *** Cass has quit IRC
[02:00:09] <hspaans> Java is going to be under GPL? wow
[02:00:43] *** Cass has joined #opensolaris
[02:00:56] <gisburn> hspaans: and then we get Gnuava!
[02:01:12] <Gman> yeah
[02:01:17] *** gonufer has joined #opensolaris
[02:01:21] <hspaans> ieks
[02:01:35] * gisburn greets gonufer
[02:01:56] <edp> Cyl, kinda cool.. coraid sells the spare internal parts for their enclosures also
[02:03:05] *** Kronuz has quit IRC
[02:03:10] <Cyl> edp: granted they are the only one that sells AoE equipment period :D
[02:03:21] <hspaans> but time for bed, g'day to you all
[02:03:24] *** hspaans has quit IRC
[02:03:33] <gisburn> Gman: I hope sun is clever enougth to create a closed-bins for the GPL'ed java that the GNU folks can't fork it ASAP
[02:03:46] <gisburn> =:-)
[02:03:49] <Gman> why would they fork it?
[02:04:07] <gisburn> Gman: do they need a reason ?
[02:04:13] <edp> Cyl, true, it would be nice if there was some competition so their enclosures would drop in price some
[02:04:14] <richlowe> Why would stopping them do anything but make Sun look like asses?
[02:04:18] <g4lt-U60> they can fork themselves
[02:04:27] <Gman> uh, usually the GNU folks are a lot more responsible than you think
[02:04:30] <gisburn> Gman: did Gnome need a reason except "KDE has a wrong license" ?
[02:04:35] <richlowe> it's either open, or it's not.
[02:04:36] <edp> i think coraid still costs more for the enclosure than nthe disks to fill it
[02:04:42] <Gman> GNOME didn't fork KDE
[02:04:48] <Gman> bash didn't fork ksh
[02:04:53] <g4lt-U60> Gman, yeah, two words: info pages
[02:04:54] <Gman> choice is good ;)
[02:05:34] <gisburn> Gman: neither gnome or bash are good examples... really... at least bash is horrible and bug-ridden. And I won't comment Gnome.
[02:05:35] <Cyl> edp: It's not that bad if you compare it against iscsi solutions though
[02:05:45] <richlowe> sheesh.
[02:05:50] <Cyl> edp: I would have purchased one if they had more solid support for Solaris
[02:05:54] <Gman> gisburn, bugzilla.grommit.com thinks otherwise about ksh ;)
[02:06:01] <Gman> back to work gisburn :)
[02:06:07] <gisburn> Gman: where ?
[02:06:14] <Gman> dude, i was joking
[02:06:21] <gisburn> oh
[02:06:22] <gisburn> sorry
[02:06:41] <gisburn> Gman: try to run bash in ja_JP.PCK
[02:06:50] <richlowe> oh boy.
[02:06:57] <gisburn> Gman: and set a native japanese speaker in front of that shell
[02:06:57] <twincest> i doubt GNU would care about a GPL'd Java, they already have their own implementation
[02:07:02] <gisburn> Gman: and then RUN
[02:08:12] <g4lt-U60> twincest, yeah, and they already had their own editor as well and they still attempted to bollix vi
[02:09:36] <gisburn> uhm
[02:09:46] <gisburn> what did change between B48 and B51 ?
[02:09:57] <gisburn> the vmware install is suddenly blazing fast
[02:10:21] *** pablo-costa has joined #opensolaris
[02:10:28] <twincest> galt: there's a GNU vi?
[02:10:35] <gisburn> vim
[02:10:41] <richlowe> is not GNU.
[02:10:48] *** pablo-costa is now known as pacx
[02:11:14] <rydis> viper-mode? ;)
[02:12:30] <pacx> hi
[02:13:09] * gisburn tries to remeber what gonufer was doing at sun...
[02:13:45] *** Gman has quit IRC
[02:14:18] *** Gman has joined #opensolaris
[02:15:48] <twincest> right, vim is not gnu, it's not even gpl
[02:16:01] <gisburn> BSD license ?
[02:16:09] <delewis> no
[02:16:11] <twincest> :help license suggests it has its own 'vim license'
[02:16:12] <delewis> the vim license
[02:16:32] <gisburn> and noone from the GPL folks tried to fork() the project yet ?
[02:16:51] <delewis> why would they?
[02:16:53] <delewis> they have emacs
[02:17:16] * gisburn chuckles
[02:17:22] <twincest> i doubt gnu wants a vi, a GNU OS could just use nvi or vim, and their have their own GNU editor
[02:18:03] *** sch0 has left #opensolaris
[02:18:09] <delewis> I was a bit surprised when Sun chose to GPL Java (though, I think that's still slightly unconfirmed)
[02:18:22] <twincest> me too, CDDL would seem to make more sense
[02:18:24] <delewis> though, I really don't care myself.
[02:18:35] <delewis> I've been using Java stuff for years, and haven't complained about license terms.
[02:18:38] <twincest> unless they intend to share development with gcj - the gcj folks could really use help with GNU classpath, i imagine
[02:18:40] <richlowe> how can something be 'slightly' unconfirmed?
[02:18:41] <richlowe> it's a binary state.
[02:18:46] <Gman> yeah, still to be confirmed, the current stuff is just rumours/leaks
[02:18:58] <twincest> (classpath seems to be the weakest part of gcj by far)
[02:19:05] <Stric> confirmed from "not sure sources" ? :)
[02:19:06] <richlowe> You're all thinking about the wrong question, damnit.
[02:19:12] <richlowe> the real thing to ask is "Will their bug database suck ass?"
[02:19:13] <richlowe> ;)
[02:19:16] <delewis> richlowe: enlighten us, Vulcan master.
[02:19:18] * richlowe makes a hasty exit
[02:19:20] <Gman> haha
[02:19:23] <twincest> actually, the real thing is 'will i have commit access'? :)
[02:19:24] * Gman ^5s richlowe
[02:19:34] <delewis> hah.
[02:19:37] <twincest> richlowe: Java actually has its own b.o.o, has for a long time before opensolaris
[02:19:37] *** solarisjon has quit IRC
[02:19:40] <Gman> i would guess they're in the same situation
[02:19:51] <twincest> and it's less sucky than ours ;-)
[02:19:52] <Gman> twincest, it's as sucky as b.o.o
[02:19:57] <Gman> actually, it's not
[02:20:02] <Gman> it's the same afaics
[02:20:02] <twincest> really?  i got the impression it at least had more details for the bugs
[02:20:05] <delewis> I'm trying to figure if I should bother learning Java or Ruby
[02:20:12] <richlowe> twincest: it does, but it's not as full featured as you'd guess from looking.
[02:20:13] <Gman> only with an option to add 'me too' like comments to each bug
[02:20:17] <twincest> not that i've done anything at all for java development..
[02:20:23] <Gman> which doesn't actually go into the real database
[02:20:40] <delewis> Java looks exciting, but everytime I get into Java I get bogged down by the massiveness of the API, J2EE, EJB, etc.
[02:20:47] <richlowe> it does, however, have many of the features that have been requested for b.o.o and have been told would take time to implement.
[02:21:00] * Gman is hopeful with java and opensolaris around, they'll help each other get where they want to be
[02:21:35] <rydis> delewis: I'd say "Take a look at Common Lisp.", rather than Ruby or Java, but that might not be entirely helpful.
[02:21:57] <twincest> delewis: yes, i actually like java as a server-side platform, but there's so much of it it's very confusing
[02:22:05] <twincest> half the time i don't even know if i'm using j2ee or j2se or what
[02:22:23] <delewis> rydis: I'm really looking for a language to do web development in that isn't PHP.
[02:22:51] <richlowe> ksh93!
[02:23:01] <delewis> richlowe: :-)
[02:23:11] * delewis wonders when gisburn will come up with a ksh93 compiler
[02:23:14] <gisburn> richlowe: nah
[02:23:18] <twincest> ksh93Python
[02:23:23] <gisburn> delewis: that is shcomp
[02:23:28] <rydis> There are quite a few who do that in CL, actually. (I don't much do web stuff, myself, so I don't know the details.)
[02:23:29] * delewis shakes head
[02:23:34] <delewis> I was just joking :-)
[02:23:50] <gisburn> delewis: really... there is a compiler
[02:24:09] <twincest> delewis: why does mplayer unpause when i switch to/from fullscreen?  (recently it's started doing it simply from clicking on it while paused)
[02:24:10] <delewis> how fast is it?
[02:24:10] <gisburn> delewis: http://bugs.grommit.com/show_bug.cgi?id=3
[02:24:22] <gisburn> delewis: you mean the compiled code ?
[02:24:33] <dwc-> twincest: I thought it did that always
[02:24:44] <delewis> twincest: it's always done that, unfortunately
[02:24:53] <delewis> that's an annoyance for me, too.
[02:24:58] <twincest> dwc: it never used to do it from clicking before i upgraded
[02:25:16] <gisburn> delewis: it depends a little on the code and other details (usage of loops etc.) - between 20%-70% performance improvement
[02:25:21] <twincest> delewis: is it a bug or just a really annoying feature?
[02:25:32] <delewis> twincest: something that nobody has bothered to fix, really.
[02:25:46] <delewis> MPlayer does a terrible job of OSD, key bindings, etc.
[02:25:57] <delewis> (which is why we still don't have libdvdnav)
[02:26:45] <delewis> twincest: if you want to, submit a bug report, but I have a feeling it's already in the bug database somewhere.
[02:28:18] <delewis> http://bugzilla.mplayerhq.hu/
[02:29:00] <delewis> hmm, i don't see anything in the bugdb
[02:29:12] <dwc-> twincest: ah... clicking... I usually get "no such binding" when I click
[02:29:43] * delewis is personally considering switching to Xine
[02:29:46] <dwc-> but it's done it on (un)full-screening for as long as I can remember
[02:29:46] <delewis> shhh, don't tell anyone. :-)
[02:30:22] <delewis> putting up with MPlayer politics has taken it's toll
[02:30:45] <gisburn> death toll ?
[02:31:04] <delewis> gisburn: :-)
[02:31:14] <delewis> you'd fit in just fine with your love for gcc :-)
[02:31:15] <Gman> mplayer has a terrible ui
[02:31:21] <Gman> same with xine ;)
[02:31:30] <delewis> Gman: we're thinking of dropping it the next release
[02:31:35] <Stric> xine tries to have a ui, but fails.. :P
[02:31:36] <delewis> it hasn't beem maintained in *years*
[02:31:41] <delewis> s/beem/been/
[02:32:04] <gisburn> delewis: My complain is that some gcc authors are either braindamaged or insane or both when it comes to their warning concept.
[02:32:06] <delewis> there's a few people putting in work for a win32 gui but no one seems to care about replacing the existing one
[02:32:23] <Gman> delewis, just being a platform?
[02:32:28] <delewis> we're all too busy debating how libdvdnav should be integrated
[02:32:37] *** cub has joined #opensolaris
[02:32:47] <delewis> (you don't even want to *know* the points of that argument)
[02:33:38] <delewis> I tend to stay out of the bicking and only drop in when something on Solaris, AIX, or SPARC in general breaks.
[02:33:43] <delewis> bickering, rather.
[02:33:58] <gisburn> delewis: is VIS working again ?
[02:34:04] <edp> gimageview is a pretty good wrapper for xine if you are viewing a lot of media files
[02:34:07] <delewis> gisburn: VIS works fine. mediaLib does not, however.
[02:34:20] <Gman> totem wraps xine too, along with gstreamer
[02:34:21] <gisburn> delewis: is there a non-VIS codepath available, too ?
[02:34:27] <delewis> gisburn: yes
[02:34:36] <delewis> you can just enable a certain macro, and it'll disable the VIS inline stuff.
[02:34:36] <gisburn> delewis: available in the same binary ?
[02:34:45] <delewis> gisburn: no
[02:34:51] <delewis> whic his actually probably something that should be implemented
[02:35:03] <gisburn> delewis: that's bad... for niagara&Fujitsu SPARC64
[02:35:16] <delewis> MPlayer does something like that MMX/SSE/SSE2 but no other SIMD instruction set, like Altivec
[02:35:23] <delewis> gisburn: yes
[02:35:37] <delewis> which is why when sun4v is detected by configure, VIS is disabled :-)
[02:35:38] <gisburn> delewis: I wait for the day when someone discovers that ksh93 uses FP math for some stuff
[02:35:44] <delewis> gisburn: :-)
[02:35:54] <gisburn> delewis: that is stupid.
[02:36:05] <delewis> gisburn: I completely agree.
[02:36:10] <gisburn> delewis: VIS should be enabled in that build.
[02:36:21] <delewis> but unfortunately, patches that do this kind of run-time checking tend to be rejected
[02:36:24] <gisburn> delewis: it should be a runtime switch.
[02:36:33] <delewis> out of the fact that "MPlayer should be built, rather than distributed in binary form"
[02:36:38] <delewis> which I also happen to disagree with.
[02:36:42] <gisburn> delewis: there is an alternative option
[02:36:49] <gisburn> delewis: make the part a shared library
[02:37:08] <gisburn> delewis: and create seperate dirs, one for the vis-enabled and one for the normal codepath
[02:37:26] <gisburn> delewis: and let isalist output select the "right" library
[02:37:59] <gisburn> delewis: niagara has VIS extensions after normal code while UltraSPARC1,2,3,4 have VIS before the normal code in the isalist
[02:38:11] <delewis> gisburn: that would work nicely, but when you're around a team full of Linux-centric developers, getting Solaris-specific fixes committed tends to be difficult.
[02:38:22] <delewis> "drop your broken hardware or fix your broken OS"
[02:38:23] <cub> how do I find out if the  sol10 box I'm on is a u1 or u2  ?
[02:38:50] <delewis> cub: prtdiag -v output should say
[02:39:13] <twincest> the mplayer binary thing annoys me
[02:39:28] <twincest> it probably made sense when CPUs were so slow you needed -O999 to play an mp3..
[02:39:36] *** pacx is now known as pabloocosta
[02:40:03] <delewis> twincest: one of the lead developer still uses a K6 for his primary workstation..
[02:40:14] <twincest> then he should feel free to build it from source
[02:40:27] <twincest> but i usually watch dvds while waiting for ON to build, so i don't really care to :)
[02:40:28] <delewis> oh, trust me he does, and he flames anyone who disagrees.
[02:41:12] <cub> delewis: can't find anything related to u2, unless it's displayed under a different thing
[02:41:32] <delewis> cub: oh, I thought you meant the processor-type
[02:41:37] <delewis> uname -a output should say it
[02:41:56] <delewis> or uname -i, specifically
[02:42:11] <cub> SUNW,Sun-Fire-V890
[02:42:20] <cub> that's the output for my  uname -i
[02:42:29] <delewis> that's definitely not an Ultra1 or Ultra2.
[02:42:30] <gisburn> cub: ahhgggrrlrlll
[02:42:32] <gisburn> cub: ahhgggrrlrlll
[02:42:38] <rydis> I usually use psrinfo -v
[02:42:40] <delewis> but I'll trade you my Ultra 2 for it
[02:42:45] <cub> no no
[02:42:49] <gisburn> cub: can I get an account on that box, please ?
[02:42:56] <cub> heehh
[02:42:57] <gisburn> cub: you have an UltraSPARC-4 machine
[02:43:15] <cub> I wanted to know if the solaris 10 software is  update 1 or update 2
[02:43:22] <twincest> cab: cat /etc/release
[02:43:23] <delewis> oh
[02:43:26] <rydis> (Doesn't say whether it's an US-I och -II, though. Sorry.)
[02:43:53] <twincest> s/cab/cub
[02:44:02] <cub> Solaris 10 1/06 s10s_u1wos_19a SPARC
[02:44:04] <gisburn> rydis: SunFire V890 takes UltraSPARC-4 CPUs
[02:44:19] * delewis wants one
[02:44:20] <cub> that's the original solaris then
[02:44:26] <delewis> no
[02:44:27] <twincest> cub: no, that's update 1
[02:44:29] <alanc> if it wears dark shades all the time and talks about saving the starving kids in Africa, it's U2
[02:44:30] <delewis> FCS is original Solaris 10
[02:44:31] <rydis> gisburn: Yeah, I was just commenting that I was mistaken about psrinfo.
[02:44:32] <twincest> cub: s10 first release was 3/05
[02:44:39] <cub> oh ok
[02:45:01] <alanc> it's update 1, since it says s10s_u1....
[02:45:24] <cub> can "smpatch"  help upgrading to update 2 ?
[02:45:37] <gisburn> cub: any objections against the idea that I can strangle you to death and take your V890 home ?
[02:45:49] *** Error_404 has quit IRC
[02:45:54] <delewis> gisburn: you couldn't lift a V890 :-)
[02:45:59] <cub> heheh
[02:45:59] <twincest> cub: no, smpatch will apply patches.  you will get some of what's in update 2 but not all of t.
[02:46:09] <twincest> s/t/it
[02:46:12] <gisburn> delewis: no, but I can move it to a car.
[02:46:15] <alanc> cub: no - you can apply patches to get existing software to the same level, but you have to do an OS upgrade to get new software that's not in patches
[02:47:34] <cub> alanc: ok, so I guess the steps to do is  downloading the ISO DVD, mount it, then use "lu" (live upgrade) , specify the location of the OS update 2 , then reboot  ?
[02:47:50] <alanc> that should work
[02:48:01] <cub> cool
[02:48:02] <twincest> cub: using lu is a bit more complicated than "use lu"
[02:48:06] <twincest> cub: it's not an in-place upgrade
[02:48:35] <cub> twincest: say again ?   lu is more complicated than "use lu"  ?
[02:49:37] <gisburn> cub: question: what do you intend to do with the machine ?
[02:49:59] <cub> gisburn:  sorry can't create an account for you on it   :)
[02:50:19] <gisburn> cub: my personal suggestion is to reinstall from scratch in most scenarios unless an update is unavoidable.
[02:51:06] *** Error_404 has joined #opensolaris
[02:51:31] <gisburn> cub: usually the cleanups and checks after an "upgrade" need more work than a clean reinstall from scratch.
[02:51:46] <cub> ummm
[02:54:24] <Error_404> there we go.. server's in the other roomm
[02:54:34] <Error_404> thank god, that thing was loudass
[02:55:01] <gisburn> Error_404: sounds you never had a V880z as workstation... =:-)
[02:55:39] <twincest> anyone know when onnv-gate hg is going to be online again?
[02:55:56] <gisburn> twincest: 6.6.2666 ...
[02:56:11] <gisburn> Santans day
[02:56:18] <gisburn> s/Santans/Satans/
[02:56:19] <richlowe> twincest: nope.
[02:56:37] <twincest> i deleted my clone before i realise it was broken :-(
[02:56:49] *** fik has joined #opensolaris
[02:56:58] * gisburn chuckles and pats his subversion tree
[02:57:06] <gisburn> rock solid
[02:57:11] <gisburn> :-)
[02:57:16] *** simford has joined #opensolaris
[02:57:21] <richlowe> Yes, because you aren't trying to bridge stuff out of teamware.
[02:57:22] <twincest> there's an svn tree that updates from onnv-gate is real time?
[02:57:25] <richlowe> if you were, you'd be just as screwed.
[02:57:33] * boyd wonders about the install architecture doc being released without any work on packaging systems
[02:57:36] <richlowe> but I commend you on the attempt, at least.
[02:58:32] *** pabloocosta has left #opensolaris
[03:03:57] *** sparc-kly is now known as sparc-kly|HOME
[03:12:24] <cub> woahh...democrats take over Senate too
[03:12:57] <gisburn> cub: evil
[03:13:33] <cub> evil or not, it's America's will
[03:14:01] <gisburn> cub: can't the replugicans take over the state and make bush "King Bush The First" ? And then declare war on europe for not partaking in the war against errorism ?
[03:14:37] <cub> hah, nice wording
[03:15:11] <gisburn> cub: and the new nuclear rearming.
[03:15:16] <gisburn> cub: very neccesary.
[03:15:28] <gisburn> hydrogen bombs everywhere
[03:15:35] <gisburn> and at least one "gaja buster".
[03:15:52] <delewis> gisburn: I trust you have similar stances towards the North Korean and Iranian weapons programs?
[03:15:58] <dwc-> cub: did they get that many recounts and lawsuits settled already?
[03:16:05] <gisburn> (e.g. 200MT+ nuke, started when someone attacks the US)
[03:16:10] <delewis> dwc-: for the most part, it would seem.
[03:16:29] <cub> haven't seen any recount yet
[03:16:30] <gisburn> cobal salting optional
[03:16:42] <dwc-> last I heard, virginia wasn't certifying their results for another few weeks ....
[03:16:49] <dwc-> do they have all their absentees counted?
[03:16:53] <delewis> ugh
[03:16:57] <delewis> damn Diebold.
[03:17:20] <dwc-> not just diebold....
[03:17:53] <gisburn> delewis: the problem is that noone should use nuclear material at all. Think about it: We (germany) have enougth weapons.grate material to build 20-30 nukes ad-hoc to nuke someone lext week.
[03:18:20] *** Gman has quit IRC
[03:18:31] <gisburn> delewis: we just don't do it.
[03:18:41] *** Gman has joined #opensolaris
[03:18:44] <dwc-> in my county, the touch screen voting machines were hardly the pinnacle of usability
[03:19:14] <dwc-> and the printers which they added to add audit and accountability ran out of paper
[03:19:23] <OnkelSchorsch_> hehe. nice :)
[03:19:24] <gisburn> heh
[03:19:25] <dwc-> and that's when they were working
[03:19:44] <dwc-> the user interface that they stuck on those things was horrible
[03:19:59] <dwc-> part of the reason they ran out of paper was people would press the wrong button, and it'd invalidate the ballot, to let them start over and make corrections
[03:20:22] <dwc-> the neighborhood grocery store has better, faster thermal printers
[03:20:59] <rydis> But then the supplier of the voting machines wouldn't be the only source of compatible paper.
[03:21:01] <dwc-> and they ran out of the paper optical scan ballots waaay early in the day
[03:31:21] <alanc> I just voted absentee - less hassle, and easier to read about what you're voting on while voting instead of trying to remember at the polling place
[03:33:16] <Error_404> i didn't vote
[03:34:48] <dwc-> I should have voted absentee
[03:35:07] <dwc-> but I mark the sample ballot, so I remember
[03:35:58] <Error_404> partly because i think republicans and democrats are all the same
[03:36:13] <Error_404> but mostly because i'm not eligible... being from, and in, a foreign country
[03:36:45] <Tpenta> the way i see it, the american voting public has just sent a mesg to bush & co
[03:37:26] <gisburn> yeah
[03:37:30] <gisburn> "RETIRE"
[03:37:56] *** Somethingelse has joined #opensolaris
[03:37:59] <dwc-> well, hopefully they'll work together and actually get shit done
[03:38:06] <Tpenta> and the quote from the outgoing rumsfeld "The American public does not understand the Iraq war", .. hang on, wasn't that part of *his* job as secdef?
[03:38:09] <dwc-> instead of partisan fighting .... but I'm not going to hold my breath
[03:38:28] <dwc-> nah, that's the job of white house PR
[03:39:44] <alanc> seems weird to see VALinux copyright notices in Solaris kernel sources...
[03:40:03] *** yongsun_ has joined #opensolaris
[03:42:08] *** bondolo has quit IRC
[03:42:20] <gisburn> alanc: DRI ?
[03:42:26] <alanc> yes
[03:42:39] <alanc> http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/uts/common/io/drm/
[03:42:52] <gisburn> nice
[03:43:01] <gisburn> digital rights management
[03:43:04] <gisburn> =:-)
[03:43:19] <gisburn> s/rights/restriction/
[03:43:45] <alanc> almost thought you escaped the FSF brainwashing on the evils of DRM
[03:44:12] <Error_404> not direct rendering manager?
[03:44:19] <gisburn> no
[03:44:48] <gisburn> Error_404: soon ZFS will send the md5 hashes of your files to RIIA for evaluation whether you are a conformant US citizen.
[03:44:56] <gisburn> or a media terrorist and thief
[03:45:01] <alanc> the DRM I pointed to is Direct Rendering, gisburns just having his fun
[03:45:28] <Gman> beats reinventing the wheel though :)
[03:45:47] <gisburn> alanc: based in the current discussons here in germany we may get something like that soon... ;-(
[03:46:41] <OnkelSchorsch_> hopefully we can use zfs-crypto in the near future
[03:47:48] <alanc> I don't understand why ON made them put a Sun copyright notice on top of all the DRI files though
[03:48:13] <gisburn> alanc: don't ask me what I had to add to all the ksh93 files.
[03:48:25] <gisburn> it's insane
[03:48:33] <gisburn> (sort of)
[03:49:05] <alanc> I told them I'd reject their changes from X if they even thought about cstyling them
[03:49:25] <gisburn> oh
[03:49:45] <gisburn> alanc: you have the option to reject the attempts to cstyle everything ?
[03:50:02] <alanc> last thing I want to have to do is page through hundreds of lines of useless whitespace changes when merging in new community versions
[03:50:22] * gisburn , too.
[03:50:36] <gisburn> just 1800 files which would need cstyle'ed.
[03:50:39] <gisburn> in my case.
[03:51:18] <alanc> in the lands outside ON, changing style of code we plan to keep in sync with an external community is forbidden to prevent insanity among those who have to do the merges
[03:51:23] <alanc> it's simple common sense
[03:51:46] <alanc> (but even before we did regular merges, cstyle just wasn't our thing in X)
[03:52:50] <gisburn> heh
[03:53:10] <gisburn> 3GHz AMD64 with graphical installation still installing B51
[03:53:31] <gisburn> Intel P4 2.00GHz laptop with VMware and console installer finished installation 40mins ago.
[03:56:03] <gisburn> is the graphical installer THAT bad ?!
[03:56:06] <gisburn> ;-((
[03:56:28] *** vmhobbes has quit IRC
[03:56:31] <Gman> yes
[03:56:41] <gisburn> why ?
[03:56:59] <Gman> ask dminer
[03:57:04] *** vmhobbes has joined #OpenSolaris
[03:57:09] <alanc> there's a reason it only starts the gui intaller if you only have 512mb or more of ram
[03:57:28] <gisburn> umpf
[03:57:33] <gisburn> the AMD64 machine has 2GB
[03:57:40] <gdamore> its called _gnome_. :-(
[03:57:58] <Gman> the gui installer isn't gnome
[03:58:06] <gisburn> java
[03:58:17] *** dclarke has joined #opensolaris
[03:58:26] <dclarke> hello
[03:58:32] <dclarke> did someone break useradd ?
[03:58:37] <gdamore> but most of the memory required is to support gnome. :-)
[03:58:50] <gisburn> dclarke: which version ?
[03:58:53] <dclarke> ld.so.1: useradd: fatal: libc.so.1: version `SUNW_1.22.2' not found (required by file /usr/sbin/useradd)
[03:59:02] <dclarke> Solaris 10 Update 2
[03:59:08] <dclarke> Solaris 10 6/06 s10s_u2wos_09a SPARC
[03:59:11] <gisburn> dclarke: which nevada version ?
[03:59:20] <dclarke> nah .. this is production gear
[03:59:31] <dclarke> I applied some patches .. rebooted
[03:59:32] <gdamore> that's bad.  :-/
[03:59:37] <dclarke> poof .. useradd is broken
[04:00:14] <dclarke> I can manually edit /etc/passwd etc but .. something not good took place here
[04:01:31] * gdamore suddenly regrets his boneheaded decision to try to use a Cobalt as his NIS master server...
[04:01:34] *** wangcity has joined #opensolaris
[04:01:51] <gdamore> particularly, a Cobalt running NetBSD....
[04:02:01] <alanc> TX changes broke several patches - looks like you need the libc patch with TX support
[04:02:38] <dclarke> TX ?
[04:02:45] <gisburn> trusted solaris 10
[04:02:48] <alanc> Trusted Extensions
[04:02:49] <dclarke> oh
[04:03:12] <dclarke> http://www.blastwave.org/dclarke/stuff/useradd.txt  << -- that looks familiar ?
[04:03:55] <wangcity> hi
[04:04:05] <wangcity> i just install sun os 10
[04:04:14] <dclarke> good job ..
[04:04:19] <dclarke> hows that going ?
[04:04:23] <wangcity> but i can not use msn--gaim
[04:04:34] <dclarke> where did you get gaim ?
[04:04:56] <wangcity> someone says must download a new gaim's msn plugin, where can download?
[04:05:01] <Gman> wangcity, you need to get a patch
[04:05:05] <Gman> which may or may not have been released yet
[04:05:13] <Gman> the msn protocol broke again
[04:05:23] <dclarke> what rev of gaim do you have there ?
[04:05:36] <Tpenta> gman: CR 6491606
[04:05:53] <gdamore> any libusb experts around?
[04:06:27] * dclarke looks around
[04:06:32] * dclarke looks under desk
[04:06:49] <gdamore> libusb does not expose the upstream port path.  that's unfortunate, because i'd like to have that info (portably if possible)
[04:07:07] <Gman> Tpenta, thanks - i've added erwann as RE, though he's away until early dec
[04:07:14] <Tpenta> :)
[04:07:15] <gdamore> its the best way i can think of to encode a unique key representing the device.  which i need for Sun Ray protocol. :-)
[04:07:18] <wangcity> the gaim just with the solaris10, i now know which ver it is.
[04:07:24] <dclarke> alanc : think I found the issue here : X11 6.6.2: X11 Trusted Extensions patch
[04:07:29] <Tpenta> do i have sufficient to work from in there? shoudl be reproducible
[04:07:31] <dclarke> SunOS 5.10: Trusted Solaris Attributes Patch
[04:07:44] <dclarke> those guys .. I'll back them out
[04:07:45] <Gman> Tpenta, yeah, erwann's the guy who wrote the blueprint and nimbus code
[04:07:53] <Gman> Tpenta, so he's the only one that will look at it
[04:08:00] <Tpenta> my guess is a couple of files are missing
[04:08:03] <wangcity> and how to let gaim work?
[04:08:13] <alanc> dclarke: nope
[04:08:21] <Gman> Tpenta, yeah
[04:08:22] <alanc> X11 won't touch libc at all
[04:08:29] <gdamore> wangcity: maybe grab the latest from blastwave?
[04:08:30] <Gman> wangcity, you need to install a patch from sun
[04:08:35] <Gman> or compile your own
[04:08:42] <alanc> the SunOS 5.10 one might
[04:08:46] * dclarke nudges Gman
[04:08:53] <gdamore> or switch to a nicer protocol like IRC. :-)
[04:08:56] <Gman> dclarke, hrm?
[04:09:02] <dclarke> wangcity: or download the latest from Blastwave
[04:10:27] <wangcity> how to download?
[04:10:53] <dclarke> http://www.blastwave.org/dclarke/gaim.png
[04:11:11] <dclarke> wangcity : you need to install pkg-get and then just pkg-get -i gaim
[04:11:30] <dclarke> wangcity : see http://www.blastwave.org/howto.html
[04:12:03] <dclarke> and just ask questions ..
[04:12:23] <dclarke> ld.so.1: useradd: fatal: libc.so.1: version `SUNW_1.22.2' not found (required by file /usr/sbin/useradd)
[04:12:34] <dclarke> I have to fix this .. some patch did this to me I am sure
[04:19:19] <dclarke> oh to heck with it ... I'll reinstall Sol10U2
[04:20:45] <gisburn> dclarke: this is why I avoid "upgrades" ... =:-)
[04:20:56] <dclarke> this is a fresh install anyways
[04:21:16] <dclarke> and I applied a boatload of patches AFTER the recommened cluster patch
[04:21:23] <dclarke> I have jumpstart
[04:21:31] <dclarke> a fres hOS will be in .. in a flash
[04:21:49] <boyd> richlowe: ping
[04:26:23] *** johnlev has quit IRC
[04:31:00] *** fik has quit IRC
[04:31:39] *** Somethingelse has quit IRC
[04:33:33] *** dclarke has quit IRC
[04:48:25] *** cub has quit IRC
[04:49:00] <richlowe> boyd: yes?
[04:49:47] <boyd> Just wondering how you go with using hg mq on the sx-shipping HG. I'm getting errors that seem to indicate it wants gnu patch
[04:50:06] <boyd> (which I can use, but I don't know how to tell hg to use it)
[04:50:29] <richlowe> I don't use the sx-shipping hg, other than for a while because I hadn't noticed it was happening.
[04:50:56] <boyd> Hmm.. me either untill now..
[04:51:08] <boyd> Oh well, thx
[04:52:47] <Gman> useful ;)
[04:52:58] <richlowe> Gman: apparently 0.9.1 fails tests on Nevada.
[04:53:03] <richlowe> my experience was that 0.9 did too...
[04:53:15] <richlowe> but either way, 0.9 is in Nevada, 0.9.1 is latest.
[04:57:00] *** jamesd has quit IRC
[04:57:16] *** jamesd has joined #opensolaris
[04:57:17] *** ChanServ sets mode: +o jamesd
[04:58:41] *** Xh4 has joined #opensolaris
[05:00:51] *** alanc is now known as alanc-away
[05:06:57] <gdamore> is the libusb solaris source posted anywhere on-line?  src.os.o doesn't seem to have the sfw consolidation
[05:07:16] *** OnkelSchorsch_ has quit IRC
[05:08:40] <richlowe> Hm, it used to.
[05:09:24] <gdamore> it gets an Error 404 now.
[05:09:30] <richlowe> there's tarballs on dlc.sun.com, either way.
[05:10:22] <gdamore> yikes, all 287MB.
[05:16:03] *** axisys has joined #opensolaris
[05:26:32] *** ndroux has quit IRC
[05:28:03] *** solarisjon has joined #opensolaris
[05:28:13] *** solarisjon has left #opensolaris
[05:30:54] *** Gman is now known as GmanAFK
[05:37:43] *** Kronuz has joined #opensolaris
[05:45:08] <gisburn> gdamore: ping!
[05:45:17] <gdamore> ack!
[05:45:33] <gisburn> gdamore: any news about the E450 ?
[05:45:40] <gdamore> nope.
[05:45:51] <gisburn> crap
[05:45:56] <gisburn> gdamore: ;-(((((((((((((((((((((((
[05:46:00] <gdamore> i'm still having trouble nailing down mgmt long enough to get a signed SCA.
[05:46:26] <gisburn> SCA=?
[05:46:43] <gisburn> Serve Crocodile Attack ?
[05:46:50] *** MikeE has joined #opensolaris
[05:46:59] <gisburn> Stealth Crocodile Attack ?
[05:47:18] <gisburn> *plock*
[05:47:25] <gisburn> Sun Contributor Agreement
[05:47:29] <gisburn> right ?
[05:47:33] <gdamore> right.
[05:47:54] <gdamore> sorry, i'm trying to figure out how to netinstall ubuntu (linux, ugh.) right now.
[05:48:04] <Error_404> society for creative anachronisms?
[05:48:07] <gisburn> on sparc ?
[05:48:23] <gdamore> no, on a low end VIA laptop that has no CDROM.
[05:48:25] <gisburn> Error_404: no, SPW
[05:48:32] <Error_404> ?
[05:48:33] <gisburn> Sucking Paper Work
[05:49:02] <gisburn> SPW=Sucking Paper Work
[05:49:11] <Error_404> i thought you had already gotten your contributor number?
[05:49:18] <gisburn> yes
[05:49:26] <gdamore> I have one for "personal" use, but not for Tadpole contributions
[05:49:29] <gisburn> But we need another ARC case
[05:49:40] <MikeE> I just got myself thouroughly confused with zfs delegation and local zones... anyone got a good grasp on that one?
[05:50:59] <axisys> anyone here 3510 expert? i have a T2000 with a dual port fc hba and my 3510 have 6 host ports and 2 drive ports.. i have 12 146G disks .. how do i connect the damn think?
[05:51:00] *** karrotx has joined #opensolaris
[05:51:03] <axisys> thing*
[05:51:15] <gisburn> axisys: which OS do you run ?
[05:51:20] <axisys> sol 10 u2
[05:51:35] <gisburn> axisys: could you install B51 for one week, please ?
[05:51:40] <gisburn> axisys: pleassssee...
[05:51:44] * gisburn purrs
[05:52:19] <axisys> gisburn: hehe.. sorry man i already have the OS setup on t2000 and i have to connect and test and ship it out cob tomorrow
[05:52:22] <axisys> :-(
[05:52:43] <MikeE> axisys, you end up going with PCI-X hbas or PCI-E on that bad-boy?
[05:52:45] <jmcp> axisys: have you RTFMd the se3510 documentation?
[05:52:50] <gisburn> axisys: you don't have a spare sparc around, right ?
[05:53:11] * gdamore thinks sun should just stick a T2000 out with the latest SXCR and hand out shell accounts to vetted os.o members.
[05:53:23] <axisys> jmcp: i am reading it .. but so far none of the pages talk of my setup..
[05:53:40] <MikeE> cxonet anyone? (last I checked it was on a t2000)
[05:53:41] <jmcp> axisys: perhaps you should think more generally
[05:53:46] <richlowe> gdamore: I think it'd be loaded up within minutes.
[05:53:51] <Error_404> gdamore: registered contributors
[05:53:52] <Error_404> *nod*
[05:54:01] <Error_404> there's only 51 of us...
[05:54:01] <richlowe> I'm hoping whatever they do build/test farm wise would be well thought out.
[05:54:10] <axisys> MikeE: pci e and pci x both i think.. i have console access to the t2000 now
[05:54:22] <gisburn> richlowe: once I wrote a proposal but noone cared.
[05:54:29] <richlowe> Error_404: well, with a decent amount of FSS maybe.
[05:54:33] <gdamore> richlowe: Solaris has good resource management right?  this would be kind of a proof of usability. :-)
[05:54:34] <MikeE> mixing both PCI-e and PCI-x HBAs? interesting choice...
[05:54:37] <gisburn> richlowe: basically I could setup a buildfarm witzin two days
[05:54:41] <axisys> jmcp: well check this out.. i am reading the 816-7300-20.pdf 3510 user guide)
[05:54:42] * richlowe tends to set DMAKE_MAX_JOBS high enough to make the machine regret existing
[05:54:43] <gisburn> richlowe: but I need machines...
[05:54:43] <Error_404> fss?
[05:54:56] <richlowe> Error_404: the fare share scheduler, everyone in their own project, everyone with equal shares.
[05:55:01] <richlowe> otherwise... :)
[05:55:01] <gisburn> richlowe: s/machine/disks/
[05:55:05] <MikeE> sadly: while(malloc(1)) in a local zone will still kill ya every time...
[05:55:17] <Error_404> that's what i thought you said, wasn't sure
[05:55:24] <MikeE> memsets are coming soon though... (and this time we mean it :-)
[05:55:25] <axisys> jmcp: but not a single pic w/ one host w/ a dual port fc hba
[05:55:42] <axisys> jmcp: this is what i have tried so far..
[05:55:46] <richlowe> the only memory intensive bit of the ON build is lint.
[05:55:51] <richlowe> though I'm not sure intensive is the right word for that.
[05:55:58] <Error_404> oh man is it ever
[05:56:01] <gisburn> richlowe: uhm... no.
[05:56:07] <richlowe> I'm not sure I have *polite* words for that, either.
[05:56:08] <Error_404> dmake lint kills my builds more or less every time it runs
[05:56:16] <gisburn> richlowe: I identified a serious bottleneck on my sparc
[05:56:18] <richlowe> Error_404: you're looking at at least 2G of VM.
[05:56:19] <gdamore> gcc 4 is a "pig".  i can't imagine that it is any smaller than lint.
[05:56:22] <axisys> jmcp: one port to host port fc0 and another to fc1 or fc5 and i get only 175G.. so that is wrong
[05:56:30] <gisburn> richlowe: running gcc and Sun Studio cc in parallel is a BAD idea.
[05:56:40] <richlowe> Error_404: it dumps 800M-ish in /tmp, and uses as much again in core.
[05:56:55] <axisys> jmcp: then i tried one port to fc0 and second one to loop B on the i/o expansion and i get all 12 146G disks
[05:56:57] <richlowe> (kernel crosschecks, everything else is reasonable)
[05:56:58] <axisys> jmcp: but..
[05:57:05] <gdamore> gisburn: s/and Sun Studio cc in parallel//
[05:57:16] <Error_404> i should just strip lint out of my makefiles
[05:57:26] <Error_404> obviously all it does is brick my machine
[05:57:29] <gdamore> ln -s /bin/true lint
[05:57:32] <axisys> jmcp: first disk is c4t4.... and the rest 11 are c4t2... .. so that is still wrong
[05:57:46] <gisburn> gdamore: I mean the gcc shadow build causes the gcc and cc processes to swap against each other.
[05:57:49] <axisys> jmcp: i want both controller to see all 12 disks
[05:58:02] <axisys> jmcp: is that possible
[05:58:07] <gdamore> yes, but if you have a choice, I would never choose to run gcc.
[05:58:16] <MikeE> axisys: so you ahve to put the thing into jbod mode?
[05:58:23] <gisburn> gdamore: the box is more busy with paging than doing real work.
[05:58:27] <gdamore> the studio compilers are so much vastly superior.  the gcc build is just a "sanity check" :-)
[05:58:47] <axisys> MikeE: so first port to loop a and second port to loop b?
[05:58:56] <gisburn> gdamore: well, gcc has sometimes the better error checking
[05:58:57] <gisburn> *sometimes*
[05:59:04] <gdamore> is that with gcc3 or gcc4?  gcc4 got a lot bigger than gcc3.
[05:59:11] <gisburn> and sometimes I am cursing it.
[05:59:18] <MikeE> axisys :: unsure... there are raid-controllers in there right?
[05:59:32] <MikeE> (the 3510 isn't just an a5000 in disguise)
[05:59:34] <axisys> axisys: well i like to see all 12 disks.. and configure thru zfs
[05:59:54] <axisys> MikeE: it does have a raid controller but i never consoled to the 3510 yet
[05:59:59] <MikeE> axisys: I hear ya, but if it aint a jbod array, you need to find a way to dumb it down.
[06:00:25] <gdamore> woo hoo.  my 2nd PSARC case was approved today. :-)  trivial, but /usr/bin/id will grow POSIX options.  and the xargs fix for -E has been sent for self review.
[06:00:26] <MikeE> typicall the raid-array sits between the server and the backend disks, and its presenting you some sort of LUN, not the direct disk.
[06:00:39] <gdamore> s/self review/automatic approval/
[06:00:46] <MikeE> axisys :: in format, what kind of label do the 3510 disk you do see have?
[06:00:47] *** gisburn has left #opensolaris
[06:01:20] <axisys> MikeE: right now i have fc0 and fc5 connected to the dual port fc hba and the format looks like this..
[06:01:53] <axisys> c4t600C0FF0000000000A77B07365312C00d0 <SUN-StorEdge3510-415F cyl 45352 alt 2 hd 64 sec 127>
[06:01:59] <MikeE> what you really "want" is a 3511 array, not a 3510...
[06:02:05] <axisys> /scsi_vhci/ssd@g600c0ff0000000000a77b07365312c00
[06:02:08] <MikeE> see that store-edge busines there....
[06:02:28] <axisys> i have really no choice but stick w/ 3510
[06:02:33] <MikeE> that is a LUN... (ie, the raid-controller is doing some sort of stripe/raid/what-have you on the back-end disks, and presenting this out.)
[06:03:17] <MikeE> the problem is that you need to tell the raid-controller to completely get out of the way (which it may not want to do), or have it present out essentially a set of lun-stripes (non-mirrored))
[06:04:11] <MikeE> problem is that now if you DO have a disk-failure, your physical disk utilization may not match what you think is going on. (since the raid-controller is remapping blocks etc. to the backend disks)
[06:04:39] <MikeE> You're trying to do ZFS jbod to a RAID/Lun (non-jbod) back-end.....
[06:04:40] <axisys> MikeE: i dont want raid controller .. i want zfs to manage them
[06:05:00] <MikeE> understood, but you've got yourself a raid array, NOT a jbod array
[06:05:42] <MikeE> (you also need to watch your firmware levels etc. on that 3510... make sure you're current)
[06:05:46] <axisys> MikeE: why do i get 12 disks if i use Loop B instead of fc5 for second port of the hba?
[06:06:23] <MikeE> maybe there is only 1 raid-controller on the a-channel or something? the b-channel is non-raid, since the controller isn't there?)
[06:06:32] <axisys> so i should just use the Loop Drives (drive port) instead of host port.. because drive ports are just expnasion units and have no controllers
[06:07:14] <axisys> i have i/o controller i/o expansion on the top and just a i/o expansion on the bottom
[06:07:34] <axisys> so i should just use i/o expansion board which has two loop drives..
[06:07:37] <jmcp> axisys: why do you want to see *disks* when you have a freakin' *raid engine* to manage them for you?
[06:07:55] <MikeE> axisys :: I'm NOT familiar with the 3510, I'm looking at some dox, and pretty much repeating what I'm reading.  3511 units are the expansion-units (jbod) for the 3510.... perhaps your hooking into some expansion port bypassing the onboard raid-controller.... you're starting to get into some very shady/grey areas
[06:08:05] <axisys> jmcp: i want to see all the disks. easy to see which one broke
[06:08:05] <jmcp> axisys: if you don't connect your host FC hbas to host ports then your config is unsupported by Sun
[06:08:16] <jmcp> axisys: wtf did you buy something with a raid engine then?
[06:08:27] <MikeE> jmcp :: thats exactly his problem. He wants aJBOD array (for zfs control) but he is stuck with the raid-engine...
[06:08:38] <jmcp> yes, so the question stands
[06:08:51] <MikeE> I'm with you.. scroll back... I've asked it too :-)
[06:08:52] <axisys> i have something similar to this http://sunsolve.sun.com/handbook_private/Systems/3510/component.rear_zoom.html
[06:09:06] <MikeE> yes, a 3510 RAID array...
[06:09:26] *** Burana has joined #opensolaris
[06:09:45] <axisys> little diff form the picture.. top board is like the top pic and bottom board is like bottom pic
[06:09:46] <MikeE> dot-hill action if memory servers...
[06:09:49] <MikeE> (serves)
[06:09:57] <jmcp> MikeE: correct
[06:10:16] <Burana> good morning!
[06:10:21] <MikeE> had some trouble with JBOD 3310's.... turns out even JBODS had firmware :-(
[06:10:22] <jmcp> axisys: if you cannot get a jbod then use the management interface on the 3510 to create lots of luns which you present out the HOST PORTS
[06:10:42] <MikeE> jmcp: I suggested this, but there is caution to be had here...
[06:10:48] <jmcp> true, true\
[06:11:13] <jmcp> MikeE: but since axisys is proving to have a lack of sense right now ....
[06:11:19] <MikeE> when you present out those "unprotected luns" you are no longer giving ZFS what it needs to ensure "Safe mirroring" as you are no longer "split" on physical disk boundries...
[06:11:20] <jmcp> axisys: yes, I think you're being an idiot
[06:11:43] <axisys> jmcp: i sure am..
[06:11:45] <Burana> does anybody know if it possible to boot the sun x64 servers from usb-stick?
[06:12:04] <axisys> jmcp: i have raid controller on the top board and a jbod on the bottom
[06:12:08] <jmcp> MikeE: and since axisys doesn't have 4 fc ports on his host to connect to the se3510 .....
[06:12:17] <Burana> I want to install them with jumpstart, but without using pxe
[06:12:18] <jmcp> axisys: just connect your host to the documented host ports
[06:12:21] <axisys> jmcp: correct
[06:12:31] <jmcp> Burana: x86/x64 requires pxe for netboot
[06:12:36] <jmcp> axisys: I know I'm correct dammit
[06:12:38] <axisys> jmcp: lets say i connect to fc0 and fc1
[06:13:13] <axisys> but then i see only two 175G disks in format..
[06:13:20] <Burana> jmpc: isn't it possible to load the solaris dvd into the server and start netboot from there?
[06:13:26] <axisys> jmcp: so console to the 3510 and create luns?
[06:13:27] <MikeE> axisys... those are NOT DISKS... they are LUNS
[06:13:53] <axisys> MikeE: why do the luns show 175G when i have 12 146G disks
[06:14:06] <jbk> because that's how they're configured in the array?
[06:14:09] <MikeE> axisys: let me be clear... those are NOT DISKS... (the 2x 175
[06:14:33] <jmcp> axisys: have you heard of this concept called "raid" ?
[06:14:48] <MikeE> look at format string... they are "raid-lun representations" where the 3510 is playing some sort of mirror/stripe/raid game with its disks to present you with a virtual hard-drive (aka LUN)
[06:14:50] <axisys> jmcp: yes
[06:14:57] <axisys> i use it all the time
[06:15:07] <jmcp> axisys: good. Now go and connect to the management interface and see what you've got configured
[06:15:08] <Kronuz> hey, jmcp, you're still around!
[06:15:09] <Kronuz> :)
[06:15:12] <jmcp> Kronuz: ssshhhh
[06:15:15] <Kronuz> :P
[06:15:22] <Kronuz> <.<
[06:15:23] <Kronuz> >.>
[06:15:58] <Burana> Kronuz: I'm arround too again... it is morning here again
[06:16:05] <Kronuz> O_O
[06:16:25] <Kronuz> Burana: I didn't get the chance to thank you... yesterday (I suppose) :)
[06:16:44] <MikeE> so axisy: there is only 1 raid-controller in that 3510?
[06:16:48] <Burana> Kronuz: have you placed your SAN order yet? :-)
[06:16:50] <axisys> MikeE: so as jmcp suggested, i can console to the 3510 and should be able to configure luns.. can i configure one lun per disk?
[06:16:55] <Kronuz> Burana: lol
[06:16:57] <Kronuz> not yet
[06:17:02] <axisys> MikeE: yes
[06:17:05] <Kronuz> I'm with the clustering issue now
[06:17:19] <axisys> MikeE: instead of two.. i have one raid controller and one jbod
[06:17:20] <Burana> Kronuz: What issue?
[06:17:27] <gnu2it2> what is the difference between a sunblade 1000 and 2000? when it boots the screen says 1000 but the front cover says 2000?
[06:17:29] <MikeE> axisy: if you configure 1 lun per disk DEPENDING if that trully maps each lun to 1 physical disk, you can come close to getting away with it.
[06:17:59] <Kronuz> Burana: trying to figure how to build one... it turns out it's something I might actually find more useful than the simple load balancing I was trying to do
[06:18:04] <MikeE> although you'll be running 1 side through jbod, and 1 side through "1-disk-LUNS".... It will probably work, but I doubt supported
[06:18:19] <Kronuz> since what I wanted that for was availability
[06:18:30] <MikeE> actually come to think of it there is a good change it wont work.
[06:18:31] <Kronuz> (and also performance)
[06:18:38] *** simford has quit IRC
[06:19:25] <axisys> MikeE: so should i create two luns ?
[06:19:28] <MikeE> the LUN-side and the "straight-up" jbod side might point at the same physical disk, but the LUN "enveloppe" around the disk when you go 1 path, and the lack thereof on the JBOD path is a recipe for disaster...
[06:19:36] <axisys> 6 on each
[06:19:38] <Burana> kronuz: IMHO, clustering is nice, but sometimes makes it more complicated. do you have any business requirements for availability
[06:20:06] <axisys> MikeE: ok so lets stay w/ raid controller
[06:20:07] <Kronuz> well, yes, it should be up 24/7
[06:20:26] <jbk> Kronuz: what are you trying to do?
[06:20:33] <Kronuz> Burana: that's the problem ... and I was thinking aof just using some load balancing techniques for that too
[06:20:36] <Burana> kronuz: that what all business people want... but it comes with a price tag :-)
[06:20:40] *** simford has joined #opensolaris
[06:20:51] <Kronuz> yeah... I'm seeing
[06:20:54] <MikeE> axisys : if you connect through the raid-ctrl with both paths, and create 12 1-disk luns, you should be able to have ZFS "do its thing"
[06:21:10] <Burana> kronuz: and when the business people see the price tag thei are ok with 23/7 :-)
[06:21:23] <Kronuz> I'm just trying to figure if our budget will make it too
[06:21:35] <MikeE> axisy :: but if that raid-controller goes... buh bye...  (make sure the firmware for that thing is uptodate)
[06:21:36] <axisys> MikeE: bingo! that is what I want
[06:21:39] <Burana> kronuz: what apps are you runninng?
[06:22:02] <Kronuz> we'll be running PostgreSQL, apache and mail server (basically)
[06:22:24] <axisys> MikeE: so the dual port hba to fc0 and fc1 should give me 12 or 24 disks?
[06:22:35] <axisys> luns rather
[06:22:56] <Error_404> so apparantly someone pre-drilled a hole from the back closet all the way to the other side of my apartment, to where my living room is
[06:22:58] <MikeE> axisys :: well, the other way you can do it, and I think jcmp mentioned: create 2 "channels/host-connections" once for each HBA-port. put 6x 1-disk luns on each
[06:23:10] <Error_404> perfect for running cat5
[06:23:13] <Burana> kronuz: do you have requirements to have those serveres physically separated?
[06:23:17] <jbk> Kronuz: and just want to cluster it on two boxes?
[06:23:55] <MikeE> axisys :: part of that (6/12/24) would have todo with multipathing ontop of this whole mess, and I really don't know how far you want to go with that, especially since you still have only a since (spof) raid-controller
[06:23:58] <Kronuz> jbk: well, we're planning on starting with two or three nodes and two firewall servers
[06:24:11] <Kronuz> ('cause the budget)
[06:24:20] <Kronuz> (that and the SAN)
[06:25:07] <MikeE> axisys: simplest way: create 2 host-ports. put 6x 1-disk luns on one, 6x 1-disk luns on the other. You should see a total of 12 disks in format ... 6x on c3.... 6x on c4..... then have zfs mirror between & hoep for the best....
[06:25:21] <Kronuz> jbk: and if it goes as expected we'll be getting more nodes soon, that's why I'm thinking about blade servers
[06:26:21] <axisys> MikeE: (taking notes)
[06:26:22] <Burana> kronuz: Somehow I don't like blade servers...always too small/ too many
[06:26:51] <Kronuz> jmcp: I have been trying to figure out how many switches I would need (I read most of the docs you told me about... I'll have to wait for the book tho')
[06:26:54] <jbk> well you have to be sure you have the right power/cooling
[06:27:10] <jmcp> Kronuz: how many nodes will you have?
[06:27:13] <MikeE> burana // have a look at Sun's.... its a little differnt from some of the others
[06:27:14] <Kronuz> jmcp: but I was wondering if one really need the extra "channel" for hearbeats
[06:27:21] <Burana> kronuz: having load balancing for mail and postgreSQL could be the difficult thing
[06:27:22] <jmcp> Kronuz: without a hearbeat, you ain't got a cluster
[06:27:29] <Kronuz> jmcp: probably three for start
[06:28:00] <Kronuz> jmcp: but couldn't the heartbeat be kept using the same NICs for regular networking?
[06:28:33] <Burana> MikeE: Thanks, but I prefere bigger boxes with zones :-)
[06:28:36] <MikeE> Kronuz :: depends on your clustering software... VCS doesn't like that much.. (public-hbeats can get you in interesting trouble with LLT)
[06:28:41] <axisys> MikeE: can i configure host port 0 (fc0) with all 12 luns ( 1 disk per lun ) and configure host port 1 the same way
[06:28:50] <axisys> MikeE: and then use multipath
[06:28:53] <jbk> true, it sounds like he wants to do something similar to oracle rac, but with postgresql, which i didn't think could be done..
[06:28:54] <Kronuz> MikeE: oh, I see
[06:28:57] <MikeE> Burana.. got 4x 25ks coming in next month.,. I hear ya brother :-)
[06:28:59] <jmcp> Kronuz: no. you need separate hearbeat nics
[06:29:12] *** laca has quit IRC
[06:29:19] <Kronuz> I suppose I could use Myrinet then?
[06:29:24] <jmcp> Kronuz: as I recall, the SunCluster installer will not proceed if it detects that you are trying to use public net interfaces for hearbeat
[06:29:32] <jmcp> Kronuz: how huge is your budget?
[06:29:33] <Burana> MikeE: You lucky guy!
[06:29:38] <Kronuz> all the data would be on the SAN... so I'm not sure how databases would sync ... would the data go thru the servers or directly from the SAN to the SAN ?
[06:29:40] <MikeE> axisys :: yes, you could do it that way too... but you need to be familiar with multipathing/leadville, and I have NO IDEA how well integrated the 3510 is into thats space...
[06:29:49] <MikeE> (anyone remember RDAC drivers... barf!!!)
[06:29:54] <Burana> I have to leave now for work! Bye!
[06:29:58] <Kronuz> jmcp: it isn't huge, that's the problem too... :S
[06:30:03] <jmcp> MikeE: I fixed a few bugs in those p.o.s.
[06:30:11] <Kronuz> Burana: bye, have a nice day at work :-)
[06:30:11] *** Burana has left #opensolaris
[06:30:16] <axisys> i know sol 10 u2 offer multipath
[06:30:22] *** Burana has joined #opensolaris
[06:30:37] <jmcp> MikeE: the 3510 is well supported under leadville. As long as you've got the SAN 4.4.10 patches installed
[06:30:38] <Error_404> postgres itself has HA functionality, does it not?
[06:30:39] <axisys> all i do is stmsboot -e (i may not even have to on 10 u2)
[06:30:49] <MikeE> axisys... yes, even U1 has multipathing... works GREAT with EMC symmetrix once you set it up...
[06:31:02] *** deather_ has joined #opensolaris
[06:31:12] <jmcp> multipathing is different to fibrechannel support, let's get that straight
[06:31:17] <jbk> i'm guessing for postgres, it's probably an active/passice type deal
[06:31:26] <MikeE> axisys : you have to tell it how to identify symmetric multipathing devives etc. etc. (at least on SYMM/SHARK) it is unclear how the 3510 plays in this space...
[06:31:51] <MikeE> jmcp ?
[06:32:00] <axisys> axisys: hmm now that i got some ideas let me go back to the 300 page doc of 3510
[06:32:08] <Kronuz> are Myrinet cards and switches expensive?
[06:32:12] <MikeE> axisys: yes you can do MPXIO multipathing..... its all about how you MAKE THAT   WORK on the 3510
[06:32:18] <Kronuz> I mean expensive as FC...
[06:32:19] <Kronuz> ?
[06:32:44] <MikeE> supposedly cheaper than 10ge, at least thats what the IB sales-pukes are saying... I haven't checked the pricing...
[06:32:53] <axisys> MikeE: the way i turned it on as you could see in my format is by enabling it using stmsboot -e
[06:33:02] <jmcp> MikeE: you don't need FC to do multipathing. You can do it with scsi too
[06:33:24] <jmcp> Kronuz: that's my understanding since they're relatively low volume and thus kinda "custom" hardware
[06:33:26] <axisys> c4t600C0FF0000000000A77B07365312C00d0 <-- this tells u multipath on
[06:33:31] <MikeE> jmcp // don't you go taking me down the multi-initiator rathole :-)
[06:33:44] <Kronuz> jmcp: from what you've said about hearbeats not being public I would guess those channels would need their own separate switches
[06:33:48] <MikeE> axisys // then the 3510 appears to work right of of the box. lucky you :-)
[06:34:07] <jmcp> MikeE: muahahahaha
[06:34:13] <jmcp> Kronuz: yes
[06:34:17] <axisys> MikeE: i think i read that somewhere in the doc and then turned it on
[06:34:21] <jmcp> Kronuz: or at the very least, their own vlans
[06:34:22] <MikeE> evil evil RDAC man!
[06:34:29] <Kronuz> so I guess having three or more nodes in a redundant configuration would require four NICs per node and four switches total
[06:34:44] <jmcp> Kronuz: yeah, that's about right
[06:34:50] <axisys> jmcp: let me go rtfm before jmcp calls me stupid again
[06:34:58] <MikeE> axisys // for EMC arrays and others you have to add some stinkin' vendor ID tag to some file in /kernel/drv somewhere... looks like it figures it out for itself on the 3510...
[06:34:59] <jmcp> axisys: if you're using s10u2 you *definitely* want to get those patches installed.
[06:35:25] <MikeE> if you go that route and present all 12 devices out of each port, you should get close to where you want to be... (assuming that riad-controller is solid)
[06:35:29] <jmcp> MikeE: when I was inside Sun we did a *lot* of work to make sure that Sun-sold hardware "just worked" with Solaris
[06:35:29] <axisys> jmcp: the san patches?
[06:35:51] <jmcp> axisys: yeah, for fcp/fp/fctl/fcsm, qlc+emlxs, scsi_vhci (mpxio driver)
[06:35:54] <axisys> scsi_vhci
[06:36:08] <Kronuz> hmm.. I'm not sure the IBM chassis has that many wholes as to put two FC switches and four ethernet, or two FC seitches, two ethernet and two myrinet ...
[06:36:25] <MikeE> RDAC BAD! BAD RDAC.... RDAC don't play nice with VxDMP... BAD RDAC...
[06:36:27] <Kronuz> I would probably have to have those in separate units
[06:36:33] *** PosixC has joined #opensolaris
[06:36:34] *** PosixCompliant has joined #opensolaris
[06:36:40] *** PosixCompliant has left #opensolaris
[06:36:41] <Error_404> switches are for suckers... real men use crossover cables
[06:36:54] <Kronuz> jmcp: hay, is SunCluster also free?
[06:36:55] <Error_404> hell, real men use paperclips chained together, twisted to ethernet pins
[06:37:00] <twincest> kronuz: no
[06:37:02] <MikeE> jmcp problem was that the leadvill stack was always a bolt-on.... (until 10) it was never entirely integrated and that cause a lot of pain...
[06:37:12] <jmcp> Kronuz: at least for development, I think so. Check the licensing. But it's part of the JES stack
[06:37:17] <jmcp> MikeE: it was integrated with s10
[06:37:25] <jmcp> yes, a heckuvalot of pain
[06:37:31] <MikeE> jmcp :: yup, and thats why we're using it now.
[06:37:32] <jbk> i think it's the 'free for use, but if you want support, you have to buy a license'
[06:37:37] <g4lt-U60> Error_404, real stupid men ;P
[06:38:21] <Error_404> it'd work
[06:38:22] <Error_404> maybe
[06:38:48] *** Burana has quit IRC
[06:39:13] <MikeE> ooh well.. so much for my delegated zfs filesystem (ZONES) business :-(
[06:39:38] <Kronuz> what's the "JES stack" ?
[06:39:48] <jbk> java enterprise system
[06:39:51] <MikeE> I'm going to try some more unnatural acts with zfs and zones... 'nite all
[06:39:58] <Kronuz> oh :P
[06:40:08] <jbk> basically sun's web server, app server, mail server, ldap server, cluster server, and probably a few other things
[06:40:09] <Kronuz> g'night MikeE
[06:41:08] *** MikeE has quit IRC
[06:42:35] <Kronuz> hey, who's used 10Gigabit ethernet?
[06:43:02] <Kronuz> (just wondering :P)
[06:43:08] * jmcp departs
[06:43:17] <Kronuz> bye jmcp
[06:43:22] <Kronuz> thanks again :)
[06:43:29] <twincest> we've thought about 10GE but there's no need to use it yet
[06:43:42] <jbk> heh.. we're still mostly 100mb ethernet :)
[06:44:06] <Kronuz> twincest: I suppose it's hard to get to the point when you need 10GE
[06:44:07] <jbk> and our network engineers decided that where we needed gigabit, we should use fiber instead of copper :)
[06:44:23] <Kronuz> hmm
[06:44:34] <twincest> jbk: er, you can get 1GE fibre.  we use a lot of that
[06:44:36] <Kronuz> is fiber too different from copper?
[06:44:41] <jbk> well
[06:44:50] <Error_404> Kronuz: yeah, about $20 per foot different
[06:44:51] <jbk> most everything somes with built in gige copper ports
[06:44:56] <Kronuz> lol
[06:45:15] <jbk> so if you have to go fiber, you end up wasting perhipheral slots
[06:45:20] <jbk> just for your network
[06:45:39] <twincest> that's really not true of everything
[06:45:46] <jbk> as well as having to buy additional cards
[06:45:53] <jbk> if you're talking sun servers it is :)
[06:46:03] *** |tsoome| has quit IRC
[06:46:26] <jbk> i think the v880/v890 is the only thing with built in network ports that doesn't have gige copper
[06:46:27] <twincest> oh, you want to use 10GE for servers?
[06:46:49] <Kronuz> jbk: 'cause now that we're still buying the stuff, we might want optical if it really makes a difference
[06:46:58] <Kronuz> instead of having to change it later
[06:47:11] *** deather has quit IRC
[07:09:26] <axisys> i am seeing this in the console of a x2100 http://www.rafb.net/paste/results/QoqnVZ30.html.. can anyone explain what this means?
[07:10:36] <Error_404> it means there's no file at that location
[07:12:11] <axisys> Error_404: hmmm.. i do have a disk that i built and tested here on a x2100 and then just shipped it to the site
[07:12:18] <Error_404> oh, n/m
[07:12:25] <Error_404> i was referring to the 404 error
[07:13:09] <axisys> 404? hmm.. here it the link again..  http://www.rafb.net/paste/results/QoqnVZ30.html
[07:13:20] <axisys> sorry about the dots *sigh*
[07:13:34] *** bougie has joined #opensolaris
[07:16:02] <Error_404> tried just getting in to the bios, hitting "save changes" & rebooting?
[07:16:03] *** nate_a has quit IRC
[07:16:55] <axisys> i missed the bios window.. ok i guess i will call the site and have someone to reboot the box
[07:17:09] <axisys> tomorrow that is.. thnx Error_404
[07:17:28] <axisys> Error_404: luckily its not in production yet
[07:17:56] <Error_404> yeah, seems a pretty stupid error to be getting
[07:24:08] <axisys> Error_404: agreed
[07:29:35] *** Kronuz has left #opensolaris
[07:34:23] *** lloy0076 has joined #opensolaris
[07:37:20] <lloy0076> Am I able to perform a Live Upgrade using the installer from a ON_51 consolidation on my system which was ON_46 but is now bfu'ed to ON_50?
[07:41:11] *** OnkelSchorsch has joined #opensolaris
[07:42:59] <twincest>   + 75 linda.bernal at sun dot com           - CR 6490848 Updated P5 utility/text awk, nawk do not display system errors
[07:43:09] <twincest> i'd love to know what was updated, but the main contains no actual information
[07:43:18] <twincest> so useful i received it :)
[07:43:24] *** lloy0076 has left #opensolaris
[07:52:17] *** bougie has quit IRC
[07:58:15] *** gnu2it2 has quit IRC
[07:59:26] <Cyl> so anyone in here play with jumbo frames on x86 yet? :D
[07:59:51] <noyb> timeless: ping
[08:01:03] *** Burana has joined #opensolaris
[08:05:26] <richlowe> twincest: gimme sec, and I'll give you a CR ID.
[08:05:57] <twincest> i have the id...
[08:06:26] <richlowe> twincest: nah, a different one.
[08:09:16] *** jafari has joined #opensolaris
[08:12:01] <Error_404> the mystical sun interior CRID where you can't access it from outside sun's campuses
[08:12:29] <richlowe> Nah, there's a CR related to the uselessness of what twincest was talking about.
[08:12:34] <richlowe> unfortunately, it's no longer visible.
[08:12:37] <Error_404> like the utterly useless problem resolutions that sometimes come up
[08:12:46] <dlg> can i sleep in a drivers attach routine?
[08:13:06] <Error_404> "visit galaxy.foobar to get the resolution binaries" great... good thing i work for sun
[08:13:09] <Error_404> ;)
[08:13:18] <richlowe> Error_404: the heads-ups do that too.
[08:13:24] <richlowe> at least when someone in here does it, you can berate them.
[08:13:42] <Error_404> heh
[08:14:15] <Error_404> "stevel, what's the matter with you? "
[08:14:29] <richlowe> twincest: if it's not obvious from the fields what happened, easiest way is to ask the person that changed it.
[08:14:38] <richlowe> twincest: you'll notice they made an attempt at hiding that too, but screwed it up :)
[08:17:22] *** lloy0076 has joined #opensolaris
[08:17:39] *** lloy0076 has left #opensolaris
[08:22:04] *** Risky_ has joined #opensolaris
[08:27:41] <PosixC> I am trying "pkg-get install wireshark" and getting ...
[08:27:42] <PosixC> ==> TYPE I ... done.  ==> CWD /pub/freeware/i386/5.11 ...
[08:27:43] <PosixC> No such directory `pub/freeware/i386/5.11'.
[08:27:43] <PosixC> ERROR: could not get catalog fil
[08:27:54] <PosixC> s/fil/file
[08:28:01] <PosixC> any idea what can it be?
[08:28:14] *** Risky has quit IRC
[08:28:23] <PosixC> the same is also with pkg-get install ethereal
[08:28:41] <PosixC> on x386 machine
[08:30:05] <g4lt-U60> ah yes, blastwave's insistence on making arbitrary distinctions between solaris releases even though it's the same binary for sol 8, 9, 10, and 11
[08:31:22] <PosixC> g4lt-U60, so what to do ?
[08:31:53] <PosixC> is there a way to download it from their site?
[08:31:59] <PosixC> could not find
[08:32:04] <PosixC> in
[08:32:15] <PosixC> http://www.blastwave.org/packages.php
[08:32:20] <PosixC> an idea?
[08:32:51] <twincest> funny, the bug report mails have actually got *worse*
[08:35:44] <richlowe> The intial notification should be complete.
[08:35:54] <richlowe> updates will be awful.
[08:36:12] <richlowe> if they did make it worse, that's a bug too, file it.
[08:36:40] <twincest> it's not complete, it only has the synopsis
[08:37:39] *** |tsoome| has joined #opensolaris
[08:38:49] <twincest> oh well, at least i can submit fixes now, or at least when onnv-gate comes bac
[08:40:08] *** jafari has quit IRC
[08:44:09] *** Cyrille has joined #opensolaris
[08:45:55] *** Risky_ has quit IRC
[08:46:44] *** Risky has joined #opensolaris
[08:47:37] *** lloy0076 has joined #opensolaris
[08:47:55] <lloy0076> Is Adobe Flash Plugin 7.0 the latest version of the flash plugin available?
[08:50:26] *** lloy0076 has left #opensolaris
[08:53:34] *** slowhog has left #opensolaris
[08:56:01] *** bobbyz has quit IRC
[08:56:05] *** bobbyz has joined #opensolaris
[09:03:03] <OnkelSchorsch> posixc: choose a different mirror
[09:03:33] <PosixC> OnkelSchorsch, looking for mirrors
[09:04:02] <PosixC> OnkelSchorsch, I was a bit desperate .... trying to build wireshark ...
[09:04:08] <PosixC> which needed libpcap...
[09:04:13] <PosixC> which needed flex...
[09:04:19] <PosixC> which needed m4
[09:04:44] <PosixC> and making flex failed ...
[09:04:51] <Error_404> why do they bother publishing these results? "the Conservatives at 33 percent in November compares with the election result of 36 percent for the Conservatives..... margin of error of 2.2 percentage"
[09:04:52] <OnkelSchorsch> sucks :(
[09:05:15] *** Burana has quit IRC
[09:05:22] <Error_404> so... it may have been a .8% drop... that's nothing
[09:05:29] <PosixC> OnkelSchorsch, I see many mirrors in http://www.blastwave.org/mirrors.php
[09:05:36] <OnkelSchorsch> PosixC, I'm using ftp://ftp.uni-erlangen.de/mirrors/blastwave.org/unstable
[09:05:43] <OnkelSchorsch> but I'm from germany...
[09:05:46] <PosixC> But How can I tell pkg-get install to use them ?
[09:06:02] <PosixC> did not have such problems before with pkg-get
[09:06:13] <PosixC> all installation with it worked
[09:06:20] <OnkelSchorsch> edit /opt/csw/etc/pkg-get.conf
[09:06:59] <PosixC> OnkelSchorsch, thnks
[09:07:12] <OnkelSchorsch> no problem :D
[09:07:21] <OnkelSchorsch> hopefully this safes you some time
[09:07:30] <PosixC> OnkelSchorsch, and change the url line , I assume ?
[09:07:59] *** ThEbUtChE has joined #opensolaris
[09:08:17] <OnkelSchorsch> just add one and comment the other one out.
[09:08:22] <OnkelSchorsch> that should do it :)
[09:08:29] <PosixC> thnks
[09:09:33] <OnkelSchorsch> np :D
[09:10:38] <PosixC> OnkelSchorsch, should I restart something ? I changes the URL
[09:10:47] <PosixC> but I see that it stills go to the old one
[09:10:58] <PosixC>  ftp://ftp.sunfreeware.com/pub/freeware/i386/5.11/catalog
[09:11:27] <PosixC> with a warning before:WARNING: no catalog file for site ftp.sunfreeware.com
[09:14:31] <OnkelSchorsch> did you comment the previously used server? did you run pkg-get -U
[09:16:33] *** uebayasi has quit IRC
[09:16:47] <PosixC> not to the second pkg-get -U
[09:16:51] <PosixC> will try it
[09:17:04] * dlg sigh
[09:17:24] <PosixC> hell
[09:17:25] <PosixC> ==> TYPE I ... done.  ==> CWD /pub/freeware/i386/5.11 ...
[09:17:26] <PosixC> No such directory `pub/freeware/i386/5.11'.
[09:17:27] <PosixC> ERROR: could not get catalog file
[09:17:28] <PosixC> when
[09:17:31] <PosixC> pkg-get -U
[09:17:53] *** uebayasi has joined #opensolaris
[09:18:42] <g4lt-U60> you know, all this talk about pkg-get reminded me to finally -uU for the first time this year
[09:19:17] <OnkelSchorsch> hehe.
[09:21:07] <OnkelSchorsch> posixc: which mirror did you choose? try to see, if the catalog files are there, by surfing the mirror with your browser :)
[09:21:18] <PosixC> ok
[09:21:37] <PosixC> I tried 2
[09:21:40] <PosixC> one is http://blastwave.solaris-fr.org
[09:22:14] <PosixC> and there exist http://blastwave.solaris-fr.org/stable/i386/5.11/
[09:22:58] <OnkelSchorsch> hmhm. and the catalog is there too. hm. weird
[09:23:58] <PosixC> it says Resolving blastwave.solaris-fr.org... 88.191.22.20
[09:24:18] <PosixC> OnkelSchorsch, can you please try : pkg-get -s http://blastwave.solaris-fr.org -U
[09:24:26] <PosixC> and see the result ?
[09:24:36] <PosixC> and if it resolves the name in the same way?
[09:25:28] *** _dreams_ has quit IRC
[09:25:50] <OnkelSchorsch> hehe. "ERROR: could not get catalog file" :D
[09:25:57] <PosixC> !!
[09:26:45] <PosixC> the same error I get with http://csw.bsw.dk
[09:27:40] *** jamesd_ has joined #opensolaris
[09:27:42] <PosixC> http://mirrors.usc.edu/pub/blastwave/unstable
[09:27:55] <PosixC> seem to work!!
[09:27:55] <OnkelSchorsch> that should work.
[09:28:13] <OnkelSchorsch> yep. the url has to contain  ./unstable
[09:28:18] <PosixC> pkg-get -s http://mirrors.usc.edu/pub/blastwave/unstable -i wireshark
[09:28:18] <PosixC>
[09:28:23] <PosixC> works now !
[09:28:57] <OnkelSchorsch> \o/
[09:29:29] *** jamesd has quit IRC
[09:53:20] *** delewis has quit IRC
[09:58:04] *** delewis has joined #opensolaris
[10:01:35] *** Tpenta has quit IRC
[10:01:51] *** peteh has joined #opensolaris
[10:02:31] *** Tpenta has joined #opensolaris
[10:02:42] *** ChanServ sets mode: +o Tpenta
[10:04:07] *** yongsun_ has quit IRC
[10:10:03] *** GmanAFK has quit IRC
[10:10:06] *** alanc_away has quit IRC
[10:10:52] *** alanc_away has joined #opensolaris
[10:11:06] *** damienc has joined #opensolaris
[10:12:25] *** bunker_work has joined #opensolaris
[10:15:33] *** miffe has joined #opensolaris
[10:16:38] *** rachel_ has quit IRC
[10:17:44] *** rachel_ has joined #opensolaris
[10:23:46] *** delewis has quit IRC
[10:23:57] *** GmanAFK has joined #opensolaris
[10:24:42] <Berny> morning
[10:25:02] <twincest> gman: are you really here? :)
[10:27:06] *** delewis has joined #opensolaris
[10:27:42] <GmanAFK> twincest, hey, yeah
[10:27:46] <GmanAFK> stupid stupid network
[10:29:50] <Tpenta> i see that bugtraq is catching up
[10:30:25] <GmanAFK> yeah, got spammed with a whole heap of mails
[10:30:26] *** GmanAFK is now known as Gman
[10:30:27] *** kFuQ has quit IRC
[10:30:59] *** kFuQ has joined #opensolaris
[10:34:20] *** schily__ has joined #opensolaris
[10:35:37] <raph_ael> hello
[10:41:19] *** mega has joined #opensolaris
[10:43:10] *** schily_ has quit IRC
[10:43:11] *** eboutilier has quit IRC
[10:49:08] <Snake007uk> morning all
[10:52:17] <luxh> moin
[10:53:27] <twincest> gman: i don't suppose you have a moment to delve into the bug db some more? :)
[10:55:33] <Gman> twincest, sure
[10:57:03] <twincest> apparently some comments where added to bugs i filed the other day, but of course i can't see them... (6490780, 6490789, 6490848, 6490925, 6490935 ,6490855)
[10:58:15] <Gman> ok
[11:11:07] *** Gman is now known as GmanZZZ
[11:16:14] *** miffe has quit IRC
[11:28:07] *** sil3nt|warri0r has joined #opensolaris
[11:31:12] *** salamanders has joined #opensolaris
[11:36:32] *** lplatypus has quit IRC
[11:42:46] <wangcity> hi
[11:43:17] <wangcity> when i use pkg install vim but when i reboot, can not find where is it
[11:44:20] <dlg> fix your path in your profile
[11:46:43] <wangcity> how to do this
[11:47:00] <dlg> use vim :D :D
[11:47:24] <wangcity> what? i not und
[11:47:48] <PerterB> dlg: be nice :)
[11:49:00] <OnkelSchorsch> there's a book on vim :)
[11:49:08] <dlg> PerterB: sorry, its been one of those days
[11:50:00] <PerterB> wangcity: where did you get the package from?
[11:55:09] <PerterB> ... if you got it from sunfreeware.com then vim will be in /usr/local/bin, if you got it from blastwave it will likely be in /opt/csw/bin and if you got it from anywhere else you're on your own ("pkgchk -l <pkgname>" and work it out)
[11:55:24] *** bengtf has joined #opensolaris
[11:55:54] <dlg> PerterB: you can get plists out of pkg stuff? ive been wanting to know that for ages
[11:56:47] <PerterB> uhu... play with pkgchk -l, or the simpler way is to grep /var/sadm/install/contents
[11:57:47] <dlg> i dont actually know that much about solaris
[11:58:44] <PerterB> pkgchk should be common across all SysV Unixes... /var/sadm/install/contents might be an implementation detail
[11:59:16] <dlg> i live in the bsd world
[11:59:51] *** pacx has joined #opensolaris
[12:02:51] <PerterB> is it dark and scarey there? ;)
[12:03:32] <dlg> well... no
[12:03:41] <dlg> im familiar with it
[12:03:46] <dlg> so its all sweetness and light
[12:03:49] <dlg> solaris is still uncomfortable
[12:03:51] <dlg> but im getting there
[12:07:14] <wangcity> thanks
[12:13:32] *** lplatypus has joined #opensolaris
[12:15:23] *** peteh has quit IRC
[12:17:00] *** peteh has joined #opensolaris
[12:20:20] <PosixC> is there a way to install solaris on T2000 through network ? I have a suspicion that the DVD inside is not working ok ;  I had put a DVD after checking md5sum on the media itself which was ok, I run
[12:20:33] <PosixC> boot cdrom and I get a panic
[12:20:38] <PosixC> the stack shows
[12:20:44] <dlg> you should be able to netboot them
[12:20:57] <PosixC> uata: atapi_reset_drive
[12:20:59] <PosixC> uata: ata_timeout_func
[12:20:59] <PosixC> uata: ghd_timer_newstate
[12:20:59] <PosixC> uata: ghd_timer_poll
[12:20:59] <PosixC> uata: ghd_timeout_softirq
[12:21:28] <PosixC> dlg, is netboot deals with installing
[12:21:37] <PosixC> from a network?
[12:21:58] <dlg> id be extremely surprised if you couldnt netboot the installer
[12:22:01] <Auralis> a t2000 is able to netboot and install like every other sparc system, you need a jumpstart server
[12:22:39] <PosixC> Auralis, i read a bit about installing via net but a little question:
[12:22:51] <PosixC> ALOM start and I reach the sc > prompt.
[12:22:56] <PosixC> can I netboot from there ?
[12:23:07] <PosixC> or should I do it previously somewhere?
[12:23:13] <PerterB> from an "ok" prompt
[12:23:38] <PosixC> PerterB, I can reach the "ok" prompt
[12:23:45] <PosixC> without problems
[12:23:54] <Auralis> sc> console
[12:24:05] <PerterB> then, once you have configured a jumpstart server (ask google or docs.sun.com) simply "boot net - install"
[12:24:09] <Auralis> then boot net - install after you have setup the installserver
[12:24:33] *** OnkelSchorsch has quit IRC
[12:24:40] <PosixC> Auralis, OK, thnks I will read the docs
[12:25:04] <PosixC> wonder - you did not specify ip in the boot net - install command
[12:25:08] <PosixC> I know that
[12:25:27] <PosixC> in linux kickstart you DO when installing over a net
[12:26:40] <Auralis> thats because the OBP sends out a rarp request which gets handled by the installserver
[12:26:53] <Auralis> or in newer OBP versions a dhcp request
[12:27:08] <PosixC> OK, this is smarter
[12:27:39] <PosixC> I am looking at Solaris 10 6/06 Installation
[12:27:40] <PosixC> Guide: Network-Based
[12:27:40] <PosixC> Installations
[12:27:51] <PosixC> is this jumpstart ? I am not sure
[12:28:15] <PosixC> or is there a better jumpstart manual?
[12:28:22] <PosixC> it deals with jumpstart
[12:28:49] <PosixC> but I see also there is Solaris 10 6/06 Installation Guide: Custom JumpStart and
[12:28:50] <PosixC> Advanced Installatio
[12:29:03] <PosixC> g/ Installatio/ Installation
[12:29:12] <PosixC> which should I use ?
[12:29:21] <Auralis> advanged installation has everything
[12:29:22] <PerterB> you want the first one
[12:29:44] <PerterB> the advanced one covers automated installs which you don't really need at this stage
[12:30:20] <PerterB> http://docs.sun.com/app/docs/doc/819-5776/6n7r9js4t
[12:31:47] <PerterB> uh, add "?a=view" to that :)
[12:34:22] *** sparc-kly|HOME is now known as sparc-kly|WORK
[12:36:49] <PosixC> thnks all
[12:37:32] <fgd> Is it possible to add a FS to a live running zone without rebooting the zone?
[12:38:00] *** Joker^ has joined #opensolaris
[12:38:36] <Joker^> hi all
[12:39:24] <Joker^> it's possible to mount a pcfs fs over nfs ?
[12:40:50] <PerterB> should be, as long as the NFS server can mount it then it should be able to share it
[12:40:54] <Joker^> i'd like to share a fat32 partition mounted under linux with a solaris machine
[12:41:27] <Joker^> under linux nfs server is up&running
[12:41:43] <Joker^> i export /
[12:42:14] <Joker^> i can read all except fat32
[12:42:36] <Joker^> on solaris
[12:44:39] <Joker^> i can mount only a local pcfs fs but not a remote one
[12:46:54] <PerterB> hmm, sounds like a linux issue... as far as the solaris box is concerned it's just nfs regardless of the local filesystem type at the linux end
[12:48:47] <dlg> Joker^: dd the raw partition into an image file on solaris and mount that
[12:50:10] <pacx> you will need mount with nfs3 mode on
[12:50:15] <pacx> v3
[12:50:46] <Joker^> yes ... i mount with -version=3 option
[12:51:01] <pacx> so will work with no problem
[12:53:25] <Joker^> the problem is ... I see all linux fs but not the fat32 fs mounted on linux box
[12:54:09] <Joker^> the folder is empty
[12:54:18] <pacx> hum.. i never try see fat32 fs over nfs
[12:54:38] <Joker^> i don't now if it's possible
[12:54:39] <richlowe> NFS doesn't (shouldn't) traverse remote mount points.
[12:55:22] *** calumb has joined #opensolaris
[12:55:53] <pacx> try share only the fat32 partition over nfs
[12:56:37] <pacx> richlowe: i guess youre right
[12:56:59] <Joker^> .... mount -F pcfs is for local .....
[12:58:15] <Joker^> .... mount -F nfs 'ip':/windows/...... doesn't work
[12:59:40] <pacx> you also can "mount" using smb
[13:00:12] <pacx> solve the problem... but I wanna know now how do this with nfs
[13:00:18] <pacx> must be one way to do
[13:00:31] <pacx> will think about
[13:04:56] <Joker^> ok ... I solved the problem
[13:05:04] <Joker^> It's very stupid ...
[13:05:29] <Joker^> first share over nfs fat32 partition ... then /
[13:05:36] <Joker^> anyway thanks all
[13:05:47] *** sithgunner has joined #opensolaris
[13:06:12] <pacx> Joker^: ?!?!
[13:06:20] <pacx> (09:55:55) pacx: try share only the fat32 partition over nfs
[13:07:03] <Joker^> yes ... i know but i need to share / fs as well
[13:07:12] *** dunc_ has joined #opensolaris
[13:07:23] <pacx> if works, great :)
[13:10:02] *** sithgunner has left #opensolaris
[13:23:17] *** jmcp has quit IRC
[13:23:29] *** Joker^ has quit IRC
[13:29:10] *** jmcp has joined #opensolaris
[13:32:24] *** vmhobbes has quit IRC
[13:58:16] * jmcp sleeps
[14:00:55] * LeftWing also. &
[14:05:22] *** wangcity has quit IRC
[14:19:13] *** calumb is now known as calLNCH
[14:27:07] *** Teknix has quit IRC
[14:33:09] *** SymmWork has joined #opensolaris
[14:35:43] *** Burana has joined #opensolaris
[14:41:22] *** Cass has quit IRC
[14:42:05] *** Burana has quit IRC
[14:42:19] <axisys> where can I get the "Sun StorEdge 3000 Family Professional Storage Manager" software from?
[14:43:58] *** raph_ael has quit IRC
[14:49:25] <Doc> sun.com/download
[14:49:36] *** Burana has joined #opensolaris
[14:49:36] *** Burana has left #opensolaris
[14:49:42] <Doc> ohh.. Professional Storage Manager? never heard of it
[14:50:05] *** Burana has joined #opensolaris
[14:54:17] *** LordKing has joined #opensolaris
[14:55:33] <Cyrille> it's missing a reference to "Java" and "System" to sound like a proper piece of Sun software, though...
[14:57:49] *** calLNCH has quit IRC
[15:00:20] *** raph_ael has joined #opensolaris
[15:02:51] <lasseoe> err .. http://www.sun.com/download/products.xml?id=453debdd
[15:03:02] <lasseoe> sounds like a load of rubbish to me, or I must have missed something :)
[15:05:01] *** pacx has quit IRC
[15:05:22] *** jbk has quit IRC
[15:06:21] *** |tsoome| has quit IRC
[15:12:03] <SymmWork> whats the command in solaris again to see memory utilization (actual swap memory used versus committed etc)
[15:17:22] <Cyrille> prstat? is that what you're looking for?
[15:17:36] <rydis> Possibly vmstat.
[15:18:05] <Cyrille> ah yes, possibly.
[15:19:02] <SymmWork> aaahh prstat was it :) though vmstat also works
[15:23:42] *** jamesd has joined #opensolaris
[15:23:42] *** ChanServ sets mode: +o jamesd
[15:25:04] *** jason has joined #opensolaris
[15:27:12] *** karrotx has quit IRC
[15:27:15] *** jason is now known as jbk
[15:31:13] *** Burana has quit IRC
[15:31:15] *** Burana has joined #opensolaris
[15:31:42] *** OnkelSchorsch has joined #opensolaris
[15:32:10] *** jamesd_ has quit IRC
[15:35:19] *** laca has joined #opensolaris
[15:41:39] <edp> question about toe..
[15:42:10] <edp> suppose there are lazy users that don't train retrain false negatives as spam
[15:42:26] <edp> how much does that affect them if in the future they start doing so?
[15:45:10] <edp> heh.. sorry, wrong channel
[15:47:12] *** axisys has quit IRC
[15:49:00] *** Burana has quit IRC
[15:50:42] *** pete1 has joined #opensolaris
[15:51:05] *** peteh has quit IRC
[15:52:30] *** r3boot has quit IRC
[16:02:30] *** mrdeviant has joined #opensolaris
[16:02:30] <sickness> has someone a web/db inventory to advice? (possibly free)
[16:05:19] *** salamanders has quit IRC
[16:05:44] *** RaD|Tz has joined #opensolaris
[16:08:40] *** vmhobbes has joined #OpenSolaris
[16:16:59] *** Burana has joined #opensolaris
[16:18:48] *** tarzeau has quit IRC
[16:22:21] *** solarisjon has joined #opensolaris
[16:23:00] *** Somethingelse has joined #opensolaris
[16:31:46] *** postwait has joined #opensolaris
[16:31:58] <postwait> If I have a box with an unknown root password.
[16:32:08] <postwait> is there a faster way than via install media to recover the box?
[16:32:19] <postwait> arguments to grub during boot perhaps?
[16:32:36] <postwait> Or is it boot media -> mount -> edit /etc/shadow?
[16:35:19] *** bougie has joined #opensolaris
[16:36:02] <bougie> hello :p
[16:37:15] <Stric> postwait: how up-to-date is it? :)
[16:37:27] *** |tsoome| has joined #opensolaris
[16:37:44] *** Somethingelse has quit IRC
[16:38:26] *** deather_ is now known as deather
[16:38:43] *** |tsoome| has quit IRC
[16:40:36] *** mega has joined #opensolaris
[16:41:48] *** |tsoome| has joined #opensolaris
[16:44:09] *** axisys has joined #opensolaris
[16:48:46] <gdamore> hi *
[16:55:05] *** Auralis has quit IRC
[16:55:10] *** ibs has joined #opensolaris
[16:55:14] *** Auralis has joined #opensolaris
[16:59:13] *** Auralis has quit IRC
[16:59:21] *** rockz has joined #opensolaris
[16:59:30] *** Auralis has joined #OpenSolaris
[17:00:27] <jbk> hmmm
[17:00:31] <rockz> Can someone assist me on burning a Solaris 10 ISO file into a DVD using Nero burning software ?
[17:00:54] <jbk> on s10u1, there seems to be an issue with nfsv4 + directories w/ acls
[17:01:17] <rockz> upon opening the ISO file in Nero, it says the image is corrupted and asking me to correct the setting, but not sure what to change to.
[17:01:43] <raph_ael> download it again rockz ?
[17:01:44] <lasseoe> rockz, then you have a corrupt ISO
[17:02:25] <lasseoe> well, if you are unable to unzip the 5 files, then it's likely to be an issue with how you're joining the files
[17:02:44] <lasseoe> err I mean, if you are ABLE to unzip them
[17:04:43] *** Auralis has quit IRC
[17:05:00] *** Auralis has joined #opensolaris
[17:05:56] <twincest> rockz: how did you create the iso?
[17:06:34] *** jamesb has quit IRC
[17:11:25] <rockz> yeah I suck....I joined the file w/o  unzipping them   :)
[17:23:16] *** bank has joined #opensolaris
[17:23:31] <bank> hi
[17:25:03] *** damienc has quit IRC
[17:26:16] *** yarihm has joined #OpenSolaris
[17:27:23] <bank> err ...
[17:28:35] <bank> I can not find php_mbstring.so extension on solaris x86 ...
[17:29:05] *** vmhobbes has quit IRC
[17:29:22] *** bunker_work has quit IRC
[17:31:36] *** brandini has joined #opensolaris
[17:37:31] *** slowhog has joined #opensolaris
[17:44:28] *** deather_ has joined #opensolaris
[17:49:26] *** rockz has quit IRC
[17:51:28] *** johnlev has joined #opensolaris
[17:52:07] <axisys> any 3510 expert here.. ? i connected a dual port fc hba t2000 to 3510's fc2 and fc3 drive loop and format shows this http://www.rafb.net/paste/results/2iSEKk56.html
[17:52:18] <axisys> am i seeing all 12 disks
[17:52:33] *** deather has quit IRC
[17:53:45] <axisys> i have console access to 3510 now
[17:54:03] <quasi> try posting cfgadm -al as well
[17:54:59] <hali> axisys: if you see the drives, whats the problem?
[17:55:00] <axisys> http://www.rafb.net/paste/results/FQSPF991.html
[17:55:35] <axisys> hali: two things.. am i seeing jbob or lun? am i seeing all 12 or same six from both fc ports?
[17:56:06] <axisys> i connected to drive loop ports fc2 and fc3 on the 3510
[17:56:15] *** junks has quit IRC
[17:56:25] <hali> axisys: you see all 12 on both cards...
[17:56:31] *** junks has joined #opensolaris
[17:56:34] <hali> s/cards/ports/
[17:56:49] <axisys> hali: this is what i see http://www.rafb.net/paste/results/2iSEKk56.html
[17:56:50] <hali> and it looks like its jbod right now..
[17:57:05] <hali> axisys: do you know if the 3510 has a raid controller or just a jbod controller?
[17:57:08] <axisys> hali: thats what i thought..
[17:57:19] <bank> pam_unix_account: cron attempting to validate locked account nobody from local host
[17:57:24] <axisys> hali: it has raid controller .. and i have console access to it now
[17:58:36] <axisys> this is how http://www.rafb.net/paste/results/Kl86mQ71.html logical drives look like
[17:58:43] <axisys> from console
[17:59:59] *** ibs has left #opensolaris
[18:01:03] *** ndroux has joined #opensolaris
[18:01:30] <axisys> CHL 0 ID 40 (Primary Control)
[18:01:38] <axisys>  0|   LD|  1|        0|    180000|RAID0
[18:02:03] <axisys> CHL 1 ID 42 (Secondary Control)
[18:02:10] <axisys>  0|   LD|  0|        0|    180000|RAID0
[18:02:21] <axisys> 40 is fc0 and 42 is fc1
[18:02:30] <axisys> 180000 MB in size
[18:02:39] *** brandini has left #opensolaris
[18:02:50] <hali> axisys: hm, i think you got it hooked up worng
[18:02:58] <axisys> so i guess i can reconfigure to increase the size from 180G to 12 x 146G for both chnl
[18:02:59] <hali> axisys: the drive ports are for additional jbod shelves
[18:03:12] <hali> i think, i haven't used the 3510 in a while
[18:03:27] <axisys> hali: :-) .. i am realizing it now as i talking
[18:04:00] <axisys> i guess i need to reconfigure fc0 (CHL 0 ID 40) with 12 luns each point to a disk
[18:04:15] <axisys> and same for fc1 (chnl 1 id 42)
[18:04:24] <axisys> which is secondary chnl
[18:05:51] <axisys> i am expanding logical drives w/ raid 0
[18:05:55] <axisys> for testing
[18:06:14] <axisys> if that works .. i will recreate luns for each disk
[18:06:21] <axisys> i am learning sooo much..
[18:06:54] <axisys> never setup 3510 before
[18:07:06] <axisys> sorry for newbie annoyeance
[18:08:54] <quasi> axisys: the cfgadm lists the disks as connected to c2 and c3 (ie the 2 paths) - format lists them on c4 which is likely to be the multipath
[18:10:13] *** Snake007uk has quit IRC
[18:10:41] *** Darwin has joined #opensolaris
[18:11:10] *** mega has quit IRC
[18:14:38] <axisys> quasi: yep.. i did a stmsboot -u
[18:15:14] <axisys> quasi: so both of them seeign all 12 disks cool!
[18:15:27] *** LordKing has quit IRC
[18:15:40] <axisys> but they are now in jbod.. i need to switch the connections to fc0 and fc1
[18:15:46] <axisys> to use the raid controller
[18:18:30] <axisys> just finished expanding the fc0 and fc1
[18:24:35] *** solist has joined #opensolaris
[18:25:24] *** RaD|Tz has quit IRC
[18:26:07] *** yarihm has quit IRC
[18:29:02] *** Kronuz has joined #OpenSolaris
[18:29:08] <Kronuz> hello
[18:29:11] <Kronuz> :)
[18:29:20] <Kronuz> jmcp! O_O
[18:29:54] <axisys> what pkg gives sccli bin?
[18:30:17] <hali> isn't that a 3310/3320 tool?
[18:30:45] <hali> oh, it works with 3510/11 as well... my bad
[18:31:03] <bank> :D axisys
[18:31:04] <Kronuz> hey, does anyone know if I really need to have a hard drive on my servers if they can boot from a SAN? I mean, what's the recommendation there... should I always have a local hard drive or would it be okay just to boot directly from the SAN?
[18:31:23] <Kronuz> what do you think?
[18:31:32] *** deather_ is now known as deather
[18:32:41] <Kronuz> 'cause my vendor is trying to convince me I also need the hard drives on my servers (proly just 'cause he want's to sell more)
[18:32:50] <jbk> well depends on how redundant your san is
[18:33:04] <jbk> regardless, you'll probably still want a cd/dvd drive
[18:33:25] <Kronuz> jbk: it would have two FC switches and two two FC HBA connected to it
[18:33:26] <jbk> though would you otherwise need san?
[18:33:57] <Kronuz> so I figured if the SAN doesn't work why would I want to boot from the servers at all :P
[18:34:02] <PerterB> also, if the boot volumes are mirrored at the SAN level and presented as one LUN, you lose the option for splitting mirrors for live upgrade, during patching etc
[18:34:23] <rydis> I'd say you might not need it, but you probably want it, just in case. Dumping a cheapo boot-drive in the server costs pretty much epsilon, no?
[18:35:26] <jbk> but you can always map additional luns for live upgrade
[18:35:38] <PerterB> true
[18:35:55] <Kronuz> rydis: yeah, you're probably right... but the thing is they wanted me to have two SCSI drives in RAID per server... and for what if I'm not going to use them
[18:36:02] <jbk> you might want to still put them under disksuite or veritas control so you can mirror online if you decide to move to a different array as well
[18:36:07] *** gonufer has left #opensolaris
[18:36:57] *** SymmWork has quit IRC
[18:38:33] <bank> does the .so is an independent platform?
[18:39:00] <bank> I am going to copy mbstring.so php extension from friend and put under dir in solaris
[18:39:48] <jbk> it's os and instruction set specific
[18:40:00] <axisys> hali: yep ... but what pkg gives me that?
[18:40:03] <Cyrille> if you're asking whether shared object files are platform independent, as in can be reused from S10 on Sparc to Linux on x86, the answer is no.
[18:40:10] <Kronuz> how do you generally configure volumes in a SAN? via web?
[18:40:10] <axisys> bank: hi
[18:40:25] <hali> axisys: download id 4436c6a0 if i remember..
[18:40:32] <g4lt-U60> .so files are usually platform dependent, they're compiled binaries
[18:40:34] <hali> axisys: http://www.sun.com/download/products.xml?id=4436c6a0
[18:41:46] *** Cyrille has quit IRC
[18:42:26] <jbk> Kronuz: ususally dependso n the array
[18:43:11] <jbk> typically you have to create the luns on the array, then present them out one or more ports on the array to the san, the configure zoning on the switches so that the server can see the luns
[18:43:12] <Kronuz> jbk: yeah, I was trying to find out by looking at the IBM's DS4700 brochure
[18:44:22] <Kronuz> jbk: but I mean do I have to install some weird software which connects to the SAN via FC to configure it or it can generally be done via HTTP...
[18:44:46] <Kronuz> I wonder if the DS4000 series has ethernet and HTTP support for configuration...
[18:44:53] <Kronuz> (or if any other, for that matter)
[18:45:12] <hali> Kronuz: pretty much all SANs have a IP based managment, telnet console and web gui
[18:45:36] *** sniffy_ has joined #opensolaris
[18:46:06] <Kronuz> hali: hmm... that's what I sould have thought so :-) ... don't see anything about that in the brochure
[18:46:31] *** sniffy has quit IRC
[18:48:01] <Kronuz> oh... I suppose it's this: "Storage Manager is designed to allow administrators to quickly configure and monitor storage from a Java technology-based Web browser interface"
[18:48:21] *** darrenm has joined #opensolaris
[18:48:45] <axisys> hali: thnx
[18:48:52] <axisys> hali: 2.3 is the latest for sol 10 sparc?
[18:49:53] *** darrenm has quit IRC
[18:52:43] <bank> I got cc1: out of memory allocating 61545888 bytes after a total of 3317760 bytes during 'make' on solaris with './configure --prefix=/export/local/php --enable-mbstring=all'
[18:53:02] <bank> but I already have 1 GB... isn't that enough ?
[18:57:56] *** Kronuz has left #OpenSolaris
[19:00:57] *** bengtf has quit IRC
[19:05:09] *** B|nTaRa has quit IRC
[19:07:26] *** bank has left #opensolaris
[19:10:32] *** alanc-away is now known as alanc
[19:18:12] <dwc-> depends on how much is unused, doesn't it?
[19:18:51] *** osdummy has quit IRC
[19:21:15] *** osdummy has joined #opensolaris
[19:27:38] *** bondolo has joined #opensolaris
[19:29:29] *** calumb has joined #opensolaris
[19:29:38] *** sniffy_ has quit IRC
[19:29:49] *** dunc_ has quit IRC
[19:36:17] *** bondolo has quit IRC
[19:37:07] *** Tpenta has quit IRC
[19:44:14] *** gnu2it2 has joined #opensolaris
[19:47:34] *** Somethingelse has joined #opensolaris
[19:48:51] *** bondolo has joined #opensolaris
[19:49:16] *** sniffy has joined #opensolaris
[19:52:52] <axisys> mkisofs -o image.iso /media/cdrom0 ignores symlinks
[19:52:59] <axisys> how do i make sure it does not
[19:53:45] <myrkraverk> just a question, on my very fyrst sol install (excr b50) my usb hdds worked, but not now - anything that screams to mind?
[19:54:10] <myrkraverk> axisys: man mkisofs - you need a flip and I don't rceall which one
[19:54:46] *** sahafeez has joined #opensolaris
[19:55:14] <myrkraverk> for the curius about my usb hdd, more info here: http://www.opensolaris.org/jive/thread.jspa?threadID=16489&tstart=0
[19:55:49] <axisys> cool `-f' follows symlink
[19:56:48] *** Tpenta has joined #opensolaris
[19:59:51] <myrkraverk> btw, what do I need to get setxkbmap in ?
[20:00:27] *** bluenerv has joined #opensolaris
[20:01:13] *** frtdx has joined #opensolaris
[20:04:33] *** dlg has quit IRC
[20:04:33] *** flynux has quit IRC
[20:04:33] *** frtdx has quit IRC
[20:04:33] *** calumb has quit IRC
[20:04:33] *** solist has quit IRC
[20:04:35] *** jbk has quit IRC
[20:04:35] *** jmcp has quit IRC
[20:04:36] *** PosixC has quit IRC
[20:04:36] *** jcsmith has quit IRC
[20:04:36] *** Andrew____ has quit IRC
[20:04:38] *** killing-joke has quit IRC
[20:04:38] *** AbeFroman has quit IRC
[20:04:38] *** switch has quit IRC
[20:04:38] *** _soylent_ has quit IRC
[20:04:38] *** pixie_ has quit IRC
[20:04:38] *** trygvis has quit IRC
[20:04:38] *** xinkeT has quit IRC
[20:04:39] *** myrkraverk has quit IRC
[20:04:39] *** twincest has quit IRC
[20:04:41] *** kb7sqi has quit IRC
[20:04:41] *** cypromis has quit IRC
[20:04:41] *** ProfMikey has quit IRC
[20:04:41] *** quasi has quit IRC
[20:04:41] *** bluenerv has quit IRC
[20:04:41] *** sniffy has quit IRC
[20:04:42] *** bondolo has quit IRC
[20:04:42] *** Somethingelse has quit IRC
[20:04:44] *** |tsoome| has quit IRC
[20:04:44] *** postwait has quit IRC
[20:04:44] *** Burana has quit IRC
[20:04:44] *** raph_ael has quit IRC
[20:04:45] *** sil3nt|warri0r has quit IRC
[20:04:46] *** schily__ has quit IRC
[20:04:46] *** ThEbUtChE has quit IRC
[20:04:46] *** Xh4 has quit IRC
[20:04:47] *** Cyl has quit IRC
[20:04:47] *** paul has quit IRC
[20:04:47] *** heffnerd has quit IRC
[20:04:48] *** stevel has quit IRC
[20:04:49] *** elektronkind has quit IRC
[20:04:49] *** DataStream has quit IRC
[20:04:49] *** sickness has quit IRC
[20:04:49] *** boyd has quit IRC
[20:04:49] *** libkeiser has quit IRC
[20:04:50] *** Trident has quit IRC
[20:04:50] *** claudiush has quit IRC
[20:04:50] *** Saltsa has quit IRC
[20:04:50] *** Kernel86 has quit IRC
[20:04:51] *** osdummy has quit IRC
[20:04:51] *** ndroux has quit IRC
[20:04:52] *** pete1 has quit IRC
[20:04:53] *** lplatypus has quit IRC
[20:04:53] *** bobbyz has quit IRC
[20:04:53] *** luxh has quit IRC
[20:04:53] *** Odin- has quit IRC
[20:04:53] *** Drone has quit IRC
[20:04:53] *** mdj has quit IRC
[20:04:54] *** richlowe has quit IRC
[20:04:54] *** noyb has quit IRC
[20:04:55] *** Berny has quit IRC
[20:04:55] *** moazamraja has quit IRC
[20:04:55] *** dvorak has quit IRC
[20:04:55] *** hali has quit IRC
[20:04:55] *** dduvall has quit IRC
[20:04:55] *** TBCOOL has quit IRC
[20:04:56] *** mazon has quit IRC
[20:04:56] *** doownek has quit IRC
[20:04:56] *** icon has quit IRC
[20:04:56] *** sparc-kly|WORK has quit IRC
[20:04:56] *** Auralis has quit IRC
[20:04:58] *** gdamore has quit IRC
[20:04:58] *** LeftWing has quit IRC
[20:04:58] *** Elendal has quit IRC
[20:04:58] *** the-decider has quit IRC
[20:04:58] *** svoboda has quit IRC
[20:04:59] *** Peanut has quit IRC
[20:04:59] *** phips has quit IRC
[20:04:59] *** Stric has quit IRC
[20:04:59] *** lasseoe has quit IRC
[20:04:59] *** wilbury has quit IRC
[20:05:01] *** Dar has quit IRC
[20:05:01] *** spike723_ has quit IRC
[20:05:01] *** crib has quit IRC
[20:05:01] *** prg3 has quit IRC
[20:05:02] *** jbalint has quit IRC
[20:05:02] *** clee has quit IRC
[20:05:02] *** g4lt-U60 has quit IRC
[20:05:02] *** timeless has quit IRC
[20:05:03] *** Tpenta has quit IRC
[20:05:04] *** axisys has quit IRC
[20:05:04] *** solarisjon has quit IRC
[20:05:04] *** OnkelSchorsch has quit IRC
[20:05:05] *** Risky has quit IRC
[20:05:05] *** Error_404 has quit IRC
[20:05:05] *** vectra has quit IRC
[20:05:05] *** zarathustra has quit IRC
[20:05:06] *** dwc- has quit IRC
[20:05:07] *** Hunger- has quit IRC
[20:05:07] *** Sporq has quit IRC
[20:05:07] *** yusufg has quit IRC
[20:05:07] *** |joni| has quit IRC
[20:05:07] *** kirma has quit IRC
[20:05:08] *** nightswim has quit IRC
[20:05:08] *** soultan has quit IRC
[20:05:10] *** mikaeld has quit IRC
[20:05:10] *** Bart_M has quit IRC
[20:05:10] *** mlh has quit IRC
[20:05:11] *** Darwin has quit IRC
[20:05:12] *** johnlev has quit IRC
[20:05:12] *** mrdeviant has quit IRC
[20:05:13] *** rachel_ has quit IRC
[20:05:13] *** uebayasi has quit IRC
[20:05:13] *** simford has quit IRC
[20:05:13] *** linma has quit IRC
[20:05:13] *** Kernel86|Laptop has quit IRC
[20:05:14] *** hile_ has quit IRC
[20:05:14] *** edp has quit IRC
[20:05:15] *** starsoldier has quit IRC
[20:05:15] *** drio has quit IRC
[20:05:16] *** Kitty has quit IRC
[20:05:16] *** IvanR_ has quit IRC
[20:05:16] *** sparkleytone has quit IRC
[20:05:17] *** phalenor- has quit IRC
[20:05:17] *** PerterB has quit IRC
[20:05:17] *** asyd has quit IRC
[20:05:17] *** FireflyST has quit IRC
[20:05:18] *** bougie has quit IRC
[20:05:19] *** laca has quit IRC
[20:05:19] *** alanc_away has quit IRC
[20:05:20] *** sommerfeld has quit IRC
[20:05:20] *** Vanuatoo has quit IRC
[20:05:21] *** neoxed_ has quit IRC
[20:05:21] *** alanc has quit IRC
[20:05:21] *** razrX has quit IRC
[20:05:22] *** rydis has quit IRC
[20:05:22] *** polk_ has quit IRC
[20:05:22] *** fgd has quit IRC
[20:06:18] *** sommerfeld has joined #opensolaris
[20:06:18] *** frtdx has joined #opensolaris
[20:06:19] *** bluenerv has joined #opensolaris
[20:06:19] *** Tpenta has joined #opensolaris
[20:06:19] *** sniffy has joined #opensolaris
[20:06:19] *** bondolo has joined #opensolaris
[20:06:19] *** Somethingelse has joined #opensolaris
[20:06:19] *** calumb has joined #opensolaris
[20:06:19] *** osdummy has joined #opensolaris
[20:06:19] *** solist has joined #opensolaris
[20:06:19] *** Darwin has joined #opensolaris
[20:06:19] *** ndroux has joined #opensolaris
[20:06:19] *** johnlev has joined #opensolaris
[20:06:19] *** Auralis has joined #opensolaris
[20:06:19] *** axisys has joined #opensolaris
[20:06:19] *** |tsoome| has joined #opensolaris
[20:06:20] *** bougie has joined #opensolaris
[20:06:20] *** postwait has joined #opensolaris
[20:06:20] *** solarisjon has joined #opensolaris
[20:06:20] *** Burana has joined #opensolaris
[20:06:20] *** mrdeviant has joined #opensolaris
[20:06:20] *** pete1 has joined #opensolaris
[20:06:20] *** laca has joined #opensolaris
[20:06:20] *** OnkelSchorsch has joined #opensolaris
[20:06:20] *** jbk has joined #opensolaris
[20:06:20] *** raph_ael has joined #opensolaris
[20:06:20] *** jmcp has joined #opensolaris
[20:06:20] *** lplatypus has joined #opensolaris
[20:06:20] *** sil3nt|warri0r has joined #opensolaris
[20:06:20] *** schily__ has joined #opensolaris
[20:06:20] *** rachel_ has joined #opensolaris
[20:06:20] *** alanc_away has joined #opensolaris
[20:06:22] *** uebayasi has joined #opensolaris
[20:06:22] *** ThEbUtChE has joined #opensolaris
[20:06:22] *** bobbyz has joined #opensolaris
[20:06:22] *** Risky has joined #opensolaris
[20:06:22] *** PosixC has joined #opensolaris
[20:06:22] *** simford has joined #opensolaris
[20:06:22] *** Xh4 has joined #opensolaris
[20:06:22] *** Error_404 has joined #opensolaris
[20:06:22] *** jcsmith has joined #opensolaris
[20:06:22] *** Cyl has joined #opensolaris
[20:06:22] *** paul has joined #opensolaris
[20:06:22] *** vectra has joined #opensolaris
[20:06:22] *** gdamore has joined #opensolaris
[20:06:22] *** zarathustra has joined #opensolaris
[20:06:22] *** irc.freenode.net sets mode: +o sommerfeld
[20:06:23] *** LeftWing has joined #opensolaris
[20:06:23] *** Elendal has joined #opensolaris
[20:06:23] *** Vanuatoo has joined #opensolaris
[20:06:23] *** Andrew____ has joined #opensolaris
[20:06:23] *** luxh has joined #opensolaris
[20:06:23] *** Odin- has joined #opensolaris
[20:06:23] *** neoxed_ has joined #opensolaris
[20:06:23] *** heffnerd has joined #opensolaris
[20:06:23] *** linma has joined #opensolaris
[20:06:23] *** killing-joke has joined #opensolaris
[20:06:23] *** stevel has joined #opensolaris
[20:06:23] *** Kernel86|Laptop has joined #opensolaris
[20:06:23] *** dwc- has joined #opensolaris
[20:06:23] *** dlg has joined #opensolaris
[20:06:23] *** hile_ has joined #opensolaris
[20:06:23] *** alanc has joined #opensolaris
[20:06:25] *** Drone has joined #opensolaris
[20:06:25] *** Hunger- has joined #opensolaris
[20:06:25] *** mdj has joined #opensolaris
[20:06:25] *** edp has joined #opensolaris
[20:06:25] *** razrX has joined #opensolaris
[20:06:25] *** rydis has joined #opensolaris
[20:06:25] *** richlowe has joined #opensolaris
[20:06:25] *** AbeFroman has joined #opensolaris
[20:06:25] *** switch has joined #opensolaris
[20:06:25] *** quasi has joined #opensolaris
[20:06:25] *** _soylent_ has joined #opensolaris
[20:06:25] *** trygvis has joined #opensolaris
[20:06:25] *** myrkraverk has joined #opensolaris
[20:06:25] *** xinkeT has joined #opensolaris
[20:06:25] *** kb7sqi has joined #opensolaris
[20:06:25] *** cypromis has joined #opensolaris
[20:06:25] *** flynux has joined #opensolaris
[20:06:25] *** twincest has joined #opensolaris
[20:06:26] *** ProfMikey has joined #opensolaris
[20:06:26] *** pixie_ has joined #opensolaris
[20:06:26] *** Trident has joined #opensolaris
[20:06:26] *** libkeiser has joined #opensolaris
[20:06:26] *** elektronkind has joined #opensolaris
[20:06:26] *** claudiush has joined #opensolaris
[20:06:26] *** DataStream has joined #opensolaris
[20:06:26] *** Saltsa has joined #opensolaris
[20:06:26] *** boyd has joined #opensolaris
[20:06:26] *** Kernel86 has joined #opensolaris
[20:06:26] *** sickness has joined #opensolaris
[20:06:26] *** noyb has joined #opensolaris
[20:06:26] *** doownek has joined #opensolaris
[20:06:26] *** Berny has joined #opensolaris
[20:06:27] *** irc.freenode.net sets mode: +o stevel
[20:06:27] *** hali has joined #opensolaris
[20:06:28] *** moazamraja has joined #opensolaris
[20:06:28] *** dduvall has joined #opensolaris
[20:06:28] *** TBCOOL has joined #opensolaris
[20:06:28] *** mazon has joined #opensolaris
[20:06:28] *** icon has joined #opensolaris
[20:06:28] *** dvorak has joined #opensolaris
[20:06:28] *** sparc-kly|WORK has joined #opensolaris
[20:06:28] *** the-decider has joined #opensolaris
[20:06:28] *** timeless has joined #opensolaris
[20:06:28] *** wilbury has joined #opensolaris
[20:06:28] *** spike723_ has joined #opensolaris
[20:06:28] *** svoboda has joined #opensolaris
[20:06:28] *** Dar has joined #opensolaris
[20:06:28] *** g4lt-U60 has joined #opensolaris
[20:06:28] *** Peanut has joined #opensolaris
[20:06:28] *** phips has joined #opensolaris
[20:06:28] *** jbalint has joined #opensolaris
[20:06:29] *** Stric has joined #opensolaris
[20:06:29] *** clee has joined #opensolaris
[20:06:29] *** prg3 has joined #opensolaris
[20:06:29] *** lasseoe has joined #opensolaris
[20:06:29] *** crib has joined #opensolaris
[20:06:29] *** Sporq has joined #opensolaris
[20:06:30] *** |joni| has joined #opensolaris
[20:06:30] *** nightswim has joined #opensolaris
[20:06:30] *** kirma has joined #opensolaris
[20:06:30] *** soultan has joined #opensolaris
[20:06:30] *** yusufg has joined #opensolaris
[20:06:30] *** Bart_M has joined #opensolaris
[20:06:30] *** mikaeld has joined #opensolaris
[20:06:30] *** polk_ has joined #opensolaris
[20:06:30] *** starsoldier has joined #opensolaris
[20:06:30] *** drio has joined #opensolaris
[20:06:30] *** mlh has joined #opensolaris
[20:06:30] *** Kitty has joined #opensolaris
[20:06:31] *** fgd has joined #opensolaris
[20:06:31] *** IvanR_ has joined #opensolaris
[20:06:31] *** sparkleytone has joined #opensolaris
[20:06:31] *** phalenor- has joined #opensolaris
[20:06:31] *** PerterB has joined #opensolaris
[20:06:31] *** asyd has joined #opensolaris
[20:06:31] *** FireflyST has joined #opensolaris
[20:06:31] *** irc.freenode.net sets mode: +o sparc-kly|WORK
[20:06:57] <Tpenta> if you bought sun h/w from sun, the license for 8/9 was included. if otherwise, then strictly speaking, you should have had to buy a license (i think)
[20:06:57] <alanc> incentive to move people to the latest hardware
[20:06:57] <calumb> solist: you can provide patches, but right now you have to work with a sponsor from Sun to get them back into the code
[20:06:57] <frtdx> http://www.sun.com/software/solaris/get.jsp << 'real' solaris?
[20:06:57] <alanc> Tpenta: free licenses for non-commercial use were offered at various times for S8 & S9 too
[20:06:57] <alanc> S10 is free for any use
[20:06:58] <Tpenta> of course, i knew i was missing a bet
[20:06:58] <calumb> solist: (although the plan is to make that process easier...)
[20:07:02] <gdamore> alanc: not just to the latest hardware, but also to get people off of S8/S9.  it costs Sun money to support older releases.
[20:07:03] <myrkraverk> alanc: ah, I probably got one of those ;)
[20:07:09] <Tpenta> thanks alanc, (only just got out of bed)
[20:07:12] <solist> calumb, this model is more struct than linux model
[20:07:24] <solist> and it reminds me freeBSD model
[20:07:34] <alanc> gdamore: right, I meant move to the latest software - don't know why I typed hardware
[20:07:55] <gdamore> heh.  Sun also wants people to move to more recent hardware, but that's because they want to _sell_ them hardware. :-)
[20:07:58] <calumb> solist: that's why Solaris doesn't crash for years at time :)
[20:08:20] <solist> calumb, and maybe this is one of the reasons that linux became more popular than freeBSD
[20:08:21] <bluenerv> any zfs experts here? I need to know if it is possible to dual mount zfs.. or at least a workable dual pathed solution to the same data
[20:08:25] <frtdx> do i always have to use the Common Desktop Environment CDE or can i use kde / gnome ?
[20:08:27] <solist> calumb, does linux crash?
[20:08:29] <gdamore> i can make linux stay up for years at a time too.  it just won't be doing anything interesting. :-)
[20:08:41] <frtdx> solist: i have crashed linux
[20:08:43] <gdamore> you can use gnome.  Solaris 10 ships with gnome.
[20:08:54] <alanc> Sun would rather you use GNOME, CDE is going away
[20:08:56] <gdamore> its easy to crash linux. :-)
[20:09:01] <solist> frtdx, was it a driver?
[20:09:09] <gdamore> alanc: promises... promises...
[20:09:20] <frtdx> i don't know ... but it crashed more than once on me ... i used kubuntu
[20:09:21] <bluenerv> we can crash solaris10 just as easy here =)
[20:09:39] <gdamore> i crash Solaris all the time, but that's because i'm a kernel dev.
[20:09:50] <frtdx> gdamore: wow!
[20:10:15] <gdamore> my _desktop_ almost never crashes.  there is an old bug in the nge driver that still isn't fixed though. damned unbundled closed source bits.
[20:10:38] <gdamore> it causes a system hang when the cable is unplugged.  :-(
[20:10:49] *** Somethingelse has quit IRC
[20:10:54] <mrdeviant> well, that's one way of notifying the user that the link is down :P
[20:11:00] <Vanuatoo> I've got only gaim running on my JDS Desktop (Solaris Express 10/06) but X uses 145mb. Why?
[20:11:09] <frtdx> hey where can i download the solaris10 DVD iso?
[20:11:21] <sahafeez> sun.com
[20:11:22] <jbk> Vanuatoo: how much memory is in your video card?
[20:11:34] <gdamore> i remember early versions of ce that wouldn't hang, but would panic on loss of link. :-)  fortunately i think they fixed that before the driver shipped
[20:11:37] <Vanuatoo> 128 I guess
[20:11:39] <frtdx> JDS .. is it that 3D desktop ?
[20:11:48] <gdamore> no, just sun-branded gnome.
[20:11:49] <alanc> JDS is GNOME + Mozilla + StarOffice
[20:11:52] <myrkraverk> frtdx: no, it's gnome
[20:11:55] <frtdx> oh
[20:12:01] <jbk> i'm going to guess 128m of that 145mb figure is probably your video card memory
[20:12:04] <alanc> Looking Glass is 3D desktop, but that's not in Solaris yet
[20:12:12] <jbk> though i'm not an X expert
[20:12:14] <frtdx> wasnt there a 3D desktop sun was trying to implement?
[20:12:17] <frtdx> oh
[20:12:17] <gdamore> is looking glass even usable yet?
[20:12:37] <alanc> jbk: yes - ps of the X server confuses people because of the video card memory mapped into the X server
[20:12:46] <Vanuatoo> When I start firefox cpu usage going high
[20:12:58] <Vanuatoo> Even for simple pages
[20:13:05] <gdamore> firefox is a pig. :-(
[20:13:11] <Burana> there is a looking glass community on opensolaris.org
[20:13:55] <Vanuatoo> I'm going to upgrade to dual core amd but I'm afraid it won't be enough (Now I have 3500+ 939)
[20:14:00] <gdamore> alanc: that said, I often see Xorg get to ~1GB.  thanks to firefox pixmap leakage.
[20:14:19] <alanc> yes, well, firefox does that
[20:14:44] <jbk> is that by any chance fixed in 2.0?
[20:14:50] <gdamore> i'm surprised it isn't pissing anyone else off as much as it does me.  or at least it isn't pissing anyone else off who's in a position to fix it.
[20:14:51] <Burana> I would like opera to be integrated into JDS
[20:15:08] <Vanuatoo> gdamore: feel the same way
[20:15:16] <gdamore> Burana: that would be nice.  but opera isn't robust enough yet.  It crashes ~ 1/day for me.
[20:15:37] <Burana> gdamore: i haven't any problems
[20:15:54] <Vanuatoo> I switched from linux, because solaris has better fonts. But solaris is a memory,cpu hog
[20:15:58] <gdamore> do you run it on Solaris?  what version? I'm told that Opera 8 is more stable.
[20:16:09] <sahafeez> ram is cheap. get more.
[20:16:11] <Burana> i run version 9 something
[20:16:13] <gdamore> (than Opera 9)
[20:16:29] <Vanuatoo> I've been waiting to get that firefox bu fixed for half a year I guess
[20:16:38] <Burana> have you migrated old preferences from opera 8
[20:16:39] <Burana> ?
[20:16:40] <gdamore> Most of the cpu/memory hogginess comes not from Solaris, but from Firefox and Gnome.  I run xfce and save a lot memory.
[20:16:48] <jbk> i have nevada running on a sunblade 100 w/ 256mb ram and it seems to do fine for the most part
[20:16:50] <alanc> I thought Firefox 2.0 improved memory usage a bit, but not all the way
[20:16:59] <jbk> actually better than solaris 10 in fact
[20:17:13] <gdamore> i've not tested ffox 2.0.  i heard it has issues, so I've been waiting...
[20:17:30] <Vanuatoo> gdamore:Solaris on the desktop is useless for me If I don't surf the web
[20:17:43] <gdamore> heh.  i remember running Solaris on 32M sparcclassics.  Of course that was Solaris 2.4.
[20:18:09] <gdamore> Vanuatoo: yes, but the firefox pigginess is not restricted to Solaris.  You get it regardless of Linux or Solaris.
[20:18:20] <frtdx> solaris has better fonts ?? nice reason
[20:18:38] <Vanuatoo> frtdx: I do java development
[20:18:43] <Vanuatoo> So I don't care OS
[20:18:46] <gdamore> i'm actually surprised that it does. :-)
[20:18:52] <gdamore> (have better fonts I mean)
[20:19:12] <Vanuatoo> gdamore: Run Netbeans on both OSes and you will see
[20:19:27] <Vanuatoo> NetBeans editor font on linux sucks
[20:19:40] * gdamore uses Emacs for editing code.
[20:19:51] *** neoxed_ has quit IRC
[20:19:53] <frtdx> gdamore: tell me 3 reason y solaris can be considered better at thn linux ? on the top of ur head
[20:19:54] <Burana> wasn't there an adobe logo during desktop startup?
[20:20:01] * gdamore thinks GUI IDEs are a tool for wimpy javascripters. :-)
[20:20:05] <Vanuatoo> It's really shame to admit but windows is better platform for java development than solaris
[20:20:18] <gdamore> i always assumed that the adobe logo was for display postscript.
[20:20:33] <alanc> gdamore: it is - it's actually a postscript logo
[20:20:34] <jbk> heh.. does anything actually use display postscript?
[20:20:39] <alanc> not any more
[20:20:52] <Burana> mac os x uses display PDF
[20:21:10] <gdamore> anyone remember answerbook? :-)
[20:21:11] <Burana> and windows vista uses somthing similar
[20:21:21] <sahafeez> nextstep=displaypostscript
[20:21:30] <gdamore> it needed display postscript.  i was so happy when docs.sun.com switched to PDF and HTML.
[20:21:41] <Burana> I'm happy with display ASCII :-)
[20:21:42] <myrkraverk> well, there's gnustep ...
[20:21:55] <gdamore> yeah, ASCII.  True WYSIWYG. :-)
[20:22:03] <myrkraverk> gdamore: ;)
[20:22:25] <Vanuatoo> I wish sun would assign 1000 developers to desktop things, to mature looking glass and I bet many people would switch immediately
[20:22:48] *** astinus has joined #opensolaris
[20:22:57] <gdamore> hmm... 1000 developers, times ~$100k/year, == $100M investment. :-)
[20:22:57] *** calumb has quit IRC
[20:23:00] <jamesd> Vanuatoo, just as long as they aren't java coders
[20:23:04] <gdamore> per year.
[20:23:33] <frtdx> hey whts with all the segments Solaris 10 6/06 x86 DVD segment 1,2,3 etc?
[20:23:47] <Burana> if I win in eurolotto (200 million ?), I will supply the desktop
[20:24:02] <gdamore> so when your download fails part way in the middle, you can restart without a full restart.
[20:24:09] <gdamore> Burana: i'm gonna hold you to that. :-)
[20:24:30] <frtdx> is there no single Bootable ISO distribution?
[20:24:36] <jamesd> frtdx, most  webservers and browsers don't do well with  files larger than  2GB, and since if a chunk gets corrupted, people won't have to download the whole multiple gigabytes
[20:24:50] <gdamore> frtdx: you download the segments, and cat them together to make the single ISO.
[20:25:11] <Vanuatoo> frtdx: It's really simple
[20:25:30] <frtdx> cat? i use windows xp and kubuntu .. can i triple boot with solaris?
[20:25:31] <Vanuatoo> cat part1 part2 ... partn > big file
[20:25:47] <frtdx> is it like copy /b ?
[20:25:55] <gdamore> heh, its kind of a dirty secret that most of the early java coding work was done originally on Windows.  the java folks were a lot more focused on Windows at first, and Solaris was a 2nd class citizen for a long time.
[20:25:57] <Vanuatoo> exactly
[20:26:02] <frtdx> cool
[20:26:03] <frtdx> got it
[20:26:25] <Vanuatoo> gdamore: It's the reality for now
[20:26:30] * gdamore wonders how many Java developers at Sun still work primarily on Windows.
[20:26:39] <Burana> gdamore: which made sense then, because the target was desktop
[20:26:40] <frtdx> r u one of them?
[20:27:15] * gdamore programs in C, on the kernel.
[20:27:26] * gdamore does not have even one machine with Windows loaded.
[20:27:37] * gdamore uses Solaris exclusively on his desktop, and has for nearly decade.
[20:27:43] <frtdx> :)
[20:28:03] <Vanuatoo> gdamore: for running simple programs it's good
[20:28:16] <gdamore> that said, I do occasionally hack on Java (rarerly, e.g. Tadpole's wifi tool), but that was done with Emacs, and with no gui building.
[20:28:21] <gdamore> s/building/builder/
[20:28:27] <jbk> if i could get working broadcom wireless drivers, i'd pretty much be using solaris except when i need to vpn into work
[20:29:07] <gdamore> jbk: good luck.  Broadcom is _particularly_ closed minded about helping anyone not running Windows.
[20:29:18] <gdamore> (at least their wifi division.)
[20:29:23] <jbk> :(
[20:29:52] *** jafari has joined #opensolaris
[20:30:15] <razrX> wasn't Broadcom the same company which refused the OpenBSD guys to have a look into their wifi stuff ?
[20:30:21] <gdamore> we (Tadpole) went on a massive search for wifi vendors a few years ago.  Broadcom was one of the companies we were looking at that.  They flat out told us they didn't want to spend the time helping us develop drivers for Solaris, even closed source drivers under NDA.
[20:30:58] <axisys> i created a logical drive of raid 5 with 12 145G disk .. now how do i map that to a host port on 3510?
[20:31:07] <gdamore> Intersil was the easiest to work with, until they were bought by Conexant.  Even then they were pretty helpful, until they laid off most of their programmers.
[20:31:18] <jbk> funny how that works :)
[20:31:59] * gdamore wrote drivers for conexant a/b/g cards that work in Sun Ray firmware. :-)
[20:32:37] * gdamore also has drivers that could be used for same cards in x86 solaris, but lack of a robust 64-bit softmac for SPARC makes them unusable on SPARC solaris.
[20:33:01] <gdamore> so as a result, the drivers are locked away behind the Tadpole corporate firewall. :-(
[20:33:14] <quasi> gdamore: bummer
[20:34:14] <gdamore> i do have a version of a driver that works on sparc with "fullmac" cards (early 802.11G cardbus and miniPCI), but since you can't buy those cards anymore, that driver will probably never see the light of day either.
[20:34:21] <Burana> i would like a sun ray that doesn't need dhcp.
[20:34:38] <jbk> i'd be tempted to try to use the docs from the group taht has been reverse engineering the broadcom drivers (for personal use).. but it looks like a lot of the low level infrastructure isn't integrated yet (at least last time i looked)
[20:34:54] <gdamore> Burana: Comet (Sun Ray clone) doesn't need DHCP.
[20:35:20] <Burana> how does it work?
[20:35:29] <gdamore> soon (if not already) even stock Sun Rays will have manual configuration support.
[20:35:42] <gdamore> basically we allow static configuration to be stored in flash memory.
[20:35:49] <Burana> gdamore: way cool!
[20:36:08] *** solarisjon has left #opensolaris
[20:36:24] <dwc-> broadcom's busy doing WHQL testing for vista =p
[20:36:43] <Burana> Apropoz Sun Rays. Have you read Jonathan Schwartz's newest blog entry?
[20:37:14] <Burana> Jonathan doesn't like thin clients.
[20:37:22] <dwc-> gdamore: any chance of getting wpa support or cisco VPN support into the TADwifi? =]
[20:37:45] <Burana> But reading his blog, I really don't undestand, if he counts sun rays as thin clients.
[20:38:12] <gdamore> dwc: maybe.  wpa could happen with a conversion to the new nevada framework.
[20:38:21] <Vanuatoo> I'm suing x11vnc tu connect remotely to my solaris machine from home
[20:38:39] <gdamore> Burana: I've not read the blog entry, but Sun historically makes a distinction between "thin clients" and "ultra-thin clients".
[20:38:45] *** ayan has joined #opensolaris
[20:38:52] <dwc-> mm, cool
[20:38:55] <gdamore> Sun Ray is "ultra-thin".  Almost no local power at all.
[20:39:15] <Vanuatoo> At home I use tightvnc. The line is good, 3mbit I guess, but Desktop is still slow
[20:39:26] <ayan> all: is there any work being done on dbx?  if the source is available, i have a simple patch i'd like to submit.
[20:39:33] <gdamore> compared to running an Xterm or a Vnc client on a lower end PC (which is jsut a regular thin client)
[20:39:47] <Burana> gdamore: wasn't there a project to make a SW Sun Ray Protocol Client?
[20:39:49] <gdamore> ayan: probably there is.  but i'm not sure its part of this group.
[20:40:09] <Burana> And what about open sourcing the sun ray protocol?
[20:40:10] <gdamore> Burana: we (Tadpole) wrote a software client for Sun Ray. :-)  I have it running under Solaris right now.
[20:40:23] <gdamore> Burana: not going to happen.  at least not any time soon.
[20:40:39] <gdamore> the protocol is closed, and likely to remain that way for the indefinite future.
[20:40:59] <Burana> pitty
[20:41:14] <gdamore> Sun makes a lot of money on Sun Ray licenses. :-)
[20:41:25] <Burana> I would like a client for my Nokia mobile phone
[20:41:48] <dwc-> hm, no ath hal for sparc either...
[20:41:50] <gdamore> heh.  that was one of the platforms we were talking about targetting for our nextgen product. :-)
[20:42:08] <gdamore> dwc-: no, Sam needs $$ to make a Solaris-specific sparc64 hal.
[20:42:25] <dwc-> guess I'm stuck with WEP for now :P
[20:42:29] <gdamore> i've tried to get Tadpole to fund one, but no dice.
[20:42:46] <Vanuatoo> I guess dtrace is good for tracing performance and memory issues. I saw lately on alanc blog that X server probes should be integrated into the latest build. Does that mean that performance and memory bugs will be nailed down much more effectively?
[20:42:49] <dwc-> that woulda been the other way to get WPA on my tadpole... I've got an ath pcmcia card
[20:43:03] <gdamore> i keep meaning to hava shot at one of the new RAL cards, but I still need to go out and buy one.  to see if I can make it work on SPARC.
[20:43:32] <gdamore> dwc-: which model tadpole do you have?  I don't recall?
[20:43:33] <ayan> it'd be nice if dbx repeated the last command if given just a newline (like gdb).  if there is a way to do that, please let me know. :^)
[20:43:40] <dwc-> sparcle, 650mhz
[20:44:04] <gdamore> were you the one that had a bad connector for the power?
[20:44:11] <dwc-> yup
[20:44:16] <gdamore> did you get it fixed?
[20:44:55] <dwc-> not yet... I took it apart and the MB-mounted jack is in a couple more pieces than it should be
[20:45:16] <gdamore> www... buying a budget external USB 2.0 dvd drive, and finding the unit looks _nothing_ like the external packaging.
[20:45:18] <dwc-> but I did juryrig it up enough to copy the data I needed over to another computer
[20:45:31] *** Darwin has quit IRC
[20:45:42] <dwc-> get a 3.5" enclosure and take the board out?
[20:46:03] <gdamore> dwc-: if you want, I can try to make a case for Tadpole giving you a free or cheap repair.  but you'd have to send the unit in.
[20:47:09] <dwc-> send it in first, or ... ?
[20:47:37] *** jafari has quit IRC
[20:47:40] <gdamore> i'd have to get you an RMA #, but it would have to come in for repair.
[20:47:50] <gdamore> where are you located?
[20:47:57] <dwc-> san jose
[20:48:25] <gdamore> cool.  if you're willing, maybe i could even get you same day service at our location in Cupertino.
[20:48:40] <gdamore> i.e. no need to ship it.
[20:49:17] *** ayan has quit IRC
[20:49:36] <dwc-> if you can't swing it, don't worry about it.  I should be able to find the connector and self-repair...  when I get a chance.
[20:49:51] *** yippi has joined #opensolaris
[20:50:23] <gdamore> let me ask, i should have an answer for you in a day or so.
[20:50:48] <dwc-> alright, much appreciated
[20:51:05] <gdamore> np.  but also no promises.  i'm just a lowly peon engineer. :-)
[20:51:22] <gdamore> (software engineer, making it harder to get hardware folks to give me the time of day. :-)
[20:52:17] <dwc-> :p
[20:53:02] <gdamore> note to purchasers: stay away from "hival" (www.hival.com) USB dvd drives.  they are ugly repackaged/rebranded iomagic drives.
[20:53:41] <dwc-> hey, I have a hival 24x cdrom drive
[20:53:49] <dwc-> must be about 10 years old by now
[20:54:10] <Burana> gdamore: is there some plan to add UMTS support to tadpole comets? the WLAN coverage is not that high in europe
[20:54:39] <gdamore> there has been talk off and on about it.  certainly we had plans for a nextgen that would add it.
[20:55:01] <Burana> what is this nextgen?
[20:55:04] <gdamore> one possible solution we had prototyped was a bluetooth connection via a cellphone.
[20:55:14] <gdamore> (next generation comet, code named meteor.)
[20:55:36] <gdamore> right now, i can't really say much about meteor.  the project is sort of in limbo.
[20:55:52] <Burana> an i haven't signed an NDA :-)
[20:56:15] <gdamore> yep.  :-)
[20:56:40] <Burana> but it is certainly nice to hear that there is something new coming...
[20:56:58] <gdamore> meteor was the reason i spent several months of my life working on NetBSD, and why we funded Bluetooth stack development effort for NetBSD.
[20:58:48] <Burana> is there any effort in getting bluetooth into opensolaris?
[20:58:59] <gdamore> i'm not aware of any.
[20:59:22] <Burana> gdamore: go ahead :-)
[20:59:25] <gdamore> i would be interested in working on a bluetooth port starting from the NetBSD code.  but nobody is funding me to work on that. :-)
[20:59:49] <gdamore> Burana: you can fund me when you win that $200M euro lottery. :-)
[21:00:08] <Burana> I just thought the same :-)
[21:00:32] <axisys> ok so i successfully created a logical drive with 12 146G disks on a 3510 and mapped to one host lun and can see whole 1.47TB in my T2000
[21:00:35] <axisys> eehaa
[21:00:42] <axisys> this is just a test..
[21:00:51] <Burana> you have nice tests :-)
[21:00:57] <axisys> now i delete all and rebuilt with one lun per host
[21:01:14] <Burana> what FC-Adapters are you using=
[21:01:23] <jbk> speaking of bluetooth, anyone know of any plans to add bluetooth support to solaris?
[21:01:32] <axisys> can i map same lun to two different host port on 3510
[21:01:34] <gdamore> must be nice to h ave all that storage.  but I hope they are in a different room than where you are sitting. :-)
[21:01:42] <axisys> Burana: it is a fc dual port
[21:01:50] <axisys> spf
[21:01:51] <Burana> axisys: brand?
[21:01:55] <sickness> and I hope you're going to fill them with pr0n >:P
[21:01:55] <axisys> sun
[21:02:00] <gdamore> jbk: i'm not aware of any plans.  the NetBSD code would be a good place to start.
[21:02:05] <axisys> sickness: heh
[21:02:09] <quasi> gdamore: a 3510 is quiet compared to a thumper
[21:02:32] <axisys> so can i map same lun to two different host port on 3510?
[21:02:44] <gdamore> heh.  i used to work in rooms filled with E10ks.  that was noisy. :-)
[21:03:04] <gdamore> now the SB100 is the noisest thing in my room, unless I turn on a Bullfrog 2.
[21:03:05] <Burana> gdamore: animals have more rights...
[21:03:16] <quasi> gdamore: have you ever heard one of them panic?
[21:03:20] <jbk> axisys: i believe so
[21:03:29] <quasi> gdamore: when all fans go to max
[21:03:33] <gdamore> one of which? E10K?  Yes.
[21:03:40] <axisys> jbk: would u know how.. i have console and sccli access to it
[21:03:43] <gdamore> I worked on the Control Board software, including the fan controller softwar.e
[21:04:10] <jbk> axisys: off the top of my head, no it's been like 2 years since i've touched them, but i don't recall it being that difficult
[21:04:26] <gdamore> the fans go to max when the control board resets.  you can panic a domain without impacting fan speed.
[21:05:08] *** frtdx has quit IRC
[21:05:15] <axisys> jbk: well when i choose CHL 1 ID 42 and try to create a logical drive.. it says no drive available
[21:05:22] <axisys> CHL 1 ID 42 is fc1
[21:05:29] <axisys> fc0 looks good
[21:05:38] <axisys> created a lun same way
[21:05:42] <axisys> no complain
[21:06:03] <axisys> as if it is saying i already mapped the logical drive to fc0 and none available for fc1
[21:06:16] <axisys> even though fc1 is labeled as secondary chnl
[21:08:06] *** spackes1 has joined #opensolaris
[21:08:42] <gdamore> hmm... i think there is abug in some software that parses the ARC case logs.
[21:08:45] <gdamore> http://www.opensolaris.org/os/community/arc/caselog/
[21:08:58] <gdamore> note the missing titles for 2006/587 and 2006/629.
[21:09:19] *** Ivanov25 has joined #opensolaris
[21:12:58] *** GmanZZZ is now known as Gman
[21:12:59] <Gman> hey hey
[21:16:06] *** Gman has quit IRC
[21:19:44] *** Gman has joined #opensolaris
[21:25:06] *** neoxed has joined #OpenSolaris
[21:31:52] *** osdummy has quit IRC
[21:33:20] *** Kronuz has joined #OpenSolaris
[21:34:17] *** akolb has joined #opensolaris
[21:37:00] *** Symmetria has joined #opensolaris
[21:37:24] *** bengtf has joined #opensolaris
[21:42:50] *** mrdeviant has quit IRC
[21:44:30] *** jhawk has joined #opensolaris
[21:47:42] *** fik has joined #opensolaris
[21:52:07] *** jhawk has quit IRC
[21:53:47] *** pikapika has joined #opensolaris
[21:54:51] <pikapika> hello
[22:01:11] *** hell` has joined #opensolaris
[22:03:11] *** koolniczka has joined #opensolaris
[22:06:19] *** Seawolf91 has joined #opensolaris
[22:06:56] *** Symmetria has quit IRC
[22:06:59] *** Seawolf91 has left #opensolaris
[22:15:08] *** spackes1 has left #opensolaris
[22:26:27] *** koolniczka has quit IRC
[22:32:44] *** koolniczka has joined #opensolaris
[22:36:08] *** yarihm has joined #OpenSolaris
[22:38:34] <boyd> Morning, all
[22:40:10] <alanc> gdamore: complain to plocher - he is the master of all scripts ARC-related
[22:40:27] <gdamore> alanc: i was just noting it here. :-)
[22:40:42] <gdamore> at the moment i can't send e-mail.  the e-mail server is down.
[22:40:47] <alanc> yeah, but plocher doesn't IRC much
[22:43:18] <jbk> wow.. new sun backpacks are nice
[22:43:47] <axisys> how do i create a lun to a secondary controller on 3510
[22:44:12] *** bougie has quit IRC
[22:44:32] <axisys> if i try to create thru the console it says no logical drive available.. but i do have 6 left out of 12 that i created
[22:44:50] <axisys> i have 12 logical drives to 12 physical drives
[22:47:17] <axisys> funny i can create multiple luns to same drive from primary channel host port.. but cant do shit on the secondary chnl
[22:47:28] <axisys> is it setup to be builtin failover?
[22:47:35] <axisys> secondary chnl that is
[22:49:46] *** Burana has left #opensolaris
[22:52:18] <gdamore> alanc: it was more of an observation than a complaint, in any case.  kind of a nit, not a real problem.
[22:52:40] <gdamore> my guess is that some special character (e.g. "'") is responsible for the parse failure
[22:54:47] *** mdj has quit IRC
[22:56:08] *** solist has quit IRC
[22:56:23] *** Tpent1 has joined #opensolaris
[22:56:51] <boyd> Anyone heard anything on the latest hg probs? Such as, is there some prep I can be doing in my end to get ready?
[22:57:14] *** Ivanov25 has quit IRC
[23:01:24] *** koolniczka has quit IRC
[23:02:17] *** slowhog has left #opensolaris
[23:04:44] *** logic_ has quit IRC
[23:13:44] <gdamore> www... sysadmin decided to disable an IMAPS port without telling anyone.  aargh.
[23:14:15] <|tsoome|> nice:)
[23:14:42] <|tsoome|> imap is really dangerous;)
[23:15:19] <gdamore> i was using imaps (imap over SSL).  now i'm supposed to go over a VPN.  it doesn't much matter, but some warning would have been nice.
[23:15:44] <|tsoome|> and I'll bet all ports are open via vpn...
[23:16:03] <gdamore> yep, pretty much.  i'm not too impressed with our IT folk.
[23:16:41] <|tsoome|> same here. it's too dangerous to make access to service X, but feel free to have full access to office lan via vpn
[23:16:45] <gdamore> waaayyy too focused on Windows and not enough low-level understanding of TCP/IP protocols, routing, etc.
[23:17:11] <|tsoome|> vpn is obviously pptp....
[23:18:56] <gdamore> hmm... i didn't realize Philips was renaming itself NXP....
[23:19:15] *** sparc-kly|WORK is now known as sparc-kly|HOME
[23:29:37] <dwc-> why is vpn obviously pptp
[23:30:27] <gdamore> it isn't.  in fact, in my case, its just basic IPsec.
[23:31:33] <dwc-> ipsec+imaps isn't such a bad thing
[23:31:45] <dwc-> not sure why they'd turn it off, unless they need to sniff passwords or something
[23:32:08] <dlg> if you own the mailserver then you can do that anyway
[23:32:25] <dwc-> I know our exchange 2003 box here doesn't do TLS, but we do have imap over SSL
[23:32:29] *** axisys has quit IRC
[23:32:57] <gdamore> i was only using imaps (over the public net), not vpn.  its not a bad idea to turn it off, i just would have liked to know apriori.
[23:33:33] <dwc-> would have been nice
[23:34:50] <dlg> hwo is it a good idea to turn ssl off?
[23:35:20] <gdamore> ssl can be secure (as secure as VPN), but if it isn't properly configured it can be weak.  (e.g. rc4 has known weaknesses)
[23:35:31] <gdamore> and SSLv2 has known weaknesses.
[23:35:50] <dwc-> the less services you have exposed on public net the better
[23:35:53] <gdamore> SSLv3 and TLS properly configured to use AES and large certs is strong, though.
[23:36:09] <gdamore> dwc: yes.  i think that might have been one of the reasons for turning off the imaps service.
[23:36:16] <dwc-> now, if they turned off SSL'd imap, but left unsecured imap on, then that's just silly
[23:36:27] <gdamore> i was only using it because the VPN tunnel was so damned unreliable in the past
[23:36:50] <gdamore> no, they left unsecured IMAP open but only over the VPN.
[23:37:07] <gdamore> i presume they just removed the hole in the firewall for imaps. :-)
[23:37:10] *** killerdragon has joined #opensolaris
[23:38:30] *** slowhog has joined #opensolaris
[23:40:22] <myrkraverk> what do you ppl do, when usb hdds don't want to show up in solaris?
[23:41:08] <killerdragon> myrkraverk: Use Redhat Linux!
[23:41:46] <myrkraverk> killerdragon: I hate Linux - I may yet learn to hate Solaris too :P
[23:42:02] <killerdragon> myrkraverk: firewire works.
[23:42:10] *** laca is now known as lacaAFK
[23:42:15] <killerdragon> myrkraverk: USB disks are... uhm... picky.
[23:42:23] <killerdragon> or better: solaris is picky.
[23:42:35] <myrkraverk> killerdragon: this one *worked* in solaris ;/
[23:42:53] <myrkraverk> killerdragon: now, it just workes on my Linux box ;/
[23:43:02] <killerdragon> myrkraverk: the response goes from: "working" over "no disks recognized" up to "kernel freezes".
[23:44:11] <myrkraverk> killerdragon: this one went from "works purrfectly" to "no disks recognized" when I reinstalled Solaris - I'm perplexed
[23:45:41] <myrkraverk> same hw - same software
[23:45:42] * killerdragon throws a komodo dragon at myrkraverk and watches the mayhem...
[23:45:54] <myrkraverk> what's a komodo dragon?
[23:46:15] <killerdragon> myrkraverk: large lizard
[23:46:24] <myrkraverk> I know I tried some commands when I first got it to work - can anyone throw some at me to try?
[23:46:43] <killerdragon> myrkraverk: actually the largest lizand specifes which still lives on this planet
[23:46:52] <killerdragon> myrkraverk: http://en.wikipedia.org/wiki/Komodo_dragon
[23:47:11] <myrkraverk> killerdragon: ah
[23:47:34] <myrkraverk> any devfsadm magic I can try?
[23:48:21] *** Kronuz has quit IRC
[23:53:35] *** killerdragon is now known as gisburn
[23:54:02] <gisburn> myrkraverk: does the device or the usb port show up in /var/adm/messages ?
[23:54:17] <gisburn> myrkraverk: and is the cable Ok ?
[23:54:43] <gdamore> later all.
[23:54:45] *** gdamore has left #opensolaris
[23:56:06] <myrkraverk> cable should be ok - it works in linux
[23:57:09] <gisburn> myrkraverk: does the disk have a firewire port =
[23:57:16] <gisburn> s/=/\?/
[23:57:23] <myrkraverk> gisburn: no ;(
[23:57:34] <gisburn> myrkraverk: GAME OVER
[23:57:47] <gisburn> (at least in my case)
[23:57:50] <myrkraverk> gisburn: and yes, it shows up in /var/adm/messages - with no driver found
[23:57:59] <gisburn> uhm
[23:58:08] <gisburn> myrkraverk: what is the exact message ?
[23:58:13] <myrkraverk> but it *did* work - with my very first sol installation
[23:58:40] <myrkraverk> Nov 9 18:25:04 asuka usba: [ID 723738 kern.info] /pci@0,0/pci1025,90@1d,7/device@1 (usb_mid1): no driver found for device /pci@0,0/pci1025,90@1d,7/device@1/interface
[23:59:13] <myrkraverk> or one very much like it - it seems to change a bit with the port I plug it in





top