[00:00:21] <CIA-27> seb: 6479886 mipagent and related kernel interfaces should be removed (fix license list)
[00:01:10] *** Yamazaki-kun has quit IRC
[00:04:11] *** postwait has quit IRC
[00:05:38] *** sioraiocht has joined #opensolaris
[00:05:51] *** alobbs has quit IRC
[00:06:41] *** bubbva has quit IRC
[00:11:17] <coffman> haha zfs boot in solaris 10
[00:11:24] <coffman> bad joke
[00:11:28] <coffman> n8
[00:11:36] *** coffman is now known as coffman_zzz
[00:16:18] <jbk> evening
[00:23:05] *** Gekkko[PDA] has quit IRC
[00:23:28] *** Pietro_S has quit IRC
[00:26:11] *** sfire||mouse has joined #opensolaris
[00:31:05] *** EchoBinary has quit IRC
[00:42:16] *** hohum has quit IRC
[00:42:40] *** unixware has joined #opensolaris
[00:51:13] *** unixware has quit IRC
[01:00:17] *** jamesd has quit IRC
[01:04:31] *** Cass has joined #opensolaris
[01:09:01] *** Cass has quit IRC
[01:11:32] *** yippi has quit IRC
[01:12:10] *** Cass has joined #opensolaris
[01:15:37] *** jamesd has joined #opensolaris
[01:15:37] *** ChanServ sets mode: +o jamesd
[01:18:10] *** ichigo has joined #opensolaris
[01:19:35] *** cypromis has joined #opensolaris
[01:20:35] *** kimc has joined #opensolaris
[01:25:36] *** ichigo has quit IRC
[01:26:07] <asyd> sun studio for linux is available only in rpm?
[01:27:09] <stevel> http://developers.sun.com/sunstudio/downloads/index.jsp
[01:27:13] <stevel> you can download it as a tarball
[01:27:56] <asyd> afaik the tarball contains rpm :/
[01:28:06] <asyd> let me try
[01:29:47] <asyd> ouah, 250MB
[01:31:41] <jmcp> asyd: well don't chose a crappy package format :-)
[01:31:45] <jmcp> :-P
[01:32:17] <stevel> ah. it claims "This option extracts the installed version of Sun Studio software from a tarfile without using Solaris or Linux packages."
[01:35:17] *** dunc has joined #opensolaris
[01:37:54] *** detriment has joined #opensolaris
[01:40:55] *** Cass has quit IRC
[01:41:18] *** Cass has joined #opensolaris
[01:43:35] *** baijiutong has quit IRC
[01:45:11] *** cypromis has quit IRC
[01:45:57] *** uebayasi has joined #opensolaris
[01:46:02] *** jbk has quit IRC
[01:46:09] *** jbk has joined #opensolaris
[01:52:16] <kimc> [-]=h
[01:52:41] *** movement has quit IRC
[01:52:48] *** jcea has quit IRC
[01:53:28] *** Trisk[laptop] has quit IRC
[01:59:59] *** kimc has quit IRC
[02:14:44] *** theRealballchalk has quit IRC
[02:17:08] *** NikolaVeber has quit IRC
[02:17:39] *** Cass has quit IRC
[02:18:41] <Doc> hmm.. more RIF's at Sun
[02:19:20] <g4lt-U60> ouch
[02:19:47] *** danv12 has joined #opensolaris
[02:21:42] <Jondice> any idea if sun will release a T2 based workstation?
[02:21:45] <mlh> psst when is Sun releasing a T2 based system
[02:21:51] <mlh> Jondice: SNAP!!
[02:21:55] <Jondice> haha
[02:23:52] <alanc> I don't believe there are any plans for a niagara workstation of any flavor
[02:24:05] <alanc> but if there were, I probably couldn't tell you
[02:30:44] *** freakazoid0223 has quit IRC
[02:30:46] *** Cass has joined #opensolaris
[02:32:05] *** sstallion has quit IRC
[02:33:50] *** freakazoid0223 has joined #opensolaris
[02:34:34] <mlh> I'm more interested in a server; a T2000 successor.
[02:34:47] <mlh> But I guess we'll know soonish
[02:37:16] *** ShadowHntr has joined #opensolaris
[02:39:25] *** postwait has joined #opensolaris
[02:39:26] <alanc> oh, was today just the CPU launch?   thought the new servers were being announced too - guess those will come soon then...
[02:39:36] <Jondice> I was thinking it could make for a competitive solution for ray tracing workloads
[02:39:37] <alanc> or you could just go read the reg
[02:40:26] <alanc> but do many people do raytracing on their desktop workstations?  wouldn't you offload that to a server in the back room?
[02:40:40] <Doc> niagra doenst make a lot of sense for a workstation
[02:40:52] <Jondice> for final frames yeah
[02:41:09] <Doc> sunray server perhaps, but it's not going to make a good general purpose workstation
[02:41:42] <Doc> just put it in the datacenter and buy a bloody long video cable :)
[02:41:57] <Jondice> hehe
[02:42:06] *** stevel has quit IRC
[02:49:03] *** detriment is now known as movement
[02:49:09] *** sommerfeld has quit IRC
[02:49:47] <hile_> Doc: you told me once before, but I can't find my logs... what's the OBP command to send diag stuff to the RSC on a 280R/v440/v880? `diag-console rsc'?
[02:50:17] *** Cass has quit IRC
[02:52:47] *** estibi has quit IRC
[02:54:39] *** vmlemon has quit IRC
[02:58:36] <Tempt> Sunray *is* a bloody long video cable. Over ethernet.
[02:59:23] <Tempt> Hmm
[02:59:51] <Tempt> So many people ask about T1/T2 workstations. Perhaps there could be an industry rebuilding the machines into a workstation chassis.
[03:00:45] *** simford has joined #opensolaris
[03:01:56] <BatonT> OR just OEM the chip from sun
[03:03:10] <BatonT> is there any info on the T2 servers they are releasing yet (only info is about the chip at the moment)
[03:04:40] *** Tpenta has quit IRC
[03:08:00] *** eboutilier_ has joined #opensolaris
[03:08:03] *** p-a-venk` has quit IRC
[03:10:07] *** p-a-venk has joined #opensolaris
[03:14:48] <Tempt> You see, one could quite easily mount the T2000 motherboards into a desktop chassis
[03:14:56] <Tempt> Shove a display adapter in and all good
[03:15:13] <Tempt> Designing an entire system, however, would take actual engineering effort and expense.
[03:15:50] *** TomJ has joined #opensolaris
[03:16:29] <e^ipi> the price that sun wants for even a T1k would price it outside of most people's budget
[03:17:00] <e^ipi> people don't /actually/ want a t2000 in a desktop box, they want a sub-$2000 sparc workstation
[03:17:11] <e^ipi> of recent vintage
[03:17:49] <moazamraja> they do?
[03:18:02] *** Atomdrache has quit IRC
[03:18:31] <Tempt> They do?
[03:18:42] <Tempt> Then why do they ask for a T1 in a workstation? Ignorance?
[03:19:33] <alanc> 32-way workstation dude!   imagine the bragging rights!
[03:20:00] <Tempt> Hmm, ebay - Alphaserver ES40; 4 x 500; 8Gb RAM @ $199
[03:20:04] <Tempt> That sounds full of win
[03:20:20] <e^ipi> i'd take it at that price
[03:20:33] <Tempt> That'll do for my OpenVMS machine
[03:20:35] <Tempt> Excellent.
[03:20:48] *** postwait_ has joined #opensolaris
[03:23:41] *** Tpenta has joined #opensolaris
[03:24:30] *** sfire||mouse has quit IRC
[03:24:52] *** TomJ has quit IRC
[03:25:59] <Tempt> Hmm, AlphaServer GS80
[03:26:01] <Tempt> That'd be nice.
[03:26:07] *** TomJ has joined #opensolaris
[03:27:24] *** TomJ has left #opensolaris
[03:27:42] *** reidms has joined #opensolaris
[03:28:41] *** halton has joined #opensolaris
[03:30:11] *** danv12 has quit IRC
[03:31:10] *** RElling has joined #opensolaris
[03:31:11] *** lasseoe has quit IRC
[03:31:13] *** lasseoe has joined #opensolaris
[03:34:53] *** sarahj has joined #opensolaris
[03:35:59] *** postwait has quit IRC
[03:37:36] <moazamraja> i'd take a quiet T1000 1u rackmount that can be run at the desk
[03:37:43] <moazamraja> but, without the fan noise
[03:40:51] *** Atomdrache has joined #opensolaris
[03:41:43] *** Gman has joined #opensolaris
[03:48:27] <Tpenta> hey glynn
[03:49:15] *** ChanServ sets mode: +o Tpenta
[03:51:41] <Tpenta> a quiet t1000????? only if you are deaf. They sound like an industrial vacuum cleaner!
[03:52:00] <jbk> :)
[03:52:15] <jbk> one of my old coworkers used to have a V20z in his cube :)
[03:52:26] <Tpenta> seriously, that's what i thought it was when we were using one as a demo at a SOSUG
[03:52:33] <moazamraja> Tpenta: that's my point.
[03:52:37] <Tpenta> :)
[03:53:19] * Tpenta finally gets back to blogging with a simple "see dtrace is just another tool in a SA's toolkit" entry
[03:55:32] <Gman> hi Tpenta
[03:56:26] *** yongsun has joined #opensolaris
[03:58:08] <jbk> hmm.. i think i might have found a minor bug in the current sparc disassembler..
[04:01:35] *** Fullmoon has quit IRC
[04:02:47] *** rbrown has joined #opensolaris
[04:04:33] *** delewis has joined #opensolaris
[04:08:48] <jbk> any C gurus around?
[04:09:01] *** yongsun has left #opensolaris
[04:09:04] *** postwait_ has quit IRC
[04:09:15] <jbk> i can't seem to find any standards documentation on this
[04:09:28] <jbk> dealing with bit fields and how they're implemented
[04:11:19] <wesolows> umm, maybe
[04:11:24] <wesolows> it's kind of complex though
[04:11:48] <wesolows> somewhere I have the ISO C99 spec
[04:12:02] <wesolows> if your question is interesting enough I may open it up
[04:12:21] *** Shiv__ has joined #opensolaris
[04:12:33] *** xuewei has joined #opensolaris
[04:12:40] *** simford has quit IRC
[04:14:17] <jbk> well basically just if you did say 'struct foo { int8_t a:5; int8_t b:2; int8_t 1; };'
[04:14:33] <jbk> (or uint8_t, i don't care about the sign too much)
[04:15:29] <jbk> does that mean it will essentially pack them into 1 8 bit quantity, in the order given?
[04:16:00] <jbk> (assuming the values are <= to a known integral type -- i.e. 8,16,32,or 64 bits)
[04:16:55] <jbk> the only thing i've found is that implementations are free to start with the msb or bit 0 if they want, but nothing beyond that
[04:18:00] <wesolows> yes, it will pack those into 1 byte
[04:18:13] <dlg> jbk: yes
[04:18:20] *** movement has quit IRC
[04:18:33] <wesolows> uint8_t a:5; uint8_t b:2; uint8_t c:1; is equivalent to uint8_t a:5, b:2, c:1; and both declare a single byte worth of storage
[04:18:41] <dlg> be careful with byte order though
[04:18:46] <wesolows> I'm not looking at the standard; I'm telling you from practical experience.
[04:18:59] *** movement has joined #opensolaris
[04:19:02] <wesolows> Right, you need to use _BIT_FIELDS_LTOH and company if this needs to be platform-neutral
[04:19:16] <wesolows> especially if you're communicating with hardware (like for SCSI)
[04:19:21] <jbk> so i wouldn't have to worry about it doing something funky like decide to put the value for a into bits 0,1,2,6,7
[04:19:32] <wesolows> no, definitely not.
[04:19:39] <wesolows> They're absolutely contiguous, always.
[04:19:47] <jbk> no padding?
[04:19:54] <wesolows> Never any padding.
[04:19:57] <jbk> ok
[04:19:59] <jbk> cool
[04:20:10] <jbk> that'll allow me to simplify and i think make this code a bit clearer
[04:20:13] <wesolows> Types larger than 8 bits are a little trickier, but not much.
[04:20:18] <wesolows> If you can stick to 8-bit types, you'll be golden.
[04:20:22] <jbk> well this would be 32-bit
[04:20:25] <jbk> acutally
[04:20:37] <wesolows> see, now you're gonna make me go open the spec
[04:20:52] <jbk> but always all the fields would add up to 32 bits total, never any left over
[04:20:55] <jbk> haha
[04:21:15] <wesolows> I think you're still ok there, I was more worried about ordering
[04:21:27] <wesolows> Also, if the structure as a whole needs to be packed, use pragma pack(1)
[04:21:40] <jbk> well it sounds like with the define, i'd be ok
[04:21:45] <wesolows> Otherwise you may get padding between bitfields and other members.
[04:21:50] <dlg> wesolows: structures made up of u_int8_ts shouldnt need to be explicetly packed
[04:21:56] <dlg> its a nice reminder to the read though
[04:21:59] <dlg> reader
[04:22:02] <wesolows> dlg: True, if there are no other members.
[04:22:03] <jbk> dlg: what about 32-bit?
[04:22:21] <wesolows> dlg: What if you have struct mumble { uint8_t a:4, ...; uint32_t blah; } ?
[04:22:34] <dlg> jbk: its likely that if its only u_int32_t members then it will be fine
[04:22:42] <wesolows> dlg: If you want that last bit adjacent to the first bit of the 32-bit member, you need pack
[04:22:43] <dlg> wesolows: pain pain pain
[04:22:47] <wesolows> yes
[04:22:50] <wesolows> much pain
[04:22:51] <jbk> i was thinking more uint32_t a:15, b:17; (for example)
[04:22:52] <wesolows> and much danger
[04:22:59] <wesolows> jbk: Those are ok
[04:23:09] *** yongsun has joined #opensolaris
[04:23:12] <wesolows> all right, spec opened
[04:23:54] <dlg> jbk: on strict alignment archs (and a lot of others too) the compiler will put members of a struct on the archs natural word boundary
[04:24:01] <dlg> ie, on every 4th byte
[04:24:15] <dlg> so structs like { u_int8_t; u_int32_t; }; will take up 64bits of space
[04:24:26] <dlg> cos its padded the second member to the 4 byte boundary
[04:24:27] <jbk> i'm not even worried about that
[04:24:52] <jbk> just if i have stuff encoded into a single 32-bit quantity, to use bit fields to reference them instead of doing a bunch of masks & shifting
[04:24:56] *** simford has joined #opensolaris
[04:25:03] <dlg> nod
[04:25:10] <dlg> just be careful of byte order then
[04:26:02] <wesolows> oh...yeah, then you're fine.  Bit and byte order are your only worries.
[04:26:18] <jbk> ok
[04:28:29] <wesolows> This is what you wanted to know:
[04:28:41] <wesolows> 6.7.2.1(10) reads "An implementation may allocate any addressable storage unit large enough to hold a bit-
[04:28:44] <wesolows> field. If enough space remains, a bit-field that immediately follows another bit-field in a
[04:28:47] <wesolows> structure shall be packed into adjacent bits of the same unit. If insufficient space remains,
[04:28:50] <wesolows> whether a bit-field that does not fit is put into the next unit or overlaps adjacent units is
[04:28:53] <wesolows> implementation-defined. The order of allocation of bit-fields within a unit (high-order to
[04:28:56] <wesolows> low-order or low-order to high-order) is implementation-defined. The alignment of the
[04:28:59] <wesolows> addressable storage unit is unspecified.
[04:29:15] * dlg prefer macros
[04:29:16] <wesolows> (11) and (12) are also useful to know.
[04:29:39] <wesolows> Eh, I hate them both.
[04:29:53] <wesolows> Bitfields are more of a pain to write headers for; macros are more a pain to use.
[04:30:40] <wesolows> One thing I'm doing in ON is introducing a DECL_BITFIELDn macro so that you can declare 8-bit bitfields without having to do the #ifdef _BIT_FIELDS_LTOH thing and basically type the whole damn thing twice.
[04:31:00] <wesolows> Too bad there's no way to reverse __VA_ARGS__ ;-)
[04:31:01] * dlg nod
[04:31:12] <wesolows> Then you wouldn't need the n.
[04:31:37] <dlg> yeah
[04:31:48] <dlg> you could provide a little meta lang
[04:31:52] <dlg> parse it into a .h
[04:31:57] <dlg> and then include the .h
[04:31:58] <wesolows> I've thought about that too.
[04:32:14] <dlg> messy though
[04:32:27] <wesolows> Preferably the meta-language would be something like SPC3r23.pdf
[04:32:44] <wesolows> extract the structures from the tables in the standards.
[04:32:44] <dlg> heh
[04:32:51] <dlg> pdf2h
[04:33:00] <wesolows> well, scsipdf2h
[04:33:11] <wesolows> or really t10pdf2h
[04:33:12] <dlg> pdf2h -o t10.org
[04:33:17] *** simford has quit IRC
[04:33:18] <wesolows> yeah
[04:33:34] <wesolows> it's hard but not impossible
[04:33:43] <dlg> you could teach teh preprocessor extra tricks
[04:33:47] <wesolows> it might even be worth doing
[04:34:44] *** postwait has joined #opensolaris
[04:36:20] <e^ipi> are there standards issues with vim ?
[04:36:39] <dlg> e^ipi: the colors make people have fits if you scroll too fast
[04:36:49] <e^ipi> so no, then?
[04:37:47] <wesolows> since there's no standard specifying vim, I don't see how there could be
[04:37:56] <wesolows> vim replacing vi, I don't know - probably
[04:38:01] *** delphi1000 has joined #opensolaris
[04:38:14] <wesolows> I'd heard there were a few test failures in that case, but that the test was run some time ago.
[04:38:16] <delphi1000> hi everybody
[04:39:15] <delphi1000> how to install tcpdump in solaris?
[04:39:24] <alanc> alias tcpdump snoop
[04:39:48] <dlg> delphi1000: blastwave
[04:43:41] <delphi1000> what?
[04:43:50] <delphi1000> use pkgadd?
[04:43:54] <wesolows> alanc++
[04:43:57] <delphi1000> or others
[04:44:09] <wesolows> You don't need it; use snoop.
[04:47:14] <comay> delphi1000: tcpdump is also available through the companion cd
[04:47:41] <jamesd> though i can't think of a reason why you would want to use it.
[04:48:19] *** karrotx_ has joined #opensolaris
[04:48:48] <alanc> I can understand wanting ethereal/wireshark over snoop, but what does tcpdump give you that isn't in snoop?
[04:49:13] <brendang> and ethereal/wireshark can read snoop output files
[04:50:12] *** simford has joined #opensolaris
[04:50:41] <comay> tcpdump does some protocols that snoop doesn't
[04:50:48] *** derchris has quit IRC
[04:50:59] <comay> btw alanc, thanks for the packaging list
[04:51:21] <comay> is it safe to assume that the pieces that are (or will be redistributable) will also be completely open source?
[04:52:03] <alanc> the ones that I listed as are already redistributable are - they all have sources on opensolaris.org already
[04:52:26] <alanc> the rest should be, except for the Sun logo images I mentioned
[04:52:41] <delphi1000> thank you
[04:53:38] <rbrown> snoop does some nfs-rpc stuff tcpdump doesnt
[04:54:36] <delphi1000> ok
[04:54:43] <delphi1000> thank you all the same
[04:55:47] *** danv12 has joined #opensolaris
[04:58:37] *** delphi1000 has quit IRC
[05:00:17] <CIA-27> xw161283: 6544847 nge driver should support MCP04 device
[05:03:12] *** postwait has quit IRC
[05:08:43] *** ShadowHntr has quit IRC
[05:13:26] *** laca has quit IRC
[05:13:33] <e^ipi> is there an openboot variable somewhere that I ought to be setting to make this sparc machine use the framebuffer instead of the bloody serial console?
[05:13:56] <Tempt> eeprom output-device=screen
[05:14:01] <Tempt> eeprom input-device=keyboard
[05:14:04] <Tempt> init 6
[05:14:19] <Tempt> What was the problem last night? Didja fix it?
[05:14:39] <e^ipi> that was this morning
[05:14:45] <Tempt> TZ
[05:14:51] <Tempt> Anyway, the fault was ..?
[05:14:52] <jamesd> e^ipi plug in a keyboard and it will use the framebuffer and the keyboard.
[05:15:08] <e^ipi> it wouldn't boot at all this morning, i dunno what was loose but i stripped it apart and put it back together and it worked
[05:15:25] <e^ipi> jamesd: not true... evidently it will use the keyboard for input and serial for output
[05:16:15] <jamesd> e^ipi, then someone messed with the obp settings...  reset them with stop-n  or what ever the command is to reset them to defaults.
[05:17:34] <Tempt> well, just set output-device to screen and you'll use the display
[05:17:37] <Tempt> Trust me ;)
[05:19:13] *** danv12 has quit IRC
[05:23:10] *** alanc is now known as alanc-away
[05:26:16] <BatonT> is there an easy way to get motherboard model/bios from solaris (so i dont have to reboot it or go looking for receipt) ?
[05:27:32] <Tempt> does prtdiag -v tell you anything?
[05:27:54] <BatonT> yep that did it :)
[05:28:02] <BatonT> i WAS trying to remember the command
[05:28:11] <BatonT> but i couldnt remember if it also dsplayed bios
[05:28:22] <BatonT> BIOS Configuration: Phoenix Technologies, LTD ASUS M2N-E ACPI BIOS Revision 0205 06/30/2006
[05:28:35] <BatonT> eer model
[05:30:57] <Tempt> I think whether or not it displays model depends on whether your manufacturer puts that in the ident fields or whatever.
[05:31:07] <BatonT> ok
[05:31:50] *** paul has quit IRC
[05:31:57] *** paul has joined #opensolaris
[05:32:00] <e^ipi> Tempt: seems not to have worked
[05:32:01] <e^ipi> *shrug*
[05:32:46] <BatonT> ok is there a cmd to show system temp etc?
[05:33:50] <Tempt> e^ipi: Cool. If it works, it works.
[05:33:53] <Tempt> BatonT: prtdiag -v
[05:34:04] <Tempt> e^ipi: Are you happy?
[05:34:27] <e^ipi> uhh....
[05:34:35] <richlowe> Tempt: don't think prtdiag and friends have sensor info on x86
[05:34:42] * Tempt shrugs
[05:34:47] <BatonT> Tempt: nah, that just shows cpu/mem/pci slots
[05:34:55] <richlowe> sure should be nice if they did though.
[05:34:58] <Tempt> I figured it might. It should.
[05:35:00] <BatonT> yeah
[05:35:02] <richlowe> 'would', not 'should'
[05:35:03] <e^ipi> i'm still not seeing anything on the monitor, so not really
[05:36:17] *** axle_512_ has joined #opensolaris
[05:37:17] <e^ipi> i mean, i guess in general, sure I'm happy
[05:37:30] <e^ipi> but about this particular issue, not so much
[05:38:39] <Tempt> I mean, you've got your new workstation up and running; that must bring some joy.
[05:38:48] <e^ipi> indeed it does
[05:38:57] <Tempt> I always expect machines that have been shipped (especially by ebay vendors) to need a cleanup when they arrive.
[05:39:05] <Tempt> I usually get them open and reseat everything before first powerup
[05:39:18] <Tempt> (is also a chance to make sure nothing is dodgy)
[05:39:27] <Tempt> What sort of framebuffer in that thing?
[05:39:40] <delewis> when I got my SB1000 in a few years ago, both procs died within a few weeks.
[05:39:48] <Tempt> That's awful.
[05:39:51] <delewis> fortunately, the vendor had put a 90-day warranty on the system.
[05:39:56] <Tempt> That's good.
[05:39:57] <e^ipi> two of 'em, a creator3d 2 and creator3d 3
[05:40:00] <Tempt> Oh man
[05:40:03] <Tempt> You need a new FB
[05:40:10] <Tempt> XVR 500/600/1000/1200
[05:40:12] <delewis> least its not an Elite3D. :-)
[05:40:21] <Tempt> Trust me, the performance difference is monumental.
[05:40:31] <delewis> 'monumental' is an understatement.
[05:40:36] <Tempt> Okay.
[05:40:41] <Tempt> fscking monumental
[05:40:55] <Tempt> It's like going from a CG3 to an FFB
[05:40:56] <Jondice> while we're on the subject ... i ordered an ultra 60 a couple of months ago off of ebay
[05:41:04] <delewis> my SB1000 is a new system since I've replaced my Elite3D with an XVR-1000.
[05:41:10] <Jondice> first sparc system i'd ever attempted using
[05:41:11] <Tempt> I used to have an Ultra-60 for my workstation.
[05:41:20] <e^ipi> xvr-500 is still like, $100
[05:41:21] <Tempt> It was right good.
[05:41:37] <Tempt> e^ipi: Got any hardware to barter for an XVR-600?
[05:41:38] <Jondice> well, I bought the appropriate VGA/keyboard converter and such, tried different monitors, but no luck
[05:41:38] <delewis> you can get an XVR-1000 for slightly more.
[05:41:41] <delewis> Usually $150 or so.
[05:41:45] *** eboutilier_ has quit IRC
[05:41:50] <delewis> very nice framebuffers.
[05:41:53] <Tempt> Jondice: Still got the machine?
[05:42:02] <delewis> I'll be buying a second one here in a few weeks for a second display.
[05:42:02] *** eboutilier_ has joined #opensolaris
[05:42:06] <Jondice> iirc, the system did play a sound, but as i could see nothing, i dunno what was up
[05:42:17] <Jondice> Tempt, yep, but not within 20 miles at the moment heh
[05:42:22] <Tempt> Jondice: Unplug the keyboard and everything and connect via the serial port
[05:42:27] <Tempt> Jondice: It might just need some configs
[05:42:29] <e^ipi> Jondice: hang a serial term. off it
[05:42:33] <Jondice> ok
[05:43:36] *** axle_512_ has left #opensolaris
[05:43:50] <BatonT> i DO wish motherboard manufactures would give a way to update bios's in something other than DOS or windows
[05:43:55] <Tempt> HAHA.
[05:43:56] <Tempt> Yes, well.
[05:44:14] <BatonT> when i have a solaris box with no optical drive...  flashing bios is a pain in the arse
[05:44:38] <BatonT> nor does it have a floppy drive :P
[05:44:59] <e^ipi> Tempt: i have a couple of these: http://www.axiontech.com/prdt.php?item=61019&PRICECOMPARISONSID=6d0b53a64f21cf7644a78ee19346b809
[05:45:17] <e^ipi> which i found inside a POWER3 machine I bought off ebay
[05:45:47] <BatonT> e^ipi: cool
[05:46:50] <e^ipi> BatonT: you could use a processor architecture that isn't... you know.... intel
[05:47:01] <BatonT> e^ipi: yeah well :)
[05:48:02] <BatonT> a T1 rack server isnt ideal as a home server (noise/cost/limited internal drives)
[05:48:07] <Tempt> e^ipi: I'm sure if you google for dialogic and suck you'll get lots of hits. They had a reputation of making sysadmins cry.
[05:48:24] <e^ipi> Tempt: *shrug*
[05:48:33] <Tempt> e^ipi: In other words, sell 'em off!
[05:48:43] <e^ipi> it's really the only hardware i've got that's any use for trade
[05:49:01] <e^ipi> plus, i didn't pay a dime for them other than the machine they came in
[05:49:10] <Tempt> Where abouts are you located?
[05:49:15] <e^ipi> canada
[05:49:28] <Tempt> I need someone in the OS to re-ship me some cheap hardware
[05:50:01] <wesolows> OS?  Oppressive State?  You'll have to be more specific.
[05:50:04] <Tempt> US
[05:50:05] <Tempt> OS
[05:50:07] <Tempt> same thing
[05:50:13] <Tempt> Oppressive State hits the mark
[05:50:22] <Tempt> I just want a cheap POWER4 machine to play with AIX6.1
[05:50:55] <wesolows> can't help you; I'm in San Francisco, not the US.  Oh, and I don't have a Power4 machine
[05:51:38] <BatonT> and im in australia...
[05:52:04] *** Doc has quit IRC
[05:52:49] <Tempt> Another Australian.
[05:52:55] <Tempt> We're taking over the Solaris community, bit by bit.
[05:53:36] *** Doc has joined #opensolaris
[05:53:49] <Doc> hmm.. screen just royally lost the plot
[05:53:49] <Tempt> And here's another one.
[05:53:51] *** duri has joined #opensolaris
[05:54:02] <Tempt> A bug? in screen? Surely you jest.
[05:54:19] <Doc> only lost one window
[05:54:23] <Doc>     root  2260     1   0   Apr 29 ?          14:42 screen
[05:54:42] <Doc> well over 3 months. not a record, but not bad :)
[05:55:37] *** Gman has quit IRC
[05:55:42] <Tempt> Has anyone raised a case or ARC or whatever to get screen shipped by default with Solaris?
[05:56:02] <jamesd> i was totally surprised that no one on the unix team at my new job knew what "screen" was...
[05:56:15] <Tempt> BTW, graysonline.com.au has a scsi/fibre channel hard drive auction at the moment
[05:56:21] <BatonT> jamesd: gee..
[05:56:26] <Tempt> Same here
[05:56:43] <Tempt> What really surprised me was I gave them a quick demo and a tutorial and installed screen around the place ...
[05:56:47] <Tempt> ... but none of them are using it
[05:57:04] <BatonT> the team i work with have no unix skills
[05:57:18] <wesolows> I didn't start using screen until a couple years ago
[05:57:28] <wesolows> that doesn't mean I was unskilled though
[05:57:35] <BatonT> and we have to look after 2 solaris9 boxes... and management wernt clever enough to get a few people that had solaris skills
[05:57:44] <BatonT> apart from me
[05:57:46] <jamesd> one guy lives on the console , in fact his is blind, and has 20 years of unix experience... and didn't know screen.
[05:57:59] <jamesd> er he is
[05:58:18] <BatonT> ive used screen since like umm..1996 ish
[05:59:19] <jamesd> me too, perhaps a bit longer, but i never really used some of the cooler features untill recently,  re-attaching, and split screen, i have only used for the last couple years.
[05:59:40] <wesolows> re-attaching is the only reason I use it, actually
[05:59:47] <Tempt> I'm not saying people are unskilled if they don't use screen, just that having screen around is so handy.
[05:59:55] <BatonT> wesolows: yeah..
[06:01:00] <Tempt> man, grays have a physical auction coming up. No catalog yet, but there's pictures of big black meaty RS/6000s. Perhaps there will be a POWER4 machine for me.
[06:01:25] <e^ipi> ahh, here we go...
[06:01:36] <e^ipi> I yanked one of the framebuffers and it works now
[06:01:53] <e^ipi> ebay it goes then I guess
[06:01:54] <Tempt> I thought it was working before?
[06:01:59] <e^ipi> hmm?
[06:02:04] <e^ipi> no, i mean i can get display now
[06:02:19] <e^ipi> evidently only if I have a single framebuffer installed
[06:02:19] <Tempt> I thought you had display working after setting output-device to screen?
[06:02:26] <e^ipi> nope
[06:02:33] <Tempt> I've definately had multiple framebuffers working
[06:02:36] <Tempt> Multi UPAs at that.
[06:02:43] <Tempt> My Blade-1000 at home already has multiple fbs
[06:03:03] <Tempt> Let me check to see if there's anything funky in the OBP setup on that host...
[06:03:16] <e^ipi> it's really not a big deal, I only have one monitor
[06:03:47] <Tempt> I've got output-device=jfb:r1280x1024x60
[06:03:58] <Tempt> Which tells it which fb to use and which res for OBP
[06:04:03] <Tempt> You may need to do the same.
[06:04:12] <e^ipi> jfb?
[06:04:15] <Tempt> xvr600
[06:04:26] <Tempt> java frame buffer? heh.
[06:08:16] <delewis> the XVR-1000 is a Java framebuffer :-)
[06:09:13] <bda> That sounds scary.
[06:09:28] <Tempt> It's got 4Gb of RAM to house all that java goo
[06:09:34] <e^ipi> it's a bit bloated but still not particularly slow, and fairly easy to work with?
[06:10:11] <delewis> XVR-1000 used the MAJC chip (a VLIW proc that Sun designed for Java computing)
[06:10:19] <delewis> XVR-4000 uses MAJCs, as well.
[06:10:21] <Tempt> Oh, that's right.
[06:10:42] <Tempt> Never managed to get my hands on an XVR-4000, and now I've got CPUs in that slot.
[06:10:46] <Tempt> Oh well, on my next 880
[06:11:01] <delewis> the XVR-4000 is basically 3 XVR-1000s on a single board, from my understanding.
[06:11:08] <Tempt> 4 XVR-4000s
[06:11:11] <Tempt> four majc chips
[06:11:26] <Tempt> and a load of crazy-arse stuff after them
[06:11:37] <Tempt> err, 4 XVR-1000s that is.
[06:12:06] <delewis> the XVR-1000 can still hold its ground as far as Sun FBs go, though. When the XVR-1200 was released, there wasn't a huge difference between it and the XVR-1000 in terms of performance.
[06:12:27] <delewis> and as far as image quality goes, its comparable to a modern Quadro.
[06:12:50] <Tempt> The XVR-4000 is meant to deliver amazing image quality.
[06:13:00] <Tempt> There's huge amounts of clever filtering/antialiasing stuff there.
[06:13:31] <delewis> that's what the XVR-1000 was designed for. Multi-sampling anti-aliasing (basically, what RenderMan does) and some other jazz.
[06:13:38] <delewis> 30-bit color and so fourth.
[06:13:43] <Tempt> That's the one.
[06:13:56] <Tempt> Now, why can't I have an XVR-4000? The world is CONSPIRING to stop me getting the toys I want.
[06:14:50] <e^ipi> damned stuff.... costing money
[06:14:56] <delewis> I wouldn't mind an XVR-2500.
[06:15:05] <Tempt> Isn't that PCI express?
[06:15:14] <delewis> unfortunately, I think.
[06:15:19] <delewis> the last UPA framebuffer was the XVR-1000.
[06:15:20] <Tempt> Yes, well.
[06:15:27] <Tempt> You'll have to get your Ultra-45 first.
[06:15:33] <delewis> and given all my Blade 1000 has is UPA and PCI, PCI isn't an option for a framebuffer, IMO.
[06:15:47] <Tempt> Actually, PCI works well enough.
[06:16:00] <delewis> ever read the UPA Whitepaper?
[06:16:03] <Tempt> Yes.
[06:16:23] <Tempt> And while I do like the UPA framebuffers, for my use the PCI 64bit/66Mhz is fine
[06:17:06] <delewis> hopefully when I do get the cash again, Sun will have another sale equivalent to th 20th anniversary sale. :-)
[06:17:15] <delewis> the base Ultra 45 config. was actually cheaper than the Ultra 25 config.
[06:17:16] <Tempt> Pity they didn't cover Australia with that sale.
[06:17:55] <Tempt> I think IBM should have a sale. Specifically for me. Called the Tempt wants a POWER machine sale. With an entry level machine for around $1k
[06:17:56] <delewis> and AMD offerings do look tempting for a next workstation, though..
[06:18:17] <Tempt> Perhaps I can trade in my E3000 ;)
[06:19:58] <delewis> it just doesn't look like Sun is doing a whole lot of R&D nowadays on the SPARC workstation side, given I suspect the majority of customers are purchasing the next generation of workstations merely for the purpose of compatibility/development.
[06:20:07] *** rbrown has quit IRC
[06:20:25] <delewis> shame, that.
[06:20:30] <Tempt> I understand some medical applications and apps specific to the oil/gas industry use SPARC workstations
[06:20:41] <Tempt> As well as some vertical apps for military/intelligence
[06:21:00] <delewis> that's what I meant by compatibility/development (the inability to move the application to AMD64 for whatever reason -- lack of development resources, etc.)
[06:21:15] <Tempt> Lack of desire to port is a big one
[06:21:31] <Tempt> Why bother porting your $200,000/seat application to a cheaper CPU architecture?
[06:21:52] <delewis> sure, the cost offset between the Ultra 45 and the Ultra 40 is minimal compared to the cash that'd be required for porting resources.
[06:21:55] <Tempt> I used to hear people saying that being SPARC-only was a cheap and inoffensive anti-piracy measure
[06:21:57] <delewis> (validation, manpower, etc.)
[06:22:20] <delewis> Tempt, that's understandable.
[06:22:39] <Tempt> Keep it on SPARC and it means anyone pinching a copy will have to have the hardware.
[06:22:52] <delewis> with the shift of x86 vendors going to EFI, I imagine some institutions will be looking elsewhere, too.
[06:22:54] <Tempt> There's not exactly a load of torrents all over isohunt with leet solaris warez
[06:23:05] <delewis> EFI isn't an option for institutions that need some level of security.
[06:23:24] <kaiwai> delewis: IIRC alot of the new BIOS's are already based on EFI/UEFI
[06:23:28] <delewis> (LANL is refusing to use any EFI hardware)
[06:23:33] <delewis> citing security reasons.
[06:23:48] <delewis> LANL is the Los Alamos National Laboratory.
[06:26:04] <Tempt> FYI Australians, my employer is hiring for some Java developers, full time, $AU87k-$AU110k
[06:26:16] <delewis> any hardware that's not LinuxBios capable just isn't purchased. I imagine financial institutions will avoid it for the same reasons, as well (hopefully)
[06:26:24] <delewis> may see a sudden rise in SPARC workstations again. :-)
[06:26:41] <kaiwai> delewis: and SGI will make $1billion in profits too
[06:26:42] <Tempt> What's the security issue with EFI?
[06:26:47] <Tempt> I'm x86 ignorant ..
[06:26:48] *** sponix has joined #opensolaris
[06:26:55] <kaiwai> and SCO will stop litigating and actually make a decent product
[06:26:56] <delewis> Tempt, its a combination of intent and the fact it has the capabilities of OBP for the most part.
[06:27:01] <Tempt> kaiwai: SGI's stock price has increased steadily since they re-orged.
[06:27:13] <delewis> EFI can trap any IO operation and copy it into a buffer to do some DRM magic, and refuse the IO request if necessary.
[06:27:18] <sponix> jamesd !
[06:27:21] <Tempt> Oh dear.
[06:27:30] <delewis> and because EFI has a full networking stack... you can only imagine what can be done with that sort of intent.
[06:27:46] <delewis> and EFI isn't open. Any EFI vendor has to obtain a closed, binary module from Intel.
[06:27:55] <delewis> and then they can use the 'open' bit of EFI.
[06:28:02] <sponix> jamesd : know something funny, I didn't know that dclarke did most of the blastwave stuff until recently
[06:28:13] <sponix> jamesd : do you remember me ?
[06:28:29] <kaiwai> Tempt: when you slice and dice; you can do alot but you need the products coming out of R&D to have a competitive advantage; cutting R&D might cut costs but it affects long term competitiveness
[06:28:34] <jamesd> yes, but i'm heading off to bed, have to go to work in the morning
[06:29:01] <Tempt> kaiwai: Hey, I'm not recommending you buy SGI stock, I'm just saying ... they're starting to become profitable again.
[06:29:02] <delewis> http://www.fosdem.org/2007/interview/ronald+g+minnich
[06:29:09] <Tempt> kaiwai: Now that they don't have an interesting product anymore (!)
[06:29:09] <sponix> jamesd:  ok, just wanted you to know I have using the crap out of your blog material lately
[06:29:12] <delewis> read the reponse to the last question (re: EFI)
[06:29:24] <jamesd> cool
[06:29:31] <e^ipi> profitability is step 1 to being able to do R&D
[06:29:42] <e^ipi> you can only burn money for so long
[06:29:52] <sponix> anyone else in here work on blastwave stuff ? I want to do updates to vlc (and its deps) pretty bad
[06:30:03] <e^ipi> sponix: #blastwave
[06:30:04] <kaiwai> delewis: mind you, one would purchase a customisabe firmware tool from Pheonix or something
[06:30:19] <sponix> e^ipi: aww, that would make since. Thanks
[06:30:40] <sponix> what do you all think about the project indiana ?
[06:30:59] <delewis> "I am not sure what the real intent of this design is, but is is a real concern in secure environments (such as those found in governments, banks, and large search engine companies). A number of vendors and users have told me that they are not sure they can ship an EFI system they are willing to trust in a secure environment."
[06:31:09] <e^ipi> project "keep ian murdock busy"
[06:31:45] <sponix> e^ipi: think he is good for Sun as a whole ?
[06:31:54] <e^ipi> not at all
[06:32:13] <richlowe> I think it's even.
[06:32:15] <e^ipi> he's done more to dilute the Solaris brand since he's joined than millions of linux zealots could've done
[06:32:26] <richlowe> to the extent that I think it hard to believe he's worse than the present alternatives.
[06:32:43] <sponix> e^ipi: well, he is a Linux zealot -- remember
[06:32:59] <e^ipi> from my vantage, I think he's incompetent at best, and hostile to OpenSolaris at worst
[06:33:12] <Doc> blah.. they should just make solaris GPL, migrate ZFS and D-trace to Linux, and then drop solaris completely
[06:33:41] <e^ipi> didn't michael dell say something to that effect regarding Apple a few years ago?
[06:33:44] <sponix> Doc: Not really my opinion
[06:33:47] <Tempt> Can't someone local to Ian have a "conversation" with him. Outside. With a bat?
[06:33:52] <delewis> DTrace and ZFS could be migrated if the users of the GPL would actually read it.
[06:34:09] <delewis> no need to make ON GPL, though, I did detect some sarcasm in that statement.
[06:34:12] <delewis> :-)
[06:34:34] *** jamesd has quit IRC
[06:34:34] <e^ipi> the "layering violation" nonsense is just sour grapes
[06:34:43] *** jamesd_ has joined #opensolaris
[06:34:50] <Doc> sarcasm? what sarcasm?
[06:35:05] <delewis> the only restriction is that the ZFS and DTrace bits would need to be shipped separately from the Linux kernel sources, given the GPL is a source-level license only.
[06:35:13] <brendang> I wouldn't be so quick to judge -- lets see what code Indiana produces and then decide.
[06:35:24] <e^ipi> brendang: first they'd have to produce code
[06:35:38] *** lloy0076 has joined #opensolaris
[06:35:45] <kaiwai> hmm, those macbooks look tempting
[06:35:46] <brendang> e^ipi: yep
[06:35:54] <lloy0076> Anyone know of a scheme interpreter that will run nicely on SXCE b69 (x86)?
[06:36:01] <delewis> kaiwai, those Macbooks have EFI..
[06:36:05] <e^ipi> but from the sounds of it they're just going to shove the GNUserspace on top of sunOS and sun_libc
[06:36:15] <e^ipi> which is just utterly crap
[06:36:16] <kaiwai> delewis: and a nice operating system to boot
[06:36:25] <sponix> I don't see why they are going with a whole new project, you have blastwave for packages, and you have that Nexenta wanna be linux distro already
[06:36:26] <delewis> I wasn't aware the goals of Indiana had been stated in a clear form, yet.
[06:36:36] <Tpenta> e^ipi: if that's your understanding of indiana, then you need to read some more
[06:36:51] <sponix> delewis: there is a "Project Statement" posted now
[06:37:02] <Tpenta> hopefully I will shortly have glynn's talk on it up with the slides
[06:37:29] <kaiwai> delewis: from what I understand, it'll be the ubuntu/fedora of solaris
[06:37:33] <Tempt> brendang: Is there actually any code supposed to be coming from Indiana?
[06:37:45] <Tpenta> indiana is about process
[06:37:52] <Tempt> The only purpose I can see for Indiana at this stage is giving The Register something to write about.
[06:37:52] <Tpenta> for the most part
[06:37:52] <sponix> Tpenta: I caught the terms usability gaps closed, as in making installation and package management easier, more programs readily available etc
[06:37:53] <e^ipi> whatever, Indiana's being touted as a "community" distro, but if there's any contention on anything they won't even vote on it
[06:38:04] <delewis> its been noted by several others that Indiana could become a Solaris release.
[06:38:05] <e^ipi> so I really don't care about sun's new toy project
[06:38:13] <delewis> at least in the future, anyway.
[06:38:13] <brendang> Tempt: I hope so. either directly or indirectly.
[06:38:17] <Tempt> All they're doing is alienating the existing userbase at this point.
[06:38:19] <Tpenta> indiana is will be a distro
[06:38:20] <kaiwai> Tempt: how so? if it makes Solaris easier, bring smore developers and more contributors, its all good in the end
[06:38:21] <e^ipi> Tpenta: what process?
[06:38:28] <e^ipi> "ian says this, so it's the law" ?
[06:38:29] <Tempt> kaiwai: Horseshit.
[06:38:56] <Tpenta> cherry picking process that helps to produce a regularly updated distributin in a way that is acceptable to end users and developers
[06:38:59] <e^ipi> ian-diana's a joke and all it's doing is making solaris look bad
[06:39:14] <sponix> didn't mean to start a war
[06:39:15] <Tpenta> ernst, seriously you *need* to read more about it before canning it
[06:39:30] <Tpenta> you are making supposition here and calling it fact
[06:39:37] <e^ipi> what supposition ?
[06:39:37] <Tempt> Dude.
[06:39:38] <Tempt> It's IRC
[06:39:42] <e^ipi> ask about KDE
[06:39:45] <Tempt> It isn't meant to be a statement of purest fact.
[06:39:52] <e^ipi> you'll be shot down to all hell, Ian says GNOME, so that's the law
[06:40:01] <e^ipi> no votes
[06:40:02] <sponix> Tpenta: I just got warm and fuzzy when I saw xfce listed
[06:40:03] <kaiwai> hmm, but there is no KDE for Solaris
[06:40:05] <e^ipi> no discussion, just GNOME
[06:40:06] <eboutilier_> The new Solaris packaging and installation project is what will single-handedly be Indiana. Period. IMO.
[06:40:12] <kaiwai> the current KDE available is old, out of date and ugly
[06:40:13] <e^ipi> kaiwai: KDE builds just fine
[06:40:18] <e^ipi> just not with studio
[06:40:31] <kaiwai> e^ipi: in terms of packages
[06:40:40] <sponix> kaiwai: kde build is simple compared to gnome imho
[06:40:41] <eboutilier_> Other than that Indiana is basically a new release program and an acceleration of the /usr/gnu ARC case.
[06:40:42] <Tempt> kaiwai: Don't complain, make new packages
[06:40:54] <sponix> kaiwai: KDE is also listed for Indiana projects
[06:41:11] <e^ipi> sponix: yes, as an afterthought
[06:41:29] <e^ipi> SuSE users use KDE, Ubuntu users use GNOME
[06:41:36] <Tempt> I want enlightenment to be the core desktop environment.
[06:41:36] <e^ipi> people use what ships with the distro
[06:41:39] <sponix> e^ipi: xubuntu was an afterthought as well, and I still like it
[06:41:40] <Tempt> That sounds contrary enough.
[06:41:47] <kaiwai> sponix: just in the past I've found things fail to compile when trying to compile KDE
[06:41:54] <e^ipi> and KDE is a much nicer DE than GNOME from a usability standpoint
[06:41:56] <bda> Tempt: e16.5 or e17?
[06:41:59] <eboutilier_> Re: KDE, Moinak (BeleniX lead) said in no uncertain terms that BeleniX would basically become "Kindiana"
[06:42:19] <Tempt> bda: Ooh, let's have a big loud debate about it!
[06:42:38] <bda> e17 is full of shiny but doesn't work very well. Let's just use that.
[06:42:39] <sponix> kaiwai: gcc gmake automake autoconfig and a few other gnu tools need to be up front in your path for KDE to build
[06:42:44] <e^ipi> regardless, the actual choice of DE is irrelevant, so much as the fact that whoever's in charge of Iandiana seems entirely unwilling to consult the community in any meaningful fashion about it
[06:42:57] <e^ipi> so ultimately all indiana is is yet another Sun product
[06:43:00] <bda> e^ipi: ho ho, I see what you did there. "Ian-diana."
[06:43:13] <e^ipi> bda: it's his distro, it should carry his name
[06:43:38] <kaiwai> let him have his pet project - it isn't as though he can't do too much damage
[06:43:47] <e^ipi> he's already doing damage
[06:43:53] <Tempt> dba: Excellent! I like the useless shiny! GNOME isn't shiny enough (although it *is* useless enough), so making things shinier wins my vote.
[06:43:54] <eboutilier_> e^ipi: I don't think the Indiana leads will consider it a success if it's seen as "another Sun product"
[06:44:01] <kaiwai> I mean, compared to the 15months of revenue decline thanks to the SPARC bigots holding the reigns of development at sun
[06:44:02] <e^ipi> "let's make solaris more like linux" reads like "linux is awesome and solaris sucks, let's parrot them"
[06:44:04] <bda> Tempt: Man, I don't even *like* databases.
[06:44:21] <Tempt> eek.
[06:44:22] <Tempt> Heh.
[06:44:35] <Tempt> Oh yes, you do, you've had Oracle installed on your brain overnight.
[06:44:43] <LeftWing> And then your head exploded.
[06:44:59] <bda> I was wondering why I woke up with meat caked on the walls again.
[06:45:19] <e^ipi> so i've given up believing that indiana is more than just a distraction from a loudmouth Linux bigot
[06:45:55] <Tempt> Hmm, right, see, this is the problem.
[06:46:08] <Tempt> OpenSolaris community (and Sun) are beginning to look like a pack of wankers over this.
[06:46:09] <sponix> e^ipi: To me Solaris is about Stability and Throughput, things like zones and zfs. But why should adding a -v (verbose) option to cp ruin anything *Grin*
[06:46:18] <Tempt> Well, by beginning, I mean "for the last 6 months"
[06:46:29] <bda> Indeed.
[06:46:32] <kaiwai> e^ipi: and if Indiana results in more mind share - would you be willing to have a slice of this humble pie I've just baked?
[06:46:41] <richlowe> I wonder if I can get "pack of wankers" on a business card.
[06:46:50] <richlowe> it kinda needs a membership# to fit right though.
[06:46:51] <e^ipi> kaiwai: if it ruins everything that makes solaris good, no... I'll just buy more IBM kit
[06:46:57] <bda> richlowe: Lead Wanker. Pack of Wankers Dept.
[06:47:12] <Tempt> e^ipi: Exactly.
[06:47:22] <e^ipi> the kernel & libc aren't what makes solaris rock-solid, it's the whole OS
[06:47:23] <Tempt> The thing is, there's a lot of people who actually like Solaris.
[06:47:38] *** jamesd_ has quit IRC
[06:47:39] <Tempt> The current move seems to be about changing the "feel" of Solaris to make it more luser-friendly.
[06:47:47] *** jamesd__ has joined #opensolaris
[06:48:23] <kaiwai> who said anything about 'ripping out solaris components and replacing them with gnu userspace'?
[06:48:32] <Tempt> And you know, the people who like Solaris are typically the people who work with it. Every day. In production. On Sun hardware.
[06:48:41] <eboutilier_> sponix: The way the /usr/gnu ARC case works for commands like cp that collide, is the user will have to put /usr/gnu/bin ahead of /usr/bin in their path.
[06:48:46] <bda> Yay us people!
[06:49:01] <Tempt> The people who spec the hardware, decide the platform, architect environments.
[06:49:02] <sponix> kaiwai: that was more Nexenta project, the complete gnu/debian style userspace && solaris kern
[06:49:09] <Tempt> Also known as ...
[06:49:14] <Tempt> "The Paying Customer"
[06:49:14] <eboutilier_> IOW, /usr/bin/cp won't have -v, never will. But /usr/gnu/bin/cp will.
[06:49:16] <e^ipi> i like solaris, and I don't "work" (as in production) with it every day, but it's my primary OE
[06:49:23] <Tpenta> It is also worth pointing out that one of the very first slides tha gman shows when talking about this is the aim of maintaining backward compatibility above all else
[06:49:26] <e^ipi> and I only recently acquired sun hardware
[06:49:27] <richlowe> I think at this point it's clear that indy's target audience is some BSD hacker named 'Moonbeam'.
[06:49:40] <bda> haha.
[06:49:41] <richlowe> though I guess the bug text with that in it is hidden, and I can't recall who pasted the interesting bits :\
[06:49:44] <Tempt> Sun could turn Solaris into a penguin kiddie playground and get another 2 million fucking users. In their bedroom. On their dell laptop.
[06:49:55] <Tempt> Not buying support contracts, not buying hardware, not even working in the industry.
[06:49:56] <sponix> eboutilier_: yeah, or just ln the ones you like to another /early/path/bin, that way you can use some gnu, and some native
[06:49:57] <e^ipi> Tpenta: slides != actions
[06:49:58] <Tempt> Big Win.
[06:50:15] <Tempt> While IBM rolls out features in AIX that are juicy enough that Solaris diehards are getting interested.
[06:50:32] <LeftWing> I think the major fallacy is that a lot of people believe you can't have Both demographics.
[06:50:35] <sponix> eboutilier_: see, in my mind, the _alternative_ cp, ls, blah isn't bad, whats wrong with _Options_ ?
[06:50:44] <bda> Pack of wankers is basically true. People working on OSOL *right now*, both inside and outside Sun, are still going Good Work. Regardless of this other nonsense.
[06:50:49] <LeftWing> Whereas I imagine that if you're careful, you can have both.
[06:50:49] <Tempt> richlowe: Moonbeam FTW!
[06:50:52] <eboutilier_> sponix: No, only the commands that collide. There will be tons of stuff, e.g. wget, etc. etc. etc. going into /usr/bin because there's no name collision
[06:50:57] <sponix> Tempt: not I, never touched AIX
[06:50:57] <bda> And seeing as how this other nonsense is going to be *built on* the Good Work, who cares.
[06:50:58] <LeftWing> But that's probably another story and I need to go to work.
[06:51:00] * LeftWing vanishes.
[06:51:11] <richlowe> Tpenta: you could find it, which is bmc's who(1) bug with Moonbeam?
[06:51:18] <richlowe> Tpenta: also, paste it. :)
[06:51:18] <delewis> Tempt, who, to add, don't contribute anything and hardly use their computers for anything more than playing MP3s, look at dark colors, and try out new schedulers for 'glxgears' performance.
[06:51:27] <eboutilier_> I'm not talking about  Indiana by the way. I'm talking about Nevada/Solaris 11 (and therefore Indiana).
[06:51:36] <sponix> eboutilier_: yeah, I understand that, for things that don't have a native build, why not
[06:52:34] <Tempt> And the more changes there are between Solaris 10 and the next stable release, the longer the certification time, the longer it'll stay in dev and not prod, the longer the lag.
[06:52:41] <Tempt> Which is the worse for Sun.
[06:52:42] <eboutilier_> Not just native build, the default now is that *all* FOSS goes in /usr/bin, except the relatively small percentage where there's already a command by that name there.
[06:53:09] <sponix> eboutilier_: and you feel that's bad ?
[06:53:14] <eboutilier_> sponix: ?
[06:53:19] <delewis> what is FOSS?
[06:53:26] * BatonT tries again to replace 400gb drives in zpool with 500gb now that warrenty replaced 500gb drives just arrived
[06:53:31] <delewis> ON is "FOSS" and its in /usr/bin, so what's the distinction here?
[06:53:31] <Tempt> Alright, that does it, coffee and cigarette time. I'll see if I can find a linux-using emo to bitchslap while I'm out.
[06:53:32] <bda> An L away from something hygenic.
[06:53:35] <eboutilier_> I feel that's superbly wonderful.
[06:53:36] <sponix> delewis: Free/OpenSourceSoftware
[06:53:43] <sponix> delewis: gnu/gpl/blah
[06:53:52] <delewis> sponix, then say its GNU/GPL/etc.
[06:53:53] <delewis> not FOSS.
[06:54:03] <delewis> that's *your* definition of FOSS, and not necessarily everyone else's.
[06:54:08] <sponix> delewis: don't kill the messenger, its just another Buzzword
[06:54:11] <delewis> "GNU freeware will go in /usr/bin"
[06:54:16] *** Shiv__ has quit IRC
[06:54:18] <richlowe> dear me, I hope it's not a buzzword.
[06:54:23] <richlowe> it'll find it's way into JES's next name :\
[06:54:29] <bda> Hahaha.
[06:54:50] <sponix> delewis: hmm, I'm not the wikipedia of the term FOSS, but I'm sure it is defined there
[06:55:01] <sponix> was just trying to help ... you asked :P
[06:55:05] * bda "fixes" it.
[06:55:12] <delewis> and I'm sure the people that created the Wikipedia FOSS article are from the GPL cesspool.
[06:55:26] <eboutilier_> FOSS is used commonly enough I think: http://en.wikipedia.org/wiki/Free_and_open-source_software
[06:55:40] <delewis> who, mind you, don't actually the license they use.
[06:55:43] *** jamesd__ has quit IRC
[06:55:52] <delewis> understand*
[06:55:52] <lloy0076> errrk
[06:55:54] *** jamesd__ has joined #opensolaris
[06:56:01] <lloy0076> FOSS sounds like a muppet character's name.
[06:56:09] <lloy0076> FLOSS is what you clean your teeth with.
[06:56:10] <lloy0076> :(
[06:56:13] <sponix> delewis: sorry to have came from that cesspool. And I honestly don't give a crap what license things are under
[06:56:25] <e^ipi> sponix: then you don't come from that cesspool
[06:56:40] <e^ipi> the cesspool being " GPL is the /only/ license.... ever.... "
[06:56:41] <bda> lloy0076: Manamana?
[06:56:51] <sponix> I do love to see the code, even if I can't read it, it makes me sleep better at night
[06:57:12] <lloy0076> bda: Manamana?
[06:57:21] <delewis> that has little to do with the GPL, and moreso, the misguided interpretation of the GPL spawned by the FSF>
[06:57:22] <bda> lloy0076: Manamana!
[06:57:25] <delewis> s/>/./
[06:57:26] <sponix> e^ipi: that's retarded, there are plenty of licenses that are good, BSD, Apache has its own, Mozilla has one
[06:57:40] <bda> lloy0076: http://youtube.com/watch?v=KC9FtLQJoGM
[06:57:51] <e^ipi> sponix: CDDL (opensolaris' license) is a slightly modified Mozilla license
[06:58:01] <lloy0076> bda: LOL
[06:58:08] <e^ipi> which is itself an offshoot of the Apache license
[06:58:19] <lloy0076> bda: Serious, I can hear them singing FOSS/FL-LO-LO-SS
[06:58:24] <bda> lloy0076: That's what I hear when people argue licenses. Manamana!
[06:59:00] <lloy0076> bda: Gawd, and that organge haired guy even looks like Stallman with a hair colour change
[06:59:12] <bda> lloy0076: Don't ruin it for me.
[06:59:13] <delewis> oh, god.
[06:59:14] <bda> It's all I have.
[06:59:19] * delewis passes around vommit bags
[07:00:21] <bda> "The question is, what IS a manamana?" "The question is, WHO CARES?"
[07:01:56] <e^ipi> by the way, is my framebuffer ( FFB2+ ) one of the ones that will require I use XSun ?
[07:03:18] <delewis> no.
[07:03:21] <delewis> Xorg supports FFB.
[07:04:04] <delewis> the only framebuffers Xorg doesn't support are the custom Sun FBs made within the last 7-8 years.
[07:04:43] <delewis> afb is mostly supported, though, it does require that custom Sun firmware be loaded upon boot each time for certain capabilities to work.
[07:05:39] *** slowhog has joined #opensolaris
[07:07:07] <e^ipi> alan was mentioning something about a large pile of sparc graphics drivers that he's fighting with the graphics team over
[07:07:27] * Tempt tries to imagine compiz running with FFB2+
[07:07:52] <delewis> I try not to imagine compiz, period.
[07:08:38] * bda prefers his DE not act like an old episode of Batman.
[07:08:43] <bda> Bang! Zoom! Woosh! Pow!
[07:10:06] <e^ipi> some features of OSX like expose are good, because they're flashy, but stop right before they get gaudy
[07:10:15] <Tempt> Expose is great.
[07:10:19] <e^ipi> compiz pushes that over the edge and is just so flashy it looks tacky
[07:10:26] <dlg> expose exists cos alt-tab (apple-tab) is broken in macos
[07:10:33] <bda> I've never gotten into the habit of using Expose. Too used to %h.
[07:10:37] <bda> dlg: ?
[07:10:41] <BatonT> during a zpool replace... why is zfs writing of the drive that is being replaced?
[07:10:50] <dlg> it switches between apps, not windows
[07:10:55] <bda> dlg: cmd-~
[07:11:00] <bda> Also, shift-cmd-~
[07:11:15] <dlg> bda: still harder than on other wm things
[07:11:18] <bda> uhm?
[07:11:35] <Tempt> Expose is good though. Easy to just check on something with a single keystroke and then go back to what you're doing.
[07:11:42] <bda> No, it's not. It's easier. You have more granular control. You can switch between entire applications, and windows inside those applications.
[07:11:51] <Tempt> ie, "have any new logs turned up lately in that tail window?"
[07:11:58] *** Drone has quit IRC
[07:12:00] <Tempt> "has my burn finished?"
[07:12:14] <bda> Tempt: Used Spaces yet? It is basically not awful.
[07:13:51] <Tempt> I'm really not up on GUI technologies
[07:14:07] <bda> Virts in Leopards. shrug.
[07:14:11] <bda> er, s/ds/d/
[07:14:39] <Tempt> I'm not really use my Mac anymore
[07:14:42] <bda> You can bind applications to desktops, which is nice. But not, say, group windows of an application, or whatever.
[07:15:06] <bda> So I ended up cloning Terminal.app for my "chat" desktop,... which was dumb.
[07:15:13] *** pg_user has joined #opensolaris
[07:15:19] <pg_user> hi
[07:15:25] <dlg> bda: if you go to the spaces view
[07:15:30] <dlg> you can drag windows between spaces
[07:15:35] <dlg> so you can split apps over spaces
[07:15:36] <bda> dlg: uh, I know.
[07:15:46] <bda> dlg: But switch to an application and create a new window for it.
[07:15:49] <pg_user> hello.
[07:16:00] <pg_user> i compiled postgresql using sun studio 12
[07:16:04] *** p-a-venk has left #opensolaris
[07:16:05] <bda> You can get kicked to another virt.
[07:16:21] <e^ipi> pg_user: rockin'.
[07:16:51] <pg_user> I'm confused though, because it seems to be 15 % faster than gcc
[07:16:59] <e^ipi> why's that confusing?
[07:17:00] <bda> Heh.
[07:17:01] <pg_user> in execution
[07:17:14] <e^ipi> studio's got better code generation
[07:17:15] <pg_user> I think I must have done something wrong in the bench marks
[07:17:25] <e^ipi> no, gcc just sucks
[07:17:35] <pg_user> because when I ran pgbench, gcc should have come out on top
[07:17:51] <e^ipi> gcc is portable as all hell, and for that I'm thankful
[07:17:52] <pg_user> because gcc is opensource it should be faster
[07:18:29] *** libkeise1 has joined #opensolaris
[07:19:36] <pg_user> my co-worker compiled postgresql 8.2 on ubuntu using sun studio 12 and it was faster in pgbench than gcc.  Significantly faster
[07:19:46] <pg_user> he had to mess around with spinlocks though
[07:20:40] *** sponix has quit IRC
[07:21:06] <Jondice> is intel's ICC going to be ported to solaris also?
[07:21:25] <e^ipi> Jondice: with this new partnership that they announced a couple months ago, I hope so
[07:21:26] <Jondice> seems like there's some talk of new xeon systems from sun in the making
[07:21:30] <pg_user> why is gcc slower than sun studio 12?
[07:21:43] <e^ipi> pg_user: because gcc focuses on portability, not speed
[07:21:44] *** Drone has joined #opensolaris
[07:21:49] <FrostCS> Why is the sky blue?
[07:22:06] <e^ipi> studio only needs to generate x86 and SPARC on solaris ( and now linux ) code
[07:22:06] <pg_user> why does sun studio port to linux and is faster on linux?
[07:22:23] <FrostCS> Why don't ducks moo?
[07:22:25] <e^ipi> pg_user: because the architecture is still the same, the difference is the syscalls
[07:22:47] <pg_user> it seems strange though, is sun studio taking shortcuts?  I looked at the -fast switch.
[07:22:54] <e^ipi> gcc needs to generate code for almost all the architectures out there... the development effort is spread more thinly
[07:23:01] <e^ipi> pg_user: "shortcuts" ?
[07:23:19] <bda> You know, aliases.
[07:23:24] <e^ipi> plus the studio devs get paid, the gcc ones, spare time
[07:23:35] <e^ipi> though, given... with some exceptions
[07:23:43] <e^ipi> from IBM & RHAT
[07:25:20] <pg_user> shortcuts in the fact that sun studio may compile code that will fail in certain circumstances.  Read the man page for the --fast switch in studio 12
[07:26:29] <delewis> that usually just applies to floating-point exceptions.
[07:27:21] <Tempt> "<pg_user> because gcc is opensource it should be faster"
[07:27:48] <Tempt> That's a quote for the day
[07:28:02] <Tempt> Is that like "My machine is painted Ferrari red, of course it faster."
[07:28:12] <delewis> if you'll look in the #solaris channel, he's being taught what a mirror is.
[07:28:19] <e^ipi> Tempt: each "TypeR" sticker adds 3 horsepower
[07:28:23] <delewis> I wouldn't take too much stock in anything he's said over the last 15 minutes or so.
[07:28:32] <richlowe> Out of interest, are we talking about compile time, or the runtime of the result?
[07:28:58] <e^ipi> what in blazes can I do with this smartcard slot on my SPARC anyways?
[07:29:11] <richlowe> e^ipi: you could get smartcards and use them.
[07:29:24] <e^ipi> will the one on  my student ID work?
[07:29:27] <richlowe> or, you could watch rather confused acquaintences try and get a floppy in there.
[07:29:33] <richlowe> except now floppies are uncommon, that probably won't happen.
[07:29:34] <delewis> e^ipi, possibly.
[07:29:41] <delewis> there's a specification that most Smartcards follow.
[07:29:52] <e^ipi> but what purpose would it serve?
[07:29:57] <e^ipi> logins & so forth, or ?
[07:30:11] <delewis> authentication and you can program Smartcards, as well (useful if you have a network of say, Sun Rays)
[07:30:42] <e^ipi> which I do not
[07:30:50] <Tempt> I use the smartcard slot in my Blade-100 for storing business cards
[07:31:00] <e^ipi> heh
[07:31:13] *** libkeise2 has joined #opensolaris
[07:31:19] <Tempt> I use the smartcard slot in my Blade-1000 for storing my spare prox pass.
[07:31:29] <e^ipi> maybe I'll use it to store taffy
[07:31:34] <e^ipi> or gum
[07:31:48] <Tempt> And yes, you can use it for login, and you can get it working with the $2 generic smartcards if you can be bothered.
[07:31:53] <Tempt> taffy!
[07:32:00] *** Triskelios has joined #opensolaris
[07:32:24] <e^ipi> taffy is good...
[07:32:29] <Tempt> rm of / is not allowed
[07:32:32] <Tempt> How crap is that ...
[07:32:45] <bda> rm it harder.
[07:33:11] <Tempt> rm -rf /* works
[07:33:23] <bda> Good man.
[07:34:48] *** libkeiser has quit IRC
[07:36:28] <Tempt> Haven't done that in a while.
[07:36:44] *** libkeiser has joined #opensolaris
[07:37:34] *** libkeise1 has quit IRC
[07:37:49] <WickedWicky> morning
[07:38:30] <Tempt> Mornin'
[07:38:33] <Tempt> How didja go with eterm?
[07:38:40] <WickedWicky> I have the packages for you
[07:38:47] <Tempt> Coolarez.
[07:38:55] <WickedWicky> or not
[07:38:58] <WickedWicky> you run sparc I assume?
[07:39:23] <WickedWicky> it's very easy to build though, with pkgtool
[07:39:40] <Tempt> Yeah, SPARC, of course.
[07:39:53] <Tempt> I'll have to take a look at pkgtool later then.
[07:40:10] <WickedWicky> it's very very easy
[07:40:16] <moazamraja> re
[07:40:21] <moazamraja> evening/morning all
[07:40:22] <WickedWicky> download that jds-cbe tool thingy
[07:40:41] <WickedWicky> edit the file so it wont install SUNWsmbaS
[07:40:47] <WickedWicky> run ./cde-install
[07:40:52] <WickedWicky> (whichi s the file you edit)
[07:40:55] <WickedWicky> and presto
[07:41:11] *** libkeiser has left #opensolaris
[07:41:34] <WickedWicky> if i can do it, you can do it
[07:42:37] <WickedWicky> I dunno how to setup a cross compiler with Sun Pro
[07:42:38] <delewis> this sounds like Oprah.
[07:42:43] <WickedWicky> or i'd make the packages for you
[07:42:46] <WickedWicky> well
[07:42:47] <delewis> WickedWicky, you don't.
[07:42:52] <WickedWicky> I am less fat that oprah
[07:42:54] <delewis> its not a cross-compiler.
[07:43:03] <WickedWicky> delewis: that's what i thought :)
[07:43:19] * WickedWicky wants to buy a sparc
[07:43:28] <e^ipi> so buy one
[07:43:34] <e^ipi> U60 is cheap
[07:43:46] <e^ipi> U80 less so, but still very affordable
[07:43:55] <WickedWicky> yea, I'll browse around
[07:44:12] <pg_user> ouch, I just got booted from #solaris
[07:44:21] <pg_user> they are very tough in there
[07:44:37] <e^ipi> avoid the U5, U10 and blade100/150 apparently
[07:45:01] <WickedWicky> wait till Tempt turns gangsta on you, pg_user, then you'll meet tough
[07:45:03] <e^ipi> I've not have the mispleasure of having to use a u5 or 10, but the blade 150 sucks balls
[07:45:28] <pg_user> I thought the bsd groups were tough!
[07:45:31] <WickedWicky> I might ask my boss to donate me a netra 5 which is collecting dust for the last year (LITERALY) in the spare-parts department
[07:45:44] <WickedWicky> Tempt is the gangstaest gangsta of all gangstas
[07:46:25] <WickedWicky> and he calls me american at random... so :s
[07:46:50] <Tempt> Keep trolling; I'm reloading.
[07:46:59] <WickedWicky> just teasing!
[07:47:00] <e^ipi> WickedWicky: i'll assume you mean ultra5, in which case leave it where it is
[07:47:23] <pg_user> i have a question about a netra 105
[07:47:40] <pg_user> what happens when I don't have the lom password?
[07:47:41] <Tempt> Actually, Netra T1s have questions about you.
[07:47:42] <WickedWicky> no
[07:47:42] <WickedWicky> netra 5
[07:47:42] <WickedWicky> the flat grey thing
[07:47:42] <WickedWicky> 1u
[07:47:51] <e^ipi> oh, the netra 105
[07:47:53] <e^ipi> ?
[07:47:56] <Tempt> They wish you'd just leave them alone. And if they don't let you in, don't touch.
[07:48:05] <Doc> netra 5.  AKA SS5
[07:48:16] <Tempt> An uglier SS5.
[07:48:25] <pg_user> NVRAM has to be pulled?
[07:48:26] <Tempt> Who would have though it possible?
[07:48:36] <Doc> do you own a boat?
[07:48:37] <e^ipi> older than i've ever bothered to look in to then I guess
[07:48:40] <Tempt> pg_user: You can reset the password by power surging the input drivers.
[07:48:53] *** libkeise2 has quit IRC
[07:48:57] <pg_user> how is that done?
[07:49:15] <pg_user> please forgive me if this is a stupid question
[07:49:20] <Tempt> Just flick the power on and off really quickly.
[07:49:21] <Doc> if you dont own a boat, then a netra 5 isnt going to be much use to you
[07:49:33] <Tempt> Do that about 100 times and it'll go *click* and reset the passwords for you.
[07:49:43] <richlowe> Doc: I'm assuming T1-105.
[07:49:44] <WickedWicky> ok so.. netra 5 is a bad idea and I should just let it collect dust?
[07:49:50] <richlowe> Doc: close to the same result though.
[07:49:51] <Tempt> WickedWicky: It'll run Sol9
[07:50:15] <pg_user> i have two netra T1 105 boxes
[07:50:32] <pg_user> should I pull the ram and 10K SCSI drive and put it into one?
[07:50:46] <pg_user> or would I be better off running them both?
[07:51:04] <Tempt> Just wire the PCI slots together and use them as a dual-CPU box
[07:51:18] <pg_user> are you teasing, or can you really do this?
[07:51:21] <richlowe> tape them under your arms, and try to fly.
[07:51:34] <WickedWicky> pg_user: he's gangsta but just.
[07:51:37] <palowoda> Tempt: How many shots you got in that thing.
[07:51:40] <e^ipi> pg_user: what do you think?
[07:51:59] <pg_user> who knows, it may be possible.  It is a RISC chip after all
[07:52:13] <pg_user> different architectures may allow different things.
[07:52:14] <Tempt> You can actually reduce the instruction set further for increased performance.
[07:52:27] <e^ipi> Tempt: the cut-pin mods?
[07:52:32] <Tempt> Yep.
[07:52:36] <WickedWicky> haha
[07:52:49] <Tempt> A little delicate on the T1, but still most achievable.
[07:52:56] <WickedWicky> and fool-proof
[07:53:03] <pg_user> question, why is there a chat for solaris, when everyone is supposed to be an expert?
[07:53:03] <WickedWicky> a dodo can do it
[07:53:04] <Tempt> Absolutely foolproof
[07:53:27] <WickedWicky> pg_user: to kill time
[07:53:36] <pg_user> if everyone reads the RFM, no point in chatting is there?
[07:53:41] <e^ipi> pg_user: because IRC is for like-minded people to get together and chat ?
[07:53:50] <e^ipi> not a support number for dumb questions
[07:54:05] <pg_user> i guess i'm used to the linux groups
[07:54:09] <WickedWicky> yea
[07:54:14] <WickedWicky> you're not getting torched there
[07:54:17] <e^ipi> only a subset of them
[07:54:24] <pg_user> they are desperate for any user no matter how stupid and degenerate
[07:54:36] * WickedWicky is on the LKML list, dont play trix0rs on me. If you get flaimed somewhere it's there
[07:54:41] <e^ipi> and the code quality shows
[07:54:57] <pg_user> in the ubuntu forums they lick every potential user's butt.
[07:55:10] <WickedWicky> that if if they're not too busy bitching at eachother and discussing what to implement and what GPL is better
[07:55:24] <e^ipi> WickedWicky: in fairness, they're GNOME people
[07:55:25] <pg_user> it is much different here.  I got booted from #solaris because they thought i was too stupid
[07:55:32] <e^ipi> they don't implement things, they strip features out
[07:55:39] <Tempt> Anyway, just grab a couple of old ethernet cards or something and cut the slots off. Solder wires to all the pins on the component side of the board to the same spot on the other board and solder all the pins on the other side to ground and you're good to go with your CPU doubling expansion.
[07:55:42] <WickedWicky> there is nothing wrong with gnomes, I am not too tall myself
[07:55:55] <pg_user> Tempt: i'm not doing that
[07:56:05] <e^ipi> GNOME: The "That Feature Was Confusing Users" Desktop Environment
[07:56:11] <Tempt> Cut the tracks to pins 1, 19, 83, 111, and 131 to reduce the instruction set for more performance.
[07:56:32] <pg_user> yeah right
[07:56:35] <WickedWicky> ohhhh, that gnome
[07:56:38] <WickedWicky> I like it
[07:56:42] <e^ipi> GNOME must have some fscking retarded people using it or something
[07:56:44] * WickedWicky is all for confusion
[07:56:49] <e^ipi> if things like location bars confuse users
[07:56:50] <WickedWicky> yea
[07:56:52] * WickedWicky raises hand
[07:57:03] * bda is also a fan of GNOME.
[07:57:08] <WickedWicky> *cof*jds*cof*
[07:57:12] <pg_user> What about CDE?  That is even more retarded than gnome
[07:57:19] <bda> In the interest of full disclosure, however, I do go throug three dribble bibs a day.
[07:57:21] <pg_user> it is like a friggen desk!
[07:57:22] <e^ipi> pg_user: naw, CDE's fine, it's just ugly
[07:57:32] <pg_user> any moron could operate cde
[07:57:42] <WickedWicky> hence why we love it
[07:57:51] <pg_user> why don't they improve CDE's apperance?
[07:58:01] <WickedWicky> cause they're morons
[07:58:03] <e^ipi> pg_user: because then it wouldn't be CDE
[07:58:08] <WickedWicky> xactly
[07:58:13] <e^ipi> it'd be the UnCommon Desktop Environment
[07:58:17] <e^ipi> UCDE
[07:58:32] <pg_user> WickedWicky: ok, you love CDE because it is easy, yet you don't tolerate fools on @solaris???
[07:58:44] <WickedWicky> I am not just a moron
[07:58:47] <WickedWicky> I am also a jerk
[07:59:10] <WickedWicky> besides, I am not @solaris
[07:59:11] <pg_user> it is a lot easier on the ubuntu irc channel
[07:59:16] <WickedWicky> what do I care what you do there?
[07:59:24] <bda> Is @solaris a channel on silc?
[07:59:35] <pg_user> i meant #solaris
[07:59:38] <WickedWicky> looks more like a perl array to me
[07:59:40] <bda> ho ho nerd pendantic commentary.
[07:59:42] <WickedWicky> ok well
[07:59:47] <WickedWicky> ls <> Ls in Solaris
[07:59:54] <WickedWicky> so maybe that's where you go wrong
[08:00:01] <pg_user> i got booted from #solaris for being "too stupid"
[08:00:08] <WickedWicky> write what you mean, not what you meant to write
[08:00:14] <richlowe> we tend to be nicer than #solaris.
[08:00:15] <e^ipi> WickedWicky: my god... did you just use pascal syntax ?
[08:00:17] <richlowe> how much nicer is variable.
[08:00:29] <WickedWicky> *gasp*
[08:00:32] <WickedWicky> not intentionaly!
[08:00:35] <e^ipi> !=
[08:00:35] *** slowhog has left #opensolaris
[08:00:37] <e^ipi> not <>
[08:00:51] <WickedWicky> math drainage :s
[08:01:02] <richlowe> at least it's not using ^ for pointers.
[08:01:26] <bda> Is that a tinydns-data knock?
[08:01:40] <WickedWicky> ok, coffee and shower
[08:02:30] <richlowe> bda: no.
[08:02:31] <e^ipi> at the same time?
[08:02:35] <e^ipi> that'd just water down the coffee man
[08:02:48] <WickedWicky> 'f course not
[08:02:49] <WickedWicky> duh
[08:02:56] <WickedWicky> you silly CDE person
[08:03:30] <WickedWicky> pg_user: what did you EXACTLY say before you got banned anyway?
[08:04:11] <pg_user> I wanted to know how to burn multiple sessions to my cd-r
[08:04:20] <bda> pft.
[08:04:30] <bda> What about tattling to Jon S. about people on IRC being mean to you?
[08:04:37] <bda> No one likes a rat.
[08:04:45] <pg_user> i wanted to back up a zfs snapshot to cd-r
[08:04:50] <pg_user> and they made fun of me
[08:04:57] <pg_user> they said I should mirror
[08:05:19] <pg_user> bda: like Jon would listen to me.
[08:05:38] <pg_user> I probably would end up in his spam folder
[08:07:04] <pg_user> well good night everyone
[08:07:30] <delewis> I've emailed Jonathan before
[08:07:36] <delewis> and actually received a reply within an hour.
[08:07:40] <pg_user> oh
[08:07:45] <e^ipi> he's pretty good about reading his email
[08:07:51] <pg_user> i don't feel like emailing him
[08:07:52] <Tempt> http://flikr.com/photos/jurvetson/19438860/in/set-257763/
[08:07:54] <e^ipi> I've had a few short conversations with him
[08:08:08] *** uebayasi has quit IRC
[08:08:10] <Tempt> "And it has built-in cigarette lighters and ashtrays at each console".
[08:08:14] <Tempt> That's winning hardware
[08:08:34] <e^ipi> eastern european hardware?
[08:08:55] <Tempt> IBM hardware
[08:09:36] <e^ipi> for the eastern european market?
[08:10:09] <pg_user> well good night everyone
[08:10:19] <pg_user> i will be reading my manuals tomorrow
[08:10:35] <Tempt> e^ipi: For the US market.
[08:10:44] *** pg_user has quit IRC
[08:10:44] <Tempt> e^ipi: (military)
[08:11:31] <bda> Heh.
[08:13:00] <Tempt> Amazing. Troll offline.
[08:13:31] <moazamraja> hmm
[08:15:33] <lloy0076> Eh?
[08:15:38] <lloy0076> Postgres is offline, did you say?
[08:17:58] <Tempt> [16:10] pg_user [n=chatzill at 71-34-73-199 dot ptld.qwest.net] has quit ["ChatZilla 0.9.78.1 [Firefox 2.0.0.5/2007071817]"]
[08:18:52] <WickedWicky> why do people cry about being banned on IRC?
[08:19:03] <WickedWicky> give them the finger and ask your question the cdrecord lisy
[08:19:06] <WickedWicky> list
[08:19:07] <WickedWicky> jeez
[08:19:36] <WickedWicky> cdrecord isnt even different in usage from ubuntu afair
[08:20:04] <richlowe> and cdrw has so little to remember.
[08:20:15] <WickedWicky> there is always -help
[08:20:52] <Tempt> No!
[08:20:55] <Tempt> OH NOES!
[08:21:02] <Tempt> You can't read the help! I mean, that'd involve reading.
[08:21:09] <WickedWicky> that's so true
[08:21:20] <WickedWicky> but we already pointed out I am a moron and ajerk
[08:21:24] <lloy0076> It depends on what one gets thrown for.
[08:21:28] <WickedWicky> so -help would be in vain for me
[08:22:05] <WickedWicky> lloy0076: I've been an IRC server administrator for years and have seen people being banned and kicked cause they said "Blind Guardian" sucked ass
[08:23:18] <WickedWicky> if you get banned, ask your question elsewhere. Saying how the Ubuntu channels are all helpfull and such after being kicked/banned from a #solaris channel wont really help you
[08:23:43] <WickedWicky> that is: if you wanted help in the first place
[08:23:48] *** delphi1000 has joined #opensolaris
[08:26:16] <WickedWicky> ok well, I called myself a gnome, a moron, a jerk
[08:26:19] <WickedWicky> my day is set
[08:26:22] * WickedWicky will go to the office now
[08:26:33] <WickedWicky> see you all soon
[08:26:39] <razrX> later WickedWicky
[08:26:53] *** WickedWicky has quit IRC
[08:28:19] <delphi1000> hi everyone
[08:28:59] <e^ipi> ahoy
[08:29:24] <delphi1000> body
[08:30:15] <delphi1000> buddy
[08:31:07] *** delphi1000 has quit IRC
[08:35:33] *** jamesd has joined #opensolaris
[08:35:34] *** ChanServ sets mode: +o jamesd
[08:35:48] *** jamesd__ has quit IRC
[08:38:02] *** linux_user400354 has joined #opensolaris
[08:40:13] *** Dar_HOME is now known as Dar
[08:40:16] <MousePad> Morning
[08:44:19] *** nayyares has joined #opensolaris
[08:44:30] <nayyares> hi guys !
[08:47:45] <delewis> I imagine once I do this Solaris 10 11/06 install, the next release will be out sometime today.
[08:49:28] <bda> delewis: Hurry up! :P
[08:53:12] *** linux_user400354 has quit IRC
[08:53:26] *** Cyrille has joined #opensolaris
[08:53:42] *** solar-star has joined #opensolaris
[08:53:53] <jmcp> hi nayyares
[08:54:41] <nayyares> howz it jmcp !
[08:55:30] <Tempt> About due for 07/07
[08:55:40] <Tempt> delewis: Gimmeh new Solaris! ;)
[08:55:55] <jmcp> nayyares: pretty good
[08:56:01] *** andyshack has joined #opensolaris
[08:56:27] *** tsoome has quit IRC
[08:57:00] *** linux_user400354 has joined #opensolaris
[08:57:17] *** nostoi has joined #opensolaris
[08:57:29] <andyshack> greets! can someone give me some tidbits on teh sun v240' ? im currently using old ultre e450's and was wondering if it would be worth the upgrade ?
[08:58:09] <delewis> the E450 and v240 are very different systems.
[08:58:40] <delewis> v240 does have ALOM.
[08:59:19] <andyshack> lights out ?
[08:59:29] <dlg> id go the 240 just cos of that
[08:59:31] *** comay has quit IRC
[08:59:59] <andyshack> i dont really know much about sun hardware, im learning as i go although its still going to take me anohter 4 or 5 years to finsih reading all the infos at sun.com
[09:00:18] <andyshack> basically : how much "more gooderer" are the v240's ? ;)
[09:00:26] <delewis> well, like I said the E450 and v240 are very different systems that were sold for different purposes.
[09:00:27] <Tempt> 240s are fine.
[09:00:34] <delewis> E250 has 4 procs vs. the 2 procs in the v240
[09:00:36] <Tempt> You've got UltraSPARC-IIIi CPUs instead of UltraSPARC II
[09:00:39] <delewis> s/E250/E450/
[09:00:46] <delewis> and the v240 will be a much faster system
[09:01:05] <Tempt> You've only got 4 disk bays in a v240, so if you've loaded up your E450 you might need to look at an external array
[09:01:12] <Tempt> You can pack more RAM into the v240
[09:01:21] <delewis> if you're using the E450s drive bays, though, you'll need to resolve that with the v240 by somehow (attached enclosure, etc.)
[09:01:28] <Tempt> The v240 only has three PCI slots, so if you've filled your 450, that could be a problem
[09:01:38] <andyshack> yeah ive taken the arrays into consideration, its all zfs and can be moved into d100's
[09:01:43] <Tempt> You get four gigabit ethernets onboard in the 240 instead of a single 100mbit
[09:02:13] <dlg> get a v225
[09:02:24] *** sponix has joined #opensolaris
[09:03:02] <Tempt> dlg: Get a 6900
[09:03:07] <andyshack> ah im just asking re the 240 as thats all on offer today :)
[09:03:29] <andyshack> i considered an old e6000 a while back for the lulz.
[09:03:38] <dlg> Tempt: m5000 pls
[09:03:50] <Tempt> I'd rather have the 6900
[09:03:58] <Tempt> E6000 is pretty hefty.
[09:04:13] <Tempt> andyshack: you'd probably want a 6500 though to get the faster gigaplane
[09:04:14] <delewis> why not stop at the M9000?
[09:04:29] <andyshack> yeah i didnt want to feed all that juice for dev. im still a complete noob.
[09:04:29] <Tempt> Yes, m9000 is the only solution.
[09:04:40] <delewis> everybody should have an M9000 beside their desk.
[09:04:43] <delewis> its the only way.
[09:04:49] <Tempt> I guess I'll be seeing how good the m5000 is soon.
[09:04:52] <Tempt> We're about to order four of them
[09:04:57] <sponix> sorry, power loss
[09:04:58] <delewis> lucky bastard. :-)
[09:05:02] <delewis> Tempt, where do you work at?
[09:05:15] <Tempt> State Government of Victoria; State Revenue Office.
[09:05:19] <Tempt> The taxman commeth.
[09:05:26] <delewis> and you guys need M9000s?
[09:05:28] *** WickedWicky has joined #opensolaris
[09:05:31] <Tempt> m5000s
[09:05:35] <Tempt> five thousand
[09:05:42] <delewis> er
[09:05:43] <delewis> M5000s.
[09:05:45] <delewis> still.
[09:05:46] <Tempt> M9000 I wish ...
[09:05:53] <Tempt> I *want* m5000s
[09:05:56] <delewis> M5000 is still pretty high-end gear.
[09:06:01] <Tempt> Not that high-end really
[09:06:35] <Tempt> Given Sun have told us the 490 and 890 are about to go end-of-sale
[09:06:45] <Tempt> That leaves a vacuum in the product lineup.
[09:06:55] <Tempt> the 5000 fits in nicely. (We're a bit of an 890 shop)
[09:07:00] <delewis> ah, so they're replacing the mid-range line with the Fujitsu stuff?
[09:07:15] <WickedWicky> V240 is EOL as well
[09:07:35] <Tempt> Pretty much. It's a bit worrying, because there isn't the SunSolve information or anything for he Fujitsu stuff
[09:07:52] <WickedWicky> I am a bit worried too
[09:07:59] <Tempt> It looks like Sun are getting ready to ditch everything short of the E25k.
[09:08:10] <WickedWicky> especialy since I never saw a proper fujitsu thing related to computers
[09:08:19] <delewis> well, the Fujitsu stuff *is* nice.
[09:08:23] <Tempt> They'll keep the high-end (serengeti) stuff as well.
[09:08:24] <delewis> nicer than the other mid-range Sun gear.
[09:08:29] <delewis> and for the most part, the same price.
[09:08:38] <jmcp> I'm waiting for the Rock stuff to arrive
[09:08:39] <Tempt> delewis: Have you actually worked with it?
[09:08:43] <WickedWicky> I did
[09:08:55] <WickedWicky> Siemens delivered us the machines
[09:08:56] <delewis> Tempt, no, but they do have mainframe features the mid-range (and even high-end) gear don't have.
[09:08:57] <richlowe> 240 doesn't look EOL according to sunsolve.
[09:09:04] <delewis> Sun is still selling the v240.
[09:09:07] <Tempt> delewis: No proven record?
[09:09:09] <delewis> and v210, even.
[09:09:14] <richlowe> they may not sell it (I didn't looK), but they seem to still support it.
[09:09:16] <Tempt> delewis: Fujitsu have never sold SPARC into Australia.
[09:09:17] <delewis> Tempt, nope.
[09:09:26] <andyshack> WickedWicky : cheers for the EOL headup, i was jus tlooking into that
[09:09:27] <WickedWicky> then why does our spectrum contract showed EOL when we closed the contract
[09:09:34] <Tempt> delewis: and they don't have a support base, so you're reliant on Sun's band of shabby field outsourced field engineers
[09:09:36] <richlowe> and the OPL stuff is there on sunsolve.
[09:09:46] <Tempt> delewis: And nobody around here has the experience with the hardware.
[09:11:21] <andyshack> would you also mind commenting on the 3310 arrays ? i cunrrenlty use the a1000's as doorstops and am using a few d1000's and of course the bays in the e450's
[09:11:36] <WickedWicky> we use 3310s
[09:11:37] <Tempt> Anyway, the machines look nice, the feature set looks great, and I'll be happy when I get four m5000s to build my new cluster with
[09:11:37] <jmcp> andyshack: what sort of comment do you want?
[09:11:49] <delewis> surprising how well an Ultra 2 with 1.5GB of memory handles 8 zones.
[09:11:55] <WickedWicky> nothing state of the art, it can do JBOT/RAID5/RAID5+
[09:11:57] <Tempt> andyshack: The 33100 will probably shift data at least 4 times faster than your a/d1000s
[09:12:14] <delewis> with 3 zones that are actually running fairly hefty services (Sun Web Server, DSEE6, and SSGD)
[09:12:17] <WickedWicky> I'm happy about the performence, just make sure you use dual controller configuration when you buy the kit
[09:12:31] <WickedWicky> since buying the controller afterwards will set you back 10K euros
[09:13:02] <WickedWicky> (that includes an engineer comfiguring/installing the card + multipathing)
[09:13:44] <WickedWicky> andyshack: I am currently writing an implementation plan of getting two SE3310s and put ZFS on top of it with mirroring
[09:14:05] <Tempt> delewis: No it isn't. The Ultra-2 has magical powers
[09:14:17] <nayyares> i have install solaris 10 on virtual machine, now when i plug in my usd memory drive, its not showing in /dev/rdsk, how can i check its device name and mount it?
[09:15:27] <WickedWicky> does /var/sadm/messages show the drive is detected?
[09:16:25] <bda> s/sadm/adm/
[09:16:32] <WickedWicky> that
[09:17:32] <nayyares> there is a msg regarding pseudo-device: sdt0 , i am not sure whether its due to usb dirve !
[09:17:41] <nayyares> dirve/drive
[09:17:57] <WickedWicky> you could try to run devfsadm -v
[09:18:10] <Tempt> "virtual machine"
[09:18:11] *** tsoome has joined #opensolaris
[09:18:18] <Tempt> implies it's probably a fault at that point
[09:18:29] <palowoda> most problems are virtual machines
[09:18:41] <WickedWicky> well yes, still, it doesnt hurt to run devfsadm -v :D
[09:18:59] <palowoda> it doesn't hurt to use real machines either.
[09:19:16] <WickedWicky> points for you
[09:19:22] <palowoda> after you know it works than fuck it up.
[09:20:39] <WickedWicky> Dell ships computers with Ubuntu now?
[09:20:52] <e^ipi> WickedWicky, they have for months..
[09:20:59] <WickedWicky> not in europe they didnt
[09:21:50] *** blueandwhiteg3 has joined #opensolaris
[09:22:19] <jmcp> nayyares: sdt is the static DTrace probe module
[09:22:21] <WickedWicky> http://blogs.sun.com/jonathan/entry/sun_enters_the_commodity_silicon
[09:22:26] <WickedWicky> you think anyone will follow?
[09:22:31] <jmcp> yeah, I do
[09:22:57] <Jondice> i'm using one of those ubuntu laptops, except i substituted it with open solaris
[09:23:05] <Jondice> everything works that i can think of
[09:23:12] <jmcp> WickedWicky: there are fabs in China which have taken the verilog etc and are building US-T1 fpgas or asic or SoCs from it
[09:23:40] <WickedWicky> I see
[09:24:22] <Jondice> well, power management isn't the best .. but with such a long batter life, i don't care much
[09:24:23] *** uebayasi has joined #opensolaris
[09:24:31] <jmcp> and the SME unit has already got an OEM deal going with Marvell for some of the nic chips
[09:34:15] *** linux_user400354 has quit IRC
[09:35:27] <moazamraja> jmcp: I dunno, someone has licensed the nic chip, but will any OEM license the whole chip?
[09:36:17] <moazamraja> I mean, there are tons of companies who maybe SHOULD use it (citrix/netscaler, netapp, emc, etc.), but many of them run x86 chips with linux/bsd and don't want to move to sparc based chips or solaris
[09:37:12] <jmcp> moazamraja: they don't need to license the US-T1 or US-T2 because it's under GPLv2
[09:37:15] <jmcp> the nic stuff is separate
[09:37:48] <moazamraja> but will they OEM it
[09:37:54] <moazamraja> will they use it.
[09:38:03] <jmcp> "it" being what?
[09:38:13] <moazamraja> T1/T2
[09:38:16] <moazamraja> the cpu
[09:38:19] <jmcp> yes, people will
[09:38:28] <jmcp> it's already being used in uni courses on chip design
[09:38:38] <jmcp> it's being fabbed in Taiwan or southern China
[09:38:44] <moazamraja> because at that point, that company is depending on Sun for the future of Sparc and their own OEM product
[09:39:10] <jmcp> moazamraja: uh ... no
[09:39:20] <moazamraja> ...
[09:39:33] <moazamraja> let's say I'm netapp, or Netscaler...
[09:39:34] <jmcp> the verilog etc for the chip has been released under GPLv2
[09:39:38] <jmcp> so it's out there
[09:39:49] <moazamraja> yeah but these companies are not in the business of having chips made for them
[09:39:51] <jmcp> you can pull it down via a google cache if you're really desperate and can't get to opensparc.net
[09:40:03] <moazamraja> they're in the business of writing code for their platform and supporting it
[09:40:26] <jmcp> which is why they rock up to, eg, TI or Motorola or TSMC (I think that's the name), and say "build me X-hundredthousand of these
[09:40:30] <moazamraja> hence they just buy commodity x86/x64 cpus that aren't dependent on any company, and AMD/Intel is very strong
[09:40:57] *** xuewei_ has joined #opensolaris
[09:41:22] <jmcp> yes, they certainly do have volume on their side
[09:41:27] <moazamraja> if you were these companies, wouldn't you think it'd be easier to just use standard x86/x64 chips? no dependency on another company
[09:42:01] <jmcp> I call BS
[09:42:17] <jmcp> no matter what chip you use, you will always have a dependency upon anoher company for *some aspect* of your product
[09:42:18] <Jondice> the distribution terms of hardware under the gpl is interesting.  The only major entity i can think of that would (possibly) want to modify the architecure and not distribute the changes would be a government.
[09:42:37] <e^ipi> corporations /are/ governments
[09:42:40] <Jondice> which they can do legally under the gpl of course, so that's kind of cool
[09:42:47] <richlowe> you can't avoid other companies.
[09:42:53] <moazamraja> depending on x86 chips is A LOT easier
[09:43:34] <moazamraja> don't get me wrong
[09:43:41] <moazamraja> i'm super pro-UltraSparc T1/T2
[09:44:17] <moazamraja> but if i'm a OEM/vendor, I'm going to think a dozen times before depending on Sun for my core CPU right now
[09:44:25] <richlowe> only a dozen? :)
[09:44:29] <moazamraja> :P
[09:44:48] <moazamraja> i think Netscaler could benefit a HUGE amount from US-T2
[09:45:47] <jmcp> moazamraja: one would hope that features of a chip, rather than just raw price, would actually have some bearing on the architectural decisions
[09:45:48] * jmcp shrugs
[09:46:20] <delewis> US-T2 would be awesome for storage/archive vendors, given the hardware crypto functionality.
[09:46:52] <delewis> most of those vendors rely on some sort of accelerator or doing the crypto in software.
[09:46:55] <moazamraja> jmcp: unfortunately, the business decision will override the tech issue
[09:47:08] <moazamraja> Sun needs to create the server appliances themselves
[09:47:18] <moazamraja> and out-do the supposed OEMs
[09:47:27] <delewis> Sun is, but they're not using SPARC to do it.
[09:47:36] <moazamraja> Apple sorta did that with their retail stores
[09:47:43] <delewis> the StorageTek 5800 is an appliance of sorts.
[09:47:44] <jmcp> moazamraja: enlightened company management actually listen to what their engineering staff recommend
[09:47:45] <delewis> but its AMD64.
[09:47:52] <delewis> which is quite a disappointment (to me, anyway)
[09:48:13] <delewis> http://www.sun.com/storagetek/disk_systems/enterprise/5800/
[09:48:33] <richlowe> jmcp: So, who has this mythical management?
[09:48:50] <jmcp> richlowe: you know what I mean
[09:48:53] <delewis> I'm assuming (or hoping) the 5800 uses ZFS.
[09:48:57] *** KermitTheFragger has joined #opensolaris
[09:49:06] <jmcp> isn't the 5800 the Honeycomb thing?
[09:49:08] <delewis> they're advertising "Reed-Solomon RAID 6"
[09:49:20] <delewis> jmcp, its a bunch of x2100s from my understanding.
[09:49:21] <blueandwhiteg3> I'm preparing to upgrade my personal solaris storage server and I have a really interesting idea
[09:49:31] <jmcp> delewis: oh, so it's a noise factory then
[09:49:38] <jmcp> blueandwhiteg3: go on
[09:49:40] <bda> M2s are pretty quiet.
[09:49:44] <bda> The M1s are jet engines.
[09:49:54] <blueandwhiteg3> It is possible to stripe together two 250 GB, then add them as a vdev to a larger raidz2?
[09:49:59] <jmcp> bda: I don't think it was ever called an M1
[09:50:12] <blueandwhiteg3> (with the rest of the devices in the RAIDZ2 being 500 GB)
[09:50:13] <delewis> http://www.sun.com/storagetek/disk_systems/enterprise/5800/specs.xml
[09:50:16] <bda> jcsmith: Just to differentiate.
[09:50:23] <bda> er. jmcp.
[09:50:31] <delewis> but like I said, that's an applicance, and Sun isn't using SPARC for it.
[09:50:44] <richlowe> jmcp: I think STE5800 is honeycomb, yeah
[09:50:53] <oxygene> blueandwhiteg3: should work
[09:51:00] <Doc> honeycomb exists now?
[09:51:13] <jmcp> Doc: I believe os
[09:51:14] <jmcp> so
[09:51:17] <Doc> last time i saw it, it didn't exist. (wait - does that make sense?)
[09:51:32] <jmcp> uh .... perhaps
[09:51:32] <richlowe> Doc: I never saw an announcement, but sun.com/storagetek/honeycomb redirects to the 5800
[09:51:37] <richlowe> rather than being its prior page.
[09:51:42] <delewis> it exists if its the STE5800.
[09:51:43] <blueandwhiteg3> oxygene: then can i later swap out the 2 x 250 GBs for 1 x 500 with replace?
[09:51:54] <Doc> honeycomb is very, very specialised storage
[09:52:09] <oxygene> Doc: it makes sense in an NSA sort of way..
[09:52:12] <Doc> not too many ppl with their heads screwed on right to do Object-based storage
[09:52:32] <jmcp> oxygene: where NSA == No Such Array ?
[09:52:51] <oxygene> jmcp: more like no such agency
[09:52:52] <blueandwhiteg3> This seems far fetched to me, but does ZFS/RAIDZ2 deal with all the scenarios ideally? or does losing a single 250 GB drive cause the entire vdev to go down?
[09:53:05] *** peteh has joined #opensolaris
[09:53:11] <oxygene> blueandwhiteg3: uh, a stripe is killed once a part of it is dead
[09:53:26] <blueandwhiteg3> so it ends up being rebuilt totally
[09:53:36] <delewis> don't forget about ditto blocks.
[09:53:58] <delewis> just because part of the stripe is dead, doesn't mean the entire stripe is.
[09:54:03] *** kaiwai has quit IRC
[09:55:11] <blueandwhiteg3> delewis: so while the strip won't be used for pairity purposes, the recovery process will be accellerated versus recovering it all?
[09:55:31] <delewis> how on earth did you get that from what I said?
[09:55:48] <blueandwhiteg3> i put that together with what oxygene said
[09:55:59] <delewis> http://blogs.sun.com/bill/entry/ditto_blocks_the_amazing_tape
[09:55:59] <jmcp> blueandwhiteg3: and came up with 5.2341 from "1+1"
[09:56:10] <oxygene> heh
[09:58:03] <blueandwhiteg3> right
[09:58:06] <sponix> anyone in here use netbsd pkgsrc on Solaris 10/SX x86 ?
[09:58:33] <bda> Yup.
[09:58:39] <blueandwhiteg3> I'm jazzed about this idea, no more wasted drives
[09:58:45] <Jondice> I always thought Venti ( http://en.wikipedia.org/wiki/Venti ) looked appealing as a backup system, but haven't had a chance to try it yet.
[09:59:01] <sponix> bda: was your "yup" to me, and if so, has it done ok for you ?
[09:59:09] <bda> sponix: Yup to both.
[09:59:22] <sponix> bda: had any better luck with anything else ?
[09:59:43] <sponix> bda: and how did you bootstrap it, with gcc/gnu tools, or Sun Studio for builds ?
[09:59:50] <bda> Been meaning to look at metapkg, but pkgsrc hasn't pissed me off enough.
[10:00:09] *** migi has joined #opensolaris
[10:00:10] *** dmarker has quit IRC
[10:00:13] <bda> Er, openpkg. Not metapkg.
[10:00:20] *** dmarker has joined #opensolaris
[10:01:07] <bda> sponix: Initially I used SunFreeware gcc to bootstrap pkgsrc gcc, then recompiled pkgsrc gcc with pkgsrc gcc to build everything else.
[10:01:28] <bda> I had some issues compiling poorly written GNUish code with sspro.
[10:01:53] <jmcp> bda: well duh :)
[10:01:57] <bda> Some stuff works, some doesn't. I try to stick to using pkgsrc's gcc for pkgsrc pkgs.
[10:02:00] *** lloy0076 has quit IRC
[10:02:06] <bda> jmcp: Yeah, I know. Just saying, for clarity's sake.
[10:02:25] *** jamesd has quit IRC
[10:02:36] <oxygene> to be honest, there are some issues with sun headers, too (eg. unguarded "#define SS" at some point, or - my latest issue - enum version in math.h, which broke struct version {..} in wine)
[10:02:52] <bda> sponix: For things I really care about (MySQL, Postgres), I use sspro and source from Sun or the vendor.
[10:03:05] *** Giaco has joined #opensolaris
[10:03:08] * bda gets asked this enough he should blog about it. :P
[10:03:35] <Jondice> yeah, i ran in to that SS problem a while back
[10:04:01] <oxygene> Jondice: I just added #undef SS right before the first use in the code as a work around
[10:04:42] <Jondice> yeah, that would be the logical thing to do, but I renamed the variable
[10:04:50] <sponix> bda: thanks for the advice. I might give both pkgsrc and openpkg a try next week
[10:05:25] <bda> sponix: openpkg actually has paid support for Solaris, and they're a ports system, so it seems like it might be a good thing.
[10:05:42] <bda> I've just built enough tools to wrap pkgsrc into our existing infrastructure that it might be annoying to move.
[10:05:45] <bda> So uh, I haven't.
[10:06:54] <oxygene> bda: what exactly means "a ports system" for you?
[10:07:37] <bda> oxygene: You get a directory full of spec and Makefiles and patches, which can all be easily modified.
[10:07:45] <oxygene> bda: ah, okay
[10:07:54] * oxygene got a new buzzword to add to pmpkg's description
[10:08:05] <sponix> bda: I'm just learning Solaris as much as possible as a Hobby ATM (yes I'm strange), over Seas playing Army again
[10:08:14] <bda> And that is the primary interface to the packaging system, as opposed to say, Debian or RPM, where source packages are available, but not the "typical" choice.
[10:08:28] <bda> Or say, Blastwave, where you don't get Makefiles (afaik anyway).
[10:08:30] <sponix> bda: so I have 1G download a week, and a nice laptop to play with, but cash is out of the question :P
[10:08:47] <richlowe> I never did understand the preference for building stuff.
[10:09:00] <richlowe> it seems unlikely you're modifying more software than you're just building pristine.
[10:09:04] <bda> sponix: openpkg has a free service. But since they *do* have *customers*, it should mean they will *work*. :)
[10:09:05] <richlowe> at which point, why use the resources?
[10:09:08] <oxygene> richlowe: as long as applications are configured on build time, it's necessary
[10:09:18] <richlowe> oxygene: see the 2nd line.
[10:09:30] <oxygene> richlowe: that's still a pristine build
[10:09:55] <bda> richlowe: Someone else upstream is managing (hopefully) the annoyances in porting the software.
[10:09:56] <oxygene> richlowe: just with --disable-unnecessary-feature --use-alternative-lib=that_one
[10:10:17] <bda> They track the vendor source, apply patches, etc.
[10:10:25] <oxygene> richlowe: which can make a big difference and yet doesn't fit into binary package systems
[10:10:46] <bda> For me, it is not "oh shiny I must use SUPER OPTIMIZE" is it being able to offset the management load and still be able to customize software when I have to.
[10:10:47] <richlowe> it doesn't, and if you're making those changes more often than not, it makes sense.
[10:10:57] <bda> richlowe: As opposed to using what?
[10:11:14] <richlowe> that was in reply to oxygene.
[10:11:26] <richlowe> bda: binary packages v. building stuff yourself
[10:11:33] <bda> Right, using what for binary packages?
[10:11:36] <richlowe> bda: my point being, I don't understand why people do the latter when they have no need to.
[10:11:52] <oxygene> richlowe: and finally: I haven't found a good binary package provider for solaris yet, so I became my own one
[10:11:52] <delewis> I generate packages from whatever I build if it isn't supplied by the vendor.
[10:12:01] <bda> Also, a ports system gives you a really nice framework for building your own packages.
[10:12:10] <delewis> richlowe, I assume you have no problem with that, right?
[10:12:23] <bda> Though uh, obviously you have in-system tools for that. :)
[10:12:24] <richlowe> bda: that's not mutually exclusive with actually using binary packages if they exist.
[10:12:31] <richlowe> delewis: I have no problem with any of it, I just don't understand it. :)
[10:12:57] <delewis> I just don't prefer to use someone's random package.
[10:13:00] * Jondice agrees with bda about wishing more package systems offered better build customization
[10:13:08] <delewis> especially, if its something that I'll be using.
[10:13:18] <delewis> now, if its shipped by Sun or some other vendor, sure, I'll use it.
[10:13:31] <sponix> bda: looks nice, but already has 2 strikes -- RPM package management and no VLC
[10:13:32] <delewis> but will I use some arbitrary package found on someone's site? absolutely not.
[10:13:57] <bda> sponix: Yeah, but I imagine pkgsrc also sucks for GUI stuff on Solaris.
[10:14:04] <bda> (guessing, I've never tried it)
[10:14:09] *** sarahj has quit IRC
[10:14:30] *** sarahj has joined #opensolaris
[10:14:55] *** WickedWicky has quit IRC
[10:15:03] <sponix> bda: its almost up to date on vlc, at 8.5patch claims not at 8.6 due to security issues
[10:15:07] <oxygene> sponix: I think openpkg's aim is to unify package management on servers with different operating systems within an organisation
[10:15:38] <sponix> oxygene: true, I can do that with pkgsrc as well though, if it functions decent
[10:15:43] <oxygene> sponix: and choosing rpm is as good as choosing any other package system at that point
[10:15:46] <bda> I agree re: using Sun pkgs when possible, except for a few cases (SunSSH vs recent OpenSSH, etc).
[10:16:06] <oxygene> bda: I think sunssh recently merged in openssh updates
[10:16:33] <bda> That doesn't make me want to change. :)
[10:16:47] <sponix> bda: Yeah, guess I could search for a totem-xine codecs package for SX
[10:16:50] <delewis> SunSSH is necessary if you're doing Kerberos.
[10:17:01] * bda isn't.
[10:17:03] <delewis> OpenSSH's GSS support is broken
[10:17:25] <tsoome> what is the reason you cant use sun ssh?
[10:17:39] <tsoome> I have never felt I need to replace sun ssh
[10:17:50] <Tpenta> how about "If it has a problem I can call sun support" ?
[10:17:51] <tsoome> just trying to get the point
[10:18:02] <sponix> as long as scp works I'm happy
[10:18:03] <deather> hello folks - i'm missing glib-object.h when trying to compile ON 20070730, does anyone know in which Sun package I can find it? (i'm on Nevada b68)
[10:18:37] <Tpenta> SUNWgnome-base-libs-devel
[10:18:38] *** sarahj has quit IRC
[10:18:39] <bda> Tpenta: I'm not really sure how to respond to that one.
[10:18:47] <deather> Tpenta: thanks!
[10:18:57] <Giaco> what can I use to make rdr on tcp ? for example from myip:110 to anotherip:110 ?
[10:18:58] <Tpenta> tsoome was after a good reason to use the one shipped, I supplied one
[10:19:14] <bda> Tpenta: He was asking why NOT to use it, I gathered.
[10:19:15] <Tpenta> oops the other way around
[10:19:22] <Tpenta> long day
[10:19:24] *** sarahj has joined #opensolaris
[10:19:29] <tsoome> yes - why are ppl replacing it?
[10:19:40] <Tpenta> a good reason TO use the sun supplied one is the support reason
[10:19:44] <tsoome> just because its *sun* ssh and not "open" ?
[10:19:59] *** sarahj has quit IRC
[10:20:02] <Tpenta> historically we were not real good at keeping it up to date
[10:20:14] <bda> No. Some of my co-workers use relatively recent features (from what I understand).
[10:20:19] *** sarahj has joined #opensolaris
[10:20:20] <bda> I myself am used to scp's -l option being available.
[10:20:41] <delewis> what does -l do?
[10:20:54] <bda> Having a single unified behavior across our three operating systems, where two of those platforms are the same already, makes a fair amount of sense.
[10:20:58] <Tempt> l for longer. It makes your transfer take longer. Handy.
[10:20:59] <bda> Bandwidth limiter.
[10:21:00] *** sarahj has quit IRC
[10:21:09] <delewis> ah.
[10:21:12] <bda> Analogous to rsync --bwlimit.
[10:21:15] <bda> But I'm used to it being there, so. :)
[10:21:21] <Tempt> dba: Jesus, I hit the mark on that one and I was only joking.
[10:21:24] <Tempt> bda, that is.
[10:21:39] * bda changes his name just to make Tempt happy.
[10:21:46] <Tempt> Just alias it ;)
[10:21:50] <bda> Actually, I won't. My mother'd kill me.
[10:22:02] <sponix> bda: I have to use the --bwlimit with rsync to my external usb drives or NAS drive, if I don't they reset themselves (strange I know)
[10:22:25] <bda> I am convinced once our colos taildrops, so -l/--bwlimit is really useful.
[10:22:54] <Tempt> sponix: rsync can really cane the I/O if you've got a lot of files there.
[10:24:24] <Giaco> is there someone who is using postfix ? I can't make it listen to my ip:25 it listen only to localhost:25
[10:26:05] <bda> Giaco: What does postconf inet_interfaces say?
[10:26:35] <sponix> Well I had better log off don't want to hit my dl restriction
[10:26:45] <bda> sponix: ta
[10:26:54] <sponix> normally I can't even get on IRC. So, it was really nice catching up with everyone
[10:27:18] <Tempt> You realise that IRC doesn't really consume much bandwidth?
[10:27:24] <Tempt> It isn't going to nail your DL destriction.
[10:27:30] <e^ipi> my uni. blocks it
[10:27:31] <oxygene> uh, why does ON need glib-object.h?
[10:27:36] <sponix> Trying to keep up with devel through blogs just isn't the same
[10:27:43] <e^ipi> oxygene, it also needs perl....
[10:27:45] <oxygene> (apart from that it also requires apache and libsnmp and various other stuff)
[10:28:07] <Giaco> bda, I've fix it, it was a master.cf syntax error
[10:28:07] <sponix> Tempt: yeah, but after the tons of pr0n I'm within megs of my limit, and still have half the week to go *Grin*
[10:28:12] <bda> Giaco: aha
[10:28:25] <Tempt> haha
[10:28:27] <Giaco> bda, the problem now is that is'nt logging anything
[10:28:29] <Tempt> Download limits are teh suck.
[10:28:50] <MousePad> Indeed.
[10:28:55] <oxygene> e^ipi: but it also seems to supply perl
[10:29:02] <sponix> Tempt: you have no idea. I'm in another country, the Net is _all_ I have, and then its restricted
[10:29:21] <MousePad> We get 1.8Mbps HSDPA wireless broadband here, but with a crap 1GB/mo limit :(
[10:29:24] <bda> Giaco: Do you have mail.* facilities set up in syslog.conf? Otherwise it might just be getting dumped to console/dmesg, hm?
[10:29:31] <Tempt> How awful.
[10:29:50] <sponix> Tempt: yeah Iraq playing Army would be ok if I had a fat connection
[10:30:16] <sponix> everything is SAT based in these parts though, so it cost major bucks
[10:30:30] <Tempt> Can't get a local dialup or something?
[10:30:49] <sponix> Tempt: something tells me you've never been to Iraq
[10:30:54] <Tempt> Indeed.
[10:31:02] <Tempt> I thought they might have a working telephone network by now though.
[10:31:30] <Giaco> bda, some information are in /var/adm/messages, but I would like to see something like mail.err
[10:31:33] <sponix> Even ATT&T Call Centers for Soldiers, all SAT/VOIP Based
[10:31:50] *** nostoi has quit IRC
[10:32:08] <bda> Giaco: Create the syslog facilities for them and `svcadm refresh system-log`.
[10:32:16] <sponix> anyway, I had better run
[10:32:20] <Tempt> later...
[10:32:23] <bda> sponix: Stay safe.
[10:32:37] *** sponix has quit IRC
[10:32:51] <bda> Giaco: Note that syslog.conf is tab-aware, if you are unfamiliar with editing it.
[10:33:02] <bda> (something that has always made me nuts)
[10:33:26] <Giaco> I think I am not able to do it (create the facility)
[10:33:33] <Tempt> I must admit I'm surprised that hasn't been changed. The tab vs space thing with syslog.conf has been a usability problem forever.
[10:34:06] <bda> Giaco: Not root?
[10:34:11] <bda> Or policy issues?
[10:34:38] <Giaco> eheh only newbe
[10:35:52] <sbahra> wow
[10:35:55] <sbahra> wtf happened
[10:35:58] <sbahra> i installed pkg-get
[10:35:58] <bda> Giaco: http://rafb.net/p/e8yPob56.html
[10:36:05] <sbahra> and now openwin won't start
[10:36:33] <bda> Giaco: You will also want to take a look at logadm(1M).
[10:36:38] <Giaco> bda, it's better /var/log or /opt/csw/var/log ?
[10:37:07] <bda> I try to keep all logs in /var/log, but do what makes sense to your environment.
[10:37:32] <bda> We have actually have two log systems at work: @syslog and socklog+rsync. :\
[10:39:44] <Giaco> bda, it works! thanks very much
[10:39:55] <bda> np
[10:40:53] <Giaco> is it sane to use webmin to administer postfix ?
[10:41:02] <bda> Not in my book.
[10:41:14] <bda> But uh, I work for an email company. So.
[10:41:23] <Giaco> I saw that webmin is a tragedy when needed to administer squid
[10:41:34] *** Teknomancer has joined #opensolaris
[10:41:53] <Giaco> I always used qmail, it's first time with postfix
[10:42:10] <bda> Postfix is great.
[10:42:17] <Giaco> I try it because was in blastwave
[10:42:37] <Tempt> Exim, man, exim. Exim is the one true MTA.
[10:42:50] <delewis> Sendmail.
[10:42:55] <bda> Not this again.
[10:43:07] <Giaco> does it work graylisting in sendmail ?
[10:43:16] <dlg> delewis: hi5
[10:43:17] <Tempt> sendmail.cf is turing complete
[10:43:24] <Tempt> You should be able to do *anything* in sendmail
[10:43:45] <delewis> and so you can. :-)
[10:43:48] <Tempt> I've seen towers of hanoi implemented in sendmail
[10:44:03] <delewis> in a virtual hosting environment, I would use something like Postfix.
[10:44:10] <delewis> Sendmail leaves something to be desired in that area.
[10:44:17] <delewis> but for your everyday MTA, Sendmail is the way to go.
[10:44:19] <Tempt> Sendmail used to be great.
[10:44:31] <Tempt> It'd handle mail, AND provide a way to run things remotely.
[10:44:32] <Stric> until anything else came along
[10:44:44] <delewis> Tempt, :-)
[10:44:53] <dlg> Tempt: thats so 10 years ago
[10:45:06] <Tempt> Hence "used to be" great. They took the best features out ;)
[10:45:11] <tsoome> like all other mail sw is bugfree
[10:45:23] <delewis> don't tell me Postfix doesn't have its own problems. (or Exim)
[10:45:55] <Tempt> Of course, all software has problems.
[10:46:03] *** yarihm has joined #OpenSolaris
[10:46:27] <Tempt> Exim just works for me though, so I'll give it a quick pitch.
[10:46:37] <Tempt> As opposed to qm**l
[10:46:44] <Giaco> mmm I think that I didn't a great job, I see only Relay access denied
[10:47:19] <delewis> any Sendmail administration worth the air he or she breathes could troubleshoot that one.
[10:47:24] <delewis> s/administration/administrator/
[10:48:14] <Stric> My vote goes to Postfix.. Greylisting is easy, config file is readable, stuff is pretty sane by default
[10:48:26] <Tempt> I got a response from IBM regarding pricing on their entry level POWER5 machines.
[10:48:47] *** MattMan has joined #opensolaris
[10:48:48] <Tempt> It basically said "We don't recommend you use Series p and AIX and would recommend a Linux machine instead"
[10:49:03] <tsoome> there are great docs for sendmail all around. if someone tells sendmail is hard to configure - this person most certainly cant read or understand the readed text.
[10:49:15] <delewis> or doesn't have the "Bat" book.
[10:49:22] *** xuewei_ has quit IRC
[10:49:28] <delewis> Tempt, that's the same way Sun treats it smaller customers.
[10:49:45] <delewis> IBM will give you an AIX/pSeries system if you don't ask about pricing. :-)
[10:49:46] <Tempt> Nope.
[10:49:57] <Tempt> Sun will give you the details of the nearest reseller and say "Enjoy!"
[10:49:58] <delewis> Tempt, no, they just refuse to sell to you.
[10:50:07] <asyd> tsoome: hmm, so I guess you speak weel ruleset lang? :p
[10:50:08] <delewis> that's very region-dependent from my experience.
[10:50:10] <Giaco> bda, what's a right way to start to setup a mail system ?
[10:50:32] <Tempt> Well, in Australia Sun just passed me over to FrontLine (my local reseller) and they looked after me. And yes, that was for a one-off purchase as an individual
[10:50:42] <bda> Giaco: Context? Not sure what you mean.
[10:50:46] <delewis> doesn't work that way in the US or EU for the most part.
[10:50:51] <tomww> tsoome: I would not say, that the rules-syntax of sendmail is something to love for an average admin...
[10:50:54] <Teknomancer> hmmm lots of undocumented calls being used by Qemu for solaris
[10:51:02] <delewis> most of the partners I've dealt with will take their precious time calling you.
[10:51:33] <Giaco> bda, I am trying to setup some mx's and smtp's as zones in a sun cluster enviroment
[10:52:03] <trochej> Elo
[10:52:05] <trochej> Coffee?
[10:52:34] <tsoome> tomww: i dont wanna know what level this "average" is:P
[10:52:51] <Tempt> delewis: I'm glad I've got the reseller support here then.
[10:53:31] <bda> Giaco: Ok? I don't use Sun Cluster (yet), but I have plenty of Postfix instances living in zones.
[10:53:40] <tsoome> well, sun does not like resellers as well;)
[10:53:53] <quasi> trochej: no thanks - I just downed 4 espressos - should keep me for a while
[10:54:33] *** Pietro_S has joined #opensolaris
[10:55:03] <e^ipi> espresso has a lot less caffeine than brewed coffee
[10:55:06] <e^ipi> just so you know...
[10:56:33] <quasi> e^ipi: that's why I get a quad espresso
[10:57:20] *** Tpenta has quit IRC
[10:57:21] <bda> QUAD DAMAGE.
[10:58:00] *** Tpenta has joined #opensolaris
[10:58:09] <quasi> e^ipi: http://coffeetea.about.com/library/blcaffeine.htm
[10:58:18] <Tempt> trust me
[10:58:34] <Giaco> does exist something like rdr command for linux ?
[10:58:35] <Tempt> a overextracted cup of cheap robusta has plenty of caffeine
[10:58:36] <quasi> bda: QUAD joy
[10:58:41] <nayyares> the default mozilla browser in solaris is java enabled?
[10:59:10] <e^ipi> quasi: who the hell drinks 8oz of coffee?
[10:59:23] <e^ipi> of course a double espresso will have more than a quarter cup of drip
[10:59:33] <Giaco> Tempt, is it possible to setup a sun cluster without using shared storage ?
[11:01:08] <e^ipi> on the plus side though, drip sucks, and it ruins perfectly good coffee beans
[11:01:20] <e^ipi> so you're better off with the espresso just from an aesthetic standpoint
[11:01:57] *** WickedWicky has joined #opensolaris
[11:02:12] <WickedWicky> Hey all
[11:02:55] <e^ipi> what up
[11:03:35] <WickedWicky> bug hunting
[11:04:00] <e^ipi> bored?
[11:04:08] <WickedWicky> no, necessaty
[11:04:23] *** aramdune has joined #opensolaris
[11:04:35] <bda> Oh. Wonderful. MTemple iPhone mgmt interface.
[11:04:38] * bda spins finger in air.
[11:04:57] <bda> How about you guys make my gridserver faster instead of working on poopy Web2.duh apps. :(
[11:05:31] <Tempt> Giaco: Not that I know of.
[11:05:35] <bda> "Buy a domain, reboot your server, add emails, pay your bill, get support and more - all from your iPhone, only at (mt)." # If it had a fucking SSH client, I could reboot servers from iPhone, thanks.
[11:05:53] <WickedWicky> haha
[11:05:54] <bda> I thought Sun Cluster didn't require shared storage as of 3.2?
[11:06:01] <WickedWicky> oh?!
[11:06:03] <WickedWicky> that'd be cool!
[11:06:07] <Tempt> I haven't worked with 3.2
[11:06:18] <WickedWicky> safes me spending $ on shared storage
[11:06:19] <Tempt> Never get any fucking time any more.
[11:06:29] <WickedWicky> how's fucking related to sun cluster?
[11:06:34] <bda> I THOUGHT I read that. Maybe I misunderstood.
[11:06:41] <quasi> bda: you could use a single node cluster or quorum server to avoid shared storage
[11:06:49] <quasi> with sc 3.2
[11:07:03] <Tempt> Just because you *CAN* run without shared storage doesn't mean it's a good idea
[11:07:04] <bda> quasi: Quorum server is a term that rings a bell, aye.
[11:07:08] <WickedWicky> ah yes, 3.2 has quorum over TCP/IP
[11:07:21] * bda has no idea what best practices for Sun Cluster are.
[11:07:43] <Giaco> quorum server is another phisycal server ?
[11:07:48] <WickedWicky> oui
[11:07:57] *** cmihai has joined #OpenSolaris
[11:08:04] <WickedWicky> Tempt: could be a good idea when you wanna use NAS instead of SAN
[11:08:16] <Tempt> aah, NetApp fanbois!
[11:08:19] <quasi> Giaco: could be a zone as well
[11:08:29] <Tempt> I believe the last update of 3.1 introduced quorum-on-nas
[11:08:32] <WickedWicky> quasi: didnt think of that yet, doh
[11:08:46] <WickedWicky> 3.1u4 didnt, afaik
[11:08:51] * WickedWicky checks
[11:09:02] * Tempt goes hunting for a cisco console cable.
[11:09:18] <quasi> WickedWicky: I read it in the docs somewhere - possibly under what's new for 3.2
[11:09:38] <quasi> WickedWicky: says something like zones being able to act as nodes in a cluster
[11:09:45] <Giaco> I need this http://www.penguin-soft.com/penguin/man/1/redir.html for solaris
[11:09:54] *** whaq has quit IRC
[11:10:13] <WickedWicky> Tempt
[11:10:15] <WickedWicky> you're right
[11:10:19] <WickedWicky> u4 does support NAS
[11:10:29] <WickedWicky> What is the type of device you want to use?
[11:10:32] <WickedWicky>     1) Directly attached shared disk
[11:10:33] <WickedWicky>         2) Network Attached Storage (NAS) from Network Appliance
[11:10:39] *** whaq has joined #opensolaris
[11:10:44] <quasi> single node cluster doc is http://docs.sun.com/app/docs/doc/817-6543/6mlu3j9fp?a=view
[11:10:51] <WickedWicky> thanks quasi
[11:11:18] <bda> http://learningsolaris.com/archives/category/sun-cluster/ # Seems full of useful stuff.
[11:11:51] <BatonT> are zpool cksum errors a sign of a bad disk?
[11:12:03] <BatonT> i just did a zpool replace and getting those errors on new drive
[11:12:20] <quasi> WickedWicky: or http://www.opensolaris.org/os/community/ha-clusters/ohac/Documentation/installsinglenode/
[11:13:08] <WickedWicky> cheers
[11:13:10] <bda> quasi: What's the status on the "Geographic Edition" (2005 post on my URL)?
[11:13:17] <bda> Boy, that'd be keeno.
[11:14:00] <bda> Oh, well, it's downloadable. So. woot.
[11:14:03] <quasi> bda: what about it? afaik it has been released with 3.2
[11:14:12] <bda> Aye.
[11:14:33] <BatonT> how can i get more details about zpool checksum errors?
[11:14:36] * bda wonders when he'll get to put that on his whiteboard.
[11:16:59] *** Fullmoon has joined #opensolaris
[11:19:34] <blueandwhiteg3> BatonT: Checksum errors are bad
[11:19:41] <BatonT> :((
[11:19:49] <BatonT> what makes things worse is..
[11:19:57] <blueandwhiteg3> BatonT: Check your cabling. Check your controller. Reboot. If that fails, RMA the drive.
[11:20:04] <BatonT> i started doing a zpool replace on another 'good' drive  with a new one
[11:20:05] *** delphi1000 has joined #opensolaris
[11:20:12] <BatonT> so i have a failing drive  + a drive being replaced
[11:20:23] <BatonT> so if that replace also causes error i loose my data
[11:20:24] <blueandwhiteg3> BatonT: Scary, but the checksum errors could just be a cable thing.
[11:20:33] <BatonT> BOTH disks are already been RMAed!
[11:20:47] <blueandwhiteg3> BatonT: As long as they stay under a certain error rate, you will be alright.
[11:20:48] <BatonT> had 2 critical failures soon as i plugged them in
[11:20:56] <WickedWicky> BatonT: SCSI black plane
[11:21:05] <BatonT> its a sata hotswap enclosure
[11:21:09] <quasi> sure sounds like a bad controller
[11:21:11] <BatonT> so it has a backplane
[11:21:17] <BatonT> onboard nvidia
[11:21:25] <delphi1000> well
[11:21:38] <oxygene> hmm.. are there actually any non-bad sata controllers out there?
[11:21:45] <Tempt> LSI SAS controllers
[11:21:49] <WickedWicky> I had a failing disk on a blue wednesday, got it replaced, other disk broke during metasync, Sun replaced the other disk, two days alter, both disks borked again, then they replaced the backplane
[11:22:04] <quasi> Tempt: those suck quite a bit
[11:22:09] <WickedWicky> oxygene: sure, the internal disks of a V440 are SCSI
[11:22:12] <Tempt> Not sucking for me at the moment.
[11:22:32] <BatonT> its ok to shutdown in middle of a zpool replace right?
[11:22:47] <oxygene> WickedWicky: uh, what does that have to do with "sata controller"?
[11:22:48] <delphi1000> zfs is a good choise
[11:23:06] <WickedWicky> oh, I misread
[11:23:07] <Tempt> Man, the serial cable fairy has raided my office and stolen all my cisco console cables *AND* all my 9 pin sun backshells
[11:23:12] <WickedWicky> I read non-bad no-sata
[11:23:13] <WickedWicky> :D
[11:23:22] <oxygene> WickedWicky: wishful thinking
[11:23:33] <BatonT> i have a stack of cisco console cables
[11:23:44] <BatonT> after doing an install of about 40 cisco switches :)
[11:24:01] <Tempt> I had at least six of the little varmints.
[11:24:11] <delphi1000> hi buddy
[11:24:24] <Giaco> you can use an utp with a rj45 to db9 converter
[11:25:26] <e^ipi> much more effective than an rj45->mains power converter
[11:25:56] <e^ipi> but I have one of those in a drawer, just in case
[11:27:28] <Tempt> found one
[11:27:30] <Tempt> Woohoo!
[11:27:43] <Tempt> Time to install that pesky V210
[11:27:51] <delphi1000> who knows SAN?
[11:28:03] <e^ipi> Bo?
[11:28:05] *** boro has joined #opensolaris
[11:28:15] <Tempt> I eat SANs for breakfast.
[11:28:16] <boro> hello
[11:28:44] * quasi kills SANs for lunch
[11:28:58] * tomww uses SANs for everything
[11:29:10] <boro> how network is configured for solaris xen dom0 - domu connections, is there already bridging support...
[11:30:41] *** LuckyLuke has joined #opensolaris
[11:32:43] <BatonT> anyhow... any way i can get more info about the zpool errors?
[11:35:30] <delphi1000> lookup log
[11:36:00] *** delphi1000 has quit IRC
[11:36:21] *** bnitz has joined #opensolaris
[11:37:00] <trochej> http://chfbs.org/high_school/high_sch_math.htm
[11:38:23] <Giaco> I go to make some motocross
[11:38:26] <Giaco> bye all and thanks
[11:38:56] <Tempt> trochej: Holy dog shit.
[11:39:16] *** Giaco has quit IRC
[11:40:11] *** LuckyLuk2 has joined #opensolaris
[11:41:43] <Tempt> trochej: Is that a joke site?
[11:41:56] *** Erwann has joined #opensolaris
[11:42:17] <Tpenta> scarily i suspect not
[11:42:18] <Berny> Tempt: probably not
[11:42:30] <Berny> there's enough of those wierd people around
[11:42:31] <trochej> Seems like not :)
[11:42:49] <Berny> one should send them the flying spaghetti monster :-)
[11:42:54] <Tpenta> look thru some of teh other courses, like science and english
[11:42:57] <trochej> Becoming Ph.D. in math can do that to a mere human being.
[11:44:41] <Cyrille> to be honest, it looks a bit like landoverbaptist (which is a fake joke site)
[11:44:46] *** LuckyLuk1 has quit IRC
[11:45:36] *** dunc has quit IRC
[11:46:15] <Berny> one of my friends hosts a nice satirical site on nearly everything (chistians, satanists etc) he gets quite a number of threats and stuff :-) though it's pretty obvious his site is a joke
[11:46:50] <Tpenta> School Mascot "The Eagles", ... obviously not the eagles that I think of when reading that
[11:49:59] *** uebayasi has quit IRC
[11:51:36] *** Gman has joined #opensolaris
[11:52:24] <Tpenta> omg read the cheerleading guidelines
[11:52:37] <Tpenta> hey glynn
[11:54:44] <Berny> where's that stuff hidden?
[11:54:57] <Tpenta> go to just the top site
[11:55:44] *** LuckyLuke has quit IRC
[11:55:48] <Gman> hey Tpenta
[11:57:58] <Gman> Tpenta, amusing use of dtrace :)
[11:58:06] * Tpenta points glynn at his IM
[11:58:11] <Tpenta> i was feeling lazy ;)
[11:58:18] <Gman> IM?
[11:58:30] <Tpenta> Instant Messenger, private mesg etc
[11:58:45] <Gman> oh, i haven't seen anything on my end :/
[11:58:50] <BatonT> crap zpool scrub bringing up a stack of checksums on 2 drives in a raidz1 :(
[11:59:07] *** ChanServ sets mode: +o Tpenta
[12:02:31] <Tempt> Man, that site was amusing.
[12:02:36] <Tempt> Cheer guidelines included.
[12:02:51] <Berny> yeah they're a fun bunch :>
[12:03:28] <Berny> if you understand german try rasputin.de :-)
[12:04:07] <Cyrille> well I recommend landoverbaptist if you don't already know it too.
[12:05:58] *** kaiwai has joined #opensolaris
[12:06:29] <Tempt> Oh, I know landoverbaptist.
[12:06:39] *** Erwann is now known as ErwFood
[12:07:23] <kaiwai> the sad part, there are people who are like landoverbaptist
[12:07:31] <bnitz> chfbs is probably real and so not as funny.
[12:07:55] <bnitz> Or if chfbs is a satire, it pales in comparison to reality.
[12:07:56] *** movement has quit IRC
[12:08:34] *** Corpuscule has joined #opensolaris
[12:08:41] *** movement has joined #opensolaris
[12:08:52] * Tempt watches his v210 jumpstart ...
[12:09:01] *** boro has quit IRC
[12:12:57] *** blueandwhiteg3 has quit IRC
[12:14:31] *** cypromis has joined #opensolaris
[12:16:12] *** nayyares has quit IRC
[12:16:24] *** act has joined #opensolaris
[12:17:13] * WickedWicky has a v240 under his desk now with a broken ALOM
[12:17:36] <Berny> how do you break the alom?
[12:17:48] <WickedWicky> I asked the dude who put the server under my desk
[12:17:51] <WickedWicky> you power on the server
[12:18:02] <WickedWicky> it scrubs the memory
[12:18:14] <WickedWicky> then tries to load the image
[12:18:22] <WickedWicky> and does that over and over again
[12:18:42] *** simford has quit IRC
[12:18:53] <WickedWicky> my first reaction was "WTF"
[12:19:09] <WickedWicky> then I called Sun and asked me if I could flash the firmware
[12:19:16] *** act is now known as edwardocallaghan
[12:19:17] <WickedWicky> which I tried, and it tells me it cant detect the hardware
[12:19:32] <WickedWicky> on which Sun said: your alom board is broken
[12:20:39] <edwardocallaghan> Hey all
[12:20:46] <edwardocallaghan> whos around
[12:21:14] *** linux_user400354 has joined #opensolaris
[12:26:56] <Tempt> WickedWicky: That's might crap.
[12:27:03] <Tempt> WickedWicky: Did you try flashing it from within Solaris?
[12:27:22] <Tempt> edwardocallaghan: Me, building a new webserver.
[12:28:04] <Doc> yes, your ALOM board is broken
[12:28:19] <Doc> you could try re-seating it - sometimes they come loose and do weird shit like that
[12:28:41] <Doc> but it's probably broken.  Sun should just send you a new one
[12:28:54] <andyshack> WickedWicky : glad youre still about. re : 3310's you mentioned the dual raid card is better. is that just for throughput and fast access or is the second actually required to enable the the use of the other 6 bays.
[12:29:15] <Doc> with only one controller you have a huge SPOF
[12:29:29] <andyshack> stupid pile of shite ?
[12:29:43] <Doc> plus you really shouldnt enable the write-behind cache with a single controller, 'cos if one fails you lose the cache and corrupt all of your data
[12:29:47] <Doc> Single Point of Failure
[12:30:00] <andyshack> nice
[12:30:09] *** jpdrawneek has joined #opensolaris
[12:30:14] <Doc> same goes for any hardware raid product
[12:30:14] <andyshack> ah failure is my networks middle name
[12:30:31] <andyshack> ok cool, cheers for that.
[12:31:06] <edwardocallaghan> Tempt:Hey
[12:31:40] <edwardocallaghan> Tempt:I am still looking for people to help me with my Magazine
[12:32:02] *** halton has quit IRC
[12:32:06] <edwardocallaghan> Would you be intersted?
[12:32:10] <Doc> magazine? want riters?  i right good!
[12:32:37] <edwardocallaghan> Doc:You in?
[12:32:50] <edwardocallaghan> Great ! Can you pm me please
[12:34:08] <edwardocallaghan> Anyone, going once, going twice..
[12:34:57] <Doc> what do i get to rite good about?
[12:35:05] <Tempt> edwardocallaghan: Email me some login details and I'll write an article for you.
[12:35:06] <edwardocallaghan> Just need people to blog about, "this is how to fix this, or this is how to get a basic webserver setup etc.."
[12:35:26] <Tempt> edwardocallaghan: I'll give you either an overview of SunCluster or vmstat for dummies
[12:35:41] <edwardocallaghan> Tempt:Fantastic
[12:35:52] <Doc> in solaris?
[12:35:54] *** Gekkko has joined #opensolaris
[12:36:03] *** calumb has joined #opensolaris
[12:36:10] <Doc> hasnt anyone actually worked out that i dont like solaris yet?
[12:36:21] <Tempt> Doc: Actually, I was thinking of writing an article for a Solaris magazine about vmstat on IRIX. That'll teach 'em all.
[12:36:22] <edwardocallaghan> Tempt:I can add a user now, I just need a username/pass and email you would like to use
[12:36:36] <Tempt> msg'd.
[12:36:41] <Gekkko> hey
[12:36:51] <Doc> "Microsoft SMS comes with a web server, thus you don't need to set it up"
[12:37:16] <cmihai> I wrote a Solaris article in a Linux magazine once
[12:37:21] <cmihai> All it got me is a flood of stupid questions.
[12:37:27] <cmihai> "Why isn't Solaris Linux?"
[12:37:32] <cmihai> "What kind of Linux is Solaris?"
[12:37:33] <cmihai> god dammit.
[12:37:43] *** Gekkko` has joined #opensolaris
[12:38:02] <Tempt> Hmm
[12:38:20] <Gekkko> cmihai: that amuses me
[12:38:22] <Tempt> I could take a camera to a Linux function of some sort and write a linux-spotting documentary explaining the failings of the people in the photos
[12:38:34] *** Gekkko` has left #opensolaris
[12:38:43] *** Gekkko has left #opensolaris
[12:38:48] <edwardocallaghan> Well this site is a OpenSolaris site only for all them HWOTOs people look for when starting out with Linux. Only its for OpenSolaris SXCE etc..
[12:38:52] *** Gekkko has joined #opensolaris
[12:38:55] <Gekkko> wrong button ><_>!
[12:39:04] <Tempt> Gekkko: Man, that's tacky as hell: ["http://bbqsrc.org - Now less gay."]
[12:39:13] <Gekkko> extremely tacky lol.
[12:39:28] <Gekkko> Don't you think that's the point?
[12:40:12] <cmihai> Soyent Green - Now with more Girl!
[12:40:49] <edwardocallaghan> victoredwardocallaghan _ At _ yahoo _ . _ c0m [.] au
[12:40:50] <Gekkko> wanna here somehting shit?
[12:40:51] <Tempt> Kinda like the domain though.
[12:40:52] <aramdune> cmihai, that myl article?
[12:40:55] <Gekkko> my internet got capped to 6kb/s
[12:40:59] <Gekkko> half way through downloading opensolaris
[12:41:00] <Gekkko> >_>
[12:41:02] <cmihai> aramdune: yeah
[12:41:18] <edwardocallaghan> please email me Tempt as I forgot your email :p my bad
[12:41:31] <edwardocallaghan> You too Doc
[12:41:32] <Tempt> edwardocallaghan: re-msg'd
[12:42:02] <edwardocallaghan> Oh, sorry did not see that
[12:42:03] <edwardocallaghan> thanks
[12:42:31] <Doc> edward: i tried that address - it bounced
[12:43:58] *** aruiz has joined #opensolaris
[12:44:41] <edwardocallaghan> hehe
[12:45:00] <cmihai> edwardocallaghan: so, is this yet another pdf web magazine?
[12:45:20] <Doc> are the underscores actually _ or do i replace them with -'s ?
[12:45:49] <Gekkko> spaces i'd presume
[12:45:49] *** Corpuscule is now known as Corpuscule[AWAY]
[12:46:20] <cmihai> I thing he means <spambot> !add edwardocallaghan at yahoo dot com.au :P </spambot>
[12:46:24] <Tempt> I think it'd reasonably obvious that everything to the right of the username is @yahoo.com.au with no underscores, spaces, or other excitement
[12:46:25] <cmihai> (Just kidding :P)
[12:46:46] <Gekkko> Tempt: obviously not :P
[12:47:24] <edwardocallaghan> cmihai:Yes I will need some input on the matter
[12:48:11] <cmihai> So, you want actual articles?
[12:48:59] <edwardocallaghan> cmihai:Yes and if you don't want to do that I am looking for people just to check spelling/content
[12:49:06] <cmihai> lol
[12:49:11] <cmihai> ispell/aspell
[12:49:16] <Tempt> edwardocallaghan: You need to find a graphics person to do a skin for that.
[12:49:19] *** Cass has joined #opensolaris
[12:49:28] <cmihai> You need a Quark XPress geek.
[12:49:42] <cmihai> Or, sigh, at least Scribus.
[12:50:03] <edwardocallaghan> At the moment I need 1 graphics  and 1 other person to help with th CMS side of things
[12:50:18] <cmihai> And what exactly do you have so far?
[12:50:19] <edwardocallaghan> Tempt:I do indeed
[12:50:33] <Tempt> Managing joomla is a snap. I've done a few deployments.
[12:50:42] <cmihai> If it's just tutorials or whatever, you need a wiki. If it's saying stuff, you need a blog. If it's a pdf magazine.. it's more complex
[12:50:56] <Tempt> Nobody reads PDF magazines
[12:50:59] <edwardocallaghan> I am talking to a graphics guy at the moment, but its hard to get though to artists that I am not a artist
[12:51:06] <Tempt> They'll read print magazines because they can read them away from a computer
[12:51:17] <Tempt> They'll read web content (especially RSS syndicated) because it doesn't require effort
[12:51:34] <cmihai> Tempt: my point exactly, too little content, too hard to read.
[12:51:39] <edwardocallaghan> Its going to come in more that one form
[12:51:50] <cmihai> That's silly. Just start a blog or a wiki, you're better off.
[12:52:01] <cmihai> And like Tempt put it, if you give them Atom or RSS, it's much better.
[12:52:08] <cmihai> Just as long as you can add content.
[12:52:40] <cmihai> I suggest you start some blog and blog your experiences. Besides, articles are more like general presentations, not technical tutorials.
[12:52:47] <edwardocallaghan> A PDF of all the things that have come together after x weeks and RSS feed of new HOWTOs and a search though a WiKi like artical area
[12:52:50] <cmihai> Start a blog or a wiki, and see how it goes from there.
[12:53:02] <cmihai> Nobody reads PDF tutorials mate.
[12:53:07] <cmihai> Especially in elongated article form.
[12:53:14] <Tempt> I'll probably put my blog online when I get this new server racked. Just so I've got some load on the host to burn it in.
[12:53:26] <Tempt> PDF tutorials are appropriate for some things.
[12:53:30] <Tempt> I like IBM's AIX stuff
[12:53:48] <cmihai> Yeah, I keep an internal blog too, for notes and stuff... started to publish some content on blogspot... easier then having to pastebin every time.
[12:53:49] <Tempt> w00t, install finished
[12:53:50] <Tempt> now patching ...
[12:53:59] <edwardocallaghan> Well its more complex than what your saying
[12:54:04] <cmihai> Oh crap, brb :P
[12:54:05] <Tempt> cmi: I find a wiki works better for notes and stuff
[12:54:11] <cmihai> Have a consultant waiting for me hehe
[12:54:26] <Berny> consultants can wait :-)
[12:54:46] <Berny> that way they get paid for doing no damage... thats fair enough
[12:54:47] <Doc> i wrote a PDF document once telling people how to install Acrobat Reader
[12:54:52] <Berny> lol
[12:54:54] <Tempt> Excellent.
[12:54:56] <Doc> it turned out not to be all that popular... :(
[12:54:56] <edwardocallaghan> WiKi's are great but its some times overwhelming conent..
[12:55:20] <Tempt> Wikis lack attribution
[12:55:21] <edwardocallaghan> lol
[12:55:36] <Tempt> People writing technical tutorials etc deserve at least one line of recognition.
[12:55:50] <Tempt> OR, give people a chance to judge the credibility of the article by the author.
[12:56:39] <edwardocallaghan> Yes they do
[12:56:41] <Tempt> For example, a dtrace article by Brendan gets more notice. Doc would be ideal for writing articles about flaming lamers on aus.computers.sun
[12:56:57] <edwardocallaghan> I need as much feedback as I can get on this, good or bad
[12:57:14] <Tempt> If we saw kaiwai writing an article about whining about iPods, we'd note it to be a very good article about whining
[12:57:30] <edwardocallaghan> lol
[12:57:49] <Doc> aus.computers.sun still exists?
[12:58:18] <edwardocallaghan> Well the CMS only came online yesterday
[12:58:33] <kaiwai> Tempt: hmm, I don't own one :P
[12:58:36] <edwardocallaghan> So there are many things to be sorted out on that side of the story
[12:58:59] <Tempt> Doc: And Craig Dewick still posts there looking clueless; yes.
[12:59:10] <edwardocallaghan> If some one would like to help with that I would be very thankful :)
[12:59:28] <Doc> he still selling 2nd hand trains and driving suns?
[12:59:32] <Doc> or is that the other way around
[13:00:00] <Tempt> pkgrm SUNWsndmu SUNWsndmr
[13:00:04] <Tempt> That's better...
[13:00:33] <Tempt> Doc: I think he's crying because nobody wants to pay $200 for an IDPROM for a sparcstation and got retrenched from the train driving
[13:01:06] <Tempt> Doc: He now hangs around comp.unix.solaris bitching and moaning that the Blade-1000 that he bought for $AU20 (literally) has a dead fibre channel interface
[13:01:58] <edwardocallaghan> what! I would keep my mouth shut, how cheap is that!
[13:02:51] <Doc> hmm.. you're making me wish i still worked as Sun so i could get the lawyers onto him again
[13:03:20] <Tempt> He always irritated me.
[13:03:29] *** deather_ has joined #opensolaris
[13:06:02] <Berny> hmm, on a 16000kbit dsl line what speed would one expect to see? isn't 1073kbit enough reason to complain?
[13:06:03] <Doc> he always irritated everyone
[13:06:18] <Doc> how are you measing it?
[13:06:25] <Doc> measuring  too
[13:06:48] <Berny> value reported by router... that's what it synced itself to
[13:07:02] <Berny> router/dsl modem combo that is
[13:08:04] <Tempt> Berny: distance to exchange?
[13:08:33] <Berny> get the same speed since they activated my dsl line
[13:09:09] <Berny> though the confimation on my order says 16000kbit :-\
[13:09:14] <Doc> sounds like you need a call to support
[13:09:22] <Berny> yeah :-\
[13:09:25] <Doc> 16,000 will be a theoretical max - most ppl wont get to that
[13:09:29] *** deather has quit IRC
[13:09:32] *** deather_ is now known as deather
[13:09:32] <Doc> but you should get more than 1000
[13:09:43] <Berny> if i'm lucky that techie hops about our block today...
[13:09:53] <Doc> mine is up to 24Mbps, but i get about 10Mbps
[13:10:07] <Berny> i wouldn't expect exact 16000 but 6000 would be fair to keep me quiet
[13:11:12] *** jpdrawneek has quit IRC
[13:11:17] <Gekkko> I hate the Linux smitten people.
[13:11:32] <Tempt> mine is up to 24Mbps, but i get about 24.5Mbps
[13:11:40] <Tempt> I can see the exchange from my front door though
[13:11:51] <Berny> grrr
[13:11:56] <edwardocallaghan> boyd said you had a good DSL Tempt
[13:12:27] <Tempt> I used to work in that exchange, too. I've seen my DSLAM.
[13:12:30] <edwardocallaghan> Were is boyd these days anyways ?
[13:12:37] <Tempt> Boyd is in India.
[13:12:55] <edwardocallaghan> Ah yes, I think he said something about it
[13:13:06] <edwardocallaghan> I forgot :P
[13:13:09] <Berny> i walk past my dsl on the way to my car :-)
[13:13:22] <Berny> dslam i mean
[13:13:34] <edwardocallaghan> Doc:I did not get a pm or email from you yet mate/?
[13:14:04] <edwardocallaghan> sell your DSLAM on ebay, your get $$$ :D
[13:14:10] <Doc> i tried to send an email to victoredwardocallaghan _ At _ yahoo _ . _ c0m [.] au but it bounced
[13:14:17] <Doc> are the _ supposed to be -'s ?
[13:14:33] <Gekkko> >_>
[13:14:42] <edwardocallaghan> no, just get ride of them
[13:14:57] <edwardocallaghan> Doc:Just pm me
[13:15:15] <delewis> oh, the torture.
[13:15:28] <edwardocallaghan> delewis:Hey
[13:15:31] *** Corpuscule[AWAY] is now known as Corpuscule
[13:15:42] <edwardocallaghan> Hows that torture going?
[13:15:51] <delewis> much better now, unfortunately.
[13:16:05] <edwardocallaghan> ha!
[13:16:54] *** whaq_ has joined #opensolaris
[13:17:26] <edwardocallaghan> whats the OpenSolaris main RSS feed URL ?
[13:17:47] <dlg> hg clone
[13:18:19] <Gekkko> all the best tech news comes from apcmag.com in my opinion.
[13:18:23] *** whaq has quit IRC
[13:18:29] <MousePad> http://opensolaris.org/jive/rss/rssmessages.jspa?forumID=76
[13:20:09] <edwardocallaghan> Tahnks
[13:20:13] <edwardocallaghan> All fixed up
[13:23:56] *** coffman_zzz is now known as coffman
[13:26:01] <Tempt> Sweet; just got myself out of having to reinstall Linucks on my craptop
[13:26:44] <edwardocallaghan> lol
[13:27:17] <edwardocallaghan> Let me guess; The RPM db decided to poo its self ?
[13:27:45] <Tempt> No, my girlfriend is installing it for me
[13:27:54] <edwardocallaghan> LOL !
[13:28:10] <edwardocallaghan> I got my RSS feed working on the side of my site
[13:28:24] <Tempt> I can't wait for OpenSolaris to be usable on that craptop.
[13:28:47] <edwardocallaghan> Tempt:You have to put that as our cover story ~
[13:29:12] <edwardocallaghan> thats jokes
[13:29:27] <edwardocallaghan> Oh; I am doing Java classes now at CIT
[13:29:38] *** Tpenta has quit IRC
[13:29:41] *** Gman has quit IRC
[13:29:46] <edwardocallaghan> So I made a start on a Blastwave GUI front end
[13:30:04] <edwardocallaghan> however I plan for it to be useable with other sites as well
[13:30:43] <edwardocallaghan> i.e. Java frontend/shell script backend kind of thing
[13:30:53] <migi> edwardocallaghan, there is such project
[13:31:02] <edwardocallaghan> I need to learn more OOP though
[13:31:17] <edwardocallaghan> migi:Panels ?
[13:31:30] <migi> edwardocallaghan, http://www.opensolaris.org/os/project/jds/tasks/jpack/
[13:32:11] <edwardocallaghan> ok thats new!
[13:32:17] <edwardocallaghan> I did not know that
[13:32:36] *** linux_user400354 has quit IRC
[13:32:37] <migi> edwardocallaghan, and you are free to help working on that :) :)
[13:33:06] *** MousePad has quit IRC
[13:33:32] <migi> edwardocallaghan, did you have any working solution for your GUI project?
[13:34:03] <kaiwai> hmm, I assume they're going to get rid of that horrid thing called prodreg
[13:34:06] <edwardocallaghan> Not really
[13:34:39] <edwardocallaghan> I will join up to the project as soon as I had some more Java classes
[13:34:42] <kaiwai> who ever decided to use write prodreg in java should be given a special place deep in the bowels of hell
[13:34:42] <migi> actually I have started the JPack thing...
[13:34:54] <edwardocallaghan> and there is less load on me from my site
[13:35:03] <edwardocallaghan> Ah !
[13:35:18] <andyshack> http://security.rackhost.net.au/ibm.jpg <-- is anyone able to idenitfy that system for me ? i took a really bad photo of it and didnt realise. i THINK it said it is an AS400. any ideas ? oh, its about 4.5ft tall :)
[13:35:19] <edwardocallaghan> Well migi we have lots to talk about
[13:35:30] <migi> edwardocallaghan, any time you want :)
[13:35:55] <edwardocallaghan> .. I got a little time left here
[13:36:39] <migi> as everyone probably
[13:36:58] <edwardocallaghan> hehe
[13:37:07] <edwardocallaghan> So what have you done in summary ?
[13:37:11] <kaiwai> hmmm, rackhost - wouldn't that be called a bra?
[13:37:15] <edwardocallaghan> and what in ?
[13:37:25] <deather> andyshack: I think it's an AS/400 too, though not absolutely sure
[13:37:28] <Tempt> kaiwai: If it was particular strong rack hosting, it could be a corset
[13:37:48] <edwardocallaghan> its a ThinkBrick
[13:39:55] <andyshack> edwardocallaghan : any idea what sort of think brick it is ?
[13:40:30] *** dunc_ has joined #opensolaris
[13:40:42] <edwardocallaghan> _laptop_ :P no clue sorry
[13:42:53] *** rbrown has joined #opensolaris
[13:44:05] <andyshack> cool
[13:48:26] <kaiwai> dear god the US is protectionist
[13:48:35] * kaiwai watching BBC World
[13:51:43] <edwardocallaghan> Doc:You still around ?
[13:53:01] *** tombhadAC has quit IRC
[13:54:43] <edwardocallaghan> OK well I got to get off my mates computer and have a beer
[13:54:52] <edwardocallaghan> See you lads laters :D
[13:54:58] *** tombhadAC has joined #opensolaris
[13:55:01] *** edwardocallaghan has quit IRC
[13:57:36] <kaiwai> well, that just confirmed it, I'm emo
[13:57:56] *** calumb is now known as calLNCH
[13:59:44] <trochej> kaiwai: You just found out? I think whole channel knows it for long time
[14:02:59] <Tempt> Something to do with the bitching and whining?
[14:03:46] <Tempt> I gather nobody has fixed PCA yet?
[14:05:12] <quasi> Tempt: is it broken?
[14:05:45] <Tempt> Yep
[14:05:55] <Tempt> Sun changed things and it doesn't work currently.
[14:06:24] <quasi> damn, it was such a nice way to get patches without a contract
[14:06:25] <kaiwai> trochej: just really surprised about the foreign ownership rules of domestic airlines
[14:06:35] <rbrown> http://www.rodrickbrown.com/docs/images/myspace.png
[14:06:42] <rbrown> god myspace sucks haah
[14:07:05] <quasi> rbrown: then don't use it
[14:07:49] <hile_> pca works fine for me
[14:08:19] <quasi> my old version doesn't seem to work
[14:08:32] <nightswim> well the pca maintainer is on a holiday as mentioned on the site innit
[14:08:47] *** laca has joined #opensolaris
[14:08:49] <nightswim> so I bet it'll be looked into as soon as he gets back
[14:09:08] <quasi> Tempt: have you tried downloading patchdiag.xref by hand?
[14:09:20] <Tempt> Oh, I'll get around to it soon
[14:09:26] <Tempt> That is the fix - there was a discussion on comp.unix.solaris
[14:09:29] *** ichigo has joined #opensolaris
[14:09:52] *** master_of_master has joined #opensolaris
[14:10:12] <quasi> looking at the file I got, it was the bloody accept terms page that you get every time you hit sunsolve
[14:12:25] *** Gekkko has quit IRC
[14:12:52] *** dunc_ has quit IRC
[14:13:00] *** bahadunn has joined #opensolaris
[14:19:50] <Tempt> quasi: Yep, downloading the file by hand works
[14:21:41] *** master_o1_master has quit IRC
[14:22:05] <hile_> i find that it still works for me if I use my sunsolve login in the script.. Or am I mistaken?
[14:22:43] <quasi> hile_: I tried that, but no luck
[14:22:44] <nightswim> perhaps your xref file is cached
[14:22:48] <nightswim> or recent enough
[14:23:07] <nightswim> I think the downloading of the patches is fine
[14:23:08] <quasi> hile_: however the login page seems to appear pretty randomly
[14:23:13] <nightswim> oh
[14:26:40] <Tempt> pca has required a sunsolve login for patches since sunsolve did
[14:26:51] <Tempt> but sunsolve used to let you get the xref file without being logged in
[14:28:01] *** WickedWicky has quit IRC
[14:29:12] <jmcp> Tempt: you're showing your age
[14:29:33] <PerterB> I had problems with pca yesterday... it didn't want a login to get the xref file, but wanted you to click past that lame "Accept these terms" page they have everywhere now
[14:29:53] <quasi> PerterB: yep, that's exactly what we're seeing now
[14:29:59] <Tempt> jmcp: Indeed.
[14:30:11] <jmcp> :)
[14:30:12] <PerterB> oh sorry, I should have scrolled back more
[14:30:25] <Tempt> Sunsolve has sadly become an abortion
[14:30:31] <Tempt> Sun should be ashamed.
[14:30:40] <quasi> abortion?
[14:30:47] <PerterB> yeah
[14:30:52] <jmcp> quasi: it's bug a mere shadow of its former self
[14:30:59] <quasi> but yeah, they should be ashamed
[14:31:10] <Tempt> Is the internal stuff better?
[14:31:11] *** perlmonk has quit IRC
[14:31:11] <Tempt> Less crap?
[14:31:12] <quasi> bloody incompetents ;)
[14:31:16] <jmcp> Tempt: unsure, really
[14:31:31] <Tempt> Hang on...
[14:31:34] <Tempt> You're an employee.
[14:31:38] <Tempt> Go look and tell us!
[14:31:40] <jmcp> no, I'm a contractor
[14:31:47] <Tempt> Contractors don't get access to internal sites?
[14:32:28] <jmcp> oh, we do... but there's a slight difference between contractor and employee
[14:32:29] * jmcp pedants
[14:32:51] <quasi> I'm sure they want to make sunsolve so bad that people will give up and buy SUCS
[14:33:30] <Tempt> SUCS?
[14:33:37] <Tempt> Update Connection
[14:33:38] <Tempt> ?
[14:33:39] <jmcp> Sun Update Connection Service
[14:33:40] <quasi> yeah
[14:33:47] <Tempt> I like SunSolve for the information
[14:33:52] <Tempt> You know, getting hardware info and stuff
[14:33:54] <jmcp> quasi: SunUC does *not* replace sunsolve
[14:33:56] <quasi> s/Service/Sattelite/
[14:34:02] <Tempt> I wish they'd just ship the fucker on a CD/DVD
[14:34:04] <quasi> jmcp: I know
[14:34:19] <Tempt> I'd swear the reason they stopped was Craig Dewick republishing it
[14:34:26] <iMax> Tempt: good idea...although suns search engine needs improvement :)
[14:34:42] <Tempt> I think they stopped to sell "i.RunBook" modules
[14:36:01] <LeftWing> Apparently the OSC stuff is being replaced towards the end of the next US Summer.  Perhaps it'll all get better.
[14:36:50] <Doc> bawhawhahwa
[14:36:55] <Doc> bawhawhahwa
[14:36:56] <Doc> bawhawhahwa
[14:36:59] <LeftWing> Oh noes, it's Doc.
[14:37:04] <LeftWing> He knows better.
[14:37:06] <Doc> sorry.. but there were 3 distinct things that needed to be laughed at in that statement
[14:37:17] <Doc> it's being replaced on the 23rd of July
[14:37:22] <LeftWing> ...
[14:37:23] <Doc> no.. wait..  the 8th of August
[14:37:28] <delewis> Doc, have you ever been told that you're a pessimist?
[14:37:29] <Doc> no.. wait.. the end of August
[14:37:33] <LeftWing> Doc: hahaha
[14:37:34] <LeftWing> Awesome.
[14:37:37] <Doc> no.. wait.. October sometime
[14:37:48] <Doc> (hint: those are the official dates, not just ones I've made up)
[14:38:03] <LeftWing> Yeah, well.
[14:38:17] <LeftWing> Perhaps one day.
[14:38:21] <Doc> as far as getting better.. umm.. good luck.  personally i'll just be impressed if it actually works
[14:38:28] <LeftWing> That would be nice.
[14:41:01] *** cypromis has quit IRC
[14:41:31] <delewis> ugh, everytime I build rxvt, I forget to enable mousewheel support.
[14:41:33] *** WickedWicky has joined #opensolaris
[14:42:35] <Doc> sucks to be you, apparently
[14:42:37] <Tempt> You should build it once and then package it ;)
[14:42:51] *** kaiwai has quit IRC
[14:43:45] *** yongsun has quit IRC
[14:44:52] <delewis> Tempt, newer version.
[14:44:57] *** ErwFood is now known as Erwann
[14:44:58] <Tempt> Aah
[14:45:03] <Tempt> Then you should maintain packages for us all ;)
[14:45:08] <Berny> and while you're at it... create a spec file and drop in the sfe repo ;-)
[14:45:34] <delewis> a specfile? I only work with pkginfo/prototype files :-)
[14:46:18] <jmcp> delewis: ah, a Real Man(tm)
[14:47:56] <Berny> oh well then drop of a binary package ;-)
[14:48:10] <Cyrille> open a hex editor and write the package stream manually.
[14:48:35] *** calLNCH has quit IRC
[14:48:37] <dlg> real men use cat > foo.pkg
[14:51:01] <hile_> delewis, you use pkginfo/prototype on AIX too?
[14:51:22] <delewis> yep.
[14:51:37] <delewis> I always install the System V compatibility tools, which includes the SVR4 packaging stuff on an AIX system.
[14:51:52] <delewis> no way in fuck am I creating .bff's.
[14:52:19] <jmcp> no bff's because you've got no best friends?
[14:52:30] * jmcp endures a reality shakeup .....
[14:52:51] <hile_> oh cool.
[14:53:40] *** Gekkko[PDA] has joined #opensolaris
[14:53:53] <Gekkko[PDA]> eww non-irssi
[14:56:09] *** sniffy has joined #opensolaris
[15:00:58] <Tempt> delewis: gimmeh POWER4 machine
[15:01:30] <trygvis> are there any books covering smf and fm?
[15:01:47] <trygvis> their APIs in particular
[15:06:52] *** nostoi has joined #opensolaris
[15:07:20] *** victor has quit IRC
[15:07:27] *** victor__ has joined #opensolaris
[15:07:38] *** perlmonk has joined #opensolaris
[15:09:05] *** postwait has joined #opensolaris
[15:10:32] <Gekkko[PDA]> hmm.
[15:11:52] *** axle_512 has joined #opensolaris
[15:17:55] <Doc> docs.sun.com
[15:18:02] <Teknomancer> docs.sun.com is damn slow
[15:18:09] <Doc> still?
[15:18:17] <Teknomancer> always slow for me
[15:19:13] <quasi> Teknomancer: all the sun peeps were so happy about the new "lightning" fast docs. last week
[15:19:35] <Teknomancer> hmm
[15:19:36] <quasi> Teknomancer: along with the search that is now even more broken than it used to be
[15:19:47] <Doc> it's been upgraded.  nfi if it's any better, but people seems ot be claiming it was
[15:19:58] *** logic has joined #opensolaris
[15:20:01] <Doc> google does a better job of searching it than it's own search engine
[15:20:07] <quasi> Doc: the first couple of days it did seem faster
[15:20:25] <Teknomancer> well, why don't they use some traditional doc/html thing .. i'm sure there are fast ones around
[15:20:27] <Teknomancer> for such online docs
[15:20:33] *** logic has quit IRC
[15:20:43] <quasi> Doc: they've removed the option to search within a collection to be able to use the sun.com search engine
[15:21:12] <Teknomancer> why does it always throw duplicate search entries is another question
[15:21:25] <Teknomancer> i see 10 links for the same "ddi_umem_alloc"
[15:21:46] <Doc> because it sucks harder than a $10 whore?
[15:21:51] <Gekkko[PDA]> lol
[15:21:56] <Gekkko[PDA]> 2c.
[15:22:07] <mlh> google does it better job than  99% of websites own search
[15:22:16] <Doc> gekko: prices have gone up.  oh so i'm told anyway! :)
[15:23:25] <Gekkko[PDA]> lol Doc
[15:24:08] <Teknomancer> IRC from PDA ? hmm, man i feel ancient!!
[15:24:08] <Gekkko[PDA]> Embed google search >.>
[15:24:25] <Gekkko[PDA]> irssi from pda usually
[15:24:31] <Gekkko[PDA]> onscreen kbd.
[15:24:45] *** mrdeviant has joined #opensolaris
[15:25:03] *** logic__ has joined #opensolaris
[15:25:04] <Teknomancer> well i just used my first wireless net connection about a couple of weeks back, so yeah i'm still more in the past than u guys :P
[15:25:24] <Gekkko[PDA]> me was 5 months ago
[15:25:35] <Teknomancer> wireeless expensive there also?
[15:25:42] *** cmihai has quit IRC
[15:25:44] <Gekkko[PDA]> i had a P3 till 2006
[15:25:51] <Gekkko[PDA]> fsck yes.
[15:25:52] <Gekkko[PDA]> lol
[15:26:16] <Gekkko[PDA]> i used it,not paid for it
[15:26:16] <Teknomancer> P3s aren't so bad...
[15:26:24] <Teknomancer> haha
[15:26:25] <Gekkko[PDA]> lol at unguarded wifi.
[15:26:54] *** sarah has joined #opensolaris
[15:26:55] <Gekkko[PDA]> 8mb voodoo3's are.
[15:27:01] <Gekkko[PDA]> 128mb ram was
[15:27:14] <Gekkko[PDA]> cpu was alright?
[15:27:20] <Gekkko[PDA]> . not ?
[15:27:28] <Teknomancer> ?
[15:27:45] <Gekkko[PDA]> cpu was alright%.
[15:27:58] <Gekkko[PDA]> not question, statement.
[15:28:04] <Teknomancer> ah k
[15:28:11] <Gekkko[PDA]> :P
[15:28:39] <Gekkko[PDA]> vi with an onscreen kbd, hard
[15:28:46] <Gekkko[PDA]> emacs, suicide.
[15:28:54] <coffman> hah
[15:29:13] <Teknomancer> i never did like vi
[15:29:14] <coffman> buy yourself something with a keyboard. nokia n6x something
[15:29:25] <Gekkko[PDA]> asus eee!
[15:29:43] <Gekkko[PDA]> I need so much money >.>
[15:29:59] <Gekkko[PDA]> I've only saved $140 so far
[15:30:30] <Teknomancer> i'm still undecided if i should spoil myself by buying a personal media-player like Archos, whether i really will use it or not ...
[15:31:05] <Gekkko[PDA]> you'll get over the novelty
[15:31:30] <Gekkko[PDA]> my pda plays video, why buy an mp4  player ...
[15:31:37] *** logic_ has quit IRC
[15:31:40] <Teknomancer> how many GB u have on your PDA?
[15:31:49] <Gekkko[PDA]> 2gb
[15:31:52] <Teknomancer> err :)
[15:31:53] <Gekkko[PDA]> sd card.
[15:32:08] <Teknomancer> well yeah u can buy a lot of SD cards i suppose
[15:32:15] <Teknomancer> but a PDA is totally useless to me,
[15:32:23] <Teknomancer> don't travel, don't talk to people on the phone much
[15:32:28] <Teknomancer> most of its features will be useless
[15:32:35] <Gekkko[PDA]> neither do i
[15:32:40] <Teknomancer> don't need appointments
[15:32:45] <Gekkko[PDA]> my pda doesnt do phone
[15:32:50] <Teknomancer> don't need any word processors etc.
[15:33:02] <Gekkko[PDA]> neither do i
[15:33:04] <Gekkko[PDA]> lol.
[15:33:20] <Teknomancer> i'd rather buy a powerful media-player
[15:33:34] <Gekkko[PDA]> a pda isn't?
[15:33:37] <Teknomancer> larger screen, more space, recording capabilities, more codecs
[15:33:46] <Gekkko[PDA]> look up palm tx at least
[15:33:50] <Teknomancer> PDAs are for business people :P
[15:34:01] <Gekkko[PDA]> not all of them
[15:34:13] 
[15:34:14] *** Teknomancer has quit IRC
[15:34:28] <Gekkko[PDA]> tapwave zodiac was a gaming pda
[15:34:36] *** Teknomancer has joined #opensolaris
[15:34:40] <Gekkko[PDA]> is.
[15:34:41] <Teknomancer> sorry that was gaim's crash for the day
[15:34:53] 
[15:34:54] *** Teknomancer has quit IRC
[15:35:06] *** hile_ has quit IRC
[15:35:57] <Gekkko[PDA]> tx has a large screen
[15:36:10] <Gekkko[PDA]> nice cpu :P
[15:37:19] <Tempt> Get a sharp Zaurus CL-3200
[15:37:36] <Gekkko[PDA]> why?
[15:37:51] <Tempt> They're nice
[15:37:54] <Tempt> Usable.
[15:38:03] <Gekkko[PDA]> screw it, i need sleep
[15:38:04] <Gekkko[PDA]> ttyl
[15:38:08] <Tempt> I like them and believe everyone should like what I like because it does them good.
[15:39:04] *** Gekkko[PDA] has quit IRC
[15:39:43] * dlg heart the zaurus
[15:44:44] *** Fullmoon_ has joined #opensolaris
[15:44:54] *** KuArZo has joined #OpenSolaris
[15:54:03] *** Fullmoon has quit IRC
[16:00:37] *** bnitz has quit IRC
[16:05:39] *** sponix has joined #opensolaris
[16:07:33] *** sponix has quit IRC
[16:09:25] *** Cass has quit IRC
[16:10:55] *** Anthony_ has joined #opensolaris
[16:11:26] *** Anthony_ has quit IRC
[16:14:06] <quasi> "Considering this must mean at some level that the 8/2007 release is going gold in the next week or so"
[16:16:04] <quasi> (quoting a mail to zfs-discuss)
[16:17:22] <PerterB> I was under the impression it was due this week (forget where from)
[16:20:13] *** EchoBinary has joined #opensolaris
[16:26:59] <PerterB> irksome as I'm just building a server and would prefer not to have to liveupgrade after just a week
[16:28:54] *** cypromis has joined #opensolaris
[16:34:30] *** alanc_away is now known as alanc
[16:34:42] <alanc> s10 8/07 slipped
[16:35:18] *** Fullmoon_ has quit IRC
[16:36:13] <trygvis> weee
[16:41:58] *** hile_ has joined #opensolaris
[16:42:21] <bda> Slipped as in "slipped its leash and has been let loose upon an unsuspecting world" or slipped like "missed release"?
[16:43:12] *** Cass has joined #opensolaris
[16:44:47] *** Corpuscule_ has joined #opensolaris
[16:45:39] <quasi> alanc: yeah, from 7/07 or even more?
[16:46:48] <alanc> as in not going to be shipping next week either
[16:47:01] <quasi> damn
[16:47:18] <alanc> closer to end of the month
[16:51:00] <quasi> 9/07
[16:51:05] <bda> No crying GA and letting slip the dogs of U4, eh?
[16:51:47] *** CIA-27 has quit IRC
[16:54:55] *** Teknomancer has joined #opensolaris
[16:55:08] <Teknomancer> doesn't opensolaris define "environ" ??
[16:55:21] <Teknomancer> or have i accidentally corrupted my header fiel :(
[16:55:36] *** perlmonk has quit IRC
[16:56:04] *** _estibi_ has quit IRC
[16:56:05] *** Corpuscule has quit IRC
[16:57:06] <Teknomancer> can someone please do a search for "environ" in Unistd.h or stdlib.h or stdio.h and tell me if its' found? :(
[17:01:26] <Teknomancer> anyone? :(
[17:01:32] *** asim has joined #opensolaris
[17:02:28] <asim> hello all, this is general question.  i am new to solaris, where should i start?  i just got the media and plan to install it as a server this weekend.
[17:03:06] *** asim has quit IRC
[17:03:11] *** estibi has joined #opensolaris
[17:03:44] <PerterB> Teknomancer: nope, and I'm not sure it's defined in any header... it's an implementation detail rather than an allowed interface
[17:03:55] *** estibi is now known as _estibi_
[17:04:07] <Teknomancer> PerterB: hmm
[17:04:26] <PerterB> if you look at the source for getenv() and friends, you'll see they contain an explicit declaration for it
[17:04:37] <Teknomancer> PerterB:  where can i see the source?
[17:04:41] *** peteh has quit IRC
[17:05:02] <PerterB> http://cvs.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/lib/libc/port/gen/getenv.c
[17:05:05] <PerterB> line 50
[17:05:29] <Teknomancer> PerterB:  ok thanks
[17:05:33] <Teknomancer> let me try
[17:11:07] <Teknomancer> PerterB:  thanks, i also extern defined the environ, now it's fine.
[17:11:21] <Teknomancer> thanks a lot
[17:11:40] *** solar-star has quit IRC
[17:12:30] *** Corpuscule_ has quit IRC
[17:12:52] *** stevel has joined #opensolaris
[17:12:52] *** ChanServ sets mode: +o stevel
[17:14:15] *** NikolaVeber has joined #opensolaris
[17:17:08] *** CIA-26 has joined #opensolaris
[17:17:29] <Teknomancer> nite all
[17:17:32] <Teknomancer> and thanks
[17:17:37] *** Teknomancer has left #opensolaris
[17:20:37] *** pfa3rh has joined #opensolaris
[17:29:30] *** CIA-26 has quit IRC
[17:34:24] *** sfire||mouse has joined #opensolaris
[17:37:52] *** sommerfeld has joined #opensolaris
[17:37:52] *** ChanServ sets mode: +o sommerfeld
[17:44:44] *** Fullmoon has joined #opensolaris
[17:47:52] *** RaD|Tz has joined #opensolaris
[17:49:05] *** WickedWicky has quit IRC
[17:49:25] *** WickedWicky has joined #opensolaris
[17:51:25] *** reidms has quit IRC
[17:56:45] *** perlmonk has joined #opensolaris
[17:59:26] *** MattMan is now known as MattAFC
[18:05:41] *** nostoi has quit IRC
[18:05:57] *** sstallion has joined #opensolaris
[18:09:36] *** estibi has joined #opensolaris
[18:10:45] <Pietro_S> any ideas how to simplify/shorten this option '--find_dependencies'?
[18:11:59] <movement> Pietro_S: in what program?
[18:12:17] *** Corpuscule has joined #opensolaris
[18:12:20] <Cyrille> deps is occasionally seen instead of the full dependencies.
[18:12:37] <movement> but please use --find-deps not --find_deps
[18:12:57] <Cyrille> yes mixing dashes and underscores is usually a source for confusion
[18:13:10] <Pietro_S> thanks
[18:14:12] <Pietro_S> movement: for fronted of pkgtool
[18:15:29] *** Cyrille has quit IRC
[18:15:38] *** KermitTheFragger has quit IRC
[18:27:49] *** Erwann has quit IRC
[18:35:39] *** aruiz has quit IRC
[18:36:31] *** Dar is now known as Dar_HOME
[18:48:22] *** CIA-26 has joined #opensolaris
[18:50:39] *** alanc is now known as alanc_away
[18:51:48] *** tsoome1 has joined #opensolaris
[18:52:21] *** tsoome has quit IRC
[18:54:07] *** migi has quit IRC
[18:54:42] *** sahafeez has joined #opensolaris
[18:59:50] *** WickedWicky has quit IRC
[19:00:17] *** blueandwhiteg3 has joined #opensolaris
[19:00:56] *** Cass has quit IRC
[19:01:32] <blueandwhiteg3> Anybody know if the highpoint rocketraid works with solaris?
[19:06:44] <PerterB> check the HCL, I'm sure I saw at least one of their models in there
[19:07:03] <flyingparchment> isn't that fake raid?
[19:07:33] <PerterB> thought it could be used as vanilla SATA too?
[19:09:04] <flyingparchment> wow, they made sunsolve even more annoying
[19:09:15] <flyingparchment> now i can't even visit the patch page directly without having javascript
[19:09:20] <PerterB> impressive
[19:09:59] <holcomb> pca is also broken
[19:10:32] <bda> holcomb: Sun added a terms of service agreement page before you can download patchdiag.xref. :\
[19:10:39] <axisys> how do I detach a device from a zpool before importing it?
[19:10:42] <bda> If you grab the file and put it in place manually, it works.
[19:11:22] <holcomb> yeah
[19:11:28] <flyingparchment> are sun actively trying to annoy people who use their software?
[19:11:34] <axisys> hopefully pca author will fix that when he comes back on aug 12(?) from vacation
[19:11:40] <flyingparchment> some kind of reverse-psychology marketing?
[19:11:53] <richlowe> seemingly.
[19:12:04] <holcomb> it also works if you provide sunsolve login, but when getting the patchdiag.xref pca assumes you don't need a login
[19:12:29] <holcomb> sun should give that guy a job
[19:12:46] <axisys> is it possible to manually modify the zpool before importing the pool? i need to detach a bad disk.. if i import before detach i get tons of disks erros.. sytem stop responding "almost"
[19:12:50] <bda> holcomb: aha
[19:12:55] *** Corpuscule_ has joined #opensolaris
[19:13:33] <flyingparchment> what kind of moron designs a standard HTML form that requires javascript to submit, anyway?
[19:13:38] <PerterB> holcomb: nah, he failed to implement pca as a huge java GUI based monstrosity, there's no place for that kind of slapdash work at Sun...
[19:13:59] <flyingparchment> even though it's easy enough to work around, the sheer idiocy annoys me
[19:14:04] <axisys> PerterB: lol
[19:14:42] <holcomb> seriously.
[19:15:24] *** e^ipi has quit IRC
[19:15:25] <bda> They seem to break it on a pretty regular basis, too.
[19:15:26] <wesolows> javascript is an ordinary and expected feature of modern browsers
[19:15:27] *** e^ipi has joined #opensolaris
[19:15:45] *** rbrown has quit IRC
[19:15:47] <wesolows> there aren't many sites these days that offer full features without it
[19:15:53] *** e^ipi has left #opensolaris
[19:16:06] <flyingparchment> i don't use it, and the vast majority of sites i visit work fine (maybe missing some functionality)
[19:16:13] <flyingparchment> a few obviously need JS for what they do... sunsolve is not one of those
[19:16:24] <wesolows> considering that the alternatives are (a) java applets, (b) java webstart applications, (c) flash, and (d) activex goo, I think javascript is a great thing, and I'm happy to see people using it.
[19:16:38] <flyingparchment> the alternative?  the alternative is <input type="submit">
[19:16:40] * sfire||mouse loves noscript
[19:16:43] <wesolows> Well, that's true - there's no reason to use it where simpler stuff will do.
[19:17:12] <flyingparchment> i'm not complaining about JS, i'm complaining that JS is used here when there's absolutely no need.  and actually, even though that's been true a while, i still didn't complain.
[19:17:18] <PerterB> and presumably sun's own patching tools which also access sunsolve don't implement javascript either?
[19:17:28] <flyingparchment> the only reason i'm complaining now is that you can't even directly go to the patch page on sunsolve without submitting that form anymore
[19:17:31] <bda> They use their own magical suck.
[19:17:43] <wesolows> sunsolve has been sucking more for a long long time
[19:17:48] <bda> flyingparchment: I seem to have run into it when looking at documentation, too.
[19:18:05] <wesolows> used to be you could get patches by ftp...
[19:18:13] <bda> Glory days, eh?
[19:18:16] <wesolows> oh yeah
[19:18:35] <PerterB> *sigh* them were the days
[19:18:57] <flyingparchment> can i still order patches on tape? :)
[19:19:01] *** blueandwhiteg3 has quit IRC
[19:19:02] *** Mdx4 has joined #opensolaris
[19:19:02] <wesolows> wasn't that long ago, either - I remember doing it no more than 6 or 7 years ago.
[19:19:06] <bda> Is the mentality that if you make the free stuff more tedious to use, people will give you money for the stuff that supposedly isn't?
[19:19:17] <wesolows> flyingparchment: Sure.  How much are you willing to pay?
[19:19:19] <flyingparchment> wesolows: FTP patches went away really recently, like last year or something
[19:19:24] <PerterB> wesolows: it was only shut down recently, in the last year or so
[19:19:34] *** blueandwhiteg3 has joined #opensolaris
[19:19:51] <wesolows> bda: I'm sure there are people who think that way.  I don't work with any of them, though.
[19:19:51] *** gnut has joined #opensolaris
[19:19:54] <gnut> hi all.
[19:20:01] <PerterB> bda: but that fails to work here because it's just as tedious to get to the pay-for stuff!
[19:20:08] <flyingparchment> bda: i'd buy that, except sunsolve would still be annoying if i paid them.  :)
[19:20:19] <gnut> i'm trying to figure out why my computer is slow... and using dtrace, i noticed that Xorg calls clock_gettime quite a bit
[19:20:22] <gnut> is this normal?
[19:20:25] <wesolows> sun should basically blow up all of its web "properties" and start over
[19:20:31] <flyingparchment> (actually, i've been _trying_ to pay them, but the sun store is broken.)
[19:20:38] *** delphi1000 has joined #opensolaris
[19:20:49] <wesolows> gnut: I don't know, but have you calculated how much time is spent in those calls?  They should be very cheap.
[19:20:52] <PerterB> I'd be happy to help with that.... well, the blowing up part at least
[19:21:26] *** delphi1000 has quit IRC
[19:21:42] <bda> I did say "supposedly".
[19:21:50] <gnut> wesolows: okay. let me do a dtrace time thing
[19:22:05] *** e^ipi has joined #opensolaris
[19:22:17] <bda> Also, I was being facetious. I'm sure that isn't the mentality of the people making the decisions, but that certainly is the outcome. :\
[19:22:31] <gnut> wesolows: i noticed that when my computer feels slow, there are tons of icsw
[19:22:32] <bda> uh, except for how it doesn't make me want to give them money for it.
[19:23:00] <flyingparchment> bda: really, i'd rather believe they're (being forced to) do it because of that, than believe they want to help customers but are simply too incompetant to design a website
[19:23:01] <wesolows> bda: I don't think it's all that facetious - marketing people can be really evil.
[19:23:31] <flyingparchment> sorry, a web property
[19:23:32] <wesolows> flyingparchment: Sadly, it's almost surely the latter - otherwise store.sun.com would be much better than it is.
[19:23:49] <bda> store.sun.com got a hell of lot better this year.
[19:23:52] <bda> But it still does suck a lot.
[19:23:58] <wesolows> It did.  But it still sucks horribly.
[19:23:59] <PerterB> even so, surely someone, somewhere in charge of sunsolve must realise how much they're pissing off the paying customers too?
[19:24:08] <bda> "Oh, I want some servers. And some add-on cards -- hey, where did my servers go?"
[19:24:08] <wesolows> And there are also way too many "call sun for pricing" products.
[19:24:11] <bda> *sigh*
[19:24:11] <flyingparchment> i liked the bit when they stopped anyone living outside the US from visiting the sun store
[19:24:13] <PerterB> or do they think it's all hunky dory
[19:24:28] <flyingparchment> that made me really want to give sun money from my US organisation
[19:24:36] <flyingparchment> fortunately that's gone away now
[19:24:51] <wesolows> PerterB: I don't know.  Have you sent feedback?  Did you get a response?  If not, did you mail Jonathan?
[19:25:00] *** alanc_away is now known as alanc
[19:25:36] *** CIA-26 has quit IRC
[19:26:09] <bda> wesolows: I had a weird experience with the Sun store earlier this year. Ordered half a dozen boxes, and two minutes later a woman who had to be chemically enhanced to be as chipper as she was, called and asked if I enjoyed using the site.
[19:26:10] <PerterB> I don't see a way to send feedback, and emailing jonathan seems a bit extreme (I have whined at account managers in the past though, and got a kind of "well, what can you do" response)
[19:26:13] <axisys> so zpool expert?
[19:26:16] <bda> It creeped out everyone in the office.
[19:26:28] <axisys> any zpool expert I meant
[19:26:37] <PerterB> oh wait, there's the feedback link, duh
[19:26:47] <bda> "make site go more harder plzkthxbai"
[19:27:10] <wesolows> PerterB: You know, jis is a really good CEO.  If he hears things consistently from customers, he'll take action.  But if everyone thinks it's "too extreme" but builds up simmering resentment of Sun, who wins?
[19:27:20] <PerterB> fair point
[19:27:51] <PerterB> personally it doesn't build any resentment for Sun as an entity, it just makes me want to hurt the people responsible
[19:28:25] <RElling> axisys: assuming the zpool is redundant, have you tried removing the bad disk and importing?
[19:28:26] <wesolows> But you'll never find those people, because they're cloaked in anonymity and shielded from the consequences of their incompetence by your disinclination to call them out.
[19:28:33] *** Corpuscule has quit IRC
[19:28:50] <bda> Also, they're wrapped in enigmas.
[19:29:13] <PerterB> yeah... don't worry I shall send feedback once I get a few minutes to compose something coherent
[19:29:26] *** nostoi has joined #opensolaris
[19:30:12] <PerterB> but surely there should be internal pressure from people at Sun? Or is this a case of them not eating their own dogfood?
[19:30:24] <wesolows> It depends.
[19:30:26] <alanc> gnut: Xorg switched from gettimeofday to clock_gettime recently - it uses it for timers and such (like keeping track of how long you've been idle to determine if it's time to blank the screen)
[19:30:34] <wesolows> For sunsolve and store.sun.com, we don't eat our own dogfood.
[19:30:46] <alanc> it shouldn't be that expensive of an operation
[19:31:01] <wesolows> For docs, there's enormous pressure but the organisation is just too ineffective and incompetent to fix the problem.
[19:31:02] <PerterB> you should, maybe it would help to improve the taste
[19:31:39] <wesolows> PerterB: Well, we can't, really.  We order stuff through a tool called CAMS, because things are accounted for differently.
[19:32:00] <wesolows> PerterB: And trust me, it's no better than store.sun.com - although it is much more flexible, in ways you wouldn't want, actually.
[19:32:03] <gnut> alanc: okay. thanks. i'm trying to track down the slowdown. apparently, now my key repeat rate is amazingly slow and takes about 3 seconds to activate
[19:32:11] *** bubbva has joined #opensolaris
[19:32:32] <wesolows> PerterB: For sunsolve, I have no use for it - I only run Nevada, so I never use patches.
[19:33:31] <RElling> internally, we have a sunsolve mirror, but it doesn't have the ask-for-login bits activated
[19:33:33] *** trede has joined #opensolaris
[19:33:45] *** WickedWicky has joined #opensolaris
[19:35:40] <WickedWicky> howedy
[19:36:38] <alanc> the only thing I use the internal sunsolve mirror for is searching the full text of bugs, since it works better than bugster's full text search
[19:37:13] <alanc> docs.sun.com at least got faster in their recent rollout of a new server - usability still sucks though
[19:37:33] <wesolows> and of course no permalinks
[19:38:23] *** polk_ has quit IRC
[19:40:02] *** halton has joined #opensolaris
[19:40:15] *** halton has left #opensolaris
[19:40:20] *** WickedWicky has quit IRC
[19:40:21] *** baijiutong has joined #opensolaris
[19:40:52] *** alanc is now known as alanc_away
[19:42:12] <axisys> RElling: physcially removing? I can go ahead and do that
[19:43:06] <RElling> axisys: the timeout... wait... retrying... wait... slow mode... occurs when a disk isn't fully dead
[19:45:26] *** CIA-17 has joined #opensolaris
[19:45:52] *** WickedWicky has joined #opensolaris
[19:47:26] *** eboutilier_ has quit IRC
[19:47:30] <bda> 13:43 < helen> bda: X4100 just arrived.
[19:47:31] <bda> eee!
[19:47:44] <bda> And I think "Spook Country" is also going to be delivered to the office today.
[19:47:53] <bda> It's like Christmas!
[19:48:01] <bda> Only in August and almost 100F.
[19:49:48] <axisys> RElling: aight.. let me head down to the computer room.. brb
[19:52:45] <elektronkind> hey weslows
[19:52:46] *** nostoi has quit IRC
[19:52:52] <elektronkind> question for ya
[19:53:10] *** Corpuscule_ is now known as Corpuscule
[19:58:26] *** baijiutong has quit IRC
[20:00:06] *** dunc_ has joined #opensolaris
[20:00:17] <CIA-17> rscott: 6589236 mkbfu and bfu argue over $(PLATFORM) vs. $(CLASS)
[20:00:17] *** gnut has quit IRC
[20:02:37] *** Yamazaki-kun has joined #opensolaris
[20:11:18] *** hile_ has quit IRC
[20:14:17] <Pietro_S> Any volumteree for testing utility for finding pkgtool dependecies?
[20:14:26] <WickedWicky> sure
[20:15:25] <Pietro_S> WickedWicky: you need ruby, sqlite both they are in SFE repository
[20:16:36] <WickedWicky> building
[20:16:37] *** aruiz has joined #opensolaris
[20:19:27] <Pietro_S> also this http://rubyforge.org/frs/download.php/17096/sqlite3-ruby-1.2.1.tar.bz2 , in README you have advices how to install it (eg. ruby setup.rb ...)
[20:21:04] <WickedWicky> allrighty
[20:21:05] <laca> Pietro_S: no spec file for sqlite3-ruby?  :O
[20:21:21] <WickedWicky> hey laca
[20:21:27] <laca> hey hey
[20:21:39] <WickedWicky> all works fine since I use the 2GB partition, thank you
[20:21:48] <oxygene> Pietro_S: what exactly is your tool scanning for?
[20:22:09] <laca> WickedWicky: you fixed it, not me
[20:22:21] *** Shiv__ has joined #opensolaris
[20:23:22] <WickedWicky> sqlite is built, now ruby
[20:24:09] <Pietro_S> oxygene: it eats spec files directory and put it to database(sqlite) and then allow you to do some things over database without parsing spec files
[20:25:00] <Pietro_S> right now only 2 things are working: import and searching dependencies
[20:25:33] <oxygene> Pietro_S: that's why I asked - I thought it's for helping find dependencies to track for the spec files themselves
[20:26:36] <laca> oxygene: alanc has a script for that (check_deps.pl) i'm sure it was posted somewhere but i can't find it now
[20:27:09] <Pietro_S> but later hopefully it will be able to search patch which fix error message
[20:27:14] <oxygene> laca: it would be just out of curiosity, and to compare it with my tools
[20:27:29] <laca> heh
[20:27:38] <laca> is your tool public?
[20:28:23] <oxygene> laca: part of pmpkg
[20:28:36] <laca> oh, okay
[20:28:38] <Shiv__> Pietro_S: Hi
[20:28:46] *** RaD|Tz has quit IRC
[20:28:50] <laca> hey Shiv__
[20:28:55] <Pietro_S> Shiv__: heelo
[20:29:00] <Shiv__> Laca: Hi
[20:29:13] <Shiv__> Pietro_S: Why a database? Why not just a flat file?
[20:29:49] *** MattAFC is now known as MattMan
[20:30:11] <Shiv__> Pietro_S: Flat file makes it easier for anybody to provide additionally tooling.
[20:31:06] <oxygene> they also probably make it harder for him to write the tool in the first place ;)
[20:31:13] *** mrdeviant has quit IRC
[20:32:47] <Pietro_S> and sqlite database is faster
[20:33:40] <Pietro_S> Shiv__: making it in flat file doesn't make sence, if you want to have it in file you can search directly spec files
[20:33:50] <e^ipi> and easier to scale up to a real database if your dataset gets bigger
[20:34:15] *** g4lt-U60 has quit IRC
[20:34:18] <oxygene> e^ipi: not necessarily
[20:34:42] <oxygene> e^ipi: sqlite, by being embedded, sometimes leads to different usage patterns - some that don't translate well to client/server
[20:34:43] *** Mdx4 has quit IRC
[20:35:14] <Shiv__> Pietro_S: Not really. The final value a particular tag takes is not known if it is parameterised which is the case in all SFE specs.
[20:35:50] *** LordKing has joined #opensolaris
[20:37:22] *** cypromis has quit IRC
[20:37:26] <Pietro_S> laca: I hope that you don't mind that I named it pkgbase, I'll post message to pkgbase-discuss about it (I would like to know your opinion about it
[20:37:59] <Pietro_S> Shiv__: all variables are parsed before inserting to database
[20:39:10] <laca> Pietro_S: go for it
[20:39:17] <Shiv__> Pietro_S: Ok. Will wait for your announcement. Do you intend to provide APIs/scripts to interface with the db.
[20:39:42] <Pietro_S> Shiv__: I will public everything
[20:39:51] *** vmlemon has joined #opensolaris
[20:40:33] <WickedWicky> Pietro_S: done
[20:40:36] *** MattMan is now known as MattAFC
[20:41:06] *** cypromis has joined #opensolaris
[20:41:08] <Pietro_S> WickedWicky: here is it: http://home.zcu.cz/~sobotkap/pkgbase.tar.bz2
[20:41:43] <Pietro_S> after unpacking go to lib directory
[20:41:46] <andyshack> yo WickedWicky : re the 3310 arrays im tossing up, do they take the same generic sun style spud bracket on the drives ?
[20:42:35] <sommerfeld> 33x0 and 35x0 don't use the spud bracket
[20:42:47] <WickedWicky> Pietro_S, there
[20:43:02] <Shiv__> Pietro_S: Do I need to install anything to use it (SQlite, ruby,...)
[20:44:52] *** dkl has joined #opensolaris
[20:45:31] <Pietro_S> WickedWicky: and run: pkgtool --import path_to_your_spec_repository
[20:45:42] <Pietro_S> Shiv__: yes you need ...
[20:46:06] <Pietro_S> s/pkgtool/ruby pkgbase
[20:46:59] <WickedWicky> pkgbase.rb i assume?
[20:47:03] <Pietro_S> yes
[20:47:38] <dkl> so, has anyone managed to get iftop working in a s10 zone?
[20:49:24] <e^ipi> this flickering is going to drive me up the goddamn wall
[20:49:45] <WickedWicky> WARNING: Failed to parse SFElibopensync-plugin.spec in /export/build/specfiles/SFE directory
[20:49:45] <WickedWicky> specinfo.sh failed to get enough informations (for file /export/build/specfiles/SFE/SFElibopensync-plugin.spec) - check your jds-cbe installation
[20:49:50] <e^ipi> 13w3 -> vga converters suck
[20:50:00] <quasi> dkl: I very much doubt that would work
[20:50:29] *** iceq has quit IRC
[20:50:33] <dkl> solaris isn't my speciality, but i assume there should exist a /dev/iprb0 or /dev/iprb0:15 in the global zone?
[20:51:36] <Pietro_S> WickedWicky: it will omit this file, but can you run ./specinfo.sh /export/build/specfiles/SFE SFElibopensync-plugin.spec
[20:51:52] <WickedWicky> it is giving me this error for heaps of files
[20:52:29] <WickedWicky> ah
[20:52:37] <Pietro_S> there is " " between SFE dir and SFElib...
[20:52:38] <WickedWicky> specinfo.sh
[20:52:48] <quasi> dkl: there's special rules for access to network interfaces inside a zone - something like promisc wouldn't fly
[20:52:53] <WickedWicky> specinfo.sh should be in my path I assume
[20:53:10] <Pietro_S> no it's in lib dir
[20:53:13] <WickedWicky> well
[20:53:14] <WickedWicky> yes
[20:53:25] <WickedWicky> but your script doesnt try to execute ./specinfo.sh
[20:53:30] <WickedWicky> it tries to execute specinfo.sh
[20:53:44] <WickedWicky> PATH=$PATH:`pwd`; export PATH
[20:53:47] <WickedWicky> solves it
[20:54:07] <dkl> iftop claims to work without promisc, but i see nothing plausible in /dev
[20:54:10] <WickedWicky> $ PATH=$PATH:`pwd`
[20:54:10] <WickedWicky> $ export PATH
[20:54:10] <WickedWicky> $ ruby pkgbase.rb --import /export/build/specfiles/SFE
[20:54:10] <WickedWicky> Starting importing spec files
[20:54:10] <WickedWicky> Parsing spec files.
[20:54:23] <dkl> ifconfig reports iprb0:15 and iprb0:16
[20:54:57] <Pietro_S> ah, I have to fix it "./" should be there, I didn't know it because I have . in my PATH
[20:55:17] <WickedWicky> that's why you have testers :D
[20:55:22] <WickedWicky> let me fetch a coffee
[20:56:33] <dkl> quas: any other ideas for monitoring conn/bw within the zone?
[20:58:10] *** cypromis has quit IRC
[20:58:16] <quasi> dkl: do it outside the zone ;)
[20:58:36] <quasi> dkl: or use the os counters
[20:58:41] <dkl> ..if i've no other access :)
[20:58:46] <dkl> via dtrace?
[20:58:49] <quasi> kstat, netstat and other bits
[20:59:57] <quasi> dtrace usually isn't allowed in zones (but can be) and doesn't have very good support for network bits yet (although brendang is hard at work fixing providers for net)
[21:00:18] <CIA-17> josephb: 6582841 boot_/final_kerneheap and kernelheap_extend() nonsense needs to be ripped out, 6507265 amd64 Solaris will fail on more than 1 Terabyte of memory
[21:00:19] <CIA-17> hylee: 6585894 pktool import usage shows bogus nickname parameter, 6585913 pktool import keystore=nss does not work for PEM encoded certs
[21:00:59] <dkl> ok, so if i were to do something with kstat what would it be
[21:01:16] <dkl> it's successfully spamming my console at high speed
[21:01:35] * quasi wanders off
[21:02:00] <dkl> haha
[21:02:01] *** Murmuria has joined #opensolaris
[21:02:02] <dkl> ok, i'll read up
[21:02:06] <dkl> thanks for the pointers
[21:04:32] <e^ipi> where on earth is dmake finding these object files?!
[21:04:46] <e^ipi> it just pulls btopw
[21:04:55] <e^ipi> btowc_dense.o out of thin air
[21:05:27] <e^ipi> I don't even /have/ a copy of closed/root_sparc/lib/libc_i18n.a
[21:05:50] *** dkl has quit IRC
[21:06:40] *** hile_ has joined #opensolaris
[21:07:02] <Pietro_S> laca: SFEgtodo2.spec at line 40 : there is %endif without %if before it, it looks like a bug
[21:07:45] <WickedWicky> I am getting an error on that spec file as well
[21:07:55] <WickedWicky> but am
[21:08:06] <WickedWicky> your script implies it's the fault of my jds-cbe build
[21:08:32] <WickedWicky> WARNING: Failed to parse SFEgtodo2.spec in /export/build/specfiles/SFE directory
[21:08:32] <WickedWicky> specinfo.sh failed to get enough informations (for file /export/build/specfiles/SFE/SFEgtodo2.spec) - check your jds-cbe installation
[21:09:01] <Pietro_S> WickedWicky: on 90% it's bug in spec file
[21:09:27] <WickedWicky> so make it say "check your spec file" ;-)
[21:09:28] <Pietro_S> that error message is a bit unautorative
[21:10:43] <Pietro_S> actually it will fail if you have jds-cbe in other dir than /opt/jdsbld
[21:12:13] *** LordKing has quit IRC
[21:12:23] <WickedWicky> I have jds-cbe in /opt/jdsbld
[21:12:31] *** LordKing has joined #opensolaris
[21:13:02] *** Mdx4 has joined #opensolaris
[21:13:41] <axisys> RElling: I got lots of disk not responding alerts in console after physically removing the bad disk followed by zpool import -a .. is there a way to detach before import?
[21:15:09] <laca> Pietro_S: it's tomww's spec file
[21:15:33] <laca> tomww: an %if %build_l10n seems to be missing from line 38
[21:16:55] <laca> ... and an %else
[21:16:56] *** derchris has joined #opensolaris
[21:17:53] <richlowe> axisys: import, and replace it afterward?
[21:19:09] <RElling> axisys: AFAIK you can't detach unless it is imported.  The data about what disks are used is part of the pool.
[21:19:12] <axisys> richlowe: i get tons of disk error message and sytem gets almost halt
[21:19:40] <axisys> RElling: so there is no zpool config file where I can emergency hack to forget about a disk?
[21:20:05] <axisys> liek for svm where i choose /dev/null instead of /etc/system and modify vfstab
[21:20:07] <RElling> but I wouldn't worry too much about disk not responding, these are not as painful as a slow retry loop
[21:20:08] <richlowe> Hrm, if you pull the disk, shouldn't the pool import, but fault?
[21:20:12] <axisys> s/liek/like/
[21:20:14] <richlowe> but not stall out waiting on the disk o' doom.
[21:20:22] <richlowe> and then let you replace the disk in peace?
[21:20:37] <RElling> no, the pool configuration data is in the pool.   This is a good thing, actually.
[21:20:42] <Pietro_S> WickedWicky: when import finish, you can search for dependecies with this: ruby pkgbase.rb -f [package] or --find-deps [package]
[21:20:54] *** hile_ has quit IRC
[21:21:07] <WickedWicky> it's still importing
[21:21:18] <Pietro_S> WickedWicky: for example ruby pkgbase.rb -f SFEmplayer
[21:21:19] <WickedWicky> will it do recursive lookup?
[21:21:22] <WickedWicky> like
[21:21:29] <WickedWicky> package A has a dependency B
[21:21:29] <axisys> looks like i never attached the good disk.. damn.. only bad disk slice 7 has the zpool
[21:21:35] <WickedWicky> dependency B has a dependency C
[21:21:36] <axisys> time to slam the bad disk back in
[21:21:39] <WickedWicky> etc etc?
[21:21:50] <Pietro_S> yes recursive search over dependencies
[21:21:57] <WickedWicky> coolies
[21:23:31] <Pietro_S> it will display package names with .spec ending so you can copy it directly to pkgtool line
[21:24:07] <WickedWicky> that's nice. I was about to compile Mplayer actually so this testing comes fitting :)
[21:24:08] *** baijiutong has joined #opensolaris
[21:24:33] *** Shiv__ has quit IRC
[21:26:20] *** KuArZo has quit IRC
[21:30:21] *** hile_ has joined #opensolaris
[21:31:10] <WickedWicky> Pietro_S
[21:31:11] <WickedWicky> http://rafb.net/p/YCn3r438.html
[21:31:40] *** Gman has joined #opensolaris
[21:32:14] <richlowe> mornin' Gman
[21:32:37] <Gman> hey there
[21:33:49] <Pietro_S> WickedWicky: that's known bug, not affecting functionality
[21:35:08] <Pietro_S> right now packages which are defined in another's package spec file are not imported to database
[21:35:49] *** polk__ has joined #opensolaris
[21:36:50] <Pietro_S> I'm thinking about making for them special table, because they have lot's same tags as parent package
[21:38:21] <Pietro_S> WickedWicky: you can now try find dependencies for mplayer
[21:38:23] *** MattAFC is now known as MattMan
[21:39:53] *** BatonT has quit IRC
[21:41:14] <Pietro_S> WickedWicky: when you wiill copy dependency list, don't forget that SFEmplayer is not dependency for SFEmplayer package ;-)
[21:43:32] *** trede has quit IRC
[21:47:15] *** odix67 has joined #opensolaris
[21:49:46] *** dmesg has joined #opensolaris
[21:50:02] <odix67> ping
[21:50:02] *** dmesg has left #opensolaris
[21:50:16] *** freakazoid0223 has quit IRC
[21:51:28] <laca> Pietro_S: SFEgtodo2.spec fixed
[21:51:42] *** jamesd has joined #opensolaris
[21:51:42] *** ChanServ sets mode: +o jamesd
[21:51:49] <laca> Pietro_S: i have an awesome project suggestion for you (;
[21:52:08] <laca> can you write a tool that checks if there is a newer version of a tarball available?
[21:54:31] <oxygene> laca: how would you do it - guessing new version numbers? parsing directories? parsing changelogs?
[21:54:34] *** Triskelios has quit IRC
[21:54:53] <Pietro_S> nice feature
[21:55:17] <laca> that's the challenge (;
[21:55:50] <laca> oxygene: for some projects it's easy, you just need to find the highest version number in an ftp directory
[21:56:01] <oxygene> laca: that's the "parsing directories" bit, yes
[21:56:20] <laca> right
[21:56:27] <oxygene> for others (namely those on gforge sites), it's basically impossible, as they address files using an id in the path
[21:56:28] <richlowe> You need more resilience than that.
[21:56:34] * timeless frowns
[21:56:39] <timeless> [ Aug  8 13:49:05 Executing stop method ("/lib/svc/method/svc-syseventd stop 4815") ]
[21:56:40] <Pietro_S> oxygene: I done 1 class for taking care about version, but it needs additional spec variable ..., then it can be done easily
[21:56:41] <timeless> [ Aug  8 13:49:05 Restarting too quickly, changing state to maintenance ]
[21:56:44] <richlowe> both for projects that append alphabetic characters.
[21:56:51] * timeless wishes syseventd wouldn't do that
[21:56:54] <richlowe> and the more evil projects that will respin a tarball and not bump the version.
[21:56:56] <oxygene> ie. http://foo/download/55/bar and http://foo/download/55/baz give the same file, but with different names
[21:57:25] <oxygene> richlowe: sha1 over the directory text
[21:57:45] *** MattMan is now known as MattAFC
[21:57:58] <richlowe> oxygene: sure.
[21:58:00] <oxygene> gives some false positives, but if the directory available (and the test candidate is chosen wisely), it's a usable way
[21:58:07] <richlowe> oxygene: I was just pointing out that a remote directory walk isn't going to be enough.
[21:58:12] <timeless> richlowe, does this mean something: http://pastebin.mozilla.org/182617
[21:58:50] <Pietro_S> what about parsing website in %url, most project announce newer version at their main websites
[21:59:08] <oxygene> Pietro_S: and most announce lots of irrelevant crap, too
[21:59:40] <timeless> oh farn
[21:59:50] <timeless> my os upgraded itself!
[21:59:56] <timeless> that's bad
[22:00:01] <Pietro_S> yes but that crap doesn't have any link to tar.bz or something like it
[22:00:06] <timeless> hrm, no
[22:00:07] * timeless frowns
[22:00:09] * timeless doesn't get it
[22:00:18] <CIA-17> keerthi: 4815729 hasmntopt(3C) incorrectly finds substrings
[22:01:06] <laca> oxygene: there's no universal solution, but i suspect that a significant proportion of projects would fall into one of a handful categories
[22:01:24] <oxygene> laca: I'd just store the hash of some reference url and compare to that
[22:01:59] <oxygene> laca: that way, if that url can be chosen freely, it can be pointed to the best location - eg. directories or changelogs - anything beyond that is too much effort for the gain in my opinion
[22:02:07] <oxygene> (incidentally, that's on my todo list as well ;-)
[22:02:23] <oxygene> for pmpkg, that is
[22:02:40] *** comay has joined #opensolaris
[22:02:55] <richlowe> Hey comay.
[22:03:05] <laca> spec files do have a url field -- i usually point them at the download page
[22:03:14] <richlowe> timeless: I don't follow it either, but I have no idea what nexenta may have done.
[22:03:33] <timeless>  richlowe: it's odd, the os hasn't changed or anything
[22:03:44] <timeless> just rebooted the zone and got that spam on the console :(
[22:03:52] <timeless> (and it of course is in maintenance mode)
[22:04:05] <richlowe> also, software with a 'message.h' containing all localizable strings is quite possible the worst idea known to man.
[22:04:11] <comay> hey there richlowe
[22:04:29] <timeless> richlowe: context re message.h?
[22:04:40] <richlowe> timeless: syseventd does it. :)
[22:04:49] <timeless> heh
[22:05:00] <timeless> richlowe: i can deal w/ that
[22:05:13] *** stevel has quit IRC
[22:05:14] <timeless> it makes cross reference jumps from l10n errors to code fairly easy :)
[22:05:38] <timeless> what's worse is numerical errors w/ numbers like 42
[22:05:52] <timeless> where the numbers aren't in the code and can't easily be found anywhere else
[22:06:10] <Pietro_S> oxygene: changelog looks like easier way, hopefully most projects have changelog and lot's of them use some shelscript to create it
[22:08:44] *** andyshack has quit IRC
[22:09:07] <richlowe> Pietro_S: Yeah, but they tend to be inside the tarballs.
[22:11:11] <Pietro_S> they are also in cvs/svn/... , but I'm not sure if this one contain releases ...
[22:11:17] <WickedWicky> seems to work fine Pietro_S
[22:12:19] *** migi has joined #opensolaris
[22:12:51] <Pietro_S> WickedWicky: nice to hear, did mplayer compile well?
[22:13:19] <WickedWicky> no
[22:13:25] <WickedWicky>                    SFElibsndfile |       ERROR | Patch /lilith/user-dir/build/packages/SOURCES/libsndfile-02-cpp_test.diff not found
[22:14:10] <WickedWicky> but the dependency thingy is pretty cool
[22:15:21] *** rarruda__ has joined #opensolaris
[22:15:22] <Pietro_S> ah this package I had already installed
[22:15:35] <axisys> one of my client cannot boot from my bootserver.. http://rafb.net/p/hJRdR975.html
[22:16:01] <axisys> any idea why it says could not mount filesystem
[22:16:31] <axisys> i have done this probably 10 times for different servers that I am jumping.. this is the only one complaining
[22:17:26] <laca> WickedWicky: did you use pkgtool or pkgbuild?
[22:17:35] <WickedWicky> pkgtool
[22:17:37] <flyingparchment> what section is 7ipp?
[22:18:00] <laca> WickedWicky: hm... that should have copied libsndfile-02-....diff to SOURCES
[22:18:45] * timeless sighs
[22:18:55] <timeless> solaris's confirm shutdown prompt is kinda annoying
[22:19:18] <migi> WickedWicky, "pkgtool prep SFElibsndfile.spec" also gives this error?
[22:19:29] <WickedWicky> lemme check
[22:19:29] *** bahadunn has quit IRC
[22:20:04] <WickedWicky> no
[22:20:06] <WickedWicky> it passes
[22:20:11] <timeless> nexenta console login: syslogd: line 45: WARNING: loghost could not be resolved
[22:20:11] <WickedWicky>                          package |      status | details
[22:20:11] <WickedWicky>                    SFElibsndfile |      PASSED |
[22:20:15] * timeless wonders how to fix that
[22:20:28] <WickedWicky> put loghost in /etc/hosts
[22:20:46] <WickedWicky> #
[22:20:46] <WickedWicky> ::1     localhost       loghost
[22:20:46] <WickedWicky> 127.0.0.1       localhost       loghost
[22:22:32] <axisys> here is the snoop of that tftp http://rafb.net/p/vuQooE64.html
[22:22:50] <axisys> anything from the snoop that should tell me why the clien tfail to boot from bootserver?
[22:23:18] <WickedWicky> laca,migi where is it suposed to get that patch from?
[22:23:30] <WickedWicky> the spec file says:
[22:23:31] <WickedWicky> Patch1:                  libsndfile-01-flac-1.1.3.diff
[22:23:31] <WickedWicky> Patch2:                  libsndfile-02-cpp_test.diff
[22:23:50] <WickedWicky> but when I do a find . -name *.diff on the untar-gz-ed source there are no diff files
[22:23:57] <laca> WickedWicky: pkgtool should copy the patch from the patches subdir to SOURCE
[22:24:04] <laca> SOURCES i mean
[22:24:31] <WickedWicky> and this patches subdir is.. where?
[22:25:15] <migi> WickedWicky, /lilith/user-dir/build/packages/SOURCES/ in your case
[22:25:17] <laca> say you have your spec-files-extra specs in /home/foo/SFE, then the patches are in /home/foo/SFE/patches
[22:25:22] <timeless>       $ mdb -?      ... in your shell for a complete list of options
[22:25:25] * timeless sighs
[22:25:35] * timeless wishes that had said mdb '-?'
[22:26:13] <migi> WickedWicky, did you downloaded whole trunk from spec-files-extra?
[22:26:36] <WickedWicky> svn co https://pkgbuild.svn.sourceforge.net/svnroot/pkgbuild/spec-files-extra/trunk SFE
[22:26:53] <WickedWicky> as per http://pkgbuild.sourceforge.net/spec-files-extra/
[22:27:00] <laca> right
[22:27:05] <laca> so your patches are in SFE/patches
[22:27:15] <migi> try to find ./SFE -name "*libsndfile-02*"
[22:28:17] <WickedWicky> $ pwd
[22:28:17] <WickedWicky> $ ls *libsndfile-02*
[22:28:17] <WickedWicky> libsndfile-02-cpp_test.diff
[22:28:17] <WickedWicky> $
[22:28:20] *** aramdune has quit IRC
[22:28:48] <laca> pwd prints nothing?
[22:28:51] <WickedWicky> this is in /export/build/specfiles/SFE/patches , I did the svn co in the directory /export/build/specfiles]
[22:28:55] <WickedWicky> it does
[22:28:58] <WickedWicky> but it starts with a /
[22:29:01] <laca> oh, irc ate the / (;
[22:29:04] <WickedWicky> . /export/build/specfiles/SFE/patches
[22:29:15] <migi> and if prep goes ok the same should be in the "ls *libsndfile-02* /lilith/user-dir/build/packages/SOURCES/"
[22:29:36] <WickedWicky> $ ls *libsnd*
[22:29:36] <WickedWicky> libsndfile-01-flac-1.1.3.diff  libsndfile-1.0.17.tar.gz
[22:29:36] <WickedWicky> libsndfile-02-cpp_test.diff
[22:29:36] <laca> right, so cd .. and pkgtool build SFElibsndfile.spec should just work
[22:29:56] <laca> i mean cd .. from /export/build/specfiles/SFE/patches
[22:30:02] <migi> WickedWicky, did you make . /opt/jdsbild/bin/env.sh ?
[22:30:26] <WickedWicky> I did: '. /opt/jdsbild/bin/env.sh'
[22:30:46] <WickedWicky> Eterm built fine yesterday btw
[22:30:50] <WickedWicky> so did ruby and sqlite today
[22:30:52] *** sfire||mouse has quit IRC
[22:31:18] <migi> so maybe you are running out of space if the patch cannot be found
[22:31:23] <WickedWicky> lol
[22:31:33] <WickedWicky> this is a 1 terrabyte ZFS pool
[22:31:43] <migi> just idea ;)
[22:31:47] <WickedWicky> zpool_lilith/export   1.0T  6.4G  1.0T   1% /export
[22:32:02] <migi> and for /lilith ?
[22:32:17] <WickedWicky> zpool_lilith/build    2.0G  212M  1.7G  11% /lilith/user-dir/build
[22:32:38] <WickedWicky> I had to cap the build homedir to 2GB to avoid breakage due to large_filesystem
[22:32:41] <laca> nice machine, i'd love to have one of those here (;
[22:33:07] <WickedWicky> it's just an Athlon 2400XP+
[22:33:10] <WickedWicky> nothing fancy
[22:33:37] <migi> WickedWicky, I'll try to build it, but have to switch to vpn so be back
[22:33:46] <WickedWicky> I am not in a hurry
[22:34:02] *** reidms has joined #opensolaris
[22:34:12] <laca> WickedWicky: were you in the SFE dir when you tried to build mplayer?
[22:34:21] <WickedWicky> no
[22:34:40] <laca> you need to cd there so that .pkgtoolrc is found
[22:35:14] <WickedWicky> then that is the problem
[22:35:25] *** migi_ has joined #opensolaris
[22:36:11] <WickedWicky> I did a for x in `ruby pkgbase.rb -f SFEmplayer | grep -vi searching`; do pkgtool build .....
[22:36:32] <WickedWicky> yea
[22:36:34] <WickedWicky> it's building now
[22:37:17] *** stevel has joined #opensolaris
[22:37:18] *** ChanServ sets mode: +o stevel
[22:37:37] <migi_> WickedWicky, so what was the actual problem (I wasn't here)?
[22:37:54] <WickedWicky> the actual problem was me not being in SFE directory
[22:37:59] <migi_> :)
[22:38:16] <WickedWicky> thanks people :)
[22:38:29] <Pietro_S> WickedWicky: you should copy it manually, meanwhile I code -b --build option which invoke pkgtool in right dir ;-)
[22:38:50] <WickedWicky> yea, I tried to write a one liner
[22:39:21] <WickedWicky> so I did for x in `bla bla`; do pkgtool build /path/to/spec
[22:41:40] *** elijahwright has joined #opensolaris
[22:41:40] <laca> you can do for x in $(cd ....; bla bla); do ....
[22:41:45] <WickedWicky> yep!
[22:41:47] <WickedWicky> did that right now
[22:42:34] <Pietro_S> WickedWicky: don't forget that SFEmplayer won't be inside bla bla ;-)
[22:42:53] <Pietro_S> as it's not dependency of itself
[22:43:13] <migi_> WickedWicky, and some of the SFE does compile only on gcc (inkscape)
[22:43:52] <Pietro_S> WickedWicky: thanks for testing, laca and others thanks for help/nice chat
[22:44:39] <Pietro_S> I go to take small nap, befoe I will go for father to airport (he should arrive around 3:45am) ...
[22:44:47] <WickedWicky> Pietro_S: thanks
[22:45:38] *** hile_ has quit IRC
[22:46:47] *** Pietro_S has quit IRC
[22:50:40] <WickedWicky> migi_ when packages are gcc-minded this is normaly configured in the specfile by the CC= variable, no?
[22:52:14] *** migi has quit IRC
[22:52:29] <migi_> I am not sure, so I wan't answare
[22:53:18] <migi_> from the spec file:
[22:53:19] <migi_> %if %cc_is_gcc
[22:53:20] <migi_> %else
[22:53:20] <migi_> echo this version of inkscape uses gcc specific code
[22:53:35] <migi_> exit 1
[22:53:36] <migi_> %endif
[22:53:54] <WickedWicky> yea
[22:54:00] <WickedWicky> I saw it yesterday in libast.spec
[22:54:01] <WickedWicky> # This source is gcc-centric, therefore...
[22:54:01] <WickedWicky> export CC=/usr/sfw/bin/gcc
[22:54:01] <WickedWicky> # export CFLAGS="%optflags"
[22:54:01] <WickedWicky> export CFLAGS="-O4 -fPIC -DPIC -Xlinker -i -fno-omit-frame-pointer"
[22:54:20] <migi_> so yes
[22:55:33] <WickedWicky> ok well
[22:55:42] <WickedWicky> building mplayer and its dependencies will take a bit
[22:55:50] <WickedWicky> which gives me the opportunity to get some sleep :)
[22:55:56] <WickedWicky> thanks for your help migi and laca :)
[22:56:01] <migi_> np
[22:56:10] <WickedWicky> see you all tomorrow
[22:56:17] <laca> np, good night
[22:56:17] *** dkl has joined #opensolaris
[22:57:04] <dkl> anyone feel like debugging a rbac?
[22:58:36] *** BatonT has joined #opensolaris
[23:01:57] *** dkl has quit IRC
[23:05:50] *** eboutilier_ has joined #opensolaris
[23:06:36] *** yarihm has quit IRC
[23:08:31] *** eboutilier_ has joined #opensolaris
[23:09:50] *** cypromis has joined #opensolaris
[23:09:50] *** pbaumgartner has joined #opensolaris
[23:10:08] *** adultera1edjedi has joined #opensolaris
[23:10:28] *** odix67 has quit IRC
[23:13:20] *** noyb has joined #opensolaris
[23:15:52] *** sahafeez has quit IRC
[23:18:32] *** eboutilier_ has quit IRC
[23:19:10] *** eboutilier_ has joined #opensolaris
[23:22:39] *** tombhad-AC has joined #opensolaris
[23:24:25] *** pjlv has quit IRC
[23:24:51] *** BatonT has quit IRC
[23:24:56] *** tombhad-AC has quit IRC
[23:29:13] <coffman> i got a app that locks up my whole system when i hit a certain menu point
[23:29:24] *** tombhad-AC has joined #opensolaris
[23:29:25] <coffman> any way to dtrace that?
[23:30:53] * FireflyST is excited, his new (hopefully close to 100%) opensolaris-compatible laptop is coming :D
[23:31:46] <e^ipi> FireflyST, bought a sparcbook?
[23:33:07] <FireflyST> Heh, no, a Dell, but I had it built checking each piece of hardware carefully.
[23:34:26] <Stric> Current Sil3114 support is "if you've got the raid versiom, you're screwed", right?
[23:35:46] <e^ipi> i bought a $500 acer, i only bothered to check wifi
[23:35:58] <e^ipi> it was atheros, so i bought it
[23:35:59] <coffman> Stric: i think you allways can flash the generic raid bios
[23:36:16] <Stric> coffman: it's an EPROM on this specific card.. not EEPROM/Flash..
[23:36:17] <FireflyST> Acers are decent, I just got this because I was able to do payroll deduction on it.
[23:36:21] <e^ipi> woah, what the hell?!
[23:36:23] <Stric> (according to tech specs for the chip)
[23:36:30] <e^ipi> this machine has a speaker in it somewhere
[23:36:34] <FireflyST> which means I can have a laptop while selling my iBook
[23:36:46] <Stric> e^ipi: laptops have that..
[23:36:47] <e^ipi> sounds started coming from the inside of my blade1k
[23:36:57] <Stric> and all sun workstations do too
[23:37:03] <e^ipi> yes, evidently they do
[23:37:19] <sommerfeld> e^ipi: blade1k disks have a feeper in them, and there's a speaker connected to the audio card
[23:37:49] <coffman> Stric: urgs
[23:37:57] *** tombhadAC has quit IRC
[23:38:10] <e^ipi> it was just trippy to discover that by accident when I was previously unaware of it
[23:38:10] <coffman> so what about tracing this app problem of mine? some dtrace guru around?
[23:38:23] <brendang> maybe
[23:38:33] <brendang> what's the app?
[23:38:35] <FireflyST> e^ipi: when one of the disks in my 2k went bad, it was beeping constantly.
[23:38:43] <brendang> coffman: ie, Java app? ruby? C?
[23:38:56] <coffman> brendang: psi-im, qt C,C++
[23:39:20] <elijahwright> it hangs the whole system hard, or just seems to?
[23:39:21] <brendang> coffman: are you using the system's console?
[23:39:26] <coffman> brendang: locks up the system with while doing heavy io
[23:39:37] <coffman> brendang: yes, its my laptop
[23:39:43] <brendang> coffman: locks such that the mouse can't move at all?
[23:39:58] <coffman> brendang: locks up that i cant kill X
[23:40:13] <brendang> coffman: right, but can you move the mouse pointer?
[23:40:32] <coffman> nope, it jumps some times
[23:40:41] <richlowe> that sounds like very very heavy activity, rather than a lock.
[23:41:10] <coffman> well it jumps once in 2 mins or so, i waited like 15mins
[23:41:35] <brendang> coffman: right. I'd run vmstat 1 > logfile, and read that afterwards. It'll begin the investigation (point to low swap, high dispatcher queue length, syscalls, etc)
[23:41:39] *** yarihm has joined #OpenSolaris
[23:42:26] <brendang> coffman: once we know what type the issue is, then we know what DTrace scripts to run.
[23:42:33] <coffman> if i start it in Xephyr and the select the option i have a couple of seconds where the usage starts to rise
[23:42:47] <brendang> coffman: which "usage"?
[23:42:57] <coffman> i gues of the disk
[23:43:09] <brendang> coffman: this is why we need the vmstat output :)
[23:43:17] <coffman> ill do that :P
[23:44:06] <brendang> cool. you could also spawn an "mpstat 1 > mpstat.log &" and "iostat -xnmpz 5 > iostat.log &" as they are handy starting points too.
[23:45:13] *** elijahwright has quit IRC
[23:53:01] <pbaumgartner> anybody here using zfs+samba in production?
[23:53:55] * Stric will be in a month
[23:54:01] <Stric> hope it works fine :)
[23:54:06] <pbaumgartner> done any testing yet?
[23:54:26] <Stric> not samba.. mainly plaing with the zfs part.. homedirs for ~2700 accounts.. so 2700 filesystems..
[23:54:34] <Stric> playing
[23:55:00] <pbaumgartner> acl support seems iffy right now
[23:55:11] <pbaumgartner> acl in samba that is
[23:56:15] <Stric> http://bugs.opensolaris.org/view_bug.do?bug_id=2149388 says stuff should be fixed in s10u4 which is coming pretty soon
[23:57:15] *** MattAFC has quit IRC
[23:58:45] *** freakazoid0223 has joined #opensolaris
[23:59:04] <jbk> afternoon





top