[00:02:06] *** laserdrill has joined #opensolaris
[00:03:09] <laserdrill> hi
[00:04:14] <laserdrill> I am new to Solaris and managed to download and install it without too much trouble. Now I am trying to install applications and find I am quite lost in this area
[00:05:03] <jamesd>  www.blastwave.org
[00:05:30] <laserdrill> Specifically i am trying ti onstall firefox 2, and ADempiere, an accounting package
[00:05:44] <laserdrill> thx jamesd, will look at it
[00:06:23] *** m0le has quit IRC
[00:06:37] *** Dink has quit IRC
[00:06:49] *** m0le has joined #opensolaris
[00:08:49] *** loke_ has quit IRC
[00:10:26] <zooko> Which version did you download and install?
[00:11:30] <laserdrill> solaris 10
[00:12:47] <zooko> Thanks.
[00:13:08] <wesolows> zooko: The authoritative license is the one that comes with whatever source code you get.
[00:13:14] *** obsethryl has joined #opensolaris
[00:13:17] <wesolows> zooko: Nothing else governs.
[00:13:37] <zooko> But you know what I meant, right?
[00:13:58] <zooko> Simon Phipps is paid to advise the sun execs on what to do wrt open source licensing, if I understand correctly.
[00:14:12] <wesolows> Frankly, I don't.  Because if you're asking the question I suspect you were, you're asking people to speculate and/or prognosticate.  We can't.
[00:15:31] <zooko> I was, indeed, asking people to speculate and prognosticate.
[00:15:46] <zooko> Also to let me know if there were more tidbits such as Simon Phipp's blog entry.
[00:15:47] <wesolows> He also didn't express a position in his entry (just careful noncommital guessing), and even if he had there's no guarantee that his advice to execs, if any, is or was the same.
[00:16:30] <zooko> I know.
[00:16:37] *** loke_ has joined #opensolaris
[00:16:52] *** jmcp_ has joined #opensolaris
[00:17:52] <zooko> I will disappear soon due to battery depletion and family responsibilities.
[00:18:10] <wesolows> Battery depletion makes fools of us all.
[00:18:21] <wesolows> With hope you'll be able to return later.
[00:19:00] *** zooko has quit IRC
[00:23:40] *** triplah_ has quit IRC
[00:24:57] *** Mazon is now known as mazon
[00:58:57] *** echelog has joined #opensolaris
[00:58:59] <wesolows> You really believe that?
[00:59:12] <aruiz> e^ipi, have you contributed code to opensolaris already?
[01:00:04] <PerterB> ok, on paper it doesn't have
[01:01:22] <e^ipi> aruiz: i'm in the middle of freeing libc_i18n.a
[01:01:22] <wesolows> PerterB: If you're in Iran or China, your government will actively oppose the United States.  If you live anywhere else, your government is effectively owned by ours.
[01:01:23] <wesolows> PerterB: You don't even get to choose your Congressmen...
[01:01:23] <PerterB> oh, we have a "special relationship", which as far as I can tell means "bend over and take it"
[01:01:23] <wesolows> e^ipi: put that wad back and I'll pour an Old Potrero for you next time I see you.
[01:01:38] <aruiz> e^ipi, why wouldn't you wirte gplv3 code?
[01:01:43] <wesolows> PerterB: Yes, that's correct.  In other words, you have the same kind of relationship with the US government that we do.
[01:01:50] * jmcp_ growls on storage-discuss, again
[01:01:54] <PerterB> however, that kind of authority can't be written into a license agreement
[01:02:32] <e^ipi> aruiz: because the license is arrogant and poisonous. I won't write any code that forces other people to change the license of theirs to use it.
[01:02:46] <aruiz> e^ipi, have you singed the contribution agreement?
[01:02:48] <wesolows> PerterB: But it already has been.  International copyright treaties effectively guarantee it, and while this isn't their intent, the US "content" cartels' insistence pretty well guarantees that the GPL's terms are enforceable in your country as written in the US.
[01:03:09] <wesolows> the SCA is much more vile than the GPL.
[01:03:34] <e^ipi> aruiz: yeah, i had to to integrate some bug fixes
[01:04:04] <e^ipi> and I still have hope Sun's execs don't do something stupid like GPL'ing solaris
[01:04:37] <moazamraja> the santa clara campus? SCA?
[01:04:46] <moazamraja> and here I thought it was a nice place...
[01:04:51] <coffman> im more afraid of dual-licenz...
[01:05:04] <wesolows> e^ipi: Our execs do something stupid?  Thankfully that never happens.  No reason to worry.
[01:05:13] <e^ipi> heh
[01:05:14] *** Mazon is now known as mazon
[01:05:37] <wesolows> moazamraja: Actually SCA is lovely; for something like 150 years it's been a mental institution, and the campus helps keep them calm.
[01:05:53] <PerterB> well, that's a cheery thought.... my initial comment though was meant to say that if you want an authority to define what is and isn't derivative then it should at least be _nominally_ international and independant rather than the US Congress
[01:06:05] *** karrotx_ has quit IRC
[01:06:39] <moazamraja> wesolows: and MPK = Sun Quentin? (looks like it from afar)
[01:06:45] <wesolows> PerterB: I wouldn't bother.  An even nominally independent body writing such a definition would never get it adopted, because the US, on behalf of our corporate overlords, would just force something else on you instead.
[01:07:02] <aruiz> e^ipi, do you know what does that means?
[01:07:03] <aruiz> :)
[01:07:29] <wesolows> moazamraja: It's kind of bleak, for sure.  But that's pretty much the story of the peninsula.
[01:07:45] <wesolows> moazamraja: The buildings themselves are functional, not too bad.  Just boring.
[01:07:47] <moazamraja> mpk kicks butt tho, once you step inside the campus itself
[01:07:48] <e^ipi> yeah, it means my contributions dry up & I start using FreeBSD if sun decides to infect OpenSolaris
[01:08:09] <moazamraja> my old office had a great view of the center court area *sigh*
[01:08:12] <moazamraja> ah, the days.
[01:08:15] <wesolows> What's really unfortunate is the location - nowhere near transit, on fill, surrounded by a giant parking lot and a dike to hold back the bay.
[01:08:34] <wesolows> I figure that campus is a death trap in a large quake.
[01:08:41] <moazamraja> and not much good food nearby
[01:08:46] <PerterB> wesolows: fair point, but if I simply give up and say that it's all fucked anyway so why should I bother, then surely I'm just becoming part of the problem
[01:09:12] <wesolows> PerterB: No, when you say that you take the first step toward becoming a revolutionary - i.e., part of the solution.
[01:09:12] <moazamraja> actually, i had heard MPK is a pretty safe place to be in an earthquake, cause the buildings are built uber-strong with earthquakes in mind
[01:09:32] <PerterB> you miplset "dangerous, wanted terrorist"
[01:09:43] <wesolows> moazamraja: For the sake of the several thousand people working there, I sure hope so.  Because the ground certainly isn't going to hold up for shit.
[01:10:08] <moazamraja> <Tricky> at least I won't be getting it from a man
[01:10:10] <moazamraja> <Tricky> unlike u
[01:10:10] <moazamraja> <Tricky> ur ass (literally) got infected with anal chlamidiya
[01:10:21] <e^ipi> o_O
[01:10:22] <wesolows> PerterB: The sooner we acknowledge that we aren't stakeholders in the global economy and political universe any more, the sooner we can replace all this.
[01:10:26] <moazamraja> irc has gone bonkers
[01:10:35] <e^ipi> moazamraja: what's that from?
[01:10:36] <moazamraja> wtf, where do these ppl come up with this shit
[01:10:47] <moazamraja> another channel i was on, a TECHNICAL channel no less
[01:10:54] <PerterB> moazamraja: irc has been bonkers since forever ;)
[01:10:57] <moazamraja> 2 guys arguing about nvidia/ati
[01:11:08] <e^ipi> oooo....kiedokie
[01:11:09] <zooko> e^ipi: so apparently it is in order to avoid alienating folks like you that Sun is doing what they are doing re: the Simon Phipps web log earlier.
[01:11:33] <e^ipi> dual-licensing is even worse
[01:11:51] <e^ipi> because that's just a mess that /will/ fork, sooner rather than later
[01:12:30] <wesolows> zooko: BTW, Simon's post is factually incorrect in one regard: "While Sun holds those aggregated rights on trust, it would be a mistake to assume Sun can just change the license without any form of community discussion" - this is simply not (legally) true.
[01:12:53] <zooko> I understood him to mean "Can" not in the legal sense but in the "Can get away with it" sense...
[01:13:08] <wesolows> We can hope for dialogue, we can demand the right to be heard, but at the end of the day the SCA lets Sun do whatever it wants.
[01:13:22] <PerterB> wesolows: I've been aware of that for some years now.....
[01:13:27] <wesolows> Yes, that must be what he meant to say.
[01:13:37] <zooko> Yes, there's this other thing that constrains Sun -- the aforementioned global economy and political whatchajigger.
[01:15:05] *** henrik_ has joined #opensolaris
[01:15:14] <wesolows> Actually the only thing constraining Sun's choice of licensing regimes is the same one that constrained the choice in the first place - a fear of bad PR.
[01:16:43] <moazamraja> does sun have a PR dept?
[01:16:46] <wesolows> And, of course, any individual desire to do the right thing, or a feeling of responsibility to the shareholders (if not the contributors).
[01:16:48] <moazamraja> or a mktg dept?
[01:17:06] <wesolows> moazamraja: Sure, where do you think all those costs come from?
[01:17:19] *** loke__ has joined #opensolaris
[01:17:31] <moazamraja> i have a response to that.
[01:17:32] <moazamraja> but.
[01:17:42] <moazamraja> in polite company and all, I'll keep quiet.
[01:17:46] <henrik_> evening
[01:17:51] <wesolows> polite? IRC?
[01:18:30] <moazamraja> yes...don't you see 'henrik' saying 'evening' ? Ever so polite...
[01:18:51] <wesolows> true, true
[01:19:07] <wesolows> it's enough to make me feel like a boor
[01:19:10] <wesolows> good evening, henrik_
[01:19:35] *** hohum has joined #OpenSolaris
[01:19:41] *** karrotx has joined #opensolaris
[01:21:07] *** aruiz has quit IRC
[01:23:29] *** loke_ has quit IRC
[01:23:38] *** danv12 has joined #opensolaris
[01:24:43] *** loke_ has joined #opensolaris
[01:25:46] *** loke__ has quit IRC
[01:26:16] *** xinkeT has joined #opensolaris
[01:28:46] *** obsethryl has quit IRC
[01:32:25] *** soulie has joined #opensolaris
[01:32:40] <soulie> hello :)?
[01:32:47] <Trident> Hi.
[01:32:56] <soulie> :P may i ask some questions
[01:33:14] <Tempt> Might as well.
[01:33:17] <Tempt> Morning all.
[01:33:48] <soulie> do they still ship those starter kits?
[01:34:09] <soulie> filled in the stuff couple of weeks ago
[01:35:23] <Tempt> Might take a little while
[01:35:31] <Tempt> Probably easier to download it if you're in a hurry
[01:35:32] <soulie> thaught i might as well give it a try or get whined by brother for downloading 3+gb again :)
[01:35:43] <soulie> me never in a hurry , me lazy :P
[01:37:17] <soulie> then my next question will be, is there a code forum
[01:37:34] <Tempt> Plenty.
[01:37:43] <Tempt> Mailinglists, and of course, this channel.
[01:38:43] <soulie> I was trying out some c/c++ regular expressions but everything about #include <regex.h> fails horrible on my opnesolaris
[01:40:39] <soulie> and the regex.h it shows doesnt rly look like its finished at all ;o
[01:40:40] *** NikolaVeber has quit IRC
[01:49:00] *** jpdrawneek has quit IRC
[01:52:33] <laserdrill> Following the instructions on blastwave I downloaded te pgp key and rn the command gpg --import pgp.key and bash says: command not found. Please advise how I might correct this. Thanks much.
[01:52:52] <jmcp_> laserdrill: try /opt/csw/bin/gpg --import pgp.key
[01:56:29] <e^ipi> soulie, it's in the pile of crap i'm replacing anyways
[01:59:09] <laserdrill> Thanks, will try that.
[01:59:52] *** monkey_Hou_ has joined #opensolaris
[02:02:02] <laserdrill> jmcp, that does not work unfortunately it says /opt/csw/bin/gpg: No such file or directory
[02:02:25] <soulie> :)
[02:02:26] <jmcp_> did you install it?
[02:02:38] <laserdrill> I tried to pkg-get it but pkg-get did not find it
[02:03:10] <laserdrill> is there more to type than pkg-get install gpg ?
[02:03:49] <jmcp_> /opt/csw/bin/pkg-get -i gnupg
[02:03:56] <jmcp_> if you haven't installed it already, then you have to install it
[02:04:10] <jmcp_> fortunately you can bootstrap /opt/csw without gpg
[02:04:19] <laserdrill> thanks. trying that now
[02:05:20] *** Pietro_S has quit IRC
[02:07:14] <laserdrill> jmcp: that worked thanks again
[02:07:39] *** monkey_Ho1 has joined #opensolaris
[02:08:47] <laserdrill> In fact its still downloading and installing stuff
[02:09:30] *** tsoome has quit IRC
[02:13:43] *** aruiz has joined #opensolaris
[02:14:08] *** danny_j has quit IRC
[02:14:57] <soulie> may grub maze finally works ("_)
[02:15:23] <soulie> I must backupt his cause , never doing it again xD
[02:18:39] <laserdrill> anyone with a working installation of ADempiere, the opensource accounting package, on solaris 10
[02:18:49] <laserdrill> ?
[02:19:12] *** monkey_Ho1 has left #opensolaris
[02:19:24] <laserdrill> I am trying to get it installed on my x86 system but the instructions i found were not working
[02:19:31] <soulie> not me
[02:19:36] *** monkey_Hou_ has quit IRC
[02:21:29] *** Fish has quit IRC
[02:25:25] <laserdrill> ok
[02:26:44] <laserdrill> in the instructions was running a command called dmake to compile gnu make - which solaris said was not found.
[02:27:23] <laserdrill> So I could not compile gnu make and could make no further progress in the exercise
[02:27:28] <jmcp_> laserdrill: you need to install Sun Studio for dmake, and then it'll be in /opt/SUNWspro/bin
[02:31:39] <laserdrill> thanks. I started the download of sun studio 12 this morning with the same idea, but it stalled at 21%
[02:32:57] <jmcp_> then you might just have to wait
[02:33:39] <laserdrill>  doing that. I am using the sun download manager so it should survive interruption
[02:33:45] *** Tpenta has joined #opensolaris
[02:34:05] <jmcp_> hi Tpenta
[02:34:20] <Tpenta> hey james
[02:38:41] <e^ipi> hey Tpenta, long time no see
[02:45:10] <Tpenta> hey there
[02:46:56] *** simford has joined #opensolaris
[02:47:17] <e^ipi> how goes?
[02:47:58] <Doc> morning alan. written your talk for thursday yet?
[02:48:55] *** gaz_ has quit IRC
[02:49:42] *** soulie has left #opensolaris
[02:51:02] <SYS64738> is it sane to use sun cluster on solaris express ?
[02:51:22] <SYS64738> or I should use on solaris 10 ?
[02:51:57] <boyd> If your balance of availability and stability and features and supportability matches.....
[02:52:06] * boyd isn't sure if it even works
[02:52:26] *** hile_ has joined #opensolaris
[02:52:37] <boyd> Hehe... Ruby bindings for the Google API -> "
[02:52:39] <jmcp_> SYS64738: SunCluster isn't supported on solaris express. it might well run, but then again, it might now
[02:52:40] <jmcp_> not
[02:52:43] <boyd> Hehe... Ruby bindings for the Google API -> "Gooby"
[02:52:46] <jmcp_> SYS64738: if it does, you're on your own
[02:53:15] <boyd> Yes, That's part of what I was getting at
[02:54:07] *** delewis has quit IRC
[02:54:50] <hile_> SYS64738,  "it's supported" and "it works" are two majorly different questions :)
[02:55:18] <boyd> So is "it makes sense for the mission-criticalness of my app"
[02:55:39] <SYS64738> I would like to integrated many services (mx, smtp, dns, proxy, www) in one single server
[02:55:50] <SYS64738> at the moment I have two identical server
[02:56:12] <hile_> if this a production environment, why the hell are you running SX?
[02:56:38] <SYS64738> at the moment I am playing with zone on a: megatron # uname -a
[02:56:39] <SYS64738> SunOS megatron 5.11 snv_66 i86pc i386 i86pc
[02:57:12] <SYS64738> SX is opensolaris ?
[02:57:24] <SYS64738> or SX is solaris 11 ?
[02:57:39] <SYS64738> or opensolaris is soraris 11 ? :)
[03:00:20] <SYS64738> why don't use sx in production ? I have web proxy servers that go like a cannonball
[03:01:01] <hile_> because it's not supported
[03:01:59] <SYS64738> I think that neither windows is supported
[03:02:10] <SYS64738> like they say
[03:02:14] <boyd> Lol
[03:02:49] <SYS64738> I like zfs and zones, than should I use SX
[03:03:00] <boyd> Both are in solaris 10
[03:03:07] <SYS64738> is it free ?
[03:03:10] <boyd> Yes
[03:03:13] <Trident> SYS: yup.
[03:03:20] <boyd> (as in beer)
[03:03:25] <SYS64738> dio lupo
[03:03:37] <boyd> "Day of the wolf?"
[03:03:50] <SYS64738> dio is god
[03:04:01] <SYS64738> shit
[03:04:07] <boyd> God of the wolf?
[03:04:08] <SYS64738> I didn't knew
[03:04:13] <SYS64738> no
[03:04:18] <SYS64738> wolf god
[03:04:54] <SYS64738> than I think that I must delete something
[03:04:55] * boyd was trying to reconstruct from his limited Latin.
[03:05:08] <Trident> SYS: For production stuff that is critical you whould definently go for Solaris 10.
[03:05:13] <SYS64738> rosa, rosae
[03:05:39] <SYS64738> damn
[03:05:50] <SYS64738> I didn't knew that was free
[03:06:14] <SYS64738> opensolaris is s10 and nevada is pre-s11 ?
[03:06:20] <boyd> The OS is free. Commercial support is, well, commercial
[03:06:32] <coffman> SYS64738: i would wait a month - new solaris 10 release is coming
[03:07:00] <boyd> !tell SYS about "confused? so was i"
[03:07:28] <SYS64738> where can I look for s10 support ?
[03:07:35] <jmcp_> sun.com/software/solaris
[03:08:27] <jmcp_> http://www.sun.com/software/solaris/support.jsp, even
[03:08:32] <Trident> SYS: S10 is S10, a supported release of Sun Solaris. Nevada is a snapshot of Opensolaris with some propriatary stuff added. Opensolaris is the opensourced parts of Solaris. If I have understood it correctly?
[03:09:10] <Trident> And Nevada probably will be Solaris 11 when released in a supported version.
[03:10:20] <SYS64738> damn I must erase some things in my brain and restart it
[03:10:31] <boyd> http://whacked.net/2005/06/21/confused-so-was-i/
[03:10:54] * boyd pokes stevel about updating that to SXDE & SXCE
[03:10:59] <Trident> I say probably as we have no idea what Sun will decide to call it. But Solaris 11 is the probable choice. On the other hand Solaris 2.7 would have been a probable choice too.
[03:11:19] <boyd> Nobody is convinced the marketing people can count higher than 10
[03:12:39] <Trident> :)
[03:13:02] <Trident> They might wrap around then, Solaris 1 all over again... :)
[03:13:07] <Trident> Or Solaris 0
[03:13:09] <SYS64738> tomorrow I'll buy a book
[03:13:25] <Trident> Vut I don't think marketing starts with 0.
[03:13:32] <Tempt> Solaris Vista
[03:13:35] <Trident> s/Vut/But/
[03:13:56] <boyd> sometimes I don't think they get past 0
[03:14:08] * boyd goes back to work
[03:18:00] <CSFrost> Solaris Doors? it comes preshut? :-P
[03:18:31] *** hohum has quit IRC
[03:21:32] <SYS64738> can I use the solaris express zones in solaris 10 ?
[03:21:37] <Trident> CSFrost: Ofcourse. Until you do a door_create. :)
[03:22:22] <coffman> SYS64738: solaris 10 comes with zones
[03:22:54] <SYS64738> I mean, can I migrate the zones that now I have in sx in solaris 10 ?
[03:23:05] *** cypromis_ has quit IRC
[03:25:59] <SYS64738> damn they also  send the dvd at home
[03:26:14] <SYS64738> why I didn't see this first
[03:26:37] <coffman> SYS64738: uhm, not realy
[03:27:22] <laserdrill> to download openoffice2 from blsatwave what would be coorect pkg-get -i ooocore or pkg-get -i ooocommon ? There are a lot of ooo packages. Thanks
[03:29:52] <SYS64738> I go to bed
[03:29:57] <SYS64738> thank you all and good night
[03:30:16] <coffman> laserdrill: why do that? just pick it from opensolaris.org
[03:30:20] <coffman> ah
[03:30:28] <coffman> openoffice.org :P
[03:35:11] <laserdrill> Coffman: The installation process gets to me
[03:36:35] <laserdrill> I am in the process of installing a bunch of apps and stuff to get even one compilation to work. It appears you cant compile anything on Solaris 10 out of the box.
[03:37:16] <Doc> yah.. solaris sucks
[03:39:06] <moazamraja> laserdrill: u can, you need to install the compilers (same as linux/ubuntu)
[03:39:23] <moazamraja> Solaris Express Developer Edition comes with all the compilers pre-installed
[03:40:01] <coffman> laserdrill: weell, if you go by blastwave...
[03:40:24] <coffman> ffs, sxce aalso comes with a compiler
[03:40:28] <coffman> gcc
[03:40:48] <coffman> and i dont need stupid sxde to intstall studio
[03:40:51] <laserdrill> I am trying to install the Adempiere accounting software on Solaris 10 and the instructions I found require that I compile everything. So I started and downloaded GNU make.
[03:41:26] <laserdrill> The configure failed saying something about the compiler not being able to generate executables
[03:41:33] <moazamraja> why is sxde 'stupid
[03:41:35] <moazamraja> ?
[03:41:40] <moazamraja> cuz it comes with optimized compilers?
[03:41:51] <laserdrill> Ive been researching and asking for help and tryinf stuff since.
[03:42:09] <coffman> moazamraja: no, cause the i can install a compiler by myself
[03:42:39] <coffman> moazamraja: and i dont need a extra "tested" release for that i dont get updates
[03:42:48] <coffman> when i need such i use solaris 10
[03:43:01] <laserdrill> im downloading sun studio 12 now, hope it install s ok when I finally get it down
[03:43:03] <coffman> sxde binds the work of people at sun
[03:43:07] <coffman> for nothing
[03:43:14] <laserdrill> I expect to be able to compile stuff from there
[03:43:32] <coffman> laserdrill: you might be better of with using gcc on linux style software
[03:44:03] <laserdrill> I've been thinking about that
[03:44:06] <coffman> laserdrill: you should take a look in /usr/sfw
[03:44:15] <coffman> laserdrill: there is gmake and gcc etc
[03:44:25] <coffman> its not in the default path
[03:44:32] <laserdrill> ok, will do
[03:44:39] <CSFrost> moazamraja, sxde is stupid because it slows down community editions progress, and it redirects engineers time from doing what they should be doing.
[03:45:08] <hile_> coffman, arguably, he should fix the software so that it builds with a standards-compliant compiler
[03:45:10] <CSFrost> and, sxce also comes with optimized compilers.
[03:46:23] <laserdrill> coffman: I wonder if it is necessary to compile everything for the application to work. for example, it uses Postgresql as the backend and i though once the correct version is installed the application should work.
[03:46:23] <coffman> hile_: yes sure, but i think its easier/fast for him to use gcc
[03:46:26] *** deedaw has quit IRC
[03:47:00] <moazamraja> CSFrost: how? by spending QA/QE efforts on SXDE?
[03:47:26] <coffman> and as long as studio is not os software, i would not get to dependent on it
[03:47:28] <moazamraja> CSFrost: SXCE comes with Studio C/C++ compilers?
[03:47:37] <moazamraja> coffman: it's what the OS is built with.
[03:47:44] <moazamraja> so no matter what, you are dependent on it
[03:47:55] <hile_> why not?
[03:48:16] <coffman> well, hope some day you can build it without it
[03:48:27] <coffman> seflhosted distro would be nice
[03:48:32] <coffman> not that i dont like studio
[03:48:45] <moazamraja> then I don't understand the anti-studio sentiment
[03:49:00] <laserdrill> Coffman: as a bit of background, the instructions start by downloading GNU make and uss dmake to "make" it. dmake was not installed on Solaris so thatfiled, as did the configure. Are you saying there is another way to compile the stuff without compiling make and GCC and stuff?
[03:49:08] <moazamraja> even if the distro is self-hosted, it might still be built on studio compilers
[03:49:35] <CSFrost> moazamraja, I see no reason in doing the same work 3+ times
[03:50:01] <coffman> laserdrill: solaris 10/sx ships with gcc and gnu make
[03:50:11] <CSFrost> and the studio compilers are on the ce dvd as well.
[03:50:12] <coffman> laserdrill: its in /usr/sfw/bin
[03:50:38] <laserdrill> hmmmm, now I had no idea!
[03:50:41] <moazamraja> CSFrost: the reason would be so that customers get a full application development environment bundled with the OS
[03:51:07] <laserdrill> mind you, i'm new to solaris, so please bear with me
[03:51:14] <moazamraja> I've downloaded and installed the CE ISO before (dvd), but I don't recall Studio compilers on it, maybe I had an old version
[03:51:59] <coffman> the thing is to not depend on sun, more poolitical then technical...
[03:52:12] <CSFrost> moazamraja, then why not do it with community edition? why split teams up to work on the same projects for 3 different editions at different times..
[03:52:20] <moazamraja> CSFrost: if the amount of work is the problem, than cancel SXCE and make SXDE default
[03:52:27] <moazamraja> sure, I agree
[03:52:32] *** yongsun has joined #opensolaris
[03:52:40] <CSFrost> if CE is ahead of DE, why would you cancel the one that is ahead? :)
[03:52:44] <moazamraja> made SXCE bundle the tools by default also, then no need for SXDE
[03:52:49] <moazamraja> yes yes, you're right :)
[03:52:54] <coffman> would be nice to be able to build the distro on other systems like bsd,linux
[03:53:07] <wesolows> or just, you know, make cc and lint part of the WOS again
[03:53:13] <moazamraja> as long as it gets Studio tools bundled by default, they can call it SXCSFrost for all I care ;)
[03:53:30] <wesolows> we've been having this fight with them for literally 15 years
[03:53:39] <moazamraja> wesolows: and dmake? and C++? Fortran? dbx?
[03:53:46] <moazamraja> 'them'....
[03:53:47] <CSFrost> yea, I know it's senseless, but I just hate the idea behind DE.
[03:53:50] *** djgregor has quit IRC
[03:53:59] <wesolows> moazamraja: dmake, certainly - it's just a make compiled with different #defines
[03:54:04] <wesolows> moazamraja: CC sure
[03:54:16] <wesolows> moazamraja: dbx, I dunno - we try to avoid duplication, and mdb got there first ;-)
[03:54:24] <moazamraja> but, you do know that Studio is a seperate 'product', right?
[03:54:32] <moazamraja> as in, sold to customers, supported by sun, etc.
[03:54:37] <wesolows> I know all about this problem.
[03:54:40] <moazamraja> of course :)
[03:54:48] <wesolows> The point is that it doesn't *need* to be.
[03:54:53] <wesolows> And, ages ago, it wasn't.
[03:54:55] <moazamraja> gotta make money somehow
[03:54:57] <moazamraja> I know
[03:55:10] <wesolows> You can't make money selling a product for $0.  You have to make it some other way.
[03:55:16] <moazamraja> let me ask you then
[03:55:27] <wesolows> If you're going to sell two products for $0, it costs you nothing to make them a single $0 product.
[03:56:14] <moazamraja> if Studio was to be absorbed into Solaris, would that org still concentrate on making Studio tools that are focused towards *application* development?
[03:56:18] <wesolows> And actually...since simply having part numbers has a cost, it would probably yield some savings.  Especially once the Studio marketing people get the axe.
[03:56:20] <galt> wesolows, tell that to the SXDE/sparc team ;P
[03:56:42] <wesolows> moazamraja: I don't see why anything about the programs themselves would need to change...
[03:56:53] <moazamraja> but the focus on development going forward might change
[03:57:07] <wesolows> It could anyway.  I don't see the point.
[03:57:35] <wesolows> Why would something be less "application focused" (whatever that means) solely because it's part of the WOS?
[03:58:06] <moazamraja> .....because org owners (vps) have their own interests
[03:58:07] <galt> moazamraja, I've alwasy hated the neologism "development going forward"  is there a corresponding "development going rearward"?
[03:58:30] <moazamraja> galt: no, the correspondent is development going NOWHERE, i.e., stalled
[03:58:40] <moazamraja> the same toolset 10 years later
[03:58:47] <moazamraja> while the competitors pass us by
[03:58:50] <wesolows> moazamraja: You do know that the same VP that's responsible for this "separate product" is also responsible for delivering, say, make to the WOS, right?
[03:58:51] <moazamraja> thats the correspondent
[03:59:09] <laserdrill> coffman: i looked in /usr/sfw and sure enough GCC and GMAKE are there. I don't understand why the ./configure of make failed although now it seems I don't really need it.
[03:59:11] <wesolows> moazamraja: As much as we try to expose our organisational structure via our product architecture, sometimes we fail. :-)
[03:59:29] <moazamraja> *sigh*
[03:59:47] <moazamraja> I'm still a fan of SXDE, or at least a packaged OS that comes with Sun tools
[03:59:55] <moazamraja> tools that are used to build the OS
[03:59:58] <wesolows> I don't really see anyone else out there with tools anywhere near as good as the stuff that's in the WOS...though I do see stuff that's a whole lot more *popular* (see "Eclipse") than the stuff that's *not*.
[04:00:18] <laserdrill> So how would i proceed, forget about compiling  make, the build utilities and GCC and compile the other stuff using gmake and GCC that are in /usr/sfw ?
[04:00:25] <moazamraja> wesolows: from a kernel/driver dev viewpoint, or a application development viewpoint?
[04:00:48] <moazamraja> wesolows: we faltered on app development for a long time
[04:00:50] <moazamraja> we = sun
[04:01:07] <wesolows> moazamraja: Since I spend over 80% of my time in userland these days, I guess I'm talking about both.
[04:01:30] <wesolows> moazamraja: Find something, anything, a competitor offers that can compete with USDT.
[04:01:31] <moazamraja> I will give you this, our tools are overly complicated in naming, structure, etc.
[04:01:32] <coffman> laserdrill: you should replcace make with gmake and link cc and CC to gcc
[04:01:45] <wesolows> moazamraja: The Studio tools are, certainly.
[04:01:46] <moazamraja> but the same can be said for Sun opensolaris/solaris express
[04:01:47] *** aruiz has quit IRC
[04:01:54] <wesolows> Yep
[04:02:12] <wesolows> But blame marketing for that.
[04:03:23] <laserdrill> coffman: Great. How do I do that, I am not much into compiling apps as you might hve laready figure. Is there somewhere I could read ip the instrcutions or could you give me the core instuctions to do iit?
[04:04:04] <wesolows> coffman: eh?  do CC=gcc MAKE=gmake not work?
[04:04:07] <laserdrill> bI am refering to linking CC and cc to GCC and such
[04:04:31] <wesolows> CC=gcc MAKE=gmake ./configure [args]; gmake; gmake install
[04:04:33] <laserdrill> ok, will try that then
[04:04:38] <wesolows> nearly always works
[04:05:35] * galt throws dogfood at wesolows
[04:05:46] <laserdrill> thanks, I am learning a lot herebut I do like this stuff. Hope I can help in some areas too or help someone in my predicament later on
[04:05:49] <moazamraja> wesolows: at the end of the day, the 2 main points need to be that 1.) customers get an OS which they can develop on 'out of the box', 2.) Sun does not stagnate on application development tools
[04:05:53] <wesolows> galt: /me eats it, happily
[04:06:19] <wesolows> moazamraja: Sure, I'll agree with both points.
[04:06:31] <moazamraja> however that is accomplished, so be it
[04:07:11] <wesolows> And both are served by making the *successful* and *useful* tools currently languishing as a "separate product" poorly integrated with the fantastic developer tools offered by Solaris itself a part of the OS proper.
[04:07:16] <wesolows> And discarding the rest.
[04:08:35] <moazamraja> the rest = GUI tools, i presume?
[04:08:50] <boyd> With the exception of discarding the rest, that seems to be what the SXDE is trying to achieve
[04:09:45] *** Teknomancer has joined #opensolaris
[04:10:43] *** FFForever has joined #opensolaris
[04:10:46] <FFForever> Hello
[04:11:02] <FFForever> is open solaris better then linux?
[04:11:03] *** yongsun has quit IRC
[04:11:47] <CSFrost> no, haha never
[04:11:54] <CSFrost> there, you got your answer, now shoo
[04:11:57] <laserdrill> FFForever: that's a trick question, right?
[04:12:12] *** xuewei has joined #opensolaris
[04:12:41] <FFForever> no
[04:12:52] <FFForever> i have been looking at open solaris for some time now....
[04:12:58] <wesolows> FFForever: Welcome to the Coke Drinker's Club of Greater Atlanta, GA.  Your question was "Is Coke better than Pepsi?"  A: Well, we think so.  That's why we're here, ya dig?  Thanks.
[04:13:17] <moazamraja> food time, bbl
[04:13:33] <FFForever> yeah i get it...., but i am just wondering from a user/security stand point is it more secure....
[04:14:03] <wesolows> The difference in security between one shop's policies and procedures and another's far, far outweighs the differences in the software itself.
[04:14:08] *** yongsun has joined #opensolaris
[04:14:34] <FFForever> well i take security as my number one goal
[04:14:37] <wesolows> We have more features that good administrators can use to improve security, certainly.  And I feel they're better implemented, documented, and integrated.
[04:15:05] <wesolows> But you don't have to use them, so you can put some box on the Internet with an empty root password and you'll be cracked in days...if not sooner.
[04:15:16] <FFForever> why do i have 2 register with sun and give them ALL of my info to get open solaris?, am i going 2 get junk mail and that is how they make it free?
[04:15:21] <wesolows> So I assume you're chatting with me now over IPsec?
[04:15:29] <wesolows> Oh, that.  You don't.
[04:15:36] <wesolows> You only have to register to get Solaris Express.
[04:15:39] <FFForever> IPsec?
[04:15:50] <FFForever> what is the difference?
[04:16:01] <FFForever> and witch one should i get?
[04:16:02] <wesolows> Solaris Express is a Sun product based on OpenSolaris.
[04:16:09] <wesolows> OpenSolaris is a set of building blocks.
[04:16:27] <boyd> I set up an IRC server that only supported SSL over IPSec from previously trusted IPs once. Nobody ever came to talk to me :(
[04:16:28] <coffman> wesolows: ehm, sorry meant that...
[04:16:43] <FFForever> boyd i wonder why.....
[04:16:59] <wesolows> coffman: oh, ok.  I was worried that some poor soul was going to end up with a very confused environment.
[04:17:18] <boyd> Then I realised that security shouldn't be my number 1 goal
[04:17:38] <wesolows> FFForever: That was my point - if security were really your number one priority, you would never use unencrypted pipes, nor use an IRC client.  Because these things all increase your risk.
[04:18:12] <FFForever> ok..., security the best i can get while not giving things up :)
[04:18:15] <wesolows> And then you'd realise that you actually mean that your number 1 priority is to get the things done that you care about, and doing so securely is a constraint.
[04:18:26] 
[04:18:27] *** Teknomancer has quit IRC
[04:18:46] <coffman> and my terminal is still bork
[04:18:47] <coffman> gar
[04:18:59] <coffman> shity utf8, shity gnome terminal
[04:19:06] <FFForever> what should i get to get me started with open solaris?
[04:19:11] <wesolows> Ahh, Justice Bork.  What could have been.
[04:19:54] <wesolows> FFForever: Well, I recommend Solaris Express (I work for Sun), but if you're not comfortable giving Sun your personal information, consider one of the other distributions like Belenix, MarTux, or Nexenta.
[04:20:29] <FFForever> i was reading the side panels that says they give out my info?, does son profit from my info?
[04:20:36] <wesolows> If you're just looking to get started, you'll need one of them, because OpenSolaris by itself isn't usable as an operating environment.  Of course, if you want to use it as a base to roll your own distribution, you're welcome to do so.
[04:21:17] <wesolows> How should I know what they do with it?
[04:21:31] <wesolows> I'm sure they only do what it says they can do.
[04:21:38] <wesolows> Beyond that, I honestly have no idea.
[04:22:05] <FFForever> so for a basic user what one would u recommend for me?, i just use my computer for word processing and browsing the internet with firefox XD
[04:22:08] <wesolows> Sell it to spammers?  I doubt that very much.  Give it to their marketing people?  Maybe.  Keep it in a database somewhere?  Very likely.
[04:22:11] <FFForever> ohhh and chat on ims
[04:22:32] <wesolows> I do all those things as well, and I'm using Solaris Express.
[04:22:33] <CSFrost> dude, just register as bill gates or mickey mouse then
[04:22:34] <CSFrost> jeez
[04:22:59] <FFForever> i don't like lieing on online apps that ask for meh info, isn't the fraud?
[04:23:10] <FFForever> by lieing on them
[04:23:35] <wesolows> It depends on local law, I suppose.  Since I'm not a lawyer, I can't give you legal advice.
[04:23:48] <e^ipi> not legally anyways
[04:24:00] <wesolows> e^ipi: Well, that would be a legal opinion. ;-)
[04:24:13] <e^ipi> drats! they caught us in a loophole
[04:24:17] <e^ipi> shifty lawyers
[04:24:18] <boyd> Well, it's not an *illegal* opinion.
[04:24:28] <boyd> (yet)
[04:24:46] <wesolows> FFForever: Look, if this bothers you that much, don't use it.  And be sure to send Sun your feedback telling them that you're not using it and that's why.
[04:24:54] <CSFrost> I play mickey mouse on plays, I was just on stage when I registered.
[04:25:06] <CSFrost> at the time of registration I was legally within my entity.
[04:25:15] <wesolows> FFForever: Otherwise, go download Solaris Express Community Edition, install it, and start using Firefox, StarOffice 8, and gaim ir whatever you want to do the things you described.
[04:25:16] <boyd> I reckon that Mr Mouse has been a heavy downloader
[04:25:17] <FFForever> what boot loader does os use?, and i just finished registration :D
[04:25:27] *** karrotx has quit IRC
[04:25:30] <wesolows> On x86, grub.  On SPARC, the firmware knows how to boot.
[04:25:31] <FFForever> wow staroffice?, i thought u had 2 pay for it?
[04:25:35] <CSFrost> boyd, him and bill gates were top downloaders
[04:25:42] <CSFrost> you thought a bit wrong
[04:25:53] <FFForever> i thought you could open get openoffice free?
[04:26:04] <jmcp_> you can
[04:26:05] <FFForever> you could get*
[04:26:12] <boyd> Unless you count running Solaris as paying.....
[04:26:15] <wesolows> You can get OOO source, not StarOffice source.  But the binaries are included in SX at no charge.
[04:26:28] <FFForever> so kool :)
[04:26:35] <boyd> s/Solaris/SXCE/
[04:26:49] <FFForever> one last idiotic question :), during the installer can i tell it not to install a boot manager?
[04:26:50] <wesolows> boyd: I count *not* running Solaris as paying.  Does that count?
[04:27:09] <wesolows> FFForever: Ah...no.  It would, well, not boot then, right?
[04:27:14] <boyd> I can't count all the counts. (good thing you're a Mayor, not a Count)
[04:27:15] <FFForever> i have grub :)
[04:27:29] <wesolows> boyd: Indeed so.
[04:27:41] <FFForever> (from ubuntu and kinda just wanted to install this as a site partition), do i have 2 wipe my whole drive to use this?
[04:27:50] <wesolows> boyd: Though I recently got promoted - I'm now the Viceroy of Crazytown
[04:28:10] <FFForever> on a side*
[04:28:11] <boyd> FFForever: The Solaris GRUB understands the Solaris UFS filesystem. Others don't, generally. It only puts it in the Solaris FDISK partition IIRC
[04:28:13] <wesolows> FFForever: Only if your entire disk is consumed by a single BIOS/fdisk partition.
[04:28:28] <wesolows> You need one BIOS/fdisk partition for Solaris.
[04:28:51] <wesolows> If there's free space and a free primary partition in which to create that, you're fine.  Otherwise, you'll need to repartition.
[04:28:54] <FFForever> kk, how big of a space to i need 2 take out from linux for solaris?
[04:29:06] <wesolows> Whether that's destructive to data depends on what kind of filesystems you use and what tools you have.
[04:29:10] <FFForever> do*
[04:29:22] <FFForever> ext3 so i should be fine :D
[04:29:26] <wesolows> A full install with some space for a small home directory should fit in 10GB.
[04:29:37] <FFForever> so 15gb it is :D
[04:30:24] <boyd> wesolows: Viceroy? Nice, your excellency. Thing is, a viceroy normally represents a monarch. Who's your boss?
[04:30:52] <wesolows> boyd: Technically, Ed Hunter.  But operationally, mws himself. :-)
[04:31:09] <boyd> :)
[04:31:44] <wesolows> Someday I hope to be promoted again to a fully independent governor.
[04:31:53] <wesolows> And then eventually to Imperator.
[04:31:57] <coffman> FFForever: solaris can not read ext3 btw...
[04:32:17] <FFForever> coffman, i don't pull files off of other os's
[04:32:17] <wesolows> coffman: Is that true?  I thought we had a r/o ext2 implementation somewhere.
[04:32:18] <FFForever> :D
[04:32:37] <wesolows> Mr. Sullivan wrote it.  Maybe ext3 broke it though.
[04:32:41] <FFForever> i run 2 linux distros, 1 windows, and now solaris :D
[04:33:21] <FFForever> can solaris read ntfs?
[04:33:33] <jmcp_> not that I'm aware of
[04:33:40] <FFForever> what about fat 32?
[04:33:50] <wesolows> that works, I think - pcfs
[04:34:19] <jmcp_> as much as pcfs can be called a filesystem
[04:34:24] <wesolows> with write support even
[04:34:27] <FFForever> pcfs?
[04:34:28] <wesolows> yeah, true
[04:34:44] <wesolows> pcfs is the module/filesystem name that supports various FAT implementations.
[04:35:26] <FFForever> good D:
[04:35:30] *** Teknomancer has joined #opensolaris
[04:37:58] <FFForever> why does sun use a .zip for the part files?
[04:38:13] <wesolows> No one knows.
[04:38:26] <e^ipi> to make them smaller
[04:38:29] <e^ipi> *shrug*
[04:38:39] <wesolows> As opposed to gzip, I suppose.  Or bzip2.
[04:38:54] <wesolows> My answer to this type of question will henceforth be "top spite you."
[04:39:27] <FFForever> woot only 8hours till i can install solaris.....
[04:39:45] <FFForever> 9*
[04:40:17] <e^ipi> wander down to a university campus, you'll have it in 20 mins
[04:40:25] <boyd> I think the zip files might be so that they don't scare away all the windows users who would be put off by bzip2.... but are otherwised unfazed by installing a completely new os :)
[04:41:11] <FFForever> wesolows, what do you do for sun?
[04:41:23] * boyd listens :)
[04:42:00] <CSFrost> you can read ext3, just not write.
[04:42:17] <CSFrost> ntfs is also readable
[04:42:20] <wesolows> FFForever: Whatever it asks of me.
[04:42:31] <wesolows> FFForever: My job title is "Staff Engineer"
[04:42:38] * boyd groans and eats popcorn
[04:42:55] <wesolows> boyd: Sorry, would you have preferred "secret stuff"?
[04:43:14] <boyd> :)
[04:43:27] <FFForever> where is sun located?
[04:43:41] <wesolows> Almost everywhere, we're a global company.
[04:43:50] <wesolows> HQ is in Menlo Park, I think.
[04:44:00] <Auralis> near the gravitational center of the solar system :)
[04:44:15] <wesolows> It moves fairly often, for reasons I don't know.
[04:44:26] <FFForever> lmao @ Auralis
[04:44:32] <wesolows> Auralis: Yes, that's a better answer, I think.
[04:45:09] <boyd> Well, if you calculate the average position of all Sun offices over a year....
[04:46:23] <jmcp_> wesolows: Santa Clara, according to sun.com
[04:46:48] *** delewis has joined #opensolaris
[04:47:03] <wesolows> jmcp_: Yeah maybe so.  It's odd - I can never remember which building houses the execs but I thought it was MPK10
[04:47:26] <CSFrost> for investor details, they like to say Santa Clara
[04:47:45] <CSFrost> but for actual work getting done, it doesn't matter so much
[04:47:50] <FFForever> how many ppl in here work 4 sun?
[04:47:58] <boyd> Is there actually a legal requirement for anything specific to be at the "headquarters"?
[04:48:12] <CSFrost> only to please the stockholders.
[04:48:21] <wesolows> I don't know.  Receiving mail, I suppose.
[04:48:24] <FFForever> isn
[04:48:25] <CSFrost> and where the tax paperwork is sent from
[04:48:30] <FFForever> 't the board normally located there
[04:48:40] <CSFrost> nope
[04:48:47] <zooko> wesolows: I hope you're designing some sturdy new staves.
[04:48:54] <CSFrost> boards these days can be strong across several countries.
[04:49:27] <FFForever> i wish ms would buy adobe :D
[04:49:34] <CSFrost> I'd sure as hell hope not
[04:49:39] <wesolows> zooko: For making barrels of whiskey?
[04:49:41] <boyd> Anyway, they have to be somewhere that the chairs can legally dump the people into lava
[04:49:50] <FFForever> btw does flash work on os?
[04:49:50] <CSFrost> and I don't think we'd have to worry about that.
[04:49:55] <zooko> Err, I meant "staffs" ?
[04:50:10] <CSFrost> FFForever. google "flash solaris"
[04:50:15] <galt> boyd I hear yellostone is good for that....
[04:50:22] <wesolows> zooko: Ah, that.  Well, I don't have a degree in Staff Engineering for nothing!
[04:50:29] <CSFrost> staffs? I just ordered a new cane recently, if your shopping and looking around..
[04:50:40] <coffman> hmm
[04:51:03] <sbahra> So until now, I haven't seen any big advantage with Solaris lgrp's compared to Linux's trivial first touch policy.
[04:51:06] <coffman> about dladm, it lacks about showing the channel
[04:51:08] <sbahra> s/big/noticeable/
[04:51:19] <coffman> i mean wifi
[04:51:35] <CSFrost> besides.. sun's staffs are made in oregon and scotland I believe.
[04:52:08] <CSFrost> hrm, is that plural right?
[04:52:33] <wesolows> No, he had it right - staves.
[04:52:36] <boyd> "staves" more correctly, but I think you'll find it's acceptible
[04:52:39] <boyd> able
[04:52:52] <CSFrost> yea, staves is plural of the staff as a walking tool..
[04:53:08] <CSFrost> staff is plural of staff as in a employment term..
[04:53:25] <CSFrost> I thought it looked funny :-P
[04:53:29] <boyd> Staff as in employment is a mass noun, like "sand"
[04:53:40] <CSFrost> or deer
[04:53:55] <zooko> "Look at that tub of deer you got there."
[04:54:10] <CSFrost> it's new on KFC's menu.
[04:54:15] * zooko laughs.
[04:54:29] <boyd> ew
[04:54:56] <boyd> CSFrost: seriously, though, no, sheep and deer are not (normally) mass nouns
[04:55:01] * sbahra walks away
[04:55:16] <wesolows> sbahra: I care, but I'm not really an lgrp expert.
[04:55:41] <wesolows> sbahra: And it seems that the guys who are aren't here.  Have you posted in the performance group's lists?
[04:55:50] <sbahra> wesolows, http://www.kerneled.org
[04:55:59] <sbahra> I'm doing a comparison test now, to see if some QoS can show good performance improvement.
[04:56:14] <boyd> ... since you can can "a sheep" or "two deer" but you can't have "a sand" or "two margarine"
[04:56:17] <sbahra> Another factor is that my microbenchmarks are single-threaded. Haven't gone to multi-threaded benchmarks yet.
[04:56:44] <CSFrost> moose buckets..
[04:56:54] <CSFrost> crispy crunchy..
[04:58:07] <CSFrost> boyd, well you have "sand" as any amount, or you have "sand" as in beaches..
[04:58:09] <FFForever> what is the difference from star office and open office?
[04:58:33] <e^ipi> sbahra, solaris comes alive when you start threading like crazy
[04:58:34] <CSFrost> ones open-source, ones not?
[04:58:42] <boyd> CSFrost: sorry?
[04:59:01] <sbahra> e^ipi, in this case, lgrp's should show noticeable improvements :-)
[04:59:01] <CSFrost> boyd, "a grain of sand"
[05:00:07] <FFForever> e^ipi, that is all?
[05:00:12] <e^ipi> ?
[05:00:17] <sbahra> There is some other problem too.
[05:00:23] <sbahra> I can't pinpoint it
[05:00:30] <sbahra> but for example, cpu0 always has high latency
[05:00:39] <CIA-26> fb209375: 6571758 scsa2usb:scsa2usb_pkt_completion code is bcopying beyond their allocated buffer boundary
[05:01:12] <boyd> Yes, that's "a portion of a mass" e.g. a spoonful of sand, a molecule of water, but it makes it none the less mass
[05:01:53] <CSFrost> but if I find one grain of sand in my shoe, I can say "I have sand in my shoe"
[05:02:05] <CSFrost> right? :-)
[05:02:12] <sbahra> I think the issue is that
[05:02:19] *** Hunger- has quit IRC
[05:02:23] <sbahra> Solaris is associating both CPU *and* memory with an lgrp.
[05:02:33] <boyd> Yes. But you wouldn't say "I have a sand in my shoe", which you would if it were, for example, stones we were talking about
[05:02:34] <sbahra> Making QoS stuff kind of hard.
[05:02:45] <FFForever> are the commands the same on solaris and linux?
[05:02:55] <wesolows> sbahra: That's the expected result, I'd think - since the memory is attached to a CPU
[05:03:13] <boyd> FFForever: some commands are the same some are similar some are different
[05:03:15] <wesolows> sbahra: At least that's the case for both opterons and the big SPARC boxes (memory is on a CPU board)
[05:03:20] <sbahra> wesolows, but for example, topology would be more useful.
[05:03:40] <sbahra> hrm
[05:03:50] <sbahra> I wish we had Solaris on our 8-socket AMD system ;[
[05:03:58] <wesolows> If you say so - as I said, I'm really not an expert and have spent very little time thinking about this problem.
[05:04:07] <boyd> I wish *I* had solaris on your 8 socket AMD system
[05:04:15] <sbahra> wesolows, lgrp in a way does propagate topology, but it might be useful to pin on another CPU in specific while associating with memory from another cpu.
[05:04:26] <sbahra> ogate
[05:04:38] *** Hunger- has joined #opensolaris
[05:04:41] <boyd> AFAIK (sparc at least) that's always slower on current HW
[05:05:06] <CSFrost> I just wish amd would rush out the quads :-P
[05:05:09] <sbahra> boyd, not necessarily. Assume there is no memory on the node.
[05:05:23] <sbahra> CSFrost, I actually ran some of the microbenchmarks on one of their latest quads from fab.
[05:05:28] <sbahra> CSFrost, well, this AMD guy was nice enough
[05:05:37] <sbahra> CSFrost, *excellent* performance, 5% performance drop max due to ccNUMA
[05:06:01] <CSFrost> I've been waiting since I ordered these boards in Jan.. :-P tell him to hurry up
[05:07:00] <sbahra> boyd, or shared data
[05:07:41] <sbahra> boyd, in which case a CPU might be closer in terms of hops to all on another lgroup.
[05:08:35] <CSFrost> well, I am off.. need to atleast get some rest so my left arm can unswell :-p
[05:09:14] <boyd> sbahra: I see your point, but (at least on the sunfire arch) once you hit the x-bar they're all equally far.
[05:09:27] * boyd goes to do some paid work :)
[05:10:53] * libkeiser seconds sbahra's suggestion. we really do need a better notion of topology
[05:11:02] <rbrown_> well its official I've switched from boring ass compsci to cis programming track
[05:11:13] <rbrown_> two semesters of compsci had to be the worse ever
[05:11:28] <rbrown_> man the US compsci courses need a major overhaul
[05:12:13] *** djgregor has joined #opensolaris
[05:13:02] *** pguser has joined #opensolaris
[05:13:23] <pguser> hi, what does it mean when you get a "no os instance found"?
[05:13:57] <pguser> I booted via my solaris 10 disk 1, and normally it finds the solaris instance
[05:14:44] <pguser> and then I do an /sbin/installgrub -m /boot/grub/stage1 \ /boot/grub/stage2 /dev/rdsk/c0d0s0
[05:14:53] <pguser> but now I don't see anything!
[05:14:57] <pguser> :-(
[05:15:14] <pguser> yet when I do an fdisk -l in linux I see the solaris partition!
[05:15:21] <pguser> what's going on?
[05:15:31] <pguser> more importantly, how do I fix it?
[05:15:50] <pguser> oh, looks like this group is dead.....
[05:17:35] *** pguser has quit IRC
[05:18:15] <jmcp_> how impatient. he'll never make it as an Ent
[05:20:34] <boyd> lol
[05:20:47] <sbahra> libkeiser, :-D
[05:21:26] <galt> hey at least he stayde three minutes here, he stayed one minute in #solaris
[05:21:33] <libkeiser> sbahra: i've been complaining to myself about this exact problem for months
[05:21:59] <boyd> Oh, I don't mean to say it's not a good idea, I just mean to say that on current HW I don't know if it helps
[05:22:14] <sbahra> libkeiser, do you have specific applications that have suffered from this?
[05:26:27] *** Reidms has joined #OpenSolaris
[05:26:52] <sbahra> libkeiser, also, wondering if you've had time to look at that paper since the move
[05:26:57] <libkeiser> sbahra: well, one of my side projects has been to push a numa-awareness abstraction into openafs (our existing trivial about of numa-awareness is irix-only), and my new abstraction works by reporting topology to callers.  on solaris, i have to punt.  yes, i agree many mainstream vendors are punting on the problem by flattening out their topologies, but i still don't think that absolves vendors from reporting richer data
[05:27:54] <libkeiser> sbahra: i've read through about three quarters of it.  i've been unexpectedly swamped the past two weeks. sorry about that.
[05:29:17] <libkeiser> ok, enough bitching on my part. it's a small issue in the whole scheme of things (power4 being a notable exception)
[05:29:56] <sbahra> NP :-)
[05:30:17] <sbahra> libkeiser, http://www.opensolaris.org/os/community/performance/numa/8opteron/
[05:30:22] <sbahra> libkeiser, we're not the only ones that agree.
[05:32:46] *** jcsmith has quit IRC
[05:33:42] *** jcsmith has joined #opensolaris
[05:45:26] *** apokayi has joined #opensolaris
[05:47:23] <sbahra> Alaska sux
[05:48:45] <apokayi> hehe
[05:58:11] *** halton has joined #opensolaris
[06:03:06] *** sahafeez has joined #opensolaris
[06:04:02] *** delewis has quit IRC
[06:05:41] <FFForever> btw do i need a code 2/registration number for os express edition?
[06:06:42] <moazamraja> no
[06:14:33] *** xinkeT has quit IRC
[06:19:18] *** Triskelios has quit IRC
[06:32:49] *** Triskelios has joined #opensolaris
[06:35:36] <laserdrill> question from a Solaris 10 newbie. I am looking at the instructions to install PL/Java for Postgresql and it says: cd /usr; ln -s /usr/jdk/instances/jdk1.5.0 java; which creates a sym link. But I already ahve a symling named java in the user directory as i have JDK and the latest JRE installed (as well as the previous JRE. Would this be necessary or is it that they would be assuming that there was no java symlink in /usr? Can any
[06:40:16] <jmcp_> laserdrill: does the PL/Java stuff work with java6, or does it require java5 ?
[06:40:28] <laserdrill> Incidentally, my file browser quit and all the desktop icons have vanished ! Haven't seen this happen since Windows 95 I think! Any ideas what might cause that? The system seems to be ok otherwise but i haven't tried to use the file browser since.
[06:41:01] <laserdrill> It does not mention either one.
[06:41:48] <jmcp_> laserdrill: so you've already got a symlink there, and I hope PL/Java would just work
[06:41:56] <jmcp_> restart nautilus
[06:43:11] <laserdrill> I must say that third party documentation for the Solaris 10 platform seems nothing like the Sun documentation which is usually very clear and detailed so the average person could follow it. Tis stuff looks almost like a prescription.
[06:43:48] <laserdrill> jmcp: thanks.
[06:44:05] *** dlynes_laptop has joined #opensolaris
[06:45:31] <jmcp_> laserdrill: for the project I've been working on, we've had to add some new chapters to the mpxio admin guide, and re-edit other chapters .... the amount of effort we go to to make sure that there are no mistakes and that damned near anybody can pick it up and follow it easily,..... it's amazing. otoh, the effort pays off
[06:47:58] <laserdrill> jmcp:I completely understand as i handle a fair amount of document creation and editing myself
[06:48:25] *** The-spiki has quit IRC
[06:48:38] <jmcp_> It's Got To Be Right(tm)
[06:48:46] <jmcp_> that's how we approach documentation
[06:48:53] <jmcp_> :)
[06:49:11] <laserdrill> I wouldn't mind helping with documentation for this project, but I don't know how and need to learn a lot about how some of this stuff works.
[06:49:58] *** stevel has joined #opensolaris
[06:49:58] *** ChanServ sets mode: +o stevel
[06:50:16] <laserdrill> I am with you on that - its a philosophy I support wholeheartedly.
[06:51:33] <laserdrill> Is there some way I can help with documentation, jmcp?
[06:52:38] *** WickySolaris has quit IRC
[06:53:35] <stevel> you could join the docs community/lists
[06:54:03] <stevel> http://opensolaris.org/os/community/documentation
[06:54:30] <laserdrill> ok. I'll see about it now.
[06:55:29] *** sahafeez has quit IRC
[07:05:32] <laserdrill> Stevel: I've bookamrked that. Will definitely get involved with that. Seems I could do something useful there.
[07:07:14] <stevel> laserdrill: great, thanks; i'm sure the docs community will appreciate some more help
[07:07:28] *** Murmuria has joined #opensolaris
[07:07:43] *** WickedWicky has joined #opensolaris
[07:07:46] <WickedWicky> morning
[07:07:59] *** sahafeez has joined #opensolaris
[07:11:25] <laserdrill> stevel: Will do my best. My system has become unstable and I cannot, at this point, tell why. May have to restart soon.
[07:12:08] <laserdrill> Nautilus won't restart either
[07:13:03] <jmcp_> try a simple logout/login rather than restarting
[07:13:11] <jmcp_> and please #define unstable, too
[07:13:27] <boyd> #define unstable 1<<13
[07:14:12] <boyd> (sorry)
[07:14:42] *** WickedWicky has quit IRC
[07:15:08] <laserdrill> well by unstable I mean, typing nautilus at the command prompt does not result in any visisble activity for as lung as it runing, and when i return to this applet i sometimes cannot type anything in the text box even though the cursor is there.
[07:15:15] *** WickedWicky has joined #opensolaris
[07:16:05] <laserdrill> I have to copy and paste somethng there, then it starts working again and I can type text. Stuff like that.
[07:21:44] *** The-spiki has joined #opensolaris
[07:22:33] *** laserdrill has quit IRC
[07:29:58] *** Laserdrill has joined #opensolaris
[07:30:40] *** karrotx has joined #opensolaris
[07:31:32] <Laserdrill> Just restarted. Seems to be ok now
[07:39:31] *** karrot-x has quit IRC
[07:42:25] <Teknomancer> weird, i've got sunddi.h included, but i still get an incomplete type for "dev_info_t" ...
[07:45:08] *** loke__ has joined #opensolaris
[07:50:50] *** Reidms has quit IRC
[08:03:55] *** Murmuria has quit IRC
[08:05:41] *** loke_ has quit IRC
[08:06:55] <jmcp_> Teknomancer: #include <sys/dditypes.h> as well
[08:08:18] <Teknomancer> jmcp_:  i tried that ... as well
[08:09:12] <jmcp_> Teknomancer: what are you trying to build?
[08:09:31] <Teknomancer> a simple kernel module...
[08:09:42] <Teknomancer> i'm not getting a Header not found error
[08:10:00] <jmcp_> what's your compilation command line look like?
[08:10:03] <Teknomancer> but i'm getting an error with dev_info_t not being found (compiler error)
[08:11:11] <Teknomancer> must debug more...
[08:11:18] <Teknomancer> brb
[08:24:22] *** Laserdrill has quit IRC
[08:28:37] *** apokayi has quit IRC
[08:28:40] *** mazon is now known as Mazon
[08:32:25] *** WickedWicky has quit IRC
[08:34:30] <Teknomancer> k it builds using custom command "gcc" but fails within the custom build environment
[08:34:38] <Teknomancer> so its not a bug in the code as such ...
[08:38:31] <jmcp_> figured as much
[08:38:44] <jmcp_> now if you'd care to actually answer my question from earlier we might be able to help you
[08:38:45] *** NikolaVeber has joined #opensolaris
[08:39:28] <Teknomancer> jmcp_:  well i don't know if my company would like me to reveal what i'm coding for solaris :(
[08:40:29] <Teknomancer> that's the problem, don't wanna get sued :)
[08:42:46] <jmcp_> I asked you what your compilation command line was
[08:43:08] <jmcp_> if your corp thinks that *that* is confidential, then .... you get what you deserve, which certainly ain't assistance from me!
[08:43:23] <jmcp_> I don't give a rats what you're coding, but you had a problem with your command line
[08:43:24] <Teknomancer> err, the compilation command has the name of the proggie
[08:43:29] <jmcp_> so elide it
[08:43:40] <jmcp_> or, I dunno, substitute something like "programname"
[08:43:41] <Teknomancer> k, will change that and show u
[08:43:42] <jmcp_> duh
[08:44:08] <Teknomancer> sry for being slow, i'm just in conversation with around 4 people or so ...
[08:44:26] * boyd thinks this is not helping....
[08:44:40] <boyd> jmcp: How's Brisvegas?
[08:45:20] <jmcp_> boyd: doing quite nicely, thankyou
[08:45:55] <boyd> Nice to hear you're settling in. Funny how the little things help (like cable fixing)
[08:46:19] <jmcp_> yup, it sure is
[08:46:35] <jmcp_> planning on replacing a deadlock and an interior door handle tonight
[08:46:37] <Teknomancer> modyfing
[08:46:48] <jmcp_> boyd: still trying to work out an appropriate front door lock replacement
[08:46:55] <jmcp_> the current one is .,...... not good, shall we say
[08:46:57] <boyd> Doberman? :)
[08:47:46] <jmcp_> doberman and codercat don't go together :)
[08:47:53] <boyd> Hehe
[08:48:08] <boyd> Incompatible OSs huh?
[08:48:24] <jmcp_> yup
[08:48:46] <jmcp_> nah, the problem is that the existing lock is basically insecure and if you've got a screw driver you're in
[08:49:16] <jmcp_> which we don't find very reassuring, for some reason
[08:49:23] <Teknomancer> jmcp_ http://pastebin.com/d35000f08
[08:49:40] <boyd> Of, course, in many places, if you have a brick, you're in.
[08:50:05] *** JonathanW has joined #opensolaris
[08:50:39] <jmcp_> boyd: very true
[08:52:39] *** danv12 has quit IRC
[08:55:43] * jmcp_ checks the doco for studio cc
[08:56:29] *** rbrown_ has quit IRC
[08:57:41] *** yongsun|wfh has joined #opensolaris
[08:58:09] *** danv12 has joined #opensolaris
[08:58:51] *** JWheeler has quit IRC
[08:59:04] <jmcp_> Teknomancer: from the gnu cpp help output :::   -MP                         Generate phony targets for all headers
[08:59:07] <jmcp_> I don't think you really want that
[09:01:00] <Teknomancer> jmcp_ hmm... ok, i will ask why our guys have put that there in the first place. thanks :)
[09:01:20] <jmcp_> you're welcome
[09:01:22] *** dme has joined #opensolaris
[09:01:59] <jmcp_> you might also want to ask them why they're not using a separate "make depend" target in the makefile - it's my experience that separating dependency generation out makes your whole compilation process much more streamlineed
[09:02:22] *** tsoome has joined #opensolaris
[09:02:25] <Teknomancer> u mean platform specific makefiles ?
[09:03:09] <jmcp_> not necessarily
[09:03:13] <jmcp_> actually no
[09:04:45] *** Cyrille has joined #opensolaris
[09:05:30] <Teknomancer> lunch ,bbiab
[09:05:59] <jmcp_> for each .c and .h file that you have in your project, you can get make (or /usr/openwin/bin/makedepend iirc) to generate dependencies so that when the compiler gets fired up, it can save some parsing activity
[09:10:22] *** yongsun has quit IRC
[09:17:40] *** sahafeez has quit IRC
[09:19:58] *** jmcp_ has quit IRC
[09:21:43] *** Snake007uk has joined #opensolaris
[09:21:48] *** cydork has joined #opensolaris
[09:31:13] *** nostoi has joined #opensolaris
[09:34:46] *** dlynes_laptop has quit IRC
[09:42:16] *** monkey_Hou has joined #opensolaris
[09:43:30] *** ichigo has joined #opensolaris
[09:44:04] *** alobbs has joined #opensolaris
[09:47:13] *** mega has joined #opensolaris
[09:57:09] *** yatesy_ has joined #OpenSolaris
[09:57:26] *** yatesy has quit IRC
[09:57:56] *** axxl has joined #opensolaris
[09:59:15] *** NikolaVeber has quit IRC
[10:00:06] *** dmarker has quit IRC
[10:00:15] *** dmarker has joined #opensolaris
[10:02:46] *** migi has joined #opensolaris
[10:02:55] *** yatesy_ is now known as yatesy
[10:04:13] *** KermitTheFragger has joined #opensolaris
[10:06:43] *** laca has joined #opensolaris
[10:12:02] *** dme has quit IRC
[10:13:16] *** dme has joined #opensolaris
[10:13:30] *** dme has quit IRC
[10:13:56] *** FFForever has quit IRC
[10:14:47] *** dme has joined #opensolaris
[10:15:18] *** mega has quit IRC
[10:17:32] *** mega has joined #opensolaris
[10:21:46] *** Murmuria has joined #opensolaris
[10:22:03] *** henrik_ has quit IRC
[10:24:14] *** solar-star has joined #opensolaris
[10:24:53] *** stevel has left #opensolaris
[10:25:29] *** Gman has joined #opensolaris
[10:32:13] *** solar-star has quit IRC
[10:32:24] *** solar-star has joined #opensolaris
[10:40:14] *** timsf has joined #opensolaris
[10:40:21] <timsf> morning everyone
[10:45:00] <Teknomancer> is this the correct way to load a kernel module in opensolaris:      copy the module to /usr/kernel/drv   then add_drv <driver_name> ?
[10:45:11] *** danv12 has quit IRC
[10:51:02] *** Dar has joined #opensolaris
[10:55:06] *** ilyxa has joined #opensolaris
[10:58:36] *** MattMan has joined #opensolaris
[11:02:41] *** _setuid_H has joined #opensolaris
[11:04:53] *** linma has joined #opensolaris
[11:15:25] *** yongsun|wfh has quit IRC
[11:21:37] *** Fish has joined #opensolaris
[11:22:04] <Fish> hello
[11:22:10] <_setuid_H> hello
[11:24:13] *** derchris has quit IRC
[11:24:29] *** derchris has joined #opensolaris
[11:26:43] *** aruiz has joined #opensolaris
[11:30:04] <tomww> Teknomancer: add the pci-ids to the command, if it is a device-driver module
[11:30:40] <Teknomancer> tomww:  nah, its just a pseudo driver to make an entry under /dev
[11:31:33] <tomww> what module/pseudo-dev is it?
[11:32:01] <Teknomancer> just a pseudo character device
[11:33:02] <asyd> \_o<
[11:34:56] *** jcea has joined #opensolaris
[11:35:22] *** jcea has left #opensolaris
[11:35:30] *** jcea has joined #opensolaris
[11:35:49] <tomww> you can explicitly load a module by: cd /kernel/drv; modload modulename  (or /kernel/amd64/drv; ...)
[11:36:07] <Teknomancer> oh , ok
[11:36:36] <tomww> or make the devicefile with the major/minor number and assignt this in /etc/name*  -  but im not shure what works for you
[11:37:00] <tomww> you didn't modload from the kernel-dir or the absolute patchname?
[11:37:12] <Teknomancer> hmm
[11:37:27] <Teknomancer> modload <modname> doesn't work (without using add_drv)
[11:37:32] <Teknomancer> if that's what u meant
[11:39:08] <tomww> what is the name of your character/device or module?
[11:39:22] <Teknomancer> sampdrv
[11:39:35] <Teknomancer> i tried modload ./sampdrv (from /usr/kernel/drv)
[11:39:44] <Teknomancer> after copying sampdrv there
[11:46:16] *** NikolaVeber has joined #opensolaris
[11:47:10] *** calumb has joined #opensolaris
[11:47:43] *** solar-star has quit IRC
[11:50:33] *** solar-star has joined #opensolaris
[11:50:34] <tomww> I can't find any references for sampdrv and solaris. So I'm wondering ...
[11:53:01] <Teknomancer> tomww:  ah no i'm writing it by look at the site/PDF docs
[11:53:56] <Teknomancer> it adds etc, just fails to attach
[11:54:01] <Teknomancer> or create /dev entry ...
[11:55:00] <tomww> ah ok :-) Do the device-driver tutorial help (or are they the ones you are actually reading) ?
[11:55:31] <Teknomancer> which one are u referring to ?
[11:55:45] <Teknomancer> i'm look at sun.com for reference, ramdisk and tonic_drivers.pdf  also
[11:55:54] <Teknomancer> ramdisk (example driver)
[11:56:41] <Teknomancer> 		if ((ddi_create_minor_node(dip, "c,raw", S_IFCHR, <-- does this mean its try to make /dev/c,raw ?
[11:56:58] *** Atomdrache has quit IRC
[11:57:23] *** ilyxa has quit IRC
[12:11:15] *** The-spiki has quit IRC
[12:14:32] *** Chihan has joined #OpenSolaris
[12:18:12] *** obsethryl has joined #opensolaris
[12:20:12] *** simford has quit IRC
[12:31:06] *** takahide has joined #opensolaris
[12:31:56] <renihs> http://www.johncore.net/sunburn/THE_ULTIMATE_CES_TSE_CHEATSHEET.html  nice and pretty complete one
[12:35:14] *** _setuid_H has left #opensolaris
[12:35:56] *** lplatypus has joined #opensolaris
[12:46:26] <cormac> I'm a bit stumped, I'm trying to setup Japanese character displaying/input
[12:46:46] <cormac> but cant seem to get anywhere, any idea's?
[12:48:45] *** nostoi has quit IRC
[12:49:14] *** halton has left #opensolaris
[12:57:41] *** Atomdrache has joined #opensolaris
[13:00:59] *** dlg has joined #opensolaris
[13:02:14] *** deather_ has joined #opensolaris
[13:04:11] *** obsethryl has quit IRC
[13:04:12] *** JonathanW has quit IRC
[13:04:14] *** Griffous has joined #opensolaris
[13:04:39] *** m0le has quit IRC
[13:11:37] <solar-star> Known error? ata_id_common: BUSY status 0xfe error 0x0
[13:12:22] *** Griffous is now known as JWheeler
[13:13:02] *** mav_eric has joined #opensolaris
[13:17:57] *** deather has quit IRC
[13:17:58] *** deather_ is now known as deather
[13:21:54] *** cmang_ has joined #opensolaris
[13:27:41] *** cmang has quit IRC
[13:31:05] *** Teknomancer has quit IRC
[13:35:17] *** Teknomancer has joined #opensolaris
[13:35:52] <Teknomancer> man... open solaris sure hates improper shutdowns
[13:36:44] <Tempt> How so? (Well, how so more than any other UNIX?)
[13:37:31] <Teknomancer> haven't really used other pure Unixs before , but it seemed quite picky the next time i boot that i must manually fix something (by entering safe mode)
[13:37:55] <Teknomancer> i thought ZFS could handle pretty much anything :P
[13:38:21] <asyd> are you have you had trouble with zfs ?
[13:38:24] <asyd> I don't think so
[13:38:58] <Tempt> Boot archive problems?
[13:38:59] <Teknomancer> not yet :P but the "recovery process" was quite slow .. i don't know what it did, scan the entire hard-disk .. not sure. but it took a while
[13:39:03] <Teknomancer> Tempt:  yes
[13:39:09] <Tempt> That's a sad part of the limitations imposed by grub and x86 BIOS.
[13:39:13] <Teknomancer> asked me to fix it, and i did ... just took quite a while
[13:39:20] <Teknomancer> without much feedback
[13:39:23] <trygvis> for some reason ufs get corrupted almost half of all the times I have to hard boot the box
[13:39:30] <trygvis> but ext3 on linux work every time :/
[13:39:45] <Teknomancer> bfs also was good
[13:39:51] <Tempt> That's a bit odd.
[13:39:59] <quasi> trygvis: I'm quite used to it being the other way around
[13:40:01] <Tempt> Have you manually disabled logging somehow?
[13:40:13] <quasi> trygvis: ext3 being broken very often
[13:40:14] <trygvis> nope
[13:40:35] <trygvis> I was a bit surprised myself as ufs is supposed to be logging and all
[13:41:53] <quasi> trygvis: you can turn off logging in ufs very easily
[13:42:15] <trygvis> but I haven't
[13:43:10] <quasi> so mount says the fs is logging?
[13:43:29] <trygvis> haven't checked, but I have not changed any mount options
[13:43:49] <quasi> do check just to make sure
[13:44:41] <trygvis> it just can't be off, it is a new installation of solaris
[13:45:17] <Stric> which solaris?
[13:45:27] <Stric> <= sol9  had it off by default
[13:45:31] <trygvis> 10 and a few builds
[13:45:43] <Teknomancer> hmm strange
[13:45:44] <Stric> then it's on..
[13:45:51] <Tempt> Man, I wish doing access-lists on routers was as easy as doing ipfilter.
[13:45:54] <Teknomancer> after the reboot add_drv gives me a different error :)
[13:45:57] *** dme_ has joined #opensolaris
[13:46:05] <trygvis> exactly
[13:46:39] *** dme_ has quit IRC
[13:46:53] <Teknomancer> now it says drvconfig: modctl failed to add major number binding. ....
[13:50:14] *** terdmonk has joined #opensolaris
[13:50:51] *** monkey_Hou has quit IRC
[13:53:11] *** mega has quit IRC
[13:53:42] *** Dink has joined #opensolaris
[13:55:20] <quasi> trygvis: humor me and check anyway - I've seen logging turned off on recent installs without having done anything myself
[13:58:08] <timsf> cormac, have you tried using /usr/sbin/localeadm ?
[13:59:12] <trygvis> will do
[13:59:19] *** hohum has joined #OpenSolaris
[14:00:09] <Teknomancer> anyone knows where solaris stores the major device number/name bindings ?
[14:01:04] <Teknomancer> etc/name_to_major perhaps...
[14:01:15] <asyd> man driver.conf
[14:01:24] <asyd> ah hmm, err
[14:01:25] <Tempt> What exactly are you trying to do that involves messing around with the drivers that much?
[14:01:38] <quasi> Teknomancer: like /etc/path_to_inst ?
[14:01:43] <Teknomancer> Tempt:  i loaded a driver that hung the system
[14:02:00] <Teknomancer> it was probably still residing in /usr/kernel/drv when i rebooted this time
[14:02:13] <Tempt> Which driver?
[14:02:19] <Teknomancer> my sample driver
[14:02:26] <Tempt> Oh dear.
[14:02:29] <Teknomancer> :P
[14:02:35] <Tempt> rm it?
[14:02:38] <Teknomancer> anyway, after fixing it
[14:02:40] <Teknomancer> yeah removed...
[14:02:44] <Teknomancer> i mean 'rm'
[14:02:52] <Teknomancer> i'm trying to re-add it using "add_drv"
[14:03:05] <Tempt> And it won't let you play.
[14:03:12] <Teknomancer> and it fails with drvconfig: modctl failed to add major number binding. ....
[14:03:13] <Teknomancer> yes exactly
[14:03:16] <Teknomancer> i need to reboot?
[14:03:37] <Tempt> /etc/driver* ?
[14:03:49] <Teknomancer> hm
[14:04:16] <Teknomancer> no trace of my driver in /etc
[14:05:49] <Tempt> path_to_inst?
[14:06:23] <Teknomancer> means ?
[14:07:07] *** master_of_master has joined #opensolaris
[14:07:10] <Teknomancer> no its not there either
[14:07:11] *** laca has quit IRC
[14:08:49] *** calumb is now known as calLNCH
[14:11:05] <mav_eric> hi all
[14:11:28] <mav_eric> could someone help me with a non-booting OpenSolaris Nevada?
[14:11:45] <cormac> timsf
[14:11:49] <mav_eric> I installed it in an Virtualization on my Mac Mini with Intel Prozessor
[14:11:57] <cormac> i did, but then i wasnt sure what to do
[14:12:06] <cormac> locale -a | fgrep ja_
[14:12:06] <cormac> ja_JP.PCK
[14:12:06] <cormac> ja_JP.UTF-8
[14:12:06] <cormac> ja_JP.eucJP
[14:12:12] <mav_eric> everything runs through but after the reboot
[14:12:13] <cormac> but how should I set these?
[14:12:29] <mav_eric> the system freezes with a sendmail error
[14:12:38] <timsf> if you logout of the system, there's a language selector on the main login box,
[14:12:46] <mav_eric> I've found the solution for this sendmail issue but I cannot edit etc/hosts
[14:13:02] <quasi> mav_eric: then boot in single user mode and fix it
[14:13:07] <timsf> check one of those that's either a *.UTF-8 locale, or any ja* locale and japanese input should work.
[14:13:16] <mav_eric> quasi:  how do I boot single users mode?
[14:13:17] <quasi> mav_eric: adding -s to the boot string
[14:13:18] <cormac> timsf: is there a way to do that from command line, caus eim using xfce
[14:13:22] <mav_eric> thanks
[14:13:23] <cormac> im not fond of gnome
[14:13:52] <cormac> like, set LC_CTYPE=ja_JP.eucJP
[14:13:53] <timsf> You need to set the locale that you're logging in with.
[14:13:57] <cormac> or such, like in linux
[14:13:58] <timsf> you want LC_ALL
[14:14:03] <cormac> hmmm
[14:14:06] <cormac> LC_ALL
[14:14:19] <cormac> that overrides everything, I'll try it
[14:14:23] <timsf>  (which is a wrapping environment variable for all the other locale ones..)
[14:14:30] <cormac> aye :)
[14:14:48] <cormac> im just worried It might override something I dont want
[14:14:54] <cormac> ok, ill try, brb
[14:14:54] <timsf> Such as ?
[14:14:57] <timsf> no worries..
[14:18:05] *** master_o1_master has quit IRC
[14:21:55] *** MattMan has quit IRC
[14:22:13] <Teknomancer> let me reboot and try
[14:26:43] *** Murmuria has quit IRC
[14:30:15] *** lplatypus has quit IRC
[14:30:18] *** Fubarovic has joined #opensolaris
[14:31:10] <Fubarovic> how do i find out which process is using a network port? i thought netstat would be able to do it (like on gnu/linux) but i haven't been able to figure it out
[14:31:53] <PerterB> lsof or pfiles
[14:32:11] <Fubarovic> i'll check those out, thanks
[14:35:47] *** lloy0076 has joined #opensolaris
[14:36:00] * lloy0076 sigh
[14:36:21] <lloy0076> I've decided I dislike the Lambda calculus despite the fact I can sort of understand it.
[14:36:41] <lloy0076> Its penchant for saying: "Oh, just replace X with Y and it'll be the same" kind of confuses me no end.
[14:37:18] *** Murmuria has joined #opensolaris
[14:37:33] *** Murmuria has left #opensolaris
[14:37:42] *** lloy0076 has left #opensolaris
[14:40:05] *** Teknomancer has quit IRC
[14:43:39] *** sioraiocht has quit IRC
[14:43:52] *** sioraiocht has joined #opensolaris
[14:44:43] *** Teknomancer has joined #opensolaris
[14:44:51] <Teknomancer> rebooting helped..
[14:45:47] *** axxl_ has joined #opensolaris
[14:46:20] *** kFuQ has joined #opensolaris
[15:00:13] <cormac> still don't work
[15:00:14] <cormac> :
[15:00:16] <cormac> :/
[15:00:26] *** cmang has joined #opensolaris
[15:00:34] <cormac> what packages do I need?
[15:00:42] <cormac> to be able to see asian fonts?
[15:02:27] <mav_eric> quasi: thanks I was able to fix problem number one
[15:02:44] *** cajetanus has joined #opensolaris
[15:02:49] <quasi> mav_eric: cool
[15:02:51] <PerterB> cormac: use localeadm, it should know these things
[15:03:04] <cormac> i did already, but didnt work
[15:03:12] *** axxl has quit IRC
[15:03:19] *** cajetanus has left #opensolaris
[15:04:10] <mav_eric> quasi: not quite. I'm again stuck with sendmail. Looks like either I have a working ethernet card or a working sendmail but not both at once
[15:05:06] <quasi> mav_eric: that seems unlikely - try to svcs -xv to find the issue with sendmail or run svcadm disable sendmail
[15:05:32] <timsf> cormac, what's your font path ?
[15:05:48] <mav_eric> quais: I alreay run svcs -xv but I'm a total newbie and the output didn't tell me anything
[15:06:22] <cormac> tsoome: appears not to be set
[15:06:23] <cormac> hmmm
[15:06:26] <cormac> timsf: rather
[15:06:42] <timsf> xset -q ?
[15:06:43] <cormac> or do you meant he Xorg font path?
[15:06:47] <timsf> The latter..
[15:06:52] <cormac> ahh
[15:07:09] <cormac> /usr/X11/lib/X11/fonts/TrueType/,/usr/X11/lib/X11/fonts/Type1/,/usr/X11/lib/X11/fonts/Type1/sun/,/usr/X11/lib/X11/fonts/F3bitmaps/,/usr/X11/lib/X11/fonts/misc/,/usr/X11/lib/X11/fonts/100dpi/,/usr/X11/lib/X11/fonts/75dpi/
[15:08:35] <timsf> Hrm, I have http://pastebin.ca/630732
[15:08:44] <timsf> Might be worth trying that.
[15:08:53] <cormac> looking now
[15:09:03] <cormac> I also jsut redid localeadm
[15:09:06] <cormac>  cat /usr/sadm/lib/localeadm/Locale_config_S11.txt |grep ja_
[15:09:11] <timsf> Better yet, login to a CDE or GNOME session, to check to see your locales are setup properly
[15:09:19] <cormac> timsf: they are
[15:09:20] <timsf> then work out why xfce isnt' working.
[15:09:21] <cormac> in gnome
[15:09:30] <timsf> right, so you have the correct packages then,
[15:09:32] <cormac> no wait
[15:09:37] <cormac> damn
[15:09:38] <cormac> sorry
[15:09:40] <cormac> I mean they are not
[15:09:44] <cormac> not working in gnome
[15:09:51] *** nachox has joined #opensolaris
[15:09:53] <cormac> its not an xfce issue,
[15:09:55] <cormac> :(
[15:09:56] <timsf> Aha, so you don't have the correct packages.
[15:10:03] <cormac> perhaps
[15:10:07] <nachox> morning people
[15:10:12] <cormac> morning
[15:10:19] <timsf> How are you verifying display of asian characters ?
[15:10:26] <cormac> using scim
[15:10:44] <cormac> http://blogs.sun.com/yongsun/entry/scim_and_uim_packages_for
[15:11:01] <cormac> yongsun updated his page with proper scim support, so I'm trying to get it working
[15:12:11] <timsf> Okay, first thing, get asian fonts appearing, next work out input method stuff
[15:12:15] *** cmang_ has quit IRC
[15:12:58] <timsf> set LC_ALL=ja_JP.UTF-8, start a new xterm, then run 'cal'
[15:13:09] <cormac> ok
[15:14:20] <cormac> its showing in english
[15:14:24] <cormac> Warning: locale not supported by C library, locale unchanged
[15:14:27] <cormac> damn
[15:14:34] <cormac> gah, I had this wokring before
[15:14:46] <cormac> but when I ran localeadm it seems to have broke it
[15:14:56] *** Gropi has joined #opensolaris
[15:15:27] <timsf> if a locale is properly installed, you should have a shared library for that locale in /usr/lib/locale/<locale name>/
[15:16:00] <cormac> /usr/lib/locale/ja_JP.UTF-8/
[15:16:03] <cormac> exists
[15:16:54] <cormac> and looks sane
[15:18:43] *** NilsAN has joined #opensolaris
[15:18:44] *** cmang_ has joined #opensolaris
[15:19:00] *** CSFrost has quit IRC
[15:19:02] * timsf isn't sure what to suggest next
[15:19:12] *** CSFrost has joined #opensolaris
[15:19:24] <cormac> yeah I'm going around in circles with it myself
[15:20:52] <cormac> maybe its cause ther is no LC_ALL in /usr/lib/locale/ja_JP.UTF-8/
[15:21:59] <timsf> Nope, that's not it - so long as there's a shared object, you shuold be okay. (compare/contrast existing working locales on your system to see)
[15:22:08] <cormac> bizarre
[15:23:54] *** cmang has quit IRC
[15:24:07] <timsf> (just localeadm'ing -a ja -d <foo> on my system now to check)
[15:24:52] *** calLNCH is now known as calumb
[15:25:12] <cormac> i installed some new fonts
[15:25:16] <cormac> adding     FontPath   "/usr/openwin/lib/locale/ja/X11/fonts/UTF8/"
[15:25:17] <cormac> now
[15:25:21] *** xushi_ has quit IRC
[15:27:35] *** calumb has quit IRC
[15:28:48] *** MikeTLive has quit IRC
[15:28:59] <cormac> for some reason locals always get me
[15:29:04] <timsf> Hmm. For any UTF-8 desktop, the system should be setting the font path from the OWfontpath I thought
[15:29:11] <timsf> eg. /usr/openwin/lib/locale/en_US.UTF-8/OWfontpath
[15:29:28] <coffman> cormac: /usr/openwin is xsun, isnt it?
[15:29:48] <cormac> coffman: yeah
[15:30:03] <timsf> bits of usr/openwin are used for all xservers I'd thought, at least that's where the fonts live
[15:30:04] *** lasseoe has quit IRC
[15:30:07] *** lasseoe has joined #opensolaris
[15:30:18] <timsf> Guess you could always xset +fp <directory> to see if that makes a different
[15:30:23] <timsf> s/different/difference/
[15:31:27] <timsf> Aha, dtlogin sources /usr/dt/config/Xsetup
[15:31:33] <timsf> which builds the appropriate fontpath
[15:31:40] *** cmang has joined #opensolaris
[15:31:51] *** dunc has joined #opensolaris
[15:33:50] <cormac> /usr/openwin/lib/locale/ has no ja_JP.UTF-8
[15:33:57] <cormac> probably why its crapping out
[15:34:47] <timsf> pkginfo SUNWjxplt ?
[15:35:24] <cormac> i think its installed
[15:35:25] <cormac> sec
[15:35:27] <nachox> timsf: you wrote the dump gui right? is it possible to have smf properties that disable some functionality? for example i want my users to be able to read the summary but not send the dump over the net
[15:35:47] <timsf> nachox, yeah.
[15:35:58] <timsf> For now, I haven't implemented the sending of dump files
[15:36:03] <timsf> (for exactly that reason!)
[15:36:15] <cormac> timsf: already installed
[15:36:19] <nachox> well, but once you have :)
[15:36:35] <timsf> I was thinking that having RBAC involved, to ensure that users would need a particular profile in order to send dumps.
[15:36:46] <timsf> that's a bit better than just an smf property I think.
[15:37:08] <Tempt> Hmm. That's an idea. Every time you have a machine panic, it automatically raises a case and uploads the dumps and an explorer.
[15:37:44] <timsf> Tempt: Yeah, to an extent I'm nervous about duplicating effort that's already going on elsewhere in Sun
[15:37:49] <timsf> (I hope!!)
[15:38:14] <timsf>  - that is, the whole  Sun control station thingy, managed services - all that corp. datacentre stuff
[15:38:27] <Tempt> I don't know if Sun would be happy about every panic getting a case raised. I think they rely on the fact that most customers just have a quick whinge and move on.
[15:38:34] <timsf> I was really thinking about laptop users here
[15:38:36] <timsf> yeah!
[15:39:17] <Tempt> Although doubling the caseload would be a plus for the Indian company that handles all that.
[15:39:18] <nachox> can i dos sun dumping storage just by repetedly dumping cores? :)
[15:40:02] <timsf> Yeah, all those questions become issues if I ever get around to adding that support
[15:40:38] <Tempt> Although I must admit nearly every panic I've seen is caused by either hardware problems or Veritas products.
[15:40:51] <timsf> I've seen the odd zfs one too ;-)
[15:41:40] <Tempt> I'd be happy to never use a veritas product again. Ever.
[15:42:03] <nachox> i've seen a few zfs ones and some related to the wpi driver, but those freeze the box, no dump there
[15:42:04] <timsf> I used veritas for a few afternoons back before ZFS was integrated,
[15:42:41] <timsf> hated it
[15:42:42] *** cmang_ has quit IRC
[15:42:46] <Stric> ZFS lacks stuff that veritas has though.. like migration stuff..
[15:42:58] <nachox> and srinking
[15:43:07] <Stric> moving to other disks, evacuating disks etc..
[15:43:08] <Tempt> I'm not sure why, but at least around these parts everyone latches onto VxVM and VCS and VVR.
[15:43:20] <Tempt> Stric: zpool export; zpool import.
[15:43:37] <Stric> Tempt: that doesn't move data from disk array #1 to #2
[15:43:43] <nachox> and zfs send/recv
[15:43:46] <Tempt> Stric: Oh, that.
[15:43:51] <Stric> that doesn't move either, it copies
[15:43:59] <Stric> I'm not talking about offline stuff
[15:44:00] <Tempt> Stric: I must admit that's a feature I wouldn't mind seeing in ZFS.
[15:44:24] <Tempt> Stric: At least the ability to play the same tricks as in SDS/SVM - mirror onto the new array, remove the old array.
[15:44:39] <Stric> Tempt: unless you're doing raidz
[15:44:58] <Tempt> I'm sure these features will arrive in time.
[15:45:05] <Stric> then you have to move to something new that has the same amount of disks
[15:45:37] <Stric> sure, but some say that vxvm has no use anymore and everyone should go zfs.. but that's not true.
[15:45:44] <Tempt> Frankly shrinking filesystems doesn't have that much appeal anymore.
[15:46:08] <Stric> not shrinking filesystems, but getting rid of disk storage you don't want to use anymore
[15:46:27] <Stric> for instance because it's getting old and might fail.. or added by mistake.. or needed for something temporary.. or ...
[15:46:30] <Tempt> The tiered storage in VxVM has uses as well.
[15:46:38] <cormac> ext4 will probably be better
[15:46:39] <Tempt> "added by mistake"
[15:46:39] <PerterB> being able to grow raidz incrementally (eg add a single new drive) is pretty widely requested
[15:46:57] * nachox smacks cormac
[15:47:00] <Stric> cormac: ext4 isn't a volume manager
[15:47:10] <Tempt> I think cormac was joking.
[15:47:25] <Stric> hahaha.. I don't get it.
[15:47:26] <Stric> :P
[15:47:47] <cormac> :)
[15:48:54] <cormac> tbh I wasnt paying any attention to the conversation, I just threw that in :]
[15:49:22] <Stric> good riddance.
[15:51:10] <nachox> there is always svm for those of us that are poor :)
[15:51:29] <timsf> I believe they're targetting the removal of devices for the end of the calendar year
[15:51:51] <timsf> (mail from Matt to zfs-discuss on Thurs. He did say it was a rough guess though...)
[15:53:20] *** monzie has joined #opensolaris
[15:53:50] <Tempt> Integrating SNDR with ZFS could be nice.
[15:54:53] <Tempt> Wouldn't it be nice to create a zpool on another host, and effective do a couple of zfs set commands and get replication.
[15:54:54] <asyd> SNDR ?
[15:55:11] <pjd-> timsf: Hi.
[15:55:21] <timsf> hey pjd
[15:55:32] <Tempt> asyd: replication...
[15:55:51] <timsf> the AVS sort of does that Tempt, doesn't it ?
[15:56:07] <asyd> ah yes
[15:56:10] <tomww> yes, i read about that...
[15:56:13] <Tempt> Yes, SNDR is part of AVS (I think, the acronyms and abbreviations change so often)
[15:56:37] <timsf> Ah.
[15:56:38] <JWheeler> is RUNPATH meant to have an unescaped -R in there? I thought they were all meant to be colon delimited?
[15:58:03] <nachox> StorEdge Network Data  Replicator
[15:58:04] *** farak has joined #opensolaris
[15:58:06] <Tempt> Ideal world would be just to set replication_mode=master; replication_target=$TARGET_IP ; replication_bearer=ssh ; replication_bandwidth=$bw on the master and replication_mode=slave ; replication_master=$MASTER ; replication_bearer=ssh on the target and watch it go for it.
[15:58:07] <nachox> that?
[15:58:34] <pjd-> timsf: I'm trying to make tests/functional/history to work and I see something like this:
[15:58:43] <pjd-> 08:56:14 ASSERTION: Internal journal records all the recursively operations.
[15:58:46] <pjd-> 08:56:15 SUCCESS: /sbin/zfs create testpool.18829/testfs.18829/fs1
[15:58:49] <pjd-> 08:56:15 SUCCESS: /sbin/zfs create testpool.18829/testfs.18829/fs2
[15:58:52] <pjd-> 08:56:15 SUCCESS: /sbin/zfs create testpool.18829/testfs.18829/fs3
[15:58:54] <pjd-> 08:56:16 SUCCESS: /sbin/zfs snapshot -r testpool.18829/testfs.18829@snap
[15:58:57] *** The-spiki has joined #opensolaris
[15:59:13] <pjd-> And then verify_history function doesn't find internal history it is looking for.
[15:59:14] <nachox> Tempt: add public key and private key options, i would hate having to write passwords every time :)
[15:59:54] <pjd-> timsf: It does something like this:
[16:00:05] <pjd-> $ECHO $line | $GREP "internal $subcmd txg:" | $GREP -w "$ds"
[16:00:32] <pjd-> Where $subcmd=snapshot, $ds=testpool.19053/testfs.19053/fs1@snap
[16:00:43] <pjd-> But $line I see are:
[16:00:50] <pjd-> LINE: [internal snapshot txg:20] dataset = 42
[16:00:50] <pjd-> LINE: [internal snapshot txg:20] dataset = 44
[16:00:50] <pjd-> LINE: [internal snapshot txg:20] dataset = 46
[16:00:50] <pjd-> LINE: [internal snapshot txg:20] dataset = 48
[16:00:50] <pjd-> LINE: zfs snapshot -r testpool.19053/testfs.19053@snap
[16:00:53] <Tempt> nachox: You get the picture though.
[16:01:21] <pjd-> So none of them have both: "internal snapshot txg:" and "testpool.19053/testfs.19053/fs1@snap" strings.
[16:01:24] <Tempt> nachox: Having a choice of bearers would be good, too.
[16:02:49] <pjd-> timsf: Am I running different version of ZFS or something? This is from the most fresh source from onnv-gate.
[16:03:55] <timsf> pjd, just checking
[16:04:51] <timsf> (it could be that the history tests are broken)
[16:10:16] <timsf> Right. The version of the test suite on opensolaris.org is old :-( The ZFS devs recently changed the code, and our test suite broke for a while
[16:10:59] <Stric> sunsolve gone bonkers?
[16:11:49] *** jpdrawnee1 has joined #opensolaris
[16:11:52] <timsf> it's fixed in our copy, but the one on opensolaris.org is still broken.
[16:12:04] *** Plouj has joined #opensolaris
[16:12:44] <timsf> The way we're doign things is that we have the official version of the zfs test suite, which is stable and baselined, which gets pushed to the other Solaris test groups (and opensolaris.org)
[16:12:48] <quasi> Stric: no
[16:13:01] <Stric> seems like it woke up again
[16:13:05] <timsf> and then the development version of the test suite, which can be a bit hit & miss,
[16:13:41] *** elijahwright has left #opensolaris
[16:14:11] <Gman> hi timsf
[16:14:57] <Plouj> "Power user Petra has a four-disk RAID5 workstation. Parity calculations make this a fairly slow set-up because of the number of writes of small files. She upgrades to ZFS and sees performance benefits, because small files are mirrored instead of included in parity calculations." - https://wiki.ubuntu.com/ZFS - is that true?
[16:15:17] <Plouj> RAIDZ avoids doing parity calculations for small files?
[16:15:21] <pjd-> timsf: Ok, that explains it.
[16:15:39] <timsf> pjd, we could put up weekly source drops of the development version if that'd help ?
[16:16:29] <timsf> Hey Gman
[16:16:33] <timsf>  you in portland yet ?
[16:16:41] <Gman> yeah arrived last night
[16:16:48] <pjd-> timsf: I don't want to add you work. How often is this stuff updated: http://src.opensolaris.org/source/xref/test/ontest-stc2/src/suites/zfs/ ?
[16:16:54] <quasi> Plouj: maybe ubuntu is not the most reliable source on info about zfs :)
[16:17:12] <timsf> pjd that's the offical stc gate afaik
[16:17:28] <Plouj> actually, I just found similar info on wikipedia: http://en.wikipedia.org/wiki/Non-standard_RAID_levels#RAID-Z
[16:17:29] <timsf> So that's the one the other test groups pick up
[16:17:46] <Plouj> "It avoids the need for read-modify-write operations for small writes by only ever performing full-stripe writes; small blocks are mirrored instead of parity protected, which is possible because the file system is aware of the underlying storage structure and can allocate extra space if necessary."
[16:17:54] *** Gman has quit IRC
[16:18:17] <timsf> It's updated about every 6 weeks or so - pretty much whenever there's sufficient new zfs functionality, or additional test cases we want the rest of the test org to test.
[16:19:02] <timsf>  (worst case scenario, would be where we putback every day to the main stc2-gate, and cause all zfs testing in PIT to grind to a halt because of a test case bug)
[16:19:29] <cormac> these locale not supported by C library errors are bizarre
[16:19:56] <timsf> However, doing a drop of the dev test suite as a tarball to the zfs community page mightn't be too compelx
[16:20:19] <cormac> I can help you test stuff
[16:20:20] <cormac> if you like
[16:20:57] *** Gman has joined #opensolaris
[16:20:58] *** Reidms has joined #OpenSolaris
[16:21:07] <cormac> timsf: what are you working on?
[16:21:16] <timsf> ZFS test development.
[16:21:42] <cormac> cool, if you want, when your ready I can test your stuff
[16:21:49] <timsf> still weird about that locale stuff
[16:22:07] <cormac> have you tried tonic?
[16:22:08] <Plouj> sweet
[16:22:21] <timsf> tonic ?
[16:22:58] *** sstallion has joined #opensolaris
[16:23:21] <cormac> yeah
[16:23:23] * sstallion sighs
[16:23:24] <cormac> sec ill like you
[16:23:26] <sstallion> im back
[16:23:54] *** danny_j has joined #opensolaris
[16:25:12] <Gman> timsf, is that crash dump thing using zenity?
[16:25:20] <timsf> Gman, yeah.
[16:25:29] <Gman> timsf, ok, thought so
[16:25:37] <Gman> not quite the right fit UI wise
[16:25:39] <timsf> (hence reluctance to add much GUI flash)
[16:25:44] <timsf> I know...
[16:25:56] * timsf doesn't want to be a gtk2 programmer...
[16:26:03] <Gman> timsf, you should knock something up in pygtk :)
[16:26:23] <timsf> I was playing around with glade last night - is that the way to go ?
[16:26:36] <timsf>  (+ python ?)
[16:26:50] <Gman> yeah, but use libglade if you're going to do anything at all
[16:26:59] <timsf> Right..
[16:27:12] <timsf> That said, I might be able to refine what I'm doing with zenity
[16:27:13] <Gman> it'll be very straightforward
[16:27:22] <timsf> after all, it's a fairly basic gui..
[16:27:50] <Gman> yeah
[16:27:58] <Gman> but zenity doesn't handle that type of thing very well right now
[16:28:08] <cormac> timsf: i was under the impression tonic was live, but apparently not, maybe :/
[16:28:17] <Gman> [ie. removing the treeview, and the radio list into a vbox]
[16:28:23] <cormac> http://test.opensolaris.org/selftest/
[16:28:26] <aruiz> timsf, I can help with the pygtk foo :-)
[16:28:26] <cormac> but it will be there
[16:28:29] <cormac> so stay tuned
[16:28:36] <aruiz> s/foo/fu/
[16:30:30] *** ichigo has quit IRC
[16:30:56] <timsf> cormac - yep, tonic is where our test suite gets deployed
[16:31:06] <timsf> thanks aruiz, I might well ping you at some stage
[16:31:29] <cormac> ahh i see, it doesnt say tonic.. strange
[16:31:32] <aruiz> timsf, sure :)
[16:31:50] <timsf> Thanks!
[16:32:23] <Gman> timsf, you should totally take him up on that offer, it's a 30 min hack, and it'll look very sweet indeed
[16:32:49] *** ichigo has joined #opensolaris
[16:34:11] <bda> SEE ALSO: http://xkcd.com/
[16:34:22] <timsf> Gman, righto, would like to have a bash myself first if I can, then call in the marines if it all goes pear-shaped..
[16:35:10] <Gman> timsf, heh
[16:36:32] *** gobbler has joined #opensolaris
[16:38:04] *** MattMan has joined #opensolaris
[16:42:23] *** hohum has quit IRC
[16:43:30] *** calumb has joined #opensolaris
[16:46:07] *** EchoBinary has joined #opensolaris
[16:48:07] *** Teknomancer has quit IRC
[16:48:43] *** jpdrawnee1 has quit IRC
[16:49:50] *** nnode has joined #opensolaris
[16:50:14] *** jgay has quit IRC
[16:50:39] *** Pietro_S has joined #opensolaris
[16:50:42] *** Somethingelse has joined #opensolaris
[16:54:31] *** pfa3rh has joined #opensolaris
[16:58:44] *** Reidms has quit IRC
[16:59:38] <Pietro_S> hello, is there any clever way how to copy 2 directories - copy only changed data ?
[17:00:08] <cormac> rsync
[17:00:12] <sickness> rsync
[17:00:15] <sickness> :)
[17:00:16] *** gobbler has quit IRC
[17:00:24] <cormac> rsync ftw
[17:00:35] <cormac> :>
[17:00:44] <sickness> rsync -cvr --progress --inplace /a/* /b/
[17:00:45] <sickness> :P
[17:01:02] <cormac> sometimes you also need to do --checksum also
[17:01:03] <sickness> or also -acvr
[17:01:04] <cormac> i find
[17:01:05] <sickness> well
[17:01:06] <sickness> yeah
[17:01:09] <nachox> did anyone mention rsybc already?
[17:01:10] <Pietro_S> thanks - looking in manual for more things :-)
[17:01:25] <cormac> bda: whats that did you mean rsync?
[17:01:26] <nachox> *rsync
[17:01:33] <cormac> nachox: rather
[17:01:34] *** soulie has joined #opensolaris
[17:01:37] <cormac> yeah rsync
[17:01:41] <sickness> I think it's even in by default in latest snvs...
[17:04:18] <cormac> yeah its on mine by default
[17:04:22] <cormac> this is 59
[17:04:27] <sickness> nice
[17:04:51] <cormac> I'm trying to bfu to 66 atm, I dont have a burner though
[17:04:56] <cormac> well no bfu
[17:04:57] <cormac> update
[17:05:03] <cormac> *not
[17:05:12] <timsf> (and if your directories are actually ZFS filesystems, then zfs send -i might be what you're after)
[17:05:21] <sickness> I usually leave another root to liveupgrade...
[17:05:32] <cormac> hmm
[17:08:04] *** Fish has quit IRC
[17:08:21] *** Fish has joined #opensolaris
[17:08:56] <cormac> I think I'll jsut try to fix 59, will be quicker
[17:09:06] <sickness> heh
[17:10:08] *** Fish has quit IRC
[17:10:44] *** Fish has joined #opensolaris
[17:12:58] *** calumb has quit IRC
[17:13:12] *** calumb has joined #opensolaris
[17:14:16] *** loke has joined #opensolaris
[17:15:15] *** loke__ has quit IRC
[17:15:23] *** dunc has quit IRC
[17:16:36] <Pietro_S> timsf: only one destination dir is zfs, I have zfs-ized /opt which lupgrade copied to ufs and then somehow modify ...
[17:19:08] <timsf> guess you need rsync then.. (zfs send -i allows for incremental sending of two snapshots, only sends the differences between two filesystems - should be very fast)
[17:19:25] <timsf> s/filesystems/snapshots/
[17:25:15] *** karrotx has quit IRC
[17:26:22] <Pietro_S> hmm, rsync ended with bunch of "skipping non-regular file ..."
[17:27:17] *** hsilva has quit IRC
[17:27:30] <Pietro_S> most of them are -r--r--r-- file in /opt/staroffice  , looks like cp need to be used again ;-)
[17:27:49] *** hsilva has joined #opensolaris
[17:29:13] *** jamesb_ has quit IRC
[17:29:17] *** jamesb has joined #opensolaris
[17:35:15] *** sartek has joined #opensolaris
[17:35:48] <EchoBinary> im having an interesting issue
[17:36:02] <EchoBinary> running a samba share
[17:36:30] <EchoBinary> trying to put data from old drives onto a new raidz2
[17:36:30] <EchoBinary> i have a gigabit network
[17:36:30] <EchoBinary> but the data transfers keep timing out
[17:37:02] <EchoBinary> transfering from linux via samba -> opensolaris
[17:37:40] <EchoBinary> i (lazily) tried it via wireless at first,and assumed the timeouts were due to the wireless being flakey
[17:37:51] *** Snake007uk has quit IRC
[17:38:07] <Pietro_S> by smbclient or nautilus?
[17:38:18] <EchoBinary> so i switched to ethernet
[17:38:18] <EchoBinary> same issue
[17:42:00] <hile_> why are you using samba for that?
[17:42:49] <asyd> btw, do you know if SE's nautilus is build with kerberos support ? (the sambra browser at least)
[17:46:03] *** Somethingelse has quit IRC
[17:49:08] *** tsoome1 has joined #opensolaris
[17:50:55] <cormac> is there a way to rebuild locals
[17:51:02] <cormac> like forcefully?
[17:51:13] <cormac> im using localeadm -a
[17:51:17] <cormac> but it says already installed
[17:51:26] <cormac> it fails trying to remove them witha syntax error
[17:51:31] <cormac> so i cant remove and then install
[17:52:37] *** tsoome has quit IRC
[17:53:59] *** Reidms has joined #OpenSolaris
[17:54:30] <timsf> pkgrm the locale packages manually I guess
[17:54:49] <cormac> messy, but i guess I ahve no choice
[17:55:05] <timsf> I agree.. I'm not too impressed with localeadm myself
[17:55:54] <_mary_kate_> hmm, a V20z (3GB RAM, 70GB disk) should have no problem with 13 sparse zones, right?  (apart from the application resources..)
[17:58:45] *** ichigo has quit IRC
[17:59:08] <EchoBinary> is it better to use nfs or samba with zfs? or does it not matter?
[17:59:35] <aruiz> EchoBinary, which clients are you going to have? unix clients?
[18:06:12] <Plouj> humm
[18:06:17] <cormac> it couldnt remove that package because I had it running in svcs,
[18:06:22] <cormac> removed now
[18:06:31] <Plouj> I could put my 3 160GB drives into a simple striped array and add a 500GB drive for redundancy...
[18:06:45] <cormac> still, syntax error. isnt the most greaceful fail
[18:07:51] *** alanc_away_ is now known as alanc
[18:08:07] <cormac> ok, now all Japanese localshave been uninstalled and freshly reinstalled from snv_59 cd, hopefully this will tify things up a bit
[18:09:30] <Plouj> since 160*3 is only 480GB, is it possible to mirror that on a 500GB drive and still be able to use the remaining 20GB from the large drive without creating slices?
[18:10:11] <cormac> timsf: i think the problem is that this wnn svcs service is in maintenance mode
[18:10:22] <cormac> for some reason
[18:10:23] <timsf> Aha. What does svcs -x tell you ?
[18:10:36] <cormac> timsf: well im reinstalling now, so i have ot wait
[18:10:45] <cormac> but i jsut noticed that while package removing
[18:10:49] <timsf> (and /var/svc/log/<wuh>.log )
[18:10:54] <cormac> it wouldnt let me cause it was a running service
[18:11:05] <timsf> Hrm
[18:11:06] <quasi> Plouj: my guess would be that you have to slice the drive
[18:11:07] <cormac> then i saw it was in maintenance
[18:11:16] <cormac> ill -vx it as soona s it reinstalls
[18:11:17] *** swmackie has joined #opensolaris
[18:11:19] * cormac plays waiting game
[18:11:24] <Plouj> humm
[18:12:08] <Plouj> I wish I could benchmark raid-z on three SATA disks but they are currently storing all of my data on the desktop :/
[18:12:13] <quasi> at least if you want to use the last bits
[18:12:42] *** migi has quit IRC
[18:12:43] *** calumb has quit IRC
[18:12:46] <Plouj> quasi: yeah, on second thought, slicing in that case is not an issue
[18:13:21] *** matsutsu has joined #opensolaris
[18:14:21] *** sommerfeld has joined #opensolaris
[18:14:22] *** ChanServ sets mode: +o sommerfeld
[18:15:14] <cormac> woah
[18:15:19] <cormac> now its all showing up crazy glyps
[18:15:27] <cormac> cause my local is swithced to japanese
[18:15:32] <cormac> but its jsut garbled
[18:15:47] <CSFrost> see what you started..
[18:15:56] *** nnode has quit IRC
[18:16:07] *** sahafeez has joined #opensolaris
[18:16:08] *** delewis has joined #opensolaris
[18:16:13] *** alanc is now known as alanc_away_
[18:16:24] *** slowhog has joined #opensolaris
[18:16:32] *** swmackie has quit IRC
[18:16:48] <timsf> what locale is your terminal running in ?
[18:17:02] <cormac> LC_ALL=ja_JP.eucJP
[18:17:09] <cormac> ill change to utf
[18:17:12] <timsf> nono,
[18:17:24] <timsf> I mean your terminal, not the shell that's running in it.
[18:17:33] <cormac> oh sorry
[18:18:15] <cormac> hmmm
[18:18:32] <cormac> I'm... note sure
[18:19:09] <timsf> pargs -e <pid of terminal> | grep LC_
[18:19:54] <cormac>  envp[20]: LC_ALL=ja_JP.UTF-8
[18:20:28] *** theRealballchalk has joined #opensolaris
[18:20:39] <timsf> there you go - you're running a UTF-8 terminal, and tyring to display codepoints from eucJP in it.
[18:20:49] <_mary_kate_> hmm, QFS and zones doesn't play nicely
[18:20:50] <cormac> ahh
[18:20:58] <timsf> hence corruption
[18:20:59] *** reflect_ has quit IRC
[18:21:01] <cormac> kk
[18:21:04] <cormac> i dig it
[18:21:08] *** reflect has joined #opensolaris
[18:21:10] <timsf> it /should/ be able to display ja_JP.UTF-8 though,
[18:21:17] <cormac> .. but I ahvnt been doing that all along,
[18:21:18] <timsf> assuming fonts are all nicely installed...
[18:21:26] <cormac> ok sec
[18:21:35] <cormac> ive been doing alot of messing around
[18:21:40] * timsf nods
[18:23:32] *** swmackie has joined #opensolaris
[18:26:39] *** KermitTheFragger has quit IRC
[18:30:04] *** Cyrille has quit IRC
[18:31:59] *** zooko has quit IRC
[18:32:45] <soulie> ;o why cant set my screensaver using the preferences->screensaver out of menu
[18:33:02] <soulie> using root
[18:33:29] <axisys> projadd -K 'rcap.max-rss=320mb' user.john caps each process not to exceed 320m .. but if the user runs 10 processes `prstat -s rss -t' shows john is using over 650mb in RSS.. and `pmap -x <PID>' of that user and add up the total numbers of Anon column shows over 620mb
[18:34:16] <axisys> how do I limit teh total number of Anon or even RSS a user can use?
[18:36:00] *** calumb has joined #opensolaris
[18:38:03] *** soulie has quit IRC
[18:39:48] <axisys> what mailing list would discuss this?
[18:40:45] <timsf> axisys - I guess http://mail.opensolaris.org/mailman/listinfo/rm-discuss
[18:41:03] <timsf> though I'm not subscribed myself - could be worth a dig in the archives to see if it's appropriate
[18:42:04] *** sartek has quit IRC
[18:42:09] *** zooko has joined #opensolaris
[18:42:26] <axisys> timsf: i got it too.. thnx
[18:42:39] <cormac> timsf:
[18:42:42] <cormac> success
[18:42:54] <timsf> Hurrah!
[18:42:57] <cormac> yah
[18:43:00] <cormac> only took me like 8 hours
[18:43:03] <timsf> Do I get beer now ? ;-)
[18:43:04] <cormac> :p
[18:43:08] <cormac> hmm
[18:43:08] <cormac> no
[18:43:12] <timsf> :-D
[18:43:13] <cormac> I jsut wasted a day
[18:43:15] <cormac> lol
[18:43:20] <cormac> so, I get a kick
[18:43:23] <cormac> :]
[18:43:50] <timsf> http://www.brainyquote.com/quotes/quotes/j/johnlennon108522.html
[18:44:56] <Pietro_S> wow, I just plug new keyboard (Genius Media Cruiser) to sxce and it works like charm, even multimedia keys worked - like changing volume, emulaion of mouse and power off button logout me from xfce! :-)
[18:45:37] <cormac> timsf: so true :)
[18:45:52] <cormac> but since I'm at work, my boss would probably have other ireas :p
[18:45:55] <cormac> *ideas
[18:50:01] <Pietro_S> does I need to reinstall ndiswrapper after lupgrade? wifi looks somehow borked
[18:55:42] *** gaz has joined #opensolaris
[18:56:52] *** WickedWicky has joined #opensolaris
[18:56:58] <WickedWicky> Hellows!
[18:57:58] *** estibi has joined #opensolaris
[18:58:40] <zooko> Does anyone know if the gigabyte m57sli-s4 motherboard will run OpenSolaris?
[18:59:17] *** perlmonk has joined #opensolaris
[18:59:18] *** obsethryl has joined #opensolaris
[19:01:41] *** linma has quit IRC
[19:02:02] <mav_eric> good bye
[19:02:03] *** mav_eric has left #opensolaris
[19:02:09] *** calumb has quit IRC
[19:02:22] *** calumb has joined #opensolaris
[19:03:21] *** jpdrawneek has joined #opensolaris
[19:03:36] <Pietro_S> zooko: I guess yes, any specialities there?
[19:04:03] <zooko> You mean -- what's special about that motherboard?  It's the first motherboard that you can buy retail that comes with linuxbios instead of a proprietary bios.
[19:04:09] <zooko> At least I *think* it comes with linuxbios.
[19:04:10] *** RaD|Tz has joined #opensolaris
[19:04:15] <zooko> If it doesn't, then I'm going to flash linuxbios myself.
[19:04:29] <zooko> Unfortunately, the process of installing solaris on linuxbios is not yet a well-trodden path...
[19:04:34] <WickedWicky> I think Pietro_S is more asking about special hardware specs that make you ask if its supported
[19:04:45] <zooko> So if that doesn't work, or if I just run out of time, then I'll install Ubuntu Linux on it instead.
[19:04:58] *** bubbva has joined #opensolaris
[19:05:02] <sommerfeld> zooko: someone has to be first...
[19:05:13] <zooko> WickedWicky: I thought that the state of the art was to assume that Solaris does *not* work on hardware unless that hardware is on the "HCL".
[19:05:29] <WickedWicky> assumptions are the.. etc etc
[19:05:54] <zooko> When I buy new hardware for Linux, I tend to google around on the linux mailing lists to find people talking about their attempt to install Linux on it...
[19:06:23] <sommerfeld> zooko: that's only half of it.  the other half is that if solaris doesn't run on popular hardware, it's worth understanding why (filing bugs, etc.,)
[19:06:47] <zooko> Good point.
[19:07:05] *** timsf has quit IRC
[19:07:50] <wesolows> If the HCL were exhaustive, that would be reasonable.  But it's woefully incomplete, and Solaris works just fine on a huge amount of hardware not listed there.
[19:07:59] <wesolows> Honestly, I never use the HCL.  It's far too narrow.
[19:08:23] <WickedWicky> I nevewr look at the HCL either, I install it, if it works, sweet, if not, I ask around or google for the specific thing I want to work
[19:08:47] <wesolows> Well, that's more haphazard than I like.  I look in usr/src/uts/*/io
[19:08:52] *** Pietro_S_ has joined #opensolaris
[19:09:12] <seanmcg> ping timsf
[19:09:19] <wesolows> Which is the authoritative source on what's actually going to work and what's not.
[19:09:48] <WickedWicky> as for working, rebooting to nv69 now on my AMD 2400+XP server, wish me luck
[19:09:51] <wesolows> Anything that attempts to condense and generalise that information is bound to be incomplete, inaccurate, out of date, or some combination thereof.
[19:11:14] *** alanc-away is now known as alanc
[19:11:36] <zooko> So what does /usr/src/utc/*/io say about the Gigabyte m57sli-s4 ?
[19:11:55] <wesolows> The mainboard itself is all but certain to work.
[19:12:08] <wesolows> Only the most exotic or grievously broken boards won't work.
[19:12:46] <wesolows> If you want to know which of its attached devices will work, you need to get the board's manual and get the numbers off the chips.
[19:13:09] *** Dar_ has joined #opensolaris
[19:13:22] *** Dar has quit IRC
[19:13:41] *** Dar_ is now known as Dar_HOME
[19:17:29] *** duri_ has quit IRC
[19:18:12] <WickedWicky> wesolows: do you know by heart if the PDC20276 ATA controller is suposed to work? It's the Promise MBFastTrak133 Lite
[19:18:29] <wesolows> probably not
[19:18:34] <WickedWicky> oh :(
[19:18:35] <wesolows> promise is bad bad bad
[19:18:50] <wesolows> In fact I think I had one of those once
[19:18:52] <WickedWicky> yes but it's also the default extra set of controllers on many K7 mainboards
[19:19:18] *** Pietro_S has quit IRC
[19:20:58] <Pietro_S_> I though that promise is 'working' -  drives are visible directtly and let you to do your software raid ...
[19:21:20] <WickedWicky> the entire controller is not detected it seems
[19:21:39] <WickedWicky> I'm using nv69
[19:22:18] <wesolows> I think it depends on whether the controller acts like an AT IDE controller.
[19:22:26] <wesolows> If so, it should be detected and should work.
[19:22:27] <_mary_kate_> do people still use promise?  i thought those died out with the end of the fake ide raid fad
[19:22:34] <wesolows> But my experience with that controller is that it does not.
[19:23:02] <WickedWicky> _mary_kate_: I use it as normal controller, not as RAID controller. I have a Via controller as primary and secundary, and use the promise controllers as third and fourth
[19:23:06] <wesolows> _mary_kate_: That was my impression as well.  Between their nonstandard stuff and lots of bugs and flakiness, it's a wonder they've stayed in business.
[19:23:40] <wesolows> Right, the 20276 is a basic ATA133 controller.  Unfortunately it seems to require a promise-specific driver.
[19:24:14] <wesolows> There was one in Linux at one point, I don't know about BSD.  I didn't have one for Solaris at the time and I'm not aware that anyone has written once since (given that the hardware is obsolete, that's not surprising).
[19:24:31] <WickedWicky> yea the controller works in linux, but sod linux
[19:24:40] <WickedWicky> I'll just plugin my Silicon ATA controller
[19:25:29] <wesolows> My only advice on Promise is never buy anything they make as an add-on, and never buy a mainboard that has their chips soldered on.
[19:25:50] <WickedWicky> wesolows: two years too late with that advice :D but I will keep it in mind, thank you
[19:25:52] <wesolows> They make SiS look good.
[19:26:03] <WickedWicky> now that is a harsh statement :P
[19:26:15] <wesolows> Maybe even Marvell.
[19:26:23] *** swmackie has quit IRC
[19:31:53] *** MattMan has quit IRC
[19:31:55] <Abe_Froman> haha
[19:32:08] <EchoBinary> primarily linux/bsd
[19:33:08] <EchoBinary> but i will have windows clients as well
[19:33:08] <EchoBinary> the windows clients will be reading media files
[19:33:08] <EchoBinary> the linux/bsd clients will be reading data, mostly source code, and databases
[19:33:32] <EchoBinary> i can find an NFS windows client for some ofn the win boxes
[19:33:35] <EchoBinary> but there is one, a modified XBox running XBMC that likely req. samba
[19:36:58] * _mary_kate_ wishes initial smf import was faster
[19:37:17] *** Jooles has joined #opensolaris
[19:37:18] *** EchoBinary has quit IRC
[19:37:33] <Jooles> hi
[19:38:37] *** EchoBinary has joined #opensolaris
[19:39:26] <zooko> Darn.  I just figured out that the motherboard that I bought doesn't actually *come* with linuxbios.
[19:39:37] <zooko> It's just that linuxbios is known to run on it.
[19:39:48] <zooko> But I will have to flash it myself, which risks bricking my new motherboard if I screw up...
[19:41:03] <sommerfeld> zooko: is the flash chip socketed? (most of them are...)
[19:41:16] <zooko> http://linuxbios.org/GIGABYTE_GA-M57SLI-S4_Build_Tutorial
[19:41:22] <zooko> "The fact that the BIOS is soldered onto the board complicates matters considerably, because it means that one flash of a faulty image will render your board unusable (it will be 'bricked'). Top Hat Flash does not work with the M57's SST 49LF040B 33-4C-NHE, but might work with other FWH."
[19:41:31] <sommerfeld> d'oh
[19:42:01] <sommerfeld> i've rescued three or four motherboards by swapping chips.
[19:42:38] <zooko> Hm.  This is not something that I have experience in.  On the other hand, it sounds kind of fun.  :-)
[19:44:10] *** Chihan has quit IRC
[19:44:33] *** EchoBinary has quit IRC
[19:44:34] *** EchoBinar1 has joined #opensolaris
[19:44:34] *** jcea has quit IRC
[19:44:43] *** Chihan has joined #OpenSolaris
[19:44:52] *** EchoBinar1 is now known as EchoBinary
[19:45:09] <sommerfeld> there are a couple vendors out there who will sell you a flash chip preprogrammed with the bios of your choice (and appropriate chip extractor tools).
[19:45:22] *** calumb has quit IRC
[19:45:34] <sommerfeld> requires only a minimal amount of manual dexterity (much less than, say, desoldering & soldering...)
[19:47:13] *** EchoBinary has quit IRC
[19:47:14] *** obsethryl has quit IRC
[19:50:06] <zooko> sommerfeld: how do I find such a vendor?
[19:50:33] <hile_> linuxbios?
[19:52:02] *** axxl_ has quit IRC
[19:52:20] <sommerfeld> i've been happy with http://www.biosman.com/
[19:52:48] <sommerfeld> (again, depends on the bios chip being socketed..)
[19:52:51] *** stevel has joined #opensolaris
[19:52:51] *** ChanServ sets mode: +o stevel
[19:53:15] *** duri has joined #opensolaris
[19:53:27] *** comay has joined #opensolaris
[19:53:27] <zooko> Thanks!
[19:57:17] *** cydork has quit IRC
[19:58:34] <Jooles> this is probably a really noob question, but where can I get a program to scan my pci slots for opensolaris? I don't seem to have either lspci or scanpci
[19:58:50] <hile_> prtpicl
[19:59:16] <hile_> isn't some scanning program for x86 is /usr/openwin or /usr/X11 or something?
[19:59:23] <nachox> scanpci
[19:59:29] <hile_> I don't use x86 hardware, so I coudln't tell you
[19:59:45] <Jooles> me neither
[20:00:19] <nachox> doesnt cfgadm also lists some info about pci slots?
[20:00:27] <CIA-26> govinda: 6578427 Further hotplug interrupts will not occur
[20:00:28] *** jcea has joined #opensolaris
[20:00:58] <Jooles> all come up with command not found...  = /.
[20:01:22] *** jcea has left #opensolaris
[20:01:25] <Jooles> Sorry.. Kinda new at the solaris thing. And this isn't even the most fun of my "getting things to work" problems
[20:01:26] *** jcea has joined #opensolaris
[20:02:15] <LuckyLuke> Jooles: prtconf?
[20:02:42] <LuckyLuke> add -v to your taste :)
[20:03:16] <Jooles> where's it live?
[20:03:56] <nachox> Jooles: "find /usr/X11 -name scanpci" what does that return?
[20:03:58] <LuckyLuke> /usr/sbin/prtconf on my SXCE
[20:04:30] <WickedWicky> you can use /usr/X11/bin/scanpci
[20:04:30] <LuckyLuke> yeah, there's also /usr/X11/bin/scanpci
[20:04:34] <Jooles> scanpci isn't there.. I've already looked
[20:04:47] <LuckyLuke> Jooles: mind that it's not in PATH
[20:05:28] <Jooles> thanks. prtconf works  = ). Now to work out why I can't compile ndiswrapper... (masochist, me?)
[20:05:43] <alanc> scanpci should be there if you have Xorg installed
[20:05:56] <LuckyLuke> wasn't ndiswrapper a linux thing?
[20:05:57] <alanc> (it's just a standalone version of Xorg's pci bus scanner)
[20:06:04] <Jooles> I do, but it's not
[20:06:12] <alanc> weird
[20:06:14] <Jooles> yeah, but there's an opensolaris version of ndis.
[20:06:34] <LuckyLuke> ah, cool, didn't know
[20:06:52] <Pietro_S_> LuckyLuke: ndiswrapper is working very well in sxce, see laptom comunity
[20:06:58] <alanc> the opensolaris ndis is a port of BSD's project evil ndis if I recall correctly
[20:07:03] <Jooles> yup
[20:07:33] <Jooles> and it works fine for them because it compiles.. for me I get sytax errors from, as far as I can tell, problem free code
[20:13:53] <Pietro_S_> alanc: yes, and that's maybe why it's working better than linux one on my laptop ;-)
[20:14:48] <Pietro_S_> Jooles: are you in right directory for your kernel type?
[20:16:38] <Jooles> you mean 32/64bit?
[20:17:57] *** comay has quit IRC
[20:18:11] <Pietro_S_> yes
[20:18:26] *** stevel has quit IRC
[20:18:40] <Jooles> I am, though I've tried the other too and get the exact same error
[20:19:04] <Plouj> what do you think will perform faster, a raid-z array of 3 160GB disks, or a striped array of 160GB disks mirrored by a 480GB disk?
[20:19:19] <Pietro_S_> what wifi driver (chip manufacter/ some id) are you using?
[20:20:03] <Jooles> *cough* broadcom *cough*
[20:20:45] <Pietro_S_> I use broadcom especially bcm4318 chip without any problems
[20:21:07] *** nachox has left #opensolaris
[20:21:43] <Jooles> It's not even getting anywhere where the chip would be important yet... this is one of the first header files it tries to compile. Seems to think "struct" is a syntax error
[20:22:27] *** Gman has quit IRC
[20:25:27] <Jooles> actually.. wrong line..
[20:25:44] <Pietro_S_> do you have /usr/sfw/bin in PATH?
[20:26:27] <Jooles> the line it doesn't like is "extern int pe_get_dos_header(vm_offset_t, image_dos_header *);" My guess is missing headers or something...
[20:26:49] <Jooles> yes I do
[20:28:34] <Jooles> (the error I get, well the first in a long list anyway, is "syntax error before image_dos_header"
[20:28:44] <Pietro_S_>  /usr/X11/bin/scanpci | grep broadcom (or bcm) should gives you your chip id, did you copied (.sys, .inf and renamed them?)
[20:28:53] <Jooles> I don't have scanpci
[20:28:59] <Jooles> and yes I did
[20:29:11] <Jooles> it's not even got to the bit where it needs those yet
[20:29:43] *** matsutsu has quit IRC
[20:29:52] <Pietro_S_> it's needed, cause ndis.inf is parsed as first thing
[20:30:20] <Jooles> well fair enough then. Even so, I do have the files and they are renamed
[20:30:38] <Plouj> how much of my root needs to be on a mirrored pool for ZFS boot?
[20:31:38] <Pietro_S_> so you have ndis.inf/sys in same dir from where you run make ndiscvt, right?
[20:31:44] <Jooles> yup
[20:32:33] <Pietro_S_> what drivers are you using? from cd or internet?
[20:33:36] <Jooles> cd
[20:33:47] <Pietro_S_> what kernel are you using build number + 32bit or 64?
[20:33:54] *** apokayi has joined #opensolaris
[20:34:28] <Pietro_S_> Jooles: it's recomended to use latest/internet drivers some cd drivers were reported as not working
[20:35:03] *** theRealballchalk has left #opensolaris
[20:35:06] <Jooles> 64 bit.. not sure how to check the build number. Unless you mean the opensolaris build in which case it's the latest one - build 68
[20:35:14] <Jooles> ok. I'll try the net then.
[20:35:34] <Jooles> oh, and it's a sparc machine
[20:37:14] <Pietro_S_> oh, I fear that it's impossible to get ndis working under sparc, but rather ask in laptop-discuss
[20:37:34] <Jooles> even if not a laptop?  = P
[20:37:35] <Jooles> ok, will do. Thanks
[20:38:01] <sommerfeld> you'd need to execute the ndis driver with an x86 emulator on sparc.
[20:38:36] <Jooles> surely the kernel handles drivers the same way... ?
[20:39:29] <PerterB> that as may be, it's still going to expect them to be written in its native instruction set
[20:39:32] <richlowe> Basically, but the CPU doesn't handle instructions the same way.
[20:39:33] <Pietro_S_> Jooles: but sys file is for X64 architecture that's *the* problem
[20:40:30] <coffman> why on earth did you bought a broadcom card for your sparc machine?
[20:41:01] <Jooles> just had the card and thought it would be worth a try
[20:41:59] <richlowe> which broadcom?
[20:42:13] <Jooles> 4306
[20:42:20] *** sfire||mouse has joined #opensolaris
[20:44:56] *** Pietro_S_ has quit IRC
[20:46:22] *** Symmetria has quit IRC
[20:48:24] *** RaD|Tz has quit IRC
[20:49:26] *** Yamazaki-kun has joined #opensolaris
[20:50:36] <WickedWicky> lads, is the Silicon Image PCI0680 supported as ATA controller?
[20:54:16] * bubbva is away: lunch
[20:59:27] <axisys> where do I get the pfil source ? i need it for one of my solaris box
[20:59:47] <richlowe> onnv.
[20:59:52] <richlowe> but you may have to look back in time a little.
[21:00:36] <CIA-26> gm209912: 6583537 sip_create_dialog_req() should add correct value to the Contact header
[21:01:09] <Plouj> WickedWicky: did you google it?
[21:01:41] <axisys> richlowe: do not see it here http://opensolaris.org/isearch.jspa?query=pfil&Submit=Search
[21:01:51] <axisys> i need the full source code of pfil
[21:01:54] <axisys> hmm
[21:02:00] <WickedWicky> Plouj: yeppers, and I see it should be supported
[21:02:08] <richlowe> Yeah, it'll be in deleted_files/
[21:02:16] <richlowe> or, if you go back far enough, in usr/src ;)
[21:02:48] <axisys> found it ftp://coombs.anu.edu.au/pub/net/ip-filter/pfil-2.1.6.tar.gz
[21:03:03] <WickedWicky> o wait
[21:03:08] <WickedWicky> lemme try this jumper thingy I read here
[21:04:29] <Stric> um.. isn't 2.1.6 close to stone age?
[21:04:53] <Stric> current is 4.1.24 .. which is close to what's in solaris10 for instance
[21:07:32] *** nergal_ has joined #opensolaris
[21:10:12] <axisys> Stric: that is pfil.. u r thinking of ipfilter
[21:10:25] <axisys> 2.17 is the latest
[21:10:30] <axisys> 2.1.7 that is
[21:12:00] <Stric> I am ;)
[21:12:44] <Abe_Froman> solaris is still 4.1.9
[21:13:00] <Abe_Froman> which is highly annoying, since 4.1.10 has a fix for stateful rules
[21:13:06] *** farak has quit IRC
[21:16:24] <axisys> Abe_Froman: i am trying to compile since there is no new pkg for sol 8 atleast
[21:16:30] <sickness> solaris should have gone with openbsd's pf, really, that way it could have become the ONLY os for firewalls too, in this way, I still need to use openbsd on firewalls :P
[21:16:40] <Abe_Froman> indeed
[21:23:50] <PerterB> solaris firewall users fall into two groups, the corporate ones who use FW-1 and ones for whom ipfilter is sufficient
[21:28:15] *** pizdec_ is now known as pizdec
[21:38:28] <coffman> isnt there a brand bsd ?
[21:39:13] <PerterB> not so far
[21:53:36] *** freakazoid0223 has quit IRC
[21:55:34] *** dclarke has joined #opensolaris
[21:56:35] *** freakazoid0223 has joined #opensolaris
[21:57:17] * dclarke searches for coffee
[21:57:38] * WickedWicky shares
[21:58:57] <dclarke> I'm looking at building GCC 4.2.1 here .. it looks to be a scary process
[21:59:09] <WickedWicky> you need more than coffee
[21:59:18] * dclarke searches for scotch
[21:59:35] <wesolows> Lagavulin is good
[21:59:41] <dclarke> ick
[21:59:44] <dclarke> no thanks
[21:59:45] <wesolows> !!
[21:59:45] <quasi> wesolows: very good
[21:59:51] <wesolows> heretic!
[21:59:54] <dclarke> bring me a Glen
[22:00:02] <dclarke> Morangie
[22:00:05] <WickedWicky> christ
[22:00:09] <quasi> wesolows: although I'm almost partial to ardbeg
[22:00:09] <WickedWicky> why doesnt this work :(
[22:00:19] <dclarke> if it ain't old enough to drink with me .. I ain't drinking it
[22:00:22] <quasi> dclarke: weak stuff
[22:00:29] <CIA-26> hiremath: PSARC 2007/323 IBTF IP Addressing, 6400920 Add IP addressing ability to IBTF interfaces
[22:00:29] <wesolows> those have no flavour
[22:00:30] <dclarke> with ice no less
[22:00:35] <quasi> eek
[22:00:39] <wesolows> the lagavulin is 16 years old, that's quite respectable
[22:00:41] <WickedWicky> my Sil680 is configured as Mass storage device, not raid, I do: drv_update -a -i "pci1095,680" pci-ide
[22:00:47] <WickedWicky> still no disks, sniff
[22:00:59] <PerterB> jesus....... you can't put ice in good scotch you heathen!
[22:01:01] <dclarke> hrmmm .. I've had my share of the lag
[22:01:12] <dclarke> I think delwhiney is another of that ilk
[22:01:17] <quasi> PerterB: but it isn't a good scotch
[22:01:31] <dclarke> delwhiney .. can't spell it but I sure uese to drink it
[22:01:35] <PerterB> it's better than average :)
[22:01:38] <wesolows> quasi: Well, true, but then again if it already has no flavour why would you water it down more?!
[22:02:01] <quasi> wesolows: only if you don't like the taste of it I imagine
[22:02:14] <wesolows> quasi: Yeah...which is understandable enough.
[22:02:27] * bubbva is back (gone 01:08:10)
[22:02:28] <dclarke> https://www.glenmorangie.com/landing/index.php
[22:02:32] *** nergal_ has quit IRC
[22:03:07] * WickedWicky is a big fan of chivas
[22:03:08] <quasi> wesolows: I suppose so for a 'merican - they're too used to having tasteremover added to their food/drink
[22:03:34] <quasi> WickedWicky: that's not even worth using to clean the windows
[22:03:45] <WickedWicky> I use cachaca for that
[22:04:22] <dclarke> speaking of scoth .. anyone want to see what I am testing here ?
[22:04:36] <WickedWicky> sute
[22:04:38] <WickedWicky> sure
[22:04:47] <dclarke> SunOS fossil 5.8 Generic_117350-47 sun4m sparc SUNW,SPARCstation-20
[22:04:54] <dclarke> Installation of <CSWseamonkey> was successful.
[22:05:02] <quasi> dclarke: fitting name ;)
[22:05:05] <WickedWicky> lol
[22:05:21] <dclarke> tis patched up to date too
[22:05:31] <dclarke> and .. I'll do a bootstrap build of GCC 4.2.1 on it
[22:05:37] <dclarke> should take a week
[22:05:57] <WickedWicky> how much coffee did you have ready you said? :P
[22:07:00] <dclarke> clearly enough and I had better stop
[22:07:09] <dclarke> really this is just a test
[22:07:35] <dclarke> once that bootstrap build is done then I can packge it up .. take it to sun4u and run a bootstrap again there
[22:07:39] *** SYS64738 has quit IRC
[22:07:46] <dclarke> run the testsuite .. submit the results
[22:07:46] *** Jooles has left #opensolaris
[22:07:50] <dclarke> the usual fare
[22:08:10] <dclarke> heyyy ... I have a question for you geeky types
[22:09:11] <sstallion> ?
[22:09:20] <dclarke> what result do you get if you take 1.000001 and square it 26 times
[22:10:32] <sstallion> 821.2980401419965396255701284542
[22:10:50] <dclarke> okay .. now square that
[22:10:57] <dclarke> sorry .. I intended 27 times
[22:11:00] <sstallion> k
[22:11:09] <sstallion> want the result?
[22:12:47] <dclarke> yes please
[22:13:00] <dclarke> 674530.470741
[22:13:19] <dclarke> did you use bc ?
[22:13:24] <sstallion> worse :(
[22:13:27] <sstallion> 674530.4707410845593826891780296
[22:13:36] <sstallion> windows calculator does arbitrary precision
[22:15:30] <dclarke> 674530.47074108455938268917802974681284444414341034202598521216209
[22:15:43] <sstallion> via bc?
[22:15:48] <dclarke> hrmmm .. it is one hell of an accuracy test
[22:15:53] <dclarke> via bc .. yes
[22:16:10] <dclarke> most calculators and most machines can not product that to six digits
[22:16:19] <sstallion> shows me... just goes to show that calc doesnt do arb. precision
[22:16:28] <dclarke> with IEEE754 based floating point you lose accurary pretty quick
[22:16:45] <sstallion> 754 is usually left for exercises in the classroom
[22:16:55] <sstallion> at least for comp. applications
[22:17:03] <dclarke> it is also the "standard" for most things numerical
[22:17:16] <dclarke> in any case .. I digress .. I was looking at GCC 4.2.1 here
[22:17:17] <sstallion> yeah, but inaccurate
[22:17:26] <sstallion> oh?
[22:17:33] <delewis> In[9]:= 1.000001^(2*27)
[22:17:38] <delewis> Out[9]= 1.00005
[22:17:42] <dclarke> ah .. spot the geek
[22:17:43] <delewis> In[10]:= 1.00001^(2^27)
[22:17:52] <delewis> Out[10]= 7.89354271334358 10^582
[22:17:57] <delewis> for laughs. :-)
[22:17:58] <dclarke> yes .. try that with bc and bc complains
[22:18:15] <dclarke> scale=99
[22:18:17] <dclarke> l(1.0000001)
[22:18:18] <dclarke> scale factor is too large
[22:18:29] <delewis> yeah, I got a stack overflow with bc trying to do that.
[22:18:32] <dclarke> take the natural log of 1.000001 and I get that
[22:18:33] <delewis> not with mathematica, though. :-)
[22:18:43] <dclarke> well no .. I spotted the format there
[22:18:46] <sstallion> meh, just depends on the egine
[22:18:47] <sstallion> err engine
[22:18:58] <dclarke> I have mathematica here also .. but .. its a Windows License
[22:19:10] <sstallion> true arbitrary precision engines suck up heap like nobody's business
[22:19:14] <dclarke> NMI .. gotta run for a sec
[22:19:19] <sstallion> good luck :0
[22:19:24] *** linux_user400354 has joined #opensolaris
[22:19:49] <Stric> matlab says "ans = Inf" pretty quickly ;)
[22:20:02] <delewis> student version or professional version?
[22:20:11] <delewis> matlab has precision limitations in the student version.
[22:20:48] <Stric> 1.00001^100000  looks suspiciously close to e
[22:21:21] <Stric> seems like I got a student license, but it's the "real" install.. for researchers etc..
[22:21:52] <delewis> interesting enough, the other day I was comparing my $MaxNumber in Mathematica on my Tadpole with 4G of memory to a friend's Sony Vaio which is 32-bit and has 2GB of memory. Surprisingly, his $MaxNumber was larger than mine by an exponential factor.
[22:22:19] <delewis> ($MaxNumber returns the largest number possible on the system)
[22:22:29] <sstallion> interesting
[22:22:34] <sstallion> hrmm... need a new calculator
[22:22:41] <sstallion> TI-89 or TI-84...
[22:22:54] <Stric> hm. I should install the 64bit matlab..
[22:23:12] * delewis has UNIX licenses for both Matlab and Mathematica
[22:23:23] <delewis> AIX, Solaris, HP-UX, IRIX, etc.
[22:25:16] *** zooko has quit IRC
[22:25:33] <Stric> bc seems to be having a hard time with 1.00001^(2^27)
[22:26:08] <delewis> $ bc
[22:26:08] <delewis> scale=20
[22:26:08] <delewis> 1.000001^(2^27)
[22:26:08] <delewis> exp too big
[22:26:08] <delewis> empty stack
[22:26:08] <delewis> save:args
[22:27:24] <Stric> 1.00001^(2^17)  takes about 15 secs here.. 2^27 is 1000x more.. so 5.5h..
[22:27:37] *** aruiz has quit IRC
[22:27:51] <delewis> Mathematica spit it on on my slow 650Mhz UltraSPARC-IIe in less than second :-)
[22:28:16] <delewis> there are several slick ways to calculate large exponents, though.
[22:28:45] <Stric> 1.00001^(2^26)  = 2.8095e+291, takes no time in matlab..
[22:28:52] <dclarke> wow .. you have Mathematica on Sparc
[22:28:55] <dclarke> thats cool
[22:29:01] <delewis> and POWER :-)
[22:29:06] <Stric> I think I have it in my office somewhere too
[22:29:07] <dclarke> that result looks wrong
[22:29:16] <dclarke> very very wrong
[22:29:21] <dclarke> hrmm .. let me think
[22:29:34] <delewis> you
[22:29:35] <delewis> er
[22:29:42] <delewis> you're missing a power of 10.
[22:29:45] <dclarke> (a^2)^2 == a^4 == a^(2^2)
[22:29:48] <delewis> 5 zeroes.
[22:30:17] <dclarke> ((a^2)^2)^2 == a^16 == a^(2^4)
[22:30:24] <dclarke> if a=1.000001
[22:30:31] <dclarke> and we need to square it 26 times
[22:30:46] <dclarke> then a^(2^26) sounds right
[22:31:18] *** Gropi has quit IRC
[22:31:27] *** polk__ has quit IRC
[22:31:28] <Stric> my pda calculator gives me the same result
[22:31:32] <dclarke> but with bc I simply state scale=80 and then a=1.000001 followed by b=a*a
[22:31:39] <Stric> although I'd trust matlab more probably ;)
[22:31:46] <dclarke> and then a=b*b
[22:31:58] <dclarke> the result is correct but .. it can not be done with logarithms
[22:32:23] <dclarke> my old old HP RPN calculator gets the correct result with logarithms but is wrong with simple squares
[22:32:50] <dclarke> its a interesting thing to run with a C compiler and basic IEEE754 math
[22:33:17] <dclarke> it was on my mind after I read this : http://developers.sun.com/sunstudio/articles/fp_errors.html
[22:33:59] <dclarke> oops .. swapping here
[22:34:10] <Stric> hm. bc -l; 1.00001; .^2  repeated 27 times  gives me about 2.8095e343
[22:34:30] <dclarke> well thats just bizarre
[22:34:41] <dclarke> with Solaris 8 on sun4m I get the correct result
[22:35:00] <_mary_kate_> 20:27:58 up 12 days, 15:19, 24 users, load average: 498.28, 511.28, 423.41
[22:35:02] <_mary_kate_> i hate users.  :)
[22:35:13] <dclarke> eek!
[22:35:19] <Stric> err. sorry. missing some decimal dots..
[22:35:22] <dclarke> what machine is that
[22:35:48] <Stric> no, I get 2.8095e291 with bc too.. scale=1000
[22:36:18] <dclarke> ummm .. that's not Solaris you have there mate
[22:36:28] <dclarke> on Solaris you can not go past scale=99
[22:36:36] <dclarke> unless that is GNU bc
[22:36:40] <Stric> it's GNU bc
[22:37:20] <Stric> and with solaris bc I get.. *shock* 2.8095e291
[22:38:13] <Stric> (2^26 that is.. the next step overflows the other calculators)
[22:39:03] <Stric> 7.89e582 for 1.00001^(2^27)
[22:40:42] <dclarke> http://rafb.net/p/Kg8dLB70.html
[22:42:03] <Stric> Ah. I see the problem.
[22:42:13] <Stric> There has been different uses of the original number here..
[22:42:18] <Stric> 22:17 (delewis) In[9]:= 1.000001^(2*27)
[22:42:19] <Stric> 22:17 (delewis) In[10]:= 1.00001^(2^27)
[22:42:25] <Stric> note the difference.
[22:42:28] <dclarke> thats not right
[22:42:31] <dclarke> ahhh
[22:42:42] <Stric> so 1.00001 != 1.000001
[22:42:47] <dclarke> yes .. its a test that fills the basic decimal regiater of a seven digit calculator
[22:42:54] <delewis> woops, typo. :-)
[22:42:56] <dclarke> its a 1 followed by six zeros
[22:43:12] <delewis> why couldn't you just express it in exponential notation, dclarke? :-)
[22:43:14] <Stric> ok, six seroes..
[22:43:16] <delewis> I'm awful with zeroes.
[22:43:17] <Stric> zeroes..
[22:43:19] <dclarke> this test was in a Scientific Ameri3can article back in the 80's
[22:43:46] <Stric> so the same number then.
[22:43:48] <dclarke> well .. in EXP notation I could have said 1 + 1x10^-7
[22:46:07] <dclarke> even on sun4m old Solaris 8 I get 674530.4707410845593826891780297468128444441434103420317423773278
[22:46:15] <Stric> that sounds about right
[22:46:18] <dclarke> but try that with C
[22:46:27] <dclarke> and its real real tough to get the correct result
[22:46:30] <Stric> but with the other uses of the original number, it gets tricky ;)
[22:47:00] <dclarke> yes .. you would need significant error and random screw ups to arrive at the correct result
[22:47:35] <Stric> the trouble was that it was >500 digits.. seems like various calculators says "inf" about that..
[22:48:04] <dclarke> you mean 500 digits to the left of the decimal point ?
[22:48:07] <dclarke> that would be large
[22:48:14] <Stric> yes. if you start with 1.00001
[22:48:25] <dclarke> hrmm .. the butterfly effect
[22:48:37] <dclarke> a slight change there and a huge result
[22:49:00] <Stric> yep.. and all I did was to remove some "insignificant" zeroes ;)
[22:49:31] <dclarke> imagine the nerve of them getting in the way eh?
[22:49:58] <dclarke> let me set my CFLAGS to CFLAGS=-xstrconst -xildoff -xarch=generic -fsimple=0 -ftrap=%all -H -Kpic -xlibmieee -xlibmil -xtime -Xa -xtransition
[22:50:05] *** sstallion has quit IRC
[22:50:08] <dclarke> and then comepile some simple C and see what I get here
[22:50:37] <dclarke> $ cc -V
[22:50:38] <dclarke> cc: Sun C 5.5 Patch 112760-18 2005/06/14
[22:50:43] <dclarke> Studio 8 there
[22:50:56] <dclarke> sorry .. it was Sun ONE Studio 8
[22:52:10] <WickedWicky> talk about going stone age.. I am gonna need a floppy drive
[22:52:18] <dclarke> it works
[22:52:21] <dclarke> real real well
[22:52:32] <dclarke> as for Stone age ?
[22:52:51] <dclarke> I have a HP-28S scientific calculator here that predates Sun
[22:52:55] *** alanc_away_ has quit IRC
[22:52:56] *** seanmcg has quit IRC
[22:52:56] *** noyb has quit IRC
[22:52:57] <dclarke> and it works real well
[22:53:16] *** seanmcg has joined #opensolaris
[22:53:18] *** alanc_away_ has joined #opensolaris
[22:53:20] *** noyb has joined #opensolaris
[22:53:43] <dclarke> Sir Issac Newton was well known to perform calculations to 13 digits with paper and pencil
[22:54:36] <WickedWicky> lol
[22:55:20] <WickedWicky> well the thing is, that I see a lot of posts saying how you need to put the sil680 in "IDE mode" by pulling JP1 on the card, which is done already but the card has a RAID bios, so i guess I need to flash the card
[22:55:32] <WickedWicky> and to flash it, I need a floppy
[22:55:34] <WickedWicky> \o/
[22:56:55] *** seanmcg has quit IRC
[22:57:15] *** seanmcg has joined #opensolaris
[22:57:46] *** perlmonk has quit IRC
[22:59:01] <dclarke> on the phone
[22:59:14] <tomww> dclarke: not that old, but still old: HP 48G
[23:00:16] <_mary_kate_> i keep getting this error in syslog: iscsi: [ID 286457 kern.notice] NOTICE: iscsi connection(14) unable to connect to target <iqn> (errno:145) .. but my iscsi lun is working fine.  what does it mean?
[23:00:52] *** Arnald has joined #opensolaris
[23:02:30] *** zooko has joined #opensolaris
[23:04:59] *** zooko has quit IRC
[23:05:47] *** dunc has joined #opensolaris
[23:18:03] <trygvis> what is the _easiest_ way to get something (FMD perhaps) to send an email when a service fails or otherwise changes state?
[23:19:21] *** takahide has quit IRC
[23:20:19] *** linux_user400354 has quit IRC
[23:24:18] *** derchris has quit IRC
[23:24:26] *** derchris has joined #opensolaris
[23:26:32] *** Chihan has quit IRC
[23:32:32] *** nostoi has joined #opensolaris
[23:32:44] *** JoRGiToFF has joined #opensolaris
[23:33:29] *** JoRGiToFF has left #opensolaris
[23:37:38] <dclarke> off the phone
[23:37:45] <dclarke> Arithmetic Exception(coredump)
[23:37:48] <dclarke> oops
[23:39:56] <dclarke> cc: Warning: illegal option -xarch=sparcv7
[23:40:05] <dclarke> hrmm .. whats the option for old type 1 sparc ?
[23:40:08] <dclarke> generic ?
[23:40:12] <dclarke> sparcv8 ?
[23:40:20] <dclarke> sparcv8plusa .. not likely
[23:40:23] *** tsoome has joined #opensolaris
[23:40:33] <alanc> sparcv8 is sun4m and later
[23:41:14] *** tsoome1 has quit IRC
[23:41:16] <dclarke> isainfo -v does not tell me much
[23:41:35] <dclarke> $ psrinfo -pv
[23:41:36] <dclarke> psrinfo: Physical processor view not supported
[23:41:48] <dclarke> hrmm .. perhaps generic is safe here
[23:42:11] <richlowe> generic, I think.
[23:42:19] <richlowe> apparently studio10 defaults to v8plus
[23:42:23] *** gdamore has joined #opensolaris
[23:42:24] <richlowe> 11, rather.
[23:42:35] <dclarke> studio 10 has to be told sparcv7
[23:42:36] <alanc> not sure Sun compilers still support pre-v8 CPU's
[23:42:50] <dclarke> they all do .. except
[23:42:53] <dclarke> studio 12
[23:42:59] <dclarke> I have not tested it however
[23:43:00] <richlowe> Studio 11 cc(1) mentions v7 being there, but obsolete.
[23:43:10] <dclarke> but .. you can not install it on Solaris 8 so .. doesn't really matter
[23:43:44] <wesolows> Since Solaris no longer supports v7 anyway, what good is it?
[23:43:58] <dclarke> exactly
[23:44:04] <wesolows> In fact I'd be fine with killing v8 support too for the same reason
[23:44:07] <alanc> even Solaris 8 only ran on v8 and later CPU's I thought
[23:44:14] <wesolows> Even 32-bit binaries should now use V9 instructions.
[23:44:16] <dclarke> please .. yes .. let's do that
[23:44:27] <dclarke> SunOS fossil 5.8 Generic_117350-47 sun4m sparc SUNW,SPARCstation-20
[23:44:29] <wesolows> so-called v8plus
[23:44:40] <dclarke> that is my last test box for sun4m and I think its sparcv8
[23:44:48] <dclarke> v8plusa
[23:44:49] <wesolows> I wonder if Roger ever finished the libc work needed for that.
[23:44:50] <dclarke> I think
[23:45:01] *** Arnald has quit IRC
[23:45:03] <alanc> v8plusa is sun4u
[23:45:10] <wesolows> no a
[23:45:14] <wesolows> a is ultrasparc only
[23:45:24] <wesolows> we still, I think, support non-ultrasparc CPUs
[23:45:27] <dclarke> oh .. its related to VIS isn't it ?
[23:45:52] <dclarke> no non .. Solaris 8 patches are still coming out but the sun4m gear was EOL'ed some time ago
[23:45:56] <alanc> neither Fujitsu nor niagara CPU'
[23:45:59] <alanc> s have VIS
[23:46:19] <dclarke> niagara CPU's barely have floating point let alone VIS extensions
[23:46:30] <alanc> though I think they both trap and emulate in software, it's just slower
[23:48:08] <dclarke> cc: Warning: illegal option -xarch=sparcv8
[23:48:12] <dclarke> geez ..
[23:48:53] <dclarke> Arithmetic Exception(coredump)
[23:49:00] <dclarke> this is not going well at all here
[23:51:00] <dclarke> ah .. thank you
[23:51:09] *** sfire||mouse has quit IRC
[23:51:51] *** eboutilier has joined #opensolaris
[23:54:30] <dclarke> it never ceases to amaze me how I have to look up the various printf format chars .. year after year
[23:55:36] *** hohum-_ has joined #OpenSolaris
[23:56:53] *** tombhadAC has joined #opensolaris
[23:58:59] *** Symmetria has joined #opensolaris





top