April 3, 2010 
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | >
April 3, 2010 [00:00:13] *** clint- has joined #opensolaris[00:00:14] <victori> ipkg is on s10 now?[00:00:16] <tsoome> but osol got pkg image-update, whish is the same, and beadm[00:00:17] <bda> No.[00:00:21] <TomJ> what does zbe mean? as in: rpool/zone/db-mysql1/ROOT/zbe 589M 440G 589M legacy[00:00:29] <bda> ZFS boot environment.[00:00:32] <TomJ> ah thanks[00:00:35] <clint-> so heh, maybe I should start working with openBSD ?[00:00:43] <clint-> maybe then I will master everything. ; ([00:00:49] <victori> /cant wait for sol11[00:01:13] <victori> would be a fine system[00:01:28] <TomJ> so I set up some zones, and put I 'set defrouter=10.32.33.65' in their config. somehow, it seems that OS added 10.32.33.65 as the default route for the global zone too - is this expected? or a known bug?[00:01:31] <clint-> pretty sad, a card, geforce FX 5200, that uses a nvidia 173xx really.. probes up in vgatext[00:01:56] <eklof> oh, 5200, that was hot... in 1990 :)[00:02:13] <clint-> hey everyone?/ theres no support for Sound Blaster Audigy SE ? in osol 2009, the cd I got from Sun, and the card came out in 2006 heh[00:02:15] <TomJ> I was installing zones, then suddenly I got kicked out of the server - when I remote console logged in, I saw that I had a new default route of 10.32.33.65, in addition to my normal internet default route, which is why I got disconnected[00:02:27] <clint-> its frickin 4 years old lmao.. I got support in every nix distro[00:02:46] <clint-> do I need to add like.. PLF repos like : P[00:02:46] <TomJ> clint-: *Solaris is primarily a server OS. not many servers care about Sound Blaster cards. If you need the card and it isn't supported, use some other operating system[00:03:17] <clint-> well, I think its about just swaping out sound cards until 1 works[00:03:25] <TomJ> Yes, that is what I would do[00:03:31] <clint-> I know the realteks are nice, but... erm, someone told me the Audigy SE is working for them the other day[00:03:42] <TomJ> audigy SE is pretty old anyway, nothing special[00:03:55] <clint-> I'm not going to stop working with this OS, just because a fu** sound vendor doesn't write drivers for open source, or unix or linux systems[00:04:03] <clint-> I been working with open source systems since 2004[00:04:09] <clint-> I know[00:04:33] *** sahafeez has joined #opensolaris[00:04:47] <clint-> I put a CMI card in, but I couldn't get past post in my machspeed, although I like Asus boards, I just been busy building everyone else a bloody nice system, then doing 1 for myself,[00:05:24] <Meths> Hmmm, apparently the Audigy/Audigy2 are supported, guess it's just the SE that doesn't work (assuming you haven't just misconfigured something)[00:06:07] *** Norrin has quit IRC[00:06:08] *** loke has quit IRC[00:06:08] *** kukhuvud has quit IRC[00:06:08] *** tehhobbit has quit IRC[00:06:08] *** DerSaidin has quit IRC[00:06:08] *** cocacola has quit IRC[00:06:08] *** x58 has quit IRC[00:06:08] *** gnibbler has quit IRC[00:06:10] *** mui_ has quit IRC[00:06:10] *** prudhvi_ has quit IRC[00:06:10] *** tomww has quit IRC[00:06:10] *** zynox_ has quit IRC[00:06:10] *** boos has quit IRC[00:06:10] *** trygvis has quit IRC[00:06:10] *** Tusk has quit IRC[00:06:10] *** bgran has quit IRC[00:06:10] *** odyi has quit IRC[00:06:10] *** hile__ has quit IRC[00:06:10] *** jbk has quit IRC[00:06:20] <clint-> so your telling me, that its not meant to run opensolaris on a desktop, or notebook, when they state that on there site[00:06:44] <clint-> I havn't miss configured anything[00:07:05] *** cybernd has quit IRC[00:07:31] <clint-> whta have I miss configured Meths[00:07:35] <clint-> what*[00:07:43] <clint-> please inform me[00:07:46] <Meths> I didn't say you had[00:07:53] <reflect> hm[00:07:57] <clint-> I'm ready to jump[00:08:00] <clint-> feel me..[00:08:04] <clint-> ; ([00:08:19] <TomJ> clint-: ok then jump[00:08:20] <Meths> I was saying the SE must be significantly different to the base models if it doesn't work as they are supported[00:08:25] * clint- jumps[00:08:26] *** odyi has joined #opensolaris[00:08:26] *** trygvis has joined #opensolaris[00:08:33] <reflect> solaris isn't, and afaik, has never been geared toward the desktop[00:08:34] <TomJ> clint-: why do you think we care? the card isn't supported. use another or use another operating system[00:08:43] <clint-> is there an alsamixer like for osol?[00:08:50] *** mui has joined #opensolaris[00:08:50] *** tehhobbit has joined #opensolaris[00:08:58] *** gnibbler has joined #opensolaris[00:09:06] *** Tusk has joined #opensolaris[00:09:11] *** tomww has joined #opensolaris[00:09:12] *** hile_ has joined #opensolaris[00:09:14] <clint-> is there a sound control for the console[00:09:19] <clint-> if so, whats the command please[00:09:22] <trygvis> reflect: solaris was initally written for the desktop[00:09:22] *** cocacola has joined #opensolaris[00:09:57] <echobinary> I like solaris for ZFS[00:10:29] <clint-> is there a sound command to get sound properties in the terminal[00:10:30] *** Tusk has quit IRC[00:10:34] <echobinary> although I recently learned a valuable lesson about raidz2's and physical redundancy isn't necc the only thing that can destroy an array :([00:10:34] <clint-> like alsamixer[00:10:38] * echobinary lost 11TB of data :([00:10:40] <clint-> for linux[00:10:41] *** Tusk has joined #opensolaris[00:10:45] <bda> echobinary: How?[00:10:56] <echobinary> pool metadata got corrupted[00:11:06] <bda> Did you talk to Viktor on zfs-discuss?[00:11:09] <reflect> trygvis: I doubt that.. but, I'm willing to admit when I'm wrong. do you have something to back that up? for the past 13 years, solaris has been more "server" than desktop, imho[00:11:11] <echobinary> no[00:11:11] <bda> import -F didn't help?[00:11:14] <echobinary> nope[00:11:17] <bda> :([00:11:17] <echobinary> tried it[00:11:19] <echobinary> yeah[00:11:20] <alanc> reflect: you mean except when Sun was a workstatin company and the desktop was it's primary product? or when it was selling Java Desktop System based on Solaris as a desktop product?[00:11:26] <echobinary> said no pools were avail to import[00:11:32] <echobinary> it was fubared[00:11:57] *** boos has joined #opensolaris[00:11:57] *** Norrin has joined #opensolaris[00:11:57] *** loke has joined #opensolaris[00:11:57] *** kukhuvud has joined #opensolaris[00:11:57] *** DerSaidin has joined #opensolaris[00:11:57] *** x58 has joined #opensolaris[00:11:57] *** prudhvi_ has joined #opensolaris[00:11:57] *** zynox_ has joined #opensolaris[00:11:57] *** bgran has joined #opensolaris[00:11:57] *** jbk has joined #opensolaris[00:12:00] <reflect> alanc: is this time within the 13 years I stated?[00:12:00] <bda> echobinary: You probably could have used some zdb magic to fix it, but it would have been non-trivial.[00:12:01] <bda> If you still have the disks, try talking to Viktor.[00:12:02] <clint-> I need to go back with supporting PC-BSD[00:12:16] <bda> How did your metadata get corrupted?[00:12:19] <clint-> later everyone,[00:12:22] <echobinary> zdb?[00:12:26] <alanc> reflect: Java Desktop System was 2005[00:12:41] <echobinary> I still have the disks - but I recreated the array as an ampty volume[00:12:44] <jdoe> echobinary: uh, how'd it get corrupted?[00:12:44] <bda> ah.[00:12:52] <bda> Well, nevermind. :)[00:12:58] <clint-> later everyone[00:12:58] <alanc> there's been lots of server work too, but the two don't have to be mutually exclusive[00:13:15] <echobinary> jdoe: a perfect storm of 1 bad disk and some corruption on 3 other disks, and then a power outage whilst trying to scrub[00:13:15] <alanc> there was even the Toshiba laptops sold with OpenSolaris preinstalled last year[00:13:24] <echobinary> upon reboot everything was fragged[00:13:30] *** clint- has left #opensolaris[00:13:31] <jdoe> echobinary: ... yeah, uh, okay that's not so surprising then ;)[00:13:35] <echobinary> indeed not[00:13:43] <reflect> alanc: which was a deviation, and it failed. I'm not saying it has to be mutually exclusive, I'm just saying that the primary work has been for servers and server features.. not for desktop users and the ease of use[00:13:44] <bda> Heh.[00:14:01] <echobinary> had the scrub been allowed to complete (I think there was also some resilvering going on) it woulda been fine.. but alas.. alack[00:14:09] <echobinary> is there a way to backup pool metadata?[00:14:40] <bda> Pool metadata should already be redundant on disk.[00:14:49] <bda> Usually that only happens when metadata gets corrupted in memory.[00:15:01] <bda> (and is then happily written to disk)[00:15:09] <bda> Are you using ECC RAM?[00:15:27] <echobinary> I am not sure[00:15:30] <bda> mm.[00:15:30] <echobinary> been a while[00:15:39] <echobinary> is there a command to check that sort of thing?[00:15:56] <bda> smbios, maybe.[00:16:17] *** boos has quit IRC[00:16:17] *** Norrin has quit IRC[00:16:18] *** loke has quit IRC[00:16:18] *** kukhuvud has quit IRC[00:16:18] *** DerSaidin has quit IRC[00:16:18] *** x58 has quit IRC[00:16:18] *** prudhvi_ has quit IRC[00:16:18] *** zynox_ has quit IRC[00:16:18] *** bgran has quit IRC[00:16:18] *** jbk has quit IRC[00:16:22] <alanc> reflect: I'll admit more work has gone into server features, but that doesn't mean none has gone into desktop[00:16:26] <bda> Should be under the system board/motherboard section.[00:16:40] <bda> YYMV though.[00:17:07] <alanc> at least, I think I did some work in the last 10 years, but some days I'm not sure[00:17:12] <reflect> alanc: I didn't say that, though :)[00:17:21] <alanc> and I'm just one of many engineers in the desktop group[00:17:51] <reflect> sorry if you took offense, it wasn't meant like that[00:18:57] <eklof> Any new tools for checking HDD temperatures in the post 111 versions?[00:19:25] *** DerSaidin has joined #opensolaris[00:19:26] *** DerSaidin has joined #opensolaris[00:19:36] *** zynox_ has joined #opensolaris[00:19:38] *** loke has joined #opensolaris[00:19:41] *** bgran has joined #opensolaris[00:19:41] *** x58 has joined #opensolaris[00:19:47] <TomJ> I'm installing zones with 'set defrouter=10.32.33.65' and every time I do it, OS adds 10.32.33.65 as the default router in the *global* zone.. which screws everything up. Why is it doing this?[00:20:05] *** Norrin has joined #opensolaris[00:20:28] *** jbk has joined #opensolaris[00:20:38] *** kukhuvud has joined #opensolaris[00:21:46] *** jbit_ is now known as jbit[00:21:55] <system5> I'm so pissed, I keep calling Oracle trying to buy Solaris support and they keep telling me to e-mail service-quote-request at sun dot com and every e-mail message I've sent there in the last 4 days gets bounced back[00:22:07] <system5> I'm e-mailing from my yahoo and gmail accounts, and it's not like I'm a spammer[00:22:12] <alanc> did you tell them it's bouncing?[00:22:45] <alanc> I've been having weird problems with some of my incoming mail since they started forwarding it from sun.com to oracle.com, but haven't dug into it[00:22:47] <system5> yeah, the Oracle sales rep told me it wasn't working on April 29th but that it should be working today and it still bounced, but took so long to bounce that I was already off the phone with the rep by then[00:22:48] <jdoe> yeah, you should send them an email and let them know.[00:22:59] <system5> can't send them an e-mail cause it bounces[00:23:09] <reflect> system5: seems strange.. if you call them, why not give them your email address, and have them email you?[00:23:10] <jdoe> whoosh.[00:23:26] <system5> try sending an e-mail to service-quote-request at sun dot com just as a test e-mail and make the subject line: "I would like to buy Solaris 10 support" and see what happens[00:24:02] <system5> I'm sure there are other people in here besides me who plan on eventually buying Solaris or OpenSolaris support, why not send a test e-mail right now to see what happens[00:25:07] <reflect> is that for other products too?[00:25:30] <system5> uhh, I called in and talked to a bunch of Oracle people about buying Solaris support and they always say:[00:25:42] <system5> send an e-mail to the Sol10 team at service-quote-request at sun dot com[00:26:14] *** boos has joined #opensolaris[00:26:14] *** prudhvi_ has joined #opensolaris[00:26:17] <reflect> I just don't want to send a mail and it doesn't bounce, you know ;)[00:26:22] <ant> I'm told that Oracle has dropped OpenSolaris support[00:26:24] <system5> and my business partner called in on March 29th and March 30th and got the same e-mail address and his e-mails bounced too[00:26:26] <ant> commercial support[00:26:40] <system5> well, I still have an active OpenSolaris support contract[00:27:06] *** moepmoep has quit IRC[00:27:17] <eklof> TomJ: It seems like that's what it's supposed to to, have you read this? http://blogs.sun.com/stw/entry/using_zonecfg_defrouter_with_shared[00:27:29] <system5> maybe I should open up a trouble ticket requesting a reboot on the brmea-mail-1.sun.com mail server since that is the server that is bouncing the e-mails[00:27:42] <reflect> ok, my mail hasn't bounced so far[00:27:50] <ant> I was called to do some contracting work with a company I've not worked with before and they said they were very hesitant to use OpenSolaris because they had tried to setup a new support contract and were told no[00:27:54] <reflect> granted, it's just been a minute..[00:27:55] <system5> might be a DNS screwup? MX record is still pointing to an old Sun server when it should be pointing to a new one[00:28:12] *** boos has quit IRC[00:28:12] *** prudhvi_ has quit IRC[00:28:13] <system5> it takes about 15 minutes for the e-mail to bounce[00:30:21] <system5> my most recent bounce was around 1:10pm PST today, it looked like this:[00:30:37] <system5> Your message was not delivered to the following recipients:[00:30:38] <system5> service-quote-request at sun dot com: 550 5.7.1 <service-quote-request at sun dot com>... Access denied[00:30:56] <system5> that was from yahoo, an earlier bounce from gmail looked like this:[00:31:18] <system5> Your message was not delivered to the following recipients:[00:31:18] <system5> Service-Quote-Request at sun dot com: 550 5.7.1 <Service-Quote-Request at sun dot com>... Access denied[00:31:21] <system5> Original-Recipient: rfc822;Service-Quote-Request at sun dot com[00:31:23] <system5> Final-Recipient: rfc822;Service-Quote-Request at sun dot com[00:31:26] <system5> Action: failed[00:31:28] <system5> Status: 5.1.1[00:31:31] <system5> Remote-MTA: dns;brmea-mail-1.sun.com[00:31:41] <system5> sorry for pasting, but it would be too confusing to try to put it all in one line[00:31:49] <reflect> well, I'm sending from a "real" address[00:31:49] <alanc> weird, it shows as a valid alias in the internal LDAP db[00:32:04] *** Javache has quit IRC[00:32:14] <system5> I think it might be an internal only e-mail address and the oracle sales reps are confused and giving it to potential customers?[00:32:15] <reflect> I guess we'll know in about 10 minutes :)[00:32:30] <system5> I'll try sending again from yahoo[00:32:39] <reflect> wait 10 mins[00:32:45] <trygvis> system5: use something like pastie.org and just paste the url. very nice, yay :)[00:32:48] *** moepmoep has joined #opensolaris[00:32:56] *** prudhvi has joined #opensolaris[00:33:30] <TomJ> eklof: thanks, reading it now. in truth, I haven't research defrouter much ; I assumed it meant that the zone had a different default router to the global, hence that the default route wouldn't be added to the global. I may be mistaken[00:33:36] <system5> I'm also amazed that there aren't any Oracle sales reps trolling the OpenSolaris.org discuss forums / mailing lists, since I know that IBM people troll those lists[00:33:38] *** Animal-X has quit IRC[00:34:04] <alanc> system5: oh, yes, the mailing list manager shows it's marked to reject mail from the internet[00:34:11] <system5> like there was this one guy on OpenSolaris discuss that was helping noobs fix there stuff and if you trace his e-mail address / IP it goes to IBM's system Z mainframe department[00:34:12] <reflect> they're new[00:34:30] *** javashin has joined #opensolaris[00:34:41] <system5> so if IBM sales droids can troll opensolaris.org, then Oracle sales droids should do the same to keep the heat on ;-)[00:35:34] *** skullone has joined #opensolaris[00:37:08] *** boos has joined #opensolaris[00:37:10] <alanc> system5: if you forward the mail you tried sending to alanc at sun dot com, I can forward it to them from an internal address[00:37:14] <echobinary> say - is OpenSolaris going to be affected at all by the sale of Sun?[00:37:25] <system5> sweet[00:37:29] <system5> will do[00:37:35] <eklof> echobinary: no one really knows at this stage.[00:37:37] <alanc> yes, it will now say "Copyright Oracle" on OpenSolaris boot, instead of "Copyright Sun"[00:38:32] <eklof> Rumours say they will probably not spill all of the solaris features down to opensolaris, and try to separate them, but that remains to be seen, just speculations.[00:38:47] <echobinary> isn't OpenSolaris "open source" as it were?[00:38:48] <jdoe> also, by installing it you agree to give Larry your firstborn. (His yacht is powered by SOULS)[00:39:08] <system5> well there was a lot of SOlaris 10 code that was encumbered, but right now OpenSolaris has all the new features, not Solaris 10[00:39:13] <eklof> echobinary: I think so yes.[00:39:37] <reflect> many parts of opensolaris is opensource, yes[00:39:39] <system5> Solaris 10 is better for backwards compatibility with Solaris 7,8,9 because of Jumpstart, Flash Archive, SysV package based, UNIX System V tools first in the $PATH, etc.[00:39:47] <reflect> afaik, there are some parts which aren't[00:39:50] *** jaywalk has quit IRC[00:40:14] <sstallion_work> reflect: the distribution does not contain encumbered packages.[00:40:30] <sstallion_work> (which is why its marked as "freely distributable")[00:40:49] <sstallion_work> ON still maintains a closed list though.[00:41:29] <alanc> the distribution contains packages for which source is not provided, but the right to redistribute is[00:41:47] <alanc> http://hub.opensolaris.org/bin/view/Main/no_source has a list of most of the binary blobs[00:42:03] <victori> heh jds[00:42:10] *** twilling has joined #opensolaris[00:42:19] <basic`> aw, there's no source provided for Flash?[00:42:25] <victori> heh[00:42:47] <sstallion_work> basic`: flash is not in the distribution; its in the extra repo[00:42:53] <sstallion_work> alanc: d'oh - I had forgotten about those.[00:42:54] <reflect> system5: no bounce so far, and it's been 15 minutes now I think[00:43:17] <reflect> hah[00:43:22] <reflect> as I typed;[00:43:28] <basic`> sstallion_work: heh, it's on the page[00:43:31] <alanc> the no_source list doesn't distinguish between main repo, extra repo, and things we've just tossed since Solaris 10 (like Xsun)[00:43:58] *** bubbva has quit IRC[00:44:07] <g4lt-mordant> alanc, I'm speculating wildly here, but I'm guessing that oracle is going to start up a forced-emancipation project for the binary blobs, given that they already said "there will be things in the pay version of solX that won't be in the free opensolaris"[00:44:08] <sstallion_work> basic`: no_source != distributed in Indiana[00:44:18] <alanc> it was based on what was in SXCE when OpenSolaris launched[00:44:41] <eklof> how much is a solaris 10 licens anyway?[00:44:43] <sstallion_work> alanc: *nod* the format is prettier though ;)[00:45:00] <alanc> g4lt-mordant: umm, kinda hard to ship a system without libc or critical hardware drivers, so I'm going to guess that would be insane[00:45:07] <sstallion_work> g4lt-mordant: we already have an emancipation project which is re-writing the items on the page alanc just posted.[00:45:21] <reflect> http://pastebin.com/PxqWCviH[00:45:36] <sstallion_work> as it is, jbk and I are rewriting: spwr, ce, pcn, and others[00:45:46] <g4lt-mordant> sstallion, yeah, webmink pushed it pretty hard, I'm just thinking it will be an "emancipate or die" level of involvement[00:45:51] <sstallion_work> which reminds me, I need to finish spwr this weekend[00:45:57] <eklof> Any news on the zfs encryption btw, seems a little bit quiet on Moffets twitter on that cause :([00:46:02] *** sergiusens has quit IRC[00:46:03] <reflect> system5 is correct, that address does not work, even when used from a "real" account, outside of gmail and yahoo[00:46:04] <alanc> g4lt-mordant: my understanding is that it's much the same as was always the case with things like the Fishworks storage appliances - the server management and dtrace analytics features of those were never opened either[00:46:22] <sstallion_work> gdamore has also just integrated a re-written elxl[00:46:43] <alanc> and the ksh93 folks have a open sed ready to go[00:46:51] <TomJ> how do I unset a defrouter setting? I have tried remove defrouter and set defrouter= and clear defrouter[00:47:59] <reflect> alanc: will you deal with that, http://pastebin.com/PxqWCviH .. make sure someone is on it?[00:48:00] <sstallion_work> TomJ: heh. rm /etc/defaultrouter[00:48:01] <alanc> drivers and basic Unix utilities are not where you differentiate your OS[00:48:08] <TomJ> sstallion_work: no I mean in zonecfg[00:48:23] <sstallion_work> TomJ: remove[00:48:30] <basic`> is there an official page/mail thread/anything official/semi official about the status of osol+oracle?[00:48:45] <TomJ> sstallion_work: zonecfg:db-mysql1:net> remove defrouter[00:48:46] <TomJ> syntax error at '\n'[00:48:58] <sstallion_work> TomJ: you are under-specifying defrouter[00:49:02] <alanc> reflect: I can send mail to the list telling them that someone is giving out their address - do you remember who it was or who you called?[00:49:18] <TomJ> sstallion_work: what do I specify then? zonecfg:db-mysql1:net> remove defrouter=10.32.33.65[00:49:20] <TomJ> syntax error at '='[00:49:31] <reflect> alanc: no, I just confirmed system5's problem from a "real" address[00:49:31] <sstallion_work> TomJ: hang on, let me fire up zonecfg[00:49:34] <alanc> basic`: google should find plenty of references to the IRC chat a month or two ago with the Director of Solaris & OpenSolaris marketing and various other statements from google[00:49:57] <basic`> i keep seeing these 'RIP OpenSolaris' articles and just assume it's bullshit[00:50:03] <TomJ> alanc: was that after the Osol head publically admonished them for saying nothing? (or maybe not hte head, but someone important)[00:50:08] <alanc> oh, sorry, right, I forgot that after my brain had context switched to the other conversations[00:50:11] <basic`> alanc: right, there are some quotes floating around, nothing very solid though[00:50:21] <alanc> TomJ: OpenSolaris has no head[00:50:33] <TomJ> ok, I was referring to: http://www.theregister.co.uk/2010/02/18/opensolaris_under_oracle/ - it was Peter Tribble who said it[00:50:38] *** hajma has joined #opensolaris[00:50:43] <TomJ> that was 6 weeks ago; was the IRC chat after that?[00:50:47] <sstallion_work> TomJ: try: remove defrouter x.x.x.x[00:50:57] <tsoome> basic`: they are very annoying about what they see from opensolaris, so they are mass spreading fud now[00:51:05] <TomJ> sstallion_work: works, thanks[00:51:06] <alanc> TomJ: yes[00:51:11] <TomJ> alanc: ok thanks[00:51:16] <basic`> so it's just a bunch of random quotes floating around the news sites still[00:51:23] *** zmisc has joined #opensolaris[00:51:29] <basic`> tsoome: and yeah, FUD everywhere[00:51:53] <alanc> TomJ: see ptribble's followup after the IRC: http://ptribble.blogspot.com/2010/02/opening-up-some-details-of-opensolaris.html[00:52:01] <TomJ> alanc: thanks[00:52:13] <zmisc> You guys talking about the uncertainty of Solaris after the acquisition by Oracle?[00:52:18] <basic`> most of the "Solaris 10 only 90 day trial" articles have some reference to OpenSolaris being "dead", with no backing[00:52:22] <zmisc> the FUD?[00:52:40] <zmisc> basic`: I don't think it will die, but I can see the concern.[00:52:41] <basic`> but i can't really find any solid information about the future of it either[00:52:46] <eklof> shouldnt oracle with all its resources be able to release 10.03 on time at least?[00:52:50] <zmisc> What would Oracle have to gain with OpenSolaris?[00:53:06] <TomJ> zmisc: OpenSolaris, not Solaris. There wasn't much uncertainty about Solaris I believe[00:53:13] <sstallion_work> zmisc: it helps to understand what OpenSolaris is before asking that question.[00:53:29] <alanc> eklof: ever read or heard of "The Mythical Man Month"? The more resources you add to a late software project, the later it becomes[00:53:36] <TomJ> alanc: good book[00:53:39] <BankerBarney> http://itmanagement.earthweb.com/osrc/article.php/3867771/OpenSolaris-Alive-and-Well-at-Oracle.htm[00:53:48] <zmisc> sstallion: I know what OpenSolaris is,[00:53:49] <basic`> eklof: i believe alanc said yesterday that the .03 is just a reference to where the release was at a certain point in time, .06 came out in july[00:53:49] <alanc> have to pull the existing people off real work to bring the new people up to speed[00:54:00] <eklof> alanc: well you hava a good point :)[00:54:04] <trygvis> alanc: so if oracle are pulling some people off opensolaris it will move faster? :)[00:54:18] *** ewdafa has quit IRC[00:54:19] <alanc> trygvis: unfortunately, it doesn't work that way either[00:54:31] <zmisc> sstallion_work: I just wanted to get up to speed on the topic before I participated in the conversation.[00:54:41] <eklof> maybe 10.04 will have gnome 2.30 then :)[00:54:45] <zmisc> sstallion_work: It also helps if you don't assume ;)[00:54:46] <reflect> so whatever you do, you're screwed.. when it comes to software development?[00:54:48] <reflect> :)[00:54:49] <eklof> or 10.05 or whatever[00:55:12] <BankerBarney> http://www.h-online.com/open/news/item/OpenSolaris-future-assured-by-Oracle-942161.html[00:55:15] <sstallion_work> zmisc: then you should realize that ON *is* Solaris (simplistically) and ON *lives* on the trunk (i.e. the OpenSolaris project).[00:55:37] <snuff-home> i am lookin forward to the next dev release after they finish 10.03 :)[00:55:56] <zmisc> sstallion_work: Understanding what OpenSolaris is, and the structure of OpenSolaris are two completely different things.[00:55:57] <sstallion_work> INOW, additions to ON happen on the tip first, and are back-ported to Solaris 10 if deemed necessary.[00:56:04] *** jaywalk has joined #opensolaris[00:56:06] <sstallion_work> zmisc: no, they aren't.[00:56:12] <CIA-21> John Beck <John.Beck at Sun dot COM>: 6733941 sort -t . dumps core on IP address list[00:56:14] <eklof> yes, at least so one can get rid of thenasty rpool mirror bug in 134 :([00:56:43] <zmisc> My understanding is that OpenSolaris is a community derived Solaris derivative.[00:57:00] <sstallion_work> zmisc: your understanding is flawed (as mentioned earlier).[00:57:08] <basic`> BankerBarney: good link, thank you[00:57:10] <zmisc> sstallion_work: what is it then?[00:57:13] <zmisc> apples?[00:57:20] <BankerBarney> :)[00:57:21] * sstallion_work digs for stevel's post.[00:57:44] <alanc> OpenSolaris is the development branch of Solaris, much like Fedora is to RHEL[00:58:06] <zmisc> I also thought that OpenSolaris is what Solaris may become.[00:58:16] <zmisc> testbed for new technologies[00:58:25] <reflect> so, most things successful in opensolaris, makes it into solaris..[00:59:00] <sstallion_work> reflect: Solaris Next. back-porting to Solaris 10 is expensive.[00:59:00] <zmisc> sstallion_work: I'm not trying to start an argument or anything, I'm just a unix hobbyist.[00:59:33] *** ozux has quit IRC[00:59:33] <zmisc> I don't have any extensive Solaris background or anything, mainly Linux and the BSDs[00:59:35] <eklof> didn't zfs / go into osol and then into solaris 10 the latest versions?[00:59:36] <reflect> sstallion_work: my comment was "rolling".. no versions mentioned[00:59:51] <BankerBarney> http://www.eweek.com/c/a/Linux-and-Open-Source/Oracle-Explains-Unclear-Message-About-OpenSolaris-444787/[01:00:34] <sstallion_work> a bit out of date, but: http://whacked.net/2005/06/21/confused-so-was-i/[01:00:36] *** javashin has quit IRC[01:00:43] <zmisc> sstallion_work: thanks for explaining it in a bit more detail, I must have misinterpreted what you meant by "understanding what solaris is"[01:00:48] <zmisc> opensolaris I mean[01:00:53] *** pizzadliveryboi has joined #opensolaris[01:01:41] <zmisc> OpenSolaris is also planned to be a community project as well?[01:01:49] *** apinski has joined #opensolaris[01:02:03] <sstallion_work> A few years ago, project Indiana assumed the name OpenSolaris which became the "blessed" distribution (this adds a bit more confusion on top of everything else).[01:02:04] <zmisc> Or are contributions pretty much Oracle only?[01:02:23] <basic`> zmisc: like alanc said, are you familiar with how Fedora works with redhat?[01:02:30] <sstallion_work> zmisc: there are plenty of external contributors... many in this channel...[01:02:34] <zmisc> sstallion_work: well it would help their cause if they made it clear?[01:02:57] <sstallion_work> zmisc: IANAM[01:02:59] <zmisc> basic`: yes, it will be used as a testbed for features and whatnot that will make it into Solaris.[01:03:09] <zmisc> IAMAM?[01:03:17] <sstallion_work> I Am Not A Marketeer[01:04:06] <zmisc> sstallion_work: I never said you were, I was referring to Oracle when I said _they_[01:04:17] <basic`> Fedora isn't just a testbed for features though, it's a free bleeding edge distribution.. things worth backporting go back into solaris 10 (like fedora->rhel). that's my understanding at least. There are a large number of redhat employees that work on Fedora, along side community developers[01:04:46] <basic`> bleeding edge vs. stable platform[01:04:50] <zmisc> wouldn't bleeding edge imply that it is somewhat experimental[01:04:54] <zmisc> i.e. testbed[01:05:04] <apinski> so more like -stable vs -unstable for debian :)[01:05:14] <apinski> or sometimes -testing[01:05:15] <zmisc> You're just rebranding what I say in your own words.[01:05:27] *** syamajala has joined #opensolaris[01:05:43] * sstallion_work sighs.[01:05:44] * sstallion_work &[01:06:25] <zmisc> Maybe I don't understand what bleeding edge is, and stable vs. testing.[01:06:36] <zmisc> I guess stable is stable, and stable might be testing if testing is stable?[01:07:38] <snuff-home> stable osol = a release.. which u pay a support contract which gets patches backported..[01:07:58] <zmisc> I know what stable is.[01:08:21] <basic`> bleeding edge osol = snv_134, dev builds, etc?[01:08:31] <zmisc> I'm just saying that my understand of OpenSolaris is to progress Solaris, I'm assuming the whole point isn't to backport features.[01:09:32] <basic`> that changed recently though, i think? 2009.06 used to get updates for free?[01:10:21] <zmisc> Who knows what Oracle is doing, if they had a clear message instead of beating around the bush maybe there'd be more interest in OpenSolaris.[01:10:46] <basic`> zmisc: or maybe they're trying to figure out how interested people are by beating around the bush ;)[01:10:49] <zmisc> I[01:11:01] <zmisc> who knows? hehe[01:11:10] * basic` just waits patiently[01:11:14] <zmisc> We might find out, we might not.. who knows?[01:11:15] <basic`> 134 is stable enough for me :)[01:11:19] <pizzadliveryboi> What language takes the cake in OpenSolaris' (and I guess Solaris') code?. Java or C?[01:11:22] <zmisc> I don't even think oracle knows.[01:11:48] <apinski> pizzadliveryboi: most likely C[01:12:28] <reflect> solaris is ancient.. so definately C[01:12:32] <Meths> Are spec files for packages in /release and /dev available somewhere?[01:13:06] <zmisc> reflect: Solaris is not ancient, it is home to a lot of exciting technologies that aren't found elsewhere.[01:13:29] <zmisc> well at least as far as OpenSolaris is concerned..[01:13:33] <basic`> 1983? :)[01:14:16] <reflect> zmisc: when it comes to operating systems, its roots are very old[01:14:20] <alanc> Meths: only the GNOME/Desktop packages in /release & /dev are built with spec files - those are available from the JDS svn repo[01:14:33] <BankerBarney> http://www.internetnews.com/dev-news/article.php/3867776[01:14:45] <zmisc> reflect: in that regard any unix-like system is[01:14:52] <zmisc> or unix system[01:15:06] <alanc> pizzadliveryboi: far more of the Solaris & OpenSolaris code is C than Java[01:15:27] <zmisc> alanc: is there any specific reason Oracle hasn't really released any information>[01:15:50] <BankerBarney> well we dont want thing exactly "unchanged"...we want snv_138[01:15:51] <alanc> zmisc: hell if I know, I'm an engineer, not an executive or marketing[01:15:54] <reflect> zmisc: no.. not gnu/linux, it's fairly new.. and if you look up the historical tree of unix, iirc solaris is fairly early[01:16:01] *** javashin has joined #opensolaris[01:16:53] <zmisc> alanc: ah ok, I'm not trying to hassle you or anything and I'm sure you guys get sick of people asking[01:17:01] <zmisc> sorry about that[01:17:03] <alanc> basic`: all OpenSolaris releases since 2008.05 have only offered free updates if you move to /dev and get development builds - the stable updates were always only for people paying for support contracts[01:17:08] <zmisc> It's just that people want to know.[01:17:13] <Meths> alanc: Ah, how are the others built then and will this have a bearing on moving packages from contrib to dev/release? (I still haven't come across any docs on this process)[01:17:36] <alanc> zmisc: I know, and yes, we are getting sick of it, and really wish we had answers to give[01:17:36] <zmisc> alanc: It's just a disappointment because this "community" that they speak of is pretty much non existant[01:17:53] <basic`> alanc: thanks for clearing that up for me[01:18:18] <basic`> alanc: someone here had told me otherwise, but i'm too new on the scene to know better[01:18:26] <zmisc> I've seen pretty much no contributions outside of Sun, doesn't seem too open to me.\[01:18:38] <zmisc> Sun/Oracle[01:19:18] <alanc> Meths: every consolidation has it's own build process - most, like ON, X, and SFW, have a build tree with all the software in that they build all at once, and then build packages from the results with pkgmk or pkgsend[01:19:37] <alanc> how stuff moves from /contrib to /dev, I have NFC[01:20:29] <zmisc> alanc: are a majority of patches from the outside rejected, or are there just very few contributions submitted?[01:20:53] *** psychicist has quit IRC[01:21:12] <alanc> zmisc: I have no way of knowing[01:21:20] <Meths> alanc: Okay, thanks. One last one for now, am I right that you mentioned /contrib is officially for /release not to be used with /dev or is that only because of the current quality of contrib and the recent package renaming?[01:21:21] <zmisc> I'd just like to know where this community is, because I don't see one.[01:21:35] <alanc> I work on X, I accept most patches submitted to me[01:21:51] <zmisc> alanc: good to hear, a lot of people do want to help out :D[01:22:17] *** ingenthr has quit IRC[01:22:18] <zmisc> I've thought about helping with documentation...[01:22:19] <alanc> I know ON (the kernel and core utilities) usually have a long backlog, but those who have done a lot of good contributions often find a fast path to get in[01:22:40] *** Mdx4 has quit IRC[01:22:51] <alanc> Meths: I don't think I ever said any such thing - I don't know all that much about /contrib, sorry[01:23:12] * alanc is not the person with all the answers![01:23:26] <BankerBarney> why not?[01:23:29] <Meths> alanc: Ah, okay, my bad. I'll see what I can get out of the mailing lists.[01:23:36] <BankerBarney> j/k? ;P[01:23:36] <Meths> alanc: Thanks.[01:23:58] <alanc> BankerBarney: because they haven't accepted me into the secret cabal yet[01:24:09] <zmisc> alanc: ok, thanks for your help.. thanks for your work on OpenSolaris, I am appreciative nonetheless even though my concern about a lack of community.[01:24:10] <BankerBarney> lolz[01:24:19] <alanc> damn! and know that I've mentioned it, they'll reject me again! There is NO Secret Cabal![01:24:57] * BankerBarney is rofl[01:25:01] <BankerBarney> doh![01:25:40] <BankerBarney> its the weekend folkz...enjoy it instead of worrying bout the future of osol[01:25:48] * BankerBarney is afk[01:25:57] <zmisc> BankerBarney: if OSOL fails my life is over @_@[01:26:10] * alanc wonders where he can hide easter eggs in the OpenSolaris code this weekend[01:26:13] <basic`> BankerBarney: its only 4:30 here, i've got 30 min of work left :([01:26:22] *** syamajala has quit IRC[01:26:29] <basic`> if you can call this work[01:26:46] <Meths> basic`: slacker ;P[01:26:59] <basic`> Meths: i know, i know ;)[01:28:31] *** javashin_ has joined #opensolaris[01:29:46] *** cfs_ has quit IRC[01:30:32] <fleyta> alanc: ksh93 now has new easter eggs again.[01:31:03] <fleyta> ksh93 --easter-egg-mode=animated[01:31:41] <alanc> heh, and I thought shpiano was the ksh93 easter egg[01:31:58] *** javashin has quit IRC[01:33:06] <fleyta> alanc: I have a simple matlabtoksh converter but I wasn't allowed to add it.[01:33:30] <tomww> Meths: I think /contrib is still seen as separate form the OS distro. So if the OS distro needs a package, an approval is needed to get is into the distro. I might happen, that the discussion around the new packages leds to a removal form /contrib[01:34:09] <tomww> Meths: *but* there is not versioning on the /contrib side, so the package may in fact not removed, old releases may still need it.[01:36:27] <Meths> tomww: By versioning you mean there aren't multiple versions of the same package in contrib?[01:36:50] *** richb has quit IRC[01:37:07] <Meths> tomww: Is the approval to get into distro process documented?[01:38:53] *** Aria has joined #opensolaris[01:39:03] <alanc> each consolidation has their own process, so you'd need to pick a consolidation that it would be going into[01:40:15] *** javashin_ has quit IRC[01:41:36] <basic`> Meths: i lied, gonna mega slack and leave early[01:42:01] <basic`> the open source lab will survive without me for 15 minutes... i hope[01:43:21] *** cfs has joined #opensolaris[01:43:52] <richlowe> several people think /contrib being a whole 'nother world is really stupid[01:43:58] <richlowe> it's precisely what left to the CCD being largely useless.[01:44:09] <richlowe> uh "left" or "lead to", but not "left to"[01:44:25] <richlowe> it's also leading to SFW being completely ungainly[01:45:40] <Meths> Is consolidation synonymous with community group?[01:46:17] <alanc> no, consolidation are the source trees that the Solaris source was long ago organized into[01:46:40] <alanc> some of them have matching community groups, some are spread across multiple ones[01:47:22] <Meths> Okay so would the Desktop Community Group likely feed into and use the X11 consolidation?[01:47:30] <lattera> http://wikis.sun.com/display/DTrace/tcp+Provider[01:47:36] <lattera> I really want to learn dtrace[01:47:53] <alanc> the Desktop community group runs the JDS consolidation[01:48:01] <richlowe> X may have its own community[01:48:06] <alanc> the X Window System community group is the X11 consolidation[01:48:19] <alanc> http://hub.opensolaris.org/bin/view/Main/downloads has a list of consolidations under "Building OpenSolaris"[01:48:25] <richlowe> though it'd make way more sense for it to be a project in Desktop[01:48:50] <alanc> richlowe: yeah, we started down the path of merging once, but never finished - I don't remember why[01:49:26] *** nanase has quit IRC[01:49:26] *** hunter has quit IRC[01:49:39] <alanc> the ON consolidation has an ON community group, but many other community groups are involved in ON as well (storage, security, dtrace, zfs, drivers, etc.)[01:49:42] <Meths> alanc: Yeah, I found that, think I just didn't parse it very well, found JDS though now, thanks.[01:50:24] <richlowe> if you're actually looking for X, that's pretty much alanc, as far as visible people go.[01:51:02] <alanc> on IRC - a couple driver people answer mailing list mail (john and niveditha do a lot, some of the others a bit)[01:51:34] <richlowe> alanc: I just like seeing people blame you for stuff.[01:51:42] <richlowe> alanc: when I try it, you always punt onto either John/nv or i18n.[01:51:48] *** Spencer_tt has joined #opensolaris[01:52:22] <alanc> richlowe: two can play that game you know, now that you're singlehandedly to blame for the ON IPS packaging[01:52:43] <alanc> (may not be singlehandedly responsible, just a convenient scapegoat 8-) )[01:53:31] <zmisc> IMO the bureacracy is leading to inefficiancies and a lack of community involvement[01:54:05] *** Spencer_tt has quit IRC[01:55:16] <zmisc> mostly the lack of involvement[01:55:17] <richlowe> alanc: that's all Liane and Mark's fault :)[01:55:27] <tomww> well, not chaning every now an then major components, thats the old Solaris. And thats what a lot of customers wanted because the had only a few fixes to do after upgrading instead of redeploying all applications (I'm simplifying)[01:56:06] <datadigger> bureaucracy is evil, but we can't do without.[01:56:25] <richlowe> It's possible what people think it's like deters them.[01:56:35] <richlowe> but I don't think many of the people who've actually dealt with it, care that much about it.[01:57:16] *** gm152 has quit IRC[01:57:18] <richlowe> It's hard to see making sure your code works and explaining what it does as a burden.[01:57:46] <tomww> if you read the discussions about new approvals, very smart people are discussing there and a lot of good results. so if someone wants to know, he if free to take part in the phone conferences and read the papers (=cases)[01:59:02] <zmisc> datadigger: it's not as much as the bureaucracy as it makes it next to impossible to become involved[01:59:20] <richlowe> It does?[01:59:23] <richlowe> Fix something.[01:59:24] <richlowe> tada.[02:00:02] <zmisc> each has their own opinion.. which is a good thing[02:01:22] <zmisc> not as far as things being fixed, they are fixed, just not as far as community goes[02:01:26] *** swa_work has quit IRC[02:01:38] <Meths> zmisc: I'm not finding any bureaucratic hurdles. At the moment it is simply a clarity and ease of access to information for community involvement on the website. I need to work out if the info isn't there (and make some) or if it's just me.[02:02:09] <zmisc> Meths: so it is getting better as far as community involvement goes in your opinion?[02:02:38] <Meths> I've found certain things on jucr to be poorly documented and some of that is being fixed and some now has new bugs open :)[02:02:38] <zmisc> if I understood the process better I would love to write documentation[02:02:42] <alanc> zmisc: richlowe would be one of those community members who has gotten many fixes in[02:02:55] <datadigger> It's hard to keep development a complex thing as an operating system consistent, Even with bureaucracy the result is so-so. Imagine what would happen without all that coordination,,,[02:03:01] <zmisc> I just don't know the process, and it doesn't seem very clear unless you are on the inside[02:03:34] <zmisc> Docs are always good, I'm going to do some research and try to help with documentation if possible.[02:03:44] <datadigger> I agree, the process is complex.[02:04:07] <zmisc> alanc: is documentation possibly community maintained if a member would like to do it and the contributions are good?[02:04:26] <zmisc> The only reason I'm asking is because it seems you work for Oracle[02:05:40] *** niq has quit IRC[02:05:58] *** gm152 has joined #opensolaris[02:06:14] <alanc> zmisc: I think the documentation community is open to community contributions[02:06:26] *** gm152 is now known as Guest31259[02:06:48] <alanc> and yes, I work at oracle as an engineer on X - that doesn't give me insight to what every single one of the other 100,000+ Oracle employees is doing though[02:06:54] <xenol> let's assume i have data zpool. if i create datasets on data zpool and i export zpool via system nfs, how can i make it that i can write into zfs partitions?[02:07:02] <Meths> In terms of community I'm not sure I have been here long enough to determine it's size but it doesn't seem to be huge and my fear is it will always remain small as long as there are no freely available security updates. If the community doesn't find some way to sort that out with Oracle then the choice will be use /dev and osol is (for want of a better term) a "hackers" OS or pick another operating system for free stability[02:07:16] <Meths> (sorry if that fills some screens)[02:07:55] *** comay has quit IRC[02:08:01] * alanc is tired of explaining that if the community wanted security updates for the stable branch to be free, they could set up a community maintained repo on a site like genunix to provide them[02:08:48] <Meths> alanc: And that is where I would like to go, but to get there I need to learn the processes that govern the community. I'm trying one step at a time.[02:10:23] *** Guest31259 has quit IRC[02:10:48] *** gm151 has joined #opensolaris[02:11:11] *** alanc changes topic to "ON 136, IPS 134 ( Release notes: http://bit.ly/d4N8N2 & x86 ISO torrent: http://bit.ly/9F12It ) | Docs: (READ THEM!) http://cli.gs/87t7SH | Sysadm Guide: http://cli.gs/54pBU9 | Linux->Solaris migrants: http://bhami.com/rosetta.html | Clipboard: http://opensolaris.pastebin.ca/ | 134 IS NOT 2010.03! | We know 2010.03 is late. We don't know when it will be released."[02:13:03] <LadySerena> does the gnome-gtk-smooth engine build on OSOL?[02:16:07] <LadySerena> ooooo, it comes with a spec file[02:17:58] *** davenz has joined #opensolaris[02:18:20] *** kpt-pejsek has quit IRC[02:18:37] *** tosh has quit IRC[02:19:16] <x58> How about BeleniX or ShilliX don't those provide security updates?[02:20:31] *** duri has joined #opensolaris[02:24:30] <hajma> belenix hasn't moved in the past 3 months or so[02:28:19] <alanc> does schillix have updates?[02:28:29] <x58> Just checked, doesn't look like it.[02:29:47] *** fleyta has quit IRC[02:30:09] <x58> Seems it has indeed been a while since the last release of BeleniX or update for that matter. Shame really.[02:31:01] <x58> There is OSUNIX as well, but their web site seems to be down.[02:32:37] <alanc> nexenta is fairly active and has updates often[02:32:55] <alanc> they have their own corporate backer though[02:34:39] <x58> updates are going to cost money.[02:34:56] <x58> Well screw me sideways, I fell for a late April Fools.[02:40:14] <alanc> system5 & reflect: got confirmation that the team behind that alias already knows of the problem and is already working through channels to get it fixed[02:44:41] <eklof> Anyone knows if there is some nexentastor-like project that is totally free and open with no storage-limits?[02:45:46] *** Odin- has quit IRC[02:46:51] <lattera> I guess this is off-topic, but I wonder why nexenta has storage limits[02:47:08] <eklof> not nexenta, but nexentastor[02:47:16] <lattera> well, right[02:47:20] <eklof> i guess they want to earn money :)[02:54:56] *** jaywalk has quit IRC[02:55:32] *** jaywalk has joined #opensolaris[02:55:32] <LaidBack_01> anyone here using nginx with osol?[02:55:57] *** TommyTheKid has quit IRC[02:56:08] <LaidBack_01> getting this on enabling the service for nginx (in logs) ld.so.1: nginx: fatal: hardware capability unsupported: 0x100 [ AMD_3DNow ][02:56:41] <LaidBack_01> I'm running this on an Intel. Does that mean that nginx was compiled with AMD hardware in mind?[02:57:12] <tsoome> yes[02:58:00] *** chendy has joined #opensolaris[02:59:06] <LaidBack_01> file for nginx gives nginx: ELF 32-bit LSB executable 80386 Version 1 [SSE2 SSE FXSR AMD_3DNow CMOV FPU], dynamically linked, not stripped[02:59:24] *** comay has joined #opensolaris[02:59:42] *** comay has quit IRC[03:02:50] <alanc> looks like that binary was compiled for cpu's with AMD's ISA extensions[03:03:22] <alanc> I don't remember if Intel ever picked those up like they did AMD64 or not[03:05:28] <alanc> looks like they didn't, at least on the Nehalem series CPU in my U27: sse4.2 sse4.1 ssse3 popcnt tscp ahf cx16 mon sse3 pause sse2 sse fxsr mmx cmov sep cx8 tsc fpu[03:06:25] <LaidBack_01> no issue... I'm just trying to compile up another one out of the spec files[03:06:35] <alanc> seems weird that a web proxy would use 3Dnow instructions though[03:06:38] <LaidBack_01> if that doesn't work, I'll use lighttpd[03:06:46] <richlowe> If the instructions are in some optional path, for instance, if nginx actually checks if the features are available, you can clobber the hwcap sections with elfedit[03:07:17] <bball> I installed onnv 134 from the iso.. I didn't see an option for setting the root password during install and "opensolaris" isn't working...[03:07:28] <x58> alanc: nginx is so much more than a web proxy.[03:07:31] <richlowe> bball: same as the password of the first user, I think[03:07:46] <bball> richlowe: ty :-)[03:07:52] <alanc> x58: that's just what came up in google, since I didn't know what it was[03:08:24] <x58> alanc: It is a very small, EXTREMELY fast web server that has FastCGI support, proxy/caching, and a whole range of other features.[03:08:33] <x58> Speed is where it accels though.[03:09:21] <alanc> okay, and that still seems like a strange place to have 3Dnow instructions used, but maybe there's something in there useful for more than just the obvious graphics/matrix operations[03:09:52] *** nitrile has joined #opensolaris[03:09:53] <cypromis> nginx is usefull for django projects[03:09:53] <x58> Sure, I wasn't arguing that :P Just pointing out that nginx is not just a web proxy.[03:10:13] <LaidBack_01> nginx is also useful for catalyst ... ;)[03:10:50] <x58> I use nginx with FastCGI PHP.[03:11:53] *** nitrile_ has quit IRC[03:13:00] <BankerBarney> whats the command to list groups on the server[03:13:04] <BankerBarney> not users...but groups[03:13:18] <BankerBarney> listusers for users[03:13:22] <x58> cat /etc/groups[03:13:29] <trygvis> getent groups[03:13:50] <BankerBarney> Unknown database: groups[03:13:50] <BankerBarney> usage: getent database [ key ... ][03:14:07] <trygvis> man getent[03:14:10] <BankerBarney> cat: cannot open /etc/groups: No such file or directory[03:14:19] <tsoome> group[03:14:32] *** cocacola has quit IRC[03:14:34] <BankerBarney> that did it[03:14:35] <BankerBarney> thx[03:16:11] *** vergil66 has joined #opensolaris[03:16:24] <x58> sorry, s/s//[03:16:57] <BankerBarney> i was trying to add the user root to the ftp group, but I have no ftp group[03:17:51] <richlowe> You're probably wanting to read 'man ftpusers'[03:18:00] <BankerBarney> k[03:18:16] <LaidBack_01> richlowe: thanks so much for the elfedit hint. It works when I removed that hardware capability bit out of the binary.[03:20:44] <LaidBack_01> nice and it's working with my catalyst app... thanks for helping![03:23:01] *** sergiusens has joined #opensolaris[03:24:18] *** _zenbalrog has joined #opensolaris[03:24:40] *** _zenbalrog has left #opensolaris[03:25:28] *** agrajag has quit IRC[03:26:00] *** agrajag has joined #opensolaris[03:30:28] <BankerBarney> [21:22] <richlowe> You're probably wanting to read 'man ftpusers' <--- I dont wanna do what I was getting ready to do...[03:30:34] * BankerBarney tries to find a different way[03:34:17] *** cocacola has joined #opensolaris[03:34:55] <lewellyn> smrt: hi[03:34:56] <smrt> Hi, I'm smrt. I try to be a helpful bot. For more information: /msg smrt help[03:35:01] <lewellyn> yay the bot's alive[03:35:02] *** miamano has quit IRC[03:35:23] <lewellyn> speaking of alive... did everyone hear that ed roberts died? :([03:35:52] <lewellyn> LaidBack_01: are you using nginx from the webstack repo?[03:36:02] <LaidBack_01> yes I am[03:36:22] <lewellyn> file a bug :)[03:36:38] * lewellyn isn't sure which tracker the webstack guys look at though[03:36:48] <LaidBack_01> Oh, okay. Didn't think it was that big a deal, but sure, I can do that.[03:37:29] <lewellyn> well, if it only runs on modern (AMD-only) chips, i'd call it a big deal, personally[03:37:30] *** javashin has joined #opensolaris[03:37:40] <alanc> had to gootle to see who Ed Roberts was[03:37:45] <alanc> google, even[03:38:14] <lewellyn> alanc: :([03:38:30] <lewellyn> we kind of have him to thank for the computing explosion[03:38:38] <lewellyn> personal computing, mostly[03:38:59] <alanc> yeah, was just reading http://www.nytimes.com/2010/04/03/business/03roberts.html[03:39:25] <lewellyn> wow. far better article than the bbc had[03:45:11] *** TommyTheKid has joined #opensolaris[03:46:02] *** vergil66 has left #opensolaris[03:46:35] *** swa_work has joined #opensolaris[03:55:10] <system5> the community is not non-existant[03:55:20] <system5> whoever said that[03:55:33] <zmisc> that was me and it was a long time ago[03:55:36] <zmisc> old news[03:55:39] <system5> how can you have people like Joerg and Martin in your forums and not call it a community[03:55:58] <lewellyn> system5: as the person who said it says it's old news, i'd suggest dropping the subject[03:56:16] *** chendy has quit IRC[03:56:23] <system5> a lot of the technology that was in the original OpenSolaris 2008.05 was borrowed from Belenix and Schillix, which were community based distros[03:56:26] <system5> ok, I'll drop it[03:59:18] <bball> I'm following the readme for building from source but I'm unatble to install the SUNWonbld package[03:59:31] <bball> pkgadd: ERROR: no packages were found in </var/tmp/onbld>[03:59:51] *** nitrile_ has joined #opensolaris[04:00:02] *** chendy has joined #opensolaris[04:00:05] <bball> any idea what I'm doing wrong ?[04:00:25] <lewellyn> bball: which readme?[04:00:37] <bball> README.opensolaris[04:00:54] <bball> that was in in the base dir where I extracted the on-src tarball[04:00:54] <lewellyn> ok. that doesn't describe much where you found said readme[04:01:01] <lewellyn> ok ON[04:01:05] <lewellyn> smrt: explain on[04:01:06] <smrt> The consolidation which delivers the OpenSolaris kernel, filesystems, some drivers and other modules, basic commands, daemons and servers, libraries, and system headers. Also known as OS/Net or OS/Networking.[04:01:11] <lewellyn> smrt: explain ON[04:01:12] <smrt> The OS and Networking consolidation - contains the userland and kernel components of OpenSolaris. Source for the current release at http://src.opensolaris.org/source/xref/onnv/onnv-gate. See also http://www.opensolaris.org/os/community/on/ and http://opensolaris.org/os/community/on/devref_toc/[04:01:22] <lewellyn> see that last url[04:01:32] <bball> ty.. reading now[04:01:35] <richlowe> More than likely all that is out of date with respect to this[04:01:44] <richlowe> SUNWonbld is no longer an SVR4 package, so isn't provided as one[04:01:49] *** delewis has joined #opensolaris[04:01:56] <richlowe> I think it's a plain tarball now.[04:02:05] * lewellyn is kinda hoping that the devref may have been updated by now[04:02:10] <richlowe> there's also developer/build/onbld, though that one will be out of date.[04:02:32] <lewellyn> since it broke to some degree with the switch to xwiki and stuff's changed since[04:02:43] *** nitrile has quit IRC[04:02:55] <richlowe> lewellyn: last time I thought about it, I didn't actually know the answers.[04:03:03] <richlowe> as maybe obvious from the way I'm hedging, not sure I do now, either.[04:03:45] <lewellyn> i gave up building ON since i don't have any build environments capable of it.[04:04:00] <richlowe> and for that matter, I'm not sure where README.opensolaris comes from.[04:04:03] *** nitrile_ has quit IRC[04:04:06] *** jiraia has joined #opensolaris[04:04:12] <lewellyn> some on-src tarball[04:04:28] <lewellyn> which i assume is a tarball of the hg repo[04:04:32] *** skeeziks has joined #opensolaris[04:04:52] *** nitrile has joined #opensolaris[04:04:55] <richlowe> lewellyn: not in this respect.[04:04:59] <richlowe> looks like it's built from a page on hub.o.o[04:05:05] <richlowe> which I guess also should be fixed[04:05:21] <lewellyn> lots of things need to be fixed. ;)[04:08:05] <bball> section 2.5 on that page has the same instructions for install SUNWonbld[04:08:20] <bball> err section 2.4[04:08:45] <lewellyn> bball: you have the svr4 package for onbld?[04:09:00] <bball> http://dlc.sun.com/osol/on/downloads/current/[04:09:04] <bball> I got the package from there[04:09:43] <bball> I've been fumbling through various build howtos and such and everthing seems to point to different places :-)[04:09:44] <lewellyn> that's not an SVR4 package[04:10:04] <lewellyn> untar it and you'll see[04:10:07] <alanc> looks like it's just a tarball now[04:10:15] <lewellyn> which kind of sucks[04:10:22] <richlowe> what would you rather it was?[04:10:27] <bball> ok.. yeah.. just looks like a tarball[04:10:41] <bball> can I just place that in /opt ?[04:10:48] <lewellyn> richlowe: at least an svr4 package. i'd rather not have unpackaged files[04:10:52] <richlowe> lewellyn: developer/build/onbld is actually enough, in almost all cases, to build your own tools. Not only that, but nightly and bldenv default to using the workspace tools anyway, so really it's only those two scripts that matter.[04:10:54] <alanc> but if you want an IPS pkg, there's one in the repo, though it's a couple builds old[04:11:02] *** Section9 has quit IRC[04:11:14] <lewellyn> richlowe: ok. tell bball that, not me, since i don't have any machines able to build ON right now :)[04:11:28] *** CodeWar has joined #opensolaris[04:11:30] <richlowe> if I could explain it clearly and concisely, I'd just fix the bloody documentation :)[04:11:59] <lewellyn> i tried:[04:12:01] <lewellyn> You are not allowed to view this document or perform this action.[04:13:46] <bball> richlowe: not sure I follow what you are saying.. are you suggesting I don't need to install SUNWonbld.i386.tar.bz2 ?[04:14:54] <richlowe> you should mv the opt/onbld in the tarball into /opt[04:15:03] <bball> hehe ok.. that's what I did :-)[04:15:29] <richlowe> it nearly always works to just use the IPS package, and rely on the workspace tools being used, or whatever, but I just realized it won't right now.[04:15:41] <richlowe> alanc: 'cos of Mark and Liane ;)[04:15:49] *** zmisc has quit IRC[04:17:23] <alanc> sure, blame them, when they're not here[04:18:35] <richlowe> so for 3a: "Otherwise, use pkgrm(1m) or pkg uninstall to remove the old package", 3b: "The SUNWonbld tarball is an archive of the /opt/onbld directory. Unpack it in a temporary directory and then mv the opt/onbld directory into /opt. For example: # cd /var/tmp # bzip2 -dc ... | tar xf - # mv onbld/opt/onbld /opt/"[04:18:52] <bball> think there is another small error on the wiki in section 2.5[04:19:00] *** nitrile_ has joined #opensolaris[04:19:02] <bball> it says you should add "/opt/onbld/`uname -p`/bin[04:19:11] <lewellyn> that sounds right[04:19:12] <richlowe> it means /opt/onbld/bin/`uname -p`[04:19:13] *** kwinz3 has quit IRC[04:19:15] <bball> to your path.. but I think it's /opt/onbld/bin/uname -p[04:19:24] <lewellyn> that doesn't sound right[04:19:26] *** sstallion has quit IRC[04:19:34] <lewellyn> but i may be wrong[04:19:39] <richlowe> lewellyn: half and half[04:19:50] <richlowe> lewellyn: it's bin/`uname -p`, but I'd say that's the wrong bit. :)[04:19:55] <lewellyn> likewise[04:21:26] *** kwinz3 has joined #opensolaris[04:21:51] *** plompi has joined #opensolaris[04:22:03] *** nitrile has quit IRC[04:22:03] <richlowe> problem with fixing the devref, is we need to stop describing bfu and all of that stuff, too.[04:22:52] <alanc> first, you have to kill bfu[04:23:01] *** plompi has left #opensolaris[04:23:11] <richlowe> alanc: jmcp took the bug, I offered him my diffs.[04:23:15] <alanc> it's still lurching around undead, but not yet alive[04:23:36] <richlowe> I don't think we ever reached agreement with our piscene friends as to the specifics.[04:23:42] *** nitrile_ has quit IRC[04:23:59] *** nitrile has joined #opensolaris[04:25:26] <alanc> wonder if the same community members shocked that SXCE downloads went away after months of warning will also be surprised by the demise of bfu[04:25:49] <richlowe> fixed the /opt/onbld/$MACH/bin/ v. bin/$MACH thing[04:25:51] *** davenz has left #opensolaris[04:26:27] <alanc> though given that the only way to get post-134 right now is the 136 bfu's, I guess it would be unfair to push too quickly to ax it[04:26:43] <DerSaidin> whats replacing bfu?[04:26:46] <richlowe> pkg[04:26:51] <alanc> pkg image-update[04:27:10] <richlowe> alanc: I have no way to save a copy and show people diffs, with xwiki, do I?[04:27:15] <alanc> well, 136 bfu's or build yourself from the gate[04:27:21] <alanc> richlowe: not that I know of[04:29:11] *** nitrile_ has joined #opensolaris[04:29:35] *** Wild_Geese has quit IRC[04:30:08] *** bklang has quit IRC[04:30:33] *** sahafeez has quit IRC[04:31:00] <bball> ok.. I think I'm getting close to being able to run "nightly" :-)[04:31:18] <bball> not sure how confident I am that it will actually succeed[04:32:13] <richlowe> wow, the devref tells you to read the gate's public/README[04:32:16] <richlowe> that's... useful[04:32:22] *** nitrile has quit IRC[04:33:44] <richlowe> http://pastie.org/private/rt9i0neim7ckxf2n9kxc2g[04:34:22] <lewellyn> the devref can use quite a bit of love[04:34:34] <richlowe> lewellyn: there's only really kupfer maintaining it.[04:35:02] <lewellyn> i gathered :([04:35:17] <richlowe> Well, now it's no longer docbook many more people easily *can*[04:35:20] <richlowe> just a matter of seeing if they will[04:35:23] *** Wild_Geese has joined #opensolaris[04:36:16] <lewellyn> they also need access to do so[04:36:36] *** CodeWar has quit IRC[04:36:38] <richlowe> lewellyn: the pastie link, above, is README.opensolaris diffs, if anyone feels like checking them.[04:36:48] <alanc> looks fine to me[04:36:49] <richlowe> uh, but without the "lewellyn:" on the front.[04:36:58] <richlowe> lewellyn: anyone who can edit ON pages can edit that.[04:37:07] <lewellyn> richlowe: which wouldn't be me. :)[04:37:12] <richlowe> not sure precisely who that is, I was pretty surprised it included me.[04:37:25] <lewellyn> anyhow, i still feel like ass, so i'll be heading back to bed now[04:41:56] <system5> hey, I got a good idea for a new OpenSolaris t-shirt[04:43:19] <system5> guess no one cares[04:43:55] <devians> hey, has anyone else found that osol cifs freaks out if you have files with non standard characters? ie if a folder contains a file with a character in the name with an umlaut, cifs wont display the contents of the directory till you remove the umlaut and restart cifs?[04:44:04] <system5> what version?[04:44:07] <system5> devians[04:44:08] <winstonw> system5: snuggle snuggle[04:44:11] <devians> 2009.06[04:44:14] <system5> what is your output for cat /etc/release[04:44:19] <alanc> we were waiting for you to tell us the idea[04:44:34] <system5> I was thinking the new OpenSolaris t-shirt could have a logo and say:[04:44:42] <system5> Rumors of my death have been greatly exagerrated....[04:44:45] <system5> what do you think?[04:44:49] <winstonw> lul[04:45:00] <devians> system5: snv_111b[04:45:28] <winstonw> I dislike all the press about organizations trying to kill off OSOL, its rather annoying and unofficial[04:45:28] <system5> hmmm, I'll google for CIFS bugs, good to know that you're having probs since I'm due to deploy my first production CIFS OpenSolaris server next week[04:47:08] <system5> but the shirt can say whatever it was that Mark Twain said[04:47:12] <lewellyn> devians: use only utf-8 characters in the filenames[04:47:21] <system5> "Reports of my death have been greatly exagerrated" or whatever[04:47:27] <lewellyn> devians: and upgrade to to post-2009.06. there are many bugs in the cifs in the release[04:47:28] *** smyth_rj has quit IRC[04:48:33] *** smyth_rj has joined #opensolaris[04:49:24] <alanc> too bad we never got a mascot picked - having the mascot wearing a kevlar vest with the bubble logo appearing as bullet impacts would be a fun picture for a shirt like that[04:49:43] <lewellyn> no mascot, just bubbles[04:50:08] * bda still likes the starfish idea.[04:52:37] *** devians has quit IRC[04:55:30] *** Jenkens has joined #opensolaris[04:56:00] <Jenkens> is anyone here using a solaris share to back up via TimeMachine in Snow Leopard?[04:56:20] *** nikolam_ has joined #opensolaris[04:59:09] *** devians has joined #opensolaris[04:59:29] *** nikolam has quit IRC[05:01:42] *** deirdres has joined #opensolaris[05:02:05] <devians> lewellyn: i was going to wait for 10.03 to update. are you saying i should move to dev?[05:03:42] *** Netwolf has quit IRC[05:04:18] *** javashin has quit IRC[05:04:19] *** Shoggoth has joined #opensolaris[05:04:55] <Shoggoth> hi all! what's the correct way to download the opensolaris source?[05:06:41] <Shoggoth> I've created an account on the opensolaris site but when I go to "Code Repositories" from the nav bar I get "Unauthorised access - You have attempted to view a page that you are not allowed to access. Please log in with an account which has the required level of privilege for the page you are trying to view. "[05:07:58] <lewellyn> devians: currently dev is similar to the next release[05:08:07] *** javashin has joined #opensolaris[05:08:07] <richlowe> Shoggoth: Yeah, that's a really bad link title[05:08:17] <devians> how do you swap to dev and update?[05:08:23] <lewellyn> that's a horrible link title, since i still don't know what lies behind it[05:08:30] <lewellyn> smrt: explain dev repository[05:08:31] <smrt> To switch to the /dev repository (OpenSolaris development releases, see the IPS version in /topic): pfexec pkg set-publisher -O http://pkg.opensolaris.org/dev opensolaris.org && pfexec pkg install SUNWipkg && pfexec pkg image-update -v # Then wait. You *must* read the release notes before starting out or you *will* be unhappy. See also: ips mirrors[05:08:56] <devians> kk, will look into it[05:08:56] <richlowe> lewellyn: don't suppose you know of a decent page listing hg trees, etc, etc?[05:08:57] <lewellyn> Shoggoth: which consolidation's source are you looking for?[05:09:01] <richlowe> kinda embarrassed to say I don't.[05:09:02] <lewellyn> richlowe: not really :([05:09:14] <lewellyn> i keep meaning to track them all down and list them on my site[05:09:33] <richlowe> dlc.sun.com/osol/ has http downloads, the mercurial stuff is ssh://anon at hg dot opensolaris.org/hg/<project>/<gate>[05:09:35] <Shoggoth> lewellyn: essentially I want to download the most current one[05:09:36] <lewellyn> since apparently hub's not an appropriate place, since there isn't already a page :)[05:09:49] <lewellyn> smrt: explain WOS[05:09:51] <smrt> Um. I seem to not know anything about WOS...[05:09:52] <lewellyn> smrt: explain wos[05:09:53] <smrt> "Wad Of Stuff," referring to the integration of all consolidations that make up the Solaris binary distribution shipped to customers.[05:09:58] <lewellyn> smrt: explain consolidation[05:09:59] <smrt> A group effort that compiles together software, tests the result and delivers the resulting packages into a release. A consolidation consists of one or more source gates, staff that manages the flow of putbacks/commits, builds the packages and generally verifies putbacks/commits are ready before they are done. An example is the ON consolidation, which consists of the kernel, libraries, and basic utility and server programs.[05:10:04] <richlowe> lewellyn: "Code Repositories" takes you to repo.o.o, which is SCM management.[05:10:13] <richlowe> it's a bloody awful link[05:10:20] <lewellyn> Shoggoth: you need to be slightly more descriptive as to what you want[05:10:33] <Shoggoth> richlowe: thanks... precisely what I was after[05:10:34] <lewellyn> richlowe: since i've never seen the other side of the link, i get annoyed when i click it :P[05:10:38] *** smyth_rj has quit IRC[05:10:49] <lewellyn> Shoggoth: there's more than just hg in use, keep in mind :)[05:10:50] <Shoggoth> lewellyn: richlowe has just provided what I was after, thanks[05:10:51] <alanc> plus repo.os.o is a horrible name for scm management, since more people associate repo with pkg repo's now[05:11:00] *** javashin has quit IRC[05:11:10] <lewellyn> alanc: that's not the SCM folks' fault ;)[05:11:19] <Shoggoth> oh that's ok... at this point in time I just want to _look_ at the sources... not build and/or modify them[05:11:51] <alanc> there's always http://src.opensolaris.org if you just want to look in your web browser[05:12:01] <richlowe> alanc: it doesn't really give you enough info to clone though.[05:12:05] <richlowe> /sc nearly does[05:12:05] <Shoggoth> I'd rather be able to do it offline[05:12:20] <lewellyn> Shoggoth: that's still just a subset of the sources[05:12:27] <Shoggoth> oh![05:12:28] *** javashin has joined #opensolaris[05:12:32] <richlowe> wherever /sc went.[05:12:34] <Shoggoth> ok... let me go the whole hog then....[05:12:38] <lewellyn> see what smrt said about wos and consolidation[05:12:55] <richlowe> "all the source" is a scary amount of stuff.[05:13:04] <lewellyn> this isn't linux where the "OS" is just a kernel[05:13:05] <Shoggoth> where is the most canonical, up-to-the-minute, publically accessable repo?[05:13:08] *** mroconnor has quit IRC[05:13:12] <lewellyn> for which consolidation? ;)[05:13:30] <lewellyn> there's no master repository for the entire project[05:13:38] <alanc> sort of amusing to see the people confused that you can't check out the source for individual packages Linux/RPM style vs the people confused you can't check out the source for the entire OS at once BSD-style[05:13:39] *** InTheWings has quit IRC[05:13:59] <Shoggoth> lewellyn: are you asking if I want 2009.06 vs something else?[05:14:02] <lewellyn> no[05:14:11] <lewellyn> smrt: explain consolidation[05:14:12] <smrt> A group effort that compiles together software, tests the result and delivers the resulting packages into a release. A consolidation consists of one or more source gates, staff that manages the flow of putbacks/commits, builds the packages and generally verifies putbacks/commits are ready before they are done. An example is the ON consolidation, which consists of the kernel, libraries, and basic utility and server programs.[05:14:18] <lewellyn> smrt: explain sfw[05:14:19] <smrt> Um. I seem to not know anything about sfw...[05:14:21] <richlowe> alanc: per-package source would be useful in some cases, not in others.[05:14:26] <lewellyn> oh that's a huge glaring omission[05:14:31] <Shoggoth> lewellyn: sorry, I don't have a handle on the opensolaris developer jargon[05:14:33] <lewellyn> smrt: explain on[05:14:34] <smrt> The consolidation which delivers the OpenSolaris kernel, filesystems, some drivers and other modules, basic commands, daemons and servers, libraries, and system headers. Also known as OS/Net or OS/Networking.[05:14:35] <alanc> do you want the kernel, the desktop, X, or other software? there's different repos for each (plus about a dozen more)[05:14:50] <Shoggoth> ahhh!!!! I see[05:14:54] <lewellyn> Shoggoth: a consolidation is just a subset of the project. you have to be specific what you want[05:15:00] <richlowe> lewellyn: 'sfw' should take care to explain the difference between sfwnv, and the companion cd (which happens to use SFW as a package prefix).[05:15:13] <Shoggoth> so there isn't a single place I can download the whole lot and re-build an ISO from source[05:15:18] *** chendy has quit IRC[05:15:20] <lewellyn> richlowe: that may be why i didn't add it when i added all the rest (mostly from the devref)[05:15:28] <lewellyn> Shoggoth: no[05:15:33] *** smyth_rj has joined #opensolaris[05:15:39] <Shoggoth> that's fair enough.... I've just gotten to used to the linux and/or bsd way of doing things[05:15:42] <lewellyn> hell, i'm not even sure how to build the ISO right now :P[05:15:50] <lewellyn> well, i never was :)[05:15:53] <Shoggoth> ok.... so what I want is the kernel source[05:15:57] *** lattera_ has joined #opensolaris[05:16:00] <lewellyn> smrt: explain ON[05:16:01] <smrt> The OS and Networking consolidation - contains the userland and kernel components of OpenSolaris. Source for the current release at http://src.opensolaris.org/source/xref/onnv/onnv-gate. See also http://www.opensolaris.org/os/community/on/ and http://opensolaris.org/os/community/on/devref_toc/[05:16:01] <Shoggoth> and maybe some of the basic userland stuff[05:16:01] <richlowe> rebuilding the ISO from source would take a long time[05:16:30] <richlowe> that'd be ON: http://dlc.sun.com/osol/on/downloads, ssh://anon at hg dot opensolaris.org/hg/onnv/onnv-gate[05:16:31] <lewellyn> yeah. i've only entertained the thought of building my own iso once. comay laughed at me and i dismissed the idea :P[05:17:00] <richlowe> lewellyn: building it from packages, or bits of your own build + packages is trivial.[05:17:02] <lewellyn> richlowe: the latter two links have download instructions i think[05:17:05] *** Jenkens has quit IRC[05:17:07] <Shoggoth> lewellyn: lol.... let me give you an insight into what I'm looking for and then perhaps you can point me at the right stuff[05:17:13] <lewellyn> richlowe: this was build 130, iirc[05:17:19] <richlowe> lewellyn: http://blogs.sun.com/lianep/entry/testing_on_changes_with_opensolaris[05:17:45] <richlowe> lewellyn: similar would apply if you changed other consolidations, except getting the packages into IPS form would be tedious.[05:17:45] <lewellyn> richlowe: which would have meant i would have build it from scratch to get it earlier than everyone else ;)[05:17:48] <Shoggoth> lewellyn: I've been looking at the following project: http://hub.opensolaris.org/bin/view/Project+devmapper/[05:18:34] <lewellyn> Shoggoth: ok. they link to their repo info too[05:18:51] <Shoggoth> and I would like to perhaps add / port some code to provide equivalent implementations of some of the dev-mapper functions that linux supports that the current solaris port doesn't... vis-a-vis serpent and twofish crypto[05:18:54] <lewellyn> http://hub.opensolaris.org/bin/view/Project+devmapper/source[05:19:08] <Shoggoth> lewellyn: yes... I've gotten that already... please see the 2nd part[05:19:33] <lewellyn> what second part? i don't see reference on that page that you need ON[05:19:35] <Shoggoth> I need to get a mental picture of how solaris does aes in the kernel and how (if?) it's possible to add other crypto[05:19:37] <alanc> richlowe: distro-import isn't all that hard once you have the SVR4 packages - getting those if you're external and want a post SXCE build is the hard part[05:19:56] <Shoggoth> lewellyn: lol... no not on that page... the 2nd part of my question above...[05:19:58] <Shoggoth> :)[05:20:00] <lewellyn> Shoggoth: you may want to start with opengrok[05:20:14] <richlowe> alanc: you'd be building them.[05:20:17] <lewellyn> sometimes it's easier to point and click your way through unfamiliar opensolaris code :)[05:20:30] <Shoggoth> ok... cool... that looks like a nice start[05:20:31] <richlowe> alanc: packaging works in the X build, right? I know it does (did?) for SFW[05:20:49] <lewellyn> Shoggoth: thank alanc for the src.opensolaris.org suggestion earlier ;)[05:21:09] <Shoggoth> damn... I think I didn't even see it with the flurry of comments :)[05:21:17] <alanc> yes - since 129 you can build every single X package from the src provided on hg.opensolaris.org[05:21:17] <Shoggoth> so I'll thank him now...[05:21:22] <alanc> you're welcome[05:21:22] <Shoggoth> alanc: thanks for the link[05:21:34] <Shoggoth> ... and I'll examine my scrollback later at my leisure :)[05:21:59] <alanc> for builds older than 129, there were missing bits in some packages, which we cut down over a couple years[05:22:04] <lewellyn> Shoggoth: smrt knows lots of useful stuff, too :)[05:22:06] <lewellyn> smrt: count[05:22:07] <smrt> 397 factlets in the database at the moment.[05:22:08] <lewellyn> smrt: hi[05:22:09] <smrt> Hi, I'm smrt. I try to be a helpful bot. For more information: /msg smrt help[05:22:10] <Shoggoth> k[05:22:25] <Shoggoth> ahhhh!!!!![05:22:45] <alanc> and JDS packages can be built from the spec-files in their svn repo except for a few[05:22:53] * lewellyn hopes Shoggoth didn't fall off a building or something[05:23:03] <Shoggoth> lmao[05:23:18] <Shoggoth> no... that was the sound of a light bulb warming up above my head[05:23:19] <Shoggoth> :)[05:23:45] *** chendy has joined #opensolaris[05:24:44] <alanc> and if I didn't spend all day on IRC, you'd be able to build the X gate directly into IPS packages by 138, but I'm probably going to miss that target by a build or two[05:25:20] <Shoggoth> lol[05:25:54] <richlowe> alanc: slacker! :)[05:25:57] <alanc> but hey, ON slipped their conversion by 7 builds, so I'm in good company[05:26:17] <richlowe> Well, that went and bit me in the ass, didn't it.[05:26:21] <alanc> 8-)[05:27:07] <Shoggoth> hey on another somewhat unrelated topic... has anyone heard anymore news wrt oracle vis-a-vis opensolaris?[05:27:27] <lewellyn> afaik, oracle has made no official announcements[05:27:42] <Shoggoth> I remember reading something this week to the effect that any _new_ tech from the closed-source product won't be making it's way into the open-source product[05:27:55] <lewellyn> considering how much of the old sun.com stuff is still tagged as "under review", that's not surprising[05:27:59] *** n3kl has quit IRC[05:28:10] <lewellyn> Shoggoth: there were never guarantees that solaris stuff would end up in opensolaris[05:28:27] <lewellyn> look at the complaints of lost functionality when sxce died as proof[05:28:44] <Shoggoth> lewellyn: true, but when the comment comes after aquisition they tend to carry a bit more weight[05:28:44] <alanc> that seemed to just be mangling the statement Dan Roberts made in IRC weeks ago that *some* new stuff would be closed - just as Sun kept the Storage Appliance ("FishWorks") software closed[05:29:00] <Shoggoth> alanc: that's good to hear[05:29:07] <alanc> not *all*, not *any*, but *some*[05:29:12] <Shoggoth> damn media[05:29:16] <lewellyn> Shoggoth: note that most of the "news" isn't from reputable sources[05:29:29] <alanc> after all, if you look at the hg repos, there's been plenty of new commits by Oracle employees[05:29:44] <Shoggoth> doesn't matter if it's the 6 o'clock news talking about a drug bust or a medical journal ... they seem to have the same attention to detail regardless[05:29:55] <lewellyn> http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6799081 is a good example of a bug that irks people now[05:29:56] <Shoggoth> ie. none![05:30:33] <lewellyn> Shoggoth: there's attention to detail and then there's actually knowing anything about what you're reporting on[05:31:06] * lewellyn wonders why the bmc driver can't go into the extra repo[05:31:07] <Shoggoth> lewellyn: unfortunately that sounds about right[05:31:13] *** Garen has quit IRC[05:31:15] <richlowe> lewellyn: You're mistaking "can't" and "hasn't"[05:31:26] <richlowe> lewellyn: if you look in /extra, you'll find, last I knew, absolutely none of ON[05:31:44] <alanc> richlowe: except the handful of wbem closed bits needed to build the open bits[05:31:52] <richlowe> alanc: they're from admin[05:31:52] *** Garen has joined #opensolaris[05:32:03] <lewellyn> so extra isn't required for ON anymore?[05:32:12] <alanc> oh are they? I guess I just associate them with ON because they're part of the osnet-dev cluster[05:32:15] <lewellyn> last time i tried, pkg install osnet required bits from extra[05:32:17] <richlowe> extra is required to build ON, but none of ON lands in extra[05:32:27] <richlowe> a bunch of it probably should, but it doesn't.[05:32:30] <Shoggoth> lewellyn: it's an unfortunate state of affairs... "regular" current affairs journalism you'd expect that from (as CP Snow described as "The Two Cultures") but from tech journo's you'd hope for something a _little_ bit better[05:33:02] <lewellyn> Shoggoth: everyone's having a field day reporting on the death of something that's still alive. take it all with a grain of salt[05:33:07] <alanc> lewellyn: last I heard they were still trying to figure out which closed drivers should go to extra or not, but other issues have been higher priority[05:33:28] <richlowe> /extra is a whole different kettle of fish[05:33:36] <lewellyn> alanc: it'd be nice if there was a poll for the community to vote on which ones they miss most :)[05:33:40] <richlowe> "redistributable" v. "not" is known (in theory) because it affects the closed-bins, too.[05:33:47] <Shoggoth> lewellyn: normally I would take it with an ocean of salt... but I don't trust leisure suit larry :)[05:33:47] <richlowe> but "Can be in /extra" probably requires re-lawyering.[05:34:13] <alanc> unfortunately the immediacy of web based publishing drives a lot of tech journalists to rush the article to beat the competition, and take less time for fact checking[05:34:19] <richlowe> lewellyn: probably bmc, at least one person misses qus[05:34:26] <richlowe> doubt most folks realize what else is absent.[05:35:11] <alanc> besides - the more controversial the story, the more advertising revenue - nothing better for getting eyeballs to ads than people passing around your link with "Have you read this crap? We need to all go correct him now!"[05:35:26] <Shoggoth> alanc: yes I kind of agree... that's the line many would draw between journalists and bloggers, but to be fair I don't think that traditional journalists have been any different over the last 20 years or so[05:35:47] <richlowe> lewellyn: bmc, isp, ifp, ksh88, spwr[05:36:14] <richlowe> lewellyn: think that's the list of bits of ON that aren't redistributable, but have been shipped before[05:36:39] <lewellyn> ksh88 doesn't need to return :P[05:37:10] <richlowe> oh, and qus.[05:37:22] <richlowe> I think qus, ifp, and isp are mostly missed by people with old-ish ultrasparc hardware[05:37:25] * lewellyn heads off now that the gf is finally home[05:37:39] <alanc> but it might be cheaper to buy me a new scsi controller than deal with qus 8-)[05:37:50] <richlowe> alanc: if that logic holds, I have an ifp around here someplace...[05:38:01] <richlowe> alanc: probably in an ultra5, of all things.[05:39:41] *** nitrile has joined #opensolaris[05:39:57] *** sergiusens has quit IRC[05:41:38] *** Edgeman has quit IRC[05:42:22] *** nitrile_ has quit IRC[05:48:01] *** syamajala has joined #opensolaris[05:48:19] *** syamajala has quit IRC[05:54:37] *** TommyTheKid has quit IRC[05:59:12] *** mroconnor has joined #opensolaris[06:07:03] <kohju> hmm. terrible, php in opensolaris does not support pspell... some php software doesn't work...[06:07:54] *** e1kg has joined #opensolaris[06:08:35] *** deirdres has quit IRC[06:09:20] <kohju> Can I build the php5 added '--with-pspell' ? php build environments are in sfw-nv?[06:10:19] <lewellyn> i think all the webstack stuff is in sfw now, yes[06:11:01] *** LaidBack_01 has quit IRC[06:11:21] *** eleusis has quit IRC[06:12:53] *** eleusis has joined #opensolaris[06:12:53] *** eleusis has joined #opensolaris[06:15:53] <kohju> Thanks, lewellyn. I don't know the building sfw-nv environment...[06:15:58] <lewellyn> nor do i[06:16:05] <lewellyn> i've not had the cajones to try it yet[06:16:45] <kohju> here? http://hub.opensolaris.org/bin/view/Project+sfwnv/install_quickstart[06:17:47] *** hsp has joined #opensolaris[06:19:13] *** cn28h has quit IRC[06:19:25] <lewellyn> kohju: perhaps. as i said, i've not tried :)[06:20:40] *** jamesd_laptop has joined #opensolaris[06:21:47] *** jamesd2 has quit IRC[06:22:19] <kohju> :)[06:24:59] *** zenjuzu has joined #opensolaris[06:33:14] *** javashin has quit IRC[06:34:02] <alanc> sfwnv is very much like onnv, but I've never tried building it either[06:35:41] *** vraa has joined #opensolaris[06:36:39] <system5> any word on whether there will ever be a working IPMI / BMC in OpenSolaris?[06:36:41] <system5> http://www.cuddletech.com/blog/pivot/entry.php?id=881[06:37:05] <system5> some people on opensolaris.org said they tried copying the files over from SXCE to OpenSolaris and it panicked the system[06:37:49] <system5> see also: http://opensolaris.org/jive/thread.jspa?messageID=464909[06:39:35] <kohju> hmm[06:40:44] <system5> we just need to work on getting a minimal core install like FreeBSD that is 100% Opensource and OpenSolaris will be all set, Oracle can make oodles of legacy cash cow money just by owning the only distro that has a real UNIX CDE desktop in it[06:40:57] <system5> while all the rest of us freetard posers use GNOME or XFCE :-)[06:41:04] <jdoe> haha. it sure would be nice to be able to easily install opensolaris without x/whatnot :P[06:41:08] <bda> CDE has been excised from ONNV.[06:41:12] <bda> It will not be in Solaris Next.[06:41:21] <system5> you sure Oracle won't put it back? AIX and HP-UX still have it[06:41:29] <bda> Ask alanc.[06:41:44] *** smyth_rj1 has joined #opensolaris[06:41:52] <alanc> technically CDE was never in ONNV[06:41:58] <bda> ah, fair.[06:42:04] <alanc> since ON doesn't have desktop software in it[06:42:10] <bda> :)[06:42:10] <alanc> but it's gone from Nevada to[06:42:12] <alanc> o[06:42:15] <system5> I think you can install OpenSolaris now without X-windows using the "textinstall" image from here: http://www.genunix.org/[06:42:30] *** xapiens has quit IRC[06:42:54] <system5> I remember hearing somewhere that installing Oracle RDBMS on OpenSolaris used to crap out because it was looking for some motif / CDE related library[06:43:02] <system5> I'm sure that will be true for lots of legacy software[06:43:13] <alanc> the Oracle installer uses a version of Java that requires Motif[06:43:23] <alanc> Motif is available in the OpenSolaris repo[06:43:31] <system5> If Oracle really wants to be "the next IBM" they'll have to hold on to those legacy cash cow customers still running archaic stuff that requires Solaris 7 or whatever[06:43:35] *** smyth_rj has quit IRC[06:43:42] <alanc> CDE libraries required for binary compatibility are provided in the OpenSoalrsi repo[06:44:00] <alanc> system5: what do you think Solaris 8/9/10 zones are for?[06:44:17] <system5> like, I think "Sun Secure GLobal Desktop" didn't work in OpenSolaris for the longest time and neither did Sun Ray stuff[06:44:19] <alanc> the CDE desktop & applications have been terminated with extreme prejudice[06:44:31] <system5> Sun Ray = Larry Ellison's network computer[06:44:39] <system5> don't know if they work now[06:44:44] <alanc> system5: I'm using a Sun Ray right now that is connected to a Sun Ray server running OpenSolaris 2009.06[06:44:49] <alanc> they've worked for quite a while[06:44:59] <system5> good to know it was integrated in 2009.06[06:45:09] <alanc> not integrated in 2009.06 - works with[06:45:17] <system5> see, I started out with 2008.05 instead of SXCE like other people did, so I'm used to expecting stuff not to work :-)[06:45:21] <alanc> Sun Ray is a separately sold product[06:45:42] <system5> what about XDMCP login, does that work easily in 2009.06?[06:45:58] *** zenjuzu has quit IRC[06:46:02] <alanc> I think so, but I never use it unless forced to debug it[06:46:03] <system5> it use to be real easy to do Windows thin client logins to Solaris 10 using XDMCP and Xming, wasn't secure but was easy to set up[06:46:36] <system5> well, I'm getting ready to deploy into a bunch of Micro$oft saturated environments, and my main selling point is that Solaris / OpenSolaris works better with Windoze than Linux does[06:46:47] <alanc> XDMCP login is provided by GDM, which the gnome team takes care of, so I don't have to deal with it[06:47:10] *** hsp has quit IRC[06:47:25] <system5> yeah, but Alan, check how easy it was to do Xming windows thin client logins to Solaris 10, this link explains:[06:47:28] <system5> http://blogs.sun.com/angad/entry/getting_solaris_up_and_running[06:47:31] <system5> literally takes less than 5 minutes[06:47:50] <system5> you have to use the "Solaris Management Console" which isn't in OpenSolaris, unfortunately[06:48:02] <system5> all the Solaris certification books always talk about the "Solaris Management Console" too[06:48:19] <lewellyn> which, thank god, is dead[06:48:23] *** Gman__ has quit IRC[06:48:36] *** Gman_ has joined #opensolaris[06:48:41] <system5> what's going to replace Solaris Management Console?[06:48:47] <system5> Are we getting a port of Smitty from AIX?[06:48:47] <lewellyn> i disable it on every machine, but somehow it still manages to start up and consume ram and cpu sometimes[06:48:55] <lewellyn> system5: don't joke about that[06:49:00] <system5> Smitt happens you know[06:49:01] <jamesd_laptop> smc is proof even sun can produce terible java code...[06:49:12] <lewellyn> system5: there are ibm folk in here, after all[06:49:20] <system5> there are[06:49:29] <lewellyn> system5: and have you bothered looking at visual panels yet?[06:49:30] <alanc> I thought sdtaudiocontrol was proof Sun could produce terrible java code...[06:49:35] <system5> you'd have to be blind and deaf not to notice all the IBM people trolling opensolaris.org discuss[06:49:36] *** jiraia has quit IRC[06:49:39] <lewellyn> smrt: explain sdtaudiocontrol[06:49:40] <smrt> ⌂ Gone as of b117. <@alanc> yah, I figure few will mind not having a 64mb java application to change their volume any more[06:49:47] <system5> especially the mainframe people[06:50:09] <lewellyn> system5: you seem to think that companies are a single entity with a central single view.[06:50:10] <jamesd_laptop> alanc, most sun users are not forced to use it.. so they came out with smc, so that everyone can experience it[06:50:18] <lewellyn> either that or you're trolling[06:50:35] <system5> yet, we don't have Oracle sales droids on opensolaris.org discuss yet... what gives? You'd think they'd use it as an avenue to push a sale[06:50:50] <lewellyn> system5: are you actually a sun/oracle customer?[06:51:00] <system5> I have one OpenSolaris contract[06:51:04] <alanc> Oracle sales people are still learning Sun products - it's not been that long[06:51:16] <system5> and I'm not giving my # away, I don't want you guys to be predudiced against my trouble tickets[06:51:34] <alanc> we don't see #'s, just customer names[06:51:34] <lewellyn> system5: if you were a regular customer, you'd know that right now buying any sun products is a bit painful[06:51:42] <lewellyn> i don't even see names[06:51:43] <jamesd_laptop> alanc, yes they are still learning to use ls instead of select * from directoryname ;[06:51:44] <system5> yes!!!!!!![06:52:04] <system5> I have been trying to buy Solaris support for 4 days, I'm going to e-mail alan later this evening and have him forward my e-mail[06:52:19] <lewellyn> system5: this is a huge acquisition. it won't be done overnight. let the company do its job and don't expect us here to have any clue as to the machinations[06:52:20] <system5> since the only way to get a quote for Solaris 10 or OpenSolaris support is to e-mail an "internal only" mailing list[06:52:23] <alanc> and the Oracle employees here are mostly engineers, so don't see trouble tickets, only bugs if the service people determine one should be filed[06:52:29] <lewellyn> a fair number of us are not affiliated with sun nor oracle[06:52:44] <system5> I'm tempted to learn IBM stuff just to get a job[06:52:48] <alanc> system5: did you see my earlier message that they did confirm it's a known problem they're working on?[06:52:59] <alanc> I can still forward until it's fixed though[06:53:10] <system5> it makes sense, all those people who have mainframe knowledge can't live forever, so when they retire I can get on that awesome legacy cash cow bandwagon[06:53:20] <system5> COBOL programmins is so easy too[06:53:21] <jamesd_laptop> system5, the problem is... aix isn't a free download and doesn't run on x86 and power is not cheap[06:53:29] <system5> true[06:53:35] <lewellyn> alanc: funny. oracle simply gave me a mind-boggling number of zeroes to make me go away till they get things sorted :)[06:53:46] <system5> I heard about that[06:53:47] *** chendy has quit IRC[06:53:52] <system5> what number did they give?[06:54:04] <system5> I had a server I was supposed to deploy today that didn't get deployed because we couldn't buy support[06:54:07] *** cn28h has joined #opensolaris[06:54:16] <lewellyn> there are many countries with a smaller gdp than the number they gave :P[06:54:18] <system5> it's too late now that it's the weekend[06:54:35] <lewellyn> but, again, it'd cover every single oracle product on every single machine we own :P[06:54:45] <system5> well, I own 3 machines[06:54:46] <alanc> I don't know that they'll help him, I just know that if I forward his mail it will be better than listening to another week of complaints about it here[06:54:52] <system5> so I'm not paying a billion dollars for support just yet[06:54:59] <system5> yes!!![06:55:09] <lewellyn> system5: in any case, we're all in the same boat here.[06:55:17] <system5> lewellyn, what number did they give?[06:55:28] <lewellyn> i'm not saying :P[06:55:43] <lewellyn> (rather, i'm not pulling up the email at 10pm on friday night to recount the zeroes[06:55:46] <system5> wonder if it was enough money to cover for a huge coke party for Safra and Phillips and big L with strippers included ;-)[06:56:00] <system5> just kidding[06:56:03] <DerSaidin> did they use a hooked pinky to mouth gesture? :D[06:56:14] <system5> LOL @ DerSaidin[06:56:17] <alanc> damn, reminds me I still have to take the mandatory ethics and sexual harassment training[06:56:23] <jamesd_laptop> system5, he doesn't need coke.. he is only interesting in financing his new boat... have to get the best to keep winning[06:56:36] <system5> true[06:56:46] <lewellyn> alanc: yes, we need you to be ethical when you sexually harrass people[06:57:11] <system5> eventually Oracle is going to get a big Naval fleet and you will all have to wear naval uniforms, like what L Ron Hubbard did to Scientology[06:57:19] <jamesd_laptop> at my current employer, sexual harrasment will not be reported but will be graded.[06:57:40] <alanc> of course, have to make sure that when you hire the stripper, she/he is not a cop in disguise so we don't get busted for bribing government officials[06:57:53] *** amgarching has quit IRC[06:57:54] <lewellyn> alanc: yes, i prefer my strippers to be in costume... ;)[06:58:09] <alanc> until they're out of it?[06:58:19] <system5> this could be where we are headed: http://en.wikipedia.org/wiki/SeaOrg[06:58:21] <jamesd_laptop> lewellyn, i prefer my strippers to be out of costume[06:58:49] <system5> Oracle's better than Microsoft and IBM though[06:58:58] <lewellyn> jamesd_laptop: what alanc said[06:59:22] <system5> but learning IBM's legacy stuff is tempring, I mean, how many people really know an AS/400 vs how many of those dang things are still chugging away at airlines and casinos[06:59:27] <system5> * is temptint[06:59:29] <system5> * tempting[06:59:31] <system5> can't type[07:00:19] * lewellyn sighs at the "10 minutes remaining" dialog keeping him from wandering off to play video games with the gf[07:00:24] <system5> heh, reading that wikipedia article.... if L.E. makes you guys sign a "billion year contract" you'll know you're in for trouble....[07:01:18] <system5> BTW, have any of you guys seen this article that says that the character of "Iron Man" is based on Larry Ellison:[07:01:21] <system5> http://en.wikipedia.org/wiki/SeaOrg[07:01:22] <system5> ah shoot[07:01:24] <system5> wrong link[07:01:30] <system5> this one:[07:01:34] <system5> http://blog.cratchit.org/2008/03/larry-ellison-is-iron-man.html[07:01:38] <system5> found it![07:01:44] <system5> it's a pretty compelling argument[07:02:02] <system5> Tony Stark = a poorly disguised alias for Larry Ellison[07:02:31] <system5> There you have it. Larry's face, Larry's voice, Larry's cars, Larry's money... Hey! Larry flies fighter jets. And those computers of "Tony Stark's" sure as hell aren't running Windows.[07:03:15] <system5> I need to go FSCK a stoopid Linux server in a data center three blocks away, be back later....[07:03:55] <alanc> that explains http://www.oracle.com/us/ironman2/index.html[07:04:28] <alanc> and http://www.oracle.com/marvel/index.html[07:04:49] *** Gman_ has quit IRC[07:06:04] *** Shoggoth has quit IRC[07:14:24] *** chendy_ has joined #opensolaris[07:16:42] *** Gman_ has joined #opensolaris[07:17:03] *** chendy_ is now known as chendy[07:24:48] <bball> the deref manual seems to sugest that I don't need to set "ROOT" doesn't need to be set if I use bldenv[07:24:57] <bball> s/deref/devref[07:25:14] *** chendy has quit IRC[07:25:18] *** lattera_ has quit IRC[07:25:18] <bball> but when I run nightly it it says ROOT must be set[07:27:34] *** chendy has joined #opensolaris[07:31:05] *** vraa has quit IRC[07:32:46] *** nitrile_ has joined #opensolaris[07:35:42] *** nitrile has quit IRC[07:43:32] *** andersenep has quit IRC[07:48:01] *** sponix has quit IRC[07:58:27] *** Zauberpo1y has joined #opensolaris[07:59:25] *** Mousk4r has quit IRC[07:59:55] *** Zauberpony has quit IRC[08:01:01] *** relling has quit IRC[08:06:00] *** nanase has joined #opensolaris[08:06:54] *** gm151 has quit IRC[08:12:41] *** hajma has quit IRC[08:19:08] *** chendy has quit IRC[08:23:50] <bball> how can I install the gcc package on onnv 134 ?[08:24:21] <bball> my pkg publisher is http://pkg.opensolaris.org/dev[08:24:25] <Aria> pkg install gcc-dev ?[08:28:26] <echobinary> I just installed 134 (thanks for the help earlier) and I need to do manual network config - so I followed the instructions for switching to manual by disabling NWAM.. but now I can't ping out to the net (name resolution? route?) can anyone point me in the right direction?[08:29:04] <Aria> Set up /etc/nsswitch.conf -- you probably want /etc/nsswitch.dns copied to /etc/nsswitch.conf[08:29:15] <Aria> And you need nameservers -- make sure /etc/resolv.conf has them listed[08:29:28] <bball> err.. my http_proxy was messed[08:30:10] <DerSaidin> echo 123.routerIP.1 > /etc/defaultrouter[08:30:18] *** Edgeman has joined #opensolaris[08:32:20] *** swa_work has quit IRC[08:32:27] <echobinary> DerSaidin: was that meant for me?[08:33:52] <DerSaidin> yes[08:34:36] *** chendy has joined #opensolaris[08:34:58] <echobinary> it didn't help (what is that supposed to do anyways?)[08:38:32] *** nitrile_ has quit IRC[08:38:58] *** nitrile has joined #opensolaris[08:41:07] *** jenky has joined #opensolaris[08:41:20] <jenky> whats the easiest way to get Flash going with firefox?[08:45:00] <smyth_rj1> https://pkg.sun.com/register/[08:45:09] <smyth_rj1> their extras repo has a plugin[08:45:16] <jenky> thanks[08:45:44] <smyth_rj1> np[08:46:39] *** skrewla has joined #opensolaris[08:46:58] *** jenky has quit IRC[08:47:18] <system5> echo binary, what is your output for this command: ifconfig -a[08:47:41] *** purserj has quit IRC[08:48:14] *** purserj has joined #opensolaris[08:48:15] <system5> you need to make a hostname.interface file (like on OpenBSD or UNIX System V), so if you have an intel e1000g0 network card, it would be /etc/hostname.e1000g0[08:49:11] <system5> or for a 3com elxl card: /etc/hostname.elxl0 for a broadcom: /etc/hostname.bnx0[08:50:33] <system5> you also need an /etc/netmasks file that has your subnet and netmask in it: i.e. 192.168.1.0 255.255.255.0[08:51:03] <system5> and you need an /etc/defaultrouter file with the ip address of the router that is used as a default gateway[08:51:25] <system5> dns servers go in /etc/resolv.conf[08:51:53] <system5> and the hosts line in the /etc/nsswitch.conf file needs to go like this:[08:51:55] <system5> hosts: files dns[08:52:16] <system5> and that's static IP address configuration on OpenSolaris[08:52:33] <system5> or any Solaris UNIX OS as far as I know[08:56:02] <skeeziks> Wow[08:56:08] <skeeziks> OSol adds daemon().[08:56:09] <skeeziks> Huh.[09:02:19] <skrewla> i was trying to fix a user not being able to access my smb share and the directory got a + on it and i'm unable to access it, im trying to google for this but it's been difficult finding this[09:17:29] *** RedSun has quit IRC[09:21:38] *** ziyourenxiang has joined #opensolaris[09:23:48] *** q5sys has quit IRC[09:27:23] *** Tusk has quit IRC[09:27:25] *** Mdx4 has joined #opensolaris[09:31:49] *** Aria has quit IRC[09:39:05] *** Tusk has joined #opensolaris[09:39:08] *** zazenrasta has joined #opensolaris[09:41:42] *** ozux has joined #opensolaris[09:45:31] <echobinary> I think I solved my problem... for whatever reason I had the ipaddress of the interface in /etc/hostname.rge0 and /etc/hostname.bge0[09:45:42] <echobinary> and also the entries weren't in /etc/hosts[09:46:01] <echobinary> so I reset everythign back to autoconfig, and started from scratch[09:46:14] <echobinary> it _seems_ to be working - we'll see if it still works after this reboot[09:46:25] <echobinary> (I expect it will)[09:46:38] <echobinary> thanks for pointing me in the right direction y'all :)[09:47:27] <echobinary> frak :-\[09:47:33] <echobinary> not workign after a reboot[09:49:31] *** kpt-pejsek has joined #opensolaris[09:50:09] <echobinary> odd - both my interfaces have the same manual ip after a reboot - 10.10.100.100[09:50:27] <echobinary> they should be rge0=10.10.100.100 bge0=10.10.100.101[09:50:30] <echobinary> O_o[09:50:36] <echobinary> why would something like that happen?[09:50:58] <echobinary> I reset them to the proper IPs manually and now its workign again[09:51:05] <echobinary> but I don't wanna have to do that on every reboot[09:58:41] <DerSaidin> skrewla: the + means it has ACL permissions[09:59:26] <DerSaidin> skrewla: /usr/bin/ls -v[09:59:44] <DerSaidin> by default the path has /usr/gnu/bin/ls first, which doesn't support ACLs[10:00:25] <DerSaidin> echobinary: did you mention why you couldn't just use NWAM?[10:00:46] <echobinary> because the computer is a server - and it needs a static IP[10:01:07] <DerSaidin> smrt: explain static ip[10:01:07] <smrt> There is no reason to disable NWAM just to configure a static IP address. NWAM deals just fine with this scenario, even though there is no GUI widget for it. http://blogs.sun.com/PlasticPixel/entry/nwam_static_ip_address_for (Alternatively, add a reservation to your DHCP server.) See also: http://wikis.sun.com/display/OpenSolarisInfo200906/How+to+Manually+Configure+Network+Settings[10:02:15] <echobinary> awesome![10:02:19] <echobinary> yes its true[10:02:21] * echobinary is a noob[10:02:29] <echobinary> thanks I'll give that a try[10:02:39] *** Yu\2 has joined #opensolaris[10:07:39] <RoyK> hm... I read zfs won't use the drive's cache unless it's working with a "physical device" e.g. not a partition - what exactly is a physical device by naming? is c7t2d0 a physical device?[10:08:40] <DerSaidin> RoyK: I think that means an entire disk, so d0 is all of a disk (although I'm not totally sure)[10:12:38] <RoyK> seems it is - c7t2d0p0 would be partition 0 and c7t2d0s0 would be slice 0, and c7t2d0 the whole drive[10:13:03] <DerSaidin> yeah, that is what those names mean[10:13:36] *** nikolam_ is now known as nikolam[10:13:48] <DerSaidin> but I'm unsure about how the drives cache is used and when[10:15:26] *** chendy has quit IRC[10:15:55] <RoyK> another thing - http://www.solarisinternals.com/wiki/index.php/ZFS_Best_Practices_Guide#Storage_Pools says "Avoid creating a RAIDZ, RAIDZ-2, RAIDZ-3, or a mirrored configuration with one logical device of 40+ devices. See the sections below for examples of redundant configurations." - what do they mean by that? 40+ devices in a single raidz[123] set or 40+ devices in a pool regardless of raidz[123] sets?[10:18:56] *** nikolam has quit IRC[10:22:14] <DerSaidin> they mean striping across 40-N drives with N drives parity (ie, raidzN)[10:22:50] *** ziyourenxiang has quit IRC[10:24:39] <DerSaidin> mmm, actually I'm not sure[10:25:57] <causality> probbaly because the performance will suck due to iops constraints[10:26:15] <causality> zfs isn't the most efficient file system when it comes to outright speed[10:27:39] *** ziyourenxiang has joined #opensolaris[10:28:38] *** Javache has joined #opensolaris[10:29:32] *** Javache has quit IRC[10:29:32] *** Javache has joined #opensolaris[10:29:53] *** chendy has joined #opensolaris[10:34:37] *** chendy has quit IRC[10:37:01] <RoyK> causality: raid-0 on linux is rather good at speed :þ[10:37:03] *** chendy has joined #opensolaris[10:45:57] <skrewla> is there a way to symlink something on one zfs fs to another zfs fs[10:46:59] <CosmicDJ> ln -s doesn't work?[10:49:17] <RoyK> :)[10:50:14] <RoyK> skrewla: symlinks are just smart text files, really[10:52:30] *** jamesd_laptop has quit IRC[10:53:18] *** jamesd_laptop has joined #opensolaris[11:01:12] *** m4rku5 has joined #opensolaris[11:04:42] *** ziyourenxiang has quit IRC[11:04:47] *** raichoo has joined #opensolaris[11:07:29] *** p3rror has joined #opensolaris[11:08:36] *** master_of_master has quit IRC[11:10:26] *** master_of_master has joined #opensolaris[11:15:34] *** vivid has quit IRC[11:17:58] *** vivid has joined #opensolaris[11:25:31] *** hsp has joined #opensolaris[11:26:02] *** nikolam has joined #opensolaris[11:30:51] *** ozux has quit IRC[11:35:03] *** dalibor has joined #opensolaris[11:43:42] *** tsoome has quit IRC[11:47:26] *** jamesd_laptop has quit IRC[11:52:29] *** raichoo has quit IRC[11:52:37] *** chendy has quit IRC[12:02:34] *** jollyd has joined #opensolaris[12:02:39] *** jollyd has left #opensolaris[12:03:00] *** amgarching has joined #opensolaris[12:03:37] *** tsoome has joined #opensolaris[12:04:32] *** chendy has joined #opensolaris[12:08:31] *** hsp has quit IRC[12:13:26] *** Odin- has joined #opensolaris[12:16:23] *** fOB has joined #opensolaris[12:27:02] *** Nofer has joined #opensolaris[12:28:13] *** milek_ has joined #opensolaris[12:31:38] *** ivo_ has quit IRC[12:36:16] *** jmcp_ has joined #opensolaris[12:36:16] *** ChanServ sets mode: +o jmcp_[12:37:07] *** dalibor has quit IRC[12:38:46] *** jmcp_ is now known as jmcp_laptop[12:39:22] *** dalibor has joined #opensolaris[12:39:42] *** thistle_ has quit IRC[12:41:27] *** datadigger has quit IRC[12:44:06] <nikolam> hm, will there be 134b or something available on IPS?[12:44:59] <jmcp_laptop> no[12:45:16] <jmcp_laptop> not unless that turns out to be the final build which is the next OSOL release[12:46:09] <nikolam> jmcp, well, how I can test it then. Since 131 I have X freezes. Is there any way to test?[12:46:28] <nikolam> jmcp_laptop, sry[12:46:47] <jmcp_laptop> #define "it" please[12:46:58] <palowoda> Java is in the same boat no more releases unless you buy a contract.[12:47:26] <nikolam> I am thinking about testing changed 134, newest, before that release[12:47:54] <jmcp_laptop> nikolam: so when 134 hits pkg.opensolaris.org, image-update[12:47:54] *** dajune has joined #opensolaris[12:48:34] <jmcp_laptop> palowoda: you are way off[12:48:56] <nikolam> jmcp_laptop, well, I did, few weeks ago . And still got screen freezes. Is there 134 changing over time, is that what you are trying to tell me?[12:49:13] <jmcp_laptop> 134 is not changing[12:49:22] <jmcp_laptop> 134 is a fork point for The Next Release Of OpenSolaris[12:49:24] <palowoda> jmcp_laptop: Good it might be a starting point.[12:49:47] <jmcp_laptop> ie, there are certain stopper bugs which have been identified since 134 was closed, which need to be pulled into a respun build of 134[12:49:54] <jmcp_laptop> these will become 134a, 134b.....[12:50:10] <nikolam> jmcp_laptop, I know that but I can only use 131 since with newer I have screen freezes and can not test latest[12:50:29] <jmcp_laptop> so I suggest that if you can't wait for whenever The Next Release pokes its head above the parapet, that you keep going on the /dev train[12:50:33] <jmcp_laptop> have you logged bugs, btw?[12:50:57] <nikolam> ah, ok then I can only wait for 134a etc.[12:51:01] <nikolam> yep.[12:51:15] <jmcp_laptop> no, you won't get 134a etc *unless* that becomes The Next Release[12:51:22] *** dalibor has quit IRC[12:52:26] <nikolam> jmcp_laptop, , so no way to say, change intel driver with something newer in 134 and use it and test it.[12:53:06] <jmcp_laptop> I don't know what your constraints are[12:53:12] <nikolam> I am worried release will still have freezes . That I am worrying about.[12:53:51] <jmcp_laptop> (a) you won't know unless you try, (b) we won't know unless you log bugs, (c) the release has not been cut yet[12:54:12] <nikolam> a) did b) did c) I know[12:55:21] *** kevku has joined #opensolaris[12:55:38] <palowoda> Need to talk about something new I guess.[12:56:13] *** thistle_ has joined #opensolaris[13:02:59] *** thistle_ has quit IRC[13:05:00] <sickness> Broadcom BCM4315 <- any hope this wifi card will ever be supported?[13:06:08] <palowoda> No way Broadcom is totally is proprietary/closed source when it comes to Opensolaris.[13:06:22] <sickness> :((([13:07:01] <sickness> this sux, other broadcom wifi cards have open (bsd licensed) drivers, for example under openbsd![13:08:27] <nikolam> sickness, contacted Broadcom?[13:09:35] <sickness> nah, I didn't purchase the card from them, it's inside a compaq mini 110, so I suppose they'd say to contact hp, and hp will say that they support windows only...[13:10:00] <sickness> so my only hope now is to sell this netbook or try to change the internal mini-pci wifi[13:10:18] <sickness> funny thing is that it works in osx, which it isn't even supposed to be installed on this thing, lulz[13:10:21] <sickness> (!!!)[13:11:10] *** datadigger has joined #opensolaris[13:18:07] <palowoda> sickness: I'd sell it. Compaq and HP laptops are known to reject other wifi cards in their bios if you try to replace the wifi card.[13:19:07] <sickness> palowoda: yeah, I'm reading also about this issue on some mini 110 forums...[13:27:17] *** tosh has joined #opensolaris[13:33:05] *** Netwolf has joined #opensolaris[13:34:28] *** ewdafa has joined #opensolaris[13:48:25] *** jamesd2 has joined #opensolaris[13:58:49] *** Odin- has quit IRC[13:59:52] *** TomJ- has joined #opensolaris[14:00:56] *** TomJ has quit IRC[14:04:06] <RoyK> nikolam: if you have X freezes, perhaps porting a bug report can be a start?[14:05:09] <RoyK> palowoda: that's evil![14:08:48] <nikolam> RoyK, did that[14:10:19] *** chendy has quit IRC[14:11:28] *** chendy has joined #opensolaris[14:11:34] <RoyK> nikolam: k[14:16:06] *** chendy has quit IRC[14:21:30] <smyth_rj1> Dell do that with 3rd party HDD's[14:21:40] <smyth_rj1> get an annoying bios error[14:22:13] *** sergiusens has joined #opensolaris[14:28:06] *** chendy has joined #opensolaris[14:30:25] *** javashin has joined #opensolaris[14:33:07] *** Odin- has joined #opensolaris[14:40:54] *** dalibor has joined #opensolaris[14:43:25] *** micols has quit IRC[14:44:12] *** micols has joined #opensolaris[14:46:13] *** jmcp_laptop has quit IRC[14:49:14] *** Yu\2 has quit IRC[15:02:01] *** javashin has quit IRC[15:06:22] *** _matze has joined #opensolaris[15:06:28] <_matze> hi[15:07:42] *** Edgeman has quit IRC[15:08:55] *** biant92 has joined #opensolaris[15:12:30] *** Edgeman has joined #opensolaris[15:12:42] *** eleusis has quit IRC[15:12:45] *** chendy has quit IRC[15:13:38] *** eleusis has joined #opensolaris[15:18:54] *** Yu\2 has joined #opensolaris[15:22:47] *** fOB has quit IRC[15:24:53] *** _matze has quit IRC[15:27:36] *** g4lt-mordant has quit IRC[15:27:39] *** TomJ- has quit IRC[15:29:24] *** zenjuzu has joined #opensolaris[15:31:52] *** RoyK has quit IRC[15:35:58] *** thistle_ has joined #opensolaris[15:46:57] *** g4lt-mordant has joined #opensolaris[15:49:13] *** psychicist has joined #opensolaris[15:52:44] *** Edgeman has quit IRC[15:54:55] <echobinary> AZIZ! LIGHT!![15:57:22] *** raichoo has joined #opensolaris[16:02:06] *** RoyK has joined #opensolaris[16:06:14] <reflect> That brings back memories :)[16:13:58] *** kimc has joined #opensolaris[16:15:37] <kimc> with b134 the audiohd driver is not installed according to the 'Device Driver Utility'[16:16:07] <kimc> whats the best way install that?[16:16:34] *** raichoo has quit IRC[16:19:23] *** yarihm has joined #opensolaris[16:19:36] <Meths> Is Parfait Oracle internal only or can others use it?[16:24:07] *** xapiens has joined #opensolaris[16:24:49] *** iFire has quit IRC[16:25:52] *** kukhuvud has quit IRC[16:25:53] <DerSaidin> Meths: internal, but you may be able to apply for early access[16:26:22] *** Norrin has quit IRC[16:26:29] *** lblume has joined #opensolaris[16:29:55] *** kukhuvud has joined #opensolaris[16:31:12] <kimc> Looks like the problem is not a missing audiohd driver.. its 1414 E Maple Road[16:31:12] <kimc> Troy, MI 48083[16:31:22] <kimc> opps sorry this[16:31:37] *** myosound has joined #opensolaris[16:33:00] <joshua_> ah, that is a standard problem[16:33:45] <kimc> yeh :) the actual problem is: No volume control GStreamer plugins and/or devices found.[16:34:26] <kimc> looking at the output from prtconf -v it appears the driver for the mixer and audio output devices are there[16:34:41] *** kukhuvud has quit IRC[16:35:01] <kimc> dev_link=/dev/sound/audiohd:0[16:35:33] <kimc> i don't know about GStreamer[16:38:08] *** ewdafa has quit IRC[16:39:49] *** ewdafa has joined #opensolaris[16:39:49] *** ewdafa has joined #opensolaris[16:43:01] <kimc> joshua_: is the GStreamer a standard problem?[16:44:35] *** ewdafa has quit IRC[16:46:24] *** fleyta has joined #opensolaris[16:46:30] *** milek_ has quit IRC[16:46:32] <fleyta> jmcp: Are you available?[16:47:35] <joshua_> kimc, I was just making fun of you for claiming that the problem is 1414 E Maple Road ;)[16:47:44] <joshua_> I have no idea what your actual problem is[16:48:50] <kimc> yeh right thats what i thought you were doing but then i thought hmm..[16:54:24] *** leito has quit IRC[16:55:16] *** leito has joined #opensolaris[16:56:20] <fleyta> jmcp: Please read the mail I send you. Thanks.[16:56:28] *** fleyta has quit IRC[16:56:46] *** dtech has joined #opensolaris[16:57:57] <RoyK> omg. someone on the zfs-discuss list got a replacement drive that was a few sectors smaller than the others in the pool, so instead of returning the drive and asking for a better one, he keeps asking on how to use slices to make the pool less vulnerable to such problems ........ I wish I had all the spare time this guy must have[17:00:47] *** eleusis has quit IRC[17:00:59] *** dtech has quit IRC[17:01:19] *** eleusis has joined #opensolaris[17:02:41] *** unhygienix has joined #opensolaris[17:06:28] <crazed> what man page has information about doing nfs+kerberos on opensolaris?[17:07:00] <tsoome> kerb5.conf[17:07:10] <tsoome> then share_nfs[17:07:25] <crazed> already have kerb setup just need to get nfs working with it[17:07:32] <tsoome> depends if you are kerberos client or need to set server as well[17:07:36] *** GhostInTheShell has quit IRC[17:07:40] *** GhostInTheShell has joined #opensolaris[17:07:56] <tsoome> you need at least host principal, but i think nfs as well[17:08:11] *** lattera_ has joined #opensolaris[17:11:35] <crazed> ok those can be done, and i see the sec= option for nfs but no mention of kerberos hm[17:12:07] <crazed> ah there's nfssec[17:12:26] <tsoome> its done with gss[17:13:22] *** Odin- has quit IRC[17:14:47] *** delewis has left #opensolaris[17:15:31] *** dtech has joined #opensolaris[17:15:50] <dnaumov> RoyK: he has a point, harddrive manufacturers make no guarantees on amount of sectors[17:17:37] <RoyK> dnaumov: but iirc it's a sun system, and sun has given him unsusable drives for that, which is somehow bad[17:18:08] <RoyK> newer builds (since 117) allow for using slightly smaller drives for replacements in a pool[17:24:08] *** g4lt-mordant has quit IRC[17:24:10] *** milek_ has joined #opensolaris[17:26:49] <lblume> Sun does guarantee the number of sectors in ther HD. That is why they are so expensive.[17:28:43] *** g4lt-mordant has joined #opensolaris[17:31:39] <tsoome> well thats done by remapping[17:32:29] <tsoome> in vtoc lable. but i guess you are on your own if zfs will use efi on them...[17:34:19] *** dtech has quit IRC[17:34:25] <RoyK> tsoome: this message on zfs-discuss was about getting a _smaller_ drive - I wonder how you can remap to something larger.... :)[17:34:39] <tsoome> you wont[17:35:39] *** milek_ has quit IRC[17:38:41] <tsoome> they are mapping to smallest. old sun 18GB disks were in fact 16GB[17:39:47] *** lblume has quit IRC[17:41:17] *** lblume has joined #opensolaris[17:45:10] *** RoyK has quit IRC[17:49:02] *** lateau has joined #opensolaris[17:51:54] *** Mousk4r has joined #opensolaris[17:52:48] *** gerard13 has quit IRC[17:55:08] *** kimc has quit IRC[17:58:38] *** RoyK has joined #opensolaris[18:10:14] *** dtech has joined #opensolaris[18:10:45] *** dtech is now known as kukhuvud[18:10:58] *** cocacola has quit IRC[18:11:26] <RoyK> hi. zfs get compressratio somepool reports a number - is this for somepool itself or somepool including all 'child pools'?[18:11:49] <RoyK> eg including somepool/backup etc[18:11:57] <RoyK> erm[18:12:22] <RoyK> wrong - does it report compressratio for all _datasets_?[18:12:24] <RoyK> not pool[18:12:26] <RoyK> my bad[18:13:01] *** nikolam_ has joined #opensolaris[18:15:15] *** raichoo has joined #opensolaris[18:15:50] *** nikolam has quit IRC[18:16:05] *** myosound has quit IRC[18:19:28] <BankerBarney> like I started with 2TB in diskspace...created my pool/datasets allotted 180G when I created the pool..then after the pool it said it was 1.8T, but when I did a zfs list it only showed 1.37 available. So I was wondering where my other space went. Someone told me it had something to do with "overhead". 500gb is pretty much not showing up..[18:20:33] *** nage_ has joined #opensolaris[18:21:47] <tsoome> zfs list -t all ?[18:22:09] <BankerBarney> raidz[18:22:38] <tsoome> raidz with 500gb drives?[18:22:55] <tsoome> 3+1 setup?[18:23:29] <tsoome> what you think, where the parity is stored?:P[18:23:30] <BankerBarney> http://pastebin.com/xX90n261[18:23:40] <BankerBarney> yes tsoome, I have been copying stuff over[18:23:55] <BankerBarney> 4x500GB drives, 1x120[18:24:15] <BankerBarney> rpool is on the 120[18:24:28] <BankerBarney> shiloh datasets are on the 4x500[18:25:02] *** nachox has joined #opensolaris[18:26:18] <BankerBarney> compression is enabled on the pool and datasets[18:26:31] <BankerBarney> somewhere I am missing 500gb of space[18:26:58] <CosmicDJ> just to get this right, you've added 4x 500GB disks to a raidz pool and now you're confused that the pool is not 2TB in size?[18:27:10] *** unhygienix has quit IRC[18:27:11] <BankerBarney> it sais I have copied over 465gb, but only 374 is showing up on the shiloh pool[18:27:28] <BankerBarney> correct CosmicDJ[18:27:38] <CosmicDJ> 19:27 < tsoome> what you think, where the parity is stored?:P[18:27:47] <BankerBarney> I am curious why 2TB is physically available, and only 1.37 is being reported[18:28:40] <crazed> hm.. how do i get osol to use kerb for the password store? i've followed examples in man pam_kerb5 but seems that it is using the passwords stored in ldap instead[18:29:29] *** andersenep has joined #opensolaris[18:29:46] <tsoome> with raidz and 4x500GB you have effective usable space 1.5TB, with raidz2 and same disks , you have usable space 1TB[18:30:28] <BankerBarney> ok..so what is zfs doing with 500gb of space if it is not being used?[18:31:03] <BankerBarney> I lost 500gb of physical hard disk space..but where did it go..for what cause is it being used for if not usable space?[18:31:23] <tsoome> if you set up mirror with 2x500GB disks, you get usable space 500GB, what do you think, where is another 500GB going?:P[18:31:25] *** Mousk4r is now known as LinuxBMW[18:31:30] *** unhygienix has joined #opensolaris[18:32:15] <BankerBarney> its all being used of course[18:32:25] <tsoome> for what?[18:33:01] <BankerBarney> usable space?[18:33:09] <BankerBarney> im not quite sure how mirrors work[18:33:13] <Meths> BankerBarney: You need to read up on the technologies you are using. http://en.wikipedia.org/wiki/RAID and http://en.wikipedia.org/wiki/ZFS should answer your questions.[18:33:24] <tsoome> with morror, half of your disks are used for copy of your data. with raidz (which is basically raid5), you loose 1 disk for parity[18:33:38] <tsoome> with raidz2 you loose 2 disks for parity[18:33:57] <BankerBarney> oh I see...so its taking a picture of my data for backup purposes[18:34:06] <BankerBarney> to self heal and what not[18:34:09] <BankerBarney> i gotcha[18:34:14] <tsoome> :)[18:34:24] <BankerBarney> prett slick[18:34:31] <BankerBarney> boy I hope this usb expansion card works[18:34:45] <BankerBarney> i need more hd space...[18:34:52] <BankerBarney> for snapshots and what not[18:35:16] <tsoome> but the whole point is, if you have, say, 8 disks, you make raidz, it makes 7+1 setup, 7 disks for data, 1 for parity.[18:35:28] <BankerBarney> [12:37] <Meths> BankerBarney: You need to read up on the technologies you are using. http://en.wikipedia.org/wiki/RAID and http://en.wikipedia.org/wiki/ZFS should answer your questions. <---True...I will do that..[18:35:51] *** PhlowX_ has quit IRC[18:35:57] <BankerBarney> ok so in this scenario 1 disk is setup for parity[18:36:01] *** jamesd_laptop has joined #opensolaris[18:36:21] <BankerBarney> if a hard disk goes bad with zfs does it shift data to good parts of the disk[18:36:24] <BankerBarney> or the pool rather[18:36:25] <tsoome> in fact, the parity is spread over all disks[18:36:51] *** jamesd2 has quit IRC[18:36:51] <BankerBarney> at least try if enough parity is avail..[18:37:36] *** yarihm has quit IRC[18:37:47] <Meths> BankerBarney: no, the point is it doesn't have to shift data. It runs in degraded mode until you replace the bad hard drive. If another disk goes bad before you replace the first, you test your backup process.[18:38:07] <BankerBarney> ok[18:40:59] *** dalibor has quit IRC[18:43:11] <BankerBarney> i can see why zfs would be important in a corporate environment[18:43:25] <BankerBarney> the "craze"[18:43:46] <duckinator> uh..wtf...i'm getting endless keyring requests from pidgin.. does anyone know what that's about? :\ i keep putting the password and clicking ok, and it prompts again[18:44:01] <duckinator> after about 10 times it changed from "The application 'Application'..." to "The application 'Pidgin'..." in the message[18:44:01] <BankerBarney> for personal use, it does not make much sense unless you are just trying to learn how it works[18:44:27] *** unhygienix has quit IRC[18:45:01] <BankerBarney> i dunno tho some people have important personal info..so it may have its usefulness :P[18:45:09] <BankerBarney> pRon is important to some people[18:45:41] <BankerBarney> important is uhh "speculative"[18:46:49] *** ivo_ has joined #opensolaris[18:47:12] *** rune- has joined #opensolaris[18:47:28] *** nikolam__ has joined #opensolaris[18:47:48] *** Rune_ has quit IRC[18:48:02] *** unhygienix has joined #opensolaris[18:48:23] <BankerBarney> but would it not it take up less space if you just took a screenshot of your information instead of dedicating 500gb towards it...something to think about for advanced zfs technology ;P[18:48:36] *** vdi_ has joined #opensolaris[18:48:38] <BankerBarney> lolz[18:48:47] <BankerBarney> newbz[18:48:57] <vdi_> help needed with this error "PHP 5.x CLI Must Be Installed To Run The Phoronix Test Suite. Only PHP For The Command Line Is Needed, Not A Complete PHP Web Server Installation."[18:49:09] <vdi_> currently new to opensolaris[18:49:34] <vdi_> can someone please teach me how to install the php 5.x cli?[18:49:48] <duckinator> how can i disable gnome-keyring? it's doing nothing but trip over itself, as it always does[18:50:31] <RoyK> vdi_: pkg search php is a start[18:50:34] *** nikolam_ has quit IRC[18:51:15] <vdi_> currently no idea on how to install the php-cli or php5-cli[18:51:27] <vdi_> can someone please guide me on this?[18:51:36] <vdi_> help would be greatly appreciated[18:52:02] * RoyK repeats[18:52:04] <RoyK> vdi_: pkg search php is a start[18:53:43] <duckinator> wow, really -.-[18:53:56] <duckinator> i just uninstalled gnome-keyring, and it prompted me for the default keyring password[18:56:34] *** gjl has joined #opensolaris[18:59:06] *** jdparker_ has quit IRC[19:00:23] <duckinator> ...heh[19:00:41] <duckinator> `rm ~/.gnome/keyrings/default*` and i set everything up again. sure am glad i'd only saved 4 passwords :P[19:01:25] *** Aria has joined #opensolaris[19:03:19] *** vdi_ has quit IRC[19:04:38] *** lblume has quit IRC[19:06:35] *** ivo_ has quit IRC[19:06:36] *** lblume has joined #opensolaris[19:07:32] *** cocacola has joined #opensolaris[19:07:45] *** cocacola has left #opensolaris[19:07:50] *** syamajala has joined #opensolaris[19:08:31] <BankerBarney> There is also RAID-Z2 which doubles the parity structure to achieve results similar to RAID 6: the ability to sustain up to two drive failures without losing data.[5] In July 2009, triple-parity RAID was added to OpenSolaris[19:08:40] <BankerBarney> interesting..2 drive failures..[19:09:14] <Aria> They happen[19:09:18] <Aria> With alarming frequency.[19:09:24] <BankerBarney> indeed[19:09:30] <Aria> Especially when your drives are all the same age.[19:09:36] <Aria> And all on the same power feed.[19:10:01] *** kimc has joined #opensolaris[19:10:55] <BankerBarney> so each step up in this technology you sacrifice a bit more hd space to support your data..not so bad really when hd space is so cheap[19:11:22] <taemun> its also because as hard drives get bigger, the time to resilver gets longer[19:11:29] <taemun> ie, its more likely that another will die[19:11:55] <taemun> another reasoning for raidz3 was that hard drives have loads of space set aside on-platter for error correcton etc[19:12:01] *** ozux has joined #opensolaris[19:12:02] <BankerBarney> yea..this technology has to advance when hd space gets larger per disk..it is kina a hand in hand thing[19:12:19] <BankerBarney> more data on a disk.."more risk"[19:12:27] <taemun> if you had custom firmware, you could theoretically have the full (or more) disk capacity available, with a higher bit-error rate[19:12:33] <taemun> and have raidz3 fix those media errors[19:12:39] *** derchris has joined #opensolaris[19:12:43] <Aria> I've heard that theory advanced before.[19:12:51] <Aria> I'd really love to see how it works out in practice.[19:12:58] <taemun> iirc, the error recovery accounts for 30+% of space used[19:13:10] <Aria> Maybe even let the OS see the RS codes.[19:13:22] <taemun> on my hitachi 2tb drives, they quote a media transfer rate above 200MB/s and only around 130MB/s to the external world[19:13:49] <tsoome> also remember that every parity addet up will reduce perfomance, as disk capacity grows but perfomance does not...[19:13:51] <taemun> 4k sectors are a step in the other direction I guess[19:14:05] <tsoome> added*[19:14:38] <tsoome> which basically means, you need more ram to keep up with perfomance[19:16:04] <RoyK> more ram, why?[19:19:58] *** Mousk4r1 has joined #opensolaris[19:21:03] <BankerBarney> forget the ram...technology should just use something like swap[19:21:09] <BankerBarney> but not swap..[19:21:26] <BankerBarney> maybe a separate hard disk for cache[19:21:49] <RoyK> a mirrored pair of SSDs is nice[19:21:51] <taemun> .... l2arc?[19:22:12] <RoyK> no, let's all go back to ext3 or ufs[19:22:23] <RoyK> zfs is just bullshit[19:22:30] *** LinuxBMW has quit IRC[19:22:31] <taemun> ah huh[19:22:41] *** biant92 has quit IRC[19:22:42] <BankerBarney> well...while hd space is so cheap..and you are right there anyway...why reach to another part of the mobo for memory when you could just sacrifice a bit per pool for ram[19:22:45] *** medar has joined #opensolaris[19:22:57] <BankerBarney> just an idea[19:23:34] <BankerBarney> zfs2[19:23:35] <taemun> I'm not sure you appreciate the performance difference between RAM <=> CPU and SATA[19:23:41] <BankerBarney> or 3 or something like that[19:23:44] <taemun> let alone real world SATA implementations[19:24:03] <BankerBarney> well the technology throughput would have to change[19:24:04] <taemun> what would you change in zfs that requires a complete re-qrite[19:24:06] <taemun> ?[19:24:08] <BankerBarney> but it is anyway so..[19:24:17] <BankerBarney> at least for hard disks[19:24:42] <RoyK> BankerBarney: using zfs instead of getting more ram is stupid - you want to speed things up or slow them down?[19:25:11] <BankerBarney> im saying use the hard disk for extra cache[19:25:23] <RoyK> don't![19:25:23] <taemun> BankerBarney: I'm saying its called L2ARC[19:25:26] <BankerBarney> leave the ram alone[19:25:27] <Aria> Cache of what, exactly?[19:25:36] <taemun> it exists[19:25:36] <taemun> today[19:25:37] <BankerBarney> thats my take on it[19:25:38] <taemun> in ZFS[19:25:49] <BankerBarney> well..this technology would act like ram[19:25:57] <BankerBarney> im not saying it exists yet[19:26:01] <tsoome> ram accesss speed is in nanoseconds, disk access speed is in milliseconds.[19:26:10] <RoyK> BankerBarney: say you have 10 drives in a raidz2, then, adding a single drive for cache would be limiting the access time quite a bit[19:26:16] <tsoome> there is a bit of difference:P[19:26:19] <BankerBarney> that is what I just got finished saying[19:26:26] <BankerBarney> throughput to the drives would have to change[19:26:28] * Aria laughs.[19:26:35] <Aria> I think we call this technology ... RAM[19:26:47] <taemun> BankerBarney: go read http://blogs.sun.com/brendan/entry/test[19:26:50] <RoyK> and there's this new thing called SSDs[19:26:57] <taemun> RAM technology is also improving[19:27:24] *** Aria has quit IRC[19:27:50] *** plompi has joined #opensolaris[19:27:50] <RoyK> unless someone tries to make a RAM disk with some chips with an SSD drive as backup and places it on a 1x PCI-ex bus[19:28:00] <taemun> RoyK: its been done[19:28:15] <taemun> and a "super" capacitor for the event of power failure[19:28:38] <RoyK> I just saw one of those - 200MB/s throughput - lots of IOPS, sure, but still, 200MB/s to a RAM-based drive?[19:28:43] *** plompi has left #opensolaris[19:28:58] <BankerBarney> in my scenario you would not pull from the hard disk for major processes for ram[19:29:02] <taemun> err why a 1x PCIe bus?[19:29:04] <BankerBarney> you would pull from RAM for that[19:29:22] <taemun> PCIe 1.1 1x is only 250MB/s sync[19:29:25] <BankerBarney> for smaller processes you could pull from the HD[19:29:37] <RoyK> taemun: I didn't make it[19:29:38] <taemun> 200MB seems a reasonable throughput given that no implementation is perfect[19:29:51] <taemun> there are ones around that are 16x and PCIe 2.0, in any case[19:30:15] <BankerBarney> at least until throughput to the drives got better[19:30:20] <BankerBarney> anyway..[19:31:18] <RoyK> BankerBarney: drive throughput has been more or less stable for quite some years[19:31:19] <taemun> BankerBarney: if they are smaller processes, they aren't going to use ("waste") much ram[19:31:25] <taemun> so its not really worth it[19:33:24] <BankerBarney> [13:35] <RoyK> BankerBarney: drive throughput has been more or less stable for quite some years <--- Yea I am definitely not suggesting my scenario with the current hardware available[19:34:46] <BankerBarney> ;)[19:35:13] <BankerBarney> however in solaris world devlopment builds are ass backwards..they are more stable then the stable[19:35:20] <BankerBarney> so maybe I am incorrect..here[19:36:24] <gosx> taemun: ZIL doesn't need bandwidth[19:36:40] <jamesd_laptop> solaris for the last year has been having major issues because they are walking on egg shells, and cutting costs and cutting people especially far too many in QA...[19:36:56] <taemun> gosx: he's talking about mapping process's memory space directly to hard drives[19:37:09] <gosx> oh, didn't scroll back far enough[19:37:15] <taemun> np[19:37:31] *** pjfloyd has joined #opensolaris[19:39:01] *** ivo_ has joined #opensolaris[19:39:07] <BankerBarney> people here are like..wtf..??[19:41:59] <taemun> yes, its a nonsensical idea[19:42:19] <RoyK> BankerBarney: just use an SSD for L2ARC and another for ZIL and you'll be quite happy[19:42:41] <taemun> RoyK: again, he's talking about mapping program memory to a hard drive, directly[19:42:44] <BankerBarney> RAM will probably always be faster, but there can be much more of it and we can do many more things if we pull from the HD for maybe most of smaller processes[19:42:59] <BankerBarney> just an idea[19:43:30] <BankerBarney> is arc and l2arc zfs specific..[19:43:37] <taemun> yes[19:43:41] <RoyK> taemun: heh - that's what they do with virtual memory, really[19:43:42] <BankerBarney> wow! nice[19:44:04] <RoyK> map a truckload of gigs to memory, but use ram first[19:44:07] <taemun> you're kinda in a channel where we assume that you're talking about relevant things :P[19:44:19] <taemun> anyway[19:44:23] <taemun> its 03:48:30 ... sleep[19:44:50] <RoyK> BankerBarney: have you osol installed on a dedicated box?[19:44:50] <gosx> if you did a large MMIO file, would L2ARC cache it?[19:44:57] <BankerBarney> yea RoyK[19:45:26] <RoyK> try getting a cheap ssd for l2arc[19:45:32] <BankerBarney> k[19:45:35] <RoyK> don't use a cheap ssd for zil, though[19:45:49] <ivo_> the 900GB encrypted file used as a filesystem over lofiadm is working almoust fine[19:46:02] <RoyK> almost?[19:46:06] <ivo_> if I write or read from multiple programs[19:46:07] <tomww> or two of them to get them mirrored if it's a little more then plain testing[19:46:10] <ivo_> it is damn slow[19:46:16] <ivo_> and the media is not fluent[19:46:38] <RoyK> tomww: yeah, for zil, a mirror is needed[19:46:47] <RoyK> for l2arc, not really needed[19:47:02] <RoyK> a read failure will just make zfs read from disk[19:47:55] <BankerBarney> wow! ssd's are not cheap[19:47:57] *** datadigger has quit IRC[19:48:17] <BankerBarney> 64gb for 257 bux free shipping[19:48:47] <BankerBarney> 259bux rather[19:48:56] <RoyK> you may not need that much[19:49:15] <RoyK> also, for l2arc, cheap SSDs will suffice[19:49:30] <RoyK> keep in mind how much data is actually used[19:49:34] <BankerBarney> maybe half that[19:49:42] *** datadigger has joined #opensolaris[19:49:46] <BankerBarney> but yea..thats what I am talking about[19:50:01] <RoyK> are you accessing 32 gigs regularly?[19:50:08] <BankerBarney> no[19:50:11] <bball> can I use BFU to updatea be that I created with beadm ?[19:50:24] <BankerBarney> but...if..I slapped my hd space on this..it will eat more ram[19:50:34] <BankerBarney> so I may have to go with ssd[19:50:46] <BankerBarney> anyway...[19:50:55] <RoyK> bball: what's bfu?[19:51:19] *** lateau has quit IRC[19:51:40] <BankerBarney> zfs is a memory hog[19:51:45] <RoyK> it is[19:51:48] <tomww> bfu is used to update a development build (ony core files) to the latest available version[19:51:50] <RoyK> of good reason[19:51:55] <bball> blindingly fast update :-) some tool to install ON user/kernel bits from your own ON build tree[19:52:16] <BankerBarney> i can get ssd in ide interface too[19:52:19] <alanc> for a few more builds until the final removal of bfu once everyone has converted to IPS[19:52:28] <tomww> explain bfu[19:52:32] <tomww> smrt: explain bfu[19:52:33] <smrt> Blindingly Fast Upgrade a.k.a. Bonwick/Faulkner Upgrade a.k.a. Bad For Users aka Brick Factory Upgrade. This is a way to upgrade the subset of a system's binaries that are delivered by the ON consolidation that uses cpio(1) archives instead of packages to improve speed. You do not want to do this. ESPECIALLY if you like pkg(5).[19:53:10] <bball> from the devref it's one of 2 options[19:53:39] <alanc> but if you do use bfu, creating a new BE for it and doing the bfu to it is recommended by many ON developers to allow easy rollback if something goes wrong - I don't have any links handy to instructions though[19:54:33] *** ivo_ has quit IRC[19:55:27] <bball> alanc: OK.. I'll play around with it.. just wondering if it made sense.. sounds like other people do it[19:56:23] <CIA-21> Roger A. Faulkner <Roger.Faulkner at Oracle dot COM>: 6939349 RFE: Update ksh93 to ast-ksh.2010-03-09, 6877392 ksh93 regresses 'uniq -c' performance, 6887363 Korn shell 93 sometimes mishandles return value of its child process, 6900314 (while true ; do true|true ; done) hang in ioctl() with SIGTTOU, 6904557 wc no longer counts number of bytes correctly, 6904575 cut -d with multibyte character no longer works, 6904597 paste -d no longer works with multibyte chara[19:56:41] <alanc> that should make fleyta happy[19:57:03] <gosx> BankerBarney: we have a file server with about 600GB of active content, but our L2ARC only fills up with the 5GB of most-hit stuff[19:57:20] <bball> alanc: will moving to IPS fix some of the drawbacks of BFU ?[19:57:24] <gosx> the stuff that gets hit every day is in L2ARC, our miss rate is like 0.001%[19:57:28] <BankerBarney> wow gosx[19:57:37] <BankerBarney> thats a lot less than I figured[19:57:56] <BankerBarney> so I could most likely get by with 32gb[19:58:14] <gosx> we use a 32GB X25-E for L2ARC, and don't use a ZIL[19:58:35] <alanc> bball: yes - it will keep the packaging system in sync and automatically do the new be for you, won't require conflict resolution and other workarounds[19:58:39] <BankerBarney> thats good too cuz I can just go with a small 2.5" drive for it...velcro it to the side of the case[19:58:46] <BankerBarney> ide[19:58:55] <gosx> i wouldn't use IDE if you can avoid it[19:59:04] <gosx> IDE doesn't have any data integrity on the command interface[19:59:12] <BankerBarney> i cant..I dont have any more pci slots[19:59:16] <BankerBarney> for sata cards[19:59:21] <BankerBarney> this is an old box[19:59:26] <gosx> oh[19:59:35] <RoyK> heh - 32bit with zfs?[20:00:03] <bball> alanc: does it allow easy rollback in case your ON build won't boot :-)[20:00:05] <RoyK> http://www.solarisinternals.com/wiki/index.php/ZFS_Best_Practices_Guide#System.2FMemory.2FSwap_Space <-- see first line[20:00:10] <BankerBarney> what I could do is buy an isa nic and free up a slot I guess[20:00:16] <BankerBarney> put a sata card in it[20:00:27] <RoyK> Burana: http://www.solarisinternals.com/wiki/index.php/ZFS_Best_Practices_Guide#System.2FMemory.2FSwap_Space <-- see first line[20:00:27] <gosx> ISA?[20:00:30] <BankerBarney> were talking old technology here[20:00:33] <gosx> holy shit[20:00:36] <gosx> thats old[20:00:36] <BankerBarney> lolz[20:00:39] <BankerBarney> yea...[20:00:46] <BankerBarney> but nix is good on old hardware[20:00:56] <gosx> IMO, you're crazy to spend $$$ on an SSD to accelerate older hardware[20:01:03] <RoyK> zfs isn't designed to be used on museum hardware[20:01:18] <gosx> you'd be better off with a modern 64-bit Atom CPU system[20:01:31] *** russiane39 has quit IRC[20:01:32] <RoyK> which doesn't cost much......[20:01:40] * BankerBarney cranks his pc up with a lawn mower string[20:02:02] <gosx> i built an atom box a few weeks ago as a router, and the NIC cost as much as the mobo+cpu+RAM+case[20:02:07] <BankerBarney> sorry bout that it died on me[20:03:00] <BankerBarney> thats what I am saying gosx..why not just go with ide[20:03:08] <BankerBarney> yaknow?[20:05:38] <jamesd_laptop> really for home you are better off with amd, the new cheap phenom sound interesting... and can give you virtual expension to do vbox or xen for other tasks to use spare cpu on your fileserver[20:06:44] <gosx> BankerBarney: you're spending current dollars on improving an old system and that seems silly to me[20:07:00] <gosx> just buy the cheapest 64-bit CPU and PCI-e motherboard you can[20:07:47] <BankerBarney> not really..i spent past dollars on new hardware and I kept it around to play with nix on...I would not be out much money for anything new really...plus I can carry what I buy now over to something newer anyway..[20:07:59] <gosx> but IDE is deprecated[20:08:21] <BankerBarney> this machine I have I can turn into a server and it is ide/sata[20:08:24] <BankerBarney> best of both worlds[20:08:34] <jamesd_laptop> ide wont die for a while because many machines use ide for cdrom/dvd so you are safe for a while... but sata makes more sense..[20:08:36] <BankerBarney> I can upgrade this to something newer and carry that garbage over..[20:08:44] <BankerBarney> but I was talking about going sata anyway[20:09:01] <BankerBarney> and not ide...just spend a few bux on a isa nic and get a sata card[20:09:20] <gosx> depending on where you live, the older hardware will probably cost you more $$$ in electricity than you'd spend on newer hardware =P[20:09:47] <BankerBarney> those new swirly lightbulbs help..I dont live in california[20:09:52] <jamesd_laptop> intel nics are the best option in my opionion... but having working on board gigabit is a win.. and more dimms and max supported ram is good best is support for more than 8GB of ram[20:10:44] <g4lt-mordant> gosx, actually, newer hardware usually increases your power bill[20:10:54] <gosx> g4lt-mordant: not for me[20:11:04] <gosx> my i7-860 sips power as a desktop box[20:11:18] *** javashin has joined #opensolaris[20:12:02] <g4lt-mordant> think of it, the original IBM AT had a 130W PSU. can you imagine how much you'd have to not use to keep your consumption down to 130W on a modern peecee?[20:12:31] <g4lt-mordant> they make 2 and 3 kW PSUs now FFS[20:12:35] *** syamajala has quit IRC[20:12:45] *** nage_ has quit IRC[20:13:00] *** sactodave has quit IRC[20:13:09] *** CodeWar has joined #opensolaris[20:13:19] <gosx> being 3KW capable doesn't mean that you're sucking 3KW all the time though[20:13:33] <RoyK> g4lt-mordant: most PCs don't require more than 300-400 watts[20:13:51] <g4lt-mordant> RoyK, still twice the MAX power of the AT[20:14:17] <jamesd_laptop> for the home market the max is about 1500, since you need more than a standard outlet to power more than 1525watt psu[20:14:17] <gosx> g4lt-mordant: ok yes, for a 20-yr-old PC today's stuff generally consumes more power[20:14:30] <gosx> but it does 10,000 more[20:14:32] <RoyK> that is, most won't require even that, but then, a modern CPU uses 100W where the old 80286 used something like 10W or less[20:15:04] <RoyK> g4lt-mordant: would you like my old 80286 machine?[20:15:24] <g4lt-mordant> ...yet you tell me that newer hardare uses less power. news flash, the numbers don't add up[20:15:26] <RoyK> I doubt you can even run the day of the tentacle on it[20:15:38] <jamesd_laptop> i have a 80386sx-25 ... why waste the power for a floatingpoint unit...[20:15:48] <RoyK> :D[20:16:13] <CodeWar> processors and hw _is_ getting better at power savings actually .. both idle power and peak power[20:17:00] <RoyK> a 12-core AMD processor uses some 120 watts, well, ok, but it delivers a fair amount of processing power for that[20:17:05] <g4lt-mordant> CodeWar, well, given its long track record of getting worse, I submit taht the reason is "even a broken clock is right twice a day"[20:17:21] <RoyK> I think the old Pentium 60MHz used 60 watts[20:17:49] <CodeWar> RoyK, the 12 core AMD s this is the idle power rating or what exactly[20:18:02] <RoyK> CodeWar: full use[20:18:16] <bball> if I boot from an alternate BE will that become my root file system ?[20:18:23] <RoyK> yes[20:18:49] <RoyK> so a changed password won't be visible in there[20:18:51] <CodeWar> well to be fair you want to include their idle power usage .. i m sure they ll use different pstates/c-states / deep c states block gating and what not[20:19:13] <bball> ok.. so if use BFU to update the BE I'm currently booted it and that bricks that BE I can always boot back into another BE ?[20:20:15] <CodeWar> reminds me whatever happened to project tesla .. done adn incorporated already in OSOL?[20:21:42] <CosmicDJ> CodeWar: IIRC the kernel is not tickless, yet; but a bunch of code was comitted[20:21:51] *** eviljames has quit IRC[20:22:00] <CosmicDJ> committed[20:22:34] <CodeWar> 2009.06 didnt have deep cstate support hopefully thats gotten better .. there are two tricks here[20:22:56] <CodeWar> make the best out of idle states .reduce leakage.... and <b> force sleep despite there being work present[20:23:29] <CosmicDJ> http://www.c0t0d0s0.org/archives/6111-Tickless.html[20:25:43] <CodeWar> actually for laptops I m not sure tickless is sufficient .. windows and mac both do far better here..[20:26:23] <CodeWar> essentially the idea is to turn off the block (not low c-state ) and not let it wake up even if work comes for an interval .. aka batch work and sleep for longer[20:27:32] <lewellyn> CodeWar: pm-discuss has the chatter for that i think[20:32:35] * tomww is playing around with pulseaudio to get audacity make sound in a SunRay environment eventually[20:33:42] *** nage_ has joined #opensolaris[20:33:52] * lewellyn cries at the thought of pulseaudio on solaris[20:34:00] <bball> how do I add dirs to my MANPATH without getting rid of the default dirs searched by man[20:34:06] *** Yu\2 has quit IRC[20:34:43] <Meths> bball: add them in .profile or "export MANPATH=$MANPATH:/new/dir"[20:34:56] <bball> that's what I'm doing now[20:35:04] <bball> when I login MANPATH is empty[20:35:20] <bball> but it finds all the default man page entires (e.g. ls)[20:35:26] <Meths> what is your manpath line?[20:35:34] <bball> once I set MANPATH I only get stuff that's in MANPATH[20:36:01] <bball> export MANPATH=$MANPATH:/opt/SUNWspro/man:/opt/onbld/man[20:36:43] <bball> that's what is in my .profile[20:36:48] <Meths> You need to include the default locations. When your .profile is read MANPATH hasn't been created[20:37:09] <bball> but if I leave MANPATH empty it already searches the default dirs[20:37:17] <Meths> The default is /usr/share/man:/usr/X11/share/man:/usr/gnu/share/man include that in your line[20:37:29] *** ewdafa has joined #opensolaris[20:37:43] <bball> it seems like there must be a way to list the current default dirs[20:37:46] <Meths> bball: If you leave it empty it uses $PATH to extrapolate man directories[20:37:55] <bball> ahh[20:40:00] <Meths> Is there anyone here from the KDE project?[20:40:24] * RoyK hints slightly about #kde[20:40:54] *** sndcrb has joined #opensolaris[20:41:19] *** TomJ has joined #opensolaris[20:41:36] <Meths> RoyK: Sorry, I mean KDE on OSol, not base KDE. The guys who populate the BionicButton repo.[20:41:49] <Meths> BionicMutton*[20:42:27] <RoyK> k[20:42:47] * RoyK just uses gnome if in a gui on osol, meaning only on test machines[20:42:59] <RoyK> wtf wants a gui for a server?[20:43:08] <lewellyn> Meths: #kde4-solaris or somesuch[20:43:48] <Meths> Ah, thanks.[20:48:52] *** datadigger has quit IRC[20:49:32] *** datadigger has joined #opensolaris[20:49:33] *** RoyK has quit IRC[20:52:36] *** sndcrb has quit IRC[20:56:11] <CIA-21> meem <Peter.Memishian at Sun dot COM>: 6901161 IP Datapath Refactoring ipmp.c nits, 6913726 shutting down an exclusive stack generates noise, 6934331 SIOCSLIFMTU does not refresh the IPMP interface MTU[20:59:08] <tomww> lewellyn: choose your poison. wait on a oss style audio on sunray, not running you application, or put pulseaudio inbetween and run you app on SunRay[20:59:12] *** m171c0 has joined #opensolaris[20:59:25] <m171c0> Brasilian people must be happy!!!Eduardo Kislanski and Franklin Ronald are making a new Solaris Portal with a lot of articles, news, forums, tips and tricks.As soon as possible, the link of the Portal www.openyoursource.com[20:59:38] *** nanase has quit IRC[20:59:44] *** Yu\2 has joined #opensolaris[21:02:28] *** zenjuzu has quit IRC[21:04:20] *** sndcrb has joined #opensolaris[21:04:30] *** communicator has joined #opensolaris[21:04:31] *** jollyd has joined #opensolaris[21:04:42] *** jollyd has left #opensolaris[21:05:59] *** ma3x has joined #opensolaris[21:06:05] <reflect> isn't it about time that pulseaudio was fixed, though? it's been around for long enough to find the bugs.. it must be better with the next version, right? :)[21:07:17] <tomww> well, pulseaudio is a bit difficult on (Open)Solaris. I trief 0.9.15 to compile (compiled, but didn't test it),[21:07:35] <BankerBarney> www.openyoursource.com <-- in english[21:08:01] <tomww> then 0.9.16 needed newer ltdl, then I defaulted to my *old* SFEpulseaudio.spec 0.9.5 because that has to suffice my test on SunRay.[21:08:07] *** sponix has joined #opensolaris[21:08:31] <tomww> the so well now upgrade-hell-because on software needs another software with a minimum revision.[21:09:07] <tomww> something I tried to get rid of by using (Open)Solaris.[21:09:14] <reflect> tomww: we used sunrays about 10 years ago.. then they required a bit of bandwidth, as I understand they've overcome.. how does it work for you?[21:09:33] <tomww> what is overcome? SunRay? :-)[21:09:47] <reflect> the bandwidth[21:10:17] <tomww> ah well. bandwith with WAN usage, there is some improvement.[21:10:46] <reflect> WAN was out of the question when we tried it[21:11:02] *** swa_work has joined #opensolaris[21:11:07] <tomww> if you want video plaing in a LAN, then easiy you hit the 100Mbit network limit. playing vlc / mplayer with a window 300x200 pixels or so works fine.[21:11:19] *** sndcrb has quit IRC[21:11:32] * tomww sorry for the typos :-)[21:11:39] <reflect> don't worry :)[21:11:57] *** javashin has quit IRC[21:12:15] <tomww> the nex gen SunRay uses gigabit network, so I would expect help for large screens and/or better multimedia playback.[21:12:49] *** dalibor has joined #opensolaris[21:12:57] *** eviljames has joined #opensolaris[21:13:21] <tomww> my family uses sunray, so only one big box consuming power (SunRay 1)[21:13:31] *** sndcrb has joined #opensolaris[21:13:57] <tomww> and now I want audacity (but this doesn't play with SunAudio, therefore I try putting pulseaudio in between)[21:15:06] <tomww> reflect: I think most important improvement is with SunRay, that the servers got multiple times faster CPUs (and there is firefox and gnome today ...)[21:16:20] <reflect> iirc, you can run other OSes than solaris on your sunray server, no?[21:19:19] <reflect> I mean, to present to your sunray client[21:19:34] *** vergil66 has joined #opensolaris[21:21:03] *** gebi has joined #opensolaris[21:21:54] *** dalibor has quit IRC[21:22:07] <BankerBarney> audacity needs a buffer allocation like glame[21:22:14] <BankerBarney> that would help it a great deal[21:22:28] <BankerBarney> audacity records like crap[21:22:40] <BankerBarney> its good for editing files tho[21:23:40] <BankerBarney> err..when you are recording from a line-in source I mean[21:24:10] *** vergil66 has quit IRC[21:27:28] *** nikolam_ has joined #opensolaris[21:30:50] *** nikolam__ has quit IRC[21:31:03] <BankerBarney> Size hint for audio buffers[21:31:03] <BankerBarney> This number specifies the default size of audio fragments that get passed through the audio processing networks of GLAME. Choose large values (1024) for slow machines, low values (128) for low latency networks. The minimum internal latency can be calculated as size hint divided by the samplerate (128 / 44100 Hz = 2.9 ms). Note that the audio fragment size is usually limited by your audio hardware, so this is only a hin[21:31:04] <BankerBarney> t and specifying values less than 32 will probably only hurt performance.[21:31:16] <BankerBarney> http://glame.sourceforge.net/manual.var[21:33:37] *** dalibor has joined #opensolaris[21:34:26] *** ewindisch has quit IRC[21:35:26] *** nikolam_ is now known as nikolam[21:36:00] *** raichoo has quit IRC[21:36:22] <BankerBarney> of course the best is windows appz...Adobe Audtion, and now Adobe Sound Forge[21:37:29] * BankerBarney druelz[21:37:36] <BankerBarney> http://www.adobe.com/products/audition/[21:38:39] <BankerBarney> adobe audition now used to be cool edit pro[21:38:47] <BankerBarney> good app[21:39:49] <BankerBarney> audition and this app..make for a dreamy dj compilation[21:39:52] <BankerBarney> http://www.mixmeister.com/[21:40:01] * BankerBarney is a music freak[21:40:03] <Cobi_> How do I kill a process that does not want to die with kill/kill -9? It's in the S state.[21:40:54] <BankerBarney> kill -9 processid[21:41:13] <BankerBarney> thats force kill[21:41:21] *** m4rku5 has quit IRC[21:41:22] <reflect> Cobi_: often, a process that won't die even with -9, waits for something.. I/O or similar[21:41:46] <reflect> I don't think they can be killed[21:41:48] <BankerBarney> Cobi_..ps -a or top for process ids if you are from console[21:42:15] <Cobi_> BankerBarney: I already said it does not die with kill -9.[21:42:22] <Cobi_> reflect: Hrm.[21:42:30] <CodeWar> pfexec kill?[21:42:44] <Cobi_> CodeWar: As root.[21:42:56] <CodeWar> yes try it[21:42:59] <BankerBarney> cobi_ power button works...[21:43:05] <BankerBarney> or uhh..get new hardware...[21:43:15] <BankerBarney> those are the other osol answers usually[21:43:29] * BankerBarney chuckles[21:43:40] <Cobi_> BankerBarney: It's not running directly on hardware. It's running on a Xen VM server.[21:43:48] <BankerBarney> power off then[21:43:49] <reflect> BankerBarney: that's just wrong.[21:43:52] <BankerBarney> lol[21:44:20] <reflect> most of the time, there's no choice but to reboot, once you get into that state[21:44:20] *** sstallion has joined #opensolaris[21:44:22] <CodeWar> SunRay3 claims to be able to drive 5120x1600 two displays. Does it really provide good user experience for basic 2d and minimal 3d / video at that resolution?[21:44:36] *** sndcrb has quit IRC[21:45:03] <BankerBarney> or the other typical answer is man yourcrapisnotdying..this man can usually be found nowhere..did you know it even existed[21:45:10] <sstallion> SunRay3?[21:45:51] <BankerBarney> typically I get answers to manuals that exist..but I never knew they were there[21:46:36] <BankerBarney> then once you read the manual, you have to get familiar with jumbled up, alien coded, slapped together in some kina fashion reading material..[21:46:58] <Cobi_> reflect: Hrm ... the process in this state is cp, and it is copying from a samba share to a local RAID-Z array.[21:47:00] <BankerBarney> once you get the hang of this kind of manual then you have no issues really..it just takes some getting used to is all[21:47:15] <Cobi_> The samba share and local zpool both seem fairly responsive.[21:48:14] <reflect> Cobi_: then it would seem that at some point, the process couldn't reach the share, and from that, stopped responding[21:48:28] <reflect> I'm not sure how to get it to respond, I'm afraid[21:48:32] <sstallion> ahh very cool[21:48:43] <sstallion> I wonder how much faster than it is over a ray 2[21:49:50] <reflect> Cobi_: you could try kill -HUP <pid>[21:50:03] <reflect> but.. since -9 didn't work..[21:50:31] <BankerBarney> i have never had a process that would not die on -9[21:50:35] <tsoome> pstack pid[21:50:55] <tsoome> look what is it waiting for.[21:51:01] <BankerBarney> but I have not extended my research capabilities far enough to find that crappy piece of software that did not wanna die on -9[21:51:08] <reflect> BankerBarney: then I ask how long have you been running bsd/unix and in what kind of environments?[21:51:44] <BankerBarney> im running a 500pc with 768mb of ram and everything dies fine on it so far[21:51:58] <BankerBarney> but it isnt[21:52:13] <reflect> BankerBarney: then that is the answer.. you're not in a network environment[21:52:39] <reflect> once your system relies on filesystems from other machines..[21:52:53] <reflect> you will bump into this problem sooner or later[21:52:56] *** not-me-guv has joined #opensolaris[21:53:26] <reflect> and it's not about the quality of applications, I can promise you that[21:53:55] <CodeWar> "relies on filesystems from other machines.. problem sooner or later" remind me again ? what problem is that[21:54:22] <reflect> CodeWar: that some processes can't be killed[21:54:28] <CodeWar> because?[21:54:36] <BankerBarney> hung on the network supposedly[21:54:58] <Cobi_> tsoome: # pstack 1729[21:54:58] <Cobi_> pstack: cannot examine 1729: unanticipated system error[21:55:22] <reflect> CodeWar: because the other machine doesn't respond..[21:55:23] <BankerBarney> the only time I have had a process that would not die, is when I could not access the console..the app was froze in gnome on linux or something[21:55:24] <CodeWar> a blocking read across the network issued and its not returning back coz of n reasons.. hence the process cannot be killed until it times out? yuckk..[21:55:41] <CodeWar> thats extremely poor I/O management on the OS part ... even Vista gets it right http://channel9.msdn.com/shows/Going+Deep/The-Advancement-of-Windows-Narayanan-Ganapathy-Windows-Vista-IO/[21:55:50] <BankerBarney> where I had no ability to get to the console to shut it down..[21:56:32] *** m171c0 has quit IRC[21:56:41] <sstallion> reflect: process images are loaded into memory; a process isn't killable because its executable may be rooted on a shared filesystem...[21:57:04] <sstallion> s/killable/unkillable/[21:57:49] <Cobi_> It should still be able to be dropped out of the scheduler and out of memory.[21:58:35] <reflect> sstallion: yes, exactly. I'm trying to explain why a process won't die with -9, and that is has nothing to do with "bad code"[21:58:48] <sstallion> reflect: thats retarded.[21:59:01] <reflect> what is?[21:59:39] <sstallion> are you suggesting that SIGKILL may not be delivered if you are running an executable rooted in a shared filesystem?[22:00:17] *** dacek has joined #opensolaris[22:00:35] *** sergiusens has quit IRC[22:00:44] <BankerBarney> reflect I am on your side believe me[22:01:00] <BankerBarney> but I am only as good as the information given to me..[22:01:08] <BankerBarney> see what I have to put up with?[22:01:49] <reflect> I'm saying that a process recieving SIGKILL just can't die.. often due to the kernel waiting for something.. this is often in relation to nfs or other remote filesystems[22:02:00] <Cobi_> sstallion: The process is `cp`, definitely rooted locally. The process is reading from a samba share and writing to local disk.[22:02:06] <reflect> it doesn't happen often[22:02:11] <reflect> but it does happen[22:02:21] <dacek> there is a wpa support for intel pro/wirless driver?[22:02:45] <reflect> and calling it retarded is a bit strong, imho[22:03:04] <sstallion> reflect: one does not follow to the other; a process is loaded into memory prior to execution - it doesn't matter where its located. If you are in a wedged system call, that is a different matter[22:03:19] <sstallion> look at the stack of the wedged lwp - it should be pretty obvious what is going on[22:04:01] <sstallion> SIGKILL by definition cannot be caught or ignored...[22:04:42] <reflect> yeah, cannot..[22:04:48] *** gebi has quit IRC[22:04:56] *** cybernd has joined #opensolaris[22:05:05] <BankerBarney> if I hand you a manual only with the word "the" in it, but I wrote the manual and have all the rest of the information for the manual..that manual is not necessarily any good to anybody is it? These manuals on nix naturally assume you know a great deal of stuff before you come to the table..[22:05:55] <BankerBarney> so what you have to do is fill in what you "dont know" by asking silly questions[22:05:56] <sstallion> reflect: see: /usr/include/iso/signal_iso.h:60[22:06:06] <g4lt-mordant> BankerBarney, you have a weird definition of "these manuals on nix". Ever read a Sun bound manual?[22:06:09] <BankerBarney> and you get the arrogant run around in here usually on those questions[22:06:22] <reflect> sstallion: Cobi is the one you should talk to[22:07:09] <BankerBarney> g4lt-mordant, if I read a manual on zfs that supposes I know something about raid right?[22:07:20] <CodeWar> I think the uber question is having issued a blocking read call that goes over the network and gets stuck for whatever reason ... can I kill this process and ask the kernel to *cancel the I/O*[22:07:39] <sstallion> BankerBarney: value judgments made upon anecdotal evidence hardly qualifies as a reasonable response to an open question - if a SIGKILL is truly unresponsive, the correct approach is to gather information and file a bug report - not wax philosophical over man pages.[22:07:43] <g4lt-mordant> BankerBarney, no, if you read the zfs admin guide off docs.sun.com, it mqakes almost no assumptions about your knowlege[22:08:05] <sstallion> CodeWar: kill -KILL <pid> should be sufficient - if its not, pstack output would be very helpful.[22:08:29] <BankerBarney> g4lt-mordant, you say that having prior knowledge[22:08:44] <BankerBarney> approach the manual with absolutely no knowledge of a computer[22:08:58] <reflect> BankerBarney: you can't expect every manpage to treat you like you were new to computers, though[22:09:02] <g4lt-mordant> BankerBarney, no, I say that having GOTTEN said knowlege frroom a Sun manual[22:09:09] <BankerBarney> its algebra and long haired..it makes no sense to a person who reads newpapers all day[22:09:32] <Cobi_> sstallion: What do you want me to do to get more information?[22:09:40] <g4lt-mordant> BankerBarney, have you actually READ the zfs admin guide?[22:09:45] <sstallion> Cobi_: pstack <pid> > /var/tmp/stack.out[22:09:48] <BankerBarney> I dont need the manual to treat me good, I only need less arrogance to silly questions[22:09:53] <BankerBarney> thats all[22:10:00] <BankerBarney> dont suppose we know this stuff[22:10:03] <sstallion> Cobi_: post it either via pastebin, or somewhere we can get a look at it[22:10:08] <BankerBarney> when you help dont be arrogant about it[22:10:10] <g4lt-mordant> BankerBarney, have you actually READ the zfs admin guide?[22:10:16] *** ma3x has quit IRC[22:10:38] <Cobi_> pstack: cannot examine 1729: unanticipated system error[22:10:54] <g4lt-mordant> you're making assumptions about documentation you haven't even read, and call US arrogant?[22:10:54] <sstallion> Cobi_: sounds like you've hit a scheduling bug of some sort[22:10:57] <BankerBarney> I have a video tutorial about zfs..but thats just it..these guys sat down in a classrom/manual environment and learned zfs[22:10:57] *** m171c0 has joined #opensolaris[22:11:06] <sstallion> (which isn't surprising given that SIGKILL is doing nothing)[22:11:08] <BankerBarney> you cant just throw a zfs manual at someone and say here..learn this..[22:11:12] <BankerBarney> it does not work that way[22:11:24] <BankerBarney> hands on must go with the manual[22:11:28] <sstallion> what does ps -fp <pid> show?[22:11:43] <BankerBarney> all anybody asks for in here is less arrogance when they need for assistance with the hands on part of it[22:11:54] <BankerBarney> thats what this channel mostly does, it helps with the hands on part of it[22:12:08] <g4lt-mordant> BankerBarney, and how does your not having hands on experience make us arrogant for telling you to get it?[22:12:33] <sstallion> BankerBarney: wrt your current thread - if you have a question, ask it simply and wait for a response. If such a response references documentation - read it before assuming knowledge is simply being withheld.[22:12:42] <BankerBarney> g4lt you dont understand your position here...your position is mostly from a teaching standpoint...[22:12:48] <BankerBarney> not from the manual standpoint[22:12:53] <Cobi_> sstallion: http://pastebin.com/wJWLWDtv[22:13:18] <BankerBarney> classrooms/manuals/teachers all go together in pkg install knowsomethingaboutsolaris[22:13:18] <g4lt-mordant> BankerBarney, that totally didn't make sense. I didn't understand *my* position?![22:13:48] <sstallion> Cobi_: what is pid 538?[22:14:00] <BankerBarney> the natural position of anyone in here is from the "hands on" approach..because people come in here needing answers to their problems in the "hands on" phase[22:14:08] <BankerBarney> not in the text book phase[22:14:10] <Cobi_> sstallion: root 538 1 0 18:43:23 ? 0:00 -bash[22:14:15] <g4lt-mordant> actually, of those three, I've studied one, manuals. never had any classroom/teacher-led learning about solaris[22:14:25] <sstallion> gotcha[22:14:34] <reflect> BankerBarney: while I agree that humility is a good thing when it comes to channels like this, I also think that before asking a question, it's good if the person has done some homework and understands things[22:14:51] <sstallion> Cobi_: is this a subshell of some sort? its odd you don't have a TTY allocated.[22:14:51] <BankerBarney> reflect thats fine if someone knows where to look[22:15:05] <BankerBarney> and if the man is actually worth something[22:15:11] <reflect> BankerBarney: if that is the case, then there's a tried and true way of getting help[22:15:14] <BankerBarney> most nix manuals assume you know something coming to the table[22:15:20] *** rmorse has joined #opensolaris[22:15:30] <Cobi_> sstallion: Well, it was a SSH session, but then when it froze, I ~.'d and connected again.[22:15:36] <sstallion> ahh okay[22:15:39] <BankerBarney> if nix manuals explained everything folkz...there would 1 be no need for this channel..2 no need for an osol forum for tech support[22:15:40] <sstallion> that makes more sense[22:15:46] *** dalibor has quit IRC[22:15:47] <BankerBarney> this is pretty easy to understand[22:15:52] <sstallion> alright... lets get some specifics... what is your uname -a ?[22:15:54] <reflect> "I'm trying to do X with hardware Y, and I've tried searching a bit using google, however I'm not sure about the terms to search for.. could someone give me a hint?"[22:15:55] <BankerBarney> those manuals dont explain things well enough[22:16:01] <BankerBarney> they would if they covered all of the known issues[22:16:02] <BankerBarney> but they dont[22:16:05] <Cobi_> SunOS opensolaris 5.11 snv_111b i86pc i386 i86xpv Solaris[22:16:11] <g4lt-mordant> BankerBarney, no, because there's always someone that fails to RTFM[22:16:18] <sstallion> Cobi_: isainfo -k[22:16:31] <Cobi_> i386[22:16:32] *** jbit has quit IRC[22:16:42] <BankerBarney> also if it was just as easy for people to read a manual for tech support on their own they would not hire solaris tech support and spend great deals of money on it[22:16:56] <BankerBarney> lets be truly practical about what these manuals really are[22:16:59] <sstallion> Cobi_: is this a 64-bit host (not guest) ?[22:17:02] <BankerBarney> be for real about it[22:17:09] <Cobi_> sstallion: No.[22:17:29] <sstallion> Cobi_: ouch.[22:17:36] <g4lt-mordant> BankerBarney, they're things you have never read by your own admission. how do you know what's in them?[22:17:51] <sstallion> Cobi_: comfortable with dbx at all?[22:18:00] <g4lt-mordant> in fact, what qualifies you to characterize them at all?[22:18:08] <Cobi_> sstallion: It's a Xen hypervisor, with Dom0 being a Linux system. The machine is a ProLiant DL360 G3, with 8 gigs of ram, and 2 xeon processors at 2.8GHz.[22:18:15] *** jbit has joined #opensolaris[22:18:24] <sstallion> That doesnt make any sense[22:18:33] <sstallion> those specs are not compatible with a 32-bit architecture[22:18:51] <Cobi_> Hmm?[22:18:52] <sstallion> My question was asking of dom0 was 64-bit and your pv guest was 32-bit[22:19:14] <sstallion> (You can only address 4GB given 32-bit)[22:19:27] <sstallion> s/of/if/[22:20:08] *** Aria has joined #opensolaris[22:20:34] <sstallion> it sounds like you might be having some issues with mixing 32-bit and 64-bit - ISTR something in the distant past where mixing the two wasn't tolerated very well by Solaris and/or Xen[22:21:02] <Cobi_> Linux Square 2.6.26-2-xen-686 #1 SMP Wed Aug 19 08:47:57 UTC 2009 i686 GNU/Linux[22:21:02] <Cobi_> Definitely 32-bit.[22:21:19] <sstallion> If you want, I can talk you through attaching a debugger and seeing if we can get a trace, but odds are, you are going to have to reboot to flush your hosed process.[22:21:38] <sstallion> Cobi_: you are *really* underusing your hardware then.[22:21:57] <sstallion> its possible you have a chip which supports PAE I suppose[22:22:06] <Cobi_> The host sees 8GB.[22:22:28] <sstallion> PAE it is then ;)[22:23:22] *** ma3x has joined #opensolaris[22:23:29] <sstallion> Cobi_: give me a few minutes - I can walk you through debugging this if you like. As mentioned before; you are likely going to need to reboot the guest anyway to flush the corrupted state.[22:23:41] <sstallion> if you don't care about filing a bug report, you can save some pain and just reboot ;)[22:23:58] <Cobi_> "PAE is provided by Intel Pentium Pro (and above)." -- Wikipedia, so ... yeah, I would assume I do have PAE :)[22:24:09] <Cobi_> Sure, what do I need to do to debug this?[22:25:28] <crazed> anyone get kerberos as a password store working in opensolaris? users are in an ldap store[22:25:57] <reflect> Cobi_: so, is it the solaris or the linux client that stalls with the cp command?[22:26:16] <crazed> http://opensolaris.pastebin.ca/1856033 is my pam.conf[22:27:04] <Cobi_> reflect: The solaris is a Xen guest. The host is a linux machine. The cp is running on opensolaris, copying from another linux machine over a samba share.[22:27:34] <reflect> wow, that's some mix you have there :)[22:27:46] *** Odin- has joined #opensolaris[22:29:26] <sstallion> Cobi_: have you any experience with dbx or mdb?[22:29:42] <BankerBarney> [16:21] <g4lt-mordant> BankerBarney, they're things you have never read by your own admission. how do you know what's in them? <--- This approach is rather shallow...who reads the small manual to the car before they buy it...hardly no one who has already driven the car already...Do you suggest I read every manual to osol before I decide to buy into it?[22:30:08] <Cobi_> mdb, barely, I had to follow some instructions when I set up opensolaris to boot the install's kernel under xen, but ... other than that, not really.[22:30:29] <sstallion> Cobi_: mdb -p <pid>[22:30:32] <Cobi_> I haven't used dbx before.[22:30:43] <sstallion> either would be fine for this particular problem[22:30:46] <BankerBarney> no..i go with what I know already..because I have been in IT for a long time..I may not know everything about IT but I know enough to get me by...osol is supposed to sell itself to me..I am not supposed to sell myself to it[22:31:00] <BankerBarney> is it worthwhile product..yes or no? How practical is it?[22:31:07] <Cobi_> mdb: cannot debug 1729: unanticipated system error[22:31:07] <Cobi_> mdb: failed to initialize target: No such file or directory[22:31:07] <BankerBarney> see my point..[22:31:36] <sstallion> Cobi_: likely not worth it at this point[22:31:46] <sstallion> Cobi_: see if you can create a core[22:32:07] <sstallion> gcore -o /var/tmp/core.<pid> <pid>[22:32:30] <crazed> BankerBarney: as a file server it is extremely practical[22:32:34] <Cobi_> gcore: cannot grab 1729: unanticipated system error[22:33:06] <sstallion> Cobi_: I was afraid of that. At this point, all you can really do is work through the kernel debugger - something that would take a little time[22:33:32] *** pjfloyd has quit IRC[22:33:35] <sstallion> Cobi_: you've managed to corrupt the scheduler - be proud! :D[22:33:48] <BankerBarney> yea but you dont know everything about what is under the hood of your car before you buy it...what qualifies g4lt-mordant to buy a car because he does not know about every part along its whole body..[22:34:03] <BankerBarney> his approach is rather arrogant and shallow[22:34:21] <sstallion> g4lt-mordant: may as well stop feeding the troll.[22:34:53] <BankerBarney> whats he gonna do..force everybody to put the thing together from start to finish before he sells it?[22:34:56] <Cobi_> Hmm.[22:35:03] <BankerBarney> what an idiot..[22:35:52] <sstallion> Cobi_: if this is a one off, you can probably forget about it and just move on[22:35:54] <BankerBarney> then when his breaks down he takes it to the dealer and the dealer sais get fucked rtfm idiot[22:36:02] <BankerBarney> then does not wanna work on his car[22:36:24] <BankerBarney> "Lets be practical about this"[22:36:28] <Cobi_> sstallion: One other thing before I reboot -- Is snv_111b the latest version, or should I try to upgrade?[22:36:28] <robinbowes> BankerBarney: The difference is you pay for the car, and it includes warranty.[22:36:34] <sstallion> otherwise, you can download SCAT and dump kernel state[22:36:35] <Cobi_> (Since I am going to be powering it off anyway)[22:36:40] <CodeWar> BankerBarney, if you need a well documented humane OS pay for it and get a Mac/Win7 and al you re saying will hold true[22:36:44] <BankerBarney> no do your own service jackasses[22:36:52] <BankerBarney> rtfm...[22:36:53] <sstallion> Cobi_: its the latest release version. 2010.03 should be soonish[22:37:02] <sstallion> At this point, I would just wait for the next release.[22:37:12] <Cobi_> OK.[22:37:13] <crazed> use /dev[22:37:17] *** artabros has joined #opensolaris[22:37:28] <BankerBarney> see this is what I mean..and so the guy looks at the manual and has to put his crap back together all on his own..and the manual was written specifically by the mechanic[22:37:32] <artabros> hi[22:37:39] <BankerBarney> well sure the mechanic knows how to work on the car[22:37:53] <BankerBarney> and he wrote it thinking you were like him[22:37:55] <BankerBarney> but are you?[22:37:56] <BankerBarney> really[22:38:04] <robinbowes> BankerBarney: take it to /dev/null, please.[22:38:07] <sstallion> BankerBarney: you are in a channel full of mechanics. I would suggest visiting http://www.opensolaris.com if you are looking for user-oriented documentation. Otherwise, give it a break.[22:38:52] <crazed> the man pages are you friend[22:38:53] <BankerBarney> thats fine..but dont be arrogant prick mechanics to someone who asks for help when their crap breaks down[22:39:16] <BankerBarney> and dont throw them the manual like throwing a paper to a dog to go fetch when it does[22:39:32] <BankerBarney> treat people with a little respect thats all[22:39:36] <sstallion> BankerBarney: I don't recall being a "prick" - take if off-channel.[22:39:49] <sstallion> s/if/it/[22:40:09] <robinbowes> BankerBarney: there's only one guy I can see being a "prick" here.[22:40:50] <BankerBarney> im sure I do look like a prick to a bunch of "pricks" but only after you were made aware of how stupid you look[22:40:54] <Cobi_> BankerBarney: Keep in mind that people in this channel don't necessarily get paid to be here and help.[22:41:18] <BankerBarney> so now that you know..dont do it anymore[22:41:23] <BankerBarney> pretty simple fix[22:41:23] <reflect> BankerBarney: you had a point, you made it..[22:41:23] <robinbowes> ha ha ha ha ha ha[22:41:30] *** wdp__ has joined #opensolaris[22:42:18] <BankerBarney> thats just like a bunch of mechanics treating women like idiots because they dont know about vehicles[22:42:20] <g4lt-mordant> so's /mode +q[22:42:23] <BankerBarney> dont do that shit to women..thats wrong[22:42:35] <BankerBarney> dont do it to me either[22:42:49] *** ozux has quit IRC[22:42:51] <reflect> BankerBarney: you made your point.[22:43:42] <CodeWar> I have just one question .. how on earth did you land up with OpenSolaris of all OSes .. only mechanics and low lifes/ sysadmins use it .. seriously its not a popular desktop OS so let it go :-)[22:43:58] <causality> BankerBarney: most women dont know about cars though.[22:44:13] <CodeWar> thank God they dont[22:44:25] <robinbowes> Ah, that explains it...[22:44:31] <robinbowes> BankerBarney is a woman[22:44:53] *** wdp_ has quit IRC[22:45:32] *** lattera_ has quit IRC[22:46:57] <eklof> Anyone else keep getting "Error creating vdisk" when installing in xvm using virt-manager?[22:47:19] <eklof> I've checked that i have write permissions to the directory.[22:48:31] <BankerBarney> see if this was the bible there would be no excuse because the mechanic wrote the book specifically for all men to be able to comprehend what it said...it is the most commonly sold book in the world because of its ease of understanding...its not a proprietary system.[22:49:14] <reflect> BankerBarney: the.. bible?[22:49:37] *** ozux has joined #opensolaris[22:49:39] <BankerBarney> not just "a bible", "THE BIBLE"[22:50:17] <g4lt-mordant> can someone please silence this chump?[22:51:48] *** milek_ has joined #opensolaris[22:52:00] <reflect> .. religion just doesn't belong in a tech channel, imho..[22:52:19] *** BankerBarney was kicked by lewellyn (you're getting way too offtopic)[22:52:26] <g4lt-mordant> thank you[22:52:45] <CodeWar> ~lewellyn++[22:52:51] <reflect> yes, thank you.[22:52:52] *** BankerBarney has joined #opensolaris[22:52:53] <sstallion> Damn[22:52:54] <sstallion> http://www.amazon.com/OpenSolaris-Bible-Wiley-Nicholas-Solter/dp/0470385480/ref=sr_1_1?ie=UTF8&s=books&qid=1270328201&sr=8-1-spell[22:52:57] <sstallion> I had that ready to go ;)[22:53:04] <reflect> hehe[22:53:05] <lewellyn> i get back to my computer and that's what i see going on? not fun for a saturday :P[22:53:30] <lewellyn> BankerBarney: mind your etiquette or it'll be more permanent[22:55:09] <lewellyn> ok. now to play the 15 minute game i came to the computer for[22:55:23] <sstallion> alanc: ping[22:55:39] *** dalibor has joined #opensolaris[22:57:38] *** SmaSh2 has joined #opensolaris[22:59:11] *** sergiusens has joined #opensolaris[23:00:15] *** CodeWar has quit IRC[23:01:00] *** psychicist has quit IRC[23:03:05] *** lattera has quit IRC[23:04:48] *** dacek has quit IRC[23:09:29] *** RoyK has joined #opensolaris[23:09:56] *** longcat has joined #opensolaris[23:10:50] *** LaidBack_01 has joined #opensolaris[23:11:04] <longcat> kstat says clock_MHz 3200, but only current_clock_Hz 2403000000 and supported_frequencies_Hz 1203000000:1603000000:2003000000:2403000000[23:11:30] <Stric> psrinfo -pv ?[23:12:08] <longcat> http://pastebin.com/fqEAtfgv[23:12:40] <Stric> did you perhaps overclock it in bios?[23:13:25] <Stric> the label (E3200 @ 2.4GHz) is factory made.. the various cpufreq tables are also factory made.. the current freq can be altered by bios..[23:13:34] <longcat> maybe i did[23:14:28] <Stric> I tried overclocking an A64 from 2.0 to 2.1GHz just for fun.. it's fine, until I enable cpufreq scaling in linux, then it chooses between 1, 1.2 and 2.0 or so.. until I disable cpufreq, then it goes back to 2.1 (which isn't a valid freq according to tables in cpu)[23:14:59] <longcat> seems scaling is enabled by default according to powertop[23:17:03] <Stric> yep; scaling_available_frequencies is '2000000 1800000 1000000' .. but once I disable automatic cpu freq scaling, it pops back to 2.1GHz[23:17:54] <longcat> kinda surprised i got a non-flaming answer[23:18:41] *** ivo_ has joined #opensolaris[23:19:07] <jaywalk> evening all[23:19:19] <jaywalk> Im trying to learn networking in osol, making progress, slowly. Now I made a IP NAT that's partially working. :D[23:19:38] <ivo_> Hey guys do you think I should "convert" a zpool containing to mirrored pools to a zpool containing one raidz[23:20:18] *** wdp__ has quit IRC[23:20:21] <jaywalk> ip forwarding active on the nic, and ipnat rule added. I can see a daemon on one zone making outbound connections, but cannot ping from the local network or seem to make any connections out myself[23:20:33] <jaywalk> havent added ipf rules for any inbound traffic yet[23:20:48] *** kevku has quit IRC[23:21:16] *** W0rmDrink has joined #opensolaris[23:22:13] <longcat> I set everything in power.conf that was enabled or default to disabled. powertop indicates P-states are staying at max now, but only at 2.4GHz, and kstat shows the current_clock_Hz is 2.4GHz. Maybe i need to save the settings and reboot so they're never touched in the first place[23:25:58] *** dalibor has quit IRC[23:31:41] <CosmicDJ> longcat: what's your cpu (again?! sry, just jumped in)[23:32:26] <tomww> longcat: I played around on my AMD 7700 and have this: autopm enable / autoS3 default / cpu-threshold 5 / cpupm enable poll-mode / cpu_deep_idle enable[23:32:51] <tomww> if the system is really idle, it switches down to the lower speed[23:36:03] <reflect> lewellyn: do you look at your msgs.. ?[23:36:51] <CosmicDJ> longcat: nevermind[23:40:27] *** fleyta has joined #opensolaris[23:40:44] <fleyta> How can I get back to the text console from gdm?[23:40:47] <Meths> alanc: Turn's out there is a security repo which seems to have little to no docs pointing to it and so consequently may or may not be being maintained. http://chototsumoushinp.dip.jp/projectkyoto/home.html[23:42:04] <fleyta> Does anyone know how can I get back to the text console from gdm?[23:42:50] <zazenrasta> fleyta: svcadm disable gdm?[23:43:20] <fleyta> zazenrasta: I don't have ssh access and I ruined /etc/hosts and Gnome crashes from that.[23:43:44] *** hecsa has joined #opensolaris[23:43:50] <hecsa> Hi![23:44:21] <longcat> tomww: do you overclock it? it seems that since the overclocked-by-the-bios speed isnt part of some table in the cpu, when power management is on, it's never used[23:44:27] <longcat> the bios set frequency rather[23:45:13] *** Killsudo has quit IRC[23:45:26] <fleyta> Anyone familar here with X11?[23:46:44] <longcat> reboot, edit kernel prompt, add -s for single user or something like that[23:47:22] <fleyta> longcat: Why does the boot menu not have such an entry? Danger for new bies?[23:48:00] <longcat> basically, the minimal is done, anything else only gets done from a bug report[23:52:02] *** gjl has quit IRC[23:53:16] *** sergiusens has quit IRC[23:54:02] *** Odin- has quit IRC[23:54:07] <fleyta> which files do I have to edit to use e1000g manualy?[23:54:31] <fleyta> I tried /etc/hosts, /etc/netmasks and /etc/defaultrouter and now localhost is gone. WTF?[23:55:57] *** eleusis has quit IRC[23:56:14] *** gothos has quit IRC[23:57:30] <nachox> fleyta, to use e1000g manually?[23:58:01] *** eleusis has joined #opensolaris[23:58:01] *** eleusis has joined #opensolaris[23:58:08] <fleyta> nachox: without nwadm[23:58:46] <nachox> you need dhcp?[23:59:39] *** gothos has joined #opensolaris