April 27, 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 27, 2010 [00:02:13] *** sergiusens has quit IRC[00:03:11] *** InTheWings has quit IRC[00:09:48] *** hunter has quit IRC[00:12:19] *** gebi has quit IRC[00:14:15] *** dalibor has quit IRC[00:16:26] *** dalibor has joined #opensolaris[00:23:44] <sponix> So, why 2010.03 in the title Wouldn't it be 2010.04 or 2010.05 ? *Grin* As long as Oracle/Sun isn't relicensing the whole thing, or changing the EULA to do mandatory support contracts, I'll be happy either way :P[00:23:45] *** PeterT_ has joined #opensolaris[00:23:45] <kmays> vraa: Check with the EON developer as he may have some other tips of the field and is a resident CIFS/Samba expert..same with Nexenta folk.[00:24:31] <lewellyn> sponix: all the hype had said 2010.03 and no one has a firmer date yet[00:25:36] <TommyTheKid> OpenSolaris NEXT ? :p[00:25:50] *** davenz has joined #opensolaris[00:26:09] *** jthunder_ has joined #opensolaris[00:26:15] <sponix> lewellyn: can someone paid by Sun (shit Oracle) confirm (like Alanc or jmcp maybe) that to their knowledge the hold is for coding/fixing and not for _other_ reasons ? :)[00:26:38] <lewellyn> how would i know if they can? :)[00:26:43] <lattera> http://feedproxy.google.com/~r/UnixSystem/~3/uCh4Hrf9f8Y/some-light-on-horizon.html[00:26:46] <lattera> :)[00:26:47] <sponix> TommyTheKid: Not sure, but after what they did with the EULA for "Solaris" I am a bit scared[00:27:01] <sponix> lewellyn: fair enough *Grin*[00:27:14] *** fleers has quit IRC[00:27:48] *** PeterT_ has quit IRC[00:27:49] <Dganitch> thanx samba up and working on OpenSolaris, smb kernel needs some work had alot of issues[00:27:53] *** moepmoep has quit IRC[00:27:55] <sponix> http://gizmodo.com/5524843/police-seize-jason-chens-computers[00:28:02] <vraa> Dganitch told you :)[00:28:40] <Dganitch> yeah well I should have went with Samba in the first place so much easier[00:28:58] <tsoome> what build btw?[00:29:22] <lewellyn> Dganitch: it has lots of known issues in 2009.06, fixed since then[00:29:49] <jmcp> sponix: people who hang out here and who are paid by Oracle (or Sun, in countries where combination has not occurred yet), are NOT in a position to comment on your question[00:29:51] <Dganitch> 'opensolaris 5.11 snv_111b[00:30:00] <jmcp> either with authority, or otherwise[00:30:05] <vraa> Dganitch why not 134? why 111, that's so old[00:30:08] <tsoome> well, if you are using that old build, you get what you are asking for;)[00:30:21] <Dganitch> I downloaded it from site[00:30:28] <vraa> Dganitch genunix.org ?[00:30:37] <lewellyn> vraa: 111b is the current release[00:30:38] <Dganitch> opensolaris site[00:30:46] <vraa> lewellyn LOL i'm so stupid![00:30:54] <vraa> sorry my mistake, i guess it's not that old :([00:30:54] <Dganitch> welll then[00:31:00] <Dganitch> np[00:31:02] <bdha> It's almost a year old.[00:31:06] <bdha> So... pretty old.[00:31:07] <lewellyn> vraa: it's not reasonable to expect people to know not to run it for specific uses :)[00:31:16] <lewellyn> smrt: explain dev repository[00:31:16] <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[00:31:22] <TommyTheKid> quick question about fast reboot... does it "look at" the menu.lst file to determine what OS to boot to, or does it just reboot the same kernel?[00:31:26] <lewellyn> Dganitch: you can upgrade to 134 like that[00:31:37] <TommyTheKid> i haven't had an upgrade yet to know :)[00:31:41] <lewellyn> TommyTheKid: whatever bootadm does[00:31:44] <Dganitch> sticking with SAMBA until someone writes goot how to on smb kernel?CIFS[00:31:45] <jmcp> TommyTheKid: as far as I'm aware it uses bootadm to find $current[00:32:01] <jmcp> so if your menu.lst is what you want it to be, then I would expect you to reboot to that BE[00:32:04] <TommyTheKid> dang, I wish it would read the whole kernel$ line (including console=XXX)[00:32:04] <sponix> jmcp: ouch, sorry if I stepped on any toes, just a bit worried about the direction things are moving, and of OpenSolaris will continue as a community driven project, or will be gobbled up by a commercial goal(s) . I do understand _confidential_ information though especially in my line of work[00:32:09] *** fleers has joined #opensolaris[00:32:10] <lewellyn> Dganitch: in a domain or a workgroup?[00:32:11] <Dganitch> how do you upgrade?[00:32:20] <lewellyn> Dganitch: read what smrt said[00:32:22] <Dganitch> workgroup for now[00:32:28] <lewellyn> (the release notes are in topic)[00:32:35] <lewellyn> smrt: explain active directory workgroup[00:32:35] <smrt> Um. I seem to not know anything about active directory workgroup...[00:32:37] <jmcp> sponix: the lack of outbound communication is an issue, yes[00:32:38] <lewellyn> bah[00:32:57] <lewellyn> smrt: explain cifs workgroup[00:32:58] <smrt> In addition to domain mode, the CIFS server supports workgroup mode. See http://blogs.sun.com/timthomas/entry/solaris_cifs_in_workgroup_mode for more information. (See also: cifs active directory)[00:32:59] *** moepmoep has joined #opensolaris[00:33:04] <Dganitch> doamin, I have made it work before but for some reason it was a bear this time and no go[00:33:07] <TommyTheKid> lol, OpenSOlaris found d0ze installed on a CF disk on my blade .. oops :)[00:33:12] <jmcp> sponix: a lot of people inside the firewall have been letting exec-level people know, loud and *very* clearly, that this is not good, and requesting that this be fixed[00:33:40] *** sndcrb has quit IRC[00:33:43] <lewellyn> Dganitch: you need a recent dev build for domain support to work properly. and if you uare using 2008 (not r2), you need to install a microsoft hotfix.[00:33:46] <jmcp> sponix: *I* do not know. If I did know, I could get fired for telling you. As with other Oracle policies, you talk to your sales rep to get The Word(tm)[00:34:28] <Dganitch> link for latest dev build[00:34:42] <lewellyn> Dganitch: did you read what smrt said about the dev repository?[00:34:54] <lattera> link for latest dev build is in the /topic[00:34:56] <sponix> jmcp: I understand, and _almost_ partially support the call made to do what took place with the "Solaris" EULA.. BUT, would quickly search for another project to file bug reports for if similar fell on OpenSolaris[00:35:08] *** dalibor has quit IRC[00:35:08] <lewellyn> lattera: not to upgrade though[00:35:14] <lewellyn> lattera: he has 111b already[00:35:21] <lattera> ah[00:35:28] <jmcp> sponix: all that you and I have to go on right now is what Dan Roberts said during the Community agm[00:35:31] * jmcp looks for the link[00:35:39] <sponix> jmcp: But, I would _Never_ want to see anyone risk their job[00:35:40] <Dganitch> got it just being lazee[00:35:44] <stallion_work> sponix: patience is a virtual, and is usually rewarded...[00:35:51] <stallion_work> virtue rather[00:35:54] <longcat> rewarded by what, is the question[00:36:08] <jmcp> sponix: http://itmanagement.earthweb.com/osrc/article.php/3867771/OpenSolaris-Alive-and-Well-at-Oracle.htm[00:36:14] <stallion_work> longcat: by not annoying the shit out of those people who actually work on the project for starters.[00:36:17] <sponix> jmcp: I will look that up, since I wan't able to attend, thank you for pointing me to some information though[00:36:19] <lewellyn> longcat: well, the major bugs i was watching appear to all be fixed in 134a :)[00:36:22] *** twisti has quit IRC[00:36:25] <jmcp> yw[00:36:44] <jmcp> note that it can be summarised as "OpenSolaris stays free, and we reserve the right to do closed funky stuff on top"[00:37:25] <stallion_work> jmcp: IIRC thats really no different than Sun's original policy either.[00:37:29] <jmcp> which is "hey, Fishworks is closed funky stuff, makes lots of money (and we likes teh moneez), but the core that it's built on is still CDDL licensed with all the requirements of the CDDL"[00:37:36] <jmcp> stallion_work: exactly[00:37:46] <jmcp> people in general don't seem to remember that[00:37:56] * stallion_work grumbles.[00:38:07] * lewellyn grumbles too, but at different things[00:38:19] <sponix> jmcp: understood, I have fishworks on my Unified Storage box at work, and the Gov paid Big Bucks for it ;)[00:38:19] *** tg has quit IRC[00:38:32] *** tg has joined #opensolaris[00:38:39] <sponix> jmcp: And I see no Harm in that money going to sun for it either, that is why I picked the box :)[00:38:53] <jmcp> good-o[00:39:09] <sponix> jmcp: Thank you, to the best you can, answering my question(s) the a proper manor[00:39:15] <alanc> and Sun always said that Solaris Next would be OpenSolaris plus the non-redistributable bits we couldn't include in OpenSolaris[00:39:16] <jmcp> yw[00:40:07] *** sergiusens has joined #opensolaris[00:40:16] *** Trident has quit IRC[00:42:46] <lattera> sponix, http://feedproxy.google.com/~r/UnixSystem/~3/uCh4Hrf9f8Y/some-light-on-horizon.html[00:42:58] <lattera> looks like things are moving for the 2010.1H release[00:43:41] *** skeeziks has joined #opensolaris[00:44:52] <skeeziks> I'm using the Solaris LDAP client in self credential mode. Is it true that if a non-Kerberos user (imagine a local service account) queries NSS for information that the host principal will be used rather than a user principal?[00:45:14] *** ewdafa has quit IRC[00:45:52] <TommyTheKid> skeeziks: I assume that would reference nsswitch.conf[00:46:14] *** MuffiXXL has quit IRC[00:46:23] <TommyTheKid> "files ldap" means that the local files will answer first, right?[00:47:43] <alanc> yes[00:47:44] *** achester has joined #opensolaris[00:47:58] *** radsy_ has joined #opensolaris[00:48:16] <skeeziks> TommyTheKid: Well yes, but assuming it hits LDAP for the lookup, will it use the host principal for LDAP authentication since there is no user principal in this case?[00:48:29] <TommyTheKid> ldap files would be ldap answers first, but if the answer is not in ldap it will search files[00:49:00] *** radsy has quit IRC[00:49:00] <TommyTheKid> I would strongly recommend "files ldap" ... but thats just my opinion :)[00:49:01] <kmays> lattera: first user test is to install Oracle on it. :D[00:49:31] <skeeziks> TommyTheKid: "files ldap" is my configuration, but it's out of scope for my question.[00:49:33] <TommyTheKid> are you talking about using LDAP Naming Service?[00:49:58] <skeeziks> The system has been set up using `ldapclient init' using a profile on the LDAP server.[00:50:02] <TommyTheKid> ok[00:50:12] <TommyTheKid> just making sure we were on the same chapter (now for the page)[00:50:46] *** Trident has joined #opensolaris[00:51:07] <skeeziks> When I have Kerberos tickets, NSS lookups that hit LDAP work great. When I have no tickets, I'm getting GSSAPI errors referring to the lack of a credentials cache. I was under the impression that the host principal (in /etc/krb5/krb5.keytab) would be used in this case, but that is not what I'm experiencing.[00:52:09] <skeeziks> I should hit up the bug reports. I know this has worked in the past for me on Solaris 10.[00:52:19] <skeeziks> (I'm on snv_134 right now)[00:53:26] <TommyTheKid> oh, you are using kerberos to auth to the ldap for NSS? that is something I haven't tried before. WHen I did it in the past, I used SSL/Simple auth to the directory server[00:53:50] <TommyTheKid> the default for ldapclient init is no auth to the directory (which is sad, really)[00:54:00] <skeeziks> Yep, and yep.[00:54:43] <CIA-21> Srikanth Suravajhala <srikanth.suravajhala at oracle dot com>: 6940745 work structures without a pmcs_cmd_t need to be cleaned up after hot reset, 6945802 potential null dereference in pmcs_create_one_phy_stats()[00:54:45] <CIA-21> Thuy Fettig <Thuy.Fettig at oracle dot COM>: 6926134 Files printed from the global zone should reflect the subject's label[00:54:48] <CIA-21> Haik Aftandilian <Haik.Aftandilian at Sun dot COM>: 6943088 Disable memory DR after a domain migration[00:56:09] *** projectmoon has joined #opensolaris[00:56:19] <TommyTheKid> my assumption would be that if the system itself (root?) made a request, it would use the host credential, otherwise if it were an application user (httpd/named/etc) it would have to have its own user credential... but I am really shooting in the dark[00:56:45] <skeeziks> TommyTheKid: I'm OK with that, and in fact I like it. I just want it to be consistent :)[00:56:49] <skeeziks> I think I'm hitting a bug right now.[00:57:26] *** hajma has joined #opensolaris[00:58:17] <sponix> lattera: thanks for the link as well[00:58:49] *** _madmax_ has joined #opensolaris[01:00:15] *** ismell has quit IRC[01:01:57] *** Dganitch has quit IRC[01:01:57] *** eleusis has quit IRC[01:03:21] *** eleusis has joined #opensolaris[01:03:21] *** eleusis has quit IRC[01:03:21] *** eleusis has joined #opensolaris[01:03:33] *** jpd1234 has quit IRC[01:05:44] *** myosound has joined #opensolaris[01:06:39] *** ismell has joined #opensolaris[01:07:48] *** piwi has quit IRC[01:08:18] *** _madmax_ has quit IRC[01:09:35] *** cypromis has quit IRC[01:09:36] *** Andrew3 has quit IRC[01:09:53] *** cypromis has joined #opensolaris[01:10:32] *** lattera has quit IRC[01:13:40] *** radsy_ is now known as radsy[01:13:52] *** timsf has quit IRC[01:15:15] *** dover has joined #opensolaris[01:15:44] *** Andrew3 has joined #opensolaris[01:17:05] *** Odin- has joined #opensolaris[01:19:16] *** hunter has joined #opensolaris[01:19:21] *** hunter has quit IRC[01:19:46] *** fleers has quit IRC[01:22:06] *** rcarcamo has joined #opensolaris[01:26:09] *** snuff-work has quit IRC[02:07:49] *** echelog has joined #opensolaris[02:07:49] -zelazny.freenode.net- [freenode-info] channel flooding and no channel staff around to help? Please check with freenode support: http://freenode.net/faq.shtml#gettinghelp[04:24:56] *** echelog has joined #opensolaris[04:24:56] <tester_> 512ecc[04:25:14] <Triskelios> that is not enough for regular desktop use[04:25:19] <tester_> yup...[04:25:39] <tester_> not enough money..lol[04:26:06] <tester_> to buy a faster machine...[04:26:12] <Triskelios> I've gotten many newer systems for free[04:26:15] <ivo_> you need a lot more memory[04:26:32] <CodeWar> OT: is there a script/way to compare two channels on IRC and find out how many users are common. I m using xchat[04:27:29] <bdha> You could redirect /who or /names to two files and use your standard text comparison tools.[04:27:33] <tester_> thanks guys for the advice...but gotta step out for a while...cya later...ciao[04:27:37] <tester_> :D[04:27:46] <CodeWar> good[04:28:13] <bdha> You could probably write an irssi plugin to do that trivially enough. Assuming one has not been written.[04:28:39] <CodeWar> bdha, didnt know about the /who command (dumb I know) thanks[04:28:43] <bdha> np[04:29:07] <lblume> Hmmmm, guys, remind me, the dedup problem impacts even zfs destroy on fs where dedup was not enabled?[04:29:29] *** tester_ has quit IRC[04:30:16] <Triskelios> lblume: if they share blocks with another fs, probably[04:30:21] <Spencer_tt> Triskelios: that's a trick I need, just the quantity variable[04:30:40] <lewellyn> lblume: things working ok for you still?[04:31:47] <lblume> lewellyn: Depends which things you're talking about.... I've got a destroy running for 15 minutes and thrashing the disks for no good reason :-/[04:32:07] <lewellyn> i meant your image. :)[04:32:21] <lewellyn> if you haven't run image-update yet, don't right now. :)[04:32:30] <lblume> More than enough for my needs, thanks :-)[04:32:47] <lblume> I haven't yet, what's wrong?[04:32:58] <lewellyn> i've got to fix an apparent networking issue[04:33:04] <lewellyn> after dinner, i've got that scheduled.[04:34:42] <lblume> Ok, I can have lunch then?[04:35:06] <lewellyn> only if you share :D[04:35:39] *** e1kg has quit IRC[04:35:47] <ivo_> lblume, lunch??[04:35:53] <ivo_> in what time zone do you live?[04:37:13] <lblume> Triskelios: Oh well, maybe not. It was just slow and took 20 minutes while blocking all disk I/O, but it finished.....[04:37:26] *** leito has quit IRC[04:37:38] <lblume> ivo_: More like breakfast actually, right.[04:38:17] *** deet has quit IRC[04:38:24] <CodeWar> out of curiosity if anybody writes network/storage drivers on Solaris ... are your ISRs split into the classic top half register / disable interrupt on the device and bottom half actually process the interrupt or is it just one flat routine taht does both[04:38:57] <lblume> lewellyn: Got some frozen 饺子 :-)[04:39:46] *** leito has joined #opensolaris[04:40:50] <ivo_> lblume, I am thinking about cornflakes[04:40:58] <ivo_> :D[04:41:32] <lblume> Cornflakes are soooo boring. And expensive.[04:41:44] <ivo_> or French fries + steak[04:42:02] <alanc> lblume: do they have to import them from the US?[04:42:04] *** Beket has joined #opensolaris[04:42:15] <lewellyn> ivo_: if he makes them, they're by definition french fries ;)[04:42:26] <ivo_> :D[04:42:45] <freetown2> flip...what are 饺子?[04:42:48] <alanc> certainly not freedom fries in Beijing[04:42:49] <ivo_> the problem is that here is 4:47AM[04:42:58] <lewellyn> lblume: so... frozen 饺子... they're wheat flower, eh? :([04:43:01] <ivo_> and I am staring[04:43:01] <lewellyn> flour even[04:43:05] <ivo_> and I am starving[04:43:07] <freetown2> oh dumpling.[04:43:19] <bdha> Wife made some of the best mashed pototes I've ever had tonight.[04:43:20] <freetown2> sure love simplified chinese characters[04:43:24] <bdha> Er, potatoes.[04:43:34] <freetown2> roasted?[04:43:40] <ivo_> roasted[04:43:43] <freetown2> oh sorry[04:43:51] <ivo_> with rost beaf[04:43:52] <lblume> http://image.baidu.com/i?tn=baiduimage&ct=201326592&lm=-1&cl=2&word=%BD%C8%D7%D3[04:43:55] <bdha> No, mashed. Lots of cream. And garlic yogurt.[04:43:55] <bdha> No, mashed. Lots of cream. And garlic yogurt.om nom.[04:43:58] * freetown2 loves mashed potatoes[04:43:58] <bdha> er.[04:44:01] <bdha> That was odd.[04:44:03] <freetown2> aw[04:44:11] <alanc> damn, now I want dinner[04:44:19] <ivo_> http://en.wikipedia.org/wiki/Roast_beef[04:44:19] <bdha> alanc: Steak, too. :)[04:44:27] <ivo_> I love steak[04:44:29] * freetown2 wonders what his lunchbox contains...[04:44:45] <ivo_> I recently eat a 500g steak at once[04:44:56] <ivo_> wit a potato :)[04:44:58] <freetown2> that's a pounder right?[04:44:59] <ivo_> nice ...[04:45:01] <alanc> there's roast beef in the fridge at home, I just need to drive across the bay to get to it[04:45:29] <Triskelios> CodeWar: the DDIs enforce most of the structure[04:45:42] <freetown2> food, glorious food[04:45:55] <lblume> Hmmmm, I should go to the market and buy some potatoes......[04:47:08] *** deet has joined #opensolaris[04:47:18] <bdha> Throw two russets in with your golden potatoes. Better for texture.[04:47:21] <CodeWar> sorry meaning what exactly ..[04:47:21] <bdha> And of course, lots of cream. :)[04:47:23] <ivo_> my doctor adviced me not to eat so much beef or other red meat, and instead eat fish and birds[04:47:36] <ivo_> but birds are just not so teasty[04:47:45] <freetown2> and fish are full of metals..[04:47:49] <lblume> I could do some expat fries....[04:48:19] <lblume> ivo_: Try some Beijing duck :-9[04:48:22] <freetown2> lblume, you go to the wet market or to a supermarket?[04:48:33] <ivo_> I love Beijing duck[04:48:41] <ivo_> but is so dfat is not so healty[04:48:52] *** TomJ has quit IRC[04:48:55] * ivo_ loves chines food[04:49:08] * ivo_ loves steak :)[04:49:18] * freetown2 loves anything with loads of fat and salt[04:49:18] <lblume> freetown2: Either, depending on my needs :-)[04:49:43] <freetown2> did I forget sugar?[04:50:22] <freetown2> lblume, so what do you do to delace your veggies of pesticide and stuff?[04:50:31] <lblume> freetown2: Sweet & sour pork sounds like your favorite dish :-)[04:50:50] <freetown2> lblume, or teriyaki meat patties :D[04:50:58] <lblume> I wash them :-)[04:51:25] <freetown2> yeah...i SCRUB mine...especially at the stalks[04:51:55] <freetown2> Beijing probably has a different source than HK anyway[04:52:00] * ivo_ is going to his favorite steak house![04:52:08] <ivo_> in about 3 hours[04:52:22] * freetown2 starts the patience game.[04:52:40] <freetown2> Let's see how can wait the longest before breaking for food[04:52:49] <freetown2> s/how/who/[04:53:16] <lblume> freetown2: I got kumquats handy......[04:53:25] *** TomJ has joined #opensolaris[04:53:32] *** milek has quit IRC[04:54:19] <freetown2> sorry, can't guess what kumquats are....small tangeries?[04:54:43] <lblume> Yes, kind of.[04:54:59] <freetown2> cor, is that how they pronounce that in Mandarin?[04:55:07] <freetown2> kinda close to Cantonese[05:06:32] *** echelog has joined #opensolaris[05:06:32] -zelazny.freenode.net- [freenode-info] if you're at a conference and other people are having trouble connecting, please mention it to staff: http://freenode.net/faq.shtml#gettinghelp[05:07:28] *** ingenthr1 has joined #opensolaris[05:07:46] *** lateau has joined #opensolaris[05:07:53] *** ingenthr has quit IRC[05:14:51] <ivo_> so I can't wait anymore!!![05:15:06] <ivo_> just got fish + French fries[05:16:51] *** davenz has left #opensolaris[05:18:07] <freetown2> Fish'n'Chips. How I'd like to try that again[05:19:18] <ivo_> hmmmmmmmmmmmm[05:19:19] <ivo_> http://upload.wikimedia.org/wikipedia/commons/f/f2/Steak_4_bg_083103.jpg[05:31:17] *** myosoun_ is now known as myosound[05:35:55] *** CodeWar has quit IRC[05:36:56] *** tosh has quit IRC[05:37:22] *** vraa_ has joined #opensolaris[05:41:08] *** vraa has quit IRC[05:43:12] *** leito has quit IRC[05:43:59] *** sergiusens has joined #opensolaris[05:49:10] *** DerSaidin has quit IRC[05:49:38] *** DerSaidin has joined #opensolaris[05:49:38] *** DerSaidin has joined #opensolaris[05:50:43] *** leito has joined #opensolaris[05:51:20] *** Calibellus has joined #opensolaris[05:52:27] *** corretico has quit IRC[05:53:05] *** stevel has joined #opensolaris[06:01:58] *** dunc has quit IRC[06:05:41] *** timeless_mbp has joined #opensolaris[06:05:41] *** timeless_mbp has joined #opensolaris[06:09:28] *** purserj has quit IRC[06:10:58] *** purserj has joined #opensolaris[06:16:10] *** jthunder has joined #opensolaris[06:17:30] *** leito has quit IRC[06:19:52] *** strawf has quit IRC[06:19:55] *** strawf has joined #opensolaris[06:20:29] *** cnu has quit IRC[06:21:45] *** cnu has joined #opensolaris[06:21:51] *** Jondice has joined #opensolaris[06:23:36] *** derchris has quit IRC[06:23:41] *** derchris has joined #opensolaris[06:29:49] *** radsy has quit IRC[06:34:06] *** solar_sea has joined #opensolaris[06:34:34] *** solar_sea has left #opensolaris[06:37:08] *** xapiens has quit IRC[06:38:24] *** myosound has quit IRC[06:38:45] <trochej> Coffee[06:39:04] <TommyTheKid> coffee is for grown ups[06:42:16] <jthunder> ok question: in my /var/svc/manifest/network/dns/multicast.xml it lists the service system-log as grouping="optional_all" I replaced the system-log service, would this affect the ability for this service to function?[06:43:05] <jthunder> I currently get the error "DNSService call failed -65537"[06:43:37] <jthunder> when I try to browse or register a service using dns-sd[06:44:08] *** Calibellus has quit IRC[06:44:31] <TommyTheKid> jthunder: pastebin the output of svcs -xv[06:45:07] <TommyTheKid> if dns/multicast is there of course :)[06:45:20] <jthunder> http://pastebin.org/183880[06:45:36] *** Shoggoth has joined #opensolaris[06:45:44] <jthunder> I created a service to register my local services automatically[06:46:01] <jthunder> and that is failing due to the registration errors mentioned[06:46:52] *** dijenerate has quit IRC[06:47:05] <jthunder> I will add the commands and log file as well.[06:47:30] <TommyTheKid> I haven't messed with avahi on Solaris yet, I can't tell you anything about the service itself...[06:48:12] <TommyTheKid> so long as its "online" in svcs -l SERVICENAME, I don't think the syslog thing is messing with you[06:49:33] <jthunder> http://pastebin.org/183893[06:49:48] <jthunder> yeah that's what I thought[06:50:07] <jthunder> I don't know how to debug the mdns service any further[06:51:28] <TommyTheKid> (of course last I did this, it was on linuxppc) ... can't you just add the service to the avahi-services.conf[06:51:38] *** nevdull has joined #opensolaris[06:51:56] <TommyTheKid> technically the afp service itself should do the registering if you are going to do it separately[06:52:07] <TommyTheKid> otherwise, you might as well just configure it statically?[06:52:38] <TommyTheKid> I am "nobody" .. fscking nfsv4 :)[06:52:45] <jthunder> so I created the dnssd_afp to register the afp (netatalk) service with mdns[06:53:07] *** nevdull has quit IRC[06:53:26] <TommyTheKid> svcs -a dns/multicast[06:53:28] <jthunder> at the time (8 months or so ago) I didn't know about avahi-services.conf[06:53:29] <TommyTheKid> i mean[06:53:34] <TommyTheKid> svcs -l dns/multicast[06:54:03] <jthunder> it is online[06:54:26] <jthunder> hold on[06:55:42] <jthunder> http://pastebin.org/183923[06:56:09] *** duri has quit IRC[06:56:27] <TommyTheKid> yep, it all looks right... I just enabled it on my sunray server and while it says online, I am not so sure it is[06:57:49] <jthunder> just the dependency for system-log[06:57:59] <TommyTheKid> so, does your site/dnssd_afp depend on both dns/multicast and your afp service ? (is that why you didn't just make /etc/avahi/service/afp.conf (or whatever)[06:58:15] <jthunder> I wouldn't think it would fail if that was missing, and it would likely not start either[06:58:23] *** Zauberpony has quit IRC[06:58:38] <jthunder> not sure let me check[06:59:39] <jthunder> it just requires dns/multicast and system/filesystem/local[06:59:46] <TommyTheKid> when I run "avahi-browse -at" I get "Failed to create client object: Daemon not running"[06:59:48] <TommyTheKid> you?[07:00:10] *** Zauberpony has joined #opensolaris[07:00:19] <jthunder> service_type_browser failed: Operation failed[07:00:23] *** dijenerate has joined #opensolaris[07:00:30] *** duri has joined #opensolaris[07:00:37] <TommyTheKid> you should make it depend on your afp service too that way it doesn't register if afp doesnt start, and it de-registers if afp dies, otherwise using SMF is rather a lot of work for little gain[07:02:24] <jthunder> yeah I would like to get rid of the dnssd_afp service, but I can't figure out how to get the afp service registered any other way[07:02:46] <TommyTheKid> just put the configuration into /etc/avahi/services, they load automatically from there IIRC[07:03:19] <jthunder> I did that I have a multi.serivce entry in there but it isn't getting read[07:03:44] <TommyTheKid> i am not conviced that its all there (all of avahi I mean)[07:04:14] <jthunder> what do you get with an avahi-discover[07:05:02] <TommyTheKid> a big python dump[07:05:11] <jthunder> sorry that's probably a gtk app[07:05:24] <TommyTheKid> I am in a desktop session[07:05:31] *** ingenthr1 has quit IRC[07:05:33] <jthunder> how about avahi-browse -av[07:05:46] *** ingenthr has joined #opensolaris[07:06:03] <TommyTheKid> "daemon not running"[07:06:28] <jthunder> I get "service_type_browser failed: Operation Failed"[07:07:58] <TommyTheKid> nice that we get entirely different errors :)[07:08:08] <TommyTheKid> mine is a fresh install of 134, yours?[07:09:01] <jthunder> nope upgraded from a 2 year old install to 134[07:10:35] <TommyTheKid> oh hahaha I dont have a multicast interface maybe[07:11:31] <TommyTheKid> no they say "MULTICAST" in the flags[07:11:39] * TommyTheKid shrugs[07:12:17] <TommyTheKid> I had not enabled avahi-bridge-dsd[07:12:44] <TommyTheKid> now my avahi-browse -at outputs a bunch of interesting services :)[07:14:35] <jthunder> arg[07:14:40] <jthunder> probably as it should[07:14:42] <TommyTheKid> you too :)[07:14:48] <jthunder> nope[07:14:54] <TommyTheKid> aw, I was hoping it was easy[07:15:46] <TommyTheKid> I suggest removing the package and re-installing it, maybe somewhere along the lines, the upgrades have gone awry ... I know that doesn't help anyone else having the same problem, but it might work for you[07:15:52] <jthunder> would avahi-bridge-dsd interfere with the way dns-sd would operate?[07:15:59] <lewellyn> TommyTheKid: you mean pkg fix?[07:15:59] <TommyTheKid> they both need to be enabled[07:16:00] <lewellyn> ;)[07:16:05] <TommyTheKid> oooh yea what he said[07:16:16] <TommyTheKid> pkg fix Avahi[07:16:25] <jthunder> so just avahi?[07:16:34] <TommyTheKid> cap A I think to get the "set"[07:16:52] <jthunder> should I disable my custom dnssd_afp for now too then?[07:17:29] <TommyTheKid> it should be configured to depend on more stuff, so yea disable it.. but it should be in maint anyhow, so its fine[07:17:44] <jthunder> k[07:18:08] <TommyTheKid> it might just be pkg:/system/network/avahi now, it seems like things have vastly simplified in the new package naming[07:18:50] *** Shoggoth has quit IRC[07:19:00] <TommyTheKid> woot, my _nfs4idmapdomain TXT "DOMAIN.com" seems to have done the trick, I am not nobody anymore :)[07:19:13] <TommyTheKid> I still cant SSH to any of my OS boxes :([07:20:24] <TommyTheKid> I can "utswitch" to them (sun ray) .. and login, and everyhting seems fine, but ssh just hangs ... last message is debug1: Next authentication method: public key (right after failing gssapi)[07:20:57] <jthunder> so a "pfexec pkg fix pkg:/system/network/avahi" ran very quickly with no output[07:21:01] <bdha> Run sshd in debug?[07:21:25] <TommyTheKid> bdha: I am not sure its not the client yet :)[07:22:19] <TommyTheKid> I get the same error ssh-ing to other systems now all the sudden[07:22:46] <TommyTheKid> even a campus sunray[07:23:17] <TommyTheKid> it goes through all the key exchanges and even displays the "issue" .. but never gets to a password prompt[07:23:53] <TommyTheKid> hmm I wonder if its waiting on /dev/random or something like that[07:25:12] *** sponix has quit IRC[07:25:39] <TommyTheKid> pkg fix wont "fix" any editable file tho[07:26:33] <jthunder> when I run pkg fix with no args, it lists tons of files where things seem to be missing / different[07:26:46] <TommyTheKid> wow[07:26:55] <TommyTheKid> when it lists them it should be "fixing" them, right?[07:26:59] <jthunder> across lots of packages[07:27:20] *** nickw has quit IRC[07:27:33] <lewellyn> jthunder: that's normal :P[07:27:47] <TommyTheKid> mine, freshly installed is complaing about several python pkgs :)[07:27:47] <jthunder> ok - I guess?[07:28:04] <lewellyn> TommyTheKid: does pkg uninstall uninstall edited editable files?[07:28:16] <TommyTheKid> erm, I dont know, likely not[07:28:23] <TommyTheKid> but it might at least warn?[07:28:26] <lewellyn> so why would you expect pkg fix to touch them? ;)[07:28:29] <TommyTheKid> the old pkgrm would warn[07:28:31] <lewellyn> haha. pkg warning :P[07:28:50] <TommyTheKid> it would say it cant remove the config dir cause its not empty or something[07:28:53] <TommyTheKid> I think? :-/[07:29:02] <lewellyn> yeah. but that's not pkg :P[07:29:12] <TommyTheKid> true[07:29:55] <jthunder> so is there any safe way to remove and/or reinstall the mdns binaries and configuration?[07:31:23] <TommyTheKid> pkg contents avahi ?[07:31:32] <TommyTheKid> manually remove any editable file?[07:31:34] *** sponix has joined #opensolaris[07:34:40] *** Aria has quit IRC[07:35:02] <TommyTheKid> pkg fix mdns[07:35:17] *** Shoggoth has joined #opensolaris[07:35:19] <jthunder> no output[07:35:24] <TommyTheKid> here either[07:35:42] <TommyTheKid> and no config other than the manifest[07:36:10] <jthunder> can I pkg uninstall mdns[07:36:18] <jthunder> then pkg install mdns[07:36:32] <TommyTheKid> i dont see why not, I would make sure you did both avahi and mdns[07:36:39] *** Shoggoth has quit IRC[07:37:05] <TommyTheKid> and make sure that /etc/avahi is gone[07:37:50] <jthunder> k[07:38:07] <TommyTheKid> though mine is empty anyhow :)[07:39:33] *** stevel has quit IRC[07:39:43] <TommyTheKid> http://www.sunwfrk.com/blog/2009/04/05/specify-the-finder-icon-with-avahi-on-opensolaris-200811/[07:39:51] <jthunder> so the mdns service allows for multicast dns[07:39:54] *** Shoggoth has joined #opensolaris[07:40:08] *** Mazon has joined #opensolaris[07:40:09] <TommyTheKid> mdns is the "client side" I believe[07:40:10] <jthunder> so I should be able to use it to see an apple remote share that's being published[07:40:45] <TommyTheKid> ... I should clarify, the NSS stuff, that allows you to ssh myMac.local[07:41:51] *** Shoggoth has quit IRC[07:42:16] *** Shoggoth has joined #opensolaris[07:42:49] *** Shoggoth has quit IRC[07:42:52] <jthunder> what switch to uninstall the dependencies as well?[07:43:02] <TommyTheKid> um, are you sure you want to do that?[07:43:07] <TommyTheKid> its -r, but beware[07:43:11] *** Shoggoth has joined #opensolaris[07:43:11] *** nickw has joined #opensolaris[07:43:13] <jthunder> or ignore the dependencies[07:43:39] <TommyTheKid> i dont htink it lets you do that[07:43:43] <jthunder> avahi is used by a bunch of gnome stiff[07:43:45] <jthunder> stuff[07:43:49] <TommyTheKid> oh yea[07:47:02] <jthunder> I'll just go for it[07:47:12] <jthunder> and then reinstall any other broken stuff after[07:47:36] <jthunder> Tommy that link is what I'm trying to accomplish[07:48:05] <TommyTheKid> jthunder: put that file contents in /etc/avahi/services/afpd.service .. works here[07:48:16] <jthunder> k[07:48:25] <TommyTheKid> I actually followed one ubuntu site[07:48:39] <TommyTheKid> of course your avahi-browse -at needs to be sorta working first I think?[07:48:57] <TommyTheKid> yout "avahi-daemon-bridge-dsd"[07:49:29] <TommyTheKid> once thats working properly either the afpd.service or the "avahi-pubish" should work[07:50:25] <TommyTheKid> fwiw, once you put that file in place, either svcadm restart avahi-daemon-bridge-dsd or "avahi-daemon-bridge-dsd -r" to re-load the static service definitions[07:50:44] <TommyTheKid> i need to drift off to sleep very soon now[07:50:55] <TommyTheKid> before I fall face first on the keyboard[07:51:37] *** sponix has quit IRC[07:55:01] <jthunder> I get a daemon not running[07:56:22] <TommyTheKid> daemon not running means you should just need to enable it (disabled by default I think)[07:58:00] *** hajma has quit IRC[07:58:50] <jthunder> trying to start the avahi-bridge-dsd - Reason: Start method exited with $SMF_EXIT_ERR_FATAL.[07:59:25] <jthunder> lets reboot...[07:59:29] *** jmcp has quit IRC[08:01:05] <TommyTheKid> ok its creepy as hell that the logs of this conversation are already in google :p[08:01:38] *** perlmongo has quit IRC[08:01:51] <jthunder> I know - that is wired[08:01:53] <jthunder> wierd[08:03:07] <jthunder> so not avahi-browse -at gives "service_type_browser failed: Operation failed"[08:03:25] <jthunder> same as before[08:03:35] <jthunder> something screwy here[08:03:38] *** CodeWar has joined #opensolaris[08:03:51] *** sysspoof has joined #opensolaris[08:03:58] <jthunder> time to reinstall - and save myself some time[08:04:06] <jthunder> nice learning process[08:04:15] <jthunder> thanks Tommy for the help[08:04:15] *** bonsaikitten has joined #opensolaris[08:04:35] <TommyTheKid> you are welcome[08:04:36] <TommyTheKid> nite ![08:05:28] *** Mace_N900 has joined #opensolaris[08:12:11] *** FattyAcid has joined #opensolaris[08:17:01] *** hsp has joined #opensolaris[08:19:29] *** jmcp has joined #opensolaris[08:19:29] *** ChanServ sets mode: +o jmcp[08:23:07] *** tg has quit IRC[08:23:47] *** CodeWar has quit IRC[08:24:20] *** Beket has quit IRC[08:28:07] *** tg has joined #opensolaris[08:32:24] *** Macer has quit IRC[08:33:18] <madwizard> Coffee[08:34:10] *** Mace_N900 has quit IRC[08:38:46] *** moepmoep has quit IRC[08:40:56] *** Mace_N900 has joined #opensolaris[08:43:57] *** comay has quit IRC[08:46:33] *** raichoo has joined #opensolaris[08:46:59] <lewellyn> tea[08:47:30] <tehhobbit> beer !![08:47:55] <hsp> bock beer[08:48:37] *** Mace_N900 has quit IRC[08:50:17] *** timeless_mbp has quit IRC[08:50:25] *** comay has joined #opensolaris[08:51:05] *** sysspoof has quit IRC[08:54:12] *** Dagobert has left #opensolaris[08:54:51] *** Dagobert has joined #opensolaris[08:57:44] *** Macer has joined #opensolaris[08:59:27] *** Mace_N900 has joined #opensolaris[09:00:36] *** phimic has joined #opensolaris[09:03:21] *** hsp has quit IRC[09:03:54] * lewellyn still kind of wonders why mouse cursors are messed up when connecting to windows 2008 r2 servers with tsclient[09:05:43] *** sysspoof has joined #opensolaris[09:08:08] *** Twigathy has quit IRC[09:11:24] *** Erwann has joined #opensolaris[09:13:02] *** FattyAcid has quit IRC[09:14:46] *** mecatsun has quit IRC[09:24:39] *** merzo has joined #opensolaris[09:25:59] *** timeless_mbp has joined #opensolaris[09:25:59] *** timeless_mbp has joined #opensolaris[09:31:02] *** jgracin has joined #opensolaris[09:34:15] *** hajma has joined #opensolaris[09:35:58] <seanmcg> smrt: how silly are mouse cursots on w2kr2 ?[09:36:09] *** CIA-21 has quit IRC[09:36:14] *** nikolam_ is now known as nikolam[09:36:37] *** kevinw has quit IRC[09:36:48] <seanmcg> smrt: hello[09:36:49] <smrt> Hi, I'm smrt. I try to be a helpful bot. For more information: /msg smrt help[09:36:56] <seanmcg> smrt: how silly are mouse cursots on w2kr2 ?[09:37:07] *** Edgeman has quit IRC[09:37:15] * seanmcg shrug, smrt knows better than to answer that[09:37:43] *** gebi has joined #opensolaris[09:37:46] *** madwizard has quit IRC[09:37:50] *** madwizard has joined #opensolaris[09:37:59] *** kevinw has joined #opensolaris[09:38:29] *** miamano_ is now known as miamano[09:39:16] *** miamano has quit IRC[09:40:09] *** miamano has joined #opensolaris[09:41:15] *** icarus902 has quit IRC[09:41:19] *** CIA-33 has joined #opensolaris[09:41:43] *** Triskelios has quit IRC[09:41:48] *** Edgeman has joined #opensolaris[09:41:59] <nikolam> so this is new era. No /dev but only ON.[09:42:09] *** Triskelios has joined #opensolaris[09:42:31] <lewellyn> nikolam: just like last release, yes.[09:44:03] <nikolam> lewellyn, ever made it orom ON?. Like made your IPS, or just made IPS available from SFE?[09:44:03] <nikolam> from[09:44:24] <lewellyn> nikolam: IPS is on hold till there's a release (or release candidate)[09:44:35] <lewellyn> if there's a release candidate, it'll probably end up in /dev[09:44:45] <lewellyn> and you can't downgrade from (e.g.) 137 to 134a[09:45:22] *** timeless_mbp has quit IRC[09:46:06] <nikolam> yep. I understand. Just would like to know if you think separace seource of packages could be made from Sfe`s , separate from this[09:46:22] <lewellyn> sfe has nothing to do with any of the consolidations[09:46:25] <nikolam> I understand there is no /pending anymore for some time[09:46:34] <lewellyn> pending is still there[09:46:44] <nikolam> Not thinking about consolidations.[09:46:56] <nikolam> Someone told me it is not there anymore.[09:47:15] <lewellyn> http://jucr.opensolaris.org/pending/en/index.shtml comes up fine here[09:48:24] *** lblume has quit IRC[09:48:24] *** tsoome has quit IRC[09:48:56] *** nikolam has quit IRC[09:49:02] *** comay has quit IRC[09:49:21] *** nikolam_ has joined #opensolaris[09:51:08] *** lblume has joined #opensolaris[09:53:31] *** twisti has joined #opensolaris[09:53:34] *** nikolam_ is now known as nikolam[09:56:29] <nikolam> lewellyn, but as the repository there is no pending as it was some time before.[09:56:52] <lewellyn> there sure is. how do you think we test the stuff we build in jucr?[09:57:34] <lewellyn> the *old* pending is long-gone. but it's so old as to be irrelevant anyhow.[09:58:27] *** jlliagre has joined #opensolaris[09:59:27] *** dalibor has joined #opensolaris[09:59:30] <nikolam> when where is it now.[09:59:45] <lewellyn> 00:52 @lewellyn: http://jucr.opensolaris.org/pending/en/index.shtml comes up fine here[10:00:08] <nikolam> well how you apply it as repository/publisher?[10:00:18] <lewellyn> first, why do you want to?[10:00:51] <lewellyn> there's lots of stuff which may never leave pending. and that's with good reason ;)[10:01:01] <nikolam> lewellyn, You told me I can do that. To add it as a publisher. I can do that or I can not?[10:01:21] <lewellyn> you can. but unless you're contributing to jucr, you probably don't want to.[10:01:29] *** timeless_mbp has joined #opensolaris[10:01:29] *** timeless_mbp has joined #opensolaris[10:01:32] *** Mace_N900 has quit IRC[10:01:44] *** Aesdyn has joined #opensolaris[10:01:54] *** Mike has joined #opensolaris[10:02:01] <lewellyn> (and installing a package via jucr gives you the pending repo)[10:02:21] <Aesdyn> Hey Lewell[10:02:23] *** Mike is now known as Guest76026[10:02:43] <Aesdyn> I was thinking about learning Solaris... seems like a pretty neat OS.[10:03:13] <nikolam> Well, to do testing i suppose, to see in what state packages are etc[10:03:13] <nikolam> Wel, how to do that.[10:03:14] <nikolam> http://pkg.opensolaris.org/contrib/ gives nothing[10:03:15] <Aesdyn> Do you like it?[10:03:17] <nikolam> so there is no pending repo?[10:03:42] <nikolam> Aesdyn, better learn Opensolaris. it is open and even more nastier ;)[10:03:58] <Aesdyn> I like to tinker a bit, I'm getting real bored with Windows[10:04:07] <lewellyn> nikolam: install them via the jucr interface[10:04:24] <lewellyn> Aesdyn: chanserv should have given you some urls when you joined the channel. and there are more in /topic :)[10:04:36] <nikolam> So there is No repooo lewellyn thats the point[10:04:49] <lewellyn> nikolam: there is. it's part of jucr now. did you not click the link i gave you twice now?[10:04:50] <Aesdyn> I'm downloading osol-0906-x86.iso[10:05:15] <nikolam> yes and I see no path for a publisher[10:05:17] <Aesdyn> Lew, how long have you used Solaris?[10:05:19] <lewellyn> Aesdyn: at this point, it may be better to grab the development build. there's a new release brewing and it's based on the current dev build :)[10:05:27] <Aesdyn> ohh nice[10:05:41] <lewellyn> see /topic for the torrent and the release notes (which has other download links, if you don't torrent)[10:05:54] <lewellyn> and i've used solaris for probably about 15 years now.[10:06:06] <Aesdyn> I'd really like to learn more about filesystems and kernel drivers.[10:06:11] <lewellyn> nikolam: just install via jucr and you'll be fine. seriously.[10:06:13] <Aesdyn> I've stayed in the realm of C# and some C++[10:06:24] <Aesdyn> Does Solaris have a decent OpenGL support?[10:06:32] <lewellyn> afaik :)[10:06:37] <Aesdyn> nice[10:06:50] <nikolam> lewellyn, so is there a way to build a repo out of selected packages that are there?[10:06:50] <lewellyn> nikolam: Not here.[10:06:54] <nikolam> Or it can not be done[10:07:07] <lewellyn> nikolam: i have no clue what you're really trying to ask[10:07:22] <lewellyn> and i need to finish up with my monday work now that it's tuesday[10:07:35] *** merzo has quit IRC[10:07:44] <Aesdyn> o[10:07:49] <Aesdyn> thanks for your help lew[10:08:05] <lewellyn> Aesdyn: feel free to lurk :)[10:08:46] <nikolam> lewellyn, If one , say port a package(s) and they made it through SFE. And they are in pending. Can one make Repository to distribute that package(s) to people that oculd do testing or not?[10:08:56] <lewellyn> sfe has nothing to do with pending.[10:09:02] <lewellyn> sfe is sfe. pending is jucr.[10:09:33] <lewellyn> pending is jucr's not-yet-finished repo. where packages live before moving to contrib[10:09:50] <nikolam> So pending is a result od jucr automated building. And sfe is set of specs for building on its own machine?[10:10:34] <lewellyn> pending is where unfinished things live until they end up in contrib[10:10:41] <lewellyn> sfe is just a bunch of build recipes[10:11:11] <nikolam> So if someone is porting something he should first do that through: Sfe and after testing, put it to build through Jucr ? Right?[10:11:52] <lewellyn> sfe and jucr have nothing to do with each other[10:12:05] <lewellyn> some people contribute to one, some to both.[10:13:05] *** CodeWar has joined #opensolaris[10:13:36] <nikolam> Aha. So, could one make repository/publisher containing packages that he made with Sfe?[10:13:50] <lewellyn> sure. there are a few[10:14:14] <nikolam> You know some of them, maybe where they list them etc[10:14:27] *** KermitTheFragger has joined #opensolaris[10:14:40] <lewellyn> they all tend to be private. especially the ones in .us, since the encumbered stuff is kind of illegal here[10:14:48] <nikolam> I believe that is a way for testing etc?[10:15:30] <nikolam> So noone is actually publishing anything form Sfe to the public?[10:15:55] <lewellyn> honestly, i don't know what your goal is. and right now, i really need to finish up what i'm working on. i have housework left after this pile of work's done, and the gf has been nagging me to go to bed for 3 hours.[10:16:12] *** dalibor has quit IRC[10:16:12] <richlowe> sleep is for cowards.[10:16:29] <lewellyn> richlowe: i get very little these days :P[10:16:38] <nikolam> ok lewellyn take care. My goal is to understand porting process and how to connect with people that could test packages[10:16:49] <lewellyn> sw-porters-discuss[10:16:50] *** tsoome has joined #opensolaris[10:17:56] <nikolam> I was using PPA repos intensively on Ubuntu. So basically ono osol one needs to make its own server running for that.[10:18:33] *** merzo has joined #opensolaris[10:18:33] <nikolam> ok[10:18:59] <lewellyn> the closest there is right now is jucr[10:19:33] <nikolam> ok See you :)[10:24:19] <AukeF> I read about Splunk on Ben Rochwoods' blog, it looks cool. I'm wondering if anyone is running it on opensolaris? The downloadable release seems Solaris9/10 only.[10:25:38] *** miamano_ has joined #opensolaris[10:26:03] <zedrich> hey all[10:26:07] *** miamano_ has quit IRC[10:26:14] *** miamano has quit IRC[10:26:24] <zedrich> I tried to delegate all related ACLs to a user... but that user can still not create a volume in zfs[10:26:29] <zedrich> any idea what I'm missing?[10:26:36] <zedrich> http://blogs.sun.com/marks/entry/zfs_delegated_administration[10:26:57] <zedrich> root is able to make the volume, tho[10:27:26] <tsoome> you did set create and mount permissions?[10:27:32] <zedrich> yeahp[10:28:24] <zedrich> zfs allow vm create,destroy,mount,snapshot,send,receive,clone,reservation,quota,share,allow tank/vm/test[10:29:55] *** axisys has quit IRC[10:30:24] <tsoome> can it be it is confusing vm for group?:P[10:30:29] <zedrich> does something like work for you?[10:30:38] <zedrich> isnt group like @vm ?[10:30:42] *** wdp_ has quit IRC[10:31:16] *** wdp_aao__ has quit IRC[10:31:17] <zedrich> and even then.. if there is a vm group.. my user vm might not be in it?[10:31:21] <tsoome> no, @vm would define permission set, which you can assign to user or group[10:32:08] *** niq has joined #opensolaris[10:32:15] <zedrich> ah right[10:32:38] <zedrich> so a zfs allow tank/vm/test shows L+D perms on path user vm allow,clone,create...[10:32:50] <zedrich> is that enough for a user to create a volume, is the questiuon[10:32:59] <zedrich> looking on google doesnt give me a precise answer[10:34:48] <tsoome> tsoome@kass:~$ pfexec zfs allow -u tsoome create,mount data[10:34:48] <tsoome> tsoome@kass:~$ zfs create data/test[10:34:48] <tsoome> cannot mount '/data/test': failed to create mountpoint[10:34:49] <tsoome> filesystem successfully created, but not mounted[10:34:55] *** CodeWar has quit IRC[10:35:19] <zedrich> so.. no?[10:35:20] <zedrich> :)[10:35:20] <tsoome> got error because this used does not have permission to write into data[10:35:24] <tsoome> user*[10:35:34] <zedrich> write into data?[10:35:55] <richlowe> /data/ to create the mountpoint, I guess.[10:36:04] <tsoome> tsoome@kass:~$ pfexec chown tsoome /data[10:36:04] <tsoome> tsoome@kass:~$ pfexec zfs destroy data/test[10:36:05] <tsoome> tsoome@kass:~$ zfs create data/test[10:36:15] <richlowe> I assume similar things apply to zvols[10:36:16] <tsoome> all is working as documented[10:36:25] <zedrich> I chowned it to the user too..[10:36:35] <tsoome> zvols are not mounted[10:36:59] <richlowe> tsoome: Right, but device nodes appear, magic happens.[10:37:34] <tsoome> aha, ye, with volume i do get the error[10:37:52] <zedrich> tsoome: is that documented?[10:37:58] <tsoome> ioctl(3, ZFS_IOC_CREATE, 0x08046190) Err#1 EPERM [sys_mount][10:38:03] <zedrich> that a user with full perms and ACLs cant create a volume?[10:38:17] <zedrich> yeah thats what i get precisely when calling the libzfs library (javazfs)[10:38:32] <tsoome> need to have sys_mount permission[10:38:34] <zedrich> then I tried it by hand... a bit surprised really cuz that path is fully owned[10:38:39] <zedrich> sys_mount.. perms..[10:38:47] <zedrich> how are those configured?[10:39:11] *** smrt has quit IRC[10:40:33] *** smrt has joined #opensolaris[10:41:50] <zedrich> holy god thats complex heh[10:41:54] <zedrich> http://www.archivum.info/zfs-discuss at opensolaris dot org/2007-05/00713/[zfs-discuss]-sys_mount-problem.html[10:42:27] *** dover has quit IRC[10:42:32] *** Aesdyn has quit IRC[10:42:37] *** wdp_ has joined #opensolaris[10:43:16] *** wdp_aao has joined #opensolaris[10:44:24] <tsoome> well, he (in that post) is using different way to delegate zfs administration[10:44:49] <zedrich> perhaps Ill wait til daylight[10:44:56] <zedrich> nice that its just me..[10:45:03] <zedrich> i guess i can run my stuff as root but its sorta rude[10:47:42] <tsoome> you can grant sys_mount with usermod[10:48:29] <tsoome> usermod -K defaultpriv=basic,sys_mount username[10:48:31] *** twisti has quit IRC[10:48:35] <richlowe> but be careful when you do it, it's not additive, follow the example in the manual page.[10:48:46] *** twisti has joined #opensolaris[10:50:09] <zedrich> not additive[10:50:11] <zedrich> ?[10:50:32] <tsoome> imo it still sounds like bug; man zfs does not mention you need to set any extra bits with zfs allow[10:50:39] *** perlmongo has joined #opensolaris[10:50:42] <zedrich> for volumes, yes[10:50:44] <richlowe> It's fairly common (I've done it, seen other people do it) in trying to give yourself an extra priv, to forget to type 'basic'[10:50:48] <richlowe> so you have your extra priv, and *no other*[10:50:49] <zedrich> and I verified that it doesnt work in 133 either[10:50:55] <zedrich> it acts the same[10:51:01] <tsoome> usermod/rolemod/groumod commands will not add attributes, they replace it[10:53:35] *** gazza has joined #opensolaris[10:53:48] <Alasdairrr> I've potentially uncovered an issue with dedup[10:54:03] <zedrich> whoa that did work[10:54:19] <zedrich> richlowe: as well as from java / javazfs library[10:54:25] *** jameslegg has joined #opensolaris[10:54:31] <Alasdairrr> ZFS Sending 5TB of data from host A to host B. ZFS Send starts off fast, but then gets slower and slower and slower until it's practically doing nothing.[10:54:47] <Alasdairrr> Host B has dedup enabled. Disable dedup and it stays consistantly fast.[10:54:56] <zedrich> Alasdairrr: out of interest what protocol are you sending it iwth[10:54:59] <Alasdairrr> netcat[10:55:09] <zedrich> interesting[10:55:14] <tsoome> how much ram does host B have?[10:55:26] <gazza> hi there, has anybody managed to get in touch with their oracles reps yet over oracles reported cancellation of HP oem agreements[10:55:37] <Alasdairrr> Host A has 16GB, host B has just a meagre 4GB[10:56:18] <Alasdairrr> Host A is snv_127 and host B is snv_134[10:56:19] <tsoome> 5 TB data dedupe with 4GB of ram..... i would not do it[10:56:43] <tsoome> dedupe needs lots of ram for speed[10:56:53] <Alasdairrr> Yeah that seems like a possible cause[10:57:13] <tsoome> as with every write you need to scan for checksums, if there is already one[10:57:39] <tsoome> 128k recordsize?[10:57:48] *** jameslegg has quit IRC[10:57:48] <Alasdairrr> Yeah[10:58:01] <tsoome> smaller would be even worse:)[10:58:40] <Alasdairrr> So it would seem likely it went quickly until the table no longer fitted in RAM then it nosedived in speed and got progressively worse[10:58:55] <tsoome> yes[11:00:10] <tsoome> dedupe will enable you to save disk space, but there is an cost - you need to have lots of ram to make dedup to work with decent speed[11:00:27] *** dalibor has joined #opensolaris[11:00:40] <tsoome> that also will imply that you wanna see relatively high dedupe rate to justify that cost[11:00:40] <Alasdairrr> I remember the days when 4GB was a lot of ram[11:00:57] <Alasdairrr> Very high dedupe rate here - backups of identical machines[11:01:01] *** jameslegg has joined #opensolaris[11:01:12] <Alasdairrr> well, mostly identical :)[11:01:13] <tsoome> well, when 4GB was a lot, the 5TB was like huge amount of space[11:01:33] <Alasdairrr> aye[11:01:34] <Alasdairrr> brb[11:01:58] *** RoyK has joined #opensolaris[11:01:59] <tsoome> yes. and with high dedupe rate, we are talking about rates like 6:1 to 10:1[11:01:59] <zedrich> Im guessing with say.... 50GB ..[11:02:02] <zedrich> it works fine[11:02:25] <zedrich> 50gb send sizes that is[11:03:55] <tsoome> there have been some calculations around how mush ram you need, obviously it depends on record size as well.[11:04:10] *** chrisgh has joined #opensolaris[11:04:43] *** it has joined #opensolaris[11:05:29] *** it has quit IRC[11:06:34] *** Beket has joined #opensolaris[11:07:52] *** timsf has joined #opensolaris[11:09:35] *** kwinz3 has quit IRC[11:09:55] *** MuffiXXL has joined #opensolaris[11:13:36] *** ewdafa has joined #opensolaris[11:19:56] *** OclkdMan has joined #opensolaris[11:21:01] *** calum has joined #opensolaris[11:23:11] *** timeless_mbp has quit IRC[11:24:29] *** OclkdMan has quit IRC[11:24:42] *** timeless_mbp has joined #opensolaris[11:24:42] *** timeless_mbp has joined #opensolaris[11:24:47] *** dalibor has quit IRC[11:28:43] *** dalibor has joined #opensolaris[11:29:06] *** Shoggoth has quit IRC[11:32:48] <chrisgh> Framework error: code: 6 reason: Couldn't resolve host 'pkg.opensolaris.org'[11:32:56] <chrisgh> is that common?[11:33:10] <Beket> yes[11:33:20] <chrisgh> why is is like that?[11:33:28] <Beket> dunno, try again later[11:33:36] <Beket> it usually comes back within hours[11:34:04] *** omry|work has quit IRC[11:34:22] *** timeless_mbp has quit IRC[11:34:34] *** jenky has joined #opensolaris[11:35:51] <lewellyn> define a mirror[11:35:55] <lewellyn> smrt: explain ips mirrors[11:35:56] <smrt> At the time of the last factlet update, there is one IPS mirror located at pkg-eu-2.opensolaris.org . The list of officially-sanctioned IPS mirrors is in the pkg(5) FAQ: http://hub.opensolaris.org/bin/view/Project+pkg/FAQ (pkg-na-2 is down indefinitely.)[11:39:30] <taemun> lewellyn: the linked FAQ site says that pkg-na-2 is up[11:39:34] <taemun> or I'm totally misreading it[11:40:36] *** timeless_mbp has joined #opensolaris[11:40:36] *** timeless_mbp has joined #opensolaris[11:41:03] *** kwinz3 has joined #opensolaris[11:42:27] *** dalibor has quit IRC[11:43:11] <chrisgh> what file do I change pkg mirror in?[11:43:41] <chrisgh> ok[11:43:42] <chrisgh> saw it[11:43:44] <timsf> no file, use pkg set-publisher[11:44:25] <lewellyn> taemun: bug the pkg folk. i don't maintain their faq :)[11:44:31] *** causality has left #opensolaris[11:44:32] <taemun> lol[11:44:32] <taemun> oki[11:44:40] <lewellyn> (and i can't get to hub.os.o right now for some reason)[11:44:48] <taemun> it said last updated march 22nd, I presumed that your factlet was older[11:45:24] <lewellyn> the factlet was last updated about 10 hours ago[11:45:34] <lewellyn> i don't monitor the faq ;)[11:45:38] <taemun> apologies, I can't see that :P[11:45:51] <lewellyn> nor can i. but i asked #pkg5 about the mirrors earlier :)[11:46:07] *** victori has quit IRC[11:47:32] <madwizard> Coffee[11:48:19] <lewellyn> hm. anyone remember how to get pkg to use a proxy?[11:48:47] <taemun> export http_proxy= ?[11:48:49] <richlowe> export HTTP_PROXY=http://host:port[11:48:50] <taemun> works here[11:48:53] <lewellyn> the man page says it honors http_proxy[11:48:56] <chrisgh> pkg set-publisher -m http://pkg-eu-2.opensolaris.org/dev opensolaris.org[11:48:59] <lewellyn> richlowe: ah. it takes the wrong case?[11:49:00] <chrisgh> is that the correct syntax?[11:49:11] <lewellyn> chrisgh: i think so[11:49:14] <tsoome> man pkg.[11:49:18] <lewellyn> chrisgh: does it complain? :)[11:49:27] <taemun> export http_proxy='http://proxy.hostname:88'[11:49:28] <taemun> works here[11:49:30] <tsoome> and then /proxy[11:50:00] <lewellyn> "and then /proxy"?[11:50:02] <taemun> err 8080[11:50:14] <tsoome> lewellyn: man pkg ;)[11:50:14] <taemun> missed the zero key :P[11:50:28] <lewellyn> tsoome: 02:54 @lewellyn: the man page says it honors http_proxy[11:50:35] <lewellyn> but it obviously is not ;)[11:50:45] <chrisgh> pkg2 eu does not answer either[11:51:07] *** niq has quit IRC[11:51:26] <taemun> http://pkg-eu-2.opensolaris.org/ talks to me[11:51:30] <taemun> http://pkg-na-2.opensolaris.org/ << doesn't[11:51:31] *** LouisJB_ has joined #opensolaris[11:52:06] <lewellyn> taemun: as smrt said, na is not up[11:52:16] <taemun> was talking to chrisgh here ^[11:52:33] <lewellyn> taemun: he was complaining that pkg.os.o wasn't responding. so i suggested a mirror. ;)[11:52:38] <taemun> ah[11:52:39] <taemun> my bad[11:52:40] <taemun> :([11:53:04] <lewellyn> but yeah. if neither works... can you hit other external hosts by name?[11:53:15] <richlowe> In that instance you don't want a mirror, you want a second origin[11:53:30] <richlowe> since a mirror would still use the regular origin for metadata[11:53:38] <lewellyn> richlowe: is all this documented well somewhere?[11:53:43] <richlowe> (worst. terminology. ever.)[11:53:45] <lewellyn> (since the man page obviously does not)[11:54:25] <richlowe> lewellyn: Don't know, I can't find it either. :\[11:54:39] <richlowe> "mirrors don't have metadata, origins do"[11:55:20] <lewellyn> so is using pkg-*-*.os.o a mirror or origin? how much is cached? *sigh* :([11:55:51] <richlowe> mirrors have all the file data and not much else.[11:55:56] <richlowe> pkg-*-*.os.o are full origins[11:56:06] <richlowe> pkg set-publisher -g http://pkg-eu-2.opensolaris.org opensolaris.org[11:56:08] <richlowe> or whatever.[11:56:36] <richlowe> they'd work as mirrors too, it just wouldn't let you survive the main host being actually down or whatever.[11:56:57] <lewellyn> so they're trying to deliberately mislead us by having called them mirrors :P[11:57:19] <lewellyn> that's like the whole reason for a mirror: to survive pkg.os.o's constant fail :P[11:57:26] <richlowe> it's deliberate and misleading but not deliberately misleading. :)[11:57:42] <richlowe> I think they screwed up and used the term 'mirror' when they shouldn't have, and now are stuck with it.[11:57:50] <lewellyn> pkg set-publisher -m http://pkg-na-2.opensolaris.org/release opensolaris.org # from the faq[11:58:01] <lewellyn> they mention -g, but don't actually say why you might want it[11:58:27] *** tosh has joined #opensolaris[11:58:27] <lewellyn> didn't they change definitions of lots of things once already? why not do the same again? :)[11:59:11] *** Yu\2 has joined #opensolaris[11:59:13] <chrisgh> I must be doing something really wrong[11:59:21] <lewellyn> chrisgh: 02:58 @lewellyn: but yeah. if neither works... can you hit other external hosts by name?[11:59:22] <chrisgh> nonone of those commands work when I try them[11:59:29] <lewellyn> ping www.google.com[12:00:28] <chrisgh> thanks lewellyn[12:00:36] <lewellyn> chrisgh: did it work?[12:00:37] <chrisgh> something is strange, nslookup works but ping doesnt[12:00:43] <lewellyn> did you disable nwam?[12:00:46] <chrisgh> Did not work :|[12:02:05] <lewellyn> the a in nwam means automagic. there's lots of fiddly bits if you disable it, such as this.[12:02:56] <chrisgh> nwadm is not active[12:02:57] <taemun> mv /etc/nslookup.dns /etc/nslookup.conf ##this is probably wrong[12:03:04] <taemun> its something like that[12:03:55] <taemun> told ya[12:03:57] <lewellyn> not mv[12:04:05] <taemun> cp /etc/nsswitch.dns /etc/nsswitch.conf[12:04:16] <lewellyn> chrisgh: this is why you don't disable nwam ;)[12:04:26] <chrisgh> garg[12:04:27] <lewellyn> solaris networking isn't trivial ;)[12:04:28] <chrisgh> ok found it[12:04:40] <chrisgh> lol I thank everyone in the channel for help[12:04:42] <chrisgh> :)[12:04:47] <taemun> lewellyn: if there were robust single step documentation for how to static properly in nwam that didn't need 15 pages of man to explain, I'm sure people would do that[12:04:56] <chrisgh> thanks taemun and lewellyn! :)[12:05:05] <taemun> np[12:05:06] <lewellyn> taemun: pfexec vi /etc/nwam/llp[12:05:16] <madwizard> khm[12:05:26] <chrisgh> my nsswitch.conf was wrong[12:05:27] <lewellyn> s/dhcp/static 123.456.789\/24/[12:05:29] <chrisgh> garg[12:05:31] <madwizard> :default can be described in few sentences[12:05:38] <madwizard> Coffee[12:06:06] <taemun> lewellyn: damn you for knowing that[12:06:07] <taemun> and[12:06:08] <taemun> http://www.google.com.au/search?q=opensolaris+static+ip[12:06:10] <lewellyn> madwizard: but there are fiddly bits. like nsswitch and resolv.conf and route -p add default[12:06:18] <taemun> there needs to be some google bombing done there[12:06:19] <lewellyn> smrt: explain static ip[12:06:19] <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[12:06:31] <lewellyn> taemun: it's documented by the nwam project :P[12:06:36] * lewellyn usually takes the dhcp route[12:07:08] *** betchou has joined #opensolaris[12:07:17] <lewellyn> the d for dynamic refers to the configuration, not that the addresses can't be static :)[12:07:19] <taemun> if googling "opensolaris static ip" doesn't get a usable result, then ... how do you expect people fresh to osol to know to look for/find NWAM documentation etc lol[12:07:29] <lewellyn> taemun: ask the nwam folks[12:07:44] <lewellyn> taemun: poke me this weekend and i'll start writing my nwam docs ;)[12:07:49] <taemun> lol[12:08:28] <calum> (Incidentally, there is a whole new NWAM GUI b135+ that does let you define static IPs etc...)[12:08:38] *** piwi has joined #opensolaris[12:08:50] <calum> er, "in b135+"[12:09:22] <lewellyn> calum: yes. :P[12:09:22] <chrisgh> pkg image-update its creating a plan :)[12:09:27] <lewellyn> calum: in phase 1[12:10:19] <chrisgh> By the way what is the news on Opensolaris 2010 ? Will it be released and when?[12:10:26] *** lateau has quit IRC[12:10:48] <calum> It will be released. Official date is still "first half of the year".[12:10:51] *** kwinz3 has quit IRC[12:11:07] <lewellyn> chrisgh: teh respin's going through RE. if it looks good, it'll either be a release candidate or release. (i don't know what the release plan is, as i'm not privy to that info any more than anyone else.)[12:11:36] <chrisgh> :) Thanks[12:12:19] *** Rune^ has quit IRC[12:12:44] <CosmicDJ> I like my "ask/mail uncle larry" answer more ;)[12:13:13] <lewellyn> CosmicDJ: followed by "be sure to send him a toy yacht for the bathtub"?[12:14:12] *** victori has joined #opensolaris[12:14:47] *** MuffiXXL has quit IRC[12:15:33] <CosmicDJ> so he had to give up his "big" yacht? ah yeah I remember: we're pumping more money than sun in everything...[12:17:40] *** kinkie has joined #opensolaris[12:18:14] *** dalibor has joined #opensolaris[12:18:22] <betchou> chrisgh: This is a recurrent question and the answer is in the topic.[12:19:20] <ivo_> shouldn't 2010.03 be renamed to 2010.05?[12:19:57] <betchou> calum: I think I will love this new GUI. :) There will be order in SSID!! It means you can have two access point available and have the connexion working. :)[12:20:33] <lewellyn> ivo_: they haven't announced the name. be patient.[12:20:55] <calum> ivo_: Not necessarily... previous releases haven't always appeared in the month that their name suggests...[12:21:01] <ivo_> lewellyn let the force be with you too :)[12:21:45] <lewellyn> calum: yes. sun has a long history of releases not appearing quite when expected. good to see the tradition is continuing :)[12:22:01] * CosmicDJ votes for the name Solaris Express 11g[12:22:11] <ivo_> :D[12:22:30] *** sipior has joined #opensolaris[12:22:58] <calum> heh[12:24:06] <calum> betchou: yeah, the new GUI should certainly address a lot of the problems with the current version.. pity it didn't make it into 2010.03, though :/[12:24:49] <ivo_> calum, what ne GUI?[12:25:07] <calum> ivo_: The new NWAM GUI that went into b135[12:25:08] <ivo_> ach I sow it[12:25:44] <calum> (well, it's not just the GUI that's new, it's whole new version of NWAM, really)[12:25:50] <lewellyn> calum: i think phase 1 was ready too late for 2010.03, and the changes are still very new and untested :P[12:26:05] <betchou> yep bad luck[12:26:17] <calum> lewellyn: yeah, it was probably the right decision, but we were originally hoping to make it in.[12:26:18] <richlowe> lewellyn: racey, too, unless that got fixed.[12:26:27] <lewellyn> it's not as simple as say gdm where you can drop a new one into 130 and know it'll be fine by 134 ;)[12:26:37] <lewellyn> calum: so was i :)[12:26:45] <richlowe> I know if I boot phase1 bits in virtualbox I commonly either have no network interface, or have ipfilter rules I shouldn't.[12:26:46] <richlowe> or both.[12:26:46] <lewellyn> richlowe: that is a corollary ;)[12:27:03] <richlowe> since, somehow, the ipfilter service times out adjusting its ruleset.[12:27:12] *** swift has joined #opensolaris[12:27:38] <betchou> I was thinking about to send the most able bug I could right during this waiting time hoping it could be included in the release.[12:27:48] <betchou> write!![12:27:51] <betchou> sorry[12:30:23] *** kwinz3 has joined #opensolaris[12:31:28] <nikolam> I dont understand How 135, 137 could be developed when Ips is on 134.[12:32:02] <nikolam> How do they people deveop it , and do they make .iso s ofr it or update packages or what and how[12:32:18] <tomww> the process is similar to the old build system.[12:32:33] <madwizard> nikolam: Internally they know how to build it[12:32:38] <tomww> there is a bunch of big machines, compiling the ineer parts by a script.[12:32:55] <tomww> then exporting the produced binaries/files to the IPS repo in engineering.[12:32:57] <nikolam> madwizard, and noone ever build anything aoutside oracle?[12:33:00] <madwizard> I wouldn't surprised if within Oracle there are repos for every tagged changeset[12:33:10] <tomww> other parts like gnome and such is added then and ready is the build[12:33:17] <madwizard> nikolam: Yes, people build all the time[12:33:43] <nikolam> madwizard, well, would like to get to know who and where they are.[12:35:36] <nikolam> I think I was reading while ago that IPS have some kind of bugs that make it writable by anyone using it or something. Anyone heard about that?[12:35:41] <madwizard> nikolam: I guess moinak builds his own. :)[12:36:16] <madwizard> nikolam: Go around and ask. I kknow at least one person that builds and boots latest+greatest from hg repo, butI'm not sure they would approve of sharing their name here[12:36:28] *** moepmoep has joined #opensolaris[12:36:35] <lewellyn> i was building prior to the switch to ips ;)[12:37:38] <nikolam> lewellyn, so basically you installed packages directly, without ips before that?[12:37:43] <nikolam> And now you have your own Ips?[12:38:52] <lewellyn> no. my workstation is still sxce[12:40:12] <nikolam> so lewellyn you actually do not use osol?[12:40:13] <nikolam> I dont understand[12:41:45] <lewellyn> i use a now-discontinued distro on my workstation[12:43:05] <nikolam> lewellyn, I was thinking about zou werw building on your own, before switching to ips[12:43:18] <lewellyn> i think that didn't fully translate[12:44:04] <tsoome> i was using sxce as well; and since i had separate disks for my data, my switch over was relatively painless[12:44:26] <lewellyn> tsoome: in my case, i'm waiting for a new machine before switching[12:44:35] <tsoome> fair enough:)[12:45:06] *** hsp has joined #opensolaris[12:45:12] <lewellyn> then i'll just zfs send the datasets i need most urgently and deal with the rest as needed[12:47:20] *** SYS64738 has quit IRC[12:47:45] *** SYS64738 has joined #opensolaris[12:51:08] *** Dagobert_ has joined #opensolaris[12:52:07] <madwizard> Coffee[12:52:25] <madwizard> God, I have a voice just right for death metal band[12:52:33] <madwizard> If it keeps like that I might form one[12:52:48] <madwizard> I know how to play few instruments and vocal talents are not important anyway. :)[12:54:38] *** Dagobert has quit IRC[12:54:38] *** Dagobert_ is now known as Dagobert[12:54:53] * jmcp wanders in[12:54:55] <jmcp> evenin' all[12:55:03] <tsoome> hi[12:55:20] <nikolam> hello[12:55:49] <madwizard> 'lo[12:59:59] *** nitrile has quit IRC[13:02:14] *** nitrile has joined #opensolaris[13:03:18] *** nitrile has quit IRC[13:05:44] *** sergiusens has quit IRC[13:06:03] *** twilling has joined #opensolaris[13:09:00] *** nitrile has joined #opensolaris[13:09:16] *** OclkdMan has joined #opensolaris[13:11:48] *** MuffiXXL has joined #opensolaris[13:15:55] *** nitrile has quit IRC[13:17:17] *** Norrin has quit IRC[13:18:35] *** timeless_mbp1 has joined #opensolaris[13:20:56] *** SYS64738 has quit IRC[13:21:21] *** SYS64738 has joined #opensolaris[13:22:16] *** timeless_mbp2 has joined #opensolaris[13:22:45] *** timeless_mbp has quit IRC[13:23:05] *** timeless_mbp1 has quit IRC[13:24:17] *** OclkdMan has quit IRC[13:29:39] *** calum is now known as calAFK[13:30:37] *** xenol has quit IRC[13:30:49] *** dunc has joined #opensolaris[13:31:06] *** SYS64738 has quit IRC[13:31:33] *** SYS64738 has joined #opensolaris[13:32:34] *** xenol has joined #opensolaris[13:40:39] *** ivo_ has quit IRC[13:42:48] *** merzo has quit IRC[13:47:59] *** zynox has left #opensolaris[13:48:04] *** andersenep has quit IRC[13:48:46] *** merzo has joined #opensolaris[13:51:38] *** calAFK is now known as calumb[13:51:59] *** kwinz3 has quit IRC[13:56:52] *** vmlemon has quit IRC[14:00:38] *** jameslegg has quit IRC[14:02:12] *** TomJ has quit IRC[14:05:54] *** dunc has quit IRC[14:12:28] <fleebailey33> what is the live user and pass?[14:12:46] <trygvis> jack/jack IIRC[14:12:57] *** vmlemon has joined #opensolaris[14:13:03] *** dunc has joined #opensolaris[14:13:04] <trygvis> though I would expect it to switch to scott/tiger soon :P[14:13:12] <asyd> huhu[14:13:13] <fleebailey33> yea[14:13:17] <fleebailey33> that works, thanks[14:13:40] <fleebailey33> is there a cli installer?[14:14:13] *** vraa_ has quit IRC[14:15:21] *** olivier_54 has joined #opensolaris[14:18:34] *** TomJ has joined #opensolaris[14:18:45] *** ziyourenxiang has joined #opensolaris[14:20:01] *** nitrile has joined #opensolaris[14:21:09] *** jollyd has joined #opensolaris[14:21:13] *** jollyd has left #opensolaris[14:21:25] *** mroconnor has quit IRC[14:22:16] <fleebailey33> meh, it boot vesa mode the third time odly[14:22:20] <nikolam> fleebailey33, there is now. get text install .iso from genunix.org[14:22:35] <fleebailey33> nikolam: ill bookmark for the future, thanks[14:22:55] *** nachox has joined #opensolaris[14:22:59] <fleebailey33> so opensolaris is compatible with solaris binaries right?[14:23:47] *** kwinz3 has joined #opensolaris[14:24:09] *** sysspoof has quit IRC[14:25:03] <seanmcg> fleebailey33, generally yes, iff they use the same libraries and correct interfaces.[14:25:29] *** nachox_ has joined #opensolaris[14:27:47] *** nachox has quit IRC[14:30:49] *** robo has joined #opensolaris[14:31:41] <fleebailey33> coolio[14:32:19] <seanmcg> fleebailey33, eg I've a solaris binary compiled on s8, still runs fine on latest osol[14:32:29] <fleebailey33> sweet[14:33:44] *** andersenep has joined #opensolaris[14:36:27] <chrisgh> noob question[14:36:41] <chrisgh> how can I query what versions of the os I have installed in dev?[14:36:51] *** KermitTheFragger has quit IRC[14:36:52] <chrisgh> Like I went from 2009.06 to snv_132 to snv_134[14:36:58] <jmcp> chrisgh: beadm list[14:39:10] <chrisgh> and if you screw up and want to revert you # beadm activate oldname ?[14:39:53] *** biant92 has joined #opensolaris[14:40:15] <jmcp> chrisgh: yes, beadm activate will do that for you[14:40:48] <chrisgh> what would happen if a driver bleeps up in the new version[14:40:51] <chrisgh> and it wont boot[14:41:01] <chrisgh> how would one recover in that scenario?[14:41:31] <jbit> you select the old version from grub[14:41:38] <jbit> grub lists all the boot environments[14:41:59] <chrisgh> aha! :)[14:42:01] <chrisgh> thanks :)[14:42:28] <chrisgh> I come from a Linux/Unix bg and still feel like a Opensolaris noob...[14:43:50] *** nikolam has quit IRC[14:44:15] <jmcp> chrisgh: the learning curve can be steep, but the docs are good[14:44:30] <jmcp> also, we like to think that here in #opensolaris you'll generally get the help you need[14:44:45] <jmcp> bearing in mind, of course, that sometimes the help you need is a pointer to TFM[14:44:50] <jmcp> so take notes :-)[14:45:58] *** nikolam has joined #opensolaris[14:46:25] <chrisgh> lol[14:46:51] <chrisgh> I will take note on that TFM, thanks for the friendly help :)[14:46:52] <jmcp> seriously[14:46:55] <jmcp> you're welcome[14:46:59] <jmcp> there's a bunch of help in /topic[14:47:12] <jmcp> also, docs.sun.com is an excellent place to read up on the technology[14:47:58] *** robinbowes has quit IRC[14:48:04] *** sysspoof has joined #opensolaris[14:48:25] <jmcp> http://www.opensolaris.com/learn/howto_guides.html[14:49:23] *** dalibor has quit IRC[14:51:11] <nikolam> Oh, this so-called multitasking.. I am installing some os on virtualbox, using only one cpu core and osol 131 user interface beaves like a pig with a heart attack(s). Seems like with high disk usage, Any system I used so far behaves erratically (linux, osol, etc)[14:51:22] *** dunc has quit IRC[14:51:47] *** dunc has joined #opensolaris[14:52:07] *** mroconnor has joined #opensolaris[14:52:07] *** mroconnor has joined #opensolaris[14:52:18] <Stric> that's because latency spikes in cpu scheduling will go unnoticed, because they are not long enough for you to notice..[14:52:35] <Stric> but with disks, you get lots of milliseconds here and there, which you can notice[14:53:02] <Stric> get a storage subsystem that's anywhere close to your cpu speed, and you'll stop noticing these problems..[14:53:13] *** xapiens has joined #opensolaris[14:53:26] <tsoome> virtualisation is not technology to be used in every even unthinkable place.[14:53:27] <Stric> (and no, such a system won't be cheap.. disk performance has been waay outrun by cpu performance)[14:53:31] *** hsp has quit IRC[14:53:34] <nikolam> Stric, isnt it more obvious that man sitting at the desktop, typing or clicking on something shoul have highest prioroty, then some background processes or something[14:53:49] <jmcp> nikolam: no, it's not[14:53:51] *** andersenep has quit IRC[14:54:29] <jmcp> nikolam: if you want extra, uber-fast interactive performance, disable all those otherwise unnecessary services that are running on your system. add more ram. configure more swap. tune[14:54:40] <nikolam> well one should be able to set it. its like not only on osol.[14:54:46] <Stric> jmcp: um. more swap? to add interactive performance..?[14:54:54] <jmcp> and get a really good video card[14:54:55] <Stric> I'd say disable swap.[14:54:56] <jmcp> Stric: yes[14:55:00] <nikolam> I have free Ram, atm[14:55:01] <tsoome> Stric: ofc[14:55:09] <jmcp> Stric: only if you've got bazillions of Gb of ram[14:55:19] <Stric> jmcp: that's your "add more ram" part ;)[14:55:22] *** sergiusens has joined #opensolaris[14:55:22] <jmcp> :-)[14:55:27] <tsoome> you wanna have your ram to be used for active data, not for figabytes of unused code[14:55:33] <tsoome> gigabytes*[14:55:46] <nikolam> I think its a problem with scheduler battle desktop vs servers, that is going on on linux, too for some time.[14:55:58] <jmcp> nikolam: you should investigate the IA scheduling class then[14:56:07] <jmcp> and invest in a copy of Solaris Internals[14:56:08] <tsoome> fs cache versus 99% of unused in firefox.[14:56:15] <tsoome> unused code*[14:56:19] <Stric> nikolam: the main issue is that disk subsystems are slow and "unpredictable"[14:56:31] <Stric> tsoome: and then you want to use that "unused code", but it's in swap..[14:56:54] *** timeless_mbp2 has quit IRC[14:56:56] <tsoome> its not swapped out. its paged out[14:57:05] <tsoome> by like 8k chunks[14:57:13] <Stric> you say potato, I say fries[14:57:14] <tsoome> it will get out pretty fast;)[14:57:19] <nikolam> Stric, well, if i dont change disks 3 times a day, Os could see how it behaves.[14:57:25] <tsoome> and yes, sure, there is an trade off[14:58:04] <jmcp> Stric: chips![14:58:25] <Stric> nikolam: still, it's trivial to starve the io subsystem in a regular machine[14:58:44] <nikolam> I was thinking more then once. If there is setting priority for cpu usage for a task, there should be such priorities on disk usage bandwith and time, too.[14:58:47] <Stric> and when that happens.. well.. there's not enough iops to make it seem interactive[14:59:44] <nikolam> There is such thing for network usage in osol. why not for disk, too.[14:59:55] <tsoome> if you have higher requirements than "regular" user, you wont buy "regular" machine.[15:00:14] <jmcp> nikolam: QoS for disks is a much more difficult problem to solve correctly[15:00:40] <Stric> nikolam: due to abstraction layers, caches, blah, whatnot.. it's not always trivial to find out "who" is actually doing the io..[15:00:45] *** calumb is now known as calLNCH[15:01:12] <Stric> nikolam: I'm not arguing against you, I'm just saying that it's a known problem that many has tried to solve.. but it's not solved yet ;)[15:01:33] *** hsp has joined #opensolaris[15:01:49] <tsoome> if i wanna my movie encoder to get all io and i dont mind my minesweeper to lag while i am waiting for encoding to finish;)[15:01:49] <nikolam> Stric, and such problem exist on all known OS`es more or less.[15:01:57] *** |AbsyntH| has joined #opensolaris[15:02:27] <Stric> nikolam: which would give you an additional hint that it's not trivial to fix[15:02:39] <nikolam> but as I understand Zfs have some kind of managing of bandwith inside[15:02:43] *** hsp has quit IRC[15:03:45] <jmcp> that's generally not something you should be trying to tweak[15:03:48] <lewellyn> nikolam: did you give your guest enough ram?[15:03:50] <jmcp> it's not what I'd include in a list of tunables[15:04:00] <lewellyn> if IT is swapping, you'll see performance problems on the host[15:04:13] <lewellyn> that's regardless of host os :)[15:04:36] <nikolam> lewellyn, 1Gig. But I think that it also might have something to do with that Zfs is compressed where it is installing.[15:04:53] <lewellyn> maybe maybe not[15:05:16] <nikolam> It is curious that after a while it returns to more or less normal. I supose its all about disk[15:05:25] <jmcp> nikolam: I think you might need to get a better understanding of the OpenSolaris kernel before making suggestions like that[15:05:29] <tsoome> give it 2 or 4 and see the difference:P[15:05:40] <nikolam> jmcp, I think so, too ;)[15:05:53] <nikolam> tsoome, ;)[15:05:57] <tsoome> anyhow, if you wanna see perfomance, drop the VM layer;)[15:06:42] *** Odin- has joined #opensolaris[15:06:59] <nikolam> well, vm should not much affect other parts , in my theory ;) ok. when.[15:07:24] <lewellyn> nikolam: it does. on any host.[15:07:42] <lewellyn> you're making one computer pretend to be more than one.[15:08:42] <nikolam> lewellyn, and taking vm aside. Did you get tuning disk usage priority per app sometimes. Ever heard of such thing etc[15:10:02] <lewellyn> never heard of such a thing[15:10:12] *** robinbowes has joined #opensolaris[15:10:13] <lewellyn> there's tuning the system for an app, but that's not the same[15:10:25] <snuff-home> mmm not long now till 139 window closes.. wonder if there will be a big surge overnight..[15:11:13] <nikolam> snuff-home, what version do you have installed etc[15:11:29] <snuff-home> heh whatever IPS is.. 134 :)[15:11:36] <snuff-home> i just like watchin the logs go by[15:11:49] <snuff-home> also see that pkg got tagged @ 1H2010 :)[15:11:53] <tsoome> well, it depends a bit. if i make data pool with 8k recordsize and redo pool with 128k recordsize and apply some more tuning and specific spindle setup for my oracle, its the IO tuning for specific app.[15:12:14] *** andersenep has joined #opensolaris[15:12:25] <nikolam> snuff-home, And you are just watching logs. Not making something with it or something.[15:12:46] *** Andrew3 has left #opensolaris[15:13:18] <snuff-home> nope dont have nice pc to build it.. or the real 'wanting' to..[15:13:45] <nikolam> tsoome, ok, that is nice. Only if layer for disk is separated to NAS, then he might see usage as apps and manage it better?[15:13:47] <snuff-home> i'm happy to test the '/dev' pkg builds.. much easier to go 'pkg image-update'[15:14:15] <nikolam> snuff-home, question it. If you want it, could you ;)[15:15:07] *** p3rror has quit IRC[15:15:24] *** hsp has joined #opensolaris[15:15:38] <tsoome> generic filesystems are generic and will try to serve all io, its quite different resource than CPU for example (where you can apply different scheduler policies and cpu groups)[15:16:46] *** hsp has quit IRC[15:17:25] <snuff-home> i'm not sayin its impossible.. i'm saying i = lazy..[15:19:25] * snuff-home gets some zzz's[15:19:48] * nikolam gets some vrrm vrrm[15:20:16] *** timsf has quit IRC[15:20:25] *** sk0rd has joined #opensolaris[15:21:01] *** MuffiXXL has quit IRC[15:21:06] <chrisgh> is there anyway to script # format list or some equivalent command[15:21:42] <chrisgh> that does not require ctrl-c to terminate?[15:21:42] *** gtirloni has joined #opensolaris[15:23:16] <tsoome> format < /dev/null[15:23:18] <nikolam> lol http://www.youtube.com/watch?v=pQHX-SjgQvQ[15:23:23] <tsoome> or echo|format[15:24:24] <chrisgh> thanks tsoome :)[15:25:29] <tsoome> if you device setup is stable, it may be just enough to ls /dev/dsk/*s2 for example;)[15:31:01] *** bklang has quit IRC[15:32:02] *** hsp has joined #opensolaris[15:32:04] *** dalibor has joined #opensolaris[15:32:30] *** nikolam has quit IRC[15:33:21] *** nikolam has joined #opensolaris[15:34:37] *** Openfree has joined #opensolaris[15:36:45] *** p3rror has joined #opensolaris[15:39:46] *** calLNCH is now known as calumb[15:40:09] *** jthunder has quit IRC[15:40:26] *** nikolam has quit IRC[15:41:13] *** nikolam has joined #opensolaris[15:43:08] <RoyK> Is there a way to do a 'nice' scrub without wasting too much disk performance?[15:43:16] *** Openfree has quit IRC[15:44:06] <Andys^> there's a thingy you can change[15:44:22] *** timsf has joined #opensolaris[15:44:24] <Andys^> zfs_scrub_limit[15:44:28] <Andys^> change it to 1 i think[15:44:32] <Andys^> slows down scrub slightly[15:44:48] <madwizard> Coffee[15:46:40] *** nikolam has quit IRC[15:47:13] <strawf> madwizard! you once guided me with non-global zone as router.. yet it remained unclear should I define a vnic bonded to the physical nic for Zone or use the physical directly instead :o[15:47:57] *** hsp has quit IRC[15:48:21] *** nikolam has joined #opensolaris[15:48:37] *** nickw has quit IRC[15:49:20] *** hsp has joined #opensolaris[15:49:27] *** MuffiXXL has joined #opensolaris[15:52:53] *** lattera has quit IRC[15:53:31] *** pablomh has quit IRC[15:54:44] <CIA-33> Erik Nordmark <Erik.Nordmark at Sun dot COM>: 6940729 gate machine panic in ip.c: assertion failed: pkt_len == msgdsize(mp), 6936196 Missing IXAF_IS_IPV4 flag causes punchin to go boom[15:55:40] *** Openfree has joined #opensolaris[15:57:30] *** nikolam has quit IRC[15:58:27] *** nikolam has joined #opensolaris[15:59:23] *** panos has quit IRC[16:02:22] *** nachox_ is now known as nachox[16:03:22] *** InTheWings has joined #opensolaris[16:04:14] *** nickw has joined #opensolaris[16:05:42] <InTheWings> I'm stuck with snv_130. pkg image-update freezes the system when 80%. Is there any log hidden somewhere outside /var/pkg ?[16:05:49] *** Animal-X has joined #opensolaris[16:06:23] <InTheWings> update is done in single-user milestone[16:07:07] *** pablomh has joined #opensolaris[16:07:15] <tsoome> did you update pkg itself first?[16:08:30] <InTheWings> no update required[16:11:20] *** jgracin has quit IRC[16:12:38] *** Openfree has quit IRC[16:12:38] <chrisgh> Do you need fcode when using LSI controllers?[16:12:58] *** niq has joined #opensolaris[16:12:58] *** niq has joined #opensolaris[16:13:06] <chrisgh> I'm trying to update the LSI1068 BIOS[16:13:15] *** omry|work has joined #opensolaris[16:13:19] <chrisgh> it prompts for fcode and EFI bios filename which I do not have :|[16:13:42] *** tsoome has quit IRC[16:15:43] *** GhostInTheShell has quit IRC[16:16:05] <lblume> InTheWings: Why do you do it in single-user? What's wrong with the normal way of updating?[16:16:33] *** pablomh has quit IRC[16:16:38] *** GhostInTheShell has joined #opensolaris[16:16:39] <InTheWings> lblume: I just doesn't ended in multiuser.[16:16:43] <Alasdairrr> chrisgh: You can grab those from lsi's website[16:17:06] <Alasdairrr> Sorry just read, don't need fcode or efi[16:17:11] <InTheWings> lblume: If freezes in single & multi mode. ssh also goes down.[16:17:15] <Alasdairrr> Unless you're on SPARC which I have no idae about ;-)[16:17:22] *** pablomh has joined #opensolaris[16:17:24] <Alasdairrr> But on x86 just the PC BIOS is fine[16:17:31] <lblume> InTheWings: So there is another problem preventing you to get to multiuser, and rather to try to fix it, you do a random update?[16:17:59] *** lattera has joined #opensolaris[16:18:36] *** sysspoof has quit IRC[16:18:52] <InTheWings> lblume: No problem in multiuser, but as update crashes the system, I try in a minimal mode[16:19:50] <lblume> It sounds more like a hardware issue, you should run at least memtest and check if there are no obvious problems like overheating.[16:20:33] <lblume> pkg has had many different ways to crash, but so far, killing the kernel has not been one of them.[16:21:07] <InTheWings> kernel seems ok, as network is still up . Just services goes down & consoles/X freeze[16:21:56] <InTheWings> and it's vboxed[16:21:57] <lblume> pkg does not interact with services.[16:22:00] *** baza11215 has joined #opensolaris[16:22:54] *** sysspoof has joined #opensolaris[16:23:43] <timsf> er, it does a bit[16:23:55] <lblume> Well, I'd say that vbox can be have trouble with bad memory all the same as a real system :-) Or there might be something else rotten. But again, it's very doubtful that pkg is responsible for your issues.[16:24:03] <timsf> but yes, I'd agree.[16:24:29] *** perlmongo has quit IRC[16:25:19] *** macros73 has joined #opensolaris[16:27:20] <lblume> timsf: Well, ok, I'm sure there's some little interaction, but only for its own service, right? Modiofying any other would be forbidden :-)[16:27:48] <InTheWings> And pkg does interact. -v gave a list of things restarted in Actuators:restart_fmri: svc:/system/rbac:default .......[16:28:06] <timsf> right - the actuator stuff[16:28:38] <timsf> but it doesn't do anything that a sysadmin wouldn't be able to do on their own[16:28:43] <InTheWings> But as I can't get a log, I can't tell if the package is related to restart[16:28:52] <InTheWings> I'll try to syslog to network[16:29:27] <timsf> You'd be better off dropping out to kmdb when the system freezes to see if you can tell what processes are running[16:29:32] *** jthunder has joined #opensolaris[16:29:39] *** afrasi has joined #opensolaris[16:29:39] *** Wes-- has joined #opensolaris[16:29:52] <timsf> Have you run a zpool scrub recently btw?[16:30:19] <InTheWings> I'll have a look in kernel debugger[16:31:13] <timsf> What nic are you using?[16:33:03] <RoyK> Andys^: zfs_scrub_limit where?[16:33:07] *** jthunder_ has joined #opensolaris[16:33:31] *** sysspoof has quit IRC[16:35:41] *** jthunder has quit IRC[16:36:14] *** theron has joined #opensolaris[16:37:21] *** jthunder_ has quit IRC[16:41:28] *** Odin- has quit IRC[16:45:16] *** raichoo has quit IRC[16:50:18] <RoyK> Is there a way to do a 'nice' scrub without wasting too much disk performance?[16:50:50] * seanmcg wonders how to 'waste' disk performance :)[16:51:52] <RoyK> "zpool scrub tank" is a good start[16:52:15] *** sergiusens has quit IRC[16:52:28] <TommyTheKid> scub only takes idle cycles i thought[16:52:43] <chrisgh> yep uses lower io priority[16:53:35] *** timeless_mbp has joined #opensolaris[16:53:37] *** timeless_mbp has joined #opensolaris[16:54:14] <RoyK> doesn't really matter what it intends to, i/o speed falls dramatically when scrub is running[16:57:31] *** phimic has quit IRC[17:00:52] *** klii has joined #opensolaris[17:04:12] <chrisgh> does zpool care about physical device names[17:04:26] <longcat> sure, when creating it. otherwise it scans all devices when importing[17:04:37] <longcat> so, they can move around and zfs knows by metadata[17:04:38] <chrisgh> I can clear persistent drive mapping in the RAID controller utility[17:04:43] *** klii has left #opensolaris[17:04:49] <chrisgh> but I do not know how that affects the zpool[17:08:56] *** nsuperbus has joined #opensolaris[17:09:11] *** webchat has joined #opensolaris[17:09:20] <TommyTheKid> chrisgh: I dont think zpool cares much... except for "rpool" and then you just need to make sure that the bios can boot to those disks still[17:09:42] <webchat> Anyone knows when Opensolaris 2010.03 will be released?[17:09:46] <webchat> Hi[17:09:49] <longcat> nobody here is telling if they know[17:09:53] <longcat> the effective answer is no[17:10:09] <webchat> very kind[17:10:20] <webchat> this answer is more than i expected[17:10:23] <TommyTheKid> chrisgh: there was a demo when zfs was new of someone taking a pile of usb thumb drives and creating a zpool on them, then exporting it, and re-inserting them at any order and it mounts it right back up[17:10:34] <longcat> Some people say june based on some stuff but if it was official it would be in the topic[17:11:16] <TommyTheKid> right, any date you hear at this point is just rumor[17:11:21] <betchou> longcat: this is officially in the topic : We know 2010.03 is late. We don't know when it will be released.[17:11:24] <TommyTheKid> it will be ready when its ready[17:11:40] <webchat> I have read that there are 2 bugs waiting to be solved,[17:12:00] <lblume> Hmmm, I'm using a font in OpenOffice that I don't see with fc-list, how come?[17:12:00] *** thepickle has joined #opensolaris[17:12:03] <webchat> one related to zfs mirrror rpool and other I don remember[17:12:51] <webchat> Ok I see my boss near me ... I have to close and act as I am working ...[17:12:56] <webchat> :)[17:12:57] *** p3rror has quit IRC[17:12:59] <longcat> run top, and dmesg in a loop[17:13:13] *** astra-x has joined #opensolaris[17:13:18] <webchat> Bye[17:13:21] *** webchat has quit IRC[17:14:57] *** calumb is now known as calAFK[17:15:22] <alanc> longcat: I know enough to know that no one knows[17:15:41] <betchou> like this one :D[17:15:50] <seanmcg> alanc and thats almost a dangerous amount of knowledge..[17:16:27] <alanc> seanmcg: highly dangerous, since I have to constantly remember how much of it I'm supposed to know, and how much I'm allowed to let others know[17:16:41] <seanmcg> 8-)[17:17:31] * betchou would like a star and a pseudo beginning with many a too :P[17:18:21] *** humpalla has joined #opensolaris[17:18:25] <lblume> Weird, this font is available on the repo, but not currently installed, where does it come from?[17:20:43] *** tsoome has joined #opensolaris[17:21:38] <alanc> I know StarOffice included some additional fonts - don't remember if OpenOffice does[17:22:34] *** betchou has quit IRC[17:22:53] *** romero has joined #opensolaris[17:23:16] <lblume> I thought about it, but didn't find them[17:23:29] *** Bateau has joined #opensolaris[17:23:37] <lblume> And now that I installed the font from the repo, they do appear in fc-list[17:24:35] *** dalibor has quit IRC[17:25:19] *** psychicist has joined #opensolaris[17:26:10] *** lateau has joined #opensolaris[17:26:24] *** lateau has quit IRC[17:26:28] *** lateau has joined #opensolaris[17:26:46] *** skeeziks has quit IRC[17:28:14] *** calAFK is now known as calumb[17:28:18] <lblume> alanc: Dang. I found them. Not in /opt/openoffice.org3, but in /opt/openoffice.org, where there is, for some exceedingly mysterious reason, some odd bits for databases.[17:29:11] *** _setuid_H has joined #opensolaris[17:29:32] *** dalibor has joined #opensolaris[17:30:01] *** timeless_mbp has quit IRC[17:33:43] *** timeless_mbp has joined #opensolaris[17:33:44] *** timeless_mbp has joined #opensolaris[17:35:37] <lblume> What would be a good SATA HBA for an external enclosure those days?[17:35:47] <baza11215> I have a system running snv_134, it has 24 sata disks connected to 3 controllers, the zpool is stripped across 11 mirrors. under heavy disk io the system reports pci errors and just freezes for a bit. It runs fine when booted into snv_131. Any ideas on where to look for the underlying problem?[17:35:54] *** stevel has joined #opensolaris[17:38:28] *** |Inthewings| has joined #opensolaris[17:38:42] *** swift_ has joined #opensolaris[17:39:30] *** swift has quit IRC[17:39:33] *** swift_ is now known as swift[17:40:44] <RoyK> lblume: won't a sas expander be better for that?[17:41:07] *** InTheWings has quit IRC[17:42:06] *** aszeszo has quit IRC[17:45:07] <gebi> lblume: you can get the 8port intel sas controller (with LSI 1068e chip) here for ~130 euro (8port pci-e)[17:45:09] <lblume> You mean, a port multiplier? I've never used those, heard about some issues sometimes[17:45:20] *** Odin- has joined #opensolaris[17:45:36] <lblume> gebi: What kind of connectors does it use?[17:45:48] <gebi> lblume: 4x mini sas[17:46:04] <gebi> SFF-8087[17:46:58] <lblume> Do you know an online vendor in Europe?[17:47:13] *** aszeszo has joined #opensolaris[17:48:00] *** sndcrb has joined #opensolaris[17:49:20] *** vraa has joined #opensolaris[17:50:18] <RoyK> lblume: supermicro has their 8-port pci-x sata controller, but I don't think it has e-sata[17:50:36] <gebi> http://geizhals.at/a385443.html - i usually use mylemon.at, they are really good in austria and also deliver into EU (though no idea about quality at eu deliveries)[17:51:24] <|woody|> http://geizhals.at/eu/a385443.html for complete eu stuff :)[17:51:55] <RoyK> gebi: no external connections on that[17:52:09] <gebi> RoyK: there are long cables[17:52:29] *** yippi has joined #opensolaris[17:52:37] <gebi> and both sas and esata have the same voltage differenze (double than that of sata)[17:52:37] *** yippi has quit IRC[17:53:04] <mbreitba> Quick question - any way to destroy zpools that I created on another box, but don't want to import?[17:53:14] *** yippi has joined #opensolaris[17:53:22] <bdha> You could wipe the disks.[17:53:25] <gebi> |woody|: lol ok, thx :)[17:53:31] *** raichoo has joined #opensolaris[17:53:32] <mbreitba> any quicker way? ;)[17:53:40] <RoyK> mbreitba: zpool create[17:53:42] <RoyK> :)[17:53:48] <lblume> Thanks all, will look into that.[17:54:28] <gebi> lblume: the only problem with the intel controller is for FW upgrades, which are only possible on intel efi server boards[17:54:29] <lblume> mbreitba: You don't need to wipe the whole disk, only the first blocks should be enough to make the zpool disappear[17:54:50] <mbreitba> would creating a new zpool then deleting it wipe those blocks too?[17:54:56] <lblume> gebi: Does it actually need those upgrades?[17:55:12] <CIA-33> Gerald Jelinek <Gerald.Jelinek at Sun dot COM>: 6927649 merge common code from solaris10 and sn1 brands (fix 32-bit x86)[17:55:26] *** Beket has left #opensolaris[17:55:28] <gebi> maybe for extended raid modes, but not for solaris and zfs[17:56:43] <RoyK> mbreitba: if you want a secure erase, just dd /dev/zero over them[17:56:49] *** bobbyz has joined #opensolaris[17:57:08] <RoyK> with modern drives, single overwrite suffices[17:57:32] <rkeene> Alternatively, you can do what we do...[17:57:35] <longcat> that's what we're told[17:57:35] <RoyK> the density is far too high to allow for electron microscope recovery[17:57:38] <rkeene> (It's much more fun)[17:57:48] <rkeene> Melt them down :-P[17:57:54] <RoyK> hehe[17:57:57] *** twisti has quit IRC[17:58:12] <mbreitba> I just don't want opensolaris to see them as having been part of a pool previously[17:58:27] <RoyK> mbreitba: just create a new pool on them[17:58:47] <RoyK> zpool create -f dev1 dev2 ....[17:59:05] <RoyK> add a pool name in there too, and it'll be happy[17:59:10] *** twisti has joined #opensolaris[17:59:16] <lblume> gebi: How do you get the ports to the drives in an external enclosure? I've been using a multimae cable for 4 disks so far, I rather like it, anything similar?[17:59:52] <Wes--> Does anybody know exactly what /bin/sh is on OpenSolaris? It's sure not the same shell as 5.10 -- aside from the bash-like command-line editting, "typeset -F[17:59:56] <RoyK> I'd recommend using esata for external stuff[17:59:57] <Wes--> " no longer works[18:00:12] <longcat> lrwxrwxrwx 1 root root 9 2010-01-28 14:07 /bin/sh -> i86/ksh93[18:00:23] <Wes--> longcat: AH! Thanks. :D[18:00:54] *** gazza has quit IRC[18:00:56] <lblume> Well, eSata means one cable/disk, right?[18:01:26] *** theron has quit IRC[18:01:27] *** nachox_ has joined #opensolaris[18:02:59] <RoyK> lblume: unless you use a port multiplier, yes[18:04:11] *** dalibor has quit IRC[18:04:26] *** nachox has quit IRC[18:04:38] <RoyK> lblume: SAS expanders are quite nice for that[18:06:13] <lblume> Well, first I'd need to bring those SFF-8087 out anyway, I'm googling that.[18:06:52] <|Inthewings|> no luck with kmdb. Process tree shows pkg running and 0x...::findstack is too big :/[18:07:12] *** ziyourenxiang has quit IRC[18:08:24] <lblume> How much memory did you allocate to the VM?[18:08:50] *** blizzow has joined #opensolaris[18:11:03] <|Inthewings|> 1.2Gb[18:11:06] <|Inthewings|> 1.2GB[18:13:27] <lblume> should be enough.....[18:13:37] <|Inthewings|> lblume: not what ::memstat says[18:14:06] <|Inthewings|> kernel 50%, anon 49%, Free 1% (6mB)[18:14:18] <|Inthewings|> lblume: thanks for ur help[18:15:15] <lblume> That it uses the memory doesn't mean that it's lacking it. Zfs uses as much as possible but relinquish it to the apps.[18:16:33] <|Inthewings|> ZFS file data: 1%[18:16:54] <|Inthewings|> Page Cache 0%[18:17:13] <|Inthewings|> Free (freelist) 1% (6MB) . So that's OOM[18:18:01] *** MuffiXXL has quit IRC[18:18:08] *** raichoo has quit IRC[18:20:09] *** olivier_54 has quit IRC[18:20:13] <Alasdairrr> Okay dedup is rather fruity.[18:20:28] <Alasdairrr> Deleting a multi-TB zfs dataset with dedup on is taking more than 4 hours[18:20:31] <RoyK> fruity?[18:20:55] <RoyK> Alasdairrr: how much memory / l2arc do you have in the box?[18:21:29] *** perlmongo has joined #opensolaris[18:21:36] <Alasdairrr> A meagre 4GB, not enough I know. I suspect the dedup table not fitting in ram is the cause for a lot of this - but there's no paging going on[18:21:47] <RoyK> dedup eats about 150 bytes per data block, that is 1,1GB per 1TB if all are 128kB blocks, which they aren't[18:21:57] *** yippi has quit IRC[18:21:58] <Alasdairrr> But the disks storing the data zpool are thrashing atm[18:22:16] <RoyK> just don't reboot the system now[18:22:31] <RoyK> it'll hang with "mounting zfs filesystems" until it's done[18:22:40] <Alasdairrr> *nods*[18:23:04] <RoyK> our 30TB box hung for like 16 hours last something like that happened[18:23:11] <RoyK> I've heard of boxes hanging for days[18:23:15] <Alasdairrr> I take it a delete of a dedup'd filesystem requires counting lots of references[18:23:35] <RoyK> how big is the pool?[18:23:37] <Alasdairrr> Would more ram fix that or is that just "how it works" and to be expected?[18:23:40] *** derchris has quit IRC[18:23:45] *** derchris has joined #opensolaris[18:23:46] <Alasdairrr> 10TB, 6x2TB in RAIDZ1[18:23:48] <lattera> that's why I don't use dedup anymore, lol[18:24:28] <RoyK> Alasdairrr: more memory will probably fix that, but I filed a bug about it and a fix is said to be in progress[18:24:39] <RoyK> it's about multithreading ZFS's guts[18:24:50] <Alasdairrr> Yeah - dedup would be super awesome for our backups server but I think it needs a bit more work.[18:25:10] <RoyK> L2ARC is said to help too - using it?[18:25:41] <Alasdairrr> No - never needed to. It's just a backup box.[18:25:54] <Alasdairrr> It's write heavy[18:26:28] <RoyK> I have a 12TB (8x2TB in RAIDz2) test system at work - getting more memory tomorrow, so I'll do some more testing with dedup then[18:26:34] <Alasdairrr> Cool[18:26:37] *** Erwann has quit IRC[18:27:10] <RoyK> I guess I'll be getting another two X25-Ms for l2arc/zil as well[18:27:30] <RoyK> I know -m isn't the best for zil, but it'll probably help a bit[18:27:38] <Alasdairrr> I've heard putting the ZIL on consumer grade SSDs is bad because they lie about fsync[18:27:42] <jamesd2> if you are using dedup, and not enough ram.. even a slow usb flash drive will help.. because disk performance is that bad...[18:27:47] <Alasdairrr> even the X25E[18:28:12] <RoyK> x25e is good if you turn off caching[18:28:40] <RoyK> I guess the same applies to all SSDs[18:29:10] <RoyK> x25e is fast, but it lies about cache flush (that is, it ignores them)[18:32:34] *** LaidBack01 has joined #opensolaris[18:36:06] *** hajma has quit IRC[18:40:47] *** lblume has quit IRC[18:41:30] *** bubbva has joined #opensolaris[18:45:45] *** Dagobert has quit IRC[18:47:05] *** theron has joined #opensolaris[18:47:15] <RoyK> Alasdairrr: the workaround is to wait until the zfs destroy finishes, shut down the machine, fill up your server with SSDs and memory, configure Zil and L2ARC and try again :D[18:48:44] *** timsf has quit IRC[18:49:33] <dnaumov> RoyK: -M is a bad idea for ZIL, because they have a mere 90mb/s write, ie any modern, even shit SATA disk will be faster[18:49:38] *** bklang has joined #opensolaris[18:50:01] *** LouisJB_ has quit IRC[18:50:44] *** dunc has quit IRC[18:51:04] *** swift has quit IRC[18:51:36] <RoyK> dnaumov: 90MB/s is about top speed for a modern SATA drive (perhaps 110MB/s outer rim), but even x25m has a far lower seek time than spinning drives, and with a bunch of (slow) 2TB drives in the back, it'll help[18:52:26] <dnaumov> RoyK: my 2 year old 1,5tb seagate does over 105mb/s[18:52:27] *** swift has joined #opensolaris[18:52:45] <dnaumov> RoyK: WD Blacks and Raptors do well over 110mb/s[18:53:36] <RoyK> dnaumov: that's just outer rim, not seek included[18:53:42] <RoyK> inner-rim is half of that[18:53:48] <doggiee> I evaluated X25-E and X25-M as ZIL and in my environment -E was just 15% better than -M[18:53:59] <eviljames> yeah, practical throughput is probably closer to 60MB/s[18:54:32] <dnaumov> RoyK: I am failing to see how/why seek latency would have a meaningful influence of performance of a ZIL vdev[18:54:33] <vraa> hi, can someone explain where the problem is in this http://osol.pastebin.com/sN8Gg1Sv <-- is zfs the culprit, or rsync, or osol?[18:54:38] <dnaumov> RoyK: for l2arc, sure[18:54:47] <CIA-33> Peter Rival <Frank.Rival at oracle dot com>: 6778289 vm locks need to scale with the size of system (strands/memory size) (fix debug panic)[18:55:21] *** wrapster has joined #opensolaris[18:55:21] <RoyK> dnaumov: even Zil is read/written[18:55:40] <wrapster> during the installation of osol.. how does the date and timezone get set?[18:55:52] <tsoome> zli slog is read only in case of crash recovery[18:56:04] <wrapster> I mean after i select a particular timezone how is it going to set it..Which script will do that job?[18:56:20] <wrapster> im facing weird date issues.. and my processes are going bonkers.[18:56:30] <tsoome> in normal cases you wont see reads from zil slog[18:56:59] *** swift has quit IRC[18:57:35] <vraa> seems like all i had to do was "touch filename" and that resolved the issue[18:58:46] *** nikolam_ has joined #opensolaris[18:59:47] *** nikolam has quit IRC[19:00:06] *** lateau has quit IRC[19:00:16] *** timeless_mbp has quit IRC[19:00:34] *** sipior has quit IRC[19:00:50] *** skeeziks has joined #opensolaris[19:02:24] *** fOB has joined #opensolaris[19:03:35] *** niq has quit IRC[19:07:03] *** n1c0b has joined #opensolaris[19:07:10] *** n1c0b has quit IRC[19:07:47] *** |AbsyntH| has quit IRC[19:11:04] *** Tusk has quit IRC[19:11:21] *** Tusk has joined #opensolaris[19:12:47] <Stric> dnaumov: ZIL isn't all about write throughput, but write latency.. and l2arc is used for random reads, where x25-* is quite good and a consumer sata drive sucks bigtime[19:13:11] *** deet has quit IRC[19:14:46] *** bobbyz has quit IRC[19:14:58] *** xapiens has quit IRC[19:19:29] *** twisti has quit IRC[19:19:38] *** fOB has quit IRC[19:19:43] *** twisti has joined #opensolaris[19:20:45] *** n1c0b has joined #opensolaris[19:22:42] *** calumb has quit IRC[19:27:38] *** deet has joined #opensolaris[19:29:30] <wrapster> anyone willing to look at my query pls[19:35:38] *** cybernd has joined #opensolaris[19:37:24] <gebi> doggiee: don't forget to disable write cache in your x25-e ZIL otherwise you have data corruption in case of powerloss[19:38:10] <gebi> and yes, this will cut your write iops in half, including a possible severe hit on your ssd lifetime (which only intel knows)[19:39:13] *** n1c0b has left #opensolaris[19:39:22] <RoyK> tsoome: for users, yes, but it's read from after it was written to[19:40:21] <Triskelios> gebi: the ZIL is checksummed so the worst case ought to be that you lose some transactions[19:41:11] <tsoome> RoyK: well, you wanna explain why my log device does not have any reads?:P[19:41:37] <gebi> Triskelios: i wouldn't bet too much on this, intels have what 128K block size? so the chance is quite high that they even destroy already written daten[19:41:40] <TommyTheKid> zil is only READ if the system fails (power?) before the data is written to disk[19:41:46] <TommyTheKid> as I understand it[19:42:06] <gebi> Triskelios: and even if it's checksummed, missing things in zil means lost data[19:42:12] <RoyK> tsoome: obviously the kernel must read from zil after it's written to it[19:42:15] <longcat> sounds like a wastefully high block size[19:42:17] <tsoome> no[19:42:45] <TommyTheKid> if its just written to zil, its still in ram, then it starts writing to "disk"[19:42:56] <TommyTheKid> it doesn't re-read it from the zil to write it to disk[19:42:57] <tsoome> zil is written to log to make sure its on non-volatile storage, then its played on real data luns[19:43:03] *** chrisgh has left #opensolaris[19:43:09] <longcat> ive found using an external zil that lots of stuff i do doesnt even touch it[19:43:14] <TommyTheKid> ... unless of course bad things happen between the ZIL write and the DISK write[19:43:14] <RoyK> so with insufficient memory, zil is worhless?[19:43:26] <Stric> RoyK: zil is used for sync writes only[19:43:31] <tsoome> the whole point of zil slog is to make sure the transaction log is present at recovery[19:43:48] <RoyK> ok[19:44:04] <Stric> which then turns it into a sync write into zil, and also put into regular disk write cache to be written async..[19:44:05] <TommyTheKid> mail servers tend to make sync writes... many many small sync writes[19:44:16] <RoyK> what was the recommended zil size again? half of ram or so?[19:44:30] <longcat> i think it's a function of how much your disks can write in x time[19:44:31] <tsoome> thats maximum[19:44:33] <Triskelios> gebi: sure, it extends the window of transactions that would be lost but it's only a matter of window size. if performance/longevity is more important to you, that may be an okay tradeoff[19:44:41] <Stric> if that goes well, then it's all fine and dandy.. if the server crashes between zil has gotten its stuff, but the async write hasn't happened yet - then the zil will be read at start[19:45:00] <Stric> that's the only time zil will be read[19:45:27] <TommyTheKid> i think one of those battery backed ram disks would make a neat zil ... it would probably not need to be more than a few gigs?[19:45:34] <tsoome> this log on my redo logs is currently 168k in use[19:45:43] <tsoome> and redos are using 128k blocks[19:45:47] <Stric> I think the recommendation is half your ram[19:45:49] <longcat> can your disks write a few gigs in 5 seconds?[19:46:31] <stallion_work> longcat: thats a neat trick.[19:46:45] <gebi> Triskelios: it's not clear if disabling writecache has a lifetime impact on the x25-e[19:46:59] <tsoome> the size of slog depends how big sync writes are, as up to 128k data blocks are written in slog with transaction log[19:47:19] <gebi> though it would be interesting to know[19:47:27] <TommyTheKid> disabling write cache shouldn't change the lifetime impact, its still going to be written to the disk one way or another?[19:47:36] <RoyK> tsoome: and it won't cache more than half you RAM?[19:47:57] <TommyTheKid> it will just slow it down[19:48:35] <RoyK> but why the limit?[19:48:36] <Stric> TommyTheKid: the flash has an internal block size.. if you're writing 1 byte, it still needs to rewrite the entire block[19:48:49] <TommyTheKid> eww[19:48:56] *** bobbyz has joined #opensolaris[19:48:57] <Stric> TommyTheKid: so the write cache will turn 10000 x 1 byte writes, into a single write[19:49:00] <Stric> instead of 10000 of them[19:49:37] <TommyTheKid> hopefully your application is not making 1 byte sync writes, but I wouldn't be surprised :)[19:49:38] <tsoome> http://www.solarisinternals.com/wiki/index.php/ZFS_Best_Practices_Guide#Separate_Log_Devices[19:50:13] <tsoome> The maximum size of a log device should be approximately 1/2 the size of physical memory because that is the maximum amount of potential in-play data that can be stored.[19:50:17] <Stric> TommyTheKid: probably not, but what if the flash block size is 512k or so? then your 4k writes will surely make lots more writes than it would need to[19:50:41] <wrapster> guys pls look into my issue..[19:50:47] <wrapster> would really be helpful[19:51:10] <stallion_work> wrapster: please stop spamming the channel. If you have specific questions, ask, otherwise do a little googling.[19:51:18] <stallion_work> (you've been at this for a couple of weeks now)[19:51:27] <wrapster> stallion_work: i already asked..[19:51:38] <wrapster> and no amount of googling is helping me[19:51:47] <Stric> wrapster: probably /etc/TIMEZONE[19:51:49] <RoyK> zfs best practice guide doesn't list dedup and its requirements[19:51:55] <wrapster> and when i ask here ppl are just rejecting this..[19:51:56] <TommyTheKid> the date and timezone are set interactively during the install[19:52:03] <TommyTheKid> on the screeen with the big map[19:52:26] *** niq has joined #opensolaris[19:52:26] *** niq has joined #opensolaris[19:52:28] <wrapster> Stric: i know all about that.. But my point is all these are giving me vague results..[19:52:34] <longcat> For a target throughput of X MB/sec and given that ZFS pushes transaction groups every 5 seconds (and have 2 outstanding), we also expect the ZIL to not grow beyond X MB/sec * 10 sec. So to service 100MB/sec of synchronous writes, 1 GBytes of log device should be sufficient.[19:52:43] <Stric> wrapster: maybe due to your vague question[19:52:47] <stallion_work> wrapster: http://tinyurl.com/3ycr9hr[19:53:18] <wrapster> Stric:/ TommyTheKid : I've tried all the cmds.. set the /etc/default/init file manually to the timezone I want and yet when i issue the 'date' cmd i see horrible results.[19:53:20] <tsoome> wrapster: timezone is in /etc/default/init, whish is linked with /etc/TIMEZONE[19:53:47] <Stric> wrapster: "horrible results" is very technical and to the point..[19:53:48] <tsoome> after you change timezone in /etc/default/init, you need to reboot[19:53:58] <tsoome> as this file is read by init at start[19:54:12] <longcat> recently while fixing time issues, i had to learn about 'rtc'[19:54:13] <Stric> and the environment variable will be inherited through the processes[19:54:16] <RoyK> tsoome: can't you change timezone without rebooting?[19:54:37] <longcat> it's like a libc security upgrade. you can upgrade it, but cannot be sure everything is using the updated library until a reboot[19:54:42] <TommyTheKid> of course you could set it temporarily using (for example) --> TZ=US/Mountain date[19:54:44] <wrapster> tsoome: yes even then i see the same issue.. even after reboots the time is not reported properly.[19:54:46] <tsoome> no, unless you restart all apps /which is same as reboot/[19:55:03] <longcat> wrapster: maybe you need to look into rtc[19:55:05] <longcat> 'rtc'[19:55:14] <wrapster> TommyTheKid: that exact cmd i've tried a lot of time on my machine...[19:55:18] <wrapster> without luck.[19:55:22] <longcat> rtc (1m)[19:55:22] <wrapster> let me just do a pasite.[19:55:25] <stallion_work> tsoome: is this in a VM?[19:55:30] <wrapster> you will all understand.. pls hold on[19:55:33] <tsoome> ?[19:55:44] <philsnow> i think he means pastie ?[19:56:18] <tsoome> wrapster: man rtc[19:56:36] <wrapster> longcat: looked at it weeks ago with no help.[19:56:38] <tsoome> and more /etc/rtc_config[19:56:53] <Stric> wrapster: please get technical and show the "horrible result"[19:57:04] *** _setuid_H has quit IRC[19:57:06] <longcat> wrapster: ah. well 'rtc' fixed my timezone issues.[19:57:09] <Stric> not just saying it's wrong, it's bad, it's horrible, you're out of luck etc.. that's not helping.[19:57:20] <wrapster> Stric: http://pastie.org/937673[19:57:27] <wrapster> that pastie was taken just now.[19:57:38] <RoyK> if I want a raid10 setup, what would the zpool create command look like?[19:57:59] <stallion_work> wrapster: you idiot.[19:58:03] <tsoome> mirror disk1 disk2 mirror disk3 disk4 and so on[19:58:06] <RoyK> I mean a mirror of striped drives, not a bunch of mirrors[19:58:10] <TommyTheKid> stallion_work: settle down[19:58:12] <Stric> wrapster: .. and.. what's horrible about it?[19:58:23] <wrapster> Stric: did you see the time?[19:58:24] <tsoome> you cant do raid01[19:58:26] <stallion_work> TommyTheKid: I've quickly losing patience.[19:58:31] <Stric> wrapster: yeah. your clock is wrong.[19:58:33] <wrapster> the current time in PDT is 11 AM[19:58:34] <stallion_work> wrapster: you need to adjust your clock.[19:58:38] <TommyTheKid> it looks like your click may be set wrong wrapster, other than that, I don't see any problems.. it says PDT (pacific daylight time)[19:58:48] <RoyK> tsoome: meaning what?[19:58:59] <stallion_work> s/I've/I'm/ rather[19:59:05] <TommyTheKid> wrapster: try ntpdate pool.ntp.org[19:59:09] <tsoome> you cant mirror stripes. you can stripe mirrors only:)[19:59:09] *** twisti has quit IRC[19:59:29] <RoyK> ok[19:59:35] <tsoome> its documented in man zpool in vdevs explanation[19:59:45] <RoyK> guess it works out more or less the same[20:00:04] <stallion_work> RoyK: not really. One is safe, the other is not.[20:00:08] <Stric> RoyK: actually, no.. striped mirrors offers way more redundancy than mirrored stripes[20:00:09] <TommyTheKid> i never knew 64MB was the minimum vdev size, what if I wanted to use these 8mb CF disks they sent with my camera :p[20:00:16] *** gebi has quit IRC[20:00:29] <stallion_work> TommyTheKid: you need a newer camera ;)[20:00:29] <wrapster> TommyTheKid: did that[20:00:32] <Stric> TommyTheKid: then you should get a CF from this side of the 1700's[20:00:35] *** twisti has joined #opensolaris[20:00:41] <RoyK> stallion_work, Stric how_[20:00:43] <RoyK> ?[20:01:00] <tsoome> wrapster: check out your bios clock and rtc setting:P[20:01:00] <Stric> RoyK: in your mirrored stripe setup.. if one disk fails, you're totally out of redundancy[20:01:03] <stallion_work> RoyK: there is no redundancy between stripes; if a strip is corrupted, then mirroring it only mirrors the corruption.[20:01:04] <TommyTheKid> wrapster: pastebin the results of the ntpdate command and then your command again[20:01:04] <wrapster> TommyTheKid: and now the time is displayed correctly.[20:01:14] <TommyTheKid> ok then, your time was just set incorrectly[20:01:22] *** xapiens has joined #opensolaris[20:01:36] <RoyK> true - perhaps I was thinking raid10 was like raid01[20:01:37] <Stric> wrapster: next time.. be specific. be technical.[20:01:50] <stallion_work> TommyTheKid: ntpdate is obsolete btw.[20:01:58] <TommyTheKid> and pastebin right away, thats absolutely the most helpful to SEE the problem[20:01:59] <stallion_work> You should be using ntpd[20:02:06] <Stric> RoyK: it all depends on how people define "raid10" and "raid01", because some people mix them up[20:02:14] <RoyK> is there a way to setup raidz of mirrors?[20:02:15] <TommyTheKid> ntpdate is the "first" command, then you setup ntp[20:02:18] <Stric> RoyK: but you want what zfs can do, you don't want the other useless kind[20:02:34] <Stric> RoyK: don't think so[20:02:34] <TommyTheKid> ntpd tries to be too clever adjusting the clock slowly[20:02:35] <stallion_work> TommyTheKid: I assure you, it is not.[20:02:45] <tsoome> SVM creates raid01 which in fact behaves like raid10, but with zfs you only can create raid10[20:02:55] <RoyK> raid[56]1 in raidz would be nice[20:02:59] <tsoome> raid 10, raid 50 and so on[20:03:04] <stallion_work> TommyTheKid: read the bottom of ntpdate(1M)[20:03:09] <TommyTheKid> ntpdate is a one time sync, which involves no other configuration files, right ;-/[20:03:21] <stallion_work> TommyTheKid: ntp -q works just as well.[20:03:21] <RoyK> or raid1[56][20:04:10] <TommyTheKid> thats fine[20:04:13] <wrapster> Stric: http://pastie.org/937682[20:04:15] <tsoome> tbh, ntpd footprint is quite small[20:04:50] <Wes--> tsoome: That's been true of SVM since ~ SDS 4.2 (where possible). Can you think of any case where you *wouldn't* want that?[20:05:02] <tsoome> and with time, its always best not to jump it but correct in small steps, thats why ntpd is best[20:05:20] <tsoome> Wes--: never told i dont want that;)[20:05:22] <Stric> tsoome: but ntpd wouldn't have fixed wrapster's time.. it was too much off..[20:05:27] <tsoome> and yes, thats nice[20:05:34] <tsoome> Stric: true[20:05:45] <TommyTheKid> ntpdate is there, ntp -q (ntpd -q is not) ... probably somethig hoakey like in.ntpd?[20:06:39] <wrapster> TommyTheKid: yes i only have ntpq and ntpdate[20:06:48] <TommyTheKid> oh nevermind, its hidden in /usr/lib/inet[20:07:02] <TommyTheKid> very clever sun, very clever ;)[20:07:17] <tsoome> you have smf service for in.ntpd[20:07:19] *** jacotton has joined #opensolaris[20:07:53] <stallion_work> TommyTheKid: hokey? Thats quite common for SYSV actually.[20:08:07] <TommyTheKid> I agree stallion_work it does appear to be deprecated, but Oracle should move ntpd to a reasonable place before removing ntpdate or just alias it or something[20:08:28] <stallion_work> TommyTheKid: there is nothing unreasonable where the inet daemons are placed.[20:08:31] <tsoome> you dont run ntpd manually in normal cases[20:08:48] <tsoome> just as you dont run in.telnetd[20:09:24] *** nsuperbus has quit IRC[20:09:39] <tsoome> extreme example, i know:P[20:10:05] <TommyTheKid> so, it seems like the problem might be that running ntpd -q doesn't follow someone's standard of what a daemon is and how it should be run?[20:10:39] <wrapster> ok guys just to verify it now.. I changed the TZ from US/Pacifc to Europe/Berlin and rebooted.. and then again i see the date cmd gives me this.. "Tue Apr 27 02:16:09 PDT 2010" ..[20:10:51] <tsoome> in solaris, since solaris 10 was introduced 5 years ago, you manage services with smf.[20:10:54] <wrapster> I used rtc -z Europe/Berlin; rtc -c ; reboot[20:11:04] *** TBCOOL has quit IRC[20:11:25] <wrapster> but after the machine came up i could see that /etc/default/init still had US/Pacific written.. why so..[20:11:25] <wrapster> ?[20:11:35] <tsoome> wrapster: did you set /etc/default/init to Europe/Berlin as well?[20:11:49] <wrapster> tsoome: no.. does rtc not set it?[20:11:50] <tsoome> did you even bother to read man rtc[20:11:55] <wrapster> yes[20:12:10] <tsoome> rtc manages only how bios clock is mapped to solaris clock[20:12:16] <tsoome> not how timezone is set[20:12:34] <TommyTheKid> we usually keep our RTC set to GMT[20:12:51] <wrapster> ok.[20:12:57] <wrapster> let me give this another shot.. one sec.[20:13:10] *** bobbyz has quit IRC[20:13:12] <tsoome> the issue with PC is that different OS'es will assume different bios time setup[20:13:30] <tsoome> which is problem with multiboot[20:14:18] <TommyTheKid> only with certain OS's made my certain companies based in Redmond, WA ?[20:14:37] <cfs> Yes.[20:14:56] *** ivo_ has joined #opensolaris[20:15:10] <bonsaikitten> ah, gameloaders[20:15:32] <wrapster> again the date is wrong.. even after changing it in the /etc/default/init to 'Europe/Berlin' the date is 'Tue Apr 27 11:20:04 CEST 2010' while the time should have been 8:20 PM[20:15:33] <tsoome> but regarding to TZ, yes, i agree its silly you need to reboot for global TZ change.[20:15:36] *** Mdx4 has joined #opensolaris[20:15:39] *** olivier_54 has joined #opensolaris[20:16:05] <tsoome> wrapster: did you check your bios time?[20:16:18] <wrapster> yes.[20:16:24] <wrapster> its configured to my local time PDT[20:17:22] <tsoome> but you just told your system its CEST with rtc, did you not?[20:17:27] <TommyTheKid> after making all your changes, probably just run ntpdate pool.ntp.org again to make sure its right[20:17:55] <tsoome> tbh, thats exactly the reason why its good idea to set bios to GMT[20:18:10] <TommyTheKid> the rtc command is a rather "large hammer" .. only use it if you multi-boot and need the clock to be in a timezone other than GMT[20:18:41] <wrapster> tsoome: yes my bios reflects the PDT time which is where i live. TO verify if the changes that you all had specified , would work or not.. I changed the TZ to Berlin and rebooted. and there its failing..[20:19:05] <wrapster> tsoome: so your telling me everything tehre is a TZ change i should also go and change my BIOS time[20:19:20] <tsoome> if you have GMT you dont have to;)[20:19:27] <TommyTheKid> ok, so if BIOS time is set to PDT, then make sure "rtc" shows that too[20:20:12] <TommyTheKid> admittedly, "rtc" on my current sunray server (just installed a day or two ago) shows US/Mountain... I thought we used GMT, guess not :)[20:21:00] <tsoome> i think it defaults to "default" TZ you set at install time[20:21:24] <TommyTheKid> which is reasonable, espcially on x86 live CD installs[20:21:25] <tsoome> its referring to my local tz here too[20:21:58] <wrapster> TommyTheKid: ahh. ok so let me just clarify.. if my bios is set to PDT , then im not suppose to change the TZ using rtc cmd? I should only edit the /etc/default/init file to change the TZ and reboot.. this will give me the new date/time of whatever TZ i choose? Is that correct?[20:22:28] <TommyTheKid> right, /etc/default/init is the Solaris timezone, /etc/rtc_config (rtc) is the BIOS clock timezone[20:22:41] <tsoome> set rtc to PDT then do env TZ=CEST date and verify its correct one[20:22:43] <ivo_> I am wondering if opensolaris will run on AMD chipset 890FX with Phenom x6[20:22:55] <TommyTheKid> /etc/default/init is the system wide timezone[20:23:13] <TommyTheKid> each user can set their own timezone by setting "TZ=whatever" in their .bash_profile (or whatever)[20:26:00] <TommyTheKid> you can even temporarily set it like "TZ=Europe/Berlin date" or if you want it for the whole shell session, try "export TZ=Europe/Berlin" ... then all future commands will have that timezone till you change it or exit that shell[20:27:46] *** afrasi has quit IRC[20:28:24] <TommyTheKid> all this time discussion has led me to a problem I have... its 12:33 and I havent started lunch yet :)[20:28:25] <TommyTheKid> later[20:28:44] *** Qiller has joined #opensolaris[20:29:25] <wrapster> TommyTheKid: nope still off by huge margin.. /etc/default/init is set to Europe/Berlin and rtc_config is set to US/Pacific .. i rebooted the machine and saw that the time was "Tue Apr 27 11:33:46 CEST 2010 "[20:29:44] <wrapster> But the current Berlin time is 8:33 PM[20:29:56] <eviljames> But the current time in Vancouver is 11:33AM[20:30:05] <eviljames> Which is the time time that matters. Didn't you know that the world has standardized on PDT?[20:30:43] <tsoome> wrapster: ntpdate again and check with date[20:30:57] *** Qiller has quit IRC[20:31:18] <wrapster> tsoome: yes now its fine.[20:31:56] <tsoome> your bios time got reset then[20:32:00] <wrapster> tsoome: but why is this happening..? running ntpdate everytime is very inconvenient.. and i cannot afford to do this[20:32:36] <tsoome> you have pdt in bios and in rtc config now?[20:32:39] <wrapster> tsoome: how can that be .. I just checked before entering the machine.. its not possible[20:33:40] <tsoome> well, something did change it. and ntpdate did fix value for you now.[20:35:46] *** ingenthr1 has joined #opensolaris[20:35:46] *** ingenthr has quit IRC[20:36:53] *** nachox__ has joined #opensolaris[20:37:54] *** TomJ has quit IRC[20:39:23] *** p3rror has joined #opensolaris[20:39:45] *** nikolam_ is now known as nikolam[20:40:03] *** mikaeld has quit IRC[20:40:04] *** nachox_ has quit IRC[20:42:06] *** mikaeld has joined #opensolaris[20:42:21] *** hsp has quit IRC[20:42:48] *** jpd1234 has joined #opensolaris[20:42:48] *** lattera has quit IRC[20:43:16] *** Fullmoon has quit IRC[20:43:45] *** mikaeld has quit IRC[20:44:02] *** mikaeld has joined #opensolaris[20:45:06] *** nachox__ is now known as nachox[20:45:48] *** TomJ has joined #opensolaris[20:46:08] *** easyE has joined #opensolaris[20:47:05] *** Odin- has quit IRC[20:49:10] *** steven43126 has joined #opensolaris[20:53:54] *** bobbyz has joined #opensolaris[20:54:04] *** LaidBack01 has quit IRC[20:54:59] *** hajma has joined #opensolaris[20:55:14] <CIA-33> Rafael Vanoni <rafael.vanoni at oracle dot com>: 6947255 early boot panic trips clock.c:2474 ASSERT(nsec_per_tick > 0)[20:58:29] *** TBCOOL has joined #opensolaris[21:00:05] <wrapster> tsoome: there is something worng. here..[21:01:08] <wrapster> i wanted to check the time in Dublin so added that entry in /etc/default/init as told by TommyTheKid and left /etc/rtc_config to US/Pacific as usual.. which was my bios time...[21:01:56] <wrapster> but after a reboot isee PDT only. which means this is wrong.. what goes into /etc/rtc_config is what is being displayed so /etc/default/init is somehow useless.[21:02:17] <twilling> /quit[21:02:27] <twilling> sorry[21:02:29] *** twilling has quit IRC[21:02:53] <Stric> there are easier ways to check the time in dublin than to reboot your machine..[21:03:32] <wrapster> Stric: TZ=Europe/Berlin date ?[21:03:52] <Stric> Yeah. Dublin is a part of Berlin.[21:04:11] <wrapster> ok whatever may be the case the value is still wrong !![21:04:46] <Stric> Try Dublin instead of Berlin[21:05:47] <wrapster> tried that as well. no luck[21:06:44] <Stric> You haven't learnt anything from earlier, it seems..[21:06:55] <Stric> Stop saying "wrong" "no luck" etc.. proof.[21:10:16] *** dijenerate has quit IRC[21:13:58] *** merzo has quit IRC[21:19:34] *** merzo has joined #opensolaris[21:21:02] *** leito has joined #opensolaris[21:24:04] <lewellyn> Stric: you like performing experiments where there's no data about the outcome, it appears.[21:24:11] *** dijenerate has joined #opensolaris[21:24:18] *** mlisok_ has joined #opensolaris[21:24:51] <cfs> Meanwhile, what mailing list should I be looking at that talks about the hold up with 2010.03.[21:25:02] *** pgr has joined #opensolaris[21:25:13] <Wes--> wrapster: # TZ=Europe/Berlin date[21:25:55] <Wes--> er, s/Berlin/Dublin/[21:26:03] <Wes--> How could that possibly not work for you?[21:28:05] <cfs> You should just set the OS to be in UTC and have users select their TZ variable in their .rc files[21:28:06] <Stric> it just does.. it gives "horrible results" instead.[21:28:12] <cfs> but that's just me[21:28:19] <mlisok_> Hello, does the MySQL Cluster require presence of MySQL. I have installed MySQL Cluster (pkgadd) and I can't run the binaries from its /opt directory? I'm trying to create a zone cluster following this paper: http://blogs.sun.com/hasham/entry/setting_up_mysql_cluster_using[21:28:34] <Triskelios> cfs: well, I can tell you the known technical reasons: 1) 134a hasn't gone through QA yet and 2) the pkg(5) backports for upgrading from the previous release aren't done[21:29:26] <Wes--> Stric: Maybe he was expecting the output in Gaelic? :)[21:29:49] <cfs> Tri, thx. Is the hold up due to lack of resources (read: ORCL gobbling up SUNW)? And does anyone have an idea of when that'll happen?[21:30:51] *** Wes- has quit IRC[21:31:10] *** biant92 has left #opensolaris[21:33:46] <Wes--> Why is it that sometimes when hit ^C a few times at the shell prompt on opensolaris that my terminal window closes?[21:34:15] <Triskelios> cfs: when what will happen? being distracted by the reorganisation is a big reason this is late, but it's mostly the same engineers as before and the normal processes afaik[21:34:41] <Triskelios> Wes--: are you sure you didn't accidentally hit ^D?[21:34:45] <lewellyn> there was also an already-optimistic release window. people saw it was optimistic months ago.[21:34:50] *** dijenerate has quit IRC[21:35:01] <Wes--> Triskelios: 100% - which is why I'm confused. (happening often enough that I'm paying close attention)[21:35:29] <lewellyn> the reorg certainly couldn't have helped hitting an optimistic target :)[21:35:37] <lewellyn> Wes--: your shell is broken? :D[21:36:15] <Wes--> lewellyn: Well, it's set to /bin/bash ATM, but I wouldn't expect /that/ to break it :)[21:36:35] <longcat> i'm hitting ^C to no ill effect[21:36:51] <CosmicDJ> wrapster: how about doing what everybody else is doing when using unix, set the bios time to UTC http://www.worldtimeserver.com/current_time_in_UTC.aspx[21:36:57] *** comay has joined #opensolaris[21:37:07] *** ChanServ sets mode: +o comay[21:37:08] <Wes--> Hm, I may have just figure out how to reproduce - I'm running a script, after it runs it'll do it. *minimizing test case*[21:37:24] *** jacotton has quit IRC[21:37:25] <wrapster> ok wil do that.[21:37:37] <Triskelios> Wes--: is the shell dumping core or just exiting?[21:38:03] <Wes--> Triskelios: excellent question, I'll go update coreadm and find out[21:38:16] <lewellyn> comay has impeccable timing ;)[21:39:54] *** jax__ has joined #opensolaris[21:40:43] *** AxeZ has joined #opensolaris[21:40:55] *** jax__ is now known as LaidBack_01[21:41:28] *** jurikm has joined #opensolaris[21:42:34] <LaidBack_01> hey, I'm trying to install mod_perl on a snv_134 zone. I'm running into an error : cc1: error: invalid option `t' and I see the Makefiles including a '-mt' in the line. So, I don't see where this is set, but it's all over google - well not the solution. Anyway, anyone know the fix right off?[21:43:04] <CosmicDJ> LaidBack_01: is that gcc or sun studio?[21:43:09] <LaidBack_01> gcc[21:43:10] <lewellyn> you're installing from the repo?[21:43:19] <LaidBack_01> uh, no this is from cpan[21:43:31] <Wes--> Argh, coreadm service won't come online![21:43:39] <cfs> is -mt supported by gcc[21:43:43] <cfs> I didn't think it did.[21:43:46] <LaidBack_01> no, -mt is not.[21:43:58] <cfs> I think perl assumes cc[21:43:59] <LaidBack_01> so install sunstudio then, and use that, eh?[21:44:03] <cfs> yeah[21:44:05] <cfs> might as well.[21:44:06] <melbogia> Anybody know a good place where I can see steps to create another instance of a service, say auto-snapshot service?[21:44:08] <Wes--> perl assumes whatever cc it was built with[21:44:08] <LaidBack_01> okay[21:44:21] <Wes--> which is kind of annoying when you are installing a sun-built perl and only have gcc on your box[21:44:25] <Wes--> (basically, it means you're hosed)[21:44:28] <lewellyn> LaidBack_01: i think there's a package[21:44:28] <cfs> I mean you could find the default makemaker config and change all of that.[21:44:55] <Wes--> cfs: true, you probably could do that, I just sigh and install either another compiler or another perl when this happens. ;)[21:45:23] <LaidBack_01> yeah, I've compiled perl 5.10.1 with Sun Studio... so that's why.[21:45:33] *** timeless_mbp has joined #opensolaris[21:45:33] *** timeless_mbp has joined #opensolaris[21:45:40] <LaidBack_01> but that was on a machine intened to handed out the packages for install.[21:45:47] <lewellyn> LaidBack_01: mod_perl ships with opensolaris's apache 2.2[21:45:59] <Wes--> LaidBack_01: Get the best of both worlds! Build it with GCC FSS :)[21:46:08] <LaidBack_01> yeah, 5.8.4[21:46:15] <LaidBack_01> I need a min of 5.8.8 for bricolage.[21:46:22] <cfs> Wes-- indeed, I tend to just use sunstudio with *solaris[21:46:22] <LaidBack_01> and I've just moved to 5.10.1 instead[21:46:35] <cfs> except bastard perl module writers to try to use C++[21:46:52] <Wes--> cfs: I'm on the polar extreme - I do gcc everywhere which means I don't need to worry quite as much about portability[21:46:53] <cfs> that just fucks up MakeMaker[21:47:08] * cfs <3's portability[21:47:23] <lewellyn> yeah. you REALLY want to use sun studio across the board with sun's perl. otherwise, compile your own and stick it in a private location.[21:47:53] <Wes--> yeah, putting it in /usr/bin is a really bad idea[21:47:59] <Wes--> don't ask me how I know[21:48:07] <Wes--> But let's just say, SUNWstade cares[21:48:24] <LaidBack_01> I've got my perl in /opt/perl/bin with links to /opt/bin[21:48:31] <LaidBack_01> it's pretty safe there I think.[21:49:04] *** dijenerate has joined #opensolaris[21:49:07] <LaidBack_01> actually it's in /opt/perl-5.10.1/<whatever> with links out. just actually looked... sry.[21:49:28] *** ivo_ has quit IRC[21:50:42] <LaidBack_01> do you guys use SunStudio or SunStudioExpress? I'm not sure which is the "right" one.[21:51:39] <Triskelios> the current express is a pre-release of ss12, so just use 12[21:51:43] <lewellyn> use sun studio[21:51:53] <lewellyn> pfexec pkg install ss-dev[21:52:26] <Triskelios> I think ss-dev still pulls in express?[21:52:57] <lewellyn> does it?[21:52:58] * lewellyn checks[21:53:01] <Wes--> triskelios: I've now proven that it's the /terminal/ rather than the shell that's exiting, but am not sure if it's dumping core or something else (I can't get coreadm to start)[21:53:46] <lewellyn> hm. ss-dev isn't in the current list :P[21:54:08] <Wes--> Does OpenSolaris have a gdb package? I can't seem to find it in 'pkg list'[21:54:26] <lewellyn> use dbx[21:54:28] <lewellyn> smrt: explain dbx[21:54:30] <smrt> Um. I seem to not know anything about dbx...[21:54:32] <lewellyn> lame[21:54:42] <lewellyn> but yeah. dbx > gdb[21:54:46] <Triskelios> Wes--: coreadm -g '/var/cores/core.%f.%p' && coreadm -e global[21:55:02] <Triskelios> && mkdir /var/cores[21:55:14] *** baza11215 has quit IRC[21:55:40] <Triskelios> Wes--: pkg install gdb should work[21:56:25] <LaidBack_01> okay, so installing sunstudio12u1... guess I'll see in a bit.[21:56:30] <LaidBack_01> 306mb later ;)[21:56:45] <lewellyn> Triskelios: seems you're right about ss-dev. that's lame :([21:56:54] <Wes--> Triskelios: "coreadm service not online" - log says "core_set_globa_path(): Invalid argument"[21:56:58] <Wes--> And no gdb package[21:57:03] <Wes--> Have I got a lemon OpenSolaris install?[21:57:06] <lewellyn> i'll file a bug later :P[21:57:24] <bdha> Maybe it wants "core_set_global_path.[21:57:24] <lewellyn> Wes--: are you on release or dev?[21:57:31] <bdha> Instead of "globa_path".[21:57:37] <lewellyn> oh. it didn't get renamed[21:57:41] <bdha> Also... why do you want gdb for that?[21:57:41] <lewellyn> SUNWgdb[21:57:41] <Triskelios> svcadm enable coreadm, but I'm pretty sure it's enabled by default[21:57:42] <Wes--> No clue, I didn't build this box. snv_111b in uname mean anything to you?[21:57:50] <lewellyn> http://pkg.opensolaris.org/dev/p5i/0/SUNWgdb.p5i[21:57:52] <lewellyn> that's release[21:57:56] *** PrestonConnors has joined #opensolaris[21:57:59] <lewellyn> change the dev to release in that url[21:58:15] <lewellyn> (assuming you have a gui)[21:58:18] * lewellyn heads off[21:58:40] <PrestonConnors> Hello, I am using snv_134 and I am trying to use sbdadm to create a logical unit but I keep on getting error "sbdadm: unknown error" and I'm not sure how to troubleshoot that error message.[21:58:51] <PrestonConnors> Any ideas would greatly be appreciated![21:59:21] <Wes--> oh man, now firefox won't start either[21:59:30] <Wes--> God I can't wait for my mac to get back from the shop[21:59:37] <Wes--> I never thought I'd say that, lol[21:59:39] <bdha> Heh... what'd you do to that box? :)[21:59:50] <Wes--> Dude, nothing, it's brand new[22:00:15] <Wes--> well, I mounted my homedir over NFS, it's up on NIS and I changed the display dpi[22:00:48] <Wes--> cloned a couple of repos, and tried to build a copy of spidermonkey - every time I get one step closer, it's two steps back[22:01:31] *** p3rror has quit IRC[22:01:36] <Wes--> I think I'll just ssh into one of the solaris 10 machines and live there while my desktop is out getting serviced - was hoping to do a 5.11 port while I had the opportunity[22:02:03] *** techqber1 has quit IRC[22:02:20] *** techqbert has joined #opensolaris[22:02:29] <PrestonConnors> Nevermind, I had specified a dataset that was not able to have a logical unit created from it.[22:03:09] <Wes--> Next time I get an Open Solaris machine built, is there anything I should ask for? Like the "dev" version or something?[22:03:45] *** syamajala has joined #opensolaris[22:06:50] *** p3rror has joined #opensolaris[22:07:14] <Triskelios> Wes--: probably a development release if your company doesn't have a support contract with Oracle[22:08:23] *** andersenep_ has joined #opensolaris[22:08:38] *** fleebailey33 has quit IRC[22:08:38] *** amgarching has quit IRC[22:08:38] <bdha> Wes--: Depends on your requirements.[22:08:38] <bdha> /dev certainly has newer bits (by almost a year).[22:08:38] <bdha> genunix.org has dev images.[22:08:44] <Wes--> Triskelios: Roger -- won't be buying support for this, was going to a skunkworks Open Solaris port because I thought it would be easy and some people in FOSS land might find it handy. (Irony - primary dev platform is Solaris 10)[22:09:32] <bdha> Wes--: Does sound like a hosed install, though. Haven't had those problems with 2009.06.[22:09:37] <Wes--> But, I am learning now that 5.11 is not just "the next version", like the 2.5 ->2.6 -> 7 -> 8 -> 9 and (to a lesser extent) 10 were[22:09:53] <bdha> Yeah, Major Changes.[22:09:57] * bdha is an S10 admin.[22:10:19] <Wes--> bdha: You're right, I'm starting to think that myself. You'd think you could ask somebody for a freaking Open Solaris box and get something that work right the first time.[22:10:22] * Wes-- grumbles[22:10:27] *** andersenep has quit IRC[22:10:28] *** andersenep_ is now known as andersenep[22:10:31] <Triskelios> there's no space in "OpenSolaris", btw[22:10:37] *** fleebailey33 has joined #opensolaris[22:10:39] <bdha> Wes--: Manual install is like four mouse clicks. :)[22:10:43] <bdha> Wes--: Maybe they tried to use AI, thoguh.[22:11:16] <Wes--> Triskelios: Yeah, I know, it just keeps happening, fingers are on brain-disconnected auto pilot I think.[22:11:42] <Wes--> bdha: Good question - he asked me about jumpstart and I said "NO! Just click the mouse until it comes up, then bring it to my office", lol[22:12:45] *** dijenerate has quit IRC[22:13:09] <bdha> Wes--: Well, grab an image from genunix and reinstall it. :)[22:13:22] <bdha> Get the automated AI image. No clicking.[22:13:40] <Wes--> bdha: But... then I'll be stuck on windows until it's done. (How long does it take?)[22:14:01] <bdha> Depends on the system?[22:14:17] <bdha> Don't you have Putty or whatever for Windows? :)[22:14:18] <Wes--> "stuck on windows", btw means, "Not able to work, can only answer email". Which maybe isn't so bad, except that I'm a motivated guy, y'know[22:14:21] <cfs> if you are on windows, why not get virtual box/vmware server 2/etc and install a vm[22:14:32] <Wes--> bdha: I suppose I could run emacs in a putty window, yeah. :)[22:14:44] <bdha> That's what I do when stuck on my Win7-only netbook.[22:14:53] <bdha> Which is hateful, but does not cause my ears to bleed.[22:15:15] *** darrenb` has quit IRC[22:15:43] <Wes--> When I'm stuck on my vista netbook (I lost my Win7 upgrade CD, stupid stupid stupid), I mostly watch YouTube. :D[22:15:43] <melbogia> can anybody tell me how i can copy an instance of SMF to another, thereby creating a new instance?[22:16:09] <trygvis> you can't really copy an instance, just create a new one and copy any properties that's instance specific[22:16:26] <melbogia> oh i see[22:16:34] <Wes--> cfs: Don't really want to run a vm, not when the other machine has a bigger monitor, type 5c keyboard, and way more ram/cpu[22:16:40] * Wes-- goes to look for that ai image[22:17:11] <TommyTheKid> 5c!?[22:17:21] *** niq has quit IRC[22:17:46] <Wes--> TommyTheKid: Dude, they are the best programming keyboards I have ever used[22:18:07] <TommyTheKid> melbogia: take a look at svccfg export and import ... and look around for blog entries :)[22:18:13] <Wes--> And they last ~8 years each[22:18:34] <TommyTheKid> I like the type 8 unix-layout that comes with the sunrays, and the new mouse too[22:19:00] <TommyTheKid> 5c was the old style connector wasn't it?[22:19:07] <Wes--> I never used them, honestly - what do the keys feel like? The type 6 sucks[22:19:16] <TommyTheKid> I guess I have only used a USB keyboard fo the longest time[22:19:23] *** darrenb` has joined #opensolaris[22:19:26] <Wes--> TommyTheKid: Yeah, the old 1200 bps inverted TTL keyboards. I also use the 5c mouse, but am not really tied to it[22:19:29] <TommyTheKid> the keys feel like any other keyboard, they dont CLICK[22:19:55] <Wes--> My windows box has a vintage 1991 IBM Model M. I wish it had a proper layout, because the keys feel awesome.[22:20:28] <TommyTheKid> Wes--: i don't suppose you like the touch keyboard like on the ipodtouch/iphone/ipad in that case :p[22:20:41] <TommyTheKid> the "star trek" style :p[22:20:54] <Wes--> Well, they are fine for texting. But not for real work.[22:21:06] * bdha digs the MacBook-style chiclet keyboard.[22:21:08] <Wes--> I have been known to go on strike when my ergonomic needs were not met. :)[22:21:08] *** PrestonConnors has left #opensolaris[22:21:33] <Wes--> I haven't tried that one, but the keyboard that came with my iMac was about 6mm thick and the biggest POS I have ever typed on[22:21:35] <TommyTheKid> i have gotten pretty used to the mac book pro style keyboard too[22:21:40] <Wes--> Well, excluding the keyboard for my ZX81[22:21:49] <bdha> Wes--: The new Mac keyboards are the same style as the MacBooks.[22:21:55] <Wes--> The keyboard that came with the Apple //c wasn't bad, though[22:22:02] <bdha> So if it's a Mac from the last two? years, it'll be chiclet.[22:22:07] * bdha hates clicky keyboards, though.[22:22:20] <TommyTheKid> yea, i like my keys to be mostly silent[22:22:24] <Wes--> bdha: AWEFUL. Would never use one, not in a million years. The keys don't move enough, and irritate my median nevers really badly[22:22:30] *** lattera has joined #opensolaris[22:22:39] <Wes--> The Sun Type 5c is close to silent, AND offers good feedback[22:22:49] <Wes--> I'd like to meet the engineer that designed it, and buy him lots of beer[22:22:58] <bdha> Eh.[22:23:03] <TommyTheKid> he was probably laid off :p[22:23:04] *** ikonia has quit IRC[22:23:17] <TommyTheKid> he or she I should say[22:23:26] *** Beket has joined #opensolaris[22:23:27] <Wes--> Probably retired by now, honestly. Those came out in, what, 1995?[22:23:37] <Wes--> I know they came standard on our Ultra 5s[22:23:39] <TommyTheKid> in 1995 I was in high school[22:23:42] * alanc has no idea who designed Type5 - by the time I got here, Sun keyboards were all OEM'ed[22:23:50] <Beket> Hi people. I can see bsdbf/arc4random.c in onnv-gate, but there's no manual page for it. Hm ?[22:24:13] <Wes--> alanc: The Type 4 feels even awesomer, but the layout is inferior[22:24:15] *** ikonia has joined #opensolaris[22:24:17] <alanc> lots of things in ON have no man pages, because they're not public interfaces for others to use[22:24:22] *** mroconnor has quit IRC[22:24:29] <alanc> I always hated the return key layout on the Type 4[22:24:36] *** pjfloyd has joined #opensolaris[22:24:50] *** niq has joined #opensolaris[22:24:52] <TommyTheKid> i think the type 7/8 are the same layout as my ultra 5/10 keyboard[22:24:57] <Beket> mm, why not ?[22:25:01] <TommyTheKid> iono, would have to dig the sucker out of the closet[22:25:22] <Wes--> Yeah, that was one of my big beefs with the type 4. The optical mice of the day were pretty good, though[22:25:34] *** pjfloyd has quit IRC[22:25:43] <alanc> bsdbf/arc4random.c source is part of crypt_modules - that's not a library you can link against or call from any other code[22:25:51] *** dijenerate has joined #opensolaris[22:26:17] <alanc> the crypt_modules are plugins for crypt(3c)[22:26:19] <Wes--> TommyTheKid: control beside A? esc beside 1? Tilde above backspace?[22:26:32] <TommyTheKid> yes, the unix layout[22:26:36] <Wes--> Awesome[22:26:45] <Wes--> Some joker was releasing Type 5c with a PC layout for a while[22:26:53] <TommyTheKid> they came out both ways[22:27:02] <alanc> the man page for crypt_modules/bsdbf is crypt_bsdbf(5)[22:27:23] <TommyTheKid> at least for the 6/7[22:27:51] <TommyTheKid> i lied, the sunray 2* keybaords were type 7 I guess[22:28:12] <TommyTheKid> 320-1367-02[22:28:45] *** ivo_ has joined #opensolaris[22:29:04] <stallion_work> TommyTheKid: Sun doesn't sell "special" keyboards for specific units[22:29:37] <stallion_work> You buy a keyboard - your only choice is whether you prefer a PC layout over the traditional one[22:29:40] <Beket> sad; thanks alanc I'll #ifdef against opensolaris[22:29:55] <TommyTheKid> right, the USB keyboards for like every Sun Fire/Blade/Ray/X/V/Z/etc system for the last 5+ years? :p[22:29:59] <stallion_work> its its been the same for quite a long time - this Type 7 is no different than my Type 5 15 years ago[22:29:59] <TommyTheKid> maybe 10?[22:30:00] *** hajma has quit IRC[22:30:13] <Wes--> stallion: Does the type 7 FEEL the same?[22:30:17] <stallion_work> just USB v.s. mini-DIN[22:30:25] <stallion_work> Wes--: no.[22:30:42] <stallion_work> Its just as well made, just not mechanical.[22:30:50] <TommyTheKid> the LAYOUT should be about the same if not exactly, but there are two layouts (at least for US)[22:30:53] <alanc> Beket: given you've not said what you want, just pointed at a bit of code that's not in a library, I guess that's what you'll have to do[22:31:03] *** jenky has quit IRC[22:31:17] *** hajma has joined #opensolaris[22:31:23] <stallion_work> TommyTheKid: see above.[22:31:25] <Wes--> Bummer - I love the type 5c -- but the 5c->USB adapter I use costs almost a hundred bucks[22:31:26] <TommyTheKid> right[22:31:30] <TommyTheKid> i just noticed I was behind[22:31:46] <stallion_work> Wes--: its not worth it lugging that around anymore. The Type 7's are very nice. I use one at home, and one at the office.[22:31:58] *** jurikm has left #opensolaris[22:32:09] <Wes--> stallion_work: I should try and find one locally, I would love to have a few more proper keyboards[22:32:30] <TommyTheKid> http://www.anysystem.com/type7-keyboard-mouse-kit.html#[22:32:31] <Wes--> I even have half the special keys (along the left) working with my mac :)[22:32:31] <stallion_work> Wes--: Last I checked they were about 50$ from Sun (a reasonable price IMHO)[22:32:43] <TommyTheKid> honestly they are cheaper at Sun.. heh Yea[22:33:54] <TommyTheKid> watch out, I htink that picture on anysystem is PC layout, they may not have cared since its soo small, but the color of the keys doesn't match mine[22:34:17] <stallion_work> TommyTheKid: you select the layout at time of purchase.[22:34:22] <TommyTheKid> ooh[22:36:05] <TommyTheKid> https://shop.sun.com/store/search -- fill out sun ray clients and then sun ray 3 plus, they are $50 RETAIL from Sun[22:36:15] <TommyTheKid> X3701A-UNIX[22:36:38] <Triskelios> er, syscall::execve:entry no longer works?[22:36:43] *** niq has quit IRC[22:37:13] <stallion_work> Triskelios: dunno, I don't usually probe userland.[22:37:47] <TommyTheKid> you need more magic for your probity-probe[22:38:42] <stallion_work> Triskelios: personally I prefer using -F and dropping entry/exit when looking at stacks like that[22:38:46] <cfs> I'm using a Type-7 US/PC on my mac mini right now.[22:39:31] <TommyTheKid> blasphemy! burn the PC keyboards[22:40:04] <TommyTheKid> sorry[22:40:34] <TommyTheKid> i like that on my mac, I can make the stupid key to the left of A be control, ~ is still in the wrong spot, but I use it a lot less than I do control[22:40:50] <cfs> I don't use emacs often and the symmetric control keys are pref now.[22:41:02] *** niq has joined #opensolaris[22:41:06] <cfs> well rather, the compose key is understood by the mac as control.[22:41:38] <Triskelios> stallion_work: too many processes here; I only want to know which one is trying to exec a certain file (which fails) and why[22:41:44] <TommyTheKid> never tried the compose key, when my macbookpro is on the KVM I can use that keyboard, but usually my SunRay is focused[22:41:58] <Triskelios> stallion_work: thanks though[22:42:24] <stallion_work> Triskelios: hrmm... you should be able to narrow by process... do you have a script written or is this a one-liner?[22:43:33] <TommyTheKid> Triskelios: could you just use like syscall:::entry and findo ut which one you want?[22:45:08] <stallion_work> TommyTheKid: ... that would be painful.[22:45:21] <Triskelios> stallion_work: basically a one-liner. there's a bug that's only reproducible by starting gnome-session, so there's a large number of candidates[22:45:29] <stallion_work> gotcha[22:45:34] *** niq has quit IRC[22:45:45] <stallion_work> Triskelios: have you considered just using good ol' fashioned truss?[22:46:28] *** longcat has left #opensolaris[22:47:33] *** olivier_54 has quit IRC[22:47:41] <Triskelios> stallion_work: need to get the stack, anyway, so this will be quicker[22:47:57] <Wes--> TommyTheKid: Compose key *rocks*, but not supported anywhere but Sun hardware in my experience[22:48:11] <Triskelios> looks like syscall::exece:entry is what I want[22:48:30] <stallion_work> Wes--: keycodes are keycodes.[22:48:40] <TommyTheKid> exec* :)[22:48:40] <Triskelios> yay, it works[22:48:52] <Wes--> stallion_work: Yeah, but you ever see another driver support compose?[22:48:52] <Triskelios> TommyTheKid: I tried that first, no matches[22:49:01] <TommyTheKid> ouch, it matches two entries here[22:49:07] <Wes--> stallion_work: composition is done in software, not on keyboard[22:49:09] <TommyTheKid> Wes--: compose works on Sun Ray[22:49:24] <Wes--> Right, which is sun hardware[22:49:29] <TommyTheKid> the light doesnt come on but comppose n ~ works[22:49:32] <stallion_work> Wes--: exactly. It doesn't matter whether or not the host is a Sun machine or not.[22:50:18] <TommyTheKid> oh sorry, I misread that wes ;)[22:50:29] *** niq has joined #opensolaris[22:50:34] <Wes--> stallion_work: Except - have you ever seen it work? Not "could it work if you re-wrote the keyboard driver" -- And when I said "sun hardware", I'm basically including x86 running solaris, I know that wasn't very clear. :)[22:50:36] <stallion_work> I would imagine there is a keyboard layout somewhere that provides compose support thats not SunOS[22:51:25] <Triskelios> TommyTheKid: strangely, I tried exec* and it works now... bizarre[22:51:31] *** nachox has quit IRC[22:51:38] <wrapster> thanks for all your help guys.. its time issue is finally sorted.[22:51:42] <Wes--> stallion_work: I've actually looked - surprisingly never found anything else, which is unfortunate, because, I mean, it's just awesome. Certainly not available on Leopard or Windows. Maybe some non-standard linux?[22:51:58] *** jaywalk has quit IRC[22:52:02] <wrapster> and one last question.. is thre a way i can convert date from one format to another.? any toolkit that provides this support?[22:52:11] *** dalibor has joined #opensolaris[22:52:12] <stallion_work> Wes--: http://tinyurl.com/3yxf5eq[22:52:17] <stallion_work> Wes--: you haven't looked hard.[22:52:18] *** dalibor has quit IRC[22:52:19] <Wes--> wrapster: date[22:52:21] *** xapiens has quit IRC[22:52:25] *** radsy has joined #opensolaris[22:52:39] *** robo has quit IRC[22:52:41] <wrapster> eg: i specifiy the date in one UTC i want solaris to display the corresponding IST time!![22:52:42] <Wes--> stallion_work: You're right, I just plugged the keyboard in and tried it[22:53:05] <Wes--> stallion_work: SHould have looked to see what what possible on every possible distro instead of giving up after tree[22:53:09] <Wes--> s/tree/three[22:53:46] <Wes--> (I was hoping for something that "just worked" - I'm getting old and crotchety, and pretty data entry applicances to rube goldberg machines. ;) )[22:53:58] <Wes--> OpenSolaris "just works" in that regard, though. :)[22:54:36] <CIA-33> Jerry Gilliam <Jerry.Gilliam at Sun dot COM>: 6945467 AI image fails to boot in x86 machines due to devfsadm -r core dump in snv_138[22:54:38] <CIA-33> jmcp <James.McPherson at Sun dot COM>: Added tag onnv_139 for changeset f199783d527ae58eb6d926889a8f4bdae1d67f72[22:55:09] *** dalibor has joined #opensolaris[22:55:28] <stallion_work> jmcp: bah.[22:56:37] <TommyTheKid> thats a nice tag :p[22:59:54] *** sk0rd has quit IRC[23:02:28] <Chipdancer> TommyTheKid: unfortunately, it just reminds those of us outside the wall that we're a couple of months behind the fun[23:03:18] *** pgr has quit IRC[23:03:35] *** relling has joined #opensolaris[23:04:27] <stallion_work> Chipdancer: what? not at all.[23:04:40] <stallion_work> CIA-33 is wired to the public hg repo.[23:04:53] <Chipdancer> stallion_work: well, I'm ont willing to build ON... I'm waiting for IPS packages ;)[23:05:25] <stallion_work> Chipdancer: wait for the release and quit kvetching about walls; there is a reason why things are running behind...[23:05:42] <Chipdancer> stallion_work: oh, that's fine.. just waiting.. waiting :)[23:05:55] <pogs> I don't suppose there's been talk of a SXCE/S10 --> opensolaris upgrade path, has there?[23:06:11] <pogs> or at least an IPS for s10?[23:06:27] <stallion_work> pogs: I'm sure its on a list somewhere; but most likely it will be for Solaris 10 to Solaris Next (not OpenSolaris)[23:06:42] <stallion_work> SXCE has been toast for 9 builds now, so no upgrade path there beyond re-installing[23:06:59] <pogs> stallion_work: OH! Is there a Solaris Next on the horizon? :D[23:07:11] <Chipdancer> pogs: always[23:07:16] <stallion_work> pogs: probably a safe bet.[23:07:25] <alanc> IPS packages were built for S10 for adding on additional software, but there's no plan to repackage Solaris 10 in IPS format - such a major change would make it something new, like Solaris 11, not Solaris 10 any more[23:09:12] <pogs> I see, that would be nice.[23:09:54] <pogs> Esp since Sun was acquired by something called "oracle" :P[23:10:00] <Triskelios> weird, I got "dtrace: no probes specified" when I try to use a destructive action (stop())[23:11:19] <alanc> well yes, it's now "Oracle Solaris Next" (where "next" is a code name for "probably 11, but not yet officially named", not the actual name)[23:11:29] *** robo has joined #opensolaris[23:11:47] <DesiJat> Oracle Solaris 11g.[23:13:49] *** Animal-X has quit IRC[23:19:04] *** smrt has quit IRC[23:20:01] <TommyTheKid> I have heard that 137 was a little rough around the edges from someone who built it... remember that the last few builds going to 134 were locked down to new changes (only stopper bug fixes), so there were a bunch of changes queued up for 135++[23:20:09] *** brougham has quit IRC[23:20:27] <TommyTheKid> (this is my theory of course)[23:21:17] <TommyTheKid> ... I may stay at 134 (a???) for a while before I find something I can't live without in /dev again :)[23:24:13] *** GdeLeo has joined #opensolaris[23:24:23] <alanc> and as CIA just announced, it only took ON an extra 15 hours to get 139 closed because they were waiting for a couple panic-on-boot fixes[23:24:25] *** syamajala has quit IRC[23:24:59] *** brougham has joined #opensolaris[23:25:50] <CosmicDJ> mmhh 139..[23:26:34] <stallion_work> jmcp: does c-team still require tip +/- 2 for integration?[23:26:53] <stallion_work> err - 2 I guess ;)[23:27:28] <CosmicDJ> I'd love to try PSARC/2010/029 Improved ACL interoperability http://bit.ly/afNfz6[23:27:47] *** xapiens has joined #opensolaris[23:30:16] *** macros73 has quit IRC[23:31:31] *** macros73 has joined #opensolaris[23:32:01] <jmcp> stallion_work: build within 2 (inclusive), yes[23:32:30] *** brougham has quit IRC[23:32:37] *** GdeLeo has left #opensolaris[23:32:44] *** robo has quit IRC[23:33:50] *** smrt has joined #opensolaris[23:34:32] *** wrapster has quit IRC[23:35:33] *** radsy has quit IRC[23:35:51] <stallion_work> jmcp: gotcha. Looks like efe is finished - I just need a NICDRV battery and to get gdamore to spearhead the PSARC case to replace spwr[23:35:59] <jmcp> excellent![23:36:19] <stallion_work> I'll have to update my snv_134 sandbox though - not a bad thing since I'll be able to pull in the official nemo interfaces[23:36:33] <stallion_work> I think that integrated somewhere between 136 and 138[23:36:36] <lewellyn> alanc: let's speculate. how likely is it that ir solaris next ships at the same time as the next version of the oracle database, it may end up being called Oracle Solaris 12g instead of 11? ;)[23:37:04] <cybernd> you mean 12c[23:37:37] * lewellyn has never understood oracle's letters too much[23:37:44] <lewellyn> i just install what DBAs tell me to[23:37:49] *** brougham has joined #opensolaris[23:37:54] <alanc> lewellyn: damn, you removed the "speculation free zone" from the topic just to ask that, didn't you?[23:38:07] <cybernd> i = internet, g = grid - something like that .. c could be cloud or cluster or so[23:38:46] <lewellyn> was i the one who removed that? :P[23:39:05] * stallion_work gets a /kick ready...[23:39:16] * alanc doesn't actually remember - probably just fell off the end from our topic length limit[23:39:20] *** ChanServ sets mode: +q *!*@greenviolet/lewellyn[23:39:28] *** AxeZ has quit IRC[23:39:29] <stallion_work> haha[23:39:31] *** lewellyn sets mode: -qn #opensolaris!*@*[23:39:56] <cfs> I think it'll be named: OracleOS YouPayOutTheWazoo 11$£¢€[23:39:57] <lewellyn> alanc: i actually think i replaced that by "We're not *-discuss"[23:40:19] <lewellyn> cybernd: also, you haven't heard larry's thoughts on the cloud, i take it?[23:40:56] <cybernd> lewellyn : nope but its one of the currently hyped themes[23:41:17] <lewellyn> regardless. it's one of the places i'm in total agreement with mr larry[23:41:42] <lewellyn> now. time for me to figure out why i can't get this vm working[23:42:11] <alanc> his thoughts mainly seem to be that cloud is just the new buzzword for client-server + clusters, things that have been done for decades, just now with outsourcing[23:42:43] <pogs> "National Weather Service reports that if you see computers in the clouds, to seek shelter immediately"[23:43:00] *** Beket has quit IRC[23:43:07] <lewellyn> alanc: and that it's a silly name and that clouds aren't something one tends to look forward to[23:43:15] <cybernd> lewellyn : okay i think i found his talk "cloud is already the past - we done this since 10 years"[23:43:55] *** dalibor has quit IRC[23:44:07] <lewellyn> cybernd: it's how things have been done for 50 yeras[23:44:09] <lewellyn> years even[23:44:37] <cybernd> its just the same thing which happened to other technologies as well - for example ajax was out for years .. it was just a new buzz compliant term for the same thing[23:45:04] <lewellyn> ajax wasn't around for longer than the people promoting it have been alive ;)[23:45:34] <cybernd> the concept behind ajax was :P[23:45:48] <lewellyn> you're comparing ajax to cics?[23:45:53] *** twisti has quit IRC[23:46:02] * lewellyn watches people cringe[23:46:20] *** radsy has joined #opensolaris[23:46:42] *** sndcrb has quit IRC[23:47:11] <TommyTheKid> how is zfs-crypto doing?[23:47:36] <stallion_work> TommyTheKid: ask zfs-discuss[23:47:39] *** fleers has quit IRC[23:47:46] <alanc> we don't know, the status report was stored on a zfs-crypto system and no one told us the key to unscramble it[23:47:55] <TommyTheKid> heh[23:48:11] <stallion_work> alanc: keyrot[23:48:18] <Triskelios> or zfs-crypto-discuss[23:49:35] *** thepickle has quit IRC[23:49:57] <TommyTheKid> so it sounds like its not just part of WOS ?[23:51:06] <alanc> last I heard it was still a work in progress and not ready for integration[23:51:13] <TommyTheKid> ok[23:51:38] <alanc> but as stallion_work said, the zfs lists would know more than random people on IRC[23:52:09] <TommyTheKid> the random response I have received here was more than enough[23:52:26] <Triskelios> TommyTheKid: it's a separate repo, externally available[23:52:27] *** Yu\2 has quit IRC[23:52:42] <TommyTheKid> someone was complaining that there was no way to have an ecrypted filesystem on Solaris (for example on an external USB disk)[23:52:58] <TommyTheKid> I was going to say that the coming release of OpenSolaris had ... but yea ;)[23:53:15] <lewellyn> the coming release doesn't even have nwam phase 1[23:53:34] *** fleers has joined #opensolaris[23:54:26] <Triskelios> TommyTheKid: lofi encryption works[23:54:36] <jbit> hasn'T solaris had encyrpted block support for a while[23:54:40] <jbit> Triskelios: ah[23:55:02] <tsoome> jbit: with lofi, yes, and only opensolaris[23:56:45] *** macros73 has quit IRC