September 28, 2012 
< | 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 | >
September 28, 2012 [00:03:32] *** tsoome_ has joined #openindiana[00:03:32] *** tsoome has quit IRC[00:03:33] *** tsoome_ is now known as tsoome[00:05:05] *** Seony has quit IRC[00:05:11] *** betchou has joined #openindiana[00:05:39] <betchou> Hy all,[00:06:19] <betchou> I would like to know how to debug a wifi driver and where to find the sources? It is the rum driver.[00:12:27] <Triskelios> betchou: http://src.illumos.org/source/xref/illumos-gate/usr/src/uts/common/io/rum/ - there are device driver development guides for Solaris which are applicable, as well as information on using mdb and DTrace for kernel debugging[00:12:35] *** alanc has joined #openindiana[00:12:36] *** ChanServ sets mode: +o alanc[00:12:42] *** BonzTM has joined #openindiana[00:13:13] <Triskelios> betchou: in general, you'd want #illumos for driver development questions[00:14:55] <betchou> Triskelios: Thanks a lot!!! :)[00:28:30] <hunter> How much suffering would I be in for if I tried pulling sendmail out and switching it for postfix?[00:30:41] *** Seony has joined #openindiana[00:33:10] <Triskelios> on your own system? probably not much, I used to use Postfix[00:44:47] *** betchou has left #openindiana[00:47:15] <hunter> I have 6 boxes as fileservers and just added a 7th as a database server.[00:48:06] <hunter> So not many, but I'd love to just stick with the postfix config we know works for us rather than duping that into sendmail.[00:48:26] <hunter> But none of these boxen will be "mailservers" so it might not be worth the effort.[00:57:26] *** patdk-lap has quit IRC[01:00:36] *** halldorh has joined #openindiana[01:12:27] *** tsukasa has joined #openindiana[01:12:28] *** tsukasa has joined #openindiana[01:13:04] *** ozquera has quit IRC[01:15:32] *** tsukasa` has quit IRC[01:15:49] *** classix has quit IRC[01:16:48] *** classix has joined #openindiana[01:47:10] <oninoshiko> hunter: you shouldn't have too much problem. just uninstall and install a package for postfix... you might have to make a package for postfix (I don't know if one exists or not), but making packages isn't too hard[01:48:40] <oninoshiko> Although I do admit there is a possibility that I have a case of stockholm syndrome....[01:49:35] <hunter> oninoshiko: There's one in SFE already.[01:49:48] <oninoshiko> oh, then it should be easy-peasy[01:49:53] <hunter> And solaris packaging is …. different[01:50:25] <oninoshiko> different, but not hard. I've been working with it at work[01:51:07] <hunter> Any good docs you've found for OI-151 and friends?[01:51:25] <hunter> (Would love to get puppet and facter packages)[01:52:57] *** Seony has quit IRC[01:52:59] <oninoshiko> while it is mostly geared to Oracle's most recent product, the Application Packager's guide is a good resource... let me find the link[01:53:14] <oninoshiko> almost everything still applies to OI.[01:54:31] <oninoshiko> I will admit to supplementing it with my personal archive of oi-dev and illumos-dev, when working on repackaging jenkins[01:56:36] <oninoshiko> http://hub.opensolaris.org/bin/download/Project+pkg/files/ipsdevguide.pdf[02:04:02] <bmercer> I heard some great things about oi on sparc![02:13:40] *** scarcry has quit IRC[02:17:41] *** scarcry has joined #openindiana[02:20:35] <TomJ> What RBAC auths are needed to install a zone? I've given a user the "Zone Security" profile which grants it solaris.zone.* authorisation - that user can zonecfg create a zone, but can't zoneadm install it[02:21:29] <TomJ> pfexec zoneadm -z rbac-testing install gives: "zoneadm: zone 'rbac-testing': only a privileged user may install a zone."[02:23:48] *** CVLTCMK0 has joined #openindiana[02:27:28] *** imaxs has joined #openindiana[02:45:57] *** Seony has joined #openindiana[02:56:49] *** kovert has joined #openindiana[03:10:31] *** SupremeOverlord has joined #openindiana[03:24:56] *** ball has joined #openindiana[03:27:34] *** LabMan has quit IRC[03:29:09] *** jellydonut has quit IRC[03:35:45] *** riccardo has quit IRC[03:36:02] *** riccardo has joined #openindiana[03:36:24] *** bruges has quit IRC[03:36:25] *** ira has quit IRC[03:37:45] *** bruges has joined #openindiana[03:41:21] *** Seony has quit IRC[03:56:29] *** szaydel has quit IRC[04:04:28] *** szaydel has joined #openindiana[04:10:42] *** szaydel has quit IRC[04:18:17] *** riccardo has quit IRC[04:27:31] *** POloser has joined #openindiana[04:38:18] *** szaydel has joined #openindiana[04:54:37] *** szaydel has quit IRC[04:56:08] *** szaydel has joined #openindiana[04:57:31] *** Seony has joined #openindiana[04:58:06] *** szaydel has quit IRC[05:16:05] *** ball has quit IRC[05:20:24] *** szaydel has joined #openindiana[05:30:20] *** szaydel has quit IRC[05:33:04] *** OMV-User has quit IRC[05:36:13] *** OMV-User has joined #openindiana[05:36:31] *** szaydel has joined #openindiana[05:36:53] *** BonzTM has quit IRC[05:37:33] *** szaydel has quit IRC[05:49:51] *** szaydel has joined #openindiana[06:14:16] *** TomJ has quit IRC[06:18:21] *** szaydel has quit IRC[07:04:29] *** EisNerd has quit IRC[07:09:47] *** jw_urodoc has joined #openindiana[07:10:00] *** LabMan has joined #openindiana[07:11:55] *** EisNerd has joined #openindiana[07:25:09] *** jw_urodoc has quit IRC[08:29:40] *** nikolam has joined #openindiana[08:34:39] <nikolam> hi, does anyone know if 320bit OI 151a6 dev release is fixed in IPS/pkg to be able to boot after update from 151a4/151a5 ?[08:34:56] <nikolam> 32bit x86 :)[08:58:10] <nikolam> Or is there a group that test OI releases prior release, to avoid such huge problems with releases.[09:01:54] *** andy_js has joined #openindiana[09:07:02] *** Webhostbudd has joined #openindiana[09:09:02] *** anikin has joined #openindiana[09:09:15] *** |AbsyntH| has joined #openindiana[09:20:37] *** ctable has joined #openindiana[09:26:05] *** loganbr has quit IRC[09:42:09] *** Micr0mega has joined #openindiana[09:53:44] *** InTheWings has joined #openindiana[09:55:59] *** tsoome has quit IRC[10:24:39] *** tsoome has joined #openindiana[10:29:18] *** nikolam has quit IRC[10:37:01] *** Whoopsie has joined #openindiana[10:37:01] *** ChanServ sets mode: +v Whoopsie[10:46:21] <jkj> btw. does anyone get the philosophy point here: https://www.illumos.org/issues/280[10:49:18] <alanc> there's about 3-4 years of history there, possibly found in the indiana-discuss at opensolaris dot org archives, possibly lost when things like the opensolaris forms & bug tracker went dark[10:49:40] <jkj> ok. have to google a bit[10:49:42] <alanc> Solaris 11 has /usr/bin first though, not /usr/gnu/bin 8-)[10:49:55] <jkj> understandable as gnu-tools lack features[10:50:10] <alanc> but they also have many other features the core tools lack[10:50:40] <jkj> true[10:51:33] <alanc> if my memory is working correctly, the original indiana plan was that /usr/gnu/bin was put first to ease users coming over from Linux or BSD, while work was supposed to begin in parallel to enhance the Solaris versions so they could be restored to primary in the future[10:52:01] <Alasdairrr> alanc: is /usr/gnu/bin still at the front on S11?[10:52:01] <alanc> part of the OpenSolaris/indiana plan for increasing adoption and easing migration[10:52:04] <jkj> sounds reasonable. nobody has had the time to do it?[10:52:15] <alanc> Alasdairrr, nope[10:52:41] <Alasdairrr> interesting[10:52:53] <alanc> got a few gnu features into core, not all of them, but obviously S11 has a different strategy than OpenSolaris did[10:53:39] <jkj> well at least a newbie i came across (who linked the bug) saw the lack of acl as the real problem and not so much different "feel" of the native tools[10:54:00] <jkj> and i tend to somewhat agree :)[10:54:13] <jkj> then again it'd be trivial just to write the missing bits in gnu tools[10:54:29] <alanc> the ones that really hit people were things like ls --color and tar uncompression support[10:54:44] <alanc> both OI & S11 have color support in /usr/bin/ls[10:55:06] <alanc> S11 got tar compression support in /usr/bin/tar - don't remember if illumos/OI did[10:55:29] <Alasdairrr> We should swap out tar with star! :-D[10:56:08] * alanc suddenly has the old "Waiting for a star to fall" song stuck in his head[10:56:33] <JT-EC> illumos tar has compression support.[10:56:41] <jkj> do we have pax?[10:56:50] <Alasdairrr> yeah[10:56:58] <trochej> Coffee[10:57:04] <jkj> coffee.[10:57:10] <Alasdairrr> i think the compression support in tar was added yonks ago in opensolaris[10:57:43] <alanc> and really, anyone who understands Unix well enough to grok ACL's should find "export PATH=/usr/bin:..." a no-brainer[10:57:53] <JT-EC> 4 months ago for j and xz support[10:58:44] <jkj> alanc: yeah. problem was that it was suprising and didn't do as the (solaris) documentation said[10:59:10] <alanc> yeah, it's annoyed a good number of people[10:59:35] <Whoopsie> Indeed ;-)[10:59:37] *** tsoome has quit IRC[11:00:37] *** tsoome has joined #openindiana[11:01:03] <alanc> anyways, I should go into the office now, so will leave the people who actually work on OI to figure out their plans here...[11:01:14] <alanc> (last day in Dublin, then finally going home!)[11:01:25] <jkj> alanc: where is home?[11:01:29] <tsoome> :)[11:01:34] <alanc> California[11:01:45] <jkj> ok. long flight :)[11:01:53] <alanc> been traveling around Europe for work the last 3 weeks[11:02:09] <alanc> 2 long flights for me - Dublin -> NWK, then NWK->SFO[11:02:36] <tsoome> 2 hops?[11:02:38] *** alanc has quit IRC[11:02:48] <tsoome> ah, he is gone:=[11:05:24] <jkj> remember one time our flight was late and we hurried to terminal and found out the next flight was going to be flown with the same airplane :P[11:06:07] <jkj> they just emptied it and also changed the crew[11:13:01] *** tsoome has quit IRC[11:13:30] *** ctable has quit IRC[11:14:33] *** tsoome has joined #openindiana[11:19:12] *** aaearon has quit IRC[11:29:25] *** aaearon has joined #openindiana[11:50:07] *** alanc has joined #openindiana[11:50:07] *** ChanServ sets mode: +o alanc[11:51:51] *** Kaishi has quit IRC[12:05:16] *** flyz_ has joined #openindiana[12:07:52] *** flyz has quit IRC[12:08:08] *** ancoron_w has quit IRC[12:12:15] *** anikin_ has joined #openindiana[12:15:02] *** anikin has quit IRC[12:15:11] *** ancoron_w has joined #openindiana[12:17:44] *** master_of_master has quit IRC[12:19:29] *** master_of_master has joined #openindiana[12:27:59] *** TomJ has joined #openindiana[12:28:00] *** TomJ has quit IRC[12:28:00] *** TomJ has joined #openindiana[12:30:14] *** movement has quit IRC[12:37:32] *** ancoron_w has quit IRC[12:41:11] *** ancoron_w has joined #openindiana[12:46:05] *** ancoron_w has quit IRC[12:46:32] *** TomJ has quit IRC[12:49:31] *** ancoron_w has joined #openindiana[12:53:25] *** ancoron_w has quit IRC[12:53:35] *** enricop has joined #openindiana[12:57:06] *** ancoron_w has joined #openindiana[13:11:35] *** andy_js has quit IRC[13:11:43] *** ozquera has joined #openindiana[13:16:32] *** TomJ has joined #openindiana[13:16:32] *** TomJ has quit IRC[13:16:32] *** TomJ has joined #openindiana[13:22:01] *** JohnG_ has joined #openindiana[13:29:22] *** jellydonut has joined #openindiana[13:31:12] *** TomJ has quit IRC[13:32:51] *** andy_js has joined #openindiana[13:32:59] *** ancoron_w has quit IRC[13:36:25] *** ancoron_w has joined #openindiana[13:37:20] *** ancoron_w has quit IRC[13:40:46] *** ancoron_w has joined #openindiana[13:41:27] *** movement has joined #openindiana[13:46:10] *** ancoron_w has quit IRC[13:49:33] *** ancoron_w has joined #openindiana[13:55:34] *** JohnG_ has quit IRC[14:06:12] *** ancoron_w has quit IRC[14:07:55] *** ira has joined #openindiana[14:09:59] *** ancoron_w has joined #openindiana[14:11:31] *** hunter has quit IRC[14:13:52] *** ancoron_w has quit IRC[14:17:59] *** ancoron_w has joined #openindiana[14:22:52] *** ancoron_w has quit IRC[14:26:24] *** ancoron_w has joined #openindiana[14:26:24] *** ancoron_w has quit IRC[14:26:30] *** Seony has quit IRC[14:28:51] *** hunter has joined #openindiana[14:29:31] *** gwr has quit IRC[14:30:18] *** ancoron_w has joined #openindiana[14:31:39] *** POloser has left #openindiana[14:33:42] *** ancoron_w has quit IRC[14:35:08] *** CVLTCMK0 has quit IRC[14:51:28] *** TomJ has joined #openindiana[14:58:40] *** Sachiru has joined #openindiana[15:15:34] *** szaydel has joined #openindiana[15:19:05] *** alanc has quit IRC[15:22:11] *** junak-michal has joined #openindiana[15:25:56] *** ancoron_w has joined #openindiana[15:26:06] <junak-michal> Hi. I have one problem related to oi networking and I was wondering if anyone here could help me solve it. I have one network card and I am trying to create a zone with virtual card, that will be using dhcp to resolve its IP address etc. I have created vnic0 (dladm create-vnic -l rge0 vnic0) and added it to my zone (add net; set physical=vnic0; in zonecfg). When I set it to use dhcp (ifconfig vnic0 dhcp start) inside my zone, it g[15:26:07] <junak-michal> ets ip address. I can ping other computers by their ip addresses, but DNS is not working (I can't ping www.google.com, but I can ping google on its ip address). dhcpinfo -i vnic0 6 returns correct entries, so I do not know where the problem is. So my question is, how to get DNS working.[15:27:09] *** Uranio has joined #openindiana[15:27:21] <tsoome> check /etc/nsswitch.conf and make sure you have dns on hosts and ipnodes lines, then check /etc/resolv.conf[15:29:31] *** ancoron_w has quit IRC[15:31:16] <junak-michal> looks promising tsoome, thanks[15:33:07] *** enricop has quit IRC[15:33:18] <tsoome> if you have changed nsswitch.conf, you may need to restart nscd (pkill nscd or svcadm restart)[15:36:02] *** KermitTheFragger has joined #openindiana[15:38:42] *** ancoron_w has joined #openindiana[15:44:52] *** hunter has quit IRC[15:44:55] *** hunter__ has joined #openindiana[15:47:50] <junak-michal> ok, it really works, thanks again tsoome.[15:50:36] <Sachiru> Read about this article: http://lwn.net/Articles/41652/ . Why does Reiser4 sound uncomfortably like ZFS?[15:51:26] *** ancoron_w has quit IRC[15:51:33] *** hunter has joined #openindiana[15:52:11] *** hunter__ has quit IRC[15:53:01] *** hunter has quit IRC[15:53:31] *** hunter has joined #openindiana[16:05:48] *** tsoome_ has joined #openindiana[16:05:49] *** tsoome_ has quit IRC[16:09:43] *** tsoome has quit IRC[16:11:28] *** APTX has quit IRC[16:11:43] *** APTX has joined #openindiana[16:19:36] *** APTX has quit IRC[16:24:07] *** |AbsyntH| has quit IRC[16:25:02] *** |AbsyntH| has joined #openindiana[16:31:40] *** Micr0mega has quit IRC[16:32:01] *** Micr0mega has joined #openindiana[16:42:41] *** anikin has joined #openindiana[16:44:34] *** anikin_ has quit IRC[16:45:21] *** Sachiru has quit IRC[16:46:02] *** gwr has joined #openindiana[16:47:41] *** Aethrs has joined #openindiana[17:03:24] *** TomJ has quit IRC[17:03:52] *** TomJ has joined #openindiana[17:03:53] *** TomJ has quit IRC[17:03:53] *** TomJ has joined #openindiana[17:04:14] *** TomJ has quit IRC[17:04:23] *** TomJ has joined #openindiana[17:04:24] *** TomJ has quit IRC[17:04:24] *** TomJ has joined #openindiana[17:06:22] *** riccardo has joined #openindiana[17:06:58] *** tsoome has joined #openindiana[17:20:17] *** TomJ has quit IRC[17:20:35] *** TomJ has joined #openindiana[17:20:36] *** TomJ has quit IRC[17:20:36] *** TomJ has joined #openindiana[17:22:52] *** Uranio has quit IRC[17:24:36] *** Uranio has joined #openindiana[17:26:02] *** jim80net has quit IRC[17:29:18] *** gvm has joined #openindiana[17:33:28] *** Whoopsie has quit IRC[17:33:32] *** anikin has quit IRC[17:47:35] *** imaxs has quit IRC[17:50:11] *** Uranio has quit IRC[17:51:38] *** Uranio has joined #openindiana[17:58:58] *** jim80net has joined #openindiana[17:59:04] *** |AbsyntH| has quit IRC[18:00:56] *** LabMan has quit IRC[18:01:45] *** LabMan has joined #openindiana[18:02:37] <Stry> Heya guys,[18:03:17] *** kdavyd has joined #openindiana[18:03:32] *** LabMan has quit IRC[18:03:33] <Stry> Just read this blog about retrieving destroyed pools. Apparently it's a small change in a couple of files...[18:04:16] *** LabMan has joined #openindiana[18:04:16] <Stry> and then you'll be able to see destroyed pools with "zpool import" and then import them using the -f flag.[18:05:16] <Stry> Maybe it's interesting to add this change to the next version?[18:05:47] <Stry> http://milek.blogspot.nl/2006/02/zfs-recovering-destroyed-pools.html[18:10:41] *** Stry has quit IRC[18:11:11] *** riccardo has quit IRC[18:25:41] *** KermitTheFragger has quit IRC[18:30:06] *** Micr0mega has left #openindiana[18:39:38] *** nightwalk has quit IRC[18:40:18] *** loganbr has joined #openindiana[18:41:06] *** nightwalk has joined #openindiana[18:54:52] *** junak-michal has quit IRC[18:57:34] *** imaxs has joined #openindiana[19:16:35] *** Stry has joined #openindiana[19:19:54] <gvm> I just rebooted my OI148 system, and it threw a console message after booting saying "savecore: [ID 570001 auth.error] reboot after panic: BAD TRAP: type=e (#pf Page fault) rp=ffffff00109aa7a0 addr=0 occurred in module "unix" due to a NULL pointer dereference". It rebooted spuriously a couple of weeks ago, with the same message. Is this really likely to be a new panic, or is it reporting...[19:19:55] <gvm> ...debris from the previous panic?[19:21:58] *** TomJ has quit IRC[19:22:32] *** ancoron_ has quit IRC[19:24:25] *** ancoron has joined #openindiana[19:36:02] <gvm> OK, looking back in the logs, it looks like it's just repeating a warning about the previous panic. Do I need to fmadm acquit this to stop the warnings?[19:48:44] *** patdk-lap has joined #openindiana[19:52:24] *** InTheWings has quit IRC[19:53:14] *** ancoron has quit IRC[19:56:18] *** APTX has joined #openindiana[20:02:52] *** ancoron has joined #openindiana[20:05:04] *** InTheWings has joined #openindiana[20:05:52] *** APTX has quit IRC[20:06:02] *** APTX has joined #openindiana[20:12:33] *** maccampus has joined #openindiana[20:13:45] *** slx86 has joined #openindiana[20:29:19] *** riccardo has joined #openindiana[20:30:19] *** andy_js has quit IRC[20:35:07] *** patdk-lap has quit IRC[20:36:06] *** patdk-lap has joined #openindiana[20:44:36] *** enricop has joined #openindiana[20:50:53] *** Seony has joined #openindiana[20:52:59] <flexo> okay.. so iostat -E doesn't update it's information after a hotswap..[20:53:06] <flexo> and diskutil is not available on OI[20:53:45] <flexo> some other place i can get some information about the new disk? (well, i found the vendor and s/n - but i'd like to know... more.)[20:55:30] *** TomJ has joined #openindiana[20:56:40] <lblume> cfgadm is probably what you are looking for. Depending on your HBA, you mayb have to tell the OS you are going to replace a disk[21:00:01] <flexo> doesn't look that way - the disk failed and cfgadm already reported the bus as unconnected / unconfigured at that point[21:01:27] <flexo> so i replaced it, -xsata_port_activate... at which point it was connected / configured => zpool replace[21:02:05] <flexo> the second time i did this since the box is running.. everything is working fine - it's just iostat which still reports information about the the old (old) disk[21:02:25] <flexo> well.. it didn't come up by itself :) i suppose it should have[21:02:39] *** ThePendulum has joined #openindiana[21:02:41] <lblume> Usually it should be -c connect/configure[21:02:47] <ThePendulum> Hey[21:02:54] <flexo> hmm... connect.. i suppose i should maybe have tried that[21:03:15] <flexo> but i don't want to trash my 10% resilver now ;)[21:03:50] <flexo> still 80 hours to go.. maybe.[21:04:05] <lblume> Thing is, -x commands are sent to the hardware, not the OS itself, so the change of state night not be noticed[21:04:15] <lblume> 80 hours? How big is it?[21:05:06] <flexo> just 6x3tb raidz2 - but a lot (lot!) of small files, some iscsi volumes, with changes happening everywhere, all the time, and everything being autosnapped...[21:06:12] <flexo> (both adding the volume containing the huge amount of files and adding the iscsi volumes increased the scrub time by a few orders of magnitude so i suppose the same goes for resilver..)[21:06:37] *** patdk-lap has quit IRC[21:07:21] <ThePendulum> I am looking for an OS for my NAS (just for home), which can be straightforward and not too hard to configure for a nut like me who's never been any good at setting permissions. Is #openindiana something to consider, or is it more an enterprise solution?[21:08:21] <flexo> lblume: but thanks for the connect hint. i wrote it down in our wiki - next time a drive fails (which should be in a month at the current rate :) i'll try it out[21:09:52] <flexo> huh.. while i'm looking at the cfgadm output..[21:09:52] *** szaydel has quit IRC[21:10:13] <flexo> i don't see the rpool drives (on another HBA)[21:10:18] <Gibheer> ThePendulum: you could take a look at freenas[21:10:47] <Gibheer> ThePendulum: but if you want to invest the time to learn more, then openindiana may be a good option[21:11:32] <flexo> ah okay. nevermind about that.[21:11:57] <ThePendulum> Gibheer: FreeNAS is exactly what I am having trouble with atm. At this point, I thought I set it so virtually anyone had every possible permission, and now no one is able to reach it.[21:12:03] <lblume> flexo: In general, allthe -x commands are for exceptional use, the regular ones are -c. Also, the -c ones are safer.[21:12:09] <ThePendulum> I guess if openindiana is a challenge upon that, I will have to continue my hunt[21:12:22] <lblume> As for the rpool, is it on a legacy mode controller? As IDE drives?[21:13:30] *** szaydel has joined #openindiana[21:13:42] <flexo> nope... some pci-e controller.. but i found it.. dynamic attachment points[21:15:28] <flexo> lblume: in any case - the internet agrees with me that you hve to reboot for "iostat -E" to update it's information.. on solaris at least[21:15:34] *** ThePendulum has quit IRC[21:16:11] <lblume> Well, I can assure that itis not true.[21:16:32] <lblume> I upgraded the firmware of 6 disks, without a reboot, and iostat -E showed the change fine[21:16:37] <flexo> hmm okay[21:16:44] <lblume> well, hmm.[21:16:59] <lblume> Okay, "it is not true on all HBA's""[21:17:04] <flexo> :)[21:17:43] <flexo> how do i find the sd?? => c?t?d? mappings?[21:17:58] <lblume> But I've not seen one not behaving properly. It should work if you do a proper cfgadm us[21:18:24] <lblume> iostat -En[21:18:44] <flexo> well.. i'm assuming that one is not telling the truth ;)[21:19:15] <flexo> also.. no? i don't see the sd?? number there[21:19:38] <flexo> ah[21:19:40] <lblume> just ignore the sd ones[21:19:41] <flexo> i see[21:20:06] <lblume> they're useless on Solaris. It's very moldy cruft that has been left over from 1988.[21:20:55] <flexo> hm[21:20:57] <Triskelios> iostat -E vs -En[21:21:13] <flexo> you really want to me to reconnect using cfgadm now, right? :([21:21:23] <flexo> well okay. better now than tomorrow[21:21:33] <lblume> Naw,your choice :-D[21:21:42] <flexo> what could possibly go wrong![21:21:52] <flexo> so.. first..[21:21:57] <flexo> somehow remove it from the zpool[21:22:35] <flexo> alright[21:23:20] <flexo> unconfigured..[21:23:34] <flexo> disconnected[21:24:11] <flexo> reconnected and configured[21:24:28] <flexo> okay. iostat -E displays the correct stuff now.[21:24:55] <lblume> See? :-)[21:25:04] <lblume> Resilver should restart where it left,too.[21:25:33] <flexo> nope :([21:25:39] <lblume> It's not a scrub, it only deals with the modified parts[21:25:56] <lblume> Well, it will start at 0%, but that will not include what was already resilvered[21:26:18] <flexo> ah[21:26:19] <flexo> correct[21:26:21] <flexo> thanks :)[21:26:30] <flexo> thanks very much.[21:26:47] <flexo> this feels much better[21:27:33] <lblume> heh, yw.[21:28:37] <lblume> It's quite a heap of data for sata, better feel comfortable.[21:29:09] <flexo> you mean my 6x3tb?[21:32:28] <lblume> yup[21:33:18] <lblume> feeling the excitement of a disk failure and replacing it on time, too :-)[21:33:54] <flexo> hehe[21:34:12] <flexo> of course the person holding the spares is on vacation for 2 weeks[21:34:19] <flexo> so i phoned him.. and told me to get some from someone else[21:34:28] <flexo> who was also on vacation for 2 weeks[21:34:31] <flexo> locked the spares away[21:34:34] <flexo> and took the key with him[21:35:29] <flexo> took me a while but i still managed to get a disk[21:35:45] <flexo> OTOH i'm glad that i now know that the email notification thingie actually works[21:36:28] <flexo> (last time i replaced the drive before it failed - i just noticed that the amount of errors was accumulating fast)[21:37:31] <lblume> it's business stuff? on sata?[21:38:01] <flexo> yea :D[21:39:28] <lblume> I'd try an SAS controller with SATA disks,at least.[21:39:49] <Alasdairrr> 7.2k RPM SAS disks are quite common now and not much more than SATA[21:40:09] <flexo> we really stress tested the controllers.. i trust them[21:40:10] <Alasdairrr> Although I think they have crippled firmware on them in the failure case[21:40:47] <flexo> (i believe this was the fourth controller was bought and tested...)[21:41:08] <lblume> Alasdairrr: crippled how?[21:41:10] <Alasdairrr> what controller are you using? sorry if you've already said, joining this late and cba reading the enormous scrollback[21:41:27] <Alasdairrr> lblume: Not failing completely[21:41:40] <lblume> flexo: Spending so much time on a SATA controller, an SAS one would surely be better :-P[21:41:51] <flexo> hm.. is there a way to find it out easily? :)[21:42:37] <lblume> scanpci should list it[21:42:52] <lblume> Alasdairrr: You mean, it keeps retrying?[21:43:16] <flexo> Marvell Technology Group Ltd. 88SE6121 SATA II Controller <= onboard, 6 ports[21:43:26] <flexo> LSI Logic / Symbios Logic SAS1068E PCI-Express Fusion-MPT SAS <= two ports, not supported by OI :([21:43:43] <flexo> Intel Corporation 82801JI (ICH10 Family) SATA AHCI Controller <= pci-e, 2x4 ports[21:43:44] <Alasdairrr> lblume: It's hard to know whether it's mpt_sas not doing the right thing or the drive not doing the right thing[21:43:57] <flexo> i believe.[21:45:01] <flexo> we have a triple (varios sizes) mirror-0 for rpool on the intel one and 6x3tb raidz2 on the onboard one[21:45:09] <lblume> flexo: the 1068 was not supported? Strange that.[21:45:32] <flexo> hm[21:45:35] <flexo> or.. wait a sec[21:45:41] <flexo> maybe i'm confusing stuff[21:45:55] <flexo> yea right[21:46:03] <flexo> it was the marvell which is actually SAS and not SATA[21:46:07] <flexo> which is not supported[21:46:16] <flexo> the intel is 6x onboard[21:46:24] <flexo> and the lsi is 2x4 pci-e[21:46:42] <lblume> Alasdairrr: Yeah, I know first hand how difficult it can be to troubleshoot that if you don't have the driver's author on your side :-/[21:46:48] *** enricop has quit IRC[21:47:19] <flexo> so we have the 3/mirror-0 on the lsi pci-e and the 6/raidz2 on the intel onboard[21:47:58] <flexo> has been running under a lot of stress (avg bandwidth is 10mb/s or something) for 6 weeks now.. stable.[21:48:12] <lblume> I'd have put the raidz on the LSI. It'd give it a small perf boost.[21:48:22] *** slx86 has quit IRC[21:48:41] <Alasdairrr> lblume: Trisk/Nexenta fixed some mpt_sas stuff[21:48:51] <flexo> lblume: yea, i would have preferred that.. the thing is we started with some unstable controller.. and to get it to work at all we had to move the low-usage rpool to the unstable controller[21:48:53] <Alasdairrr> so time will tell[21:48:57] <flexo> so the system would run it all :)[21:49:33] <flexo> oh, and we are missing one of these 1 => 4 cables[21:50:39] <flexo> soon we'll set up our offsite backup server (same hardware) - then we can mess around with the configuration[21:53:08] <lblume> Alasdairrr: Who provides code for mpt_sas, mr_sas, dr_sas, and friends? LSI? Or it; s now a separate branch with LSI talking only to Oracle?[21:57:10] <richlowe> dr_sas is mr_sas[21:57:53] <flexo> by the way.. is there something one can buy to support oi / illumos development? like.. you know.. support which doesn't actually exist.[21:57:59] <richlowe> but it's definitely now a separate branch.[21:59:25] <lblume> How comes? Don't LSI provide code for the Linux drivers?[21:59:51] *** LabMan has quit IRC[22:00:22] *** LabMan has joined #openindiana[22:05:04] <flexo> also.. http://pastie.org/private/w6vv282cgojhxnlvglclta <= can someone tell me where that 1 terabyte comes from? it just doesn't add up.[22:06:04] <richlowe> lblume: doing anything at all with LSI is never easy. I think they've been helpful to us regarding some stuff, at least[22:12:57] *** andy_js has joined #openindiana[22:14:47] *** patdk-lap has joined #openindiana[22:15:44] *** ozquera has quit IRC[22:17:53] <lblume> Ah, I see[22:18:36] *** maccampus has quit IRC[22:29:49] *** patdk-lap has quit IRC[22:30:14] *** patdk-lap has joined #openindiana[22:44:34] *** Kaishi has joined #openindiana[22:44:48] <Alasdairrr> they are petrified of Oracle[22:44:57] <Alasdairrr> Oracle are one of the largest companies in the world, who are known to sue people for fun[22:45:16] <Alasdairrr> It looks like Joyent were partially successful in getting code out of them for dr_sas[22:45:31] <Alasdairrr> Although apparently it had been redacted quite a bit[22:45:57] <Alasdairrr> with lsi even going as far as to remove portions of code that were already in mpt_sas in illumos[22:46:10] <Alasdairrr> which is silly[22:47:08] <Alasdairrr> It's within LSI's interest to help illumos, as shittonnes of ZFS based storage hangs off LSI cards/chips[22:47:22] <Alasdairrr> I guess they just have to be careful[22:49:02] <lblume> redacting code? That is weird stuff[22:49:25] *** hunter_ has joined #openindiana[22:49:35] <Alasdairrr> lblume: I suspect they co-authored it with Oracle[22:49:58] <Alasdairrr> For example with mpt_sas there are portions written by Sun evident in the onnv commit logs[22:50:03] <lblume> ah, so the copyright might be shared or ambiguous :-/[22:50:27] <Alasdairrr> Maybe not shared copyright[22:50:45] <patdk-lap> licensing is fun :)[22:50:50] <Alasdairrr> I'm more thinking that the driver was written by people who had signed up to Oracle's developer program so they saw the source to Solaris 11[22:50:59] *** Seony has quit IRC[22:51:08] <Alasdairrr> So their drivers make use of interfaces present in Solaris 11 that perhaps aren't in illumos[22:51:11] <Alasdairrr> etc etc[22:51:18] <Alasdairrr> who knows what goes on behind closed doors[22:51:59] *** hunter has quit IRC[22:52:00] *** hunter_ is now known as hunter[22:52:03] <lblume> Sometimes, very boring stuff[22:52:08] <Alasdairrr> yep[22:52:27] *** riccardo has quit IRC[22:53:44] <Alasdairrr> We're kind of lucky Oracle isn't genuinely evil with evil intentions as they could probably have forbidden LSI from working with illumos if they wanted[22:54:52] <lblume> I don't believe in evilness. Just like an elephant crushing an ant to get to his favourite banana tree isn't really evil either.[22:55:04] <Alasdairrr> :-) indeed[22:55:37] <lblume> Just, from the viewpoint of the ant, well, there's not much of a difference :-/[22:56:24] <Alasdairrr> Well Oracle thankfully just don't seem to care, and that suits everyone just fine[22:56:31] <Alasdairrr> They even let some trademarks expire i think[22:58:02] <Alasdairrr> http://tdrapi.uspto.gov/ts/cd/casestatus/sn85194050/content.html[22:58:19] <Alasdairrr> expired this time last year[22:58:35] <richlowe> I miss Sun's old trademark page[22:58:38] <richlowe> it was full of nostalgia[22:58:50] *** Uranio has quit IRC[22:59:11] <Alasdairrr> yeah :-)[22:59:14] *** pogs_ has quit IRC[22:59:15] <Alasdairrr> its probably on the wayback machine[23:01:29] *** pogs has joined #openindiana[23:03:47] <lblume> I don't think they'll go out of their way to crush ants. But they'd probably fight hard if one of them dared to take more than a couple of bites from a banana.[23:04:32] <lblume> I really can't fathom how they work. Too big, I guess.[23:05:56] <TomJ> What basis can they have for crushing anything forked from legally open sourced OpenSolaris code? Patent infringement? Or you think they'd just do it without a case to crush those who can't afford to defend?[23:07:10] *** ancoron_ has joined #openindiana[23:09:58] <lblume> If there's even the slightest ambiguity as to whom wrote the code, I guess a case could take years to decide. See SCO.[23:10:58] *** ancoron has quit IRC[23:17:09] *** andy_js has quit IRC[23:21:10] *** hunter has quit IRC[23:22:50] *** hunter has joined #openindiana[23:22:56] <ancoron_> ...that let's me believe that IBM would have been a better home for Sun (cause I just don't know nearly as much bad news from IBM as from Oracle)...[23:27:48] <lblume> Hard to say. Different, yes, better, who knows? IBM has so many platforms already.[23:40:23] <ancoron_> At least from my point of view they know that open-source is getting a bigger role in the ecosystem with every year, so they'd presumely play the game a bit nicer than Oracle (ahem, Larry) does[23:43:19] <copec> Oracle doesn't really seem to care for what they got out of Sun in terms of Solaris...only as much as to capitalize products that were already making money[23:44:32] <copec> Coming from the atypical linux/gnu world, *olaris so much better in so many ways imo[23:44:49] <copec> not that I don't have my linux/gnu love still[23:49:45] *** flyz_ has quit IRC[23:50:13] <TomJ> Keep working with Solaris, the linux/gnu love will pass in time.