November 14, 2013 
< | 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 | >
November 14, 2013 [00:00:27] *** desai has quit IRC[00:01:57] *** joltman has left #omnios[00:45:59] *** wez is now known as wez|away[00:54:14] *** wez|away is now known as wez[01:05:43] *** nikolam has quit IRC[01:07:22] *** khushildep has joined #omnios[01:58:35] *** desai has joined #omnios[01:59:10] <ashley_w> i just created my first zone, and it failed twice while installing (third time's a charm?): https://dl.dropboxusercontent.com/u/57179934/zone-error.txt[01:59:38] <ashley_w> is this normal? is there a way to have it retry a bit?[02:08:43] *** wez is now known as wez|away[02:12:08] *** desai has quit IRC[02:13:55] *** wez|away is now known as wez[02:17:44] *** khushildep has quit IRC[02:19:35] *** schleicher has quit IRC[02:20:24] *** jpeach has quit IRC[02:23:25] *** jeffrymolanus has quit IRC[02:33:29] *** gmason has quit IRC[02:52:42] *** desai has joined #omnios[03:09:25] *** wez is now known as wez|away[03:31:28] *** jpeach has joined #omnios[03:34:20] *** desai has quit IRC[03:34:44] *** Azbruh has quit IRC[03:35:16] *** desai has joined #omnios[03:35:49] *** jpeach has quit IRC[04:00:06] *** jpeach has joined #omnios[04:00:35] *** Azbruh has joined #omnios[04:04:22] *** szaydel has quit IRC[04:23:44] *** wez|away is now known as wez[04:29:04] *** szaydel has joined #omnios[04:53:06] *** Thrae has quit IRC[04:58:39] *** Thrae has joined #omnios[05:12:01] *** wez has quit IRC[05:20:55] <apeiron> ashley_w, that's an issue with our backend stuff that we're aware of[05:49:32] *** desai has quit IRC[05:50:54] *** mayuresh has joined #omnios[05:51:32] <mayuresh> hi :)[05:51:44] <mayuresh> is postwait around here at the moment?[05:51:58] <apeiron> he's at velocity London[05:52:03] <mayuresh> okay[05:56:17] <mayuresh> isn't he jesus@<company_name>?[05:57:09] *** Jeztah has joined #omnios[05:57:41] <apeiron> that should get to him, but I'm doubtful only he can answer your questions[05:58:01] <mayuresh> no worries, i'll write in a after a week, would that be fine?[05:58:11] <apeiron> it's not my inbox[05:58:34] <mayuresh> yup :)[05:59:42] <mayuresh> is there anyone with a repository willing to hold packages built by others?[05:59:54] <apeiron> doubtful[06:00:02] <apeiron> ask the list[06:00:07] <mayuresh> okay[06:00:16] <apeiron> I can tell you that omniti won't[06:00:20] <mayuresh> yup[06:00:32] <mayuresh> figured that, wanted to know if anyone else be willing...[06:01:28] <danlarkin> mayuresh: what kind of packages are you willing to build but not host?[06:01:57] <mayuresh> "cannot" host, don't have a box with a public ip... :([06:02:02] <mayuresh> packages for general usage[06:02:12] <mayuresh> as well as those from a friend building x and cde[06:02:37] <danlarkin> oh goodness[06:02:48] <mayuresh> :)[06:03:11] <apeiron> the opinions stated here do not represent those of omniti[06:03:14] <apeiron> you should be using openindiana[06:03:16] <apeiron> or better yet linux[06:03:20] <mayuresh> nope[06:03:22] <apeiron> yes.[06:03:29] <apeiron> you are forcing desktop packages into a server OS[06:03:35] <mayuresh> i'm happy with omnios, just wish to be happier ;)[06:04:02] <mayuresh> isn't omnios more a jeos instead of being a server os?[06:04:26] <thebug> you may find that you are happier building X from pkgsrc. jperkin, the joyent, and the netbsd people put quite a bit into making that sort of thing work[06:04:32] <apeiron> they're essentially the same thing[06:04:36] <apeiron> servers are best kept minimal[06:04:38] <thebug> (happier building it from pkgsrc than doing it by hand, I should say.)[06:05:02] <mayuresh> pkgsrc does tend to pull in stuff that is usually not essential[06:05:15] <mayuresh> or rather *very* essential[06:05:19] <apeiron> only if you use the default[06:05:24] <mayuresh> hmnn[06:05:43] <danlarkin> if you're building X haven't you given up on only the essentials?[06:05:48] <apeiron> ^[06:05:51] <mayuresh> :D[06:05:53] <thebug> I think you find that it follows from KYSTY and OmniOS being 'just enough' OS, that people hosting repositories are only building what they or their organization needs[06:06:09] <thebug> rather than an ubuntu style launchpad/ppa thing[06:06:30] <mayuresh> yeah, i ain't talking about launchpad style stuff[06:06:39] <mayuresh> just someone kind enough to host third-party packages[06:06:45] <thebug> you're asking someone to host your packages[06:07:01] <thebug> in IPS-land, that implies you have free reign to twiddle around in their repo[06:07:09] <thebug> all of it[06:07:18] <mayuresh> oh :o[06:08:28] <thebug> the server speaks HTTP, but it's not just apache/nginx/lighttpd serving up a directory of files, it's 'pkg' (the same software you use to install stuff) serving up a repository[06:08:53] <mayuresh> okay[06:09:04] <apeiron> pkg.depotd specifically[06:09:13] <apeiron> which is a special httpd that supports specific URIs, verbs, and caching[06:09:42] <mayuresh> okay[06:10:07] <mayuresh> danlarkin: x can be minimalist too[06:10:24] <mayuresh> danlarkin: check out the stuff that openbsd guys have built[06:10:43] <mayuresh> yeah, it's *big*, but it bundles only the essential stuff ;)[06:11:31] <apeiron> >it can be minimalist[06:11:34] <apeiron> >it's *big*[06:11:35] <apeiron> wat.[06:12:05] <mayuresh> it means, minimalisim is relative, isn't it? ;)[06:13:20] <apeiron> evidently[06:13:24] <apeiron> if you think X can be minimalist[06:13:29] <apeiron> meanwhile I have a task to kill dbus and hal[06:13:51] <danlarkin> yes! down with dbus[06:14:48] <apeiron> down with freedesktop altogether[06:14:55] <mayuresh> yes, i agree :)[06:15:00] <apeiron> but you want X11[06:15:08] <mayuresh> xorg to be precise[06:15:12] <mayuresh> :)[06:15:12] <apeiron> X11 is a freedesktop project these days, no?[06:15:52] <mayuresh> http://www.x.org/wiki/[06:16:32] <mayuresh> they supposedly do it in "conjunction" with the freedesktop folks[06:16:39] <mayuresh> what ever that means... ;)[06:17:15] <apeiron> it means "it burns get it away from me precious"[06:17:24] <mayuresh> :D[06:17:40] <mayuresh> i've got 5 months of spare time[06:17:56] <mayuresh> let's see how far i get along the x on omnios personal project[06:18:23] <mayuresh> even if it fails, no one loses[06:18:50] <mayuresh> but if it works out well, then i guess a bunch of people would be happy.[06:20:39] *** szaydel has quit IRC[06:27:03] <mayuresh> see you all later[06:27:05] *** mayuresh has quit IRC[07:00:03] *** nefilim has quit IRC[07:22:36] *** slx86 has joined #omnios[07:28:19] *** jpeach has quit IRC[07:28:53] *** berend has quit IRC[07:28:55] *** Clory has quit IRC[07:29:03] *** TBCOOL has quit IRC[07:29:03] *** rj has quit IRC[07:29:06] *** Gibheer has quit IRC[07:29:08] *** porkotron has quit IRC[07:29:09] *** RoyK has quit IRC[07:29:10] *** Mazon has quit IRC[07:29:10] *** ashley_w has quit IRC[07:32:29] *** wez has joined #omnios[07:47:18] *** daleg_ has joined #omnios[07:47:45] *** daleg has quit IRC[07:47:45] *** daleg_ is now known as daleg[08:39:36] *** jpeach has joined #omnios[08:43:29] *** wez has quit IRC[08:43:45] *** jpeach has quit IRC[09:01:29] *** hlangeveld has joined #omnios[09:26:06] *** daleg has quit IRC[09:28:45] *** KermitTheFragger has joined #omnios[09:39:17] *** daleg has joined #omnios[09:41:21] *** ksalman_ has joined #omnios[09:43:31] *** jeffrymolanus has joined #omnios[09:44:37] *** berend has joined #omnios[09:44:38] *** Clory has joined #omnios[09:44:38] *** rj has joined #omnios[09:44:39] *** TBCOOL has joined #omnios[09:44:39] *** Gibheer has joined #omnios[09:44:39] *** porkotron has joined #omnios[09:44:39] *** RoyK has joined #omnios[09:44:40] *** Mazon has joined #omnios[09:44:40] *** ashley_w has joined #omnios[09:44:56] *** mtg`` has joined #omnios[09:46:05] *** mtg` has quit IRC[09:46:15] *** Licenser has quit IRC[09:46:16] *** ksalman has quit IRC[09:46:48] *** Licenser has joined #omnios[10:22:12] *** schleicher has joined #omnios[10:22:46] *** schleicher has quit IRC[10:42:34] *** pygi has joined #omnios[11:05:16] *** vrou has joined #omnios[11:21:09] *** mihai_omniosuser has quit IRC[12:48:12] *** bens1 has joined #omnios[13:07:35] *** khushildep has joined #omnios[13:29:15] *** ira has joined #omnios[13:32:01] *** szaydel has joined #omnios[14:01:11] *** desai has joined #omnios[14:11:05] *** desai has quit IRC[14:22:35] *** sebasp has quit IRC[14:24:04] *** pygi has quit IRC[14:24:06] *** sebasp has joined #omnios[14:24:27] *** sebasp is now known as sebasp_[14:26:58] *** sebasp_ has quit IRC[14:37:07] *** desai has joined #omnios[14:40:11] *** desai has quit IRC[14:45:18] *** sebasp has joined #omnios[14:45:19] *** pygi has joined #omnios[14:48:55] *** mayuresh has joined #omnios[14:55:39] *** sebasp is now known as sebasp_[14:57:45] *** mayuresh has left #omnios[14:59:03] *** xeyed4good has joined #omnios[14:59:14] *** xeyed4good has left #omnios[15:28:10] *** sebasp_ is now known as sebasp[15:51:24] *** rzezeski has left #omnios[15:54:56] *** khushildep_ has joined #omnios[15:56:08] *** pygi has quit IRC[15:56:44] *** khushildep has quit IRC[15:56:44] *** khushildep_ is now known as khushildep[15:58:08] *** postwait has joined #omnios[15:58:10] *** ChanServ sets mode: +o postwait[15:58:57] *** slx86 has quit IRC[16:08:26] <lotheac> now that there are more omniti guys here I'll say again: please make http://omnios.omniti.com/media/r151006y.zfs.bz2[16:08:35] <lotheac> it's 404 currently[16:09:25] <postwait> lotheac: I'm not in the office, so I can't do that right now.[16:09:35] <lotheac> don't you have lackeys that can? ;)[16:09:42] <postwait> but, FWIW, it is quite easy to make one if you've goot root in a global zone.[16:10:01] <lotheac> yeah, I guess I could generate it myself[16:10:04] <postwait> you can install an older version of kayak and just run the image maker.[16:10:17] <postwait> it will make an image of latest IPS bits.[16:10:23] <lotheac> but just a heads up that maybe someone forgot to create that image along with the new media[16:10:34] <postwait> likely true[16:12:00] <lotheac> also I'm having issues with zetaback... it says 'error: 256' at line 791 (which I assume looks at waitpid's exit status)[16:12:05] <lotheac> how can it be 256 :)[16:12:16] <lotheac> => failure zfs_do_backup elbereth.niksula.hut.fi:niksula f: error: 256 at /opt/niksula/bin/i386/zetaback line 791. at /opt/niksula/bin/i386/zetaback line 826.[16:14:05] *** bens1 has quit IRC[16:15:27] *** vrou has quit IRC[16:18:03] *** gmason has joined #omnios[16:26:59] *** nefilim has joined #omnios[16:52:08] *** postwait has quit IRC[17:05:23] *** cafhacker has joined #omnios[17:05:30] *** cafhacker has left #omnios[17:05:58] *** jpeach has joined #omnios[17:15:34] *** jeffrymolanus_ has joined #omnios[17:16:02] *** desai has joined #omnios[17:16:39] *** jeffrymolanus has quit IRC[17:16:39] *** jeffrymolanus_ is now known as jeffrymolanus[17:22:16] *** neophenix has joined #omnios[17:28:37] *** esproul has joined #omnios[18:40:02] *** KermitTheFragger has quit IRC[18:52:24] *** joltman has joined #omnios[19:08:07] *** pygi has joined #omnios[19:09:09] *** wez has joined #omnios[19:15:15] *** khushildep has quit IRC[19:30:03] *** daleg has quit IRC[20:04:36] *** pygi has quit IRC[20:07:02] *** slx86 has joined #omnios[20:14:45] *** nhubbard_ has joined #omnios[20:15:05] *** nhubbard_ has joined #omnios[20:15:41] *** nhubbard has quit IRC[20:15:41] *** nhubbard_ is now known as nhubbard[20:20:13] *** philips_ has quit IRC[20:23:42] *** nhubbard has quit IRC[20:24:16] *** philips has joined #omnios[20:25:44] *** nick__ has joined #omnios[20:26:55] *** nick__ is now known as nhubbard[20:29:00] *** philips has quit IRC[20:31:19] *** philips has joined #omnios[21:12:41] *** wez is now known as wez|away[21:21:44] *** wez|away is now known as wez[21:35:10] *** joltman has quit IRC[21:46:10] *** slx86 has quit IRC[21:47:33] *** jpeach has quit IRC[21:48:08] *** jpeach has joined #omnios[21:49:07] *** joltman has joined #omnios[21:51:52] *** neophenix has quit IRC[22:00:52] <ashley_w> the more i learn about omnios/illumos, the more i love it[22:01:15] <apeiron> it has that effect, yes[22:01:46] <danlarkin> aside from that ACPI thread on the ML, me too :p[22:02:12] <thebug> I think that acpi thread has a certain OTHER effect on any/everyone that reads it[22:05:26] <ashley_w> i say this, and yet i have a feeling my box at home just panicked[22:05:40] <apeiron> lotheac, I'm building the 151006y ZFS image presently[22:05:54] <lotheac> apeiron: whee :)[22:05:56] <lotheac> thanks[22:06:18] <apeiron> that was indeed an oversight in the jumble of getting y out and ramping up for 008[22:06:28] <thebug> ooh awesome, thanks for the heads-up apeiron[22:06:55] <ashley_w> ah, can log in again. but it rebooted[22:07:21] <thebug> well on the upside, you probably have a nice crashdump to crawl through now :)[22:08:50] <ashley_w> i now have 3 of them[22:15:41] *** wez is now known as wez|away[22:16:23] *** wez|away is now known as wez[22:21:57] <apeiron> lotheac, give http://omnios.omniti.com/media/r151006y.zfs.bz2 a try?[22:23:05] <lotheac> at least it's downloading ;) I'm not at work for today anymore, will try it in action later[22:23:09] <apeiron> ok[22:23:09] <lotheac> thanks[22:23:11] <apeiron> np[22:23:14] <apeiron> thank you for nudging me[22:25:05] <apeiron> thebug, ^ feel free to try as well[22:25:51] <thebug> hmm. I'm not running kayak but I can probably import it and run it in a zone[22:26:04] <thebug> (have not sat down and built install media before)[23:03:23] *** wuff has joined #omnios[23:06:21] <wuff> i just had what I think was a kernel panic.. and I had to reboot quickly because it's a production system (dump was at 12% when I remote connected via IPMI and taking a looong time for 1% - host has 256GB).. are there any logs I can check? As expected, nothing in /var/crash since I rebooted[23:13:26] *** nhubbard has quit IRC[23:13:51] <ashley_w> wuff: see if something is at /dev/zvol/dsk/rpool/dump[23:16:52] <iAmTheDave> hey all - AWS just announced SSD-backed EC2 instances - because of the nature of the OmniOS AMI, is this something I can't use? http://aws.typepad.com/aws/2013/11/a-generation-of-ec2-instances-for-compute-intensive-workloads.html[23:16:58] <ashley_w> or maybe dump@crash[23:17:27] <danlarkin> iAmTheDave: to what "nature" are you referring?[23:17:56] *** rjwill has joined #omnios[23:18:07] <iAmTheDave> danlarkin: in that you have to use EBS (not instance store) instances, so they can be ZFS formatted[23:18:18] <rjwill> hello all. strange problem here[23:18:31] <danlarkin> iAmTheDave: ah gotcha[23:19:40] <rjwill> just installed -z. booted into omni post-install. configured network, etc[23:19:49] <wuff> ashley_w: hmm, there is something called "dump" in /dev/zvol/dsk/rpool.. it's a symbolic link tho[23:19:51] <wuff> lrwxrwxrwx 1 root root 0 Nov 14 16:28 dump -> ../../../..//devices/pseudo/zfs@0:1[23:19:59] <rjwill> rebooted. after grub menu, wont boot[23:20:32] <rjwill> after localing '/platform..../boot_archive' ...[23:20:35] <rjwill> "compression algorithm not supported[23:20:50] <rjwill> Error 16: Inconsistent filesystem structure[23:20:57] <rjwill> press any key to continue[23:21:56] <rjwill> this machine previously had -p on it but i dont expect that would have caused anything[23:22:07] <ashley_w> wuff: i've yet to debug my panics ever. i have 3 dumps sitting around for me to look at. just giving you an idea from the screenshots i took at one crash[23:22:15] <rjwill> especially since it was able to boot post install without issues[23:22:27] <apeiron> 151006z?[23:22:29] <rjwill> yes[23:22:42] <rjwill> the latest (stable) from the website[23:23:13] <apeiron> I'm looking on the server for the installation media and there is no r151006z[23:23:20] <bdha> (he means y)[23:23:26] <apeiron> maybe[23:23:29] <apeiron> I want to be sure :)[23:23:32] <rjwill> ah my bad[23:23:36] <rjwill> it's -y[23:23:47] <apeiron> how did you do the upgrade?[23:23:50] <rjwill> OmniOS_Text_r151006y.usb-dd[23:23:54] <rjwill> i didnt upgrade[23:23:56] <wuff> i couldn't wait for the dump to finish.. and i'm pretty sure it won't fit on the boot SSD anyways (the full dump), as my boot drive only has ~60GB free with the host with 256GB RAM. I'm hesitant to just "try" what I was doing.. I had just completed configuring SRP for my vmware infrastructure, and was just spinning up a new VM[23:23:57] <rjwill> just ran a new install[23:24:08] <wuff> SRP has been running stable with the Hyper-V hosts connected to it[23:24:09] <rjwill> didnt need anything on the drive[23:24:16] <apeiron> hm[23:24:19] <apeiron> I can try reproducing[23:24:30] <rjwill> just booted -y from usb, ran through the install[23:24:34] <rjwill> all went well[23:24:48] <rjwill> rebooted, set up network, etc[23:25:04] <apeiron> and rebooted again and dead?[23:25:08] <rjwill> yes[23:25:10] <apeiron> odd[23:25:16] <apeiron> rjwill, you working with bdha?[23:25:17] <ashley_w> wuff: here's screenshots from my crash: https://plus.google.com/110871972189237957931/posts/BwcyxqgC7t9[23:25:19] <rjwill> yes[23:25:20] <bdha> apeiron: haha[23:25:26] <bdha> It's not my fault. I didn't even touch this machine.[23:25:29] <apeiron> I know[23:25:30] <rjwill> nope[23:25:32] <apeiron> but you seem to be cursed[23:25:55] <bdha> Maybe.[23:26:04] <apeiron> I'm pulling that media down presently and will try to reproduce[23:26:06] <rjwill> also "reboot" seems to hang after "rebooting..." prompt[23:26:12] <bdha> I have dozens of SmartOS boxes that run on this same model of hardware. ;-)[23:26:23] <bdha> (Though I do find cool bugs in SmartOS sometimes.)[23:26:30] <apeiron> you find cool bugs everywhere.[23:26:33] <bdha> "cool"[23:26:49] <ashley_w> wuff: my /var/crash is only 4.4G with 3 dumps, and i have 32GB of ram[23:26:56] <apeiron> Just for giggles, did you power cycle it again?[23:26:59] <rjwill> no[23:27:06] <rjwill> oh wait yes[23:27:20] <rjwill> i pulled all the other drives and rebooted[23:27:28] <apeiron> "all the other drives"[23:27:28] <rjwill> leaving only the isntall disk[23:27:32] <apeiron> you didn't set up mirroring or anything?[23:27:35] <rjwill> not yet[23:27:48] <rjwill> not actually using it on this box[23:27:49] <apeiron> hmmm.[23:27:51] <bdha> apeiron: Algo not supported sounds like a desync between the kernel and ZFS version.. only thing I can think of is somehow the zpool has lz4 enabled but the kernel doesn't have it? Which seems wrong.[23:27:56] <apeiron> yeah.[23:28:21] <apeiron> that sounds pretty reasonable[23:28:48] <bdha> But weird. Because y should have lz4 support regardless.[23:28:59] <apeiron> I'd expect that to come up on the initial boot then[23:29:04] <bdha> And the machine _was_ running sol10, but it got reinstalled anyway.[23:29:08] <apeiron> unless you enabled compression and then rebooted?[23:29:18] <bdha> apeiron: Wait, what?[23:29:24] <apeiron> hmm?[23:29:28] <bdha> We do enable compression post-install.[23:29:30] <rjwill> yes i did[23:29:39] <apeiron> welp[23:29:44] <bdha> But.. so?[23:29:45] <apeiron> guess I know what I'm doing to try to reproduce it[23:29:49] <bdha> Heh![23:30:03] <bdha> Ok, I was worried for a second that compression on rpool wasn't supported or something, suddenly.[23:30:17] <apeiron> er[23:30:17] <apeiron> heh[23:30:22] <rjwill> gzip-9[23:30:34] <apeiron> I'm using lz4 on a stable VPS[23:30:43] <apeiron> gzip-9 should definitely work[23:31:58] <rjwill> can i boot off usb and change it to something else? or can you not switch[23:32:23] <bdha> Even if you disable compression, there are compressed blocks on disk it will need to read...which may or may not trigger a bug, if that's the bug.[23:32:25] <wuff> ashley_w: i will have to remember to take screenshots next time.. going to try my "experiments" after hours so less impact to my users. Still, i think that dump would have taken 30 minutes to write by the progress it was making[23:32:40] <bdha> wuff: Dumps are slow. But super valuable.[23:32:52] <bdha> But yeah, during biz hours, not great to deal with.[23:33:23] <apeiron> we've had to abort them on our larger machines because they were going to take forever[23:33:49] <bdha> All my newer machines have 192 or 256GB RAM. Once we start getting our massive corporate discount at $dayjob, it will be .. more.[23:33:58] * apeiron nods[23:34:56] <wuff> yeah my SAN box has 256GB.. all I was doing was installing Server 2012R2 on a fresh VM connected to the new datastore (ESXi 5.5 SRP via infiniband)[23:35:32] <rjwill> assuming this is something with gzip-9 comp, is there a way to side load that into the kernel and rescue this system? or am i looking at a reinstall[23:35:55] <apeiron> hmm, not sure I have a thumb drive around[23:35:55] <apeiron> meh[23:36:06] *** desai has quit IRC[23:36:10] <apeiron> the ISO should be the same bits[23:36:46] <apeiron> rjwill, what did you do specifically on that first reboot? set compression to gzip-9 on rpool and children?[23:37:05] <rjwill> zfs set compression=gzip-9 rpool[23:37:09] <apeiron> ok[23:37:13] <wuff> My hyper-v hosts have been stable until now (70+ days in continuous operation) using SRP as well.. all I did was add another WWN to the target group and created its own host group and view. Does anyone here run esx with srp/ib?[23:37:26] <bdha> I wonder if this is also the bug from beadm activate.[23:37:41] <apeiron> the last time I used anything VMWare for work the console wanted me to use IE on Windows[23:37:46] <apeiron> I have since stayed the hell away[23:38:10] <bdha> rjwill: On gp0, compression=on but not gzip-9.[23:38:25] <bdha> rjwill: Not sure when that got enabled, though, post test reboot or not.[23:38:43] <wuff> lol @apeiron[23:38:52] <apeiron> I can reboot with just doing zfs set compression=gzip-9 rpool[23:38:56] <apeiron> unsurprisingly[23:38:59] <apeiron> I suspect your bug is elsewhere[23:39:22] <bdha> Oh well.[23:39:38] <apeiron> I know it's horrible, but... try a reinstall?[23:39:47] <bdha> huh.[23:39:54] <bdha> "The gzip compression algorithm is not supported on root pools"[23:39:56] <bdha> From an Oracle doc.[23:40:03] * apeiron browfurrow[23:40:11] <bdha> Old. Talking about LU.[23:40:12] <apeiron> hmm[23:40:22] <apeiron> rjwill, you didn't do anything that would have written a new boot archive, would you?[23:40:24] <rjwill> pretty sure we have other machines using it[23:40:27] <apeiron> or a new BE[23:40:28] *** sebasp is now known as sebasp_[23:40:34] <bdha> rjwill: ha, yeah, but can they reboot. ;-)[23:41:17] <rjwill> i just ran "reboot"[23:41:25] <rjwill> i didnt run any pkg updates, etc[23:41:29] <apeiron> I mean package installations ... ah[23:41:29] <apeiron> hmm[23:42:48] <bdha> rjwill: Reinstall, no compression -> reboot, compression=on -> reboot, compression=gzip-9 -> reboot?[23:43:06] <apeiron> I know we have machines with compression=on for rpool. that sets it to lzjb according to zfs(1M)[23:43:12] <bdha> apeiron: Yup.[23:43:28] <apeiron> you might've set those up actually ;)[23:43:34] <bdha> I doubt it.[23:44:07] <rjwill> k[23:44:09] <bdha> I really hope it's not gzip/rpool... we have some systems that would be pretty annoying to reinstall.[23:44:26] *** jeffrymolanus_ has joined #omnios[23:45:18] <apeiron> esproul, ^ any wisdom here?[23:45:34] * esproul reads backscroll[23:45:46] <apeiron> essentially compression=gzip-9 on rpool == kaboom[23:45:49] <apeiron> so we think[23:45:55] <apeiron> Oracle docs say you can't do gzip onrpool[23:46:16] <esproul> ^ was my question[23:46:26] <esproul> grub needs to be able to read compressed pools[23:46:33] *** jeffrymolanus has quit IRC[23:46:34] *** jeffrymolanus_ is now known as jeffrymolanus[23:46:35] <esproul> probably doesn't support the full suite of algos[23:46:47] <apeiron> yeah, which is why I asked above if there was anything that rjwill did that would've written a boot archive or BE[23:47:24] <bdha> sol11 docs say the same thing.[23:47:25] <bdha> welp![23:47:28] <esproul> don't play with fire[23:47:34] <bdha> This might explain all sorts of bugs lately.[23:47:44] <apeiron> Could be, yeah.[23:48:16] <esproul> IIRC, Saso had to add lz4 support into grub when adding that feature[23:48:33] *** wez is now known as wez|away[23:48:48] <bdha> At this point in my career, I don't even think I have feet left to shoot off.[23:49:22] *** wez|away has quit IRC[23:49:35] <rjwill> so just go with compression=on and be done with it[23:49:49] <apeiron> yeah, sounds like[23:50:34] <bdha> I need to deal with babies.[23:50:37] <bdha> apeiron: Thanks for the help again.[23:51:04] <apeiron> no prob[23:52:30] <rjwill> yep thanks[23:52:34] <rjwill> =/[23:58:00] *** ira has quit IRC[23:59:33] <iAmTheDave> is there a decent utility to test disk read/write speed?[23:59:59] <apeiron> dd ?