[00:14:24] *** sadrul has quit IRC
[00:18:11] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "x86 generic full" from 2a451e65052c31468ca76920a9184f2b499373fe: _third_party_ at chromium dot org)'
[00:21:33] *** quannnum has quit IRC
[00:23:04] <crosbot> tree became 'Tree is open (upload timeout cycling:petermayo watching ...)(Known issues: crosbug.com/17899 - flaky chrome seg11 (x86-alex-binary, x86 generic [pfq]); crosbug.com/17866 - flakey testUpdateWipeStateful (x86-zgb-binary).'
[00:32:28] *** quannnum has joined #chromium-os
[01:01:21] *** JasonO has joined #chromium-os
[01:02:29] *** Soak has quit IRC
[01:09:32] *** |simo| has joined #chromium-os
[01:09:56] *** |simo| is now known as Guest54474
[01:11:51] *** JasonO has quit IRC
[01:22:47] *** sadrul has joined #chromium-os
[01:22:47] *** ChanServ sets mode: +v sadrul
[01:58:03] *** petermayo has joined #chromium-os
[01:58:04] *** ChanServ sets mode: +v petermayo
[01:59:56] *** Adys has quit IRC
[02:04:37] *** lipsinV2 has joined #chromium-os
[02:13:05] *** Adys has joined #chromium-os
[02:17:02] *** D|sT has quit IRC
[02:38:18] *** D|sT has joined #chromium-os
[02:52:56] *** D|sT has quit IRC
[03:21:30] *** falken has joined #chromium-os
[03:27:38] *** D|sT has joined #chromium-os
[03:57:51] *** Adys has quit IRC
[04:09:16] *** kinaba has quit IRC
[04:10:59] *** Adys has joined #chromium-os
[04:12:22] *** kinaba has joined #chromium-os
[04:49:43] <crosbot> tree became 'Tree is open (Known issues: crosbug.com/17899 - flaky chrome seg11 (x86-alex-binary, x86 generic [pfq]); crosbug.com/17866 - flakey testUpdateWipeStateful (x86-zgb-binary), crosbug.com/18465 - Upload prebuilt timeoout).'
[04:50:29] <crosbot> tree became 'Tree is open (Known issues: crosbug.com/17899 - flaky chrome seg11 (x86-alex-binary, x86 generic [pfq]); crosbug.com/17866 - flakey testUpdateWipeStateful (x86-zgb-binary), crosbug.com/18465 - Upload prebuilt timeout).'
[05:18:44] *** oc80z has quit IRC
[05:18:44] *** oc80z has joined #chromium-os
[05:45:14] <petermayo> Good night, crosbot.
[05:45:21] *** petermayo has quit IRC
[06:36:55] *** kochi has quit IRC
[06:50:51] *** Inumedia has quit IRC
[06:52:23] *** Inumedia has joined #chromium-os
[07:02:00] *** Inumedia has quit IRC
[07:02:42] *** Inumedia has joined #chromium-os
[07:35:47] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot_master" on "x86 generic pre flight queue" from 12f16c457c91791135e54caa7b31385b34254d5d: dominicc at chromium dot org <dominicc at chromium dot org@268f45cc-cd09-0410-ab3c-d52691b4dbfc>)'
[07:37:56] *** MostAwesomeDude has left #chromium-os
[07:56:00] *** kochi has joined #chromium-os
[08:07:16] *** ferringb has quit IRC
[08:30:35] *** SoulRaven has joined #chromium-os
[08:35:20] <crosbot> tree became 'Tree is open ("x86 generic pre flight queue" goes green)'
[08:44:00] *** vmil86 has joined #chromium-os
[08:47:51] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot_master" on "TOT Pre-Flight Queue" from afca327114bba429ad40ad0a3d495da5e73ac321: _third_party_ at chromium dot org)'
[09:26:13] *** Sergiu has joined #chromium-os
[09:31:20] <crosbot> tree became 'Tree is open (Known issue: crosbug.com/17899 - flaky chrome seg11 on "TOT Pre-Flight Queue")'
[09:33:54] *** Sergiu has quit IRC
[09:43:38] *** jochen__ has quit IRC
[09:43:43] *** jochen__ has joined #chromium-os
[09:43:43] *** ChanServ sets mode: +v jochen__
[10:00:10] *** Styx__ has joined #chromium-os
[10:01:51] *** patcito has quit IRC
[10:04:11] *** Styx__ is now known as Styx
[10:14:45] *** zmedico has quit IRC
[10:36:14] *** zmedico has joined #chromium-os
[10:54:37] <Jahei> any vanilla builds with broadcom drivers ?
[10:55:20] *** lipsinV2 has quit IRC
[11:39:30] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot_master" on "x86 generic pre flight queue" from f3a99b8f32982cb5feb516f9a5038dce00606b14: _third_party_ at chromium dot org, pfeldman at chromium dot org <pfeldman at chromium dot org@268f45cc-cd09-0410-ab3c-d52691b4dbfc>, tonyg at chromium dot org <tonyg at chromium dot org@268f45cc-cd09-0410-ab3c-d52691b4dbfc>)'
[13:26:02] *** superm1 has quit IRC
[13:29:05] *** jochen___ has joined #chromium-os
[13:29:06] *** ChanServ sets mode: +v jochen___
[13:30:37] *** jochen__ has quit IRC
[13:30:38] *** jochen___ is now known as jochen__
[13:41:57] *** superm1 has joined #chromium-os
[13:41:59] *** superm1 has joined #chromium-os
[14:13:00] *** SoulRaven has quit IRC
[14:37:33] *** McMAGIC--Copy has quit IRC
[14:42:43] *** McMAGIC--Copy has joined #chromium-os
[14:49:19] *** zmedico has quit IRC
[14:51:02] *** SoulRaven has joined #chromium-os
[14:56:08] *** zmedico has joined #chromium-os
[15:15:41] *** phil2 has joined #chromium-os
[15:24:32] <njw> okay, this is a very bouncy tree. Let's see if the 05:37 Easetrn closing is the same old thing.
[15:28:31] <crosbot> tree became 'Tree is open (Known issues: crosbug.com/17899 - flaky chrome seg11 x86 geeneric pfq)'
[15:29:13] <njw> yup, same old same old, and a bunch of pfq runs passed since then. Opened.
[15:50:27] <ellyjones> sheriff on duty
[15:50:33] <njw> join the club.
[15:50:37] <ellyjones> excellent
[15:50:56] <ellyjones> the internal bots: what
[15:51:15] <ellyjones> remoteFailed: [Failure instance: Traceback: <type 'exceptions.IOError'>: [Errno 28] No space left on device
[15:52:02] <ellyjones> kliegs: you awake yet? can you look at the internal pfq bot?
[15:52:52] <ellyjones> actually, all the internal bots
[15:53:15] <ellyjones> I'm closing the tree
[15:53:49] <crosbot> tree became 'Tree is closed (ellyjones: most internal bots out of disk, need a trooper)'
[15:56:18] <ellyjones> who else are the troopers
[15:56:50] <ellyjones> djmm, scottz, aluri, bradnelson, maruel, kliegs
[15:57:27] * ellyjones mails them to look at the bots
[16:05:57] <ellyjones> ok
[16:05:59] <ellyjones> maruel is lookin
[16:06:03] <ellyjones> looking
[16:06:25] <ellyjones> I got mail from petermayo telling me the internal bots are out of space on sunday, so they must have been broken for at least 16 hours by now
[16:07:59] <crosbot> tree became 'Tree is closed (ellyjones: most internal bots out of disk, maruel is looking)'
[16:08:10] *** maruel has joined #chromium-os
[16:08:10] *** ChanServ sets mode: +v maruel
[16:08:18] <ellyjones> ok
[16:08:25] <maruel> hi, having issues to access the slaves, still looking
[16:10:27] <ellyjones> ok
[16:33:29] *** vpalatin has quit IRC
[16:42:35] <ellyjones> ok
[16:42:37] <ellyjones> maruel: any luck?
[16:42:43] <maruel> no
[16:42:54] <ellyjones> alright, what next?
[16:43:04] <ellyjones> is there someone to contact for access?
[16:43:09] <maruel> I'm pinging people
[16:43:15] <ellyjones> danke
[16:45:51] *** vpalatin has joined #chromium-os
[16:45:52] *** ChanServ sets mode: +v vpalatin
[16:46:44] *** ScriptRipper has quit IRC
[17:00:35] <ellyjones> hey, the internal bots are green again
[17:00:41] <ellyjones> or at least some are
[17:01:32] *** glotov has quit IRC
[17:01:46] <ellyjones> the pfq bot was definitely not green before
[17:03:23] *** jrbarnette has joined #chromium-os
[17:20:50] <njw> back to purple. All is consistent with the world, if not exactly well.
[17:21:00] <ellyjones> consistent is good
[17:24:05] *** petermayo has joined #chromium-os
[17:24:05] *** ChanServ sets mode: +v petermayo
[17:45:23] *** grundler has joined #chromium-os
[17:45:24] *** ChanServ sets mode: +v grundler
[17:46:45] *** borios has joined #chromium-os
[17:50:33] *** Styx has quit IRC
[17:53:40] *** jrbarnette has quit IRC
[18:04:42] *** saintlou has joined #chromium-os
[18:09:49] *** jrbarnette has joined #chromium-os
[18:10:18] *** jrbarnette_ has joined #chromium-os
[18:10:55] *** cjb has joined #chromium-os
[18:11:17] *** JasonO has joined #chromium-os
[18:11:18] *** JasonO has joined #chromium-os
[18:14:29] *** jrbarnette has quit IRC
[18:14:30] *** jrbarnette_ is now known as jrbarnette
[18:22:02] <stevenjb> ellyjones: good morning. You need me to look into anything? Looks like x86 pineview is the sig 11 bug. What about arm tegra2 seaboard?
[18:26:09] <njw> I think what we need is troopers to look at the internal bots.
[18:26:48] <stevenjb> Is maruel not looking into that?
[18:28:36] <njw> last I heard is his "having issues accessing the slaves, pinging people" from a couple of hours ago.
[18:30:54] *** kerz has joined #chromium-os
[18:30:54] *** ChanServ sets mode: +v kerz
[18:33:13] <cjb> Hi, I've got an unaccelerated chromiumos build running -- is there a standard way to switch chromium over to "--disable-accelerated-2d-canvas  --disable-accelerated-compositing" on it?  (Or perhaps that's the default for the software-only build..?)
[18:36:48] <stevenjb> cjb: The flags passed to chromium are set in /sbin/session_manager_setup.sh
[18:37:49] <cjb> Thanks!
[18:41:28] *** xiyuan has quit IRC
[18:42:35] <petermayo> the internal master had no jobs running, no space to complete anyhting useful, so I've stopped it until someone can free something up.
[18:46:55] *** sergiu has joined #chromium-os
[18:47:00] <stevenjb>  djmm, scottz, aluri, bradnelson, maruel, kliegs - any troopers have a status update on the internal buildbots?
[18:47:03] *** sergiu has quit IRC
[18:47:03] *** sergiu has joined #chromium-os
[18:54:14] *** bpontes has joined #chromium-os
[18:54:15] *** ChanServ sets mode: +v bpontes
[18:54:59] <crosbot> tree became 'Tree is closed (ellyjones: most internal bots out of disk, need a trooper with access)'
[18:58:28] <ellyjones> petermayo: did you kill the botmaster?
[18:59:48] *** tbarzic has joined #chromium-os
[19:00:06] *** micahc has quit IRC
[19:00:11] *** micahc has joined #chromium-os
[19:00:11] *** ChanServ sets mode: +v micahc
[19:03:41] *** cmasone has quit IRC
[19:03:45] *** cmasone_ has joined #chromium-os
[19:03:45] *** ChanServ sets mode: +v cmasone_
[19:08:14] <ellyjones> petermayo: can you start the master up again so that we can see what slaves it has?
[19:08:52] *** saintlou has quit IRC
[19:11:00] <ellyjones> alright
[19:11:16] <ellyjones> the two troopers I can reach both can't log in, and apparently the valid set is {scottz,djmm}
[19:11:20] <ellyjones> so we wait
[19:11:59] <sosa> the master has a diff password
[19:12:08] <sosa> can't peter log in and clean it up
[19:12:14] <sosa> he clearly had access to shutdown down the bot?
[19:12:23] <ellyjones> I don't know, he's not responding
[19:12:31] <sosa> i'll take a look
[19:12:31] *** wfrichar has joined #chromium-os
[19:12:32] *** ChanServ sets mode: +v wfrichar
[19:12:34] <ellyjones> thanks
[19:17:00] <davidjames> ellyjones: chromeosbuild01-12 have lots of disk space, which bot was it you were talking about that was out of disk space?
[19:20:00] <crosbot> tree became 'Tree is closed (internal master out of disk)'
[19:20:07] <davidjames> petermayo: Could you start the master back up? We can't see the waterfall if the master is down :P
[19:20:23] <cmasone_> davidjames: peter doesn't seem to be arounfd
[19:20:58] <ellyjones> davidjames: I do not remember
[19:21:06] <ellyjones> and I do not have the waterfall open
[19:21:49] <ellyjones> try 13, though?
[19:21:54] <ellyjones> (not that it helps with the master down :P)
[19:21:58] <davidjames> ellyjones: No, that bot is fine
[19:21:58] *** lipsinV2 has joined #chromium-os
[19:22:02] *** lipsinV2 has joined #chromium-os
[19:22:06] * ellyjones does not know, then
[19:22:08] <davidjames> ellyjones: Just the master is having issues. I'm logged on to the master now
[19:22:28] <ellyjones> can you start it back up?
[19:22:33] <crosbot> tree became 'Tree is closed (chromeos-botmaster.mtv out of space)'
[19:22:42] <ellyjones> also, davidjames, how does one get credentials to log into the botmaster?
[19:22:58] <davidjames> ellyjones: Ask bevc@
[19:23:09] <ellyjones> aha, djmm exists
[19:23:41] <davidjames> ellyjones: Trying to figure out what's using the space now. And also how to restart the botmaster :)
[19:23:50] *** ScriptRipper has joined #chromium-os
[19:24:03] *** ScriptRipper has joined #chromium-os
[19:24:35] <ellyjones> cool
[19:24:39] <ellyjones> good luck and thanks
[19:24:49] <ellyjones> I'd like for the tree to be open at some point during my sheriff shift :)
[19:25:36] *** borios has quit IRC
[19:25:57] *** borios has joined #chromium-os
[19:27:30] *** McMAGIC--Copy has quit IRC
[19:31:58] *** vmil86 has quit IRC
[19:36:14] <maruel> ellyjones: I and petermayo took a look at the master's www but we don't know what can be deleted safely, nor TPMs.
[19:36:21] <maruel> I can't make that call
[19:36:48] <maruel> well, I could rm -rf * but I don't think anyone would appreciate that, even deleting older directories isn't necessarily a good idea
[19:37:25] *** BladeFreak has joined #chromium-os
[19:37:37] <ellyjones> :P
[19:38:00] <ellyjones> can we just buy a couple more disks off newegg? ;)
[19:38:09] <sosa> yeah brad and I on the other side are also looking at it
[19:41:07] <sosa> alright djmm is on the bot
[19:41:08] <sosa> and cleaning
[19:41:19] <sosa> let's leave him to have ownership on the rm -rf's
[19:41:50] <crosbot> tree became 'Tree is closed (chromeos-botmaster.mtv has too many images and is out of space, cleaning -> djmm)'
[19:41:51] <sosa> master is back up
[19:42:38] <ellyjones> rock
[19:42:43] *** McMAGIC--Copy has joined #chromium-os
[19:42:49] <ellyjones> and now all the bots show as green
[19:43:12] <ellyjones> x86-mario-binary-r13 I don't think we care for
[19:43:12] <ttuttle|work> nice
[19:44:05] <ellyjones> if djmm's build works we can open the tree :)
[19:44:55] <ellyjones> sosa: http://build.chromium.org/p/chromiumos/builders/x86%20generic%20full/builds/3060/steps/cbuildbot/logs/stdio <--- ipv6 problems?
[19:46:28] <sosa> ellyjones:  i'm not sure, either that or just flakiness again.  I thought we were supposed to have mirrored everything we depended on anyway
[19:46:59] *** vmil86 has joined #chromium-os
[19:48:35] <ellyjones> sosa: I don't suppose you have the right acls to look at that bot and get the build.log for icedtea?
[19:49:40] *** rhelmer has quit IRC
[19:54:47] *** lipsinV2 has quit IRC
[19:55:38] *** jrbarnette_ has joined #chromium-os
[19:59:04] *** jrbarnette has quit IRC
[20:00:34] *** jrbarnette_ has quit IRC
[20:01:23] *** JasonO has quit IRC
[20:01:58] *** JasonO has joined #chromium-os
[20:02:02] *** JasonO has joined #chromium-os
[20:08:14] *** saintlou has joined #chromium-os
[20:08:14] *** ChanServ sets mode: +v saintlou
[20:12:04] <crosbot> tree became 'Tree is closed (chromeos-botmaster.mtv back up, still low on disk -> djmm)'
[20:13:26] <davidjames> ellyjones: FYI chromeos-botmaster has 16G of space now, which is still tight given that a bunch of release builds will push images there soon. djmm is working on it
[20:13:46] <ellyjones> okay, cool
[20:13:56] <ellyjones> any kind of eta before we can open the tree? :)
[20:14:16] <ellyjones> djmm says over IM that we can open the tree now
[20:14:21] <davidjames> ellyjones: Sweet
[20:14:22] *** jrbarnette has joined #chromium-os
[20:14:22] *** ChanServ sets mode: +v jrbarnette
[20:14:52] <crosbot> tree became 'Tree is open (chromeos-botmaster >99% disk use, djmm is cleaning it up)'
[20:15:06] <ellyjones> go go gadget internal pfq
[20:15:29] <stevenjb> ellyjones: I'm looking at x86-mario-canary, not sure if it's a new or old/flakey issue.
[20:18:28] <stevenjb> I suspect that it is a variant of the sig 11 bug, but it's not clear.
[20:22:41] <gauravsh> we must be getting the chrome crash stack traces now, right?
[20:22:49] <gauravsh> cmasone: ^^^^^^^
[20:23:12] <stevenjb> gauravsh: I believe that we should be, I was just about to go searching for them.
[20:23:24] <stevenjb> (cmasone does not appear to be online)
[20:26:06] *** thieule has joined #chromium-os
[20:26:06] *** ChanServ sets mode: +v thieule
[20:27:22] <cmasone_> gauravsh: we're supposed to bew
[20:27:26] <cmasone_> but
[20:27:38] <cmasone_> I noticed over the weekend that something is still breaking it.
[20:27:43] <petermayo> ellyjones: sorry to act and duck ... it was purely a master issue that it was making worse by running.  Escalation and clarification will continue today and this week.
[20:28:04] <cmasone_> stevenjb: there's a case or two where we log out and don't copy the crashes.  That is on me
[20:28:10] <ellyjones> petermayo: djmm cleared some space on the master
[20:28:17] <cmasone_> stevenjb: but, there's something else that causing them to not get symbolized and stuff
[20:28:25] <ellyjones> we're green again, although if we're still green after the internal pfq catches up I'll be surprised
[20:28:52] <stevenjb> cmasone: OK. Let me know when I can look for symbol files and where to look for them.
[20:29:11] <petermayo> ellyjones:  yes, my first stop back up the stairs was scottz desk for personal updates/debriefing.
[20:29:11] *** cmasone_ is now known as cmasone
[20:29:20] <ellyjones> ok, cool
[20:29:28] <cmasone> stevenjb: ffs...why is this on me, now?
[20:30:06] <stevenjb> cmasone: I thought you were working on it. Please direct me to the right person instead.
[20:36:25] *** patcito has joined #chromium-os
[20:37:44] <stevenjb> cmasome: "Comment 11 by project member cmasone at chromium dot org, Jul 21, 2011
[20:37:44] <stevenjb> This particular random crash seems to have cleared.  Going to spend some time today working on a tool to help sheriffs get the minimum data that someone would need to investigate a problem such as this, and document the process of gathering that info and assigning an issue to the right person."
[20:38:18] <stevenjb> cmasone: Did you make any progress on that?
[20:55:25] <crosbot> tree became 'Tree is open (chromeos-botmaster >99% disk use, djmm is cleaning it up)'
[20:58:54] *** jshin has quit IRC
[21:01:30] <ellyjones> ok
[21:01:36] <ellyjones> I have a set of four interdependent CLs
[21:02:26] *** zmedico has quit IRC
[21:02:44] *** zmedico has joined #chromium-os
[21:02:51] <ellyjones> sosa: you around? I suspect pieces will need to be picked up :P
[21:02:53] <crosbot> tree became 'Tree is throttled (chromeos-botmaster >99% disk use, djmm is cleaning it up; ellyjones is landing a set of dependent changes)'
[21:03:35] <cjb> hm, chromium's crashing (Aw, Snap!) after rendering any webpage (except chrome://system) here -- /var/log/chrome/* doesn't contain anything suspicious though.  any idea where I should look next?
[21:03:55] <cjb> I do see "WARNING:zygote_main_linux.cc(127): Sending zygote magic failed"
[21:04:15] <sosa> ellyjones: hi
[21:04:47] <ellyjones> alright
[21:04:59] <ellyjones> is there any incantation to do for landing dependent CLs, or do I just push and kick the bots?
[21:05:09] <sosa> ellyjones: no not yet
[21:05:19] <ellyjones> alright then
[21:05:26] <sosa> ellyjones:  just take a look at the builders and try to get them all in one go
[21:05:37] <sosa> ellyjones: be around in case you make a mistake, that's about it
[21:05:49] <sosa> ellyjones:  usually if you commit them quick with an eye towards the tree, all goes well
[21:06:05] <sosa> ellyjones:  since you're a sheriff at least 1/2 of that shouldn't be a problem
[21:06:07] <ellyjones> ok, I'm starting comitting them
[21:06:34] <sosa> ellyjones:  the x86-gen-pfq is the main thing that matters for cros_workon packages
[21:06:47] <ellyjones> yep
[21:06:49] <sosa> ellyjones:  since it decides what to build
[21:07:35] <ellyjones> all four are in
[21:08:03] <ellyjones> and it did not kick off an x86-gen-pfq build, as far as I can tell
[21:08:13] <ellyjones> I hope it picks up all four of my changes
[21:08:52] <davidjames> ellyjones: All four of your changes are cros-workon, I hope? :)
[21:09:29] <ellyjones> actually, hrm
[21:09:43] <ellyjones> yeah, they are
[21:09:49] <ellyjones> installer, initramfs, verity, kernel
[21:09:58] <davidjames> ellyjones: Congrats! Should be easy then :)
[21:10:03] <ellyjones> I hope so!
[21:15:04] <ellyjones> aw shoot
[21:15:12] <ellyjones> x86-generic-pfq build 6779 started with two of my four present
[21:15:17] <dubroy> adlr: ping
[21:16:08] <ellyjones> sosa: where's the UI for kicking a build on a public buildbot? I am not seeing it
[21:16:25] <sosa> it's there if you're using the internal links
[21:16:27] <sosa> chromegw
[21:17:08] <sosa> http://chromegw/i/chromiumos/waterfall
[21:17:49] <ellyjones> sosa: is there a reason http://chromegw/i/chromiumos/builders/x86%20generic%20pre%20flight%20queue/ shows two current builds?
[21:19:22] <sosa> I only see one
[21:19:28] <sosa> but i have seen it before
[21:19:41] <sosa> i believe it's a side affect of backgrounding some of the work
[21:19:54] <ellyjones> do you see '#6634 ETA 0s (12:17:23) [Archive]'?
[21:19:59] <ellyjones> that build is from ages ago
[21:20:15] <ellyjones> Manifest Revision: Nothing to build!
[21:20:16] <ellyjones> argh.
[21:21:34] <sosa> yeah you didn't actually need to stop it, it picked up all your changes originally
[21:21:51] <ellyjones> oh, oops
[21:21:56] <ellyjones> it looked like it had missed two
[21:22:12] <ellyjones> but 6781 appears to have built and now be upreving? oO
[21:22:21] <ellyjones> oh, I see, it's parallelized
[21:25:54] *** grundler has quit IRC
[21:27:12] <ellyjones> sosa: if you look at x86-gen-pfq now you should see two active builds
[21:29:50] <sosa> yeah it seems to be just the master confused
[21:29:57] <sosa> it'll prob clean itself up when we restart it next
[21:31:08] <ellyjones> ok, cool
[21:32:33] *** borios_ has joined #chromium-os
[21:32:34] *** borios has quit IRC
[21:40:53] *** grundler has joined #chromium-os
[21:40:54] *** ChanServ sets mode: +v grundler
[21:56:57] *** vmil86 has quit IRC
[22:06:56] *** saintlou has quit IRC
[22:07:02] *** saintlou has joined #chromium-os
[22:07:06] *** ChanServ sets mode: +v saintlou
[22:07:07] *** saintlou has quit IRC
[22:09:19] *** vlaviano has joined #chromium-os
[22:09:19] *** ChanServ sets mode: +v vlaviano
[22:11:51] *** vmil86 has joined #chromium-os
[22:16:55] *** saintlou has joined #chromium-os
[22:17:05] *** ChanServ sets mode: +v saintlou
[22:20:24] *** ScriptRipper has quit IRC
[22:21:01] *** ScriptRipper has joined #chromium-os
[22:21:43] *** Ruetobas has quit IRC
[22:23:26] *** grundler has quit IRC
[22:26:02] *** Ruetobas has joined #chromium-os
[22:26:06] *** ScriptRipper has quit IRC
[22:26:06] *** ScriptRipper has joined #chromium-os
[22:26:44] <ellyjones> hrm
[22:26:48] <ellyjones> pfq failed
[22:26:52] <ellyjones> crosbot: stat tree
[22:26:52] <crosbot> ellyjones: tree: Tree is throttled (chromeos-botmaster >99% disk use, djmm is cleaning it up; ellyjones is landing a set of dependent changes)
[22:28:15] <ellyjones> the AU log shows that it fails to connect to the AU server
[22:28:21] <ellyjones> sosa: know anything about this?
[22:29:17] <sosa> no
[22:29:23] <ellyjones> hm
[22:29:24] <sosa> but i can interpret the log for you :)
[22:29:27] <ellyjones> woo!
[22:29:42] <sosa> It ran an update test, and upon reboot could no longer SSH into the test image
[22:30:01] <ellyjones> okay, that is bad
[22:30:10] <ellyjones> but the log for that update is absent
[22:30:27] <ellyjones> or at least, the log in test_results.tgz just shows us failing to connect to the AU server
[22:31:27] <sosa> which log are you looking at?
[22:31:47] <ellyjones> the ones at https://sandbox.google.com/storage/chromeos-image-archive/x86-generic-pre-flight-queue/0.15.837.0-a1-b6781/_index.html
[22:31:53] <stevenjb> ellyjones: Is this re: bin/cros_au_test_harness failure in x86 generic pre flight queue?
[22:32:22] <ellyjones> yeah
[22:32:22] <stevenjb> Also in x86 pineview binary
[22:32:36] <sosa> heh
[22:32:37] <sosa> no actually
[22:32:39] <sosa> i misread the log
[22:32:44] <sosa> i can't ssh into the test image at all
[22:32:48] <ellyjones> O_o
[22:32:57] <sosa> that's why the update.log is empty
[22:33:12] <ellyjones> well, that's bad
[22:33:17] <ellyjones> so the test image does not boot?
[22:33:29] <sosa> it may boot, but ssh isn't working
[22:33:36] <ellyjones> so that could be either me or jrbarnette
[22:33:50] <sosa> yes
[22:34:04] <jrbarnette> if it boots, it's not likely to be the build_image change
[22:34:20] <ellyjones> do we know that it booted?
[22:34:23] <jrbarnette> but i won't say "never", for fear of embarrassing myself
[22:34:36] <sosa> no not from those logs
[22:34:48] <ellyjones> is there a kvm instance we can ping or something?
[22:34:51] <sosa> i would prob just download the image and give it a while
[22:34:54] <sosa> whirl*
[22:34:55] <ellyjones> or a kernel log :)
[22:34:58] <ellyjones> ok
[22:35:20] <crosbot> tree became 'Tree is closed (ellyjones: x86-generic-pfq image isn't booting)'
[22:35:49] <ellyjones> er hm
[22:35:57] <ellyjones> I just realized I do not have an x86 machine
[22:36:05] <cmasone> ellyjones: VM?
[22:36:13] <ellyjones> yeah
[22:36:22] <jrbarnette> please, VM
[22:36:31] <ellyjones> alright, I will look at this soon hopefully
[22:36:31] <jrbarnette> if it's my fault, x86 ain't likely to show it
[22:36:38] <jrbarnette> ... i tested x86
[22:36:50] <ellyjones> conversely, it is more likely my fault, since I only tested on arm :)
[22:36:50] *** SoulRaven has quit IRC
[22:37:48] *** JasonO has quit IRC
[22:38:21] *** JasonO has joined #chromium-os
[22:38:21] *** JasonO has joined #chromium-os
[22:38:40] <ellyjones> hm wait
[22:38:48] <ellyjones> my change requires a host update
[22:38:51] <ellyjones> do those happen automatically?
[22:39:01] <ellyjones> without the host update it'd produce a non-bootable image
[22:40:58] <sosa> specify host update
[22:41:11] <ellyjones> verity re-merge on hostside
[22:41:27] <ellyjones> (in the 'sudo emerge verity' sense, as opposed to 'emerge-$board verity')
[22:42:09] <sosa> The build log does show it re-building verity on /
[22:43:26] <ellyjones> hrm
[22:43:32] <ellyjones> ok, I guess I will wait for the image then
[22:44:40] *** Puck` has joined #chromium-os
[22:44:42] <Puck`> hi everyone
[22:45:08] <Puck`> I'd have a noob question regarding Chromium OS. It is the same thing as Chrome OS, right? And it does run on AMD Athlon CPU?
[22:45:19] <Puck`> or, Does it run on AMD Athlon CPU?
[22:45:28] <cmasone> Puck`: this isn't really the place.  Please email chromium-os-discuss at chromium dot org
[22:45:36] <ellyjones> or ask in #chromium-os-users
[22:46:04] <Puck`> hmm, alright, sorry to disturb, thank you for the guidance (:
[22:47:08] <ellyjones> oh, shoot
[22:47:33] <ellyjones> sosa:  is that specifically the recovery test image, or is it just a generic test image?
[22:47:45] <sosa> generic test image
[22:47:50] <ellyjones> okay
[22:47:51] <cmasone> davidjames: is there a way to use portage tools to ask an ebuild what value it would set an environment variable to?
[22:48:17] <ellyjones> what is the incantation to boot a chromiumos_qemu_image.bin in qemu?
[22:48:31] <cmasone> ellyjones: cros_start_vm is the tool I use
[22:48:39] <ellyjones> thanks
[22:48:43] <cmasone> ellyjones: it has help to explain the options to give it a path to a given image and such
[22:49:00] <ellyjones> excellent
[22:50:37] <ellyjones> chromiumos_qemu_image.bin: *elephant noises*
[22:50:39] *** Puck` has left #chromium-os
[22:51:32] <jrbarnette> ellyjones: just finished a chat w/ sosa:  how about we revert my change, and watch to see if it helps things?
[22:51:48] <ellyjones> if you want to? I am skeptical but we can try that :)
[22:51:54] <ellyjones> reverting mine will be a bit of a hassle
[22:52:12] <jrbarnette> ellyjones: yes, that's why i'm suggesting revert my change for the test
[22:52:43] <ellyjones> alright :)
[22:53:04] <jrbarnette> ellyjones: that way, we have a clear path to certainty
[22:53:11] <ellyjones> sosa: why does config.txt in image.zip differe from boot.config?
[22:53:37] <jrbarnette> ellyjones: ... we should still try and debug it separately
[22:53:45] <ellyjones> yep, I am doing so
[22:53:58] * ellyjones waits for the image to download from mtv :)_
[22:54:46] <sosa> ellyjones: i don't know ... we just run build_image --test, mod_image_for_vm, go and image.zip wraps what gets put in the image dir into the archive
[22:54:50] <jrbarnette> http://gerrit.chromium.org/gerrit/#change,5091
[22:55:36] <ellyjones> jrbarnette: lgtm'd
[22:56:15] <jrbarnette> ellyjones: reverted
[22:56:26] <ellyjones> ok
[22:56:29] <ellyjones> let's watch the next run then
[22:56:45] <ellyjones> meanwhile I have 70% of the image
[22:59:37] *** borios has joined #chromium-os
[22:59:46] *** borios_ has quit IRC
[23:03:43] *** saintlou_ has joined #chromium-os
[23:03:43] *** ChanServ sets mode: +v saintlou_
[23:05:53] *** JasonO has quit IRC
[23:06:27] *** saintlou has quit IRC
[23:06:41] *** JasonO has joined #chromium-os
[23:10:24] <ellyjones> getting X to work from inside the chroot: ??
[23:11:48] <ellyjones> okay
[23:13:19] *** gspencer has joined #chromium-os
[23:13:19] *** ChanServ sets mode: +v gspencer
[23:14:21] <ellyjones> the last thing qemu said was 'booting from hard disk...', then nothing
[23:14:27] <ellyjones> I'm not sure how to debug that :)
[23:21:48] *** rush2end_ has joined #chromium-os
[23:22:13] *** nick761 has quit IRC
[23:23:03] *** rush2end has quit IRC
[23:23:19] *** rush2end_ is now known as rush2end
[23:23:21] *** nick761 has joined #chromium-os
[23:23:58] <redpig> ellyjones: maybe check partition 12
[23:24:07] <redpig> and see if the arguments for the legacy bootloaders are messed up
[23:24:24] <redpig> (though if you ran any tests in the vm it would've gone through that code - or on non-cros firmware)
[23:25:22] <ellyjones> I only ever tested on Kaen, which doesn't use legacy boot
[23:25:29] <ellyjones> so that is a good candidate for loss
[23:25:48] <redpig> doh
[23:26:14] <ellyjones> part_12 of chromiumos_qemu_image.bin is blank after unpack_partitions.sh... that seems _very_ wrong!
[23:26:33] <ellyjones> (as in almost all zeroes, with 32 bytes of ff)
[23:26:42] *** ScriptRipper has quit IRC
[23:26:59] *** ScriptRipper has joined #chromium-os
[23:27:18] <redpig> ellyjones: uh yeah :(
[23:34:00] <ellyjones> what would cause that? build_image bailing out early?
[23:37:56] <jrbarnette> build_image is responsible for creating a file system on part 12
[23:38:26] <jrbarnette> there have been changes around that area in build_image, so there could be an issue there
[23:38:44] <jrbarnette> *however*, the change happened last week, not today...
[23:39:18] *** borios has quit IRC
[23:39:23] <ellyjones> create_legacy_bootloader_templates.sh is our guy
[23:40:18] <ellyjones> but that doesn't seem failure-prone
[23:40:59] <jrbarnette> does that script monkey with part 12?
[23:41:10] *** ScriptRipper has quit IRC
[23:41:10] *** ScriptRipper has joined #chromium-os
[23:41:11] <gauravsh> ellyjones: can you give me an example of how the new kernel command line is supposed to look like? (with your verity change)
[23:41:48] <ellyjones> gauravsh: yeah - it'd be like: earlyprintk kern_guid=704b1060-3b2f-5c49-b677-2d681bb2f947 root=/dev/dm-0 rootwait ro dm_verity.error_behavior=3 dm_verity.max_bios=-1 dm_verity.dev_wait=1 dm="vroot none ro,0 1740800 verity payload=704b1060-3b2f-5c49-b677-2d681bb2f947+1 hashtree=704b1060-3b2f-5c49-b677-2d681bb2f947+1 hashstart=1740800 alg=sha1 root_hexdigest=c9f9eb6e96966195450846fb1a06cdf969f66382" panic=60 cros_recovery kern_b_hash=
[23:42:55] <redpig> panic=60?
[23:43:08] <ellyjones> that one shouldn't be in for-real commandlines :)
[23:43:12] <ellyjones> I just do that when testing sometimes
[23:43:28] <ellyjones> the changed part is the stuff inside dm=""
[23:44:32] *** saintlou_ has quit IRC
[23:44:38] <ellyjones> also the uuids are %U, usually
[23:44:43] <jrbarnette> ellyjones: create_legacy_bootloader_templates.sh doesn't operate on part_12
[23:45:09] <gauravsh> update_bootloader.sh then?
[23:45:21] <ellyjones> http://build.chromium.org/p/chromiumos/builders/x86%20generic%20pre%20flight%20queue/builds/6783/steps/cbuildbot/logs/stdio indicates that update_bootloader.sh ran
[23:45:35] <jrbarnette> gauravsh: more likely
[23:46:46] <davidjames> ellyjones: Time to revert?
[23:47:24] <jrbarnette> ellyjones: i've confirmed that update_bootloaders.sh mounts the ESP fs
[23:48:12] <jrbarnette> however, i'm not sure we can blame it as culprit...
[23:48:26] <ellyjones> yeah, jrbarnette's revert didn't fix the tree :(
[23:48:31] <cmasone> anyone know how to run unittests in the chromite directory?
[23:48:33] <ellyjones> time to revert my changes
[23:49:26] <ellyjones> huh
[23:49:29] <ellyjones> the test image has a valid part 12
[23:49:32] <ellyjones> just the qemu image doesn't
[23:49:50] <redpig> is there a conversion script being run to get the vm image?
[23:49:58] <ellyjones> yeah, image_to_vm, I think
[23:49:59] <cmasone> redpig: image_to_vm
[23:50:37] *** micahc has quit IRC
[23:51:09] <ellyjones> if image_to_vm ran, it didn't finish, since it logs at the end
[23:51:31] <davidjames> ellyjones: btw, just pinged cmp about the git hang on x86-pineview-bin, hopefully he'll login to the builder and see what's going on
[23:51:34] <redpig> it would've called cros_make_image_bootable too which calls the other
[23:51:42] <redpig> but not sure why it would differ from the test image
[23:52:02] <ellyjones> ok, well
[23:52:08] <ellyjones> I will revert and debug this without the tree closed :)
[23:52:55] <ellyjones> redpig: http://gerrit.chromium.org/gerrit/5097 http://gerrit.chromium.org/gerrit/5098 http://gerrit.chromium.org/gerrit/5099 http://gerrit.chromium.org/gerrit/5100 LGTY?
[23:53:55] <crosbot> tree became 'Tree is closed (ellyjones: x86-generic-pfq image isn't booting)'
[23:54:21] <redpig> if they are gerrit reverts, then yes :p I'll look now
[23:54:52] <redpig> will the host version of verity end up jacked though?
[23:54:57] <redpig> or was it never incremented?
[23:55:12] <ellyjones> hrm
[23:55:13] *** saintlou has joined #chromium-os
[23:55:13] *** ChanServ sets mode: +v saintlou
[23:55:25] <ellyjones> I really hope it doesn't bump host changes if the build fails?
[23:55:36] <ellyjones> if it does, we'll have to fix that explicitly
[23:55:43] <redpig> ellyjones: I wasn't sure since it needed to pull it in to properly boot anyway
[23:55:53] <ellyjones> yeah
[23:56:00] <ellyjones> I can see from the log it built with new verity
[23:56:00] <redpig> but yeah, in the past I rev'd the ebuild manually post revert but I hope that is no longer needed
[23:56:21] <ellyjones> alright
[23:56:27] <ellyjones> are you cool with those reverts?
[23:57:37] <redpig> yes and in general it's fine to TBR them, but I just +2'd them
[23:57:54] *** jshin has joined #chromium-os
[23:57:54] *** ChanServ sets mode: +v jshin
[23:58:02] <ellyjones> ok
[23:58:35] <ellyjones> all in
[23:59:38] <ellyjones> the tree should go green now
[23:59:51] <gauravsh> ellyjones: can you coordinate with me before landing the change(s) after your fixes - I'd like to land the signer changes at the same time





top