[00:00:03] <jamescook> ok
[00:00:09] <jamescook> Looking at x86 generic full
[00:01:39] <jamescook> cmasone, I'll stop arm generic PFQ
[00:04:12] *** FatDarrel has quit IRC
[00:04:59] <mononoke> ellyjones: thanks, I am trying to figure it out now.
[00:06:18] <ellyjones> mononoke: /w 2
[00:06:21] <ellyjones> er, darn it
[00:06:27] *** cooled_ has quit IRC
[00:07:33] <gauravsh> ellyjones/mononoke: http://chromeos-images/diff/report?from=0.15.892.0&to=0.15.893.0
[00:16:16] <ellyjones> woah
[00:16:22] <ellyjones> I didn't know it could do that
[00:17:26] <ellyjones> let's blame anush, then
[00:17:55] <jamescook> x86 generic full looks like SimpleTestUpdateAndVerify again
[00:19:02] <jamescook> I'm going to speculatively reopen and hope it goes green.
[00:20:23] <crosbot> tree became 'Tree is open ("x86 generic full" -> SimpleTestUpdateAndVerify, crosbug.com/18624, hopefully flake)'
[00:21:16] <ellyjones> I really hate the phrase 'hopefully flake'
[00:21:52] <mononoke> gauravsh: thanks, I found out separately about chromeos-images/diff
[00:21:54] <jamescook> I can leave the tree closed until I'm more sure it's flake :-)
[00:22:56] <mononoke> I was also told that chromeos-images/diff is not completely reliable
[00:23:14] <mononoke> in fact it's hard to imagine how a busybox change would affect this
[00:23:40] <mononoke> so I looked between 891 and 894, but it's not particularly helpful
[00:24:27] <mononoke> There is a big change to the dev_install script that might have done it
[00:25:46] *** cooled has joined #chromium-os
[00:26:47] <darinski> jamescook: you should have attached bigger section of the log to the bug report
[00:27:01] <darinski> also, it doesn't seem like this issue has anything to do with the one reported originally on Aug 3
[00:27:17] <darinski> this failure was due to the device being unable to connect back to the dev server
[00:27:38] <darinski> [0816/142954:ERROR:libcurl_http_fetcher.cc(219)] Unable to get http response code.
[00:27:45] <jamescook> ok
[00:29:04] <jamescook> I wasn't sure which of the 14 ERROR line was the actual error.
[00:29:08] <jamescook> lines
[00:29:42] <darinski> they're all the same except a few that say "This is likely harmless."
[00:29:45] <darinski> regardless...
[00:29:50] <darinski> it will probably go away
[00:30:09] <darinski> but it would be nice for somebody to figure out why the connection failed
[00:33:33] <jamescook> Perhaps it is the same as crosbug.com/16591 ?
[00:35:48] <darinski> yes, at least it manifests the same way
[00:36:07] <darinski> also, i believe crosbug.com/18624 is a duplicate of an issue that's already fixed
[00:38:17] *** grundler has quit IRC
[00:42:34] <crosbot> tree became 'Tree is open ("x86 generic full" -> SimpleTestUpdateAndVerify, possibly crosbug.com/16591 )'
[00:45:35] *** arun_ has joined #chromium-os
[00:55:58] <crosbot> tree became '?'
[00:56:13] <crosbot> tree became 'Tree is open ("x86 generic full" -> SimpleTestUpdateAndVerify, possibly crosbug.com/16591 )'
[01:04:24] *** vmil86 has quit IRC
[01:13:12] *** jennb has joined #chromium-os
[01:13:12] *** ChanServ sets mode: +v jennb
[01:14:58] *** radhermit has quit IRC
[01:30:59] *** thieule has quit IRC
[01:47:45] <satorux_> sheriffs, i just checked in a relatively tricky patch: http://gerrit.chromium.org/gerrit/#change,6058
[01:48:38] <satorux_> it shouldn't break anything but if it does, please let me know or just revert it if needed
[01:49:56] *** lipsinV2 has joined #chromium-os
[02:00:59] <jamescook> SimpleTestUpdateAndVerify happened again on x86 generic full, but we're just going to watch it for now
[02:06:34] *** saintlou has quit IRC
[02:07:15] *** saintlou has joined #chromium-os
[02:10:50] *** cooled has quit IRC
[02:11:04] <jamescook> Heading to bus, will be back online around 6:15 Pacific
[02:13:36] *** jamescook has quit IRC
[02:16:44] *** jennb has quit IRC
[02:16:55] *** sadrul has quit IRC
[02:27:13] *** ScriptRipper has quit IRC
[02:27:50] *** Keybuk has quit IRC
[02:29:53] *** sadrul has joined #chromium-os
[02:29:53] *** ChanServ sets mode: +v sadrul
[02:32:18] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot_master" on "x86 generic PFQ" from 2b49e26688ed25b818ef82c39ff353e9f826d42d: dschuff at google dot com <dschuff at google dot com@fcba33aa-ac0c-11dd-b9e7-8d5594d729c2>)'
[02:33:32] *** cooled has joined #chromium-os
[02:43:03] <crosbot> tree became 'Tree is closed (tbroch -> investigating chrome sig 11 during login_CryptohomeUnmounted)'
[03:03:03] *** Haraken has left #chromium-os
[03:04:48] <crosbot> tree became 'Tree is open (tbroch --> cycled green, chrome sig 11  http://crosbug.com/18269 and http://crosbug.com/17899)'
[03:05:27] *** tbroch has quit IRC
[03:06:35] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "tegra2_arthur-binary" from f496915667b9956ac160ece2d73385f9bb0adc47: Darin Petkov <petkov at chromium dot org>, Scott Zawalski <scottz at google dot com>, fischman at chromium dot org <fischman at chromium dot org@0039d316-1c4b-4281-b951-d872f2087c98>, jam at chromium dot org <jam at chromium dot org@0039d316-1c4b-4281-b951-d872f2087c98>, rvargas at google dot com <rvargas at google dot com@0039d316-1c4b-4281-b951-d872f2087c98>)'
[03:13:38] *** jamescook has joined #chromium-os
[03:14:14] *** wfrichar has quit IRC
[03:16:47] *** ScriptRipper has joined #chromium-os
[03:17:19] <jamescook> satorux_: you around?  tegra2_arthur-binary looks like cromo-0.0.1-r114: cromo_server.cc:7:45: error: chromeos/dbus/service_constants.h: No such file or directory
[03:17:25] <jamescook> Is that your change?
[03:17:42] *** kerz has quit IRC
[03:18:33] *** satorux__ has joined #chromium-os
[03:19:37] *** tbroch has joined #chromium-os
[03:19:37] *** ChanServ sets mode: +v tbroch
[03:20:10] <crosbot> tree became 'Tree is closed (jamescook investigating -> "cbuildbot" on "tegra2_arthur-binary", chromeos/dbus/service_constants.h not found )'
[03:20:13] <jamescook> hi tbrouch
[03:20:20] <jamescook> tbroch
[03:21:20] <tbroch> hi
[03:21:33] <jamescook> Apparently someone hit "force build"
[03:21:58] <jamescook> and didn't fill out a name ... who it was, we'll never know.
[03:29:20] <jamescook> clobbering tegra2_arthur
[03:31:56] <jamescook> clobbering tegra2_aebl also, we'll see if this gets us past the ebuild changes from satorux
[03:32:12] *** nirnimesh has quit IRC
[03:36:32] *** alecz has joined #chromium-os
[03:38:45] <crosbot> tree became 'Tree is closed (jamescook investigating -> multiple platforms, chromeos/dbus/service_constants.h not found )'
[03:39:54] *** dianders1 has joined #chromium-os
[03:40:06] *** dianders2 has quit IRC
[03:41:02] *** alecz has left #chromium-os
[03:41:39] *** jrbarnette has quit IRC
[03:42:20] <crosbot> tree became 'Tree is closed (multiple platforms, chromeos/dbus/service_constants.h not found, jamescook trying clobbers )'
[03:43:02] *** fischman has left #chromium-os
[03:59:23] *** Styx has quit IRC
[04:00:19] <jamescook> botsnack
[04:00:30] <jamescook> crosbot: botsnack
[04:00:31] <crosbot> jamescook: :)
[04:06:10] <crosbot> tree became 'Tree is open (tegra2, x86-alex should cycle green after clobber)'
[04:10:19] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "tegra2_seaboard-tangent canary" from None: )'
[04:16:27] <jamescook> sheriff going off duty, next shift picking up
[04:16:34] <jamescook> good night
[04:16:36] *** jamescook has left #chromium-os
[04:23:49] *** nirnimesh has joined #chromium-os
[04:23:49] *** ChanServ sets mode: +v nirnimesh
[04:26:39] <crosbot> tree became 'Tree is closed (investigating -> failure in syncing chrome source)'
[04:31:00] *** sadrul has quit IRC
[04:32:01] <crosbot> tree became 'Tree is open (crosbug.com/17959)'
[04:52:10] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "x86-alex canary" from None: )'
[04:55:32] *** McMAGIC--Copy has quit IRC
[04:56:02] *** sadrul has joined #chromium-os
[04:59:49] <crosbot> tree became 'Tree is open ("x86-alex canary": testUpdateKeepStateful flake -> crosbug.com/18679)'
[05:00:14] *** McMAGIC--Copy has joined #chromium-os
[05:31:59] *** radhermit has joined #chromium-os
[05:36:36] *** Inumedia has joined #chromium-os
[05:44:08] *** dianders1 has quit IRC
[05:46:34] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "x86 generic full" from ff04fb1cb63a8fa67c65f985192db039aa8556e9: _third_party_ at chromium dot org, bradnelson at google dot com <bradnelson at google dot com@fcba33aa-ac0c-11dd-b9e7-8d5594d729c2>)'
[05:52:57] <crosbot> tree became 'Tree is throttled ("x86 generic full" hungte investigating chromeos-base/chromeos-firmware ebuild)'
[05:53:37] *** jamescook has joined #chromium-os
[05:55:54] <jamescook> djkurtz: everything under control?
[05:56:25] <djkurtz> yup
[05:56:41] <jamescook> great!  have a good day
[05:56:48] <djkurtz> tty tomorrow
[05:56:55] <jamescook> bye
[05:57:09] *** jamescook has left #chromium-os
[06:05:40] *** sadrul has quit IRC
[06:10:52] <crosbot> tree became 'Tree is green ("x86 generic full" fixed by hungte: crosbug.com/19265, should cycle green)'
[06:10:58] *** Daxvex has quit IRC
[06:18:19] *** cooled has quit IRC
[06:22:02] *** cooled has joined #chromium-os
[06:23:09] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "x86 pineview full" from c3814b9effbf68c53ae921599f4303efde10c5aa: Hung-Te Lin <hungte at chromium dot org>, bradnelson at google dot com <bradnelson at google dot com@fcba33aa-ac0c-11dd-b9e7-8d5594d729c2>)'
[06:24:26] <crosbot> tree became 'Tree is green (crosbug.com/19265, should cycle green)'
[06:27:18] *** sadrul has joined #chromium-os
[06:27:18] *** ChanServ sets mode: +v sadrul
[06:28:28] *** Inumedia has quit IRC
[07:02:22] *** dhendrix has quit IRC
[07:04:55] *** Sarten-X has quit IRC
[07:11:44] *** Sarten-X has joined #chromium-os
[07:16:11] *** D|sT has joined #chromium-os
[07:16:47] *** D|sT has quit IRC
[07:22:06] *** arun has quit IRC
[07:57:22] *** vmil86 has joined #chromium-os
[08:17:44] *** sergiu has quit IRC
[08:20:03] *** arun has joined #chromium-os
[08:20:36] *** cooled has quit IRC
[08:24:07] *** cooled has joined #chromium-os
[08:54:01] *** aksr has joined #chromium-os
[08:55:48] *** aksr_ has joined #chromium-os
[08:56:15] *** aksr has quit IRC
[08:58:32] *** aksr_ has quit IRC
[09:10:33] *** jujugre has joined #chromium-os
[09:15:51] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "x86-zgb-binary" from a5c1ae42c84dd8b5071383aa4731bc4ff5bcf471: benjhayden at chromium dot org <benjhayden at chromium dot org@0039d316-1c4b-4281-b951-d872f2087c98>, erg at chromium dot org <erg at chromium dot org@0039d316-1c4b-4281-b951-d872f2087c98>)'
[09:21:54] *** Sergiu has joined #chromium-os
[09:21:55] *** Sergiu has joined #chromium-os
[09:24:00] *** magn3ts has quit IRC
[09:27:35] <crosbot> tree became 'Tree is open (x86-zgb-binary Archive Stage timeout -> crosbug.com/19266)'
[09:35:45] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "tegra2_kaen-binary" from eb6f247fe802b9a44a31eaaea8257b2ba2df2847: Dave Parker <dparker at chromium dot org>, Hung-Te Lin <hungte at chromium dot org>)'
[09:36:51] *** jochen___ has joined #chromium-os
[09:36:52] *** ChanServ sets mode: +v jochen___
[09:38:50] *** jochen__ has quit IRC
[09:38:51] *** jochen___ is now known as jochen__
[09:50:11] *** satorux__ has quit IRC
[10:01:17] <crosbot> tree became 'Tree is open (tegra2_kaen_binary sync timeout (already cycled green)'
[10:28:12] *** Styx has joined #chromium-os
[10:32:35] *** Sergiu has quit IRC
[10:33:03] *** Sergiu has joined #chromium-os
[11:17:02] *** patcito has quit IRC
[11:52:16] *** Sergiu has quit IRC
[12:45:33] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "tegra2_kaen-binary" from fc24dc42f7cff81b6d058040b299d3f33f794027: mazda at chromium dot org <mazda at chromium dot org@0039d316-1c4b-4281-b951-d872f2087c98>)'
[12:48:37] *** mazda has joined #chromium-os
[12:56:46] <crosbot> tree became 'Tree is open (tegra2_kaen_binary sync timeout (cycling green))'
[13:00:58] *** ernie_ has joined #chromium-os
[13:01:11] *** ernie_ has left #chromium-os
[13:12:50] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "tegra2_arthur-binary" from fc24dc42f7cff81b6d058040b299d3f33f794027: mazda at chromium dot org <mazda at chromium dot org@0039d316-1c4b-4281-b951-d872f2087c98>)'
[13:16:40] *** bleung has quit IRC
[13:24:20] <mazda> build bots seem flaky. repo sync is failing for some reason...
[13:25:14] *** elly has quit IRC
[13:25:49] *** elly has joined #chromium-os
[13:25:49] *** Daxvex has joined #chromium-os
[13:27:38] <crosbot> tree became 'Tree is closed (looking at the next cycle of "tegra2_arthur-binary")'
[13:33:48] <crosbot> tree became 'Tree is open (cycling green)'
[13:40:45] *** Styx has quit IRC
[14:00:08] *** D|sT has joined #chromium-os
[14:02:23] *** D|sT has quit IRC
[14:50:33] *** SoulRaven has joined #chromium-os
[14:50:53] *** vpalatin has quit IRC
[14:51:09] *** vpalatin has joined #chromium-os
[14:51:09] *** ChanServ sets mode: +v vpalatin
[14:55:08] *** cjb has joined #chromium-os
[15:08:05] *** Sarten-X has quit IRC
[15:13:01] *** bleung has joined #chromium-os
[15:13:02] *** ChanServ sets mode: +v bleung
[15:13:27] *** D|sT has joined #chromium-os
[15:14:23] *** Sarten-X has joined #chromium-os
[15:16:45] *** l9root has joined #chromium-os
[15:28:51] *** l9root has quit IRC
[15:49:17] *** rharrison_chrome has joined #chromium-os
[15:49:18] *** ChanServ sets mode: +v rharrison_chrome
[15:59:38] *** sakoman has quit IRC
[16:02:47] *** sadrul has quit IRC
[16:04:44] *** alecz has joined #chromium-os
[16:07:20] *** sakoman has joined #chromium-os
[16:13:49] *** cros_ has joined #chromium-os
[16:21:10] *** penghuang has quit IRC
[16:23:02] *** alecz has quit IRC
[16:33:40] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "x86-zgb canary" from None: )'
[16:35:28] *** Daxvex has quit IRC
[16:38:44] *** oc80z has quit IRC
[16:39:37] *** oc80z has joined #chromium-os
[16:41:04] <crosbot> tree became 'Tree is open (x86-zgb canary: testUpdateKeepStateful failed -> crosbug.com/18679)'
[16:42:12] *** SeligArkin has quit IRC
[16:42:45] *** SeligArkin has joined #chromium-os
[16:45:05] <ellyjones> I just landed http://gerrit.chromium.org/gerrit/#change,6063, which has previously broken the tree, but hopefully won't this time :)
[16:46:47] *** robster has joined #chromium-os
[16:47:31] *** jrr has left #chromium-os
[16:55:48] *** oc80z has quit IRC
[16:56:02] *** GTAXL has quit IRC
[16:56:40] *** oc80z has joined #chromium-os
[17:00:28] *** sadrul has joined #chromium-os
[17:00:28] *** ChanServ sets mode: +v sadrul
[17:08:42] *** Daxvex has joined #chromium-os
[17:08:49] *** JasonO has joined #chromium-os
[17:14:42] *** Daxvex has quit IRC
[17:15:41] *** Daxvex has joined #chromium-os
[17:17:52] *** Daxvex1 has joined #chromium-os
[17:20:56] *** Daxvex has quit IRC
[17:27:04] *** jrbarnette has joined #chromium-os
[17:35:52] *** GTAXL has joined #chromium-os
[17:36:23] *** wfrichar has joined #chromium-os
[17:36:24] *** ChanServ sets mode: +v wfrichar
[17:37:32] *** dianders1 has joined #chromium-os
[17:47:22] *** jrbarnette_ has joined #chromium-os
[17:47:22] *** ChanServ sets mode: +v jrbarnette_
[17:51:23] *** jrbarnette has quit IRC
[17:51:23] *** jrbarnette_ is now known as jrbarnette
[17:51:45] *** arun has quit IRC
[17:55:32] *** Daxvex has joined #chromium-os
[17:57:10] *** Daxvex1 has quit IRC
[18:01:36] *** arun has joined #chromium-os
[18:03:01] *** Daxvex1 has joined #chromium-os
[18:05:46] *** Daxvex has quit IRC
[18:09:58] *** jrbarnette_ has joined #chromium-os
[18:11:23] *** jamescook has joined #chromium-os
[18:11:23] *** ChanServ sets mode: +v jamescook
[18:12:01] *** saintlou_ has joined #chromium-os
[18:12:11] <jamescook> sheriff reporting for duty
[18:12:17] <jamescook> djkurtz, how did your day go?
[18:14:12] *** jrbarnette has quit IRC
[18:14:25] <satorux_> it's already midnight in his timezone
[18:14:56] *** saintlou has quit IRC
[18:23:48] <jamescook> crosbug.com is down, codesite-oncall is aware and working on it
[18:25:19] *** kerz has joined #chromium-os
[18:25:19] *** ChanServ sets mode: +v kerz
[18:25:29] <cmasone> jamescook: yeah, isn't that aggravating :-P
[18:29:33] *** GTAXL has quit IRC
[18:29:50] <satorux_> my mission of moving two files into a new repository is now completed. took like 20 patches
[18:30:03] <jamescook> wow - painful
[18:30:09] <satorux_> i'm going to try cros_sdk finally
[18:30:17] *** saintlou_ has quit IRC
[18:31:00] *** grundler has joined #chromium-os
[18:31:10] *** ChanServ sets mode: +v grundler
[18:43:26] *** dianders1 has quit IRC
[18:46:53] <zbehan> satorux_: awesome, let me know how it works for you :)
[18:47:41] *** gsam_ has quit IRC
[18:47:56] <satorux_> just created a chroot with cros_sdk. it worked for me
[18:48:35] <satorux_> i expected it to be much faster than make_chroot, but I forgot how slow make_chroot was, so I could not compare :)
[18:50:37] <satorux_> setup_board only took 22 secs.
[18:52:12] *** Styx has joined #chromium-os
[18:54:33] *** lipsinV2 has quit IRC
[18:58:32] <jamescook> Nice Google doodle today.
[19:06:42] *** rbyers_ has joined #chromium-os
[19:09:05] <cmasone> man, still can't file new issues
[19:11:13] <jamescook> Good thing we have defect-free software. :-)
[19:11:25] <cmasone> aaaaaand I can't load gmail either
[19:13:28] *** jrbarnette has joined #chromium-os
[19:13:29] *** ChanServ sets mode: +v jrbarnette
[19:14:00] *** saintlou has joined #chromium-os
[19:14:09] *** jrbarnette has quit IRC
[19:17:02] *** penghuang has joined #chromium-os
[19:17:03] *** ChanServ sets mode: +v penghuang
[19:17:22] <jamescook> sheriff -> getting coffee, brb
[19:24:09] <jamescook> Update on crosbug.com: "Service has been (mostly) restored. There's still a high level of cookieless traffic going to very expensive URLs in the issue tracker, so performance may suffer for a while longer."
[19:27:29] <ellyjones> fail :P
[19:31:17] <cmasone> yup.  still can't file an issue
[19:31:30] <robster> does BOARD=amd64-generic work well?
[19:33:01] *** jrbarnette has joined #chromium-os
[19:33:01] *** ChanServ sets mode: +v jrbarnette
[19:34:39] *** jrbarnette has quit IRC
[19:35:10] <jamescook> crosbot, botsnack
[19:35:55] *** jrbarnette has joined #chromium-os
[19:35:55] *** ChanServ sets mode: +v jrbarnette
[19:36:02] <ellyjones> jamescook: :, not ,
[19:36:13] <jamescook> crosbot: botsnack
[19:36:13] <crosbot> jamescook: :)
[19:36:27] <jamescook> sadly "," is the autocomplete default on xchat
[19:36:35] *** jrbarnette has quit IRC
[19:36:41] *** jrbarnette has joined #chromium-os
[19:36:41] *** ChanServ sets mode: +v jrbarnette
[19:36:43] <njw> what do people like for cross-compiling small or test applications in the chroot? Anything more reasonable than creating a local ebuild & co?
[19:39:05] *** jrbarnette has quit IRC
[19:39:49] <ellyjones> add it to an existing ebuild :3
[19:41:55] <njw> probably the expedient approach, but ugh.
[19:50:57] *** dtu has quit IRC
[19:51:15] <zbehan> njw: well, crosscompiling in the chroot is probably going to be needing a deployment to the target as well, right? seems like making an ebuild and using gmerge would make it a 1-step process, compared to, i don't know, a Makefile
[19:54:56] <jamescook> tbroch: I can take this one
[19:55:19] <tbroch> jamescook: sure
[19:55:27] <jamescook> You need to head out, right?
[19:55:49] <tbroch> I'll be around ( mtg cancelled )
[19:57:45] <jamescook> tar failed: RunCommandError: ('Command "[\'tar\', \'cf\', \'/b/cbuild/src/build/images/x86-generic/latest-cbuildbot/autotest.tar.bz2\', \'--checkpoint=10000\', \'--use-compress-program=/b/cbuild/chroot/usr/bin/pbzip2\', \'autotest\']" with extra env None\n', ['tar', 'cf', '/b/cbuild/src/build/images/x86-generic/latest-cbuildbot/autotest.tar.bz2', '--checkpoint=10000', '--use-compress-program=/b/cbuild/chroot/usr/bin/pbzip2', 'auto
[19:57:45] <jamescook> test'], 1)
[19:57:48] <jamescook> I wonder if the disk is full
[19:59:36] *** rbyers_ has quit IRC
[20:00:43] *** achuith has quit IRC
[20:05:20] <cmasone> jamescook: get a trooper?
[20:05:26] <cmasone> troopers at chromium dot org
[20:05:29] <jamescook> davidjames knows what it is
[20:05:42] <jamescook> flaky failure, entering bug for him
[20:06:30] <davidjames> jamescook: Yeah the bug is tests and test archiving running in parallel, I guess they can't run in parallel so I'll have to serialize those two
[20:06:58] <jamescook> "Waiting for code.google.com..."
[20:10:07] <jamescook> Filed as http://crosbug.com/19280
[20:13:23] *** Keybuk has joined #chromium-os
[20:13:24] *** ChanServ sets mode: +v Keybuk
[20:13:30] <davidjames> jamescook: Thanks, crosbug.com is so slow that I'll just take your word for it :)
[20:13:37] <jamescook> heheh
[20:17:01] *** jrbarnette has joined #chromium-os
[20:17:01] *** ChanServ sets mode: +v jrbarnette
[20:17:14] *** ukai_ has quit IRC
[20:20:26] *** McMAGIC--Copy has quit IRC
[20:21:19] *** JasonO has quit IRC
[20:21:34] *** dhendrix has joined #chromium-os
[20:21:35] *** ChanServ sets mode: +v dhendrix
[20:22:13] *** JasonO has joined #chromium-os
[20:22:13] *** JasonO has joined #chromium-os
[20:22:16] *** McMAGIC--Copy has joined #chromium-os
[20:23:26] <eggy> i see some 64bit overlays, are these usable?
[20:24:33] *** saintlou has quit IRC
[20:24:38] <marcheu> no
[20:24:41] <davidjames> eggy: They don't compile, but we could sure use help making them compile :)
[20:24:56] <eggy> mkay
[20:26:14] <cmasone> hm...why didn't the PFQs kick off again
[20:28:12] <davidjames> cmasone: I filed a bug with infrastructure team about it, there's a bug for on crbug.com but crbug.com is down
[20:31:04] *** jrbarnette has quit IRC
[20:31:28] *** jrbarnette has joined #chromium-os
[20:31:28] *** ChanServ sets mode: +v jrbarnette
[20:32:26] <jamescook> looking
[20:33:01] <cmasone> crosbot: announce +tree
[20:33:01] <crosbot> cmasone: Admin access required
[20:33:09] <cmasone> :-(
[20:33:13] <jamescook> o3d.gyp ?
[20:33:31] <davidjames> Exception: Call to 'uname -m | sed -e "s/i.86/ia32/;s/x86_64/x64/;s/amd64/x64/;s/arm.*/arm/;s/i86pc/ia32/"' returned exit status 0. while trying to load src/build/all.gyp
[20:33:39] <cmasone> ellyjones: crosbot not announce tree!
[20:33:49] *** saintlou has joined #chromium-os
[20:33:50] *** ChanServ sets mode: +v saintlou
[20:34:03] <davidjames> crosbot: stat tree
[20:34:03] <crosbot> davidjames: tree: Tree is open (x86-zgb canary: testUpdateKeepStateful failed -> crosbug.com/18679)
[20:34:21] *** saintlou has quit IRC
[20:34:23] <ellyjones> what
[20:34:26] <ellyjones> crosbot: announce +tree
[20:34:27] <crosbot> ellyjones: Admin access required
[20:34:32] * ellyjones glares at crosbot
[20:34:46] <ellyjones> crosbot: announce +tree
[20:34:47] <crosbot> ellyjones: announcing tree
[20:34:48] <ellyjones> there
[20:34:51] <ellyjones> I dunno what happened to it
[20:34:54] <ellyjones> crosbot: tree?
[20:34:55] <crosbot> ellyjones: tree: Tree is open (x86-zgb canary: testUpdateKeepStateful failed -> crosbug.com/18679)
[20:44:26] *** ukai_ has joined #chromium-os
[20:44:26] *** ChanServ sets mode: +v ukai_
[20:47:34] *** achuith has joined #chromium-os
[20:48:22] <cmasone> da's changes need to be reverted
[20:48:32] <cmasone> that's why all the PFQs died
[20:48:48] <cmasone> tbroch: jamescook ^^^
[20:49:13] *** achuith has joined #chromium-os
[20:50:06] *** achuith has quit IRC
[20:50:19] <cmasone> The following keyword changes are necessary to proceed:
[20:50:19] <cmasone> #required by chromeos-base/chromeos-dev-0.1.0-r44, required by @selected, required by @world (argument)
[20:50:22] <cmasone> >=dev-util/systemtap-1.5 ~x86
[20:50:25] <cmasone> #required by dev-util/systemtap-1.5, required by chromeos-base/chromeos-dev-0.1.0-r44, required by @selected, required by @world (argument)
[20:50:28] <cmasone> >=dev-libs/elfutils-0.152 ~x86
[20:50:56] <ellyjones> crosbot: tree?
[20:50:57] <crosbot> ellyjones: tree: Tree is open (x86-zgb canary: testUpdateKeepStateful failed -> crosbug.com/18679)
[20:51:07] <ellyjones> ok, it isn't failing to announce
[20:51:19] <cmasone> ellyjones: yeah, it's just not seein the updates
[20:52:14] <ellyjones> that is bizarre
[20:52:36] * ellyjones checks its logs
[20:53:10] <tbroch> reverted http://gerrit.chromium.org/gerrit/#change,5274
[20:55:03] *** crosbot has quit IRC
[20:55:11] *** crosbot has joined #chromium-os
[20:55:14] <ellyjones> gave it a good kick
[20:55:17] <ellyjones> crosbot: tree?
[20:55:26] <ellyjones> crosbot: tree?
[20:55:26] <crosbot> ellyjones: tree: Tree is closed (jamescook investigating "cbuildbot" on "tegra2_arthur canary", o3d.gyp ?)
[20:55:29] <ellyjones> there we go
[20:55:33] <ellyjones> dunno why it got wedged
[20:55:46] <ellyjones> crosbot: announce +tree
[20:55:46] <crosbot> ellyjones: announcing tree
[20:56:37] <crosbot> tree became 'Tree is closed (tbroch -> reverted  http://gerrit.chromium.org/gerrit/#change,5274)'
[20:57:43] <jamescook> tbroch: I think the arthur/o3d thing might be different, I guess we'll see
[20:58:12] <jamescook> I think I pulled us down a different error, which is why we didn't see da's changes
[20:59:35] <davidjames> jamescook, tbroch: FYI, arm-generic was stuck on WebKit_trimmed.git, just killed off a git process so it will continue
[20:59:43] <jamescook> ok, thank you
[21:00:29] *** steev_ has quit IRC
[21:00:37] *** reveman has quit IRC
[21:00:40] *** GodoPPL has quit IRC
[21:02:35] <davidjames> jamescook: Yeah, arthur failure is different
[21:03:39] <davidjames> jamescook: All arm internal bots are failing with  Exception: Call to 'uname -m | sed -e "s/i.86/ia32/;s/x86_64/x64/;s/amd64/x64/;s/arm.*/arm/;s/i86pc/ia32/"' returned exit status 0. while trying to load src/build/all.gyp
[21:04:37] <davidjames> jamescook: I wonder if they are failing due to vapier's locale change ? Maybe they are spotting 'warning' output from sed about locales and treating that as an error
[21:04:43] <davidjames> vapier: ^
[21:05:37] <vapier> how do i invoke the buildbot on that for arm targets ?  arm-generic-bin sufficient ?
[21:06:10] *** achuith has joined #chromium-os
[21:06:41] <vapier> sys-block/btrace-1.0.0 seems to have a parallel build issue
[21:07:42] *** Styx has quit IRC
[21:08:28] <davidjames> vapier: It's release builders that are failing
[21:09:08] <davidjames> vapier: e.g. x86-mario-release. Note that that's a really slow builder
[21:09:14] <vapier> yes, but how to reproduce locally ?
[21:09:17] *** rbyers_ has joined #chromium-os
[21:09:50] <davidjames> vapier: cbuildbot x86-alex_he-release
[21:10:07] <davidjames> vapier: (or any of the builders that are failing)
[21:10:37] <vapier> where is that magic config name coming from ?
[21:10:48] <davidjames> vapier: Top of any cbuildbot log
[21:11:03] <davidjames> vapier: E.g. http://chromeos-botmaster.mtv.corp.google.com:8026/builders/x86-mario%20canary/builds/879/steps/cbuildbot/logs/stdio
[21:14:09] *** robster has quit IRC
[21:14:13] <jamescook> Anything I can do to help, guys?
[21:14:47] <vapier> when i use cbuildbot, does it always build from source ?  or does it use binpkgs ?
[21:15:15] <davidjames> vapier: Depends on the builder, x86-mario-release builds from source
[21:16:23] <kliegs> davidjames: I'm getting an error about "The following keyword changes are necessary to proceed:" at the end of build_packages. and then it lists elfutils-0.152 ~arm and dev-util/systemtap-1.5 **
[21:16:32] <kliegs> any idea about that/how to fix?
[21:16:42] <cmasone> kliegs: read backscroll
[21:16:47] <davidjames> kliegs: Yeah that's already fixed, just sync
[21:16:49] <cmasone> kliegs: offending change already reverted
[21:17:20] <kliegs> cmasone, davidjames: thanks.  missed it in my previous backscroll read. need a better scroll parser :(
[21:17:33] <ellyjones> you need /lastlog :)
[21:17:34] <davidjames> vapier: Confirmed it's failing due to this error: /bin/sh: warning: setlocale: LC_ALL: cannot change locale (en_US.UTF-8)
[21:17:45] <davidjames> vapier: Just reproduced on the bot itself
[21:17:50] <vapier> and the script dir is up-to-date ?
[21:18:11] <kliegs> ellyjones: whoa! cool
[21:18:18] <vapier> what is the output of `locale` in the env of the buildbot ?
[21:18:20] <vapier> empty ?
[21:18:37] <davidjames> vapier: Everything is en_US except LC_ALL
[21:18:42] <davidjames> (which is empty)
[21:18:54] <kliegs> ellyjones: is there a way to specify how far back to search?
[21:19:16] <davidjames> vapier: Also note that sh -c 'uname -m | sed -e "s/i.86/ia32/;s/x86_64/x64/;s/amd64/x64/;s/arm.*/arm/;s/i86pc/ia32/"' works fine for me, no warnings, but when gclient runs it, it definitely gets warnings
[21:19:31] <ellyjones> kliegs: yes, see /help lastlog
[21:19:49] <ellyjones> crosbot: sickbots?
[21:19:49] <crosbot> ellyjones: sickbots: x86 generic full
[21:19:54] <ellyjones> hrm
[21:20:01] <kliegs> ellyjones: hmm. my lastlog may be inferior :(  it just says Usage: LASTLOG <string>
[21:20:05] <ellyjones> its heuristic is bad
[21:20:10] <ellyjones> kliegs: are you not using irssi? that'd be why
[21:20:12] <kliegs> holy crap crosbot has nifty stuff
[21:20:17] <kliegs> ellyjones: :(  xchat
[21:20:21] <vapier> davidjames: so who is setting en_US.UTF8 in the env ?
[21:20:27] <kliegs> people keep talking about irssi - maybe time to play with it
[21:20:30] <davidjames> vapier: gyp does that
[21:20:37] *** vapier has left #chromium-os
[21:20:46] *** vapier has joined #chromium-os
[21:20:46] *** ChanServ sets mode: +v vapier
[21:20:53] <vapier> blah wrong key
[21:20:55] <vapier> what is gyp ?
[21:21:01] <davidjames> vapier: gyp sets LC_ALL=en_US.UTF-8
[21:21:04] <davidjames> vapier: That's Chrome build system
[21:22:05] <davidjames> vapier: Can reproduce problem here:
[21:22:09] <davidjames>  LC_ALL=en_US.UTF-8 sh -c 'uname -m | sed -e "s/i.86/ia32/;s/x86_64/x64/;s/amd64/x64/;s/arm.*/arm/;s/i86pc/ia32/"'
[21:22:10] <davidjames> sh: warning: setlocale: LC_ALL: cannot change locale (en_US.UTF-8)
[21:22:10] <davidjames> x64
[21:22:15] *** saintlou has joined #chromium-os
[21:22:16] *** ChanServ sets mode: +v saintlou
[21:22:49] <vapier> yes, but i'm more interested in how the locale is getting set if it isnt by the tools outside of `enter_chroot`
[21:23:11] <davidjames> vapier: Or even this: LC_ALL=en_US.UTF-8 sh -c 'true'
[21:23:13] <vapier> sounds like LANG=en_US, then something runs enter_chroot, then inside of there something does LANG=en_US.UTF8
[21:23:21] *** GTAXL has joined #chromium-os
[21:23:21] <vapier> is that correct ?
[21:23:26] *** steev_ has joined #chromium-os
[21:24:18] <davidjames> vapier: Yeah Chrome build system sets LC_ALL
[21:24:27] *** rbyers_ has quit IRC
[21:24:30] <vapier> ok, then the fix is easy
[21:24:45] *** GodoPPL has joined #chromium-os
[21:24:46] <davidjames> vapier: Also note that Chrome build system runs locale-gen
[21:25:17] <davidjames> vapier: From install-chroot.sh: locale-gen $LANG en_US en_US.UTF-8
[21:25:22] <davidjames> vapier: I think we don't run that script though
[21:25:59] <vapier> where is that ?  i dont see it in depot_tools or src/scripts/
[21:26:54] <davidjames> vapier: ~/trunk/chromium/src/build/install-chroot.sh
[21:27:09] <davidjames> vapier: Again note that script probably isn't used :)
[21:27:40] <davidjames> vapier: And note that during build Chromium is checked out from svn to /var/lib/portage/distfiles-target/...
[21:27:45] *** D|sT has quit IRC
[21:28:50] <davidjames> vapier: I guess fix is to always generate en_US.UTF-8 since that's needed by Chrome ?
[21:29:02] <vapier> yes, that's what i added
[21:29:11] <vapier> but i started looking at these other files you referenced ;)
[21:30:38] <jamescook> Newbie-sheriff question: Did something change that we should revert?  And should we hold the tree closed while the canaries are broken?
[21:31:20] <davidjames> jamescook: http://gerrit.chromium.org/gerrit/5986 broke the tree, reverting it would fix canaries
[21:32:00] <vapier> i posted http://gerrit.chromium.org/gerrit/6168 to fix it
[21:32:10] <davidjames> jamescook: ^
[21:32:28] <vapier> waiting for locale buildbot to finish to verify
[21:32:49] *** D|sT has joined #chromium-os
[21:33:17] <jrbarnette> i'm seeing compile errors in build_packages for the 'btrace' package
[21:33:20] <vapier> also, i dont really understand how 5986 could have broken it when that only adds new locales ... if anything, it'd be the previous commit which dropped the hardcoding of LC_ALL=C
[21:33:34] <jrbarnette> is anyone else seeing that symptom?
[21:33:42] <cmasone> jrbarnette: what pulled that in?  Can we revert it?  Someone mentioned that above
[21:33:44] <vapier> jrbarnette: yeah, parallel build i think ... i poked the Da Zheng about it
[21:33:47] <cmasone> vapier, I think?
[21:34:05] <vapier> btrace failing is diff from the locale
[21:34:14] <cmasone> vapier: yeah, but we should revert that too
[21:34:16] <jrbarnette> I thought Da's change had a different compile error
[21:34:20] <cmasone> jamescook: tbroch ^^^
[21:34:26] <vapier> i think he posted two broken ones :)
[21:34:29] <cmasone> jrbarnette: there were multople changes
[21:34:30] <jrbarnette> but i can try reverting to find out...
[21:34:36] <vapier> cmasone: well i just posted a fix for the locale one ;)
[21:34:59] <cmasone> vapier: that's distinct from this btrace one, though, right?
[21:35:03] <vapier> yes
[21:35:22] <jrbarnette> my chroot has the revert of the 'systemtap' change; i think the btrace build problem is new...
[21:35:40] <vapier> it is
[21:35:47] <cmasone> jrbarnette: he landed two changes at teh same time.  I think they were both broken, but we only reverted one
[21:35:52] <tbroch> i'm on the revert of btrace
[21:35:57] <vapier> davidjames: `cbuildbot x86-alex_he-release` errors out early
[21:36:25] <vapier> http://pastebin.com/raw.php?i=GYXUQ3pT
[21:36:26] <davidjames> tbroch: This would be the one to revert: http://gerrit.chromium.org/gerrit/5277
[21:36:33] <jrbarnette> sure enough, there was a change to unmask btrace
[21:37:14] <davidjames> jrbarnette: Yeah, don't revert the unmask though, the better one to revert would be the one that added btrace to chromeos-dev :)
[21:37:39] <tbroch> roger ... switching to reverting the ebuild
[21:37:40] <vapier> davidjames: oh, it doesnt recover nicely from ssh denied
[21:38:29] <gauravsh> powderluv: thanks for breaking recovery, buddy
[21:38:33] <vapier> Getting manifest ... from ssh://gerrit-int.chromium.org:29419/chromeos/manifest-internal: Permission denied (publickey).
[21:38:37] <davidjames> vapier: I think your change is good let's put it in if you can verify the cmd I typed earlier
[21:38:53] <davidjames> vapier:  LC_ALL=en_US.UTF-8 sh -c 'true'
[21:39:17] <tbroch> reverted http://gerrit.chromium.org/gerrit/#change,5277
[21:39:29] <vapier> it works for me, but it's not that good of a test as my locale is normally en_US.UTF8 :)
[21:39:30] <cmasone> tbroch: thanks!
[21:39:47] <vapier> i switched away from iso8859 a while ago
[21:39:49] <davidjames> vapier:  LC_ALL=en_US sh -c 'true'
[21:40:05] <vapier> yes, that is warning free after my change
[21:40:15] <jamescook> In the good news department, codesite-oncall says code.google.com is settling down
[21:40:23] <davidjames> vapier: Sweet let's put it in, change looks harmless in any case
[21:40:42] <davidjames> vapier: Just generating en_US.UTF-8 and en_US should only be a helpful thing
[21:41:12] *** reveman has joined #chromium-os
[21:41:12] *** ChanServ sets mode: +v reveman
[21:41:46] <vapier> should i not worry about install-chroot.sh ?
[21:41:59] <davidjames> vapier: That's for debian :)
[21:42:01] <vapier> # This script installs Debian-derived distributions in a chroot environment.
[21:42:06] <vapier> yeah ok, nm
[21:42:12] <davidjames> vapier: It uses apt-get and such :)
[21:42:21] <crosbot> tree became 'Tree is closed (vapier/djames -> flushing out local issues, tbroch -> reverted  http://gerrit.chromium.org/gerrit/#change,5274 &amp; 5277)'
[21:43:03] <vapier> any idea why gerrit-int.chromium.org hates me ?
[21:43:14] <davidjames> vapier: You probably need to register your keys there :)
[21:43:26] <davidjames> vapier: It's a separate setup from gerrit.chromium.org
[21:43:30] <vapier> oh, ok
[21:45:16] <jamescook> brb
[21:45:19] *** jamescook is now known as jamescookAFK
[21:45:23] <davidjames> vapier: So I confirmed here that if the locale is completely broken then sh prints no errors
[21:45:36] <davidjames> vapier: However if it is partially working, it prints a warning
[21:45:42] <vapier> fatal: '/chromeos/manifest-internal': not a Gerrit project
[21:45:52] <vapier> it accepted my key this time
[21:46:23] *** oc80z has quit IRC
[21:46:23] *** oc80z has joined #chromium-os
[21:47:01] <vapier> davidjames: `LC_ALL=en_US perl -e ''` works
[21:47:25] <vapier> perl tends to be much pickier at startup with missing locales
[21:47:36] <vapier> (which is what all the old debian apt warnings originated from)
[21:47:50] *** Hexxeh has quit IRC
[21:50:32] *** jamescookAFK is now known as jamescook
[21:50:47] <vapier> blah looks like i signed in with the wrong e-mail on gerrit-int
[21:51:09] *** dtu has joined #chromium-os
[21:51:12] *** ChanServ sets mode: +v dtu
[21:51:17] <crosbot> tree became 'Tree is open (locale issues should be fixed. tbroch -> reverted  http://gerrit.chromium.org/gerrit/#change,5274 &amp; 5277)'
[21:52:25] *** patcito has joined #chromium-os
[21:52:39] <cmasone> what is all this perl stiff that IO'm building now???
[21:52:41] <cmasone> uh
[21:52:44] <cmasone> "perl stuff"
[21:53:19] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "x86-zgb-binary" from 367855d89bdaed9dc019c2e176e7222aa8a2a4d8: Elly Jones <ellyjones at chromium dot org>, bsalomon at google dot com <bsalomon at google dot com@2bbb7eff-a529-9590-31e7-b0007b416f81>)'
[21:54:11] <davidjames> vapier: To make it interesting, some of the bots have locale set to en_US and others have it set to en_US.UTF-8, so that's why your change works on some bots but not others. (That's just the configuration on the actual machine, not anything configured in the source tree)
[21:54:25] <vapier> classy
[21:54:38] <vapier> good thing i forced en_US and en_US.UTF8 to be safe
[21:55:55] <davidjames> cmasone: Build team OKR is to update all the packages to latest versions, including Perl which is needed for automake etc
[21:56:07] <jamescook> tbroch: x86-zbg-binary looks like updater tests with a networking problem, error: [Errno 111] Connection refused
[21:57:00] <tbroch> jamescook: wondering if its crosbug.com/18679
[21:57:08] <tbroch> but can't get to tracker
[21:58:17] <vapier> who do i ping to unscrew my gerrit-int acct ? :x
[21:58:52] <jamescook> tbroch: I'm not sure it's the same
[22:00:44] <jamescook> x86-generic-full looks like SimpleTestUpdateAndVerify
[22:01:04] <tbroch> jamescook: agreed looked at log from that failure and its nothing like connection issue
[22:01:47] <crosbot> tree became 'Tree is closed (jamescook investigating -> "cbuildbot" on "x86-zgb-binary")'
[22:02:57] *** JasonO has quit IRC
[22:03:31] *** JasonO has joined #chromium-os
[22:03:32] *** JasonO has joined #chromium-os
[22:04:17] *** ambro718 has joined #chromium-os
[22:04:38] <zbehan> vapier: try maruel
[22:04:58] <gauravsh> for gerrit-int, nsylvain or anush might be able to help
[22:05:49] <zbehan> or those, yeah
[22:05:54] <vapier> thx, will do
[22:07:03] <ambro718> Hi. I've developed a wonderful programming language for control over the operating system and its services. It could be used to elegantly implement things like booting and network interface control, replacing messy and tangled shell scripts (or the upstart thing). See http://code.google.com/p/badvpn/wiki/NCD
[22:08:04] <ambro718> My sofrware would allow a lot of system configuration and event handling to be done in a central place instead of being scattered around config files and various services.
[22:08:24] <vapier> if it isnt related to ChromeOS, sounds like you're in the wrong place ?
[22:09:07] <ambro718> vapier: I'm suggesting my software, someone here might be interested.
[22:09:09] <ellyjones> it sounds like perhaps you want #nm or something
[22:09:27] <ellyjones> we have an (extremely custom) network management daemon that does various chromeos-specific things
[22:09:31] *** yoshiki_ has quit IRC
[22:10:03] *** SoulRaven has quit IRC
[22:10:42] <jamescook> I think both x86-generic-full and x86-zgb-binary are flake related to updater tests.  Filing and reopening.
[22:10:50] <jamescook> tbroch: ^^^
[22:10:51] <ambro718> ellyjones: I know. But, if you read about my software, you'd realize that it could replace a lot of hardcoded stuff Chromium-OS currently has.
[22:11:13] <ellyjones> ambro718: it looks like it would replace it with stuff hardcoded in a different language with no tool support, though
[22:11:51] <cmasone> ambro718: A working system using your stuff would be a lot more convincing than a wiki page
[22:12:01] <ellyjones> == cmasone
[22:12:11] <cmasone> ambro718: and a lot of stuff is hard-coded because we're interested in speed, not elegance of language used for configuration
[22:12:16] *** reveman has quit IRC
[22:12:38] <ambro718> cmasone: NCD is much faster than shell scripts
[22:12:44] <ellyjones> we use C
[22:12:50] *** reveman_ has joined #chromium-os
[22:13:00] <ambro718> NCD can easily be expaned with new statements by writing modules in C
[22:13:00] <crosbot> tree became 'Tree is open ("x86-zgb-binary" testUpdateAndKeepStateful -> crosbug.com/19298, x86-generic-full SimpleTestUpdateAndVerify -> crosbug.com/16591 )'
[22:13:04] <cmasone> ambro718: then you should have no problem showing us with benchmarks that your stuff blows us out of the water.
[22:13:12] *** reveman_ is now known as reveman
[22:13:17] *** GodoPPL is now known as GodoPPL|Away
[22:14:24] <ambro718> cmasone, ellyjones: I don't have a "production system" which uses NCD. I use it on my home computers, and I'm very satisfied. (I have, however, been contacted by some developer of embedded systems who will very likely use it in a product he's developing)
[22:14:32] <ellyjones> enjoy
[22:14:40] <ambro718> I was hoping someone here could realize its high potential, too
[22:14:50] <ellyjones> unless it is measurably better than our current approach in some measurable way for what we're doing, it is not interesting to us
[22:14:59] <zbehan> ambro718: isn't calling it a programming language kind of an overstatement? looking at the wiki, its scope seems to be rather limited
[22:15:03] <ellyjones> (elegance in the mind of the designer is not a measurable way)
[22:15:18] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot_master" on "TOT Pre-Flight Queue" from 6b4e8e4d926b90d49668c0bf5189e60c60b71762: gbillock at chromium dot org <gbillock at chromium dot org@0039d316-1c4b-4281-b951-d872f2087c98>)'
[22:15:20] <cmasone> ambro718: it's not even entirely clear that it meets all our needs.
[22:15:35] <ambro718> zbehan: it is a programming language (just not a general-purpose one, but a domain-specific one)
[22:15:36] <davidjames> vapier: Should locale-gen use --keep, i.e. "Don't nuke existing locales" ?
[22:16:02] <ambro718> cmasone: how could you have determined that without looking for more than 5 minutes?
[22:16:17] *** arny91x has joined #chromium-os
[22:16:23] <vapier> davidjames: probably ... but that's the current behavior with locale-archive
[22:16:50] <ambro718> cmasone: if you haven't seen it yet, its power is well demonstrated by this http://code.google.com/p/badvpn/wiki/NCD_events_example
[22:17:28] <ellyjones> very cool but what we have now does what we need and does it fast
[22:17:38] <wfrichar> anyone looking at the current breakage?
[22:17:45] <jamescook> looking
[22:17:50] <jamescook> supplied_chrome sig 11
[22:17:50] <jamescook>   suite_Smoke/desktopui_ChromeFirstRender
[22:18:02] <jamescook> SimpleTestUpdateAndVerify
[22:18:07] <vapier> well i'm fairly confident that isnt me :)
[22:18:08] <wfrichar> thanks. Might want to update the status when you start on it.
[22:18:11] <cmasone> jamescook: can one get a stack dump from there?
[22:18:22] <ambro718> ellyjones: I do believe that. But there's always place for improvement!
[22:18:37] <crosbot> tree became 'Tree is closed (jamescook investigating: "cbuildbot_master" on "TOT Pre-Flight Queue" from 6b4e8e4d926b90d49668c0bf5189e60c60b71762: gbillock at chromium dot org <gbillock at chromium dot org@0039d316-1c4b-4281-b951-d872f2087c98>)'
[22:18:46] <ellyjones> ambro718: patches accepted, but you must show a speed or other improvement numerically
[22:18:56] <zbehan> ambro718: fwiw, chromiumos is an opensource project, you can not only demonstrate its power on a real project, but even on this very project
[22:19:07] <vapier> davidjames: how come the bar at the top of gerrit.chromium.org doesnt link you to the buildbot/waterfall page ?
[22:19:31] <ambro718> I'm not saying "throw away all you've done". Just to take a good look at my software, and see if it could be used to improve Chromium-OS, in the long term.
[22:20:07] <ambro718> (obviously integrating it into Chromium-OS would be a very big task, one which I can't "just send a patch for")
[22:20:20] <ellyjones> I have put looking at your thing into my todo list
[22:20:30] <ambro718> thank you.
[22:21:05] <jamescook> cmasone: looking, not sure where we would keep a stack trace
[22:21:14] <ambro718> and my argument is not speed, it's things like flexibility, extensibility, elegance, simplicity.
[22:21:41] <cmasone> jamescook: in the archive of test results
[22:21:48] <cmasone> jamescook: I believe it's mentioned in the FAQ
[22:21:53] <jamescook> yeah, I'm digging there now
[22:21:57] <cmasone> ambro718: speed is more important to us than those things
[22:22:16] <ambro718> cmasone: right, but still, NCD is fast :)
[22:22:35] <ellyjones> is it faster than our hand-written C thing?
[22:23:15] <ambro718> ellyjones: probably not strictly faster, but I suspect you wouldn't see a difference if you used it properly, or maybe spent some time optimizing it
[22:23:20] <ambro718> (not much time)
[22:23:45] <ellyjones> ambro718: speed is more important than any of the things you mentioned for us
[22:23:48] *** JasonO has quit IRC
[22:24:03] <ellyjones> so unless it's _strictly_ faster, you are unlikely to see us put in the engineering effort, ever
[22:24:22] <cmasone> ambro718: I have your word for these things and nothing else.  I have no reason to believe that your system meets all of our operational needs, and I have no gaurantee that your system will be supported in the event that you cease to be interested in it.  That's not really a risk that we can take, as we ship real products to real people.
[22:25:49] <ambro718> cmasone: I doubt anything could happen in the near future which woule make me abandon NCD. I'm not the kind of developer who lets his good code rot :)
[22:26:21] <zbehan> ambro718: you could get run over in a random accident
[22:26:23] <cmasone> ambro718: what if you get hit by a bus
[22:26:39] <ellyjones> bus number!
[22:26:47] <ambro718> zbehan, cmasone: that would be *extermely* unlikey
[22:26:55] <ellyjones> "I live in the country!"
[22:27:11] <ambro718> also, NCD is not *that* big some else couldn't understand it and develop it further
[22:27:24] <ellyjones> no, look
[22:27:36] <ambro718> in fact, it's very small. Which results from its simplicity and good design.
[22:27:45] <ellyjones> we ship a product on tight deadlines with little engineering time to spare; you are proposing something _slower_ than what we use now that requires additional effort to maintain and integrate
[22:27:54] <ellyjones> do you see how this is entirely a nonstarter for us?
[22:28:19] <ambro718> ellyjones: I'm proposing a long-term solution which would have long-term benefits.
[22:28:35] <ambro718> (and of course would take longer to implement)
[22:28:47] <ellyjones> we can also just do nothing, and if your program actually turns out to be better than what we have now later, switch to it when it's really better
[22:29:09] <jamescook> cmasone: Do you know where in test_results I need to look?  I'm not finding anything that looks like a stack, and I don't see anything in the CrOS sheriff FAQ that suggests where to look.
[22:29:11] <ellyjones> ('better' means 'faster' here)
[22:29:13] <tbroch> gabeblack taking over for me as sheriff ... thanks Gabe
[22:29:14] <gabeblack> Hi, I'm taking over as sheriff for tbroch
[22:29:18] <gabeblack> yep
[22:29:30] <cmasone> jamescook: I just find for *.core, basically
[22:29:35] <cmasone> or *.dmp
[22:30:00] <cmasone> I mean, this is a flaky crash, right?  We can probably re-open while we figure out which it is, or if it's new
[22:30:32] <ellyjones> hmm
[22:30:39] <ellyjones> I just got scheduling while atomic on one of my tegra boards
[22:31:09] <ambro718> ellyjones: if no one is interested in my program because existing things "work fine", it's likely its potential will never be seen and it will never be used in anything. Bad luck for me, I guess...
[22:31:32] <ambro718> It's kind of a circular problem.
[22:31:38] <crosbot> tree became 'Tree is open (jamescook investigating desktopui_ChromeFirstRender sig11 in TOT PFQ)'
[22:31:49] <cmasone> ambro718: if your system really has significant maintainability advantages, I'm sure that one of the normal linux distros would be interested
[22:32:14] <ellyjones> ambro718: the correct person to sell this to is the networkmanager maintainer, who hangs out in #nm
[22:32:16] <cmasone> ambro718: they have to deal with a lot more system services than we do
[22:32:26] <ellyjones> and they have to deal with considerably more configurations than we do
[22:32:30] <ellyjones> (i.e., more than one :P)
[22:33:56] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "tegra2_kaen-binary" from 8a3ecd0786ea715fefa5b7feddbccb97e685eb03: Andrew de los Reyes <adlr at chromium dot org>)'
[22:34:13] <ambro718> cmasone: all the normal distros seem to be concentrating on the "systemd" thing and the "networkmanager" thing, and they refuse to let go. I was thinking that my software would interest someone building a special-purpose system. It would also be easier to use it in a special-purpose system.
[22:34:54] <ellyjones> anyway
[22:35:02] <ellyjones> I will look at your thing and see if it is an improvement for us
[22:35:10] <ellyjones> I don't think there is any more useful discussion to have before I do that
[22:35:13] <ambro718> just take a good look, please :) I'll shut up not.
[22:35:19] <ambro718> *now
[22:36:01] <wfrichar> darn it! I missed the window of opportunity!
[22:36:34] <cmasone> ambro718: if you're really interested in someone adopting this, you'd do well to actually put effort into demonstrating that it will work in the first place.
[22:37:08] <gabeblack> The tree just closed because tegra2_kaen-binary failed, but it doesn't look like it failed to me
[22:37:38] <gabeblack> the builder is idle and green
[22:37:43] <gabeblack> any ideas what it's upset about?
[22:38:31] *** arny91x has quit IRC
[22:38:48] <cmasone> gabeblack: enigma wrapped in mystery?
[22:39:05] *** Hexxeh has joined #chromium-os
[22:39:27] <gabeblack> cmasone: And then wrapped in a burrito
[22:39:44] <redpig> I would like a burrito :/
[22:39:55] <ellyjones> we had burritos for lunch here :P
[22:40:01] <redpig> gah
[22:40:06] <ellyjones> (I ate LD50 of guacamole)
[22:40:11] <wfrichar> We had Thai curry. Yummy.
[22:40:15] * cmasone is pleased with the turn of conversation
[22:40:24] <redpig> at least I can take comfort knowing they were "boston" burritos and pale in comparison to anything you get elsewhere :p
[22:40:30] * wfrichar is waiting for the build to turn green again..
[22:40:45] <cmasone> redpig: yeah, it was mostly cod wrapped in a red sox tshirt
[22:40:55] <wfrichar> heh.
[22:41:01] <ellyjones> gross :P
[22:41:02] <gabeblack> so... I can open the tree because it was just confused?
[22:42:28] <wfrichar> Sure. I need to commit to two repos simultaneously, so that'll probably confuse it again.
[22:43:37] <crosbot> tree became 'Tree is open (can't diagnose imaginary failures)'
[22:43:42] <wfrichar> Pow!
[22:43:45] <ellyjones> haha
[22:43:45] <wfrichar> Pow!
[22:44:32] <gabeblack> So what's the scoop with these canaries?
[22:44:46] <gabeblack> They're red, but it doesn't look like they're running
[22:45:01] <cmasone> gabeblack: canaries run twice a day, I thinlk
[22:45:09] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "stumpy-binary" from b645116853c677aca8a316381b87441ba6004f67: Andrew de los Reyes <adlr at chromium dot org>, danno at chromium dot org <danno at chromium dot org@ce2b1a6d-e550-0410-aec6-3dcde31c8c00>)'
[22:45:17] <gabeblack> oi
[22:45:40] *** sergiu has joined #chromium-os
[22:45:40] *** sergiu has joined #chromium-os
[22:45:53] <wfrichar> haha. 92 seconds! A new record!
[22:46:25] * wfrichar is writing a shell script to submit the second the tree turns green, just in case.
[22:47:23] <ellyjones> I should have crosbot do that for me
[22:48:26] <wfrichar> actually, crosbot should do that for everyone! "crosbot: submit CL 6131"
[22:48:44] <ellyjones> too bad crosbot doesn't have credentials to commit
[22:50:16] <Stepan> wfrichar: write it in C, so you safe the time of loading a shell
[22:50:17] <crosbot> tree became 'Tree is open (bugs being filed for the flakiness)'
[22:50:36] <wfrichar> heh.
[22:57:32] <davidjames> cmasone: FYI, looked at tegra2_kaen with gabeblack to explain the imaginary failure
[22:57:42] <cmasone> davidjames: what was it?
[22:57:51] <davidjames> cmasone: tegra2_kaen has two build machines, so it was actually running two builds at once
[22:57:58] * cmasone blinks
[22:58:05] <davidjames> cmasone: So build 1450 failed after builds 1451 and 1452 succeeded
[22:58:19] <davidjames> cmasone: See http://chromeos-botmaster.mtv.corp.google.com:8026/builders/tegra2_kaen-binary
[23:02:07] <jamescook> codesite folks think code.google.com is fixed, "I'm declaring at least tentative victory. We're no longer serving errors, and some abusive traffic is being handled by dosserver "
[23:06:06] <crosbot> tree became 'Tree is open (TOT PFQ -> supplied_chrome sig 11, crosbug.com/19301 )'
[23:06:17] *** rharrison_chrome has quit IRC
[23:09:27] <jamescook> gabeblack: stumpy failure looks like hang in git, I've seen a few of those over the last day
[23:09:44] <jamescook> nothing to do for us there, bug already filed and being looked at
[23:14:22] <jamescook> brb
[23:19:52] *** achuith has quit IRC
[23:21:02] *** oc80z has quit IRC
[23:21:19] *** Hexxeh has quit IRC
[23:21:34] *** dots has quit IRC
[23:21:52] *** Hexxeh has joined #chromium-os
[23:21:53] *** dots has joined #chromium-os
[23:22:26] *** oc80z has joined #chromium-os
[23:24:22] *** achuith has joined #chromium-os
[23:28:43] *** achuith has quit IRC
[23:33:31] *** Daxvex1 has quit IRC
[23:36:32] *** Daxvex has joined #chromium-os
[23:40:00] *** Daxvex1 has joined #chromium-os
[23:42:46] *** Daxvex has quit IRC
[23:43:26] *** ambro718 has quit IRC
[23:44:45] *** Daxvex1 has quit IRC
[23:45:20] *** Daxvex has joined #chromium-os
[23:45:23] *** ambro718 has joined #chromium-os
[23:50:41] *** cros_ has quit IRC





top