mozilla :: #releng

8 Aug 2017
02:01travis-cibuild-tools#1901 (master - 54a9539 : ffxbld): The build passed. (
02:09travis-cibuild-tools#1902 (master - da9b64a : ffxbld): The build passed. (
05:33travis-cibuild-tools#1903 (master - 5b29c9d : ffxbld): The build passed. (
06:50travis-cibuild-tools#1904 (master - b9d28fc : ffxbld): The build passed. (
10:01travis-cibuild-puppet#1734 (master - 77d9e88 : Andrei Obreja): The build passed. (
10:06aobreja|builddutynick aobreja|lunch
10:09travis-cibuild-puppet#1735 (production - 637a306 : Andrei Obreja): The build passed. (
10:39nthomasthanks for that aobreja
10:59aobreja|builddutynthomas: yw
11:23igoldando we have sccache enabled on beta, right now?
12:01travis-cibuild-tools#1905 (master - d835b7a : Johan Lorenzo): The build passed. (
12:04WG9s!seen RyanVM
12:04firebotryanvm was last seen 11 hours and 19 minutes ago, saying 'jesup|laptop: 'twas inevitable - bug 1247628 was going to be the death of CZ anyway' in #developers.
13:26Tomcat|sheriffdutyCallek: ping
13:27Tomcat|sheriffdutyCallek: (hi), do you know if the update failures on beta are expected ?
13:45travis-cibuild-puppet#1736 (master - 5f0254b : Dave House): The build passed. (
13:57CallekTomcat|sheriffduty: I do not think they were expected, but I also expect they are related to whatever release is going on, and then mtabara and jlorenzo would likely have the most context
13:57Tomcat|sheriffdutyah ok thanks
14:01RyanVMyeah, those are being investigated
14:01jlorenzoI confirm they are
14:04travis-cibuild-puppet#1737 (production - bdd01ad : Dave House): The build passed. (
14:05Tomcat|sheriffdutyjlorenzo: is there a bug for this failures or shall i file one ?
14:19mtabaraTomcat|sheriffduty: bug 1388312
14:19firebot REOPENED, 56.0b1 update tests fail: MAR_CHANNEL_MISMATCH_ERROR
14:19Tomcat|sheriffdutythanks mtabara
14:55jhorakHi, I have problem with pushing to the try server. It used to work when I used mq some months ago but not now. Is someone here who could help?
14:58Callekjhorak: few steps, check if you can `ssh` to be sure that you have access ;-)
14:59Callekjhorak: then make sure you have a newer (at least 3.6 iirc) hg, for extra sanity. lastly `hg push try -f` should work, but you can also use `./mach try` (see `./mach try --help` for more information)
14:59jhorakCallek: getting: ssh 255
14:59jhorakWarning: Permanently added the ED25519 host key for IP address '' to the list of known hosts.
14:59jhorakPermission denied (publickey).
14:59Callekjhorak: then you don't have commit access right now (possibly disabled due to inactivity, or badly setup ssh config, etc)
15:00CallekI'd suggest filing a bug about it (I don't have any better knowledge here)
15:00jhorakCallek: well I can do the try run from mozreview, but that's not always the best way
15:01travis-cibuild-buildbot-configs#2802 (master - 35974d8 : Joel Maher): The build passed. (
15:01jhorakokay, I'll double check the ssh keys, is there a place where can I manage them on mozilla side.
15:21jmaherkmoir: did you attach the wrong diff to a second time? or more likely my patch is just bad
15:21firebotBug 1383789 ASSIGNED, Enable Stylo Talos tests on more desktop platforms
15:22kmoirlet me verify
15:22jlorenzorstrong: ping
15:25travis-cibuild-puppet#1739 (production - 11f27cb : Dave House): The build passed. (
15:25travis-cibuild-puppet#1738 (master - a297263 : Dave House): The build passed. (
15:35kmoirno the patch and diff look correct, let me see what might be wrong
15:35kmoirbuilder diff looks correct I mean
16:33rstrongjlorenzo: pong
16:35jlorenzorstrong: unping :) bhearsum helped us to figure out what was our signature problem
16:44jmahercatlee: I have had a lot of trouble getting windows10 tests to show up on try despite hacking; for example this try push only did debug for windows 10 (as many other try pushes have)
16:46RyanVMi see them if I show hidden jobs
16:46jmahercatlee: I also have trouble getting reftest to run for win10 at all on try
16:46jmaherRyanVM: oh
16:47jmaherI didn't think about that
16:47jmaherthanks RyanVM
16:47RyanVMyw :)
16:48jmaherand now I see reftest as well
16:48RyanVMI think they got hidden along the way because they were causing too much confusion for devs
17:00hwinefubar: ping - did you see my PM from yesterday?
17:01fubarhwine: don't think so
17:02jryansjmaher: wait, did someone already change the preseed file for stylo? or it's okay to land the tests without that?
17:03jmaherjryans: I did the preseed work and it will get updated when treeherder deploys a new update in ~3 hours
17:04jmaherjryans: I think I had it on a different bug
17:05jryansjmaher: so i see windows10-64-stylo, but there is also windows7-32-stylo... does that matter?
17:05jryansjmaher: (basically, do we need to also add windows7-32-stylo?)
17:10jmaherjryans: oh, good find
17:11jmaherjryans: I can fix that soon, after my meeting
17:11jryansjmaher: ok, i should wait until it's done?
17:11jmaherjryans: or submit a pr for the win7 bits yourself?
17:12jryansjmaher: sure, i'll try it. who reviews?
17:12jmaherwlach, myself, or emorley
17:47jmaherjryans: ok, merged in, I think we will be good to go :)
17:47jryansjmaher: great, thanks!
17:48jryansjmaher: should the bug be resolved now, or do you wait for the treeherder deploy?
17:53jmaherjryans: we will wait for the deploy and verification on trees
17:53jryansjmaher: okay! i'll proceed with landing the new tests
18:08armenzgkmoir: if I send an email to the release mailing lists; do SoftVision members doing buildduty also receive it?
18:09kmoiryes they are on
18:25lizzardlots also on release-drivers list
18:47travis-cibuild-tools#1906 (master - 0d1d9aa : Mihai Tabara): The build passed. (
19:35travis-cibuild-tools#1907 (master - 1a41925 : ffxbld): The build passed. (
19:41travis-cibuild-tools#1908 (master - 7495354 : ffxbld): The build passed. (
19:49travis-cibuild-puppet#1740 (master - 45db138 : Aki Sasaki): The build passed. (
19:53travis-cibuild-puppet#1741 (production - 8e8ca34 : Aki Sasaki): The build passed. (
19:57jmaherCallek: thanks for the feedback on the mochitest-plain->win10 patch; should I add a clause for win10-devedition as well?
19:57Callekjmaher: no need unless you're doing a windows10-.* thing, because the default is 'built-projects'
19:57jmaherI really don't understand all of that stuff
19:59Callekjmaher: I presume win10 mochitests running anywhere is the desired goal, which means you're free to not do any magic for deved
19:59jmaherok, yeah, the goal is to treat it as win7/linux/osx, i.e. default
20:01travis-cibuild-tools#1909 (master - a488c7c : Mihai Tabara): The build passed. (
20:45travis-cibuild-tools#1910 (master - 6448c51 : ffxbld): The build passed. (
20:58MattNHello, did test-windows10-64-vm machines get replaced by test-windows10-64 ones or what's the story of them?
22:03hwineanyone have a bug number for the new Authenticode signing certificate deployed today?
22:03hwineit's being used in Firefox+Setup+52.3.0esr.exe (at least)
22:07hwine!squirrel ^^
22:07hwine(in case there is no such bug)
22:08akithere was one deployed a while back i think
22:08hwinewell, it got used for the builds today
22:08hwineyeah - I have a new signature for a few weeks ago -- this new one isn't that
22:09* hwine checks what other path names
22:13akii'm not finding the bug
22:14akihwine: ?
22:14firebotBug 1366012 is not accessible
22:16hwineyeah - I knew about that one -- maybe I missed an add
22:16hwinebut I think I squirreled too early
22:18hwinesorry about that -- let me dig deeper
22:18akihwine: ?
22:18firebotBug 1320773 FIXED, Update stub installer build to include dummy certificate for Stub Attribution data
22:18akifor sha2signcodestub
22:23hwinethanks -- shouldn't be that long ago, but the signature isn't listed there
22:24nthomassounds like you're talking about the full installer anyway
22:27hwinedo we have a list of older authenticode cert signatures anywhere?
22:30aki ?
22:30akinot sure
22:32hwineaki: thanks -- shouldn't be that one either -- let me do some more digging before I waste more of your time
22:59glandiumaki: can you disable the hg extension that comments on bugzilla when you merge m-c to date?
23:00akishould the extension hardcode repos?
23:01glandiumI don't know
23:01glandiumactually, do you land things on date that are not merges?
23:02glandiumand do you care that there's a comment in bugzilla?
23:02akithe merge comments are probably a larger downside than the single push comment upside
23:03hwineaki: nthomas -- communication error on my end -- no real issue re signatures
23:03akihwine: good to hear
23:05glandiumaki: if you want comments in bugzilla, I can make pulsebot do that. And it won't comment for merges
23:06akiglandium: is automatically posting for me
23:06akii think i got that from `./mach mercurial-setup`
23:06akithis sounds like a bzpost bug
23:07glandiumgps: should we kill that extension, btw?
23:09gpswe already don't prompt for it in mercurial-setup
23:10gpssome people like it for try
23:10gpsi'm fine with changing the default behavior to only work on try
23:11gpsor making it no-op by default and requiring a config option to get today's behavior
23:11glandiumgps: I can have pulsebot do that too...
23:11glandiumif we really want try pushes to hit bugs
23:11gpssome people have strong opinions on that
23:11akiok, disabled in my .hgrc, so this specific issue should be resolved, i think
23:11glandiumpersonally, I hate it
23:11akisorry for the noise
23:12glandiumand I hate it when people do it
23:12gpsthrough the lens of "the bug tracks progress on something" then notifying the bug when a try push occurs is a good thing
23:12glandiumso my preference would be for that extension to die
23:13glandiumif we want try to produce comments on bugzilla, that should be a general thing, not one-offs on whoever happens to use an extension
23:13gpsfile a bug. needinfo a BMO admin to run a query to see how much it is actually used
23:13gpsi agree
9 Aug 2017
No messages
Last message: 13 days and 6 hours ago