mozilla :: #automation

18 Apr 2017
06:24whimboovladikoff: well, that's a startup crash
06:24whimbooyou should setup MINIDUMP_STACKWALK to get the minidump analyzed
07:46whimbooFlorinMezei: good morning. looks like the last update tests have lots and lots of failures
07:46FlorinMezeiwhimboo: good morning... yes
07:46whimboothe next beta should be much better
07:46whimboowell, for Windows and osx
07:48FlorinMezeihm OSX hasn't had problems... glad to hear about Windows :)
07:48whimbooywah the watershed failures
07:48whimboomight have been a windows only issue
08:37mhaouechwhimboo: Hello :)
08:46whimboomhaouech: good morning!
08:48mhaouechwhimboo: Sorry I have been really busy at school lately
08:48mhaouechBut I made some changes to the code
08:48mhaouechI'm not sure if that's exactly what you wanted
08:50whimboomhaouech: no worries. school is certainly more important
08:50whimboomhaouech: if you want some feedback the best is to ask me on the PR
08:54mhaouechwhimboo: Okay
08:54whimbooi was out the last days too, so i have to check a lot of emails right now
08:54mhaouechyou talked about this command "mozdownload -t daily --extension checksums" but I don't quite understand what you mean
09:01mhaouechwhimboo: Tell me if there is something that needs to change. I am not in a hurry
09:02whimboomhaouech: please add the questions to the appropriate comment
09:02whimbooso we have a flow on github
09:03mhaouechwhimboo: Ok I'll do that
09:14whimboogreat. thanks
10:14travis-cimozilla/mozmill-ci#1066 (master - 70eab18 : Henrik Skupin): The build passed.
10:14travis-ciChange view :
10:14travis-ciBuild details :
10:34FlorinMezeiwhimboo: I took offline due to repeated failures
10:34FlorinMezeithe logs seem to show that the tests run and pass, so maybe some other problem
10:34whimbooFlorinMezei: yeah. its the machine which is dying
10:34FlorinMezeioh, right :)
10:35FlorinMezeiforgot about that one
10:35FlorinMezeishould I just keep it offline?
10:35whimboobug 1344635
10:35firebot REOPENED, Unable to connect to node mm-osx-109-2 in due to HDD failures
10:35whimboomight be the best
12:16whimboodavehunt: hey. I wonder if you had the time to check for click+pageload in one of your projects yet
13:48whimboostandups: just landed click+pageload support in Marionette (bug 1335778). Lets see what's next.
13:48standupsOk, submitted #44894 for
13:48firebot ASSIGNED, Make element#click command check for page load and wait
15:42raajitr@whimboo need more info reagrding bug
15:42firebotBug 1357372 NEW, Restrict safebrowsing tests from reaching
15:44whimbooraajitr: hi and welcome to our channel!
15:44whimbooand for sure your interest to work on this bug
15:45whimbooso are you on latest mozilla-central?
15:48whimbooand you have a local build?
15:51whimbook, i assume you have a slower internet connection maybe?
15:51whimboodo you only see it for teh intial download test?
15:51whimbooif yes, just skip it
15:52whimboono changes should be necessary tehre
15:52raajitrif you think 40mbps is slower then yes its slower internet connection
15:52raajitri'm getting an error at
15:53raajitrand failing at
15:55whimbooraajitr: access location bar most likely fails because you focus another application while the test is running?
15:55raajitryes my terminal was at focus
15:56whimbooyeah leaving teh focus will make this test fial
15:56whimbooand others under the same folder
15:57whimboobut well, just concentrate here on security/test_safe_browsing
15:57whimbooand don't obey initial download
16:37mkaplyerahm: You around?
16:46erahmmkaply: yep
16:47mkaplyerahm: So I'm trying to figure out how to debug 1357219 since I can't recreate locally. Should I create a patch and push it to a try server? And just run against try (What I need is for that error to show me the values for the two parameters that are failing)
16:48mkaplyerahm: I'm wondering what is the least amount of tests I can run on try to see the problem.
16:48erahmmkaply: one sec I'll check, are you testing locally with debug builds?
16:49mkaplyerahm: Doh. It defaults to release doesn't it. Let me rebuild
16:49erahmmkaply: I think |mach test browser/components/search/test/browser_yahoo_behavior.js| is your best bet
16:49mkaplyerahm: OK, cool. Rebuilding debug now. I'll let you know
16:50erahmmkaply: thanks for looking into it!
17:41mkaplyerahm: Sadly with a debug build, I'm not getting those particular errors.
17:41erahmmkaply: what platform are you testing?
17:42mkaplyerahm: Mac
17:42mkaplyerahm: Lots of WARNING: 'NS_FAILED(rv)', file /Users/michaelkaply/Projects/mozilla-central/caps/BasePrincipal.cpp, line 380
17:42mkaplyDo we have a bug for that?
17:42erahmmkaply: yeaaaah
17:42firebotBug 1357208 NEW, 637,100 instances of "'NS_FAILED(rv)'" emitted from caps/BasePrincipal.cpp during linux64 and win32
17:43erahmI turn my back for like one months and we get some serious log spam :(
17:44mkaplyerahm: And of course the fix is to turn off the warning. Either it's a problem or it's not.
17:44mkaplyLet me pull that fix in to clean up my log a little
18:05mkaplyerahm: Can I view the mozconfig for those builds? It suddenly occurs to me that we don't normally do jar packaging on default builds, but those are jar errors. Maybe I'm missing a config?
18:06erahmmkaply: it's whatever automation uses...I'll see if I can dig it up
18:07erahmmkaply: from the build log
18:08erahm(that's linux)
18:09erahmmkaply: although can't you just do |mach package| ?
18:23mkaplyerahm: tests aren't run against the packaged version, are they?
18:24mkaplyenable debug and enable release.
18:32erahmmkaply: I guess release means like thing we package and send out (vs opt for optimized)
18:32mkaplyerahm: Probably. /building that now
20:32mkaplyI can't figure out what causes a release build to get jar packaging anymore.
20:33mkaply./mach package creates them, but then tests don't use them I think?
21:49whimboomaja_zf: hi. are you still around?
21:49maja_zfwhimboo: yep, 2ish more hours
21:50whimboomaja_zf: would you mind reviewing a small fx ui patch which might fix our most occurring update failure?
21:50whimbooi would like to have it in the next nightly
21:50maja_zfwhimboo: happy to
21:50whimbook, then i'm pushing it in a minute
21:57whimboomaja_zf: its up
22:00maja_zfwhimboo: lgtm
22:01whimboomaja_zf: thanks a lot. going to push and then heading to bed
22:01whimboomaja_zf: see you tomorrow!
22:01maja_zfwhimboo: see you, have a good night
22:01maja_zfyeesh, it's late! hope you get enough rest :)
22:03whimboomaja_zf: yeah. midnight was my deadline
22:03whimbooso 3 mins over
22:03maja_zfgo go go!
22:03whimboothanks for teh quick review!
22:58maja_zfstandups: All caught up with mail. Pushed up some work I had left-over from last week: (1) checking what needs to be fixed to get rid of legacyactions.js (A: a couple of tests need rewriting using new Actions API, nothing major)
22:58standupsOk, submitted #44924 for
22:58maja_zfstandups: (2) testing drag-and-drop for mouse actions
22:58standupsOk, submitted #44925 for
23:00maja_zfstandups: tomorrow I'll have to wrap up one remaining code-coverage task: comparison of e10s versus non-e10s results. Along the way, I'll document how to run the lcov toolchain less painfully.
23:00standupsOk, submitted #44926 for
19 Apr 2017
No messages
Last message: 10 days and 3 hours ago