1326401 - failure to run toolkit/components/extensions/test/xpcshell/test_ext_native_messaging_perf.js on ubuntu 16.04

Status: RESOLVED FIXED

(WebExtensions :: Untriaged, defect, P2)

Description

[Joel Maher ( :jmaher ) (UTC -8)]

our linux tests run in ubuntu 12.04- we want to update these and we have some running in 16.04 already!  Unfortunately some tests need some attention, luckily not too many.

it appears we are getting RTT of 20.25 (opt) / 50.4 (debug) instead of what is specified of 36/18:
https://dxr.mozilla.org/mozilla-central/source/toolkit/components/extensions/test/xpcshell/test_ext_native_messaging_perf.js?q=path%3Atest_ext_native_messaging_perf.js&redirect_type=single#12

:aswan, can you take a look at this and see if we need to increase our max rtt, or adjust something on our 16.04 image?
here is a log showing the failure:
https://public-artifacts.taskcluster.net/Be5VxoFrTQSMW6IDq-QPpw/0/public/logs/live_backing.log

and a try run:
https://treeherder.mozilla.org/#/jobs?repo=try&revision=442aa421aa40a69a124413b6546c486210b728c1&selectedJob=65308782&group_state=expanded&filter-searchStr=xpcshell  (look at xpcshell 6 (opt) and 7 (debug))

Comment 1

[Kris Maglione [:kmag]]

We should probably just increase the timeout. I'd rather not, but I can't think of any changes to the image that would improve things. It would be nice to know why the upgrade decreased the performance, though...

Comment 2

[:shell escalante]

nice if could double time on linux

Comment 3

[Andrew Swan [:aswan]]

Attachment: Bug 1326401 Bump up acceptable range for subprocess perf test

Review commit: https://reviewboard.mozilla.org/r/103986/diff/#index_header
See other reviews: https://reviewboard.mozilla.org/r/103986/

Comment 4

[Kris Maglione [:kmag]]

Comment on attachment 8825930 [details]
Bug 1326401 Bump up acceptable range for subprocess perf test

https://reviewboard.mozilla.org/r/103986/#review104680

Comment 5

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/001929d30497
Bump up acceptable range for subprocess perf test r=kmag

Comment 6

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/mozilla-central/rev/001929d30497