Closed Bug 794918 Opened 13 years ago Closed 13 years ago

Higher endurance results on Mac OS X 10.7.4 and 10.8 on Aurora branch since Sept 18

Categories

(Mozilla QA Graveyard :: Mozmill Tests, defect)

Product:
Mozilla QA Graveyard
Component:
Mozmill Tests
Platform:
x86_64
macOS
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

(Not tracked)

Status:
RESOLVED INCOMPLETE

People

(Reporter: mihaelav, Unassigned)

Details

There is a memory increase discovered by endurance tests since Sept 18 on Aurora branch. Charts: http://mozmill-ci.blargon7.com/#/endurance/charts?branch=17.0&platform=Mac&from=2012-09-10&to=2012-09-27 From reports: * 10.7.4: from Resident memory (MB): Minimum: 107 / Maximum: 628 / Average: 184 to Resident memory (MB): Minimum: 112 / Maximum: 688 / Average: 201 * 10.8 from Explicit memory (MB): Maximum: 301 / Average: 107 Resident memory (MB): Maximum: 695 / Average: 198 to Explicit memory (MB): Maximum: 380 / Average: 117 Resident memory (MB): Maximum: 759 / Average: 212 Pushlog: http://hg.mozilla.org/releases/mozilla-aurora/pushloghtml?fromchange=88b15f553474&tochange=7676a9a06403
Please include links to reports and build IDs when quoting values. Also, please include both explicit and resident values.
Looking at the following two reports for Mac OS 10.8 the test that has regressed is clearly testOpenAndCloseMultipleWindows. http://mozmill-ci.blargon7.com/#/endurance/report/671677a5d9d5ca25f3cf5ae1c4cdd334 http://mozmill-ci.blargon7.com/#/endurance/report/671677a5d9d5ca25f3cf5ae1c4d95c65 We're already working on improving the results for this test so it allows for memory to be released within each iteration (see bug 711360 and bug 788531), however it's worth looking into this. Nothing jumps out at me from the pushlog, so I'll attempt to replicate by running on qa-mb, and will attempt to narrow down using the tinderbox builds.
Unfortunately I can only run against Mac OS 10.6 on qa-mb. Would someone please try to replicate this issue locally? Something that might be likely skewing the results is that one test was skipped from September 19, so this will affect the averages.
no further updates on this bug in the last 9 months. Closing out as incomplete.
Status: NEW → RESOLVED
Closed: 13 years ago
Resolution: --- → INCOMPLETE
Product: Mozilla QA → Mozilla QA Graveyard
You need to log in before you can comment on or make changes to this bug.