If you think a bug might affect users in the 57 release, please set the correct tracking and status flags for Release Management.

[SeaMonkey 2.1] mochitest-plain-4: "leaked 360 bytes = 5 instances of jsdScript + 10 instances of nsStringBuffer"

RESOLVED FIXED in mozilla1.9.3a5

Status

Other Applications
Venkman JS Debugger
--
major
RESOLVED FIXED
8 years ago
7 years ago

People

(Reporter: sgautherie, Unassigned)

Tracking

(Blocks: 1 bug, {mlk})

Trunk
mozilla1.9.3a5
Dependency tree / graph

Firefox Tracking Flags

(Not tracked)

Details

(Whiteboard: [fixed by the "new add-ons manager UI"], URL)

(Reporter)

Description

8 years ago
Example:
{
http://tinderbox.mozilla.org/showlog.cgi?log=SeaMonkey/1270436938.1270438962.15113.gz&fulltext=1
OS X 10.5 comm-central-trunk debug test mochitests-4/5 on 2010/04/04 20:08:58

TEST-UNEXPECTED-FAIL | automationutils.processLeakLog() | leaked 360 bytes during test execution
TEST-UNEXPECTED-FAIL | automationutils.processLeakLog() | leaked 5 instances of jsdScript with size 56 bytes each (280 bytes total)
TEST-UNEXPECTED-FAIL | automationutils.processLeakLog() | leaked 10 instances of nsStringBuffer with size 8 bytes each (80 bytes total)
}

***

KaiRo, ftb, we should use a bbc config.py leak_threshold for this case.

Comment 1

8 years ago
(In reply to comment #0)
> KaiRo, ftb, we should use a bbc config.py leak_threshold for this case.

I don't see how we should need one when Firefox doesn't. We first should try to find out where the leak comes from, as it's not a systemic leak in all mochitests, it is probably confined to a certain test.

Comment 2

8 years ago
And, BTW, I see no evidence that this is a venkman thing, all those leaked objects are core code (jsd is part of the core and has a component there, IIRC).
(Reporter)

Comment 3

8 years ago
(In reply to comment #1)

> I don't see how we should need one when Firefox doesn't.

Ftb, I'm _guessing_ Firefox doesn't because it is not running Venkman.

> We first should try to find out where the leak comes from,
> as it's not a systemic leak in all mochitests,
> it is probably confined to a certain test.

But who do you think "we" is in practice?
I seem to be the (only) one most interested in test failures/leaks but I can't run debug mochitests atm.

(In reply to comment #2)
> And, BTW, I see no evidence that this is a venkman thing,

I concur, but I was looking for evidence that it's not related to Venkman first.

*****

Anyway, this leak went away with the "new add-ons manager UI" landing:
http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=b826d7ba5c45&tochange=45a1894f2f9a
though I don't know what exactly fixed it.
Status: NEW → RESOLVED
Last Resolved: 8 years ago
Resolution: --- → WORKSFORME
Whiteboard: [fixed by the "new add-ons manager UI"]
Target Milestone: --- → mozilla1.9.3a5
(Reporter)

Comment 4

8 years ago
V.WorksForMe per 3-platform SeaMonkey tinderboxes.
Status: RESOLVED → VERIFIED
(Reporter)

Comment 5

8 years ago
Leaks again after bug 562679 backout.
Status: VERIFIED → REOPENED
Depends on: 562679
Resolution: WORKSFORME → ---

Comment 6

7 years ago
Fixed with bug 562679.
Status: REOPENED → RESOLVED
Last Resolved: 8 years ago7 years ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.