Closed Bug 1054411 Opened 5 years ago Closed 5 years ago

cache2 automation: browser_keywordSearch.js tries to connect to google.com with cache2 enabled

Categories

(Firefox :: Search, defect)

defect
Not set
Points:
1

Tracking

()

RESOLVED FIXED
Firefox 34
Iteration:
34.2
Tracking Status
firefox32 --- fixed
firefox33 --- fixed
firefox34 --- fixed
firefox-esr24 --- unaffected
firefox-esr31 --- fixed

People

(Reporter: jduell.mcbugs, Assigned: MattN)

References

Details

Attachments

(1 file)

See 'bc1' failures in this try run:

  https://tbpl.mozilla.org/?tree=Try&rev=a1ea7549c4fd

I glanced at the code but see nothing obvious.  The test puts some text in the URI bar and then focuses it.  I think we may need some help from the Firefox team to quickly understand how that winds up interacting with the cache in some way that doesn't cause external network traffic in cache1 but does in cache2.
Matt, looks like you've touched this test (so did Mike de Boer but he's out until Sept 15).  Any idea how this test interacts with the HTTP cache, and why using the new cache2 causes the test to try to contact google.com?

This is blocking the new HTTP cache from landing on release, so fairly urgent.
Flags: needinfo?(MattN+bmo)
It looks like the HTTP request isn't being cancelled like related tests do. I can't repro locally on my opt build so I'm testing in my debug build now.
Assignee: nobody → MattN+bmo
Status: NEW → ASSIGNED
Flags: needinfo?(MattN+bmo)
If I switch executeSoon(nextTest); to setTimeout(nextTest, 2000); the tests fail locally. The cleanup of the tab happens so quickly on my machine that it doesn't fail normally.

This should be a one line patch… coming up
I checked that the other forks of this test already were doing req.cancel but this file didn't get the enhancement.
https://mxr.mozilla.org/mozilla-central/search?string=info%28%22Actual%20URI%3A

Jason, if you're okay with this then I'll land it.

I don't know why the new cache affects this but it seems like this patch is the right thing to do regardless. You may want to dig deeper into the different cache behaviour.
Attachment #8473826 - Flags: review+
Attachment #8473826 - Flags: feedback?(jduell.mcbugs)
Iteration: --- → 34.2
Points: --- → 1
Component: Networking: Cache → Search
Flags: firefox-backlog+
Product: Core → Firefox
QA Whiteboard: [qa?]
QA Whiteboard: [qa?] → [qa-]
Comment on attachment 8473826 [details] [diff] [review]
Cancel the HTTP requests. r=adw (in-person)

Review of attachment 8473826 [details] [diff] [review]:
-----------------------------------------------------------------

Thanks for the quick turnaround. If you're happy with this, I'm good with it.
Attachment #8473826 - Flags: feedback?(jduell.mcbugs) → feedback+
https://hg.mozilla.org/mozilla-central/rev/563f7421eeaa
Status: ASSIGNED → RESOLVED
Closed: 5 years ago
Flags: in-testsuite+
Resolution: --- → FIXED
Whiteboard: [fixed-in-fx-team]
Target Milestone: --- → Firefox 34
You need to log in before you can comment on or make changes to this bug.