645707 - Test failure in testOpenSearchAutodiscovery.js | URL mismatch

Status: RESOLVED WONTFIX

(Mozilla QA Graveyard :: Mozmill Tests, defect)

Description

[Aaron Train [:aaronmt]]

Module: testOpenSearchAutodiscovery.js
Test: testOpenSearchAutodiscovery
Error: Current URL contains the search term - got 'http://hg.mozilla.org/', expected 'firefox'
Platform: Linux Ubuntu 10.10
Branch: 3.6 (1.9.2) and 3.5 (1.9.1)

http://mozmill-release.brasstacks.mozilla.com/#/general/failure?branch=All&platform=All&from=2011-03-01&to=2011-03-28&test=%2FtestSearch%2FtestOpenSearchAutodiscovery.js&func=testOpenSearchAutodiscovery

Comment 1

[Maniac Vlad Florin (:vladmaniac)]

(In reply to Aaron Train [:aaronmt] from comment #0)
> Module: testOpenSearchAutodiscovery.js
> Test: testOpenSearchAutodiscovery
> Error: Current URL contains the search term - got 'http://hg.mozilla.org/',
> expected 'firefox'
> Platform: Linux Ubuntu 10.10
> Branch: 3.6 (1.9.2) and 3.5 (1.9.1)
> 
> http://mozmill-release.brasstacks.mozilla.com/#/general/
> failure?branch=All&platform=All&from=2011-03-01&to=2011-03-
> 28&test=%2FtestSearch%2FtestOpenSearchAutodiscovery.
> js&func=testOpenSearchAutodiscovery

I get no report when clicking on the above link. Can we get a working one?

Comment 2

[Henrik Skupin [:whimboo][⌚️UTC+1]]

This only happens on 3.6 now. It's nothing we should focus our work on. I would just say, lets get this test skipped on Linux on mozilla-1.9.2.

Comment 3

[Maniac Vlad Florin (:vladmaniac)]

Attachment: skip patch v1.0 for linux [checked-in]

initial patch for skipping the test on branch mozilla-1.9.2

Comment 4

[Henrik Skupin [:whimboo][⌚️UTC+1]]

Comment on attachment 553728 [details] [diff] [review]
skip patch v1.0 for linux [checked-in]

Sorry, have to r- because it will disable the test on all platforms and not only on Linux.

Comment 5

[Maniac Vlad Florin (:vladmaniac)]

Attachment: skip patch v1.1 for linux

Does this fix make any sense?

Comment 6

[Henrik Skupin [:whimboo][⌚️UTC+1]]

Comment on attachment 554060 [details] [diff] [review]
skip patch v1.1 for linux

No, there would be mozmill.isLinux to check for the platform. But I thought more of it and if we disable the test only on Linux, how can we declare that on Litmus? I don't want that we synch results if that single platform is only disabled.

Anthony, what would you propose? I would even be eager to say, lets skip the test on all platforms and remove the 'covered by Mozmill' flag in Litmus entirely. In such a case we can use the first patch from Vlad.

Comment 7

[u279076]

Yeah, I agree. Just disable this test outright and update the flag in Litmus.

Comment 8

[Maniac Vlad Florin (:vladmaniac)]

Then asking for r? again for patch v1.0

Comment 9

[Henrik Skupin [:whimboo][⌚️UTC+1]]

Comment on attachment 553728 [details] [diff] [review]
skip patch v1.0 for linux [checked-in]

Anthony, please check-in the patch.

Comment 10

[u279076]

Comment on attachment 553728 [details] [diff] [review]
skip patch v1.0 for linux [checked-in]

Landed:
http://hg.mozilla.org/qa/mozmill-tests/rev/517215990362 (mozilla-1.9.2)

Comment 11

[u279076]

Accidentally marked this resolved fixed -- reverting that change.

Comment 12

[Henrik Skupin [:whimboo][⌚️UTC+1]]

Well, as mentioned above just mark it as WONTFIX and leave the test disabled. Update the Litmus test and we are fine. We should not spend any of our valuable time on getting a test fixed on 1.9.2.

Comment 13

[u279076]

Not sure if I properly disabled this in Litmus -- I left it enabled due it being in the 3.6 BFT and 3.6 FFT testruns as well but removed it from the 3.6 Mozmill testrun. I also added a comment to the test:

DISABLED DUE TO TEST FAILURE (bug 645707)
This test is covered by Mozmill:
testSearch/testOpenSearchAutodiscovery.js

Comment 14

[Henrik Skupin [:whimboo][⌚️UTC+1]]

(In reply to Anthony Hughes, Mozilla QA (irc: ashughes) from comment #13)
> DISABLED DUE TO TEST FAILURE (bug 645707)

That could cause false impressions for testers so they probably assign that statement to the Litmus test itself. I have updated the wording to "This test has been disabled in Mozmill". IMO we should take that path.

Lets close this bug as wontfix.