1394081 - TEST-UNEXPECTED-FAIL | C:\slave\test\build\tests\mozmill\cloudfile\test-cloudfile-backend-hightail.js

Status: RESOLVED FIXED

(Thunderbird :: General, defect)

Description

[Jorg K (CEST = GMT+2)]

Attachment: Extract from the log showing failures

TEST-UNEXPECTED-FAIL | C:\slave\test\build\tests\mozmill\cloudfile\test-cloudfile-backend-hightail.js | test-cloudfile-backend-hightail.js::teardownTest
TEST-UNEXPECTED-FAIL | C:\slave\test\build\tests\mozmill\cloudfile\test-cloudfile-backend-hightail.js | test-cloudfile-backend-hightail.js::test_simple_case
TEST-UNEXPECTED-FAIL | C:\slave\test\build\tests\mozmill\cloudfile\test-cloudfile-backend-hightail.js | test-cloudfile-backend-hightail.js::teardownTest
TEST-UNEXPECTED-FAIL | C:\slave\test\build\tests\mozmill\cloudfile\test-cloudfile-backend-hightail.js | test-cloudfile-backend-hightail.js::test_chained_uploads
TEST-UNEXPECTED-FAIL | C:\slave\test\build\tests\mozmill\cloudfile\test-cloudfile-backend-hightail.js | test-cloudfile-backend-hightail.js::teardownTest
TEST-UNEXPECTED-FAIL | C:\slave\test\build\tests\mozmill\cloudfile\test-cloudfile-backend-hightail.js | test-cloudfile-backend-hightail.js::test_deleting_uploads
TEST-UNEXPECTED-FAIL | C:\slave\test\build\tests\mozmill\cloudfile\test-cloudfile-backend-hightail.js | test-cloudfile-backend-hightail.js::teardownTest
TEST-UNEXPECTED-FAIL | C:\slave\test\build\tests\mozmill\cloudfile\test-cloudfile-backend-hightail.js | test-cloudfile-backend-hightail.js::test_can_cancel_upload
TEST-UNEXPECTED-FAIL | C:\slave\test\build\tests\mozmill\cloudfile\test-cloudfile-backend-hightail.js | test-cloudfile-backend-hightail.js::teardownTest
TEST-UNEXPECTED-FAIL | C:\slave\test\build\tests\mozmill\cloudfile\test-cloudfile-backend-hightail.js | test-cloudfile-backend-hightail.js::test_can_cancel_uploads
TEST-UNEXPECTED-FAIL | C:\slave\test\build\tests\mozmill\cloudfile\test-cloudfile-backend-hightail.js | test-cloudfile-backend-hightail.js::teardownTest
TEST-UNEXPECTED-FAIL | C:\slave\test\build\tests\mozmill\cloudfile\test-cloudfile-backend-hightail.js | test-cloudfile-backend-hightail.js::teardownTest
TEST-UNEXPECTED-FAIL | C:\slave\test\build\tests\mozmill\cloudfile\test-cloudfile-backend-hightail.js | test-cloudfile-backend-hightail.js::test_delete_refreshes_stale_token
TEST-UNEXPECTED-FAIL | C:\slave\test\build\tests\mozmill\cloudfile\test-cloudfile-backend-hightail.js | test-cloudfile-backend-hightail.js::teardownTest
TEST-UNEXPECTED-FAIL | C:\slave\test\build\tests\mozmill\cloudfile\test-cloudfile-backend-hightail.js | test-cloudfile-backend-hightail.js::test_bug771132_fix_no_scheme

First seen: Sat Aug 26, 2017, 1:39:51
https://treeherder.mozilla.org/#/jobs?repo=comm-central&revision=8d0293462cd9ccfdddd16a0a447638c6c362b35c

M-C last good: 03d7b6dd65b93afaa6981269f69e9f7cd3
M-C first bad: 86a3c73984a23b1d524eadbd4a62b9b3fa

https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=03d7b6dd65b93afaa6981269f69e9f7cd3&tochange=86a3c73984a23b1d524eadbd4a62b9b3fa

Log at
https://archive.mozilla.org/pub/thunderbird/tinderbox-builds/comm-central-win32/1503704422/comm-central_win7_ix_test-mozmill-bm126-tests1-windows-build16.txt.gz
has many timeouts, see attachment.

Aceman, not what we wanted to do this weekend. Can you lend me a hand, please.

Comment 1

[:aceman]

A candidate bug mentioning observers would be bug 1350432.

Comment 2

[Jorg K (CEST = GMT+2)]

Hmm, I read "Initial Quantum DOM scheduler implementation, disabled by default". Can you take a look at it.

BTW, how do you disable an entire Mozmill test? The whole file?

Comment 3

[:aceman]

(In reply to Jorg K (GMT+2) from comment #2)
> Hmm, I read "Initial Quantum DOM scheduler implementation, disabled by
> default". Can you take a look at it.

Yes, building now.
 
> BTW, how do you disable an entire Mozmill test? The whole file?

I'll look for that.

Comment 4

[Jorg K (CEST = GMT+2)]

Our senior bustage spotter (who shall remain nameless) said bug 1392358. Sure we send in that test, right?

Comment 5

[:aceman]

I couldn't find .send() in that test.

Comment 6

[:aceman]

(In reply to :aceman from comment #3)
> (In reply to Jorg K (GMT+2) from comment #2)
> > BTW, how do you disable an entire Mozmill test? The whole file?
> I'll look for that.

Maybe try creating a manifest file in the test folder, like here:
https://bug732353.bmoattachments.org/attachment.cgi?id=8551767
There seems to also be a possibility of a 'disabled =' line in the manifest, similarly to xpcshell.

Comment 7

[Jorg K (CEST = GMT+2)]

(In reply to :aceman from comment #6)
> Maybe try creating a manifest file in the test folder, like here: ...
Well, maybe best to fix the test instead of disabling it.

The send is in the function that does the work:
https://dxr.mozilla.org/comm-central/rev/8d0293462cd9ccfdddd16a0a447638c6c362b35c/mail/components/cloudfile/nsHightail.js#385

Bug 1392358 removed xhr.send(nsIInputStream) which is what we use here:
https://dxr.mozilla.org/comm-central/rev/8d0293462cd9ccfdddd16a0a447638c6c362b35c/mail/components/cloudfile/nsHightail.js#1054

So the test broke because the functionality broke. May this is all we need:
https://hg.mozilla.org/mozilla-central/rev/0d9988ed1c6e#l5.12

Comment 8

[:aceman]

Ok, I only looked in the various test files ;)

Comment 9

[:aceman]

Attachment: patch

Good find. The fix worked for me locally (linux).
The threeherder is down so please check on Windows locally.
I've looked for other xhr.send() uses in c-c but didn't spot any other sending in a nsIInputStream.

Comment 10

[Jorg K (CEST = GMT+2)]

The test fortunately detects that sending/uploading stopped working. If sending e-mail stopped working, would you look for the code that broke in the test? ;-) I haven't looked at the test in detail, but I believe it triggers an upload somehow. Anyway, thanks for looking into it.

Comment 11

[Jorg K (CEST = GMT+2)]

Comment on attachment 8901459 [details] [diff] [review]
patch

I've read through bug 1392358 again and the fix here appears to be right. It fixes the issue on Windows.

No idea how I became the assignee and patch author, but I'll gladly accept ;-)

Comment 12

[:aceman]

Comment on attachment 8901459 [details] [diff] [review]
patch

Review of attachment 8901459 [details] [diff] [review]:
-----------------------------------------------------------------

You found the problem and the fix :) I do not need to be credited for the boilerplate around that :)
I'll try to push this (haven't pushed in a long time) after the try run is done:
https://treeherder.mozilla.org/#/jobs?repo=try-comm-central&revision=f3f8f99f47a02664ad1902207f4298ecd6319f8a

Comment 13

[Jorg K (CEST = GMT+2)]

(In reply to :aceman from comment #12)
> I'll try to push this (haven't pushed in a long time) after the try run is
> done:
Actually, I have another patch and there hasn't been any M-C merge since last night, so let me do it if you don't mind.

Comment 14

[:aceman]

Ok, then please fix the capitalization in the commit message.

Comment 15

[Jorg K (CEST = GMT+2)]

Already done. And we only use one dash, not two ;-)

Comment 16

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/comm-central/rev/754473839281
Port bug 1392358: Update xhr.send(nsIInputStream) to xhr.sendINputstream(nsIInputStream). r=aceman