Closed Bug 1085369 Opened 5 years ago Closed 5 years ago

Intermittent test_WebCrypto.html | Test timed out. - expected PASS

Categories

(Core :: DOM: Security, defect)

x86
Android
defect
Not set

Tracking

()

RESOLVED FIXED
mozilla38
Tracking Status
firefox36 --- fixed
firefox37 --- fixed
firefox38 --- fixed
firefox-esr31 --- unaffected

People

(Reporter: cbook, Assigned: ttaubert)

References

()

Details

(Keywords: intermittent-failure)

Attachments

(2 files)

Android 2.3 Emulator mozilla-inbound opt test mochitest-9

https://treeherder.mozilla.org/ui/logviewer.html#?job_id=3131449&repo=mozilla-inbound

05:18:58 INFO - 94 INFO TEST-UNEXPECTED-FAIL | /tests/dom/crypto/test/test_WebCrypto.html | Test timed out. - expected PASS
Moving key wrapping/unwrapping tests to a separate file.
Assignee: nobody → ttaubert
Status: NEW → ASSIGNED
Attachment #8507971 - Flags: review?(rlb)
Hmmm. It seems that I should have done some measurements first. We're actually spending most of the time here:

/tests/dom/crypto/test/test_WebCrypto.html | Test that we properly normalize algorithm names
took 1033ms

/tests/dom/crypto/test/test_WebCrypto.html | Test that we reject generating keys without any usage
took 1471ms
Should we move those to separate tests? What would the names be? Maybe:

test_WebCrypto_Normalize.html and test_WebCrypto_bug1078847.html ?
Flags: needinfo?(rlb)
Review/needinfo ping?
Comment on attachment 8507971 [details] [diff] [review]
0001-Bug-1085369-Move-key-wrapping-unwrapping-tests-to-th.patch

Review of attachment 8507971 [details] [diff] [review]:
-----------------------------------------------------------------

This looks fine, but I'm a little curious why wrap/unwrap would take so long.  Is it just the RSA-OAEP operation?
Attachment #8507971 - Flags: review?(rlb) → review+
Flags: needinfo?(rlb)
(In reply to Richard Barnes [:rbarnes] from comment #11)
> This looks fine, but I'm a little curious why wrap/unwrap would take so
> long.  Is it just the RSA-OAEP operation?

It actually doesn't. I provided some more accurate measurements in comment #3. What do you think about the suggestion in comment #4?
Flags: needinfo?(rlb)
Hey Richard, can you please give some feedback wrt comment #3 and #4?
Implemented what I suggested in comment #4 to properly fix those intermittent timeouts.
Flags: needinfo?(rlb)
Attachment #8544525 - Flags: review?(rlb)
Attachment #8544525 - Flags: review?(dkeeler)
Comment on attachment 8544525 [details] [diff] [review]
0002-Bug-1085369-Move-other-long-running-tests-to-separat.patch

Review of attachment 8544525 [details] [diff] [review]:
-----------------------------------------------------------------

LGTM. I imagine you could also make the RSA key sizes smaller, since that's likely what's taking so much time.
Attachment #8544525 - Flags: review?(dkeeler) → review+
Attachment #8544525 - Flags: review?(rlb)
(In reply to David Keeler [:keeler] (use needinfo?) from comment #35)
> LGTM. I imagine you could also make the RSA key sizes smaller, since that's
> likely what's taking so much time.

Thanks! Yeah I thought about decreasing key sizes as well, guess we can keep that in mind if we run into that sort of problem again.
https://hg.mozilla.org/mozilla-central/rev/e2c0816e8e36
https://hg.mozilla.org/mozilla-central/rev/83a293a6a859
Status: ASSIGNED → RESOLVED
Closed: 5 years ago
Flags: in-testsuite+
Resolution: --- → FIXED
Target Milestone: --- → mozilla38
You need to log in before you can comment on or make changes to this bug.