1364258 - Intermittent test_ecdsa.https.html | importVectorKeys step: ECDSA P-521 with SHA-256 using privateKey to verify - assert_unreached: importVectorKeys failed for ECDSA P-521 with SHA-256. Message: ''Operation is not supported'' Reached unreachable code

Status: RESOLVED INVALID

(Core :: DOM: Security, defect, P3)

Description

[Treeherder Bug Filer]

Filed by: wkocher [at] mozilla.com

https://treeherder.mozilla.org/logviewer.html#?job_id=98539950&repo=mozilla-inbound

https://queue.taskcluster.net/v1/task/RButRrReQmCeNbg3kDNmhA/runs/0/artifacts/public/logs/live_backing.log

Comment 1

[Intermittent Failures Robot]

22 failures in 879 pushes (0.025 failures/push) were associated with this bug in the last 7 days.   

Repository breakdown:
* mozilla-inbound: 7
* autoland: 6
* mozilla-central: 5
* try: 2
* oak: 2

Platform breakdown:
* linux64: 20
* linux64-ccov: 2

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1364258&startday=2017-05-08&endday=2017-05-14&tree=all

Comment 2

[Geoff Brown [:gbrown]]

This seems to have started with the test update in bug 1363982.

Comment 3

[Intermittent Failures Robot]

17 failures in 188 pushes (0.09 failures/push) were associated with this bug yesterday.   

Repository breakdown:
* autoland: 6
* mozilla-inbound: 5
* mozilla-central: 3
* try: 2
* graphics: 1

Platform breakdown:
* linux64: 16
* linux64-ccov: 1

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1364258&startday=2017-05-17&endday=2017-05-17&tree=all

Comment 4

[Intermittent Failures Robot]

59 failures in 777 pushes (0.076 failures/push) were associated with this bug in the last 7 days. 

This is the #16 most frequent failure this week.  

** This failure happened more than 30 times this week! Resolving this bug is a high priority. **

** Try to resolve this bug as soon as possible. If unresolved for 2 weeks, the affected test(s) may be disabled. ** 

Repository breakdown:
* mozilla-inbound: 21
* autoland: 18
* mozilla-central: 11
* try: 5
* graphics: 4

Platform breakdown:
* linux64: 52
* linux64-ccov: 5
* osx-10-10: 2

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1364258&startday=2017-05-15&endday=2017-05-21&tree=all

Comment 5

[Joel Maher ( :jmaher ) (UTC -8)]

this is a perma fail on linux64 code coverage, likewise a frequent failure on linux64 debug (both non-e10s and e10s).

Comment 6

[Joel Maher ( :jmaher ) (UTC -8)]

I see this was added to web-platform-tests:
https://github.com/w3c/web-platform-tests/commit/ea527511d7f742f5ad939339df3bad2619382a00

Philip, do you have ideas on how to resolve issues seen in this log:
https://queue.taskcluster.net/v1/task/QOWHguPmR_GqQ6PwJeuWCQ/runs/0/artifacts/public/logs/live_backing.log

specifically this error:
[task 2017-05-18T18:59:47.578247Z] 18:59:47     INFO - PID 1143 | ++DOMWINDOW == 30 (0x5708140) [pid = 1143] [serial = 30] [outer = 0x55fc4d0]
[task 2017-05-18T18:59:47.901520Z] 18:59:47     INFO - PID 1143 | ++DOCSHELL 0x2f10d80 == 11 [pid = 1143] [id = {121e3946-3626-496e-8d04-c3553baa3fe8}]
[task 2017-05-18T18:59:47.901980Z] 18:59:47     INFO - PID 1143 | ++DOMWINDOW == 31 (0x7781fc0) [pid = 1143] [serial = 31] [outer = (nil)]
[task 2017-05-18T18:59:47.937642Z] 18:59:47     INFO - PID 1143 | ++DOMWINDOW == 32 (0x33d7aa0) [pid = 1143] [serial = 32] [outer = 0x7781fc0]
[task 2017-05-18T18:59:48.086518Z] 18:59:48     INFO - PID 1143 | ++DOMWINDOW == 33 (0x38b68c0) [pid = 1143] [serial = 33] [outer = 0x7781fc0]
[task 2017-05-18T18:59:48.411575Z] 18:59:48     INFO - PID 1143 | JavaScript warning: https://web-platform.test:8443/WebCryptoAPI/sign_verify/ecdsa.js, line 478: unreachable code after return statement
[task 2017-05-18T18:59:48.414830Z] 18:59:48     INFO - PID 1143 | JavaScript warning: https://web-platform.test:8443/WebCryptoAPI/sign_verify/ecdsa.js, line 478: unreachable code after return statement
[task 2017-05-18T18:59:49.264570Z] 18:59:49     INFO - 
[task 2017-05-18T18:59:49.265041Z] 18:59:49     INFO - TEST-FAIL | /WebCryptoAPI/sign_verify/test_ecdsa.https.html | importVectorKeys step: ECDSA P-521 with SHA-512 with altered plaintext after call - assert_unreached: importVectorKeys failed for ECDSA P-521 with SHA-512. Message: ''Operation is not supported'' Reached unreachable code
[task 2017-05-18T18:59:49.265319Z] 18:59:49     INFO - run_test/</promise</<@https://web-platform.test:8443/WebCryptoAPI/sign_verify/ecdsa.js:84:17
[task 2017-05-18T18:59:49.265924Z] 18:59:49     INFO - Test.prototype.step@https://web-platform.test:8443/resources/testharness.js:1405:20
[task 2017-05-18T18:59:49.266318Z] 18:59:49     INFO - promise_test/tests.promise_tests<@https://web-platform.test:8443/resources/testharness.js:529:27
[task 2017-05-18T18:59:49.266583Z] 18:59:49     INFO - promise callback*promise_test@https://web-platform.test:8443/resources/testharness.js:525:31
[task 2017-05-18T18:59:49.266984Z] 18:59:49     INFO - run_test/</promise<@https://web-platform.test:8443/WebCryptoAPI/sign_verify/ecdsa.js:83:13
[task 2017-05-18T18:59:49.267326Z] 18:59:49     INFO - promise callback*run_test/<@https://web-platform.test:8443/WebCryptoAPI/sign_verify/ecdsa.js:67:23
[task 2017-05-18T18:59:49.267627Z] 18:59:49     INFO - run_test@https://web-platform.test:8443/WebCryptoAPI/sign_verify/ecdsa.js:66:5
[task 2017-05-18T18:59:49.267878Z] 18:59:49     INFO - @https://web-platform.test:8443/WebCryptoAPI/sign_verify/test_ecdsa.https.html:18:1
[task 2017-05-18T18:59:49.268231Z] 18:59:49     INFO - TEST-FAIL | /WebCryptoAPI/sign_verify/test_ecdsa.https.html | importVectorKeys step: ECDSA P-256 with SHA-1 using privateKey to verify - assert_unreached: importVectorKeys failed for ECDSA P-256 with SHA-1. Message: ''Operation is not supported'' Reached unreachable code


Keep in mind when we run this test on a build with gcov enabled this is a permanent failure, otherwise this shows up about one out of 20 times in normal automation.

Comment 7

[Intermittent Failures Robot]

41 failures in 891 pushes (0.046 failures/push) were associated with this bug in the last 7 days. 

This is the #32 most frequent failure this week.  

** This failure happened more than 30 times this week! Resolving this bug is a high priority. **

** Try to resolve this bug as soon as possible. If unresolved for 2 weeks, the affected test(s) may be disabled. ** 

Repository breakdown:
* mozilla-central: 14
* mozilla-inbound: 11
* try: 10
* autoland: 6

Platform breakdown:
* linux64: 31
* linux64-ccov: 10

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1364258&startday=2017-05-22&endday=2017-05-28&tree=all

Comment 8

[Geoff Brown [:gbrown]]

Attachment: disable test

Comment 9

[Joel Maher ( :jmaher ) (UTC -8)]

Comment on attachment 8872700 [details] [diff] [review]
disable test

Review of attachment 8872700 [details] [diff] [review]:
-----------------------------------------------------------------

thanks!

Comment 10

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/mozilla-inbound/rev/1f21438018cb
Disable wpt test_ecdsa.https.html; r=jmaher

Comment 11

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/1f21438018cb

Comment 12

[Intermittent Failures Robot]

13 failures in 820 pushes (0.016 failures/push) were associated with this bug in the last 7 days.   

Repository breakdown:
* try: 4
* mozilla-central: 4
* mozilla-inbound: 3
* autoland: 2

Platform breakdown:
* linux64: 9
* linux64-ccov: 3
* osx-10-10: 1

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1364258&startday=2017-05-29&endday=2017-06-04&tree=all

Comment 13

[Philip Jägenstedt [:foolip]]

Oops, I didn't notice the needinfo request in comment 6 until now. I don't know much about the Web Crypoto API, but is this test still flaky?

Handing over to James Graham as a better generic wpt contact for Gecko.

Comment 14

[James Graham [:jgraham]]

This got disabled.

Comment 15

[BugBot [:suhaib / :marco/ :calixte]]

The leave-open keyword is there and there is no activity for 6 months.
:ckerschb, maybe it's time to close this bug?

Comment 16

[Christoph Kerschbaumer [:ckerschb]]

(In reply to Release mgmt bot [:sylvestre / :calixte] from comment #15)
> The leave-open keyword is there and there is no activity for 6 months.
> :ckerschb, maybe it's time to close this bug?

Yeah, given that the test has been disabled I guess we can mark this one as invalid.