1701150 - Intermittent comm/mail/test/browser/openpgp/composition/browser_composeEncrypted.js | 2 recipient key statuses should be displayed - 0 == 2 - testEncryptedOneRecipientKeyNotAvailableMessageComposition/keyStatusDialog< :: line 439

Status: RESOLVED FIXED

(MailNews Core :: Security: OpenPGP, defect, P5)

Description

[Treeherder Bug Filer]

Filed by: geoff [at] darktrojan.net
Parsed log: https://treeherder.mozilla.org/logviewer?job_id=334522521&repo=comm-central
Full log: https://firefox-ci-tc.services.mozilla.com/api/queue/v1/task/FlEaYjfaRhibtcXELuYXFg/runs/0/artifacts/public/logs/live_backing.log

---

This is failing on all Windows builds but nowhere else.```

Comment 1

[Kai Engert [:KaiE:]]

Regression caused by bug 1693133 ?

Comment 2

[Kai Engert [:KaiE:]]

Adding Lasana who wrote this test. Do you have an idea what might go wrong now?

Comment 3

[Intermittent Failures Robot]

16 failures in 3967 pushes (0.004 failures/push) were associated with this bug in the last 7 days.

Repository breakdown:

• comm-central: 16

Platform and build breakdown:

• windows10-64: 4
  • opt: 4
• windows7-32: 4
  • opt: 4
• windows7-32-shippable: 3
  • opt: 3
• windows10-64-shippable: 3
  • opt: 3
• linux64-shippable: 2
  • opt: 2

For more details, see:
https://treeherder.mozilla.org/intermittent-failures/bugdetails?bug=1701150&startday=2021-03-22&endday=2021-03-28&tree=all

Comment 4

[Geoff Lankow (:darktrojan)]

Not sure if it's the same or not but a very similar failure just showed up on Linux.

Comment 5

[Intermittent Failures Robot]

16 failures in 647 pushes (0.025 failures/push) were associated with this bug yesterday.

Repository breakdown:

• comm-central: 16

Platform and build breakdown:

• windows10-64: 3
  • opt: 3
• linux64: 3
  • opt: 3
• windows7-32: 3
  • opt: 3
• windows7-32-shippable: 3
  • opt: 3
• linux64-shippable: 2
  • opt: 2
• windows10-64-shippable: 2
  • opt: 2

For more details, see:
https://treeherder.mozilla.org/intermittent-failures/bugdetails?bug=1701150&startday=2021-03-29&endday=2021-03-29&tree=all

Comment 6

[Lasana Murray]

I think I found the culprit behind this, maybe. Occasionally the callback here:
https://searchfox.org/comm-central/source/mail/test/browser/openpgp/composition/browser_composeEncrypted.js#436
is being called before the key status dialog is finished.

After some debugging I traced the preemptive execution to this point:
https://searchfox.org/comm-central/source/mail/extensions/openpgp/content/modules/sqliteDb.jsm#90

To me

await conn
        .execute(
          "select decision from acceptance_decision where fpr = :fpr",
          qObj
        )
        .then(result => {
          if (result.length) {
            rv.fingerprintAcceptance = result[0].getResultByName("decision");
          }
        });

is the same as

let promise = conn.execute(
          "select decision from acceptance_decision where fpr = :fpr",
          qObj
        );

await  promise
        .then(result => {
          if (result.length) {
            rv.fingerprintAcceptance = result[0].getResultByName("decision");
          }
        });

Which is to say the creation of the promise is not awaited on, potentially opening up room for racy conditions. That said, the test still fails if that is cleaned up. The key status dialog is populated asynchronously in an onload handler in any case so I'll just change the assertions to TestUtils.waitForCondition.

I should mention I could not reproduce this via an artifact build locally. Only when I did a fresh rebuild. Maybe something change in the lower layers recently.

Comment 7

[Lasana Murray]

Attachment: Bug 1701150 - Wait for key statuses to be populated in compose encrypted test. r?mkmelin

Comment 8

[Intermittent Failures Robot]

18 failures in 698 pushes (0.026 failures/push) were associated with this bug yesterday.

Repository breakdown:

• comm-central: 18

Platform and build breakdown:

• windows7-32: 4
  • opt: 4
• windows10-64: 3
  • opt: 3
• windows10-64-shippable: 4
  • opt: 4
• linux64: 2
  • opt: 2
• linux64-shippable: 2
  • opt: 2
• windows7-32-shippable: 3
  • opt: 3

For more details, see:
https://treeherder.mozilla.org/intermittent-failures/bugdetails?bug=1701150&startday=2021-03-30&endday=2021-03-30&tree=all

Comment 9

[Magnus Melin [:mkmelin] (away until Aug 10)]

(In reply to Lasana Murray from comment #6)
I don't see any difference there?

Comment 10

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/comm-central/rev/2c85f0685155
Wait for key statuses to be populated in compose encrypted test. r=mkmelin

Comment 11

[Lasana Murray]

(In reply to Magnus Melin [:mkmelin] from comment #9)

(In reply to Lasana Murray from comment #6)
I don't see any difference there?

I'm unsure of the semantics myself, it certainly is odd syntax. May be worth keeping an eye on if things get start running out of order.

Comment 12

[Intermittent Failures Robot]

34 failures in 3125 pushes (0.011 failures/push) were associated with this bug in the last 7 days.

This is the #46 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:

• comm-central: 34

Platform and build breakdown:

• windows10-64: 6
  • opt: 6
• linux64: 5
  • opt: 5
• windows7-32: 7
  • opt: 7
• windows7-32-shippable: 6
  • opt: 6
• linux64-shippable: 4
  • opt: 4
• windows10-64-shippable: 6
  • opt: 6

For more details, see:
https://treeherder.mozilla.org/intermittent-failures/bugdetails?bug=1701150&startday=2021-03-29&endday=2021-04-04&tree=all