Closed Bug 1016310 Opened 10 years ago Closed 9 years ago

Intermittent browser_console_optimized_out_vars.js | Test timed out

Categories

(DevTools :: Source Editor, defect)

Product:
DevTools
Component:
Source Editor
Platform:
x86
Windows 8
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

(firefox31 unaffected, firefox32 wontfix, firefox33 wontfix, firefox34 affected, firefox35 affected, firefox-esr24 unaffected)

Status:
RESOLVED WORKSFORME
Tracking Flags:
Tracking Status
firefox31 --- unaffected
firefox32 --- wontfix
firefox33 --- wontfix
firefox34 --- affected
firefox35 --- affected
firefox-esr24 --- unaffected

People

(Reporter: cbook, Unassigned)

References

(
URL
)

Details

(Keywords: intermittent-failure)

Attachments

(1 file, 1 obsolete file)

Wait a turn of the event loop if CodeMirror isn't settled by the time breakpoints are being set
8.96 KB, patch
vporof
: review+
Sylvestre
: approval-mozilla-aurora+
past
: checkin+
Details | Diff | Splinter Review 
WINNT 6.2 mozilla-inbound debug test mochitest-devtools-chrome-3 on 2014-05-26 12:50:10 PDT for push bda83125b312

slave: t-w864-ix-036

https://tbpl.mozilla.org/php/getParsedLog.php?id=40426274&tree=Mozilla-Inbound



TEST-UNEXPECTED-FAIL | chrome://mochitests/content/browser/browser/devtools/webconsole/test/browser_console_optimized_out_vars.js | Test timed out
The logs point to the same error as bug 1013948.
Depends on: 1013948
(In reply to Panos Astithas [:past] from comment #18)
> The logs point to the same error as bug 1013948.

Any progress here? This is currently the #4 orange on trunk.
Flags: needinfo?(past)
I'll take a look at the blocking bug.
Flags: needinfo?(past)
(In reply to Panos Astithas [:past] from comment #135)
> I'll take a look at the blocking bug.

Thank you :-)
The patch in bug 957174 should fix this.
Depends on: 957174
No longer depends on: 1013948
(In reply to Panos Astithas [:past] from comment #147)
> The patch in bug 957174 should fix this.

Almost a week since the last comment in bug 957174. Is this still on the radar? :) Given that it's one of the top oranges at the moment, I'm afraid we'll need to consider disabling the test per the Test Disabling Policy there isn't any progress made soon.
https://wiki.mozilla.org/Sheriffing/Test_Disabling_Policy
(In reply to Ryan VanderMeulen [:RyanVM UTC-4] from comment #267)
> (In reply to Panos Astithas [:past] from comment #147)
> > The patch in bug 957174 should fix this.
> 
> Almost a week since the last comment in bug 957174. Is this still on the
> radar? :) Given that it's one of the top oranges at the moment, I'm afraid
> we'll need to consider disabling the test per the Test Disabling Policy
> there isn't any progress made soon.
> https://wiki.mozilla.org/Sheriffing/Test_Disabling_Policy

Disabled browser_console_optimized_out_vars.js on all platforms for intermittent failures:
https://hg.mozilla.org/integration/mozilla-inbound/rev/f9d0af794bd2
Flags: needinfo?(past)
Keywords: leave-open
Whiteboard: [test disabled]
I'm working on it right now. An unfortunate combination of a long review cycle and PTO made this longer than expected.
Flags: needinfo?(past)
Np, thank you :-)
Now that the fix for bug 957174 landed, I've re-enabled the test:
https://hg.mozilla.org/integration/fx-team/rev/16fab8529241
Assignee: nobody → past
Status: NEW → ASSIGNED
Keywords: leave-open
Whiteboard: [test disabled]
Component: Developer Tools: Inspector → Developer Tools: Source Editor
https://hg.mozilla.org/mozilla-central/rev/16fab8529241
Status: ASSIGNED → RESOLVED
Closed: 10 years ago
Resolution: --- → FIXED
Target Milestone: --- → Firefox 33
Hitting the failures on trunk again.
Status: RESOLVED → REOPENED
Flags: needinfo?(past)
Resolution: FIXED → ---
status-firefox33: fixed → affected
Target Milestone: Firefox 33 → ---
It looks like the patch in bug 957174 decreased its frequency, but didn't eliminate it entirely. I'm investigating.
Flags: needinfo?(past)
I've been testing this both locally and on try and I haven't managed to hit the error again. This reverts the relevant part of the fix in bug 957174, because if my theory is correct, it shouldn't be needed.
Attachment #8453820 - Flags: review?(vporof)
Attachment #8453121 - Attachment is obsolete: true
Comment on attachment 8453820 [details] [diff] [review]
Wait a turn of the event loop if CodeMirror isn't settled by the time breakpoints are being set

Review of attachment 8453820 [details] [diff] [review]:
-----------------------------------------------------------------

I like this a lot more than we used to have.

::: browser/devtools/sourceeditor/debugger.js
@@ +147,5 @@
>      return;
>  
> +  let deferred = promise.defer();
> +  // If lineInfo() returns null, wait a tick to give the editor a chance to
> +  // initializa properly.

initialize
Attachment #8453820 - Flags: review?(vporof) → review+
https://hg.mozilla.org/integration/fx-team/rev/8552660728e7
Status: REOPENED → ASSIGNED
Whiteboard: [fixed-in-fx-team]
https://hg.mozilla.org/mozilla-central/rev/8552660728e7
Status: ASSIGNED → RESOLVED
Closed: 10 years ago → 10 years ago
Resolution: --- → FIXED
Whiteboard: [fixed-in-fx-team]
Target Milestone: --- → Firefox 33
Please request Aurora approval on this patch when you get a chance :)
status-firefox33: affected → fixed
Comment on attachment 8453820 [details] [diff] [review]
Wait a turn of the event loop if CodeMirror isn't settled by the time breakpoints are being set

Approval Request Comment
[Feature/regressing bug #]: it is a relatively recent regression that I papered over in bug 957174, but this is a better fix.
[User impact if declined]: oranges on tbpl and a potential for a broken JS debugger in some circumstances
[Describe test coverage new/current, TBPL]: browser_console_optimized_out_vars.js catches it.
[Risks and why]: just the general risk that comes with any change, but this one is contained to devtools code, so only developers can encounter any potential regression
[String/UUID change made/needed]: none
Attachment #8453820 - Flags: approval-mozilla-aurora?
Attachment #8453820 - Flags: approval-mozilla-aurora? → approval-mozilla-aurora+
This doesn't seem to be fixed?
Status: RESOLVED → REOPENED
Resolution: FIXED → ---
Do you know why this is still failing? It seems to be the top webconsole intermittent test failure.
Flags: needinfo?(past)
Yes, see bug 1013948 comment 7 and 8. In short, it's a race in the CodeMirror initialization that neither waiting 200 ms or a turn of the event loop fixed (see this patch and the one in bug 957174). We have to dig deeper into CodeMirror to understand why this happens, but I'm going on vacation in a couple of days, so I won't be looking at this any time soon.
Assignee: past → nobody
Flags: needinfo?(past)
Status: REOPENED → RESOLVED
Closed: 10 years ago → 9 years ago
Resolution: --- → WORKSFORME
Product: Firefox → DevTools
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: