Closed Bug 738890 Opened 12 years ago Closed 11 years ago

Intermittent Robocop testJarReader | Error getting method - got false, expected true

Categories

(Firefox for Android Graveyard :: General, defect)

All
Android
defect
Not set
normal

Tracking

(blocking-fennec1.0 -)

RESOLVED FIXED
Firefox 24
Tracking Status
blocking-fennec1.0 --- -

People

(Reporter: philor, Assigned: nalexander)

References

Details

(Keywords: intermittent-failure, Whiteboard: [test disabled][lang=java])

Attachments

(2 files, 1 obsolete file)

https://tbpl.mozilla.org/php/getParsedLog.php?id=10341393&tree=Mozilla-Inbound#error1
Android Tegra 250 mozilla-inbound opt test robocop on 2012-03-23 22:12:47 PDT for push 164c27136716

2 INFO TEST-UNEXPECTED-FAIL | testJarReader | Error getting method - got false, expected true
blocking-fennec1.0: --- → -
Attached patch Disable tests (obsolete) — Splinter Review
This disables this test and some others while we look into them. Right now I get the feeling that starring everything orange is masking any real failures we have.
Assignee: nobody → wjohnston
Attachment #609354 - Flags: review?(bugmail.mozilla)
Attached patch Disable testsSplinter Review
Ok. Talked to kats and he's fine with letting the other tests fail occasionally for now. This disables JarReader because the failures seem.... constant. I'm looking at it and will fix it as soon as I can (I tried pushing a patch that waits for Gecko:Ready to try https://tbpl.mozilla.org/?tree=Try&rev=2fb95c42f08f, but I'm not sure why that would help).

In the mean time, QA has tested the feature and we aren't seeing any other fallout from it that I know of, so I'm fine leaving the feature enabled while we fix the test.

You ok with this mfinkle?
Attachment #609354 - Attachment is obsolete: true
Attachment #609354 - Flags: review?(bugmail.mozilla)
Attachment #609359 - Flags: review?(mark.finkle)
Whoops. should not have re-enabled testFlingCorrectness. fixing locally....
Comment on attachment 609359 [details] [diff] [review]
Disable tests

r+ with testFlingCorrectness disabled
Attachment #609359 - Flags: review?(mark.finkle) → review+
https://hg.mozilla.org/integration/mozilla-inbound/rev/61e725c0d361

This bug should be left open until we find a real solution.
https://hg.mozilla.org/mozilla-central/rev/61e725c0d361

please next time annotate [leave open] in the whiteboard, the comments in a random orange bug are quite unfindable due to tbplbot spam (luckily we can collapse it).
Whiteboard: [orange] → [orange][test disabled]
Whiteboard: [orange][test disabled] → [test disabled]
Assignee: wjohnston → nobody
Whiteboard: [test disabled] → [test disabled][lang=java]
:wesj -- Can you offer any insight into the problem here? Thoughts on what we should do? This test has been disabled for over a year now. Is it worth keeping?
Flags: needinfo?(wjohnston)
I'm not sure what's up here. These worked for me at the time. I haven't had any time to look into them.

The test seems like it would be useful if we could turn it on, but I'll also note we haven't had any bugs in this component. Up to you guys if you want to keep it or kill it.
Flags: needinfo?(wjohnston)
Assignee: nobody → nalexander
OS: Mac OS X → Android
Hardware: x86 → All
No idea why this was considered intermittent orange in the first
place.  This test might have saved me Bug 884626.
Attachment #765734 - Flags: review?(wjohnston)
(In reply to Nick Alexander :nalexander from comment #92)
> Created attachment 765734 [details] [diff] [review]
> Fix testJarReader. r=wesj

Thanks Nick!

> No idea why this was considered intermittent orange in the first
> place.  

In that case, please run lots of retries through try server before landing!
(In reply to Geoff Brown [:gbrown] from comment #93)
> (In reply to Nick Alexander :nalexander from comment #92)
> > Created attachment 765734 [details] [diff] [review]
> > Fix testJarReader. r=wesj
> 
> Thanks Nick!
> 
> > No idea why this was considered intermittent orange in the first
> > place.  
> 
> In that case, please run lots of retries through try server before landing!

To be clear: at some point the underlying reflected code changed, which led to a permanent "Error getting method" NoSuchMethodException getting thrown.  That's not intermittent :)

https://tbpl.mozilla.org/?tree=Try&rev=1ac31cccc2b9 so we can retrigger to our hearts' content.
Comment on attachment 765734 [details] [diff] [review]
Fix testJarReader. r=wesj

Review of attachment 765734 [details] [diff] [review]:
-----------------------------------------------------------------

Try looks ok to me. I'm still not sure why this was failing with "Error getting method" for me then and isn't failing now....
Attachment #765734 - Flags: review?(wjohnston) → review+
(In reply to Wesley Johnston (:wesj) from comment #95)
> Comment on attachment 765734 [details] [diff] [review]
> Fix testJarReader. r=wesj
> 
> Review of attachment 765734 [details] [diff] [review]:
> -----------------------------------------------------------------
> 
> Try looks ok to me. I'm still not sure why this was failing with "Error
> getting method" for me then and isn't failing now....

The underlying code changed.  GeckoJarReader.getStream changed signature; this code adds it back.  Thanks for the review!
https://hg.mozilla.org/services/services-central/rev/c4980d251238
Status: NEW → ASSIGNED
Whiteboard: [test disabled][lang=java] → [test disabled][lang=java][fixed in services]
Target Milestone: --- → Firefox 24
https://hg.mozilla.org/mozilla-central/rev/c4980d251238
Status: ASSIGNED → RESOLVED
Closed: 11 years ago
Resolution: --- → FIXED
Whiteboard: [test disabled][lang=java][fixed in services] → [test disabled][lang=java]
Product: Firefox for Android → Firefox for Android Graveyard
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: