969446 - Backout --hide-subtests option since bug 937181 implemented the same thing but better

Status: RESOLVED FIXED

(Testing :: Mochitest, defect)

Description

[Andrew Halberstadt [:ahal]]

I'll need to test it out on Ash or somewhere to make sure the log files don't jump up to 50mb again.

Comment 1

[Andrew Halberstadt [:ahal]]

Here's the ash-mozharness backout:
http://hg.mozilla.org/users/asasaki_mozilla.com/ash-mozharness/rev/a64955bf0e14

Watching the b2g desktop mochitest retriggers of:
https://tbpl.mozilla.org/?tree=Ash&showall=1&rev=d619310b8b48

Comment 2

[Andrew Halberstadt [:ahal]]

The logs are slightly bigger, though still only 300kb, so looks like I can back it out.

Comment 3

[Andrew Halberstadt [:ahal]]

https://hg.mozilla.org/build/mozharness/rev/5f03e956749d

I'll need to wait for that to get merged to production before backing out the gecko bits.

Comment 4

[Andrew Halberstadt [:ahal]]

Attachment: backout_hide_subtests

Seems like I forgot to get around to this. There was a lot of bitrot, so here is a try run to make sure I didn't mess something up:
https://tbpl.mozilla.org/?tree=Try&rev=722f5906f0f3

Comment 5

[Andrew Halberstadt [:ahal]]

Comment on attachment 8419675 [details] [diff] [review]
backout_hide_subtests

Try run looks good. Log sizes are still small.

Comment 6

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

Comment on attachment 8419675 [details] [diff] [review]
backout_hide_subtests

Review of attachment 8419675 [details] [diff] [review]:
-----------------------------------------------------------------

ok, cool stuff

::: testing/mochitest/runtests.py
@@ +28,5 @@
>  import traceback
>  import urllib2
>  import zipfile
>  
> +from automationutils import environment, getDebuggerInfo, isURL, KeyValueParseError, parseKeyValue, processLeakLog, dumpScreen, ShutdownLeaks, printstatus

why are we removing systemMemory?

@@ +1361,4 @@
>                self.dumpScreenOnTimeout,
>                self.dumpScreenOnFail,
>                self.metro_subprocess_id,
>                self.trackShutdownLeaks,

why are we rearranging this?

Comment 7

[Andrew Halberstadt [:ahal]]

(In reply to Joel Maher (:jmaher) from comment #6)
> Comment on attachment 8419675 [details] [diff] [review]
> backout_hide_subtests
> 
> Review of attachment 8419675 [details] [diff] [review]:
> -----------------------------------------------------------------
> 
> ok, cool stuff
> 
> ::: testing/mochitest/runtests.py
> @@ +28,5 @@
> >  import traceback
> >  import urllib2
> >  import zipfile
> >  
> > +from automationutils import environment, getDebuggerInfo, isURL, KeyValueParseError, parseKeyValue, processLeakLog, dumpScreen, ShutdownLeaks, printstatus
> 
> why are we removing systemMemory?

My pyflakes informed me it was imported but unused. Just a drive-by fix up.

> @@ +1361,4 @@
> >                self.dumpScreenOnTimeout,
> >                self.dumpScreenOnFail,
> >                self.metro_subprocess_id,
> >                self.trackShutdownLeaks,
> 
> why are we rearranging this?

The original patch had re-arranged it, so the backout is just re-arranging it back to the way it was.

Comment 8

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

ok, land as is!

Comment 9

[Andrew Halberstadt [:ahal]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/6a0b9cf33604

Comment 10

[Andrew Halberstadt [:ahal]]

This was merged awhile ago, but must have been missed:
https://hg.mozilla.org/mozilla-central/rev/6a0b9cf33604