Closed
Bug 1222944
Opened 9 years ago
Closed 8 years ago
[mozrunner] Dump kernel log for B2G emulator
Categories
(Testing :: Mozbase, defect, P3)
Testing
Mozbase
Tracking
(Not tracked)
RESOLVED
WONTFIX
People
(Reporter: edgar, Unassigned)
References
Details
Attachments
(3 obsolete files)
Kernel log is helpful when debugging some test failure which is caused by emulator environment, like Bug 1202310 comment 4. It will be very handy if we could dump it when running emulator tests.
Reporter | ||
Comment 1•9 years ago
|
||
I already have a WIP patch, will attach it later.
Reporter | ||
Comment 2•9 years ago
|
||
Reporter | ||
Comment 3•9 years ago
|
||
Attachment #8684862 -
Attachment is obsolete: true
Reporter | ||
Comment 4•9 years ago
|
||
Review commit: https://reviewboard.mozilla.org/r/28113/diff/#index_header See other reviews: https://reviewboard.mozilla.org/r/28113/
Reporter | ||
Updated•9 years ago
|
Attachment #8685942 -
Attachment is obsolete: true
Reporter | ||
Comment 5•9 years ago
|
||
Try result: https://treeherder.mozilla.org/#/jobs?repo=try&revision=91647ac6cb35&group_state=expanded
Reporter | ||
Comment 6•9 years ago
|
||
Comment on attachment 8698780 [details] MozReview Request: Bug 1222944 - [mozrunner] Dump kernel log for B2G emulator Review request updated; see interdiff: https://reviewboard.mozilla.org/r/28113/diff/1-2/
Attachment #8698780 -
Flags: review?(ahalberstadt)
Comment 7•9 years ago
|
||
Comment on attachment 8698780 [details] MozReview Request: Bug 1222944 - [mozrunner] Dump kernel log for B2G emulator https://reviewboard.mozilla.org/r/28113/#review25309 This looks great! Just one issue below. ::: testing/mozbase/mozrunner/mozrunner/devices/base.py:148 (Diff revision 1) > + #save kernel log nit: space after # ::: testing/mozbase/mozrunner/mozrunner/devices/base.py:152 (Diff revision 1) > + kmsg_args = [self.app_ctx.adb, '-s', '%s' % serial, > + 'shell', 'cat', '/proc/kmsg'] 'cat' will just return right away and miss all the following output from the actual job. I think you want to use 'tail -f /proc/kmsg' instead. If you want the full file dumped from the very beginning you can use 'tail -f -n +0 /proc/kmsg' (assuming the tail binary on the emulator exists and is the same as the one on my linux machine).
Attachment #8698780 -
Flags: review?(ahalberstadt)
Comment 8•9 years ago
|
||
https://reviewboard.mozilla.org/r/28113/#review25309 > 'cat' will just return right away and miss all the following output from the actual job. I think you want to use 'tail -f /proc/kmsg' instead. If you want the full file dumped from the very beginning you can use 'tail -f -n +0 /proc/kmsg' (assuming the tail binary on the emulator exists and is the same as the one on my linux machine). Just tested, 'tail' doesn't exist by default, but comes with busybox.. So I guess we either need to come up with an equivalent to 'tail -f' or add a check to only save the kernel log if busybox is installed.
Reporter | ||
Comment 9•9 years ago
|
||
(In reply to Andrew Halberstadt [:ahal] from comment #8) > https://reviewboard.mozilla.org/r/28113/#review25309 > > > 'cat' will just return right away and miss all the following output from the actual job. I think you want to use 'tail -f /proc/kmsg' instead. If you want the full file dumped from the very beginning you can use 'tail -f -n +0 /proc/kmsg' (assuming the tail binary on the emulator exists and is the same as the one on my linux machine). > > Just tested, 'tail' doesn't exist by default, but comes with busybox.. So I > guess we either need to come up with an equivalent to 'tail -f' or add a > check to only save the kernel log if busybox is installed. Thanks for your comment, will address this in next version.
Reporter | ||
Updated•9 years ago
|
Assignee: nobody → echen
Reporter | ||
Updated•8 years ago
|
Blocks: emu-x86-kk-tests
Updated•8 years ago
|
Priority: -- → P3
Comment 11•8 years ago
|
||
I don't think we are going to move this on due to the latest B2G announcement.
Status: NEW → RESOLVED
Closed: 8 years ago
Resolution: --- → WONTFIX
Reporter | ||
Updated•8 years ago
|
Attachment #8698780 -
Attachment is obsolete: true
You need to log in
before you can comment on or make changes to this bug.
Description
•