Closed
Bug 659195
Opened 13 years ago
Closed 13 years ago
adjust installApp step to return infra error only for pre-install part
Categories
(Release Engineering :: General, defect)
Tracking
(Not tracked)
RESOLVED
FIXED
People
(Reporter: bear, Assigned: bear)
Details
(Whiteboard: [android][tegra])
Attachments
(1 file, 1 obsolete file)
8.97 KB,
patch
|
mozilla
:
review+
bear
:
checked-in+
|
Details | Diff | Splinter Review |
currently the installApp step for Android testing is an all or nothing result so any failure of the binary to run post-install will be flagged as an infrastructure failure. Adjust the code to return "retry" where necessary
Assignee | ||
Updated•13 years ago
|
Assignee: nobody → bear
Whiteboard: [android][tegra]
Assignee | ||
Comment 1•13 years ago
|
||
This patch contains two sets of changes. Change installApp to cause a RED if the install works but the push .ini fails and also dump logcat for warnings and errors. Also contains a tweak to check.py to do a PING of the tegra first before trying to contact the agent - this allows me to flag the tegra as OFFLINE|INACTIVE|active with more accuracy
Attachment #535553 -
Flags: review?(aki)
Comment 2•13 years ago
|
||
Comment on attachment 535553 [details] [diff] [review] change to installApp and also check.py Does the logcat help on install success?
Attachment #535553 -
Flags: review?(aki) → review+
Comment 3•13 years ago
|
||
Comment on attachment 535553 [details] [diff] [review] change to installApp and also check.py >--- a/sut_tools/installApp.py >+++ b/sut_tools/installApp.py >@@ -75,19 +75,21 @@ if dm.pushFile(source, target): > dm.getInfo('process') > dm.getInfo('memory') > dm.getInfo('uptime') > pid = dm.processExist(processName) > print '%s PID %s' % (processName, pid) > if pid is not None: > dm.killProcess(processName) > dm.getInfo('process') >+ print dm.sendCMD(['exec su -c "logcat -d *:W"']) > else: > clearFlag(proxyFile) >- setFlag(errorFile, "Remote Device Error: unable to push %s" % inifile) >+ setFlag(errorFile, "ERROR: unable to push %s" % inifile) >+ print dm.sendCMD(['exec su -c "logcat -d *:W"']) > sys.exit(1) Should this even be a setFlag(errorFile ? Won't that take the tegra offline, even if it's the apk that's broken? I think we might want to print instead, but you may know better.
Assignee | ||
Comment 4•13 years ago
|
||
(In reply to comment #2) > Comment on attachment 535553 [details] [diff] [review] [review] > change to installApp and also check.py > > Does the logcat help on install success? It may get removed later, but my reason to keep it is to give bmoss any debug info for a successful install to compare to when it fails.
Assignee | ||
Comment 5•13 years ago
|
||
same as previous but with setFlag(error) changed to a print
Attachment #535553 -
Attachment is obsolete: true
Attachment #535588 -
Flags: review?(aki)
Updated•13 years ago
|
Attachment #535588 -
Flags: review?(aki) → review+
Assignee | ||
Comment 6•13 years ago
|
||
Comment on attachment 535588 [details] [diff] [review] change to installApp and also check.py committed changeset 1455:790352c8011f
Attachment #535588 -
Flags: checked-in+
Assignee | ||
Updated•13 years ago
|
Status: NEW → RESOLVED
Closed: 13 years ago
Resolution: --- → FIXED
Updated•11 years ago
|
Product: mozilla.org → Release Engineering
You need to log in
before you can comment on or make changes to this bug.
Description
•