Remove mozharness' copy of virtualenv and use the one under /third_party/python instead
Categories
(Release Engineering :: Applications: MozharnessCore, enhancement)
Tracking
(firefox-esr78 fixed, firefox80 fixed, firefox81 fixed)
People
(Reporter: ahal, Assigned: hamzah18051, Mentored)
References
Details
Attachments
(3 files)
(deleted),
text/x-review-board-request
|
tomprince
:
review+
|
Details |
(deleted),
text/x-phabricator-request
|
RyanVM
:
approval-mozilla-esr78+
|
Details |
(deleted),
text/x-phabricator-request
|
Details |
Reporter | ||
Comment 1•7 years ago
|
||
Updated•7 years ago
|
Comment 2•7 years ago
|
||
Comment hidden (mozreview-request) |
Updated•6 years ago
|
Updated•6 years ago
|
Comment 4•6 years ago
|
||
mozreview-review |
Comment hidden (mozreview-request) |
Comment 7•6 years ago
|
||
Comment 8•6 years ago
|
||
Comment hidden (mozreview-request) |
Comment 10•6 years ago
|
||
mozreview-review |
Comment 11•6 years ago
|
||
mozreview-review-reply |
Comment 12•6 years ago
|
||
Comment 13•6 years ago
|
||
Comment hidden (mozreview-request) |
Comment 15•6 years ago
|
||
mozreview-review |
Comment hidden (mozreview-request) |
Comment 17•6 years ago
|
||
Comment 18•6 years ago
|
||
Comment 19•6 years ago
|
||
Comment 20•6 years ago
|
||
Updated•6 years ago
|
Reporter | ||
Comment 21•6 years ago
|
||
Assignee | ||
Comment 22•4 years ago
|
||
Updated•4 years ago
|
Comment 23•4 years ago
|
||
Comment 24•4 years ago
|
||
bugherder |
Reporter | ||
Comment 25•4 years ago
|
||
I think we should consider backing this out and fixing the regressions on our own time.
Reporter | ||
Comment 26•4 years ago
|
||
This change had a few regressions that will be easier to fix without the pressure of time.
Reporter | ||
Updated•4 years ago
|
Reporter | ||
Comment 27•4 years ago
|
||
The somewhat good news is that bug 1655666 has a patch, and the issues from bug 1655164 and bug 1655180 look the same. So I think there is only thing to fix here. I couldn't figure out why it was happening from a brief glance, so decided to back this change out in the meantime.
Here is the traceback:
[task 2020-07-20T12:13:21.829Z] 12:13:21 FATAL - Uncaught exception: Traceback (most recent call last):
[task 2020-07-20T12:13:21.829Z] 12:13:21 FATAL - File "/builds/task_1595246656/workspace/mozharness/mozharness/base/script.py", line 2137, in run
[task 2020-07-20T12:13:21.829Z] 12:13:21 FATAL - self.run_action(action)
[task 2020-07-20T12:13:21.829Z] 12:13:21 FATAL - File "/builds/task_1595246656/workspace/mozharness/mozharness/base/script.py", line 2076, in run_action
[task 2020-07-20T12:13:21.829Z] 12:13:21 FATAL - self._possibly_run_method(method_name, error_if_missing=True)
[task 2020-07-20T12:13:21.829Z] 12:13:21 FATAL - File "/builds/task_1595246656/workspace/mozharness/mozharness/base/script.py", line 2031, in _possibly_run_method
[task 2020-07-20T12:13:21.829Z] 12:13:21 FATAL - return getattr(self, method_name)()
[task 2020-07-20T12:13:21.829Z] 12:13:21 FATAL - File "/builds/task_1595246656/workspace/mozharness/mozharness/mozilla/testing/raptor.py", line 483, in install_chrome_android
[task 2020-07-20T12:13:21.829Z] 12:13:21 FATAL - self.device.shell_output("settings put global verifier_verify_adb_installs 0")
[task 2020-07-20T12:13:21.829Z] 12:13:21 FATAL - File "/builds/task_1595246656/workspace/mozharness/mozharness/mozilla/testing/android.py", line 57, in device
[task 2020-07-20T12:13:21.829Z] 12:13:21 FATAL - adb = self.adb_path
[task 2020-07-20T12:13:21.829Z] 12:13:21 FATAL - File "/builds/task_1595246656/workspace/mozharness/mozharness/mozilla/testing/android.py", line 45, in adb_path
[task 2020-07-20T12:13:21.829Z] 12:13:21 FATAL - self.activate_virtualenv()
[task 2020-07-20T12:13:21.829Z] 12:13:21 FATAL - File "/builds/task_1595246656/workspace/mozharness/mozharness/base/python.py", line 442, in activate_virtualenv
[task 2020-07-20T12:13:21.829Z] 12:13:21 FATAL - exec(open(activate).read(), dict(__file__=activate))
[task 2020-07-20T12:13:21.829Z] 12:13:21 FATAL - IOError: [Errno 2] No such file or directory: '/builds/task_1595246656/workspace/build/venv/bin/activate_this.py'
[task 2020-07-20T12:13:21.829Z] 12:13:21 FATAL - Running post_fatal callback...
[task 2020-07-20T12:13:21.829Z] 12:13:21 FATAL - Exiting -1
Reporter | ||
Comment 28•4 years ago
|
||
I just tested and the version of virtualenv we have in-tree does produce an activate_this.py script, so I have no idea why it appears to be missing. Maybe the whole venv path is wrong or something?
Comment 29•4 years ago
|
||
Assignee | ||
Comment 30•4 years ago
|
||
I think it's happening because it's not able to find abs_src_dir
or base_work_dir
in the dictionary
Comment 31•4 years ago
|
||
bugherder |
Reporter | ||
Updated•4 years ago
|
Updated•4 years ago
|
Comment 32•4 years ago
|
||
Do you also want the backout on beta, or are we ok there?
Reporter | ||
Comment 33•4 years ago
|
||
Thanks for asking! I don't think it's a huge deal to leave it, but I guess ideally it would be backed out on beta too.
Julien, would you be able to handle that for me?
Comment 34•4 years ago
|
||
I think with the fix from bug 1655666 we don't need to land it on beta, unless there are other issues it is causing. :ahal, with that fix, is there any reason this couldn't land again? I didn't investigate all the regressions, but based on comment 27, it sounds like they all had the same root cause.
Reporter | ||
Comment 35•4 years ago
|
||
I was under the impression that bug 1655666 was a separate issue from the other two, but I haven't investigated very closely. If they are all the same issue then I guess I backed it out for now reason :(.
Hamzah, can you confirm if bug 1655666 fixed the other regression?
Assignee | ||
Comment 36•4 years ago
|
||
bug1655666 fixed perma android regression(bug 1655164) , but it didn't fix bug 1655180.
I tried fixing that and did a try run . Everything seemed to work fine.
https://treeherder.mozilla.org/#/jobs?repo=try&revision=bdc11189ee0e590df5038b3db3bca67c60a44dd0
Andrew if you could please confirm this then that would be great.
Reporter | ||
Comment 37•4 years ago
|
||
(In reply to Hamzah Akhtar from comment #36)
bug1655666 fixed perma android regression(bug 1655164) , but it didn't fix bug 1655180.
Ah ok, thanks for clarifying! It looks like a separate patch landed in bug 1655164 though (aside from the one in bug 1655666). Maybe that second patch also ended up fixing bug 1655180?
I tried fixing that and did a try run . Everything seemed to work fine.
https://treeherder.mozilla.org/#/jobs?repo=try&revision=bdc11189ee0e590df5038b3db3bca67c60a44dd0
Andrew if you could please confirm this then that would be great.
That bug talks about mach raptor
being broken, and your try push doesn't contain any raptor tasks (even if it did, I think the issue was only showing up locally). So it's hard to say.
Could you answer the following please?
- Can you reproduce the
mach raptor
issue on the latest central with only the patch from this bug? (i.e excluding the fix from your try push) - If so, can you confirm that your try push from comment 36 fixes it?
Either way sounds like we're close here.
Assignee | ||
Comment 38•4 years ago
|
||
After looking at that try run that I mentioned above again, I remembered that it was related to perma Android regression, I think that is fixed now.
About the mach raptor
command being broken, yeah I was able to reproduce to it using the patch in this bug, also I was able to fix it
I'll submit a patch for that in bug 1655180
Reporter | ||
Comment 39•4 years ago
|
||
Perfect, thanks for clarifying. It might be cleanest to incorporate the changes from that fix in the patch here, but it's up to you :).
Comment 40•4 years ago
|
||
Comment 41•4 years ago
|
||
bugherder |
Comment 42•4 years ago
|
||
Comment on attachment 9162914 [details]
Bug 1408051 - Remove mozharness' copy of virtualenv and use the one under /third_party/python instead
ESR Uplift Approval Request
- If this is not a sec:{high,crit} bug, please state case for ESR consideration: Resolves
pyvenv.cfg
intermittent on MinGW builds - User impact if declined: None
- Fix Landed on Version: 81
- Risk to taking this patch: Low
- Why is the change risky/not risky? (and alternatives if risky): It's a build-only patch
- String or UUID changes made by this patch:
Updated•4 years ago
|
Comment 43•4 years ago
|
||
bugherder uplift |
Description
•