Launchpad itself

Changing a task's target using the API OOPSes as NotImplementedError

Bug #329917 reported by Ursula Junque on 2009-02-16

This bug affects 2 people

Affects		Status	Importance	Assigned to	Milestone
	Launchpad itself	Invalid	High	Unassigned
	launchpadlib	Fix Released	Undecided	Unassigned

Bug Description

As seen on OOPS-1138EC328, OOPS-1143S70.

I could reproduce trying to transitionToTarget twice. The first time it returned without error (but the target was still the old one), and the second it returned a 500 error, producing OOPS-1143S70.

In [21]: bug = lp.bugs[82110]
In [22]: bug_task = bug.bug_tasks[0]
In [23]: bug_task
Out[23]: <bug_task at https://api.staging.launchpad.net/beta/ubuntu/+source/xserver-xorg-video-intel/+bug/82110>
In [24]: bug_task.target
Out[24]: <distribution_source_package at https://api.staging.launchpad.net/beta/ubuntu/+source/xserver-xorg-video-intel>
In [25]: bug_task.transitionToTarget(target="https://api.staging.launchpad.net/beta/ubuntu/+source/xserver-xorg-input-evdev")
In [26]: bug_task.transitionToTarget(target="https://api.staging.launchpad.net/beta/ubuntu/+source/xserver-xorg-input-evdev")
...
HTTPError: HTTP Error 500: Internal Server Error

See original description

Tags:

Ursula Junque (ursinha) on 2009-02-16

description:

updated

Revision history for this message

Brian Murray (brian-murray) wrote on 2009-04-21:

I swear I was using this the other week but now it is failing for me with an AttributeError.

Revision history for this message

Martin Pitt (pitti) wrote on 2009-09-29:

For me it's still oopsing right at the first try (see my duplicate bug 342355):

  File "apport/crashdb_impl/launchpad.py", line 332, in update
    bug.bug_tasks[0].transitionToTarget(target=src)
  File "/usr/lib/python2.6/dist-packages/lazr/restfulclient/resource.py", line 460, in __call__
    url, in_representation, http_method, extra_headers=extra_headers)
  File "/usr/lib/python2.6/dist-packages/lazr/restfulclient/_browser.py", line 204, in _request
    raise HTTPError(response, content)
HTTPError: HTTP Error 500: Internal Server Error

I supplied the result of self.lp_distro.getSourcePackage(name='coreutils') as target= argument (Ursula used a string), but as I said in bug 342355 it's not actually clear what kind of argument type the target option expects (I guess getSourcePackage() should deliver a valid type, though).

Changed in malone:
status:	New → Confirmed

Brian Murray (brian-murray) on 2009-09-30

tags:

added: ubuntu-qa

Revision history for this message

Kees Cook (kees) wrote on 2009-10-01:

This appears to only fail when there is no package already assigned. I.e. if the task is "vlc (Ubuntu)" and I switch it to "linux", the API works. If the task is "Ubuntu" and I switch it to "linux", it fails.

Revision history for this message

Bryce Harrington (bryce) wrote on 2009-10-01:

test Edit (749 bytes, text/plain)

Using the attached launchpadlib script I was able to reproduce the error only if the project was set to none (i.e. "Ubuntu"), at which point it generated this traceback:

Traceback (most recent call last):
  File "./test", line 27, in <module>
    bugtask.transitionToTarget(target = target)
  File "/usr/lib/python2.6/dist-packages/lazr/restfulclient/resource.py", line 460, in __call__
    url, in_representation, http_method, extra_headers=extra_headers)
  File "/usr/lib/python2.6/dist-packages/lazr/restfulclient/_browser.py", line 204, in _request
    raise HTTPError(response, content)
lazr.restfulclient.errors.HTTPError: HTTP Error 500: Internal Server Error

Revision history for this message

Bryce Harrington (bryce) wrote on 2009-10-01:

output Edit (7.6 KiB, text/plain)

debug output from above script when it failed...

Revision history for this message

Graham Binns (gmb) wrote on 2009-10-01:

This bug is happening because the NullBugTask code is being triggered. *Why* it's being triggered, I don't know. It's possible that this is a race condition, but that seems unlikely.

This needs to be dealt with quickly, as it's now causing quite a lot of problems.

Changed in malone:
importance:	Undecided → High
milestone:	none → 3.1.10
status:	Confirmed → Triaged
tags:	added: regression
tags:	removed: regression

Revision history for this message

Eleanor Berger (intellectronica) wrote on 2009-10-01:

The reason Ursula's example is failing is because the task's URI, which is how it's being identified by the API, changes after the target changes. The workaround is to re-fetch the task (using the new URL) after the target has changed. I'm pretty sure there's a launchpadlib bug for making this workaround built-in, but I can't find it right now.

Changed in malone:
status:	Triaged → Invalid

Revision history for this message

Kees Cook (kees) wrote on 2009-10-01: Re: API cannot be used to set a package target for a bugtask when bugtask lacks a package

Bug 342355 has been split from this...

summary:	- Changing a task's target using the API OOPSes as NotImplementedError + API cannot be used to set a package target for a bugtask when bugtask + lacks a package
Changed in malone:
status:	Invalid → Confirmed
status:	Confirmed → Invalid
summary:	- API cannot be used to set a package target for a bugtask when bugtask - lacks a package + Changing a task's target using the API OOPSes as NotImplementedError

Deryck Hodge (deryck) on 2009-10-02

Changed in malone:
milestone:	3.1.10 → none

Revision history for this message

Diogo Matsubara (matsubara) wrote on 2010-04-09:

I couldn't reproduce this bug with the examples given. Might have been fixed already. Can you reproduce?

Changed in launchpadlib:
status:	New → Incomplete

Revision history for this message

Leonard Richardson (leonardr) wrote on 2010-04-13:

#10

This was probably fixed when we changed launchpadlib to re-fetch an object after a POST named operation.

Revision history for this message

Brian Murray (brian-murray) wrote on 2010-04-13:

#11

I have not had an issue with this in quite some time and have been using a bughugger plugin for assigning no package bug reports to packages so I imagine its fixed.

Revision history for this message

Gary Poster (gary) wrote on 2010-04-13:

#12

Optimistically marked as Fix Released, given comments.