SPPH.getPublishedBinaries timing out

[Colin Watson]

I'm getting repeated API timeouts from SourcePackagePublishingHistory.getPublishedBinaries. Here's a Python traceback from a script in lp:ubuntu-archive-tools, including an OOPS code:

Override component to universe
google-mock 1.6.0-0ubuntu1 in raring: main/devel -> universe
Traceback (most recent call last):
  File "/home/cjwatson/src/ubuntu/ubuntu-archive-tools/ubuntu-archive-tools/change-override", line 171, in <module>
    main()
  File "/home/cjwatson/src/ubuntu/ubuntu-archive-tools/ubuntu-archive-tools/change-override", line 167, in main
    change_overrides(options, args)
  File "/home/cjwatson/src/ubuntu/ubuntu-archive-tools/ubuntu-archive-tools/change-override", line 66, in change_overrides
    for pubtype, publication in find_publications(options, packages):
  File "/home/cjwatson/src/ubuntu/ubuntu-archive-tools/ubuntu-archive-tools/change-override", line 41, in find_publications
    for binary in source.getPublishedBinaries():
  File "/usr/lib/python2.7/dist-packages/lazr/restfulclient/resource.py", line 573, in __call__
    url, in_representation, http_method, extra_headers=extra_headers)
  File "/usr/lib/python2.7/dist-packages/lazr/restfulclient/_browser.py", line 311, in _request
    raise error
lazr.restfulclient.errors.ServerError: HTTP Error 503: Service Unavailable
Response headers:
---
-content-encoding: gzip
connection: close
content-length: 9411
content-type: text/html;charset=utf-8
date: Fri, 09 Nov 2012 11:13:52 GMT
retry-after: 900
server: zope.server.http (HTTP)
status: 503
vary: Accept-Encoding
via: 1.1 api.launchpad.net
x-lazr-oopsid: OOPS-57cb9dd835a173a0e34ef00ecd203850
x-powered-by: Zope (www.zope.org), Python (www.python.org)

[Colin Watson (cjwatson)]

SPPH.requestDeletion times out too. Thinking about it, I guess this might be a consequence of the vast numbers of BPPH rows created due to bug 1071562.

[Jamie Strandboge (jdstrand)]

What is the status of this bug? It is my understanding we cannot publish kernel security updates atm.

[Jamie Strandboge (jdstrand)]

Per IRC:
08:32 < jdstrand> hey, can someone comment on the status of bug #1076963? (I asked in the bug). I'm told we are unable to publish kernel security updates due to this bug
08:33 < czajkowski> jdstrand: yes it's being worked on and wgrant will hopfully be clearing the cruft that has mounted up so we can turn stuff back on

[William Grant (wgrant)]

This was indeed fallout from bug #1071562. Removing the 1.5 million extra publications appears to have fixed the timeouts.