not all buckets recorded in bucketsystems cf

Bug #1178012 reported by Brian Murray on 2013-05-08
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Daisy
Medium
Brian Murray

Bug Description

While working on the phased updater glue between launchpad and errors I was using two different methods to determine whether not a crash first occurred with a version of package. First I was using, first_appearance with MostCommonProblemsResource in errors/api/resources.py. Then as the backfill of sourceversionbuckets_cf completed I was switched to package-version-new-buckets which calls cassandra.get_package_new_buckets. Anyway, the set of results between the two was different.

Digging, into this the following bucket is returned for glom version 1.20.10-0ubuntu1.1 using first_appearence but not package-version-new-buckets.

u'/us/bin/glom:6:g_assertion_message:g_assertion_message_expr:gda_sql_builder_set_table:Glom::ReportBuilder::report_build_records_field:Glom::ReportBuilder::report_build_headerfooter'

Further digging revealed that the bucket doesn't exist in bucketsystems_cf at all.

In [3]: cassandra.get_package_new_buckets('glom', '1.20.10-0ubuntu1', '1.20.10-0ubuntu1.1')
> /home/bdmurray/source-trees/errors/errors/errors/cassandra.py(491)get_package_new_buckets()
    490 from ipdb import set_trace; set_trace()
--> 491 try:
    492 count = len(bucketsystems_cf.get(bucket, column_count=4))
ipdb> bucketsystems_cf.get(bucket)
*** NotFoundException: NotFoundException(_message=None)
ipdb> bucket
u'/usr/bin/glom:6:g_assertion_message:g_assertion_message_expr:gda_sql_builder_set_table:Glom::ReportBuilder::report_build_records_field:Glom::ReportBuilder::report_build_headerfooter'

The problem page for this bucket is https://errors.ubuntu.com/bucket/?id=/usr/bin/glom%3A6%3Ag_assertion_message%3Ag_assertion_message_expr%3Agda_sql_builder_set_table%3AGlom%3A%3AReportBuilder%3A%3Areport_build_records_field%3AGlom%3A%3AReportBuilder%3A%3Areport_build_headerfooter

and there we can see that this has occurred as recently as 3/16/2013 so I'd expect it to be in bucketsystems.

Related branches

Brian Murray (brian-murray) wrote :

The code for bucketsystems may have landed with RT #60417 on 4/2/2013 so it not existing makes sense now. Perhaps we should backfill bucketsystems too.

Changed in daisy:
importance: Undecided → Medium
Evan (ev) wrote :

Yes, it looks like we never created a back-population job for this. Agreed that it would be a good idea to fill in the missing data.

Changed in daisy:
status: New → In Progress
assignee: nobody → Brian Murray (brian-murray)
To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers