Closed Bug 594373 Opened 15 years ago Closed 14 years ago

BuildAPI testrun efficiency reporting

Categories

(Release Engineering :: General, defect)

x86_64
Linux
defect
Not set
normal

Tracking

(Not tracked)

RESOLVED FIXED

People

(Reporter: salbiz, Assigned: salbiz)

Details

Attachments

(2 files, 4 obsolete files)

Tracking ongoing work on adding reporting mechanism to display and visualize the time taken to complete test-runs (both unitttests and talos) to buildapi.
Attachment #473046 - Flags: feedback?(catlee)
Comment on attachment 473046 [details] [diff] [review] initial patch for adding test efficiency reporting Looks pretty good. Can you add some more comments in TestRunsQuery to explain what the different bits of the query are doing?
Attachment #473046 - Flags: feedback?(catlee) → feedback+
(In reply to comment #1) > Comment on attachment 473046 [details] [diff] [review] > initial patch for adding test efficiency reporting > > Looks pretty good. Can you add some more comments in TestRunsQuery to explain > what the different bits of the query are doing? Yes, definitely. When I have a little more time, I'd also like to re-visit it entirely and see if it can't be sped up a bit, as it crawls along quite slowly at the moment.
Added a few comments to TestRunQuery which should hopefully make it a bit easier to see what's going on. Do you have any suggestions on how to structure that code to make it inherently more readable even without the added comments?
Attachment #473046 - Attachment is obsolete: true
Attachment #474695 - Flags: feedback?(catlee)
Attachment #474695 - Flags: feedback?(catlee) → feedback+
Comment on attachment 474695 [details] [diff] [review] added_clarification_to_TestRunQuery Spent some time experimenting with ways to optimize the query. Moving some of the more complex query logic into the model/controller actually seemed to hurt initial load time (which is the real killer, since the beaker cache reduces subsequent load of groupings, etc immensely). We could try to optimize the database structure by adding some indexes to join on. That is probably out of scope for this bug though, I can file another one for that.
Attachment #474695 - Flags: review?(catlee)
Attached patch [tested]testrun_patch (obsolete) — Splinter Review
Just re-based patch against updated templates, which required some reworking of display logic to handle groupings properly.
Attachment #474695 - Attachment is obsolete: true
Attachment #479220 - Flags: review?(catlee)
Attachment #479220 - Flags: checked-in?
Attachment #474695 - Flags: review?(catlee)
Comment on attachment 479220 [details] [diff] [review] [tested]testrun_patch Looks good except for the hardcoded limit of 300 in TestRunsQuery. Can we get rid of that, or make it an argument to the function?
Attachment #479220 - Flags: review?(catlee) → review-
Attached patch [tested]testrun_patch (obsolete) — Splinter Review
That should have been dropped, it was only there for testing purposes. I've now removed it entirely. Over a date range of a few days we have less than 300 unique builders anyway, which is probably why it didn't show up during testing.
Attachment #479220 - Attachment is obsolete: true
Attachment #479381 - Flags: review?(catlee)
Attachment #479220 - Flags: checked-in?
re-applied against latest changeset on default branch
Attachment #479381 - Attachment is obsolete: true
Attachment #479582 - Flags: review?(catlee)
Attachment #479381 - Flags: review?(catlee)
Attachment #479582 - Flags: checked-in?
Attachment #479582 - Flags: review?(catlee)
Attachment #479582 - Flags: review+
Attachment #479582 - Flags: checked-in?
Attachment #479582 - Flags: checked-in+
fixed JSON output and validated request headers using newly implemented controller schema. Previous patches were based against a revision that used a different header method of getting request parameters.
Attachment #480134 - Flags: review?(catlee)
Attachment #480134 - Flags: checked-in?
Attachment #480134 - Flags: review?(catlee) → review+
Attachment #480134 - Flags: checked-in? → checked-in+
closing this out, since the fix has landed and seems to work well
Status: NEW → RESOLVED
Closed: 14 years ago
Resolution: --- → FIXED
Product: mozilla.org → Release Engineering
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: