Closed Bug 972612 Opened 10 years ago Closed 10 years ago

store gccrashes per ADU

Categories

(Socorro :: Database, task)

Product:
Socorro
Component:
Database
Type:
task
Priority:
Not set
Severity:
normal

Tracking

(Not tracked)

Status:
RESOLVED FIXED

People

(Reporter: rhelmer, Assigned: rhelmer)

References

Details

(Whiteboard: [DBChanges]) 

Per bug 915317 comment 9, this report should be gccrashes per ADU. Let's have the SP do that, there's no reason to store it in a raw form in the matview (people can access the raw data as necessary)
Blocks: 915317
Needs backfill
Whiteboard: [DBChanges]
PR: https://github.com/mozilla/socorro/pull/1888
Status: NEW → ASSIGNED
Commits pushed to master at https://github.com/mozilla/socorro

https://github.com/mozilla/socorro/commit/e56e67c3ad8307d2f71e1d613cf11c89fdb57093
fix bug 972612 - is_gc_count should be per-adu r=selenamarie

https://github.com/mozilla/socorro/commit/d27a8e996517a2e1d8a6dfeea0361e9b86200267
Merge pull request #1888 from rhelmer/bug972612-gccrashes-per-adu

fix bug 972612 - is_gc_count should be per-adu
Status: ASSIGNED → RESOLVED
Closed: 10 years ago
Resolution: --- → FIXED
Oops, adu_count can be 0 (saw this on stage) - let's filter those out.
Status: RESOLVED → REOPENED
Resolution: FIXED → ---
Commits pushed to master at https://github.com/mozilla/socorro

https://github.com/mozilla/socorro/commit/20934115eab5b07f7a231945e159694271f52111
fix bug 972612 - filter out products with 0 ADU

https://github.com/mozilla/socorro/commit/a4ce4e3abab16c8fc3faee93230b542331b74822
Merge pull request #1889 from rhelmer/bug972612-gccrashes-per-adu

fix bug 972612 - filter out products with 0 ADU
Status: REOPENED → RESOLVED
Closed: 10 years ago → 10 years ago
Resolution: --- → FIXED
Status: RESOLVED → REOPENED
Resolution: FIXED → ---
Hmm woke up too early and thought about this, shouldn't we be using build_adu here? We want "is_gc_count per ADI", but the ADI in product_adu is *all* build IDs on Firefox/30.0a1/nightly, right?

Seems like what we really want is "is_gc_count per (ADI for this Firefox/30.a1/nightly/buildID)"
Flags: needinfo?(sdeckelmann)
I concur with comment #6
Yes, build_adu is necessary for nightly/aurora builds.

Also note that since build_adu stops recording at 7 days, we should also limit the crashes to crashes reported within 7 days of their buildid.
(In reply to Benjamin Smedberg  [:bsmedberg] from comment #8)
> Yes, build_adu is necessary for nightly/aurora builds.
> 
> Also note that since build_adu stops recording at 7 days, we should also
> limit the crashes to crashes reported within 7 days of their buildid.

Got it, thanks!
Flags: needinfo?(sdeckelmann)
Commits pushed to master at https://github.com/mozilla/socorro

https://github.com/mozilla/socorro/commit/1f8d91761fc2551fd66824dd0dcf861066f4307f
fix bug 972612 - use build_adu and crashes_madu, limit crashes to w
ithin 7 days of build date r=selenamarie

https://github.com/mozilla/socorro/commit/1b39f2728fcf76bb07c67b1150073c97a0406ab7
Merge pull request #1892 from rhelmer/bug972612-gccrashes-madu

fix bug 972612 - use build_adu and crashes_madu, limit crashes to w
Status: REOPENED → RESOLVED
Closed: 10 years ago → 10 years ago
Resolution: --- → FIXED
Found one problem while backfilling - some buildIDs are way too short and can't be parsed by the build_date() function
Status: RESOLVED → REOPENED
Resolution: FIXED → ---
Target Milestone: --- → 75
Commits pushed to master at https://github.com/mozilla/socorro

https://github.com/mozilla/socorro/commit/eb34ed49af74252e5461a5f82544a8c459aa900c
fix bug 972612 - some buildIDs in reports_clean are too small to parse as dates

https://github.com/mozilla/socorro/commit/61fe65766ac8517a1f9efcbb827edd73746e1218
Merge pull request #1895 from rhelmer/bug972612-gccrashes-madu

fix bug 972612 - some buildIDs in reports_clean are too small to parse a...
Status: REOPENED → RESOLVED
Closed: 10 years ago → 10 years ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.