If you think a bug might affect users in the 57 release, please set the correct tracking and status flags for Release Management.

Reported (actual?) date range doesn't seem to update.

RESOLVED FIXED

Status

Socorro
General
RESOLVED FIXED
2 years ago
2 years ago

People

(Reporter: davidb, Unassigned)

Tracking

Firefox Tracking Flags

(Not tracked)

Details

For me, this link:
https://crash-stats.mozilla.com/topcrasher/products/Firefox/versions/42.0a1?days=3

Currently reports the date range:
'2015-07-18 through 2015-07-21'

Which seems very stale.
Same here on https://crash-stats.mozilla.com/topcrasher/products/Firefox/versions/40.0b/date_range_type/build?days=7

I see: No crashing signatures found for the period 2015-07-14 to 2015-07-21.
TCBS matviews are red on https://crash-stats.mozilla.com/crontabber-state/

Peter, is something wrong with our cron jobs?
Flags: needinfo?(peterbe)
The short answer is because cron jobs have been failing for a couple of days because of Postgres+Unicode bugs. I'm trying to find out what the "official" bug(s) for that is.
Flags: needinfo?(peterbe)
Depends on: 1168511
We need monitoring for /crontabber-state - could we get a JSON representation of that?

I'll clear out the bad unicode records so this self-heals.
breakpad=> delete from raw_crashes where raw_crash::text like '%\u0000%'::text and date_processed >= '2015-07-21';
DELETE 12

Crontabber should self-heal now, might take a few hours.
Status: NEW → RESOLVED
Last Resolved: 2 years ago
Resolution: --- → FIXED
Reports are now up to date:

https://crash-stats.mozilla.com/crontabber-state/

We are discussing ways to catch this with automated monitoring (we used to do so, but this changed in our recent AWS move.)
Looks like it's back? https://crash-stats.mozilla.com/topcrasher/products/Firefox/versions/42.0a1/date_range_type/build?days=3 shows me "2015-07-27 through 2015-07-30"
Flags: needinfo?(rhelmer)
breakpad=> delete from raw_crashes where raw_crash::text like '%\u0000%'::text and date_processed >= '2015-07-30';
DELETE 8

We have a fix for this ready, and are working on monitoring for these failures generally.
Flags: needinfo?(rhelmer)
You need to log in before you can comment on or make changes to this bug.