As a first step towards deploying Sentry to production, create a deployment-sentry machine which collects JS errors directly and backend errors via Logstash from the beta cluster.
Description
Description
Details
Details
Project | Branch | Lines +/- | Subject | |
---|---|---|---|---|
operations/mediawiki-config | master | +3 -1 | BETA: set wmgSentryDsn to null |
Status | Subtype | Assigned | Task | ||
---|---|---|---|---|---|
Declined | None | T106915 Use Sentry in production | |||
Declined | None | T106920 Integrate Sentry with beta cluster | |||
Resolved | Tgr | T84956 Create basic puppet role for Sentry | |||
Resolved | jcrespo | T112228 Need to run postgresql::user twice to set the password | |||
Declined | Tgr | T85239 Channel PHP errors from Logstash to Sentry on the beta cluster | |||
Open | None | T97133 Login integration for Sentry | |||
Open | None | T97136 Phabricator integration for Sentry |
Event Timeline
Comment Actions
I'm guessing this is something to do with the 403s I see in the JS console network tab on beta?
Comment Actions
Probably the Sentry config (specifically the API key) got reset during some maintenance of Cloud VPS boxes. I don't have the time to fix it right now; it should be harmless but if it's getting in the way you can get rid of it by unsetting $wgSentryDsn.
Comment Actions
Change 399680 had a related patch set uploaded (by Addshore; owner: Addshore):
[operations/mediawiki-config@master] BETA: unset $wgSentryDsn