You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We should have time-series metrics tracking the error rate & latency profiles of calls to db.Get. Not just calls to db.Get. Calls to pebble in general.
Similar to #11140, I'd argue this should be done in CRDB, not in Pebble itself. I think the CRDB level operations have a better correspondence to what we'd want to measure.
API boundaries are great places to add time-series metrics measuring symptoms experienced by users of that API. For example:
We should have time-series metrics tracking the error rate & latency profiles of calls to
db.Get
. Not just calls todb.Get
. Calls to pebble in general.This will compliment #1140.
The text was updated successfully, but these errors were encountered: