Conversation
dhlevi
left a comment
There was a problem hiding this comment.
Alarms look good, but the thresholds might need some tweaking based on what we see in behaviour.
Worth adding a check/alarm/canary for some other things too, like inability to hit the OAuth service (early on-prem outage warning) or for outages to any BCGW connections. This could also be added to the API healthcheck (not the readiness check for instance spin down!) for ease of checking with a simple response package. Could use a lambda
Split this into another ticket because the changes would be bigger than the scope of this PR. WFPREV-887 |
No description provided.