Database failover
Please note: if the incident relates to sensitive data, or is security related consider labeling this issue with security and mark it confidential.
Summary
There has been a database failover from patroni-07
to patroni-10
.
Service(s) affected : Database/PostgreSQL/Patroni
Team attribution :
Minutes downtime or degradation : 3
Timeline
2019-08-27
- 18:54 UTC - HAProxy web exhibited a brief but pronounced uptick in
5xx
errors - 14:24 CDT - Production Alert: Unused Replication Slots for patroni-10-db-gprd.c.gitlab-production.internal
- 14:30 CDT - Production Alert: PostgreSQL dead tuples is too large
Edited by Nels Nelson