During the system slowness incident, it took a long time for us to realize that the production database was constantly at high load. An alert had been created, but it was several days before, so we didn't correlate it with the incident.
Can we have alerts fire more consistently until they are resolved?
During the system slowness incident, it took a long time for us to realize that the production database was constantly at high load. An alert had been created, but it was several days before, so we didn't correlate it with the incident.