An incident occurred during maintenance — how do you learn about incidents, and how do you respond?

I usually learn about incidents during maintenance through monitoring and alerting systems (for example, Prometheus + Alertmanager, Zabbix, or specialized services). I also receive notifications from colleagues or through ticketing systems.

When an incident is detected, I first assess its impact and urgency. Then I proceed to diagnose: I review logs, check the status of services and infrastructure.

Next, I inform interested parties about the current status and steps being taken. If possible, I roll back changes or apply temporary solutions to minimize impact.

After resolving the issue, I conduct a root cause analysis (post-mortem) to prevent recurrence, and I update documentation and procedures.