Incident Procedure

  • Incidents

  • Escalation Plan

    1. If you are new, add your shadow to any pagerduty incident ( Delegate Incident )

    2. If the problem cannot be resolved within 15 minutes and is affecting customers, to use /incident in the #war-room so that third-level support is notified

    3. If the problem is taking longer, create an on-going incident in the Incident Manager

    4. If you need someone from SYSTEC or another team, and no one is ACK your pager-duty call/@mention in slack, escalate to your team leader

    5. In case there are issues with third-party infrastructure providers ie Heroku/Compose, use Support for External Service and Resource as a guideline

    6. Incidents need a post-mortem. In Vienna it was decided to use pager-duty post mortems. The first person coming in contact with the incident is responsible for arranging the post-mortem meeting (involving all participants part of the incident management) and for following up on post-mortem agreed steps to improve the system. Either kulo (channel related) or draven (platform related) should be invited to the post-mortem.