[22 November] Multi-Account Incident report Workflow
Hello folks,
Here is the workflow for reporting multi-account infrastructure change/create requests and incidents under Discovery DevOps.
Infrastructure Request
Currently all infrastructure request (new infra or change infra) in the shared tvlk-prod account using Github issue. For Multi-account issues are reported into specific repository based on Product Domains, because of which it is very hard to track and manage. Hence, we are changing the workflow for Multi-account issue reporting:
- For shared
tvlk-prod and tvlk-dev we will continue to use the existing workflow.
tvlk-prod Infrastructure request will be going through GitHub issue here.
tvlk-dev Infrastructure request will be gong through JIRA helpdesk.
What will change in Infrastructure Request workflow?
- For issues related to multi-account, which you want discovery DevOps to resolve should be reported in here.
- This will aid us in giving quick turnarounds and managing all multi-account request under one hood.
- For multi-account issues, the engineers wants to work themselves can be reported in the specific multi-account repository.
- Engineer can create PR along with the infrastructure request, review and execution is done by the delegates.
- The delegates can escalate to Site-infra if necessary (e.g. : out of delegate scope, delegates do not have the policy to execute it, etc)
What happen to existing Infrastructure Request on Github ?
The existing tickets which are already reported in the multi-account infra repositories will remain like that. Going forward please follow the new workflow for multi-account issues.
Incident Handling
Currently incident that cannot handled by PDA will be escalated to delegates. Since we have delegate ,Our Incident Handling workflow is PDA -> Delegates -> Site-Infra.
What will change in Incident Handling workflow?
- For
tvlk-prod,Incident report will be going through github issue here.
- These requests can be tracked along with multi-account issues in the dashboard mentioned here.
- Priority tickets will be handled by On-call person for the week. You can get the on-call person information in our slack channel #discovery-si-devops.
What should you do?
- Make sure you report all multi-account issues here.
- Create incident ticket and notify the delegates in our slack channel #discovery-si-devops.
When is this workflow implement effectively?
This workflow will be implemented effectively by 25 November 2019.