Raising issues with Reliability Engineering
Reliability Engineering are a programme in GDS, they are responsible for the infrastructure that most GDS software runs on and the underlying network configuration. They also provide shared software services used by various GDS programmes such as logging and monitoring tools.
When on 2nd line you may experience an issue with GOV.UK that requires asking Reliability Engineering for assistance.
There are Reliability Engineering tech docs for users of their systems.
If you require urgent assistance
Reliability Engineering have a Slack channel - #reliability-eng - and they have an assigned interruptible person. By posting in that channel you can get their attention. This channel can be used for general queries too so do indicate in your message that a problem is time critical.
Failing slack communication you can also walk over to Reliability Engineering desks and talk to the interruptible person directly - they are currently on the 6th floor near bank 26.
You may be advised to create a Zendesk ticket.
If a problem is not urgent
You can use the #reliability-eng slack channel for advice. If the issue you’ve identified seems like a non-urgent story you can add it the GOV.UK 2nd Line trello board in the “Proposed stories for Platform Health” column. Platform Health will then decide whether to raise this with RE, and manage the ticket through its life cycle, or to resolve this problem themselves.
Raising a Zendesk ticket with Reliability Engineering
The official way to communicate with Reliability Engineering is through Zendesk tickets.
To raise a ticket:
- Create a new ticket on Zendesk
- Enter yourself as the requester
- Set asignee to “3rd Line–Infrastructure”
- Add the 2nd Line Delivery Manager as a CC recipient
- Fill in and submit ticket
- Monitor and contribute to the ticket until it is resolved
Understanding what Reliability Engineering can assist with
There is a broad explanation of the different areas of support in GOV.UK in who do I ask for support?.
More specificially to GOV.UK these are things that fall under the responsibility of Reliability Engineering (RE):
- GOV.UK Puppet - RE are responsible for maintenance and evolution, but as GOV.UK merge changes they can too be responsible for problems
- Upgrading software packages that are end-of-life/have security issues/no longer fit for purpose
- Running and maintaining the Terraform configurations for AWS;
- Backup software such as Duplicity
- Maintaining the mirror configuration
- Keeping the CI environment running - GOV.UK are responsible for job configuration
- Fabric scripts