#productionchat channel for questions that don't seem appropriate to use the issue tracker or the internal email address for.
+GoogleCalendarbutton in the lower right of the screen when viewing the Calendar with the link here.
Production engineers work on keeping the infrastructure that runs our services running fast and reliably. This infrastructure includes staging, GitLab.com and dev.GitLab.org; see the list of nodes.
Production engineers also have a strong focus on building the right toolsets and automations to enable development to ship features as fast and bug free as possible, leveraging the tools provided by GitLab.com itself - we must dogfood.
Another part of the job is building monitoring tools that allow quick troubleshooting as a first step, then turning this into alerts to notify based on symptoms, to then fixing the problem or automating the remediation. We can only scale GitLab.com by being smart and using resources effectively, starting with our own time as the main scarce resource.
Issues in the tracker are organized into milestones to define the "workout of the week" (WoW) from one week to the next. The "week" runs from Wednesday to end of Tuesday. The other milestone in use is "Next WoW" to track items scheduled for the next week. Every week, the Production Lead renames the WoW to "WoW ending yyyy-mm-dd", and closes it; then renames "Next WoW" to "WoW". By doing this, the closed milestones provide a history of what the team has worked on, while the team only needs to be concerned with two open milestones. If issues are added to the "WoW" after the week has already started, add the
~unscheduled label (not needed if the issue is
~outage since those are by definition unscheduled).
We use issue labels within the Infrastructure issue tracker to assist in prioritizing and organizing work. Prioritized labels are:
~(perceived) data loss
We also use the
~AP3 labels as described in availability & performance priority labels. Those are mainly used to communicate priority of issues to Product Managers, for scheduling purposes.
~goals are issues that are in a WoW and we agreed as a team that we will do everything in our power to deliver them. Goal issues should fit in one WoW, that is, they are deliverable in a single week time, if they do not fit in one WoW we are probably talking about a
We use this kind of issues to indicate a general direction (generally speaking something that will take from 1 to 3 months of work) This means that a
~meta ~goal should be achievable in one quarter.
~meta issues that are not also
~goal are the tasks that are larger than what fits in a quarter, therefore they need to be sliced into actually deliverable pieces that can also become a goal.
We use some other labels to indicate specific conditions and then measure the impact of these conditions within production or the production engineering team. This is specially important from the time investment in specific parts of the production engineering team, to reduce toil or to reduce the chance of a failure by accessing to production more than enough.
Labels that are particularly important for gathering data are:
~toilRepetitive, boring work that should be automated away.
~unscheduledAn issue that became an interruption to the team and had to be handled in a WoW. It's unplanned work.
~unblocks othersAn issue that is allowing some other part of the company to deliver something.
~access requestWhen someone is requesting to get access to some part of the infrastructure.
~requires production accessEvery time someone with production access has to jump into a console to perform some manual operation like running a script in a rails console, or connecting to Redis or the database directly
We should never stop helping and unblocking team members. To this end, data should always be gathered to assist in highlighting areas for automation and the creation of self-service processes. Creating an issue from the request with the proper labels is the first step. The default should be that the person requesting help makes the issue; but we can help with that step too if needed.
If this issue is urgent for whatever reason, we should label them following the instructions above and add them to the ongoing WoW.
Ongoing outages, as well as issues that have the
~(perceived) data loss label and are (therefore) actively being worked on need a hand off to happen as team members cycle in and out of their timezones and availability. The on call log can be used to assist with this. (See link at top to on-call log).
There are 2 kind of production events that we track:
Every last Friday of the month, we have a B.A.D. "party". Two production engineers tag use this day to test our backup processes by fully restoring a backup of the database to a test instance and testing it for data integrity. We plan to continue to do this until such time as the restoring + testing can be done automatically - which is part of the longer term plans described in the production direction meta issue.
The current procedure for B.A.D is to test restore of the production database, one or two secondary databases (version, customers, etc.), PackageCloud database, and test a restore of an NFS server from Azure snapshots. We time the restore processes for each of the above to give us an idea of how long it might take to recover from any given disaster.