assigning to Dirk for triage

Actually, since this is a Blink thing, maybe the Blink infra folks should generally own it, and Ops should just have accounts as a backup?

robertma@, rbyers@, what do you think?

+Mark It seems logical to me that this service can be maintained similarly as Web Confluence, as both can fall under the predictability effort umbrella. How is Confluence currently deployed and who owns its AppEngine?

Apart from deployment/maintenance, are there development tasks? I see a few issues in the repo. Will jochen@ keep working on it?

Lastly, I notice the task is P1 and has a postmortem label. Did anything happen recently that prompted the task?

> Apart from deployment/maintenance, are there development tasks? 
> I see a few issues in the repo. Will jochen@ keep working on it?

I don't think there are any pressing dev tasks. I don't think jochen has any plans or desire to work on this (or the time).

> I notice the task is P1 and has a postmortem label. Did anything happen 
> recently that prompted the task?

Yeah, the service stopped working (was deleted) and jochen had to bring it back :). See go/postmortem97753.

> Actually, since this is a Blink thing, maybe the Blink infra folks should generally own it, and Ops should just have accounts as a backup?

We don't really have "blink infra folks" at the moment, just Robert who (as part of ecosystem-infra team) owns blink testing infrastructure.  But I don't see any other team where this would more naturally fit.  So if I'm successful in expanding ecosystem-infra so that we have at least one other person working with Robert, then owning the odd additional blink-specific infra tool like this would be OK with me (we're already in a precarious spot having critical infrastructure with a single owner).  foolip@ is the TL though, so team priorities are his call. But we agreed we'd call this blocked on me for now to grow the team.

Also related - the intent-thread-specific bits of chromestatus.com like https://github.com/GoogleChrome/chromium-dashboard/issues/469

This P1 issue has been inactive for 30 days. Even though it's not urgent I'm hesitant to call it a P2 given that it recently blew up. If we get some backup for robertma@, then this, other blink infra and upstream wpt infra is at the top of our list to spread the burden / bus factor.

Re: #c7, Confluence AppEngine is owned and maintained by markdittmer@, arobins@, and jmedley@/jpmedley@. Now that Ecosystem Infra and Feature Control teams re distinct, arobins@ wants out. It has a more complex deployment process because of data collection/syncing required before deploying the code.

I don't have access to the deployment scripts (in Drive? I think?) so it's hard for me to assess how similar the two projects are. However, the GitHub repo suggests that project is tiny, so any superficial differences (e.g., NodeJS vs. Python AppEngine) probably don't factor into the decision on whether/when Ecosystem Infra could take ownership.

It's been 30 days since the last update here. Should this still be P1? Is there any work planned here this quarter?

Yeah we've had some isolated pain, but it hasn't been terrible IMHO so I think P2 is more appropriate.  I am still trying to staff up on blink infra folks.

This came up in https://foolip.github.io/ecosystem-infra-rotation/ again today. Unchecking "Send email" just to say that P2 is still appropriate, but I don't expect action here soon. Let's see in another two months.

rbyers@, anything to report on this?