Will Lachance

Let's try to get this done sooner rather than later, in case we run into problems.

On second thought, I think new relic is probably sufficient for tracking major problems, let's move this to the next milestone.

The way we're querying the database and returning data is not currently optimized for this kind of query, so this would require some changes for it to work well (i.e....

We are now showing 14 day aggregates by default

Filed a bug to track adding this to error_aggregates: https://bugzilla.mozilla.org/show_bug.cgi?id=1441559

Yup, definitely. We actually track client counts and session hours as part of the error aggregates already, but maybe those numbers would be better off in their own series.

Yep, I was initially resistant to make missioncontrol overly general but I've come around to the idea that we need some kind of more general interface for comparing arbitrary versions...

@whd -- do you have thoughts on this? From what I remember the docker-compose setup is only used for testing, can we just go ahead and make these changes for...

@edmorley -- also, could you give some more details on what needs to be fixed in those upstream dockerfiles?

There are still some suggestions in here we could use, so I'm going to reopen for now.