Create alerts for failed and delayed GCP snapshots in Mimir
In thanos-rules/gcp-snapshots.yml
we have alerts defined for failed or missing GCP snapshots.
These alerts now spread multiple tenants (environments). When we move them to mimir, and when doing that separate them by environment.
Right now, all of these metrics are attributed to type="monitoring"
, however, they monitor disks from many different services. In a first iteration, I think we can keep these alerts global, without scoping them to the right services so we don't have to relabel the metrics now. The metrics for this alert come from stackdriver, which will be added to mimir with #2710 (closed).