A glimpse into the Bridgy Fed monitoring dashboards. Pretty conventional mix of infra, OS, and app level metrics.

Note the delay numbers. When you do something in one network, how quickly do we bridge it across? We pay a lot of attention to that, we try hard to keep it as fast as possible!

monitoring dashboard with graphs for task results, task queues, CPU, and memory
monitoring dashboard with graphs for receive protocols, types, source instances, and ATProto firehose commits
monitoring dashboard with graphs for unsupported activities and sent activities by protocol and type
monitoring dashboard with graphs for user count, receive/send/ATProto serve/ATProto consume delays, tasks run, and processing delays

I wish we could make these dashboards public! Google Cloud Monitoring doesn’t support that right now; hopefully they will eventually.