ooni infrastructure monitoring
OONI infrastructure such as collectors, agregator and probes need a monitoring system to generate time-series graphs AND alert the ops team to fix problems.
This trac ticket will function as the master monitoring ticket linking to other ticket which address specific monitoring needs such as:
- number of reports in each aggregator-pipeline-state
- is HS up?
- disk usage
- memory usage
- cpu usage
- process monitoring
- in/out network bandwidth usage
We should either use existing Tor Project infrastructure to accomplish this or create our own icinga/munin (or nagios + collectd + graphite or equivalent).