Environment
- CB Response Server: 6.0.1 and Higher
Objective
How to calculate the sensor backlog totals cluster by node when troubleshooting large aggregate sensor backlogs
Resolution
- On the master, run this psql query. Sum is in bytes.
# psql -d cb -p 5002
cb=# select node_id, sum(num_eventlog_bytes), avg(num_eventlog_bytes) from sensor_registrations where next_checkin_time > current_timestamp - (interval'5 minutes') group by node_id;
node_id | sum | avg
---------+--------------+-----------------------
6 | 4614820406 | 587201.985748823005
4 | 156842909776 | 20005473.185714285714
5 | 4606552191 | 583255.531906811851
2 | 3884693723 | 497718.606406149904
1 | 3917879845 | 504231.640283140283
3 | 4815135219 | 617641.767444843509
(6 rows)
