github.com/cockroachdb/cockroach@v20.2.0-alpha.1+incompatible/monitoring/rules/aggregation.rules.yml (about) 1 # This file contains aggregation rules, specifically: 2 # "node:X" node-level aggregation of a per-store metric X 3 # "cluster:X" cluster-level aggregation of a per-store or per-node metric X 4 # 5 # Most aggregation rules should use the "without (label1, label2, ...)" keyword 6 # to keep all labels but the ones specified. 7 8 groups: 9 - name: rules/aggregation.rules 10 rules: 11 - record: node:capacity 12 expr: sum without(store) (capacity{job="cockroachdb"}) 13 - record: cluster:capacity 14 expr: sum without(instance) (node:capacity{job="cockroachdb"}) 15 - record: node:capacity_available 16 expr: sum without(store) (capacity_available{job="cockroachdb"}) 17 - record: cluster:capacity_available 18 expr: sum without(instance) (node:capacity_available{job="cockroachdb"}) 19 - record: capacity_available:ratio 20 expr: capacity_available{job="cockroachdb"} / capacity{job="cockroachdb"} 21 - record: node:capacity_available:ratio 22 expr: node:capacity_available{job="cockroachdb"} / node:capacity{job="cockroachdb"} 23 - record: cluster:capacity_available:ratio 24 expr: cluster:capacity_available{job="cockroachdb"} / cluster:capacity{job="cockroachdb"} 25 # Histogram rules: these are fairly expensive to compute live, so we precompute a few percetiles. 26 - record: txn_durations_bucket:rate1m 27 expr: rate(txn_durations_bucket{job="cockroachdb"}[1m]) 28 - record: txn_durations:rate1m:quantile_50 29 expr: histogram_quantile(0.5, txn_durations_bucket:rate1m) 30 - record: txn_durations:rate1m:quantile_75 31 expr: histogram_quantile(0.75, txn_durations_bucket:rate1m) 32 - record: txn_durations:rate1m:quantile_90 33 expr: histogram_quantile(0.9, txn_durations_bucket:rate1m) 34 - record: txn_durations:rate1m:quantile_95 35 expr: histogram_quantile(0.95, txn_durations_bucket:rate1m) 36 - record: txn_durations:rate1m:quantile_99 37 expr: histogram_quantile(0.99, txn_durations_bucket:rate1m) 38 - record: exec_latency_bucket:rate1m 39 expr: rate(exec_latency_bucket{job="cockroachdb"}[1m]) 40 - record: exec_latency:rate1m:quantile_50 41 expr: histogram_quantile(0.5, exec_latency_bucket:rate1m) 42 - record: exec_latency:rate1m:quantile_75 43 expr: histogram_quantile(0.75, exec_latency_bucket:rate1m) 44 - record: exec_latency:rate1m:quantile_90 45 expr: histogram_quantile(0.9, exec_latency_bucket:rate1m) 46 - record: exec_latency:rate1m:quantile_95 47 expr: histogram_quantile(0.95, exec_latency_bucket:rate1m) 48 - record: exec_latency:rate1m:quantile_99 49 expr: histogram_quantile(0.99, exec_latency_bucket:rate1m) 50 - record: round_trip_latency_bucket:rate1m 51 expr: rate(round_trip_latency_bucket{job="cockroachdb"}[1m]) 52 - record: round_trip_latency:rate1m:quantile_50 53 expr: histogram_quantile(0.5, round_trip_latency_bucket:rate1m) 54 - record: round_trip_latency:rate1m:quantile_75 55 expr: histogram_quantile(0.75, round_trip_latency_bucket:rate1m) 56 - record: round_trip_latency:rate1m:quantile_90 57 expr: histogram_quantile(0.9, round_trip_latency_bucket:rate1m) 58 - record: round_trip_latency:rate1m:quantile_95 59 expr: histogram_quantile(0.95, round_trip_latency_bucket:rate1m) 60 - record: round_trip_latency:rate1m:quantile_99 61 expr: histogram_quantile(0.99, round_trip_latency_bucket:rate1m) 62 - record: sql_exec_latency_bucket:rate1m 63 expr: rate(sql_exec_latency_bucket{job="cockroachdb"}[1m]) 64 - record: sql_exec_latency:rate1m:quantile_50 65 expr: histogram_quantile(0.5, sql_exec_latency_bucket:rate1m) 66 - record: sql_exec_latency:rate1m:quantile_75 67 expr: histogram_quantile(0.75, sql_exec_latency_bucket:rate1m) 68 - record: sql_exec_latency:rate1m:quantile_90 69 expr: histogram_quantile(0.9, sql_exec_latency_bucket:rate1m) 70 - record: sql_exec_latency:rate1m:quantile_95 71 expr: histogram_quantile(0.95, sql_exec_latency_bucket:rate1m) 72 - record: sql_exec_latency:rate1m:quantile_99 73 expr: histogram_quantile(0.99, sql_exec_latency_bucket:rate1m) 74 - record: raft_process_logcommit_latency_bucket:rate1m 75 expr: rate(raft_process_logcommit_latency_bucket{job="cockroachdb"}[1m]) 76 - record: raft_process_logcommit_latency:rate1m:quantile_50 77 expr: histogram_quantile(0.5, raft_process_logcommit_latency_bucket:rate1m) 78 - record: raft_process_logcommit_latency:rate1m:quantile_75 79 expr: histogram_quantile(0.75, raft_process_logcommit_latency_bucket:rate1m) 80 - record: raft_process_logcommit_latency:rate1m:quantile_90 81 expr: histogram_quantile(0.9, raft_process_logcommit_latency_bucket:rate1m) 82 - record: raft_process_logcommit_latency:rate1m:quantile_95 83 expr: histogram_quantile(0.95, raft_process_logcommit_latency_bucket:rate1m) 84 - record: raft_process_logcommit_latency:rate1m:quantile_99 85 expr: histogram_quantile(0.99, raft_process_logcommit_latency_bucket:rate1m) 86 - record: raft_process_commandcommit_latency_bucket:rate1m 87 expr: rate(raft_process_commandcommit_latency_bucket{job="cockroachdb"}[1m]) 88 - record: raft_process_commandcommit_latency:rate1m:quantile_50 89 expr: histogram_quantile(0.5, raft_process_commandcommit_latency_bucket:rate1m) 90 - record: raft_process_commandcommit_latency:rate1m:quantile_75 91 expr: histogram_quantile(0.75, raft_process_commandcommit_latency_bucket:rate1m) 92 - record: raft_process_commandcommit_latency:rate1m:quantile_90 93 expr: histogram_quantile(0.9, raft_process_commandcommit_latency_bucket:rate1m) 94 - record: raft_process_commandcommit_latency:rate1m:quantile_95 95 expr: histogram_quantile(0.95, raft_process_commandcommit_latency_bucket:rate1m) 96 - record: raft_process_commandcommit_latency:rate1m:quantile_99 97 expr: histogram_quantile(0.99, raft_process_commandcommit_latency_bucket:rate1m)