| 1 | |
| 2 | # Common |
| 3 | |
| 4 | template: hdfs_capacity_usage |
| 5 | on: hdfs.capacity |
| 6 | class: Utilization |
| 7 | type: Storage |
| 8 | component: HDFS |
| 9 | calc: ($used) * 100 / ($used + $remaining) |
| 10 | units: % |
| 11 | every: 10s |
| 12 | warn: $this > (($status >= $WARNING) ? (70) : (80)) |
| 13 | crit: $this > (($status == $CRITICAL) ? (80) : (98)) |
| 14 | delay: down 15m multiplier 1.5 max 1h |
| 15 | summary: HDFS datanodes space utilization |
| 16 | info: summary datanodes space capacity utilization |
| 17 | to: sysadmin |
| 18 | |
| 19 | |
| 20 | # NameNode |
| 21 | |
| 22 | template: hdfs_missing_blocks |
| 23 | on: hdfs.blocks |
| 24 | class: Errors |
| 25 | type: Storage |
| 26 | component: HDFS |
| 27 | calc: $missing |
| 28 | units: missing blocks |
| 29 | every: 10s |
| 30 | warn: $this > 0 |
| 31 | delay: down 15m multiplier 1.5 max 1h |
| 32 | summary: HDFS missing blocks |
| 33 | info: number of missing blocks |
| 34 | to: sysadmin |
| 35 | |
| 36 | |
| 37 | template: hdfs_stale_nodes |
| 38 | on: hdfs.data_nodes |
| 39 | class: Errors |
| 40 | type: Storage |
| 41 | component: HDFS |
| 42 | calc: $stale |
| 43 | units: dead nodes |
| 44 | every: 10s |
| 45 | warn: $this > 0 |
| 46 | delay: down 15m multiplier 1.5 max 1h |
| 47 | summary: HDFS stale datanodes |
| 48 | info: number of datanodes marked stale due to delayed heartbeat |
| 49 | to: sysadmin |
| 50 | |
| 51 | |
| 52 | template: hdfs_dead_nodes |
| 53 | on: hdfs.data_nodes |
| 54 | class: Errors |
| 55 | type: Storage |
| 56 | component: HDFS |
| 57 | calc: $dead |
| 58 | units: dead nodes |
| 59 | every: 10s |
| 60 | crit: $this > 0 |
| 61 | delay: down 15m multiplier 1.5 max 1h |
| 62 | summary: HDFS dead datanodes |
| 63 | info: number of datanodes which are currently dead |
| 64 | to: sysadmin |
| 65 | |
| 66 | |
| 67 | # DataNode |
| 68 | |
| 69 | template: hdfs_num_failed_volumes |
| 70 | on: hdfs.num_failed_volumes |
| 71 | class: Errors |
| 72 | type: Storage |
| 73 | component: HDFS |
| 74 | calc: $fsds_num_failed_volumes |
| 75 | units: failed volumes |
| 76 | every: 10s |
| 77 | warn: $this > 0 |
| 78 | delay: down 15m multiplier 1.5 max 1h |
| 79 | summary: HDFS failed volumes |
| 80 | info: number of failed volumes |
| 81 | to: sysadmin |