This document summarizes the metrics most commonly seen on MongoLab Telemetry dashboards, and for which alert definitions are most commonly defined.
The string in column 1 represents the string that would be used as the METRIC_ID in an alert definition.
| Metric ID | Units | Description |
|---|---|---|
| commands | ops/millisecond | Rate of commands |
| deletes | ops/millisecond | Rate of deletes |
| getmores | ops/millisecond | Rate of getmores |
| inserts | ops/millisecond | Rate of inserts |
| queries | ops/millisecond | Rate of queries |
| updates | ops/millisecond | Rates of updates |
| currentConnections | - | Number of open connections |
| cpuIOWaitNorm* | percent | CPU IOWait |
| cpuSystemNorm* | percent | System CPU |
| cpuUserNorm* | percent | User CPU |
| pageFaults | faults/millisecond | Rate of page faults |
| effectiveLockPercentage | percent | Effective lock |
| operationsWaitingForGlobalLock | mumber of operations | Total queued |
| readsWaitingForGlobalLock | number of operations | Readers queued |
| writesWaitingForGlobalLock | number of operations | Writers queued |
| totalDataPlusIndexSizeWithoutLocal | bytes | Total size of data plus indexes (without 'local' database) |
| totalFileSizeWithoutLocal | bytes | Total file size (without 'local' database) |
| totalFileSize | bytes | Total file size (includes 'local' database) |
| nonMappedVirtualMemory | megabytes | Non-mapped virtual memory |
| timeFlushingToDisk | milliseconds | Background flush average |
| networkBytesOut | bytes/millisecond | Rate of network bytes out |
| networkBytesIn | bytes/millisecond | Rate of network bytes in |
| replicationLag | Milliseconds | Time that secondary is behind primary |
| replicationOplogWindow | Hours | Time between newest and oldest entries in the oplog |
- These CPU metrics are normalized to the number of CPU cores on the VM and always sum to 100%.