Understand metrics in Metricbeat logs

edit

Understand metrics in Metricbeat logs

edit

Every 30 seconds (by default), Metricbeat collects a snapshot of metrics about itself. From this snapshot, Metricbeat computes a delta snapshot; this delta snapshot contains any metrics that have changed since the last snapshot. Note that the values of the metrics are the values when the snapshot is taken, NOT the difference in values from the last snapshot.

If this delta snapshot contains any metrics (indicating at least one metric that has changed since the last snapshot), this delta snapshot is serialized as JSON and emitted in Metricbeat’s logs at the INFO log level. Here is an example of such a log entry:

{"log.level":"info","@timestamp":"2023-07-14T12:50:36.811Z","log.logger":"monitoring","log.origin":{"file.name":"log/log.go","file.line":187},"message":"Non-zero metrics in the last 30s","service.name":"filebeat","monitoring":{"metrics":{"beat":{"cgroup":{"memory":{"mem":{"usage":{"bytes":0}}}},"cpu":{"system":{"ticks":692690,"time":{"ms":60}},"total":{"ticks":3167250,"time":{"ms":150},"value":3167250},"user":{"ticks":2474560,"time":{"ms":90}}},"handles":{"limit":{"hard":1048576,"soft":1048576},"open":32},"info":{"ephemeral_id":"2bab8688-34c0-4522-80af-db86948d547d","uptime":{"ms":617670096},"version":"8.6.2"},"memstats":{"gc_next":57189272,"memory_alloc":43589824,"memory_total":275281335792,"rss":183574528},"runtime":{"goroutines":212}},"filebeat":{"events":{"active":5,"added":52,"done":49},"harvester":{"open_files":6,"running":6,"started":1}},"libbeat":{"config":{"module":{"running":15}},"output":{"events":{"acked":48,"active":0,"batches":6,"total":48},"read":{"bytes":210},"write":{"bytes":26923}},"pipeline":{"clients":15,"events":{"active":5,"filtered":1,"published":51,"total":52},"queue":{"acked":48}}},"registrar":{"states":{"current":14,"update":49},"writes":{"success":6,"total":6}},"system":{"load":{"1":0.91,"15":0.37,"5":0.4,"norm":{"1":0.1138,"15":0.0463,"5":0.05}}}},"ecs.version":"1.6.0"}}

Details

edit

Focussing on the .monitoring.metrics field, and formatting the JSON, it’s value is:

{
  "beat": {
    "cgroup": {
      "memory": {
        "mem": {
          "usage": {
            "bytes": 0
          }
        }
      }
    },
    "cpu": {
      "system": {
        "ticks": 692690,
        "time": {
          "ms": 60
        }
      },
      "total": {
        "ticks": 3167250,
        "time": {
          "ms": 150
        },
        "value": 3167250
      },
      "user": {
        "ticks": 2474560,
        "time": {
          "ms": 90
        }
      }
    },
    "handles": {
      "limit": {
        "hard": 1048576,
        "soft": 1048576
      },
      "open": 32
    },
    "info": {
      "ephemeral_id": "2bab8688-34c0-4522-80af-db86948d547d",
      "uptime": {
        "ms": 617670096
      },
      "version": "8.6.2"
    },
    "memstats": {
      "gc_next": 57189272,
      "memory_alloc": 43589824,
      "memory_total": 275281335792,
      "rss": 183574528
    },
    "runtime": {
      "goroutines": 212
    }
  },
  "filebeat": {
    "events": {
      "active": 5,
      "added": 52,
      "done": 49
    },
    "harvester": {
      "open_files": 6,
      "running": 6,
      "started": 1
    }
  },
  "libbeat": {
    "config": {
      "module": {
        "running": 15
      }
    },
    "output": {
      "events": {
        "acked": 48,
        "active": 0,
        "batches": 6,
        "total": 48
      },
      "read": {
        "bytes": 210
      },
      "write": {
        "bytes": 26923
      }
    },
    "pipeline": {
      "clients": 15,
      "events": {
        "active": 5,
        "filtered": 1,
        "published": 51,
        "total": 52
      },
      "queue": {
        "acked": 48
      }
    }
  },
  "registrar": {
    "states": {
      "current": 14,
      "update": 49
    },
    "writes": {
      "success": 6,
      "total": 6
    }
  },
  "system": {
    "load": {
      "1": 0.91,
      "5": 0.4,
      "15": 0.37,
      "norm": {
        "1": 0.1138,
        "5": 0.05,
        "15": 0.0463
      }
    }
  }
}

The following tables explain the meaning of the most important fields under .monitoring.metrics and also provide hints that might be helpful in troubleshooting Metricbeat issues.

Field path (relative to .monitoring.metrics) Type Meaning Troubleshooting hints

.beat

Object

Information that is common to all Beats, e.g. version, goroutines, file handles, CPU, memory

.libbeat

Object

Information about the publisher pipeline and output, also common to all Beats

Field path (relative to .monitoring.metrics.beat) Type Meaning Troubleshooting hints

.runtime.goroutines

Integer

Number of goroutines running

If this number grows over time, it indicates a goroutine leak

Field path (relative to .monitoring.metrics.libbeat) Type Meaning Troubleshooting hints

.pipeline.events.active

Integer

Number of events currently in the libbeat publisher pipeline.

If this number grows over time, it may indicate that Metricbeat is producing events faster than the output can consume them. Consider increasing the number of output workers (if this setting is supported by the output; Elasticsearch and Logstash outputs support this setting). The pipeline includes events currently being processed as well as events in the queue. So this metric can sometimes end up slightly higher than the queue size. If this metric reaches the maximum queue size (queue.mem.events for the in-memory queue), it almost certainly indicates backpressure on Metricbeat, implying that Metricbeat may need to temporarily stop ingesting more events from the source until this backpressure is relieved.

.output.events.total

Integer

Number of events currently being processed by the output.

If this number grows over time, it may indicate that the output destination (e.g. Logstash pipeline or Elasticsearch cluster) is not able to accept events at the same or faster rate than what Metricbeat is sending to it.

.output.events.acked

Integer

Number of events acknowledged by the output destination.

Generally, we want this number to be the same as .output.events.total as this indicates that the output destination has reliably received all the events sent to it.

.output.events.failed

Integer

Number of events that Metricbeat tried to send to the output destination, but the destination failed to receive them.

Generally, we want this field to be absent or its value to be zero. When the value is greater than zero, it’s useful to check Metricbeat’s logs right before this log entry’s @timestamp to see if there are any connectivity issues with the output destination. Note that failed events are not lost or dropped; they will be sent back to the publisher pipeline for retrying later.

Useful commands

edit

Parse monitoring metrics from unstructured Metricbeat logs

edit

For Metricbeat versions that emit unstructured logs, the following script can be used to parse monitoring metrics from such logs: https://github.com/elastic/beats/blob/main/script/metrics_from_log_file.sh.