Node Manager Metrics
Every 10 seconds, the collectd service uses a plugin to gather the following Node Manager metrics on each node in the cluster.
Name | Description |
---|---|
mapr.nm.allocated_GB | The amount of memory allocated to the Node Manager in GB. |
mapr.nm.allocated_containers | The number of containers allocated to the Node Manager. |
mapr.nm.allocated_vcores | The number of CPUs allocated to the Node Manager. |
mapr.nm.available_vcores | The number of CPUs available to the Node Manager. |
mapr.nm.available_GB | The amount of memory available to the Node Manager in GB. |
mapr.nm.containers_completed | The number of containers that have completed. |
mapr.nm.containers_failed | The number of containers that have failed. |
mapr.nm.containers_initing | The number of containers that are initializing. |
mapr.nm.containers_killed | The number of containers that have been killed by the Node Manager. |
mapr.nm.containers_running | The number of containers that are running. |
mapr.nm.containers_launched | The number of containers started by the Node Manager. |
mapr.nm.jvm.gc_count | The number of garbage collections. |
mapr.nm.jvm.gc_count_ps_mark_sweep | The number of parallel scavenge mark sweep collections. |
mapr.nm.jvm.gc_count_ps_scavenge | The number of parallel scavenge collections. |
mapr.nm.jvm.gc_time_millis | The amount of time spent on garbage collection in milliseconds. |
mapr.nm.jvm.gc_time_millis_ps_mark_sweep | The amount of time spent on parallel scavenge mark sweep collection in milliseconds. |
mapr.nm.jvm.gc_time_millis_ps_scavenge | The amount of time in milliseconds spent on parallel scavenge collection. |
mapr.nm.jvm.log_error | The total number of ERROR logs. |
mapr.nm.jvm.log_fatal | The total number of FATAL logs. |
mapr.nm.jvm.log_info | The total number of INFO logs |
mapr.nm.jvm.log_warn | The total number of WARN logs. |
mapr.nm.jvm.mem_heap_committed_m | The amount of heap memory committed to the Node Manager in megabytes. |
mapr.nm.jvm.mem_heap_max_m | The maximum amount of heap memory that can be committed to the Node Manager in megabytes. |
mapr.nm.jvm.mem_heap_used_m | The amount of heap memory used by the Node Manager in megabytes. |
mapr.nm.jvm.mem_max_m | The maximum amount of memory that can be committed to the Node Manager in megabytes. |
mapr.nm.jvm.mem_non_heap_committed_m | The amount of non-heap memory committed to the Node Manager in megabytes. |
mapr.nm.jvm.mem_non_heap_max_m | The maximum amount of non-heap memory that can be committed to the Node Manager in megabytes. |
mapr.nm.jvm.mem_non_heap_used_m | The maximum amount of non-heap memory that can be used by the Node Manager in megabytes. |
mapr.nm.jvm.threads_blocked | The number of Node Manager threads in BLOCKED state. |
mapr.nm.jvm.threads_new | The number of Node Manager threads in NEW state. |
mapr.nm.jvm.threads_runnable | The number of Node Manager threads in RUNNABLE state. |
mapr.nm.jvm.threads_terminated | The number of Node Manager threads in TERMINATED state. |
mapr.nm.jvm.threads_time_waiting | The number of Node Manager threads in TIMED_WAITING state. |
mapr.nm.jvm.threads_waiting | The number of Node Manager threads in WAITING state. |
mapr.nm.shuffle.shuffle_connection | The number of Node Manager shuffle connections. |
mapr.nm.shuffle.shuffle_output_bytes | The amount of Node Manager shuffle output in bytes. |
mapr.nm.shuffle.shuffle_outputs_failed | The number of failed Node Manager shuffle outputs. |
mapr.nm.shuffle.shuffle_outputs_ok | The number of completed Node Manager shuffle outputs. |
mapr.nm.ugi.get_groups_avg_time | The average amount of time spent by Node Manager on group resolution. |
mapr.nm.ugi.get_groups_num_ops | The number of group resolutions completed by the Node Manager. |
mapr.nm.ugi.login_failure_avg_time | The average amount of time spent by Node Manager on failed login attempts. |
mapr.nm.ugi.login_failure_num_ops | The number of failed login attempts by the Node Manager. |
mapr.nm.ugi.login_success_avg_time | The average amount of time spent by Node Manager to successfully login. |
mapr.nm.ugi.login_success_num_ops | The number of successful logins by the Node Manager. |