Yandex Monitoring metric reference
Written by
Updated at June 23, 2025
This section describes Yandex Data Processing metrics delivered to Monitoring.
The metric name goes into the name label.
Labels shared by all Yandex Data Processing metrics:
| Label | Value |
|---|---|
| service | Service ID: data-proc |
| resource_type | Resource type: cluster |
| resource_id | Cluster ID |
| zone_id | Placement zone |
| host | Host FQDN |
HDFS metrics
| Name Type, units |
Description |
|---|---|
dfs.cluster.Free_bytesDGAUGE, bytes |
Space available in HDFS |
dfs.cluster.NonDfsUsedSpace_bytesDGAUGE, bytes |
Space used by data storage subclusters (DataNode), unavailable to HDFS |
dfs.cluster.PercentRemainingDGAUGE, % |
Space available in HDFS |
dfs.cluster.PercentUsedDGAUGE, % |
Space used in HDFS |
dfs.cluster.Total_bytesDGAUGE, bytes |
HDFS size |
dfs.cluster.Used_bytesDGAUGE, bytes |
Space used in HDFS |
Disk metrics
| Name Type, units |
Description |
|---|---|
system.disk.free_bytesDGAUGE, bytes |
Space available in the system storage |
system.disk.inodes_freeDGAUGE, count |
Number of free inodes |
system.disk.inodes_totalDGAUGE, count |
Total inodes |
system.disk.inodes_usedDGAUGE, count |
Number of used inodes |
system.disk.inodes_used_percentDGAUGE, % |
Percentage of used inodes |
system.disk.total_bytesDGAUGE, bytes |
System storage size |
system.disk.used_bytesDGAUGE, bytes |
Used disk space |
system.disk.used_percentDGAUGE, % |
Used disk space |
YARN metrics
| Name Type, units |
Description |
|---|---|
yarn.cluster.activeNodesDGAUGE, count |
Number of active nodes |
yarn.cluster.allocatedMBDGAUGE, megabytes |
Allocated memory |
yarn.cluster.allocatedVirtualCoresDGAUGE, count |
Number of allocated virtual cores |
yarn.cluster.appsCompletedDGAUGE, count |
Successfully completed applications |
yarn.cluster.appsFailedDGAUGE, count |
Failed applications |
yarn.cluster.appsKilledDGAUGE, count |
Killed applications |
yarn.cluster.appsPendingDGAUGE, count |
Enqueued applications |
yarn.cluster.appsRunningDGAUGE, count |
Running applications |
yarn.cluster.appsSubmittedDGAUGE, count |
Started applications |
yarn.cluster.availableMBDGAUGE, megabytes |
Available memory |
yarn.cluster.availableVirtualCoresDGAUGE, count |
Number of available virtual cores |
yarn.cluster.containersAllocatedDGAUGE, count |
Number of allocated containers |
yarn.cluster.containersPendingDGAUGE, count |
Number of enqueued containers |
yarn.cluster.containersReservedDGAUGE, count |
Number of reserved containers |
yarn.cluster.decommissionedNodesDGAUGE, count |
Number of discontinued nodes |
yarn.cluster.decommissioningNodesDGAUGE, count |
Nodes being discontinued |
yarn.cluster.lostNodesDGAUGE, count |
Number of lost nodes |
yarn.cluster.rebootedNodesDGAUGE, count |
Number of rebooted nodes |
yarn.cluster.reservedMBDGAUGE, megabytes |
Reserved memory |
yarn.cluster.reservedVirtualCoresDGAUGE, count |
Number of reserved virtual cores |
yarn.cluster.shutdownNodesDGAUGE, count |
Number of dead nodes |
yarn.cluster.totalAllocatedContainersAcrossPartitionDGAUGE, count |
Containers allocated across partitions |
yarn.cluster.totalMBDGAUGE, megabytes |
Total memory |
yarn.cluster.totalNodesDGAUGE, count |
Total nodes |
yarn.cluster.totalReservedResourcesAcrossPartition_memoryDGAUGE |
Memory reserved across all partitions |
yarn.cluster.totalReservedResourcesAcrossPartition_resourceInformations_resourceInformation_0_maximumAllocationDGAUGE |
Maximum amount of type 0 resources reserved across all partitions |
yarn.cluster.totalReservedResourcesAcrossPartition_resourceInformations_resourceInformation_0_minimumAllocationDGAUGE |
Minimum amount of type 0 resources reserved across all partitions |
yarn.cluster.totalReservedResourcesAcrossPartition_resourceInformations_resourceInformation_0_valueDGAUGE |
Current amount of type 0 resources reserved across all partitions |
yarn.cluster.totalReservedResourcesAcrossPartition_resourceInformations_resourceInformation_1_maximumAllocationDGAUGE |
Maximum amount of type 1 resources reserved across all partitions |
yarn.cluster.totalReservedResourcesAcrossPartition_resourceInformations_resourceInformation_1_minimumAllocationDGAUGE |
Minimum amount of type 1 resources reserved across all partitions |
yarn.cluster.totalReservedResourcesAcrossPartition_resourceInformations_resourceInformation_1_valueDGAUGE |
Current amount of type 1 resources reserved across all partitions |
yarn.cluster.totalReservedResourcesAcrossPartition_vCoresDGAUGE, count |
Virtual cores reserved across all partitions |
yarn.cluster.totalVirtualCoresDGAUGE, count |
Total virtual cores |
yarn.cluster.unhealthyNodesDGAUGE, count |
Unhealthy nodes |
yarn.cluster.utilizedMBPercentDGAUGE, % |
Memory utilization |
yarn.cluster.utilizedVirtualCoresPercentDGAUGE, % |
Virtual core utilization |
Other metrics
| Name Type, units |
Description |
|---|---|
dataproc.cluster.health_statusIGAUGE |
Cluster health and technical condition. Only one metric is delivered to monitoring – it represents the cluster's state and has a value of 1. When the cluster is in a transition state, e.g., being created or stopped, the metric may not be delivered. In which case its value is displayed as -. After the cluster transitions to a new state, the old metric is replaced with a new one, also with a value of 1. |
dataproc.cluster.neededAutoscalingNodesNumberDGAUGE, count |
Yandex Data Processing default scaling metric |