Yandex Managed Service for YTsaurus metrics

Written by

Updated at April 28, 2026

Cluster resource metrics
CPU and memory metrics

This section describes the Managed Service for YTsaurus metrics delivered to Monitoring.

The name label contains the metric name.

Labels shared by all Managed Service for YTsaurus metrics:

Label	Value
service	Service ID: `managed-ytsaurus`
cluster_id	Cluster ID

Cluster resource metrics

Computing resource metrics

Name Type, units	Description
`yt_scheduler_pools_resource_limits_cpu` `DGAUGE`, count	Maximum number of CPUs available within the scheduler pool. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.
`yt_scheduler_distributed_resources_cpu` `DGAUGE`, count	Number of CPUs allocated by the scheduler for current tasks. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `yt-controller-agent` \| `` `tree`: Pool tree ID. Possible values: `default` \| `undefined_tree` \| `` \| `-` `host`: Host name. Possible values: `<host_name>`, `*` \| `-`.
`yt_scheduler_pools_resource_limits_gpu` `DGAUGE`, count	Maximum number of GPUs available within the scheduler pool. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.
`yt_scheduler_distributed_resources_gpu` `DGAUGE`, count	Number of GPUs allocated by the scheduler for current tasks. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `yt-controller-agent` \| `` `tree`: Pool tree ID. Possible values: `default` \| `undefined_tree` \| `` \| `-` `host`: Host name. Possible values: `<host_name>`, `*` \| `-`.
`yt_scheduler_pools_resource_limits_user_memory` `DGAUGE`, bytes	Memory allocation limit for user tasks within the scheduler pool. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.
`yt_scheduler_distributed_resources_user_memory` `DGAUGE`, bytes	Amount of memory allocated for user tasks by the scheduler. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `yt-controller-agent` \| `` `tree`: Pool tree ID. Possible values: `default` \| `undefined_tree` \| `` \| `-` `host`: Host name. Possible values: `<host_name>`, `*` \| `-`.
`yt_scheduler_undistributed_resources_cpu` `DGAUGE`, count	Number of unallocated CPUs available to the scheduler. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `yt-controller-agent` \| `` `tree`: Pool tree ID. Possible values: `default` \| `undefined_tree` \| `` \| `-` `host`: Host name. Possible values: `<host_name>`, `*` \| `-`.
`yt_scheduler_undistributed_resources_user_memory` `DGAUGE`, bytes	Amount of unallocated memory for user tasks available to the scheduler. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `yt-controller-agent` \| `` `tree`: Pool tree ID. Possible values: `default` \| `undefined_tree` \| `` \| `-` `host`: Host name. Possible values: `<host_name>`, `*` \| `-`.

Pool scheduler metrics

Name Type, units	Description
`yt_scheduler_pools_running_operation_count` `DGAUGE`, count	Number of `running` operations within the scheduler pool. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.
`yt_scheduler_pools_lightweight_running_operation_count` `DGAUGE`, count	Number of lightweight `running` operations within the scheduler pool. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.
`yt_scheduler_pools_max_running_operation_count` `DGAUGE`, count	Maximum allowed number of simultaneously `running` operations within the scheduler pool. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.
`yt_scheduler_pools_total_operation_count` `DGAUGE`, count	Total number of all operations within the scheduler pool. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.
`yt_scheduler_pools_max_operation_count` `DGAUGE`, count	Maximum allowed number of operations within the scheduler pool. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.
CPU
`yt_scheduler_pools_resource_usage_cpu` `DGAUGE`, count	Number of CPUs currently used for tasks within the scheduler pool. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.
`yt_scheduler_pools_resource_demand_cpu` `DGAUGE`, count	Number of CPUs required to start all pending tasks within the scheduler pool. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.
`yt_scheduler_pools_effective_strong_guarantee_resources_cpu` `DGAUGE`, count	Number of CPUs guaranteed to be available within the scheduler pool. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.
`yt_scheduler_pools_specified_resource_limits_cpu` `DGAUGE`, count	Maximum number of CPUs set in the scheduler pool configuration. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.
RAM
`yt_scheduler_pools_resource_usage_user_memory` `DGAUGE`, bytes	Current user-task memory utilization within the scheduler pool. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.
`yt_scheduler_pools_resource_demand_user_memory` `DGAUGE`, bytes	Amount of memory for user tasks required to start all pending tasks within the scheduler pool. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.
`yt_scheduler_pools_effective_strong_guarantee_resources_user_memory` `DGAUGE`, bytes	Guaranteed user-task memory within the scheduler pool. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.
`yt_scheduler_pools_specified_resource_limits_user_memory` `DGAUGE`, bytes	Amount of memory for user tasks set in the scheduler pool configuration. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.
GPU
`yt_scheduler_pools_resource_usage_gpu` `DGAUGE`, count	Number of GPUs currently used for tasks within the scheduler pool. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.
`yt_scheduler_pools_resource_demand_gpu` `DGAUGE`, count	Number of GPUs required to start all pending tasks within the scheduler pool. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.
`yt_scheduler_pools_effective_strong_guarantee_resources_gpu` `DGAUGE`, count	Number of GPUs guaranteed to be available within the scheduler pool. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.
`yt_scheduler_pools_specified_resource_limits_gpu` `DGAUGE`, count	Number of GPUs set in the scheduler pool configuration. Additional labels: `service_name`: Cluster component name. Possible values: `yt-scheduler` \| `` `pool`: Pool name. Possible values: `<Root>` \| `research` \| `robot-odin` \| `` \| `-` `tree`: Pool tree ID. Possible values: `default` \| `` `host`: Host name. Possible values: `<host_name>`, `` \| `-`.

Metrics for fault diagnostics

Name Type, units	Description
`yt_node_tracker_offline_node_count` `DGAUGE`, count	Number of nodes in `offline` status.
`yt_node_tracker_banned_node_count` `DGAUGE`, count	Number of nodes in `banned` status.
`yt_node_tracker_decommissioned_node_count` `DGAUGE`, count	Number of nodes in `banned` state.
`yt_cluster_node_alerts` `DGAUGE`, count	Number of active alerts about issues with YTsaurus cluster nodes. The additional `error_code` label may take the following values: `*` `-` `NYT::NBus::EErrorCode::SslError` `NYT::NRpc::EErrorCode::Unavailable` `NYT::NSecurityClient::EErrorCode::AccountLimitExceeded` `NYT::NTabletClient::EErrorCode::NoSuchCell` `cell_did_not_appear_within_timeout`

CPU and memory metrics

Name Type, units	Description
`yt_resource_tracker_total_cpu` `DGAUGE`, count	Total CPUs in the cluster. Additional labels: `service_name`: Cluster component name. `thread`: Stream name. `host`: Host name. Possible values: `<host_name>`, `*` \| `-`.
`yt_security_user_write_time` `DGAUGE`, ms	Total time spent on permission checks by the security system during user-initiated write operations. Additional labels: `user`: Username. `service_name`: Cluster component name. `host`: Host name. Possible values: `<host_name>`, `*` \| `-`.
`yt_security_user_read_time` `DGAUGE`, ms	Total time spent on permission checks by the security system during user-initiated read operations. Additional labels: `user`: Username. `service_name`: Cluster component name. `host`: Host name. Possible values: `<host_name>`, `*` \| `-`.
`yt_resource_tracker_memory_usage_rss` `DGAUGE`, bytes	Total actual RSS (Resident Set Size) memory usage in the cluster. Additional labels: `service_name`: Cluster component name. `host`: Host name. Possible values: `<host_name>`, `*` \| `-`.
`yt_security_user_write_request_count` `DGAUGE`, units per second	Number of user-initiated write requests processed by the security system. Additional labels: `user`: Username. `service_name`: Cluster component name. `host`: Host name. Possible values: `<host_name>`, `*` \| `-`.
`yt_security_user_read_request_count` `DGAUGE`, units per second	Number of user-initiated read requests processed by the security system. Additional labels: `user`: Username. `service_name`: Cluster component name. `host`: Host name. Possible values: `<host_name>`, `*` \| `-`.
`yt_chunk_service_weight_throttler_value` `DGAUGE`, number	Current weight throttler value. Additional labels: `user`: Username. `service_name`: Cluster component name. `host`: Host name. Possible values: `<host_name>`, `*` \| `-`.

Yandex Managed Service for YTsaurus metrics

Cluster resource metricsCluster resource metrics

Computing resource metricsComputing resource metrics

Pool scheduler metricsPool scheduler metrics

Metrics for fault diagnosticsMetrics for fault diagnostics

CPU and memory metricsCPU and memory metrics

Was the article helpful?

Cluster resource metrics

Computing resource metrics

Pool scheduler metrics

Metrics for fault diagnostics

CPU and memory metrics