Qumulo OpenMetrics API Specification

This section lists the names, types, labels, and descriptions for the metrics that Qumulo Core 5.3.0 (and higher) emits in OpenMetrics API format.

The Qumulo OpenMetrics API has a single endpoint that provides a complete view of point-in-time telemetry from Qumulo Core to monitoring systems. These systems, such as Prometheus, can consume the OpenMetrics data format that the Qumulo Core REST API emits without custom code or a monitoring agent. For more information about data formats, see your monitoring system’s documentation.

Accessing Qumulo Metrics

Qumulo metrics are available at the following endpoint.

https://<my-cluster-hostname>:8000/v2/metrics/endpoints/default/data

You can configure a monitoring system that supports the OpenMetrics Specification to use bearer token authentication to access this endpoint.

Metric Types

All Qumulo metrics belong to one of the following OpenMetrics types.

Metric Type	Description
`counter`	An integer that increases monotonically from zero, stored in `<metric_name>_count`. Note During normal operation, the value of `counter` never decreases.
`gauge`	A value that represents a single integer (similar to `counter`), stored in `<metric_name>`. Note During normal operation, the value of a `gauge` metric might increase or decrease.
`histogram`	A representation of a series of buckets, where each bucket tracks values within a specific range. A histogram has a `count` field and a `sum` field, stored in `<metric_name>_count` (the total number of samples) and `<metric_name>_sum` (the sum of all samples). Qumulo Core emits a single bucket that contains all samples. Tip You can use `histogram` metrics to keep track of averages by dividing the `sum` field by the `count` field.
`info`	Informational text about the system, stored in `<metric_name>_info`. An `info` metric always has a value of `1` and labels that contain detailed information.

For more information, see Metric Types in the OpenMetrics Specification.

Metric Labels

The OpenMetrics format allows for metric labeling for communicating additional information. To provide context for metrics, Qumulo Core emits metric-specific labels. For example, the name of a protocol operation or the url of a remote server. For more information, see Available Labels.

Available Metrics

The following table lists metric names, types, labels, and descriptions.

Note
For Qumulo as a Service, all metrics with a node_id label are unavailable because they refer to specific hardware.

Metric Name	Metric Type	Labels	Supported from Qumulo Core Version	Description
`qumulo`	`info`	`max_drive_failures` `max_node_failures` `name` `platform` `service_model` `uuid` `version` Tip Don’t confuse this label for the Qumulo Core version with the identically named label for the kernel version for the `qumulo_kernel` metric.	5.3.0	Qumulo Core information, including the cluster name, cluster UUID, and the current Qumulo Core version.
`qumulo_node`	`info`	`node_id` `node_model`	6.0.2	Information about the nodes in the cluster, including the node ID and the node model
`qumulo_ad_netlogon_request _errors`	`counter`	`domain_url` `server_url`	5.3.0	The total number of Active Directory (AD) `NETLOGON` requests that resulted in an error
`qumulo_ad_netlogon_request _latency_seconds`	`histogram`	`domain_url` `server_url`	5.3.0	The total latency for AD `NETLOGON` requests
`qumulo_ad_netlogon_requests`	`counter`	`domain_url` `server_url`	5.3.0	The total number of completed AD `NETLOGON` operations
`qumulo_cpu_crit_temperature_celsius`	`gauge`	`cpu` `node_id`	7.2.0.2	The critical temperature threshold for each physical CPU
`qumulo_cpu_max_temperature _celsius`	`gauge`	`cpu` `node_id`	5.3.1	The maximum temperature threshold for each physical CPU
`qumulo_cpu_temperature _celsius`	`gauge`	`cpu` `node_id`	5.3.0	The temperature for each physical CPU, in degrees Celsius
`qumulo_disk_endurance _percent`	`gauge`	`disk_type` `drive_bay` `node_id`	5.3.1	The remaining disk endurance value for each disk in the cluster, ranging `100` (no disk wear) to `0` (disk is worn fully)
`qumulo_disk_transport _errors`	`counter`	`disk_type` `drive_bay` `node_id`	5.3.2	The total number of communication errors between the specified drive and its host.
`qumulo_disk_uncorrectable _media_errors`	`counter`	`disk_type` `drive_bay` `node_id`	5.3.2	The total number of uncorrectable errors on the specified drive's physical media.
`qumulo_disk_is_unhealthy`	`gauge`	`disk_type` `drive_bay` `node_id`	5.3.0	The health of each disk in the cluster, ranging from `0` (the disk is healthy) to `1` (the disk is unhealthy)
`qumulo_disk_operation _latency_seconds`	`histogram`	`disk_type` `drive_bay` `io_type` `node_id`	5.3.0	The total latency for disk I/O operations
`qumulo_fan_speed_rpm`	`gauge`	`fan` `node_id`	5.3.0	The fan speed, in RPM
`qumulo_fs_capacity_bytes`	`gauge`	—	5.3.0	The total cluster space, in bytes
`qumulo_fs_directory _tree_entries`	`gauge`	`entry_type` `path`	5.3.0	The number of file system objects on the cluster, sorted by object type
`qumulo_fs_directory _used_bytes`	`gauge`	`path` `usage_type`	5.3.0	The amount of space that object types use, in bytes
`qumulo_fs_free_bytes`	`gauge`	—	5.3.0	The free space on the cluster, in bytes
`qumulo_fs_snapshots`	`gauge`	—	5.3.0	The number of snapshots on the cluster
`qumulo_ldap_lookup _request_errors`	`counter`	`domain_url` `server_url`	5.3.0	The total number of LDAP requests that resulted in an error
`qumulo_ldap_lookup _request_latency_seconds`	`histogram`	`domain_url` `server_url`	5.3.0	The total latency of LDAP requests
`qumulo_ldap_lookup _requests`	`counter`	`domain_url` `server_url`	5.3.0	The total number of completed LDAP requests
`qumulo_ldap_operation _errors`	`counter`	`domain_url`	5.3.0	The total number of LDAP operations that resulted in an error
`qumulo_ldap_operation _latency_seconds`	`histogram`	`domain_url`	5.3.0	The total latency for LDAP operations
`qumulo_ldap_operations`	`counter`	`domain_url`	5.3.0	The total number of completed LDAP operations
`qumulo_memory_correctable _ecc_errors`	`counter`	`node_id`	5.3.0	The total number of memory errors that Qumulo Core corrected automatically
`qumulo_network_interface _is_down`	`gauge`	`bond` `interface` `role` `node_id`	5.3.0	The interface status, `0` (interface is up) or `1` (interface is down)
`qumulo_network_interface _link_speed_bits_per_second`	`gauge`	`bond` `interface` `role` `node_id`	5.3.0	The negotiated link speed for the specified interface
`qumulo_network_interface _receive_errors`	`counter`	`bond` `interface` `role` `node_id`	5.3.0	The total number of receive errors on the specified interface
`qumulo_network_interface _received_bytes`	`counter`	`bond` `interface` `role` `node_id`	5.3.0	The total bytes received on the specified interface
`qumulo_network_interface _received_packets`	`counter`	`bond` `interface` `role` `node_id`	5.3.0	The total number of packets received on the specified interface
`qumulo_network_interface _transmit_errors`	`counter`	`bond` `interface` `role` `node_id`	5.3.0	The total number of transmission errors on the specified interface
`qumulo_network_interface _transmitted_bytes`	`counter`	`bond` `interface` `role` `node_id`	5.3.0	The total number of bytes transmitted on the specified interface
`qumulo_network_interface _transmitted_packets`	`counter`	`bond` `interface` `role` `node_id`	5.3.0	The total number of packets transmitted on the specified interface
`qumulo_power_supply _is_unhealthy`	`gauge`	`location` `node_id`	5.3.0	PSU health, `0` (healthy) or `1` (unplugged, removed, or unresponsive)
`qumulo_protocol_client _connections`	`counter`	`protocol`	5.3.0	The total number of clients that have connected to the specified protocol
`qumulo_protocol_client _disconnections`	`counter`	`protocol`	5.3.0	The total number of clients that have disconnected from the specified protocol
`qumulo_protocol_operation _bytes`	`counter`	`data_type` `io_type` `op_name` `protocol`	5.3.0	The total bytes that protocol operations have transferred
`qumulo_protocol_operation _latency_seconds`	`histogram`	`data_type` `io_type` `op_name` `protocol`	5.3.0	The total latency for protocol operations
`qumulo_protocol_operations`	`counter`	`data_type` `io_type` `op_name` `protocol`	5.3.0	The total number of completed protocol operations
`qumulo_quorum_node_is _offline`	`gauge`	`node_id`	5.3.0	The online status for each node in the cluster, `0` (node online) or `1` (node offline)
`qumulo_time_is_not_synchronizing`	`gauge`	`node_id`	5.3.0	The time synchronization status for each node in the cluster, `0` (time is synchronized) or `1` (time isn't synchronized)

Available Labels

The following table lists metric label names, possible values, and descriptions.

Label Name	Possible Values	Description
`bond`	`bond0` `bond1`	The bond to which a network interface belongs
`cpu`	A non-negative integer	The CPU index in the node
`data_type`	`data`: Read or write operations on the data of a file. `metadata`: Operations (such as `lookup`, `stat`, or `getattr`) unrelated to a file's data `none`: Operations that operate on neither the file data nor the metadata. Note The protocol often requires these operations for session negotiation and authentication.	The data type that an operation transfers
`disk_type`	`hdd`: Hard Disk Drive `ssd`: Solid-State Drive	The underlying storage type
`domain_url`	An Active Directory domain (for example, `my-domain.com`) or an LDAP bind URI (for example, `ldap://my-server.my-domain.com`)	The URL of the domain
`drive_bay`	A drive bay name. For example: `b3`, `1.1`	The physical drive bay in the chassis.
`entry_type`	`alternate_data_stream` `directory` `file` `other` `symlink`	The file system object type
`fan`	A fan name, for example `system fan 1`	The fan name
`interface`	An interface name, for example `eth0`	The interface name
`io_type`	`composite` `none` `read` `wait`: A blocking operation that takes an indeterminate amount of time `write`	The I/O that an operation performs
`location`	A location on the chassis, for example `left` or `right`	The location on the chassis. Note For PSU, this location is relative to the back of the node.
`node_id`	A positive integer that represents a node ID in the cluster.	A value that differentiates between the different nodes in a cluster
`op_name`	Any operation name, including NFSv3, NFSv4.1, SMBv2, SMBv3, REST, S3, replication, or FTP	The recorded operation
`path`	Slash (`/`)	The path to a directory in the file system
`protocol`	`nfs`: NFSv3 or NFSv4.1 `smb`: SMBv2 or SMBv3 `rest` `s3` `replication` `ftp`	The protocol of the recorded operation
`role`	`frontend` `backend`	The role of the interface Note `frontend` includes protocol, management, and replication traffic. `backend` includes all intra-node communications.
`server_url`	A hostname (for example, `ad.my-domain.com`) or an IP address	The URL of a remote server
`usage_type`	`data` `metadata` `snapshot`	The data type that uses space