Yandex Monitoring metric reference

Written by

Updated at December 2, 2025

CPU metrics
Disk metrics
Disk operation metrics
RAM metrics
Network metrics
Service metrics

This section describes the Yandex Managed Service for Valkey™ metrics delivered to Monitoring.

The name label contains the metric name.

Labels shared by all Yandex Managed Service for Valkey™ metrics:

Label	Value
dc	Availability zone
host	Host FQDN
node	Host type: `master`
resource_id	Cluster ID
resource_type	Resource type: `cluster`
service	Service ID: `managed-redis`
shard	Shard ID
shard_name	Shard name
subcluster_name	Subcluster name

CPU metrics

These metrics show CPU core workload.

The consumption type goes into the systag label.

Name Type, units	Description
`cpu.fraction` `DGAUGE`, %	Guaranteed vCPU performance
`cpu.guarantee` `DGAUGE`, count	Guaranteed number of cores
`cpu.guest_nice` `DGAUGE`, %	CPU core usage, `guest_nice` usage type
`cpu.guest` `DGAUGE`, %	CPU core usage, `guest` usage type
`cpu.idle` `DGAUGE`, %	CPU core usage, `idle` usage type
`cpu.iowait` `DGAUGE`, %	CPU core usage, `iowait` usage type
`cpu.irq` `DGAUGE`, %	CPU core usage, `irq` usage type
`cpu.limit` `DGAUGE`, count	Maximum number of cores in use
`cpu.nice` `DGAUGE`, %	CPU core usage, `nice` usage type
`cpu.softirq` `DGAUGE`, %	CPU core usage, `softirq` usage type
`cpu.steal` `DGAUGE`, %	CPU core usage, `steal` usage type
`cpu.system` `DGAUGE`, %	CPU core usage, `system` usage type
`cpu.user` `DGAUGE`, %	CPU core usage, `user` usage type
`cpu_utilization_by_db_15_limit` `DGAUGE`, %	Maximum VM processor core (vCPU) utilization by a database over 15 seconds (%)
`cpu_utilization_by_db_15` `DGAUGE`, %	Average VM processor core (vCPU) utilization by a database over 15 seconds (%). It ranges from 0% to the vCPU performance level.
`cpu_utilization_by_db_60_limit` `DGAUGE`, %	Maximum VM processor core (vCPU) utilization by a database over 60 seconds (%)
`cpu_utilization_by_db_60` `DGAUGE`, %	Average VM processor core (vCPU) utilization by a database over 60 seconds (%)
`load.avg_15min` `DGAUGE`, %	Average load over 15 minutes
`load.avg_1min` `DGAUGE`, %	Average load over one minute
`load.avg_5min` `DGAUGE`, %	Average load over five minutes

Disk metrics

Name Type, units	Description
`disk.free_bytes` `DGAUGE`, bytes	Free space
`disk.free_inodes` `DGAUGE`, count	Free inodes
`disk.total_bytes` `DGAUGE`, bytes	Available space
`disk.total_inodes` `DGAUGE`, count	Available inodes
`disk.used_bytes` `DGAUGE`, bytes	Used space
`disk.used_inodes` `DGAUGE`, count	Used inodes

Disk operation metrics

Name Type, units	Description
`io.avg.read_bytes` `DGAUGE`, bytes per second	Disk read speed
`io.avg.read_count` `DGAUGE`, operations per second	Number of read operations per second
`io.avg.read_merged_count` `DGAUGE`, operations per second	Number of merged read operations per second
`io.avg.read_time` `DGAUGE`, milliseconds	Total time spent reading from disk
`io.avg.weighted_io_time` `DGAUGE`, milliseconds	Total (weighted) time spent on I/O operations
`io.avg.write_bytes` `DGAUGE`, bytes per second	Disk write speed
`io.avg.write_count` `DGAUGE`, operations per second	Number of writes per second
`io.avg.write_merged_count` `DGAUGE`, operations per second	Number of merged write operations per second
`io.avg.write_time` `DGAUGE`, milliseconds	Total time spent writing to disk
`io.disk*.iops_in_progress` `DGAUGE`, count	Number of disk I/O operations in progress
`io.disk*.merged_reads` `DGAUGE`, count	Number of merged read operations for a given disk
`io.disk*.merged_writes` `DGAUGE`, count	Number of merged writes for a given disk
`io.disk*.read_bytes` `DGAUGE`, bytes per second	Read speed for a given disk
`io.disk*.read_count` `DGAUGE`, operations per second	Number of reads per second for a given disk
`io.disk*.read_merged_count` `DGAUGE`, operations per second	Number of merged read operations per second for a given disk
`io.disk*.read_time` `DGAUGE`, milliseconds	Average read time for a given disk
`io.disk*.utilization` `DGAUGE`, %	Utilization of a given disk
`io.disk*.weighted_io_time` `DGAUGE`, milliseconds	I/O wait time for a given disk
`io.disk*.write_bytes` `DGAUGE`, bytes per second	Write speed for a given disk
`io.disk*.write_count` `DGAUGE`, operations per second	Number of writes per second for a given disk
`io.disk*.write_merged_count` `DGAUGE`, operations per second	Number of merged write operations per second for a given disk
`io.disk*.write_time` `DGAUGE`, milliseconds	Average write time for a given disk

RAM metrics

The consumption type goes into the systag label.

Name Type, units	Description
`mem.active_bytes` `DGAUGE`, bytes	Active resident memory (frequently accessed and released when absolutely necessary)
`mem.available_bytes` `DGAUGE`, bytes	RAM usage, `available` usage type
`mem.available_percent_bytes` `DGAUGE`, %	Percentage of RAM usage, `available` usage type
`mem.buffers_bytes` `DGAUGE`, bytes	RAM usage, `buffers` usage type
`mem.cached_bytes` `DGAUGE`, bytes	RAM usage, `cached` usage type
`mem.commit_limit_bytes` `DGAUGE`, bytes	RAM usage, `commit_limit` usage type
`mem.committed_as_bytes` `DGAUGE`, bytes	RAM usage, `committed_as` usage type
`mem.dirty_bytes` `DGAUGE`, bytes	RAM usage, `dirty` usage type
`mem.free_bytes` `DGAUGE`, bytes	Amount of free RAM available, excluding `mem.buffers_bytes` and `mem.cached_bytes`
`mem.guarantee_bytes` `DGAUGE`, bytes	Guaranteed instance memory
`mem.high_free_bytes` `DGAUGE`, bytes	RAM usage, `high_free` usage type
`mem.high_total_bytes` `DGAUGE`, bytes	RAM usage, `high_total` usage type
`mem.huge_page_size_bytes` `DGAUGE`, bytes	RAM usage, `huge_page_size` usage type
`mem.huge_pages_free_bytes` `DGAUGE`, bytes	RAM usage, `huge_pages_free` usage type
`mem.huge_pages_total_bytes` `DGAUGE`, bytes	RAM usage, `huge_pages_total` usage type
`mem.inactive_bytes` `DGAUGE`, bytes	RAM usage, `inactive` usage type
`mem.limit_bytes` `DGAUGE`, bytes	Maximum allowed instance memory (usage limit)
`mem.low_free_bytes` `DGAUGE`, bytes	RAM usage, `low_free` usage type
`mem.low_total_bytes` `DGAUGE`, bytes	RAM usage, `low_total` usage type
`mem.mapped_bytes` `DGAUGE`, bytes	RAM usage, `mapped` usage type
`mem.page_tables_bytes` `DGAUGE`, bytes	RAM usage, `page_tables` usage type
`mem.shared_bytes` `DGAUGE`, bytes	RAM usage, `shared` usage type
`mem.slab_bytes` `DGAUGE`, bytes	RAM usage, `slab` usage type
`mem.sreclaimable_bytes` `DGAUGE`, bytes	RAM usage, `sreclaimable` usage type
`mem.sunreclaim_bytes` `DGAUGE`, bytes	RAM usage, `sunreclaim` usage type
`mem.swap_cached_bytes` `DGAUGE`, bytes	RAM usage, `swap_cached` usage type
`mem.swap_free_bytes` `DGAUGE`, bytes	RAM usage, `swap_free` usage type
`mem.swap_total_bytes` `DGAUGE`, bytes	RAM usage, `swap_total` usage type
`mem.total_bytes` `DGAUGE`, bytes	RAM usage, `total` usage type
`mem.used_bytes` `DGAUGE`, bytes	Amount of RAM currently used by running processes
`mem.used_percent_bytes` `DGAUGE`, %	Percentage of used RAM
`mem.vmalloc_chunk_bytes` `DGAUGE`, bytes	RAM usage, `vmalloc_chunk` usage type
`mem.vmalloc_total_bytes` `DGAUGE`, bytes	RAM usage, `vmalloc_total` usage type
`mem.vmalloc_used_bytes` `DGAUGE`, bytes	RAM usage, `vmalloc_used` usage type
`mem.write_back_bytes` `DGAUGE`, bytes	RAM usage, `write_back` usage type
`mem.write_back_tmp_bytes` `DGAUGE`, bytes	RAM usage, `write_back_tmp` usage type

Network metrics

Name Type, units	Description
`net.bytes_recv` `DGAUGE`, bytes per second	Network data receive rate
`net.bytes_sent` `DGAUGE`, bytes per second	Network data transmit rate
`net.dropin` `DGAUGE`, count	Dropped receive packets
`net.dropout` `DGAUGE`, count	Dropped transmit packets
`net.errin` `DGAUGE`, count	Receive error count
`net.errout` `DGAUGE`, count	Transmit error count
`net.packets_recv` `DGAUGE`, packets per second	Network packet receive rate
`net.packets_sent` `DGAUGE`, packets per second	Network packet transmit rate
`net.speed` `DGAUGE`, bits per second	Maximum network data transfer rate for this interface

Service metrics

Name Type, units	Description
`active` `DGAUGE`, count	Active resident memory (frequently accessed and released when absolutely necessary).
`available_percent` `DGAUGE`, %	Percentage of RAM usage, `available` usage type
`available` `DGAUGE`, count	RAM usage, `available` usage type
`buffered` `DGAUGE`, bytes	RAM usage, `buffered` usage type
`bytes_recv` `DGAUGE`, bytes	Size of data received
`bytes_sent` `DGAUGE`, bytes	Size of data sent
`cached` `DGAUGE`, bytes	RAM usage, `cached` usage type
`can_read` `DGAUGE`, 0/1	Host read access indicator. `1` if the hosted service is available for reads, `0` if not.
`can_write` `DGAUGE`, 0/1	Host write access indicator. `1` if the hosted service is available for writes, `0` if not.
`cmdstat_<command>_calls` `DGAUGE`, count	Number of the specified completed commands, e.g., EVAL, GET, HGETALL. The `command` special label is the command ID.
`cmdstat_<command>_usec` `DGAUGE`, microseconds	Total running time of the specified commands, e.g., EVAL, GET, HGETALL. The `command` special label is the command ID.
`commit_limit` `DGAUGE`, bytes	RAM usage, `commit_limit` usage type
`committed_as` `DGAUGE`, bytes	RAM usage, `committed_as` usage type
`count` `DGAUGE`, objects	Number of objects in a database
`dirty` `DGAUGE`, bytes	RAM usage, `dirty` usage type
`drop_in` `DGAUGE`, count	Dropped receive packets
`drop_out` `DGAUGE`, count	Dropped transmit packets
`err_in` `DGAUGE`, count	Receive error count
`err_out` `DGAUGE`, count	Transmit error count
`errorstat_<error_type>` `DGAUGE`, errors	Number of errors of the specified type. The `error` special label is the error type ID. The possible values are: `CLUSTERDOWN` `ERR` `MOVED` `NOAUTH` `NOPERM` `WRONGPASS`
`free` `DGAUGE`, bytes	RAM usage, `free` usage type
`high_free` `DGAUGE`, bytes	RAM usage, `high_free` usage type
`high_total` `DGAUGE`, bytes	RAM usage, `high_total` usage type
`hosts.ha` `DGAUGE`, hosts	Number of hosts per high-availability cluster
`hosts.subcluster.<subcluster_name>.total` `DGAUGE`, count	Number of instances per subcluster
`hosts.total` `DGAUGE`, hosts	Total hosts used by the service
`huge_page_size` `DGAUGE`, bytes	RAM usage, `huge_page_size` usage type
`huge_pages_free` `DGAUGE`, bytes	RAM usage, `huge_pages_free` usage type
`huge_pages_total` `DGAUGE`, bytes	RAM usage, `huge_pages_total` usage type
`icmp_inaddrmaskreps` `DGAUGE`, count	Number of ICMP address mask reply messages received from the polled node
`icmp_inaddrmasks` `DGAUGE`, count	Number of ICMP address mask request messages received
`icmp_incsumerrors` `DGAUGE`, count	Total IP packets with checksum errors
`icmp_indestunreachs` `DGAUGE`, count	Number of ICMP destination unreachable messages received
`icmp_inechoreps` `DGAUGE`, count	Number of ICMP echo reply messages received
`icmp_inechos` `DGAUGE`, count	Number of ICMP echo request messages received
`icmp_inerrors` `DGAUGE`, count	Number of bad ICMP packets received, excluding those with checksum errors
`icmp_inmsgs` `DGAUGE`, count	Total ICMP messages received
`icmp_inparmprobs` `DGAUGE`, count	Number of ICMP parameter problem messages received
`icmp_inredirects` `DGAUGE`, count	Number of ICMP redirect messages received
`icmp_insrcquenchs` `DGAUGE`, count	Number of ICMP source quench messages received
`icmp_intimeexcds` `DGAUGE`, count	Number of ICMP time exceeded messages received
`icmp_intimestampreps` `DGAUGE`, count	Number of ICMP timestamp reply messages received
`icmp_intimestamps` `DGAUGE`, count	Number of ICMP timestamp (request) messages received
`icmp_outaddrmaskreps` `DGAUGE`, count	Number of ICMP address mask reply messages sent
`icmp_outaddrmasks` `DGAUGE`, count	Number of ICMP address mask request messages sent
`icmp_outdestunreachs` `DGAUGE`, count	Number of ICMP destination unreachable messages sent
`icmp_outechoreps` `DGAUGE`, count	Number of ICMP echo reply messages sent
`icmp_outechos` `DGAUGE`, count	Number of ICMP echo request messages sent
`icmp_outerrors` `DGAUGE`, count	Number of ICMP messages not sent due to ICMP errors, such as buffer shortages
`icmp_outmsgs` `DGAUGE`, count	Total ICMP messages the object in question attempted to send. This includes all messages counted by `icmp_outerrors`.
`icmp_outparmprobs` `DGAUGE`, count	Number of ICMP parameter problem messages sent
`icmp_outratelimitglobal` `DGAUGE`, messages	Number of ICMP messages throttled by global rate limits
`icmp_outratelimithost` `DGAUGE`, messages	Number of ICMP messages throttled by host-specific rate limits
`icmp_outredirects` `DGAUGE`, count	Number of ICMP redirect messages sent
`icmp_outsrcquenchs` `DGAUGE`, count	Number of ICMP source quench messages sent
`icmp_outtimeexcds` `DGAUGE`, count	Number of ICMP time exceeded messages sent
`icmp_outtimestampreps` `DGAUGE`, count	Number of ICMP timestamp reply messages sent
`icmp_outtimestamps` `DGAUGE`, count	Number of ICMP timestamp (request) messages sent
`icmpmsg_intype<code>` `DGAUGE`, count	Number of received ICMP messages with the specified code
`icmpmsg_outtype<code>` `DGAUGE`, count	Number of sent ICMP messages with the specified code
`inactive` `DGAUGE`, bytes	RAM usage, `inactive` usage type
`inodes_free` `DGAUGE`, count	Free inodes
`inodes_total` `DGAUGE`, count	Available inodes
`inodes_used` `DGAUGE`, count	Used inodes
`instance_userfault_broken` `DGAUGE`, 0/1	Indicator of host failure due to user fault
`iops_in_progress` `DGAUGE`, count	Number of disk I/O operations in progress
`ip_defaultttl` `DGAUGE`, string	Default TTL value inserted into the IP header for IP packets generated in this object when the transport layer protocol does not provide a TTL.
`ip_forwarding` `DGAUGE`	IP forwarding status (SNMP): `0` for disabled, `1` for enabled
`ip_forwdatagrams` `DGAUGE`, count	Number of inbound packets not destined for this object, triggering an attempt to find a forwarding route to the final IP destination. For objects not acting as IP gateways, this count includes only those packets that were source-routed through them and where the source route option was successfully processed.
`ip_fragcreates` `DGAUGE`, count	Number of IP packet fragments generated due to fragmentation in the object in question
`ip_fragfails` `DGAUGE`, count	Number of IP packets dropped due to failed fragmentation which was required in this object, e.g., because of the `Do not fragment` flag being set
`ip_fragoks` `DGAUGE`, count	Number of successfully fragmented IP packets
`ip_inaddrerrors` `DGAUGE`, count	Number of inbound packets dropped due to an invalid destination IP address in the IP header for the object in question. This count includes invalid addresses, such as `0.0.0.0`, and addresses of unsupported classes, e.g., Class E. For objects that do not act as IP gateways and therefore do not forward packets, this counter includes packets dropped because the destination address was not a local one.
`ip_indelivers` `DGAUGE`, count	Total IP packets delivered
`ip_indiscards` `DGAUGE`, count	Number of inbound IP packets encountering no issues preventing their further processing but still dropped, e.g., due to lack of buffer space. This counter does not include packets dropped while awaiting reassembly.
`ip_inhdrerrors` `DGAUGE`, count	Number of inbound packets with header errors
`ip_inreceives` `DGAUGE`, count	Total inbound packets received from interfaces, including those received in error
`ip_inunknownprotos` `DGAUGE`, count	Number of unknown protocol packets received
`ip_outdiscards` `DGAUGE`, count	Number of outgoing IP packets encountering no issues preventing their transmission to the destination but still dropped, e.g., due to lack of buffer space. Note that the counter would include packets counted in `ip_forwdatagrams` if they met this (discretionary) drop criterion.
`ip_outnoroutes` `DGAUGE`, count	Number of IP packets dropped because no route to their destination was found. This metric tracks all packets counted in `ip_forwdatagrams` that meet this no-route criterion. This includes any packets a host cannot redirect because all its default gateways are down.
`ip_outrequests` `DGAUGE`, count	Total IP packets provided by local IP user-level protocols (including ICMP) to IP in requests for transmission. This metric does not include the packets counted in `ip_forwdatagrams`.
`ip_outtransmits` `DGAUGE`, count	Number of successfully sent IP packets
`ip_reasmfails` `DGAUGE`, count	Number of failures detected by the IP reassembly algorithm due to reasons, such as timeouts, errors, etc. This is not necessarily a count of dropped IP fragments, since certain algorithms, like the one in RFC 815, may lose count of fragments while reassembling them as they are received.
`ip_reasmoks` `DGAUGE`, count	Number of IP packets successfully reassembled
`ip_reasmreqds` `DGAUGE`, count	Number of received IP fragments requiring reassembly in the object in question
`ip_reasmtimeout` `DGAUGE`, seconds	Maximum time, in seconds, received fragments are kept while awaiting reassembly in the object in question.
`latency_percentiles_usec_<command>_p<percentile>` `DGAUGE`, microseconds	Command execution time in the given percentile. For example, `latency_percentiles_usec_get_p50` is the median execution time of the GET command, and `latency_percentiles_usec_hmset_p99.9` is the execution time of the HMSET command in the 99.9th percentile. The `command` special label is the command ID.
`latest_backup_sec` `DGAUGE`, seconds	Unix timestamp of the most recent automatic backup
`low_free` `DGAUGE`, bytes	RAM usage, `low_free` usage type
`low_total` `DGAUGE`, bytes	RAM usage, `low_total` usage type
`mapped` `DGAUGE`, bytes	RAM usage, `mapped` usage type
`memory_utilization_by_db_percent_limit` `DGAUGE`, %	Limit on memory utilization by a database
`memory_utilization_by_db_percent` `DGAUGE`, %	Memory utilization by a database
`memory_utilization_by_db_rss` `DGAUGE`, bytes	Total memory used by database processes
`memory_utilization_memory_limit` `DGAUGE`, %	Percentage of used memory in an instance relative to its limit
`merged_reads` `DGAUGE`, count	Number of merged disk reads
`merged_writes` `DGAUGE`, count	Number of merged disk writes
`n_cpus` `DGAUGE`, count	Maximum number of cores in use
`n_unique_users` `DGAUGE`, count	Number of unique users or accounts engaging with the system
`n_users` `DGAUGE`, count	Limit on the number of users
`packets_recv` `DGAUGE`, packets per second	Network packet receive rate
`packets_sent` `DGAUGE`, packets per second	Network packet transmit rate
`page_tables` `DGAUGE`, bytes	RAM usage, `page_tables` usage type
`read_bytes` `DGAUGE`, bytes per second	Read speed for a given disk
`read_count` `DGAUGE`, operations per second	Number of reads per second for a given disk
`read_merged_count` `DGAUGE`, operations per second	Number of merged read operations per second
`read_time` `DGAUGE`, milliseconds	Average disk read time
`redis_acl_access_denied_auth` `DGAUGE`, count	Number of Valkey™ connection attempts failed during authorization with an `access denied` error
`redis_acl_access_denied_channel` `DGAUGE`, count	Number of access denials during channel operation attempts due to an incorrect ACL in Valkey™
`redis_acl_access_denied_cmd` `DGAUGE`, count	Number of access denials when attempting to run commands disabled by ACL policies in Valkey™
`redis_acl_access_denied_key` `DGAUGE`, count	Number of denials of access to keys protected by ACL policies in Valkey™
`redis_aof_last_cow_size` `DGAUGE`, bytes	Size of data copied when creating an AOF file using COW (copy-on-write)
`redis_blocked_clients` `DGAUGE`, count	Number of clients waiting for a connection per cluster host
`redis_child_oom_count` `DGAUGE`, count	Number of Valkey™ child process terminations, e.g., while creating snapshots or forks, due to out-of-memory (OOM) errors
`redis_client_output_normal_hard` `DGAUGE`, bytes	Hard memory usage limit
`redis_client_output_normal_soft` `DGAUGE`, bytes	Soft memory usage limit
`redis_client_recent_max_input_buffer` `DGAUGE`, bytes	Maximum memory used to service client write connections
`redis_client_recent_max_output_buffer` `DGAUGE`, bytes	Maximum memory used to service client read connections
`redis_connected_clients` `DGAUGE`, count	Number of open connections per cluster host. In a sharded or replicated cluster, some connections will be used for data exchange between the cluster hosts. If cluster connection errors occur, perhaps, there are inactive apps that keep connections open for too long. If this is the case, update the settings by editing the Timeout value.
`redis_connected_slaves` `DGAUGE`, count	Number of secondary nodes connected to the master node
`redis_db_hashtable_overhead` `DGAUGE`, bytes	RAM used to store hash tables of all DBs
`redis_dbsize` `DGAUGE`, count	Number of keys stored in all the cluster databases
`redis_evicted_keys` `DGAUGE`, count	Number of keys deleted from memory when inserting new data
`redis_expired_keys` `DGAUGE`, count	Total number of Valkey™ keys automatically deleted after expiration of their TTL
`redis_hit_rate` `DGAUGE`, %	Percentage of requests for which Yandex Managed Service for Valkey™ finds data in the cache
`redis_instantaneous_input_kbps` `DGAUGE`, Kbps	Current incoming traffic to the Valkey™ shard in kilobits per second
`redis_instantaneous_input_repl_kbps` `DGAUGE`, Kbps	Current incoming replication traffic to the Valkey™ shard in kilobits per second
`redis_instantaneous_ops_per_sec` `DGAUGE`, operations per second	Number of requests per second
`redis_instantaneous_output_kbps` `DGAUGE`, Kbps	Current outgoing traffic from the Valkey™ shard in kilobits per second
`redis_instantaneous_output_repl_kbps` `DGAUGE`, Kbps	Current outgoing replication traffic from the Valkey™ shard in kilobits per second
`redis_io_threaded_reads_processed` `DGAUGE`, count	Number of read operations processed using threaded I/O
`redis_io_threaded_writes_processed` `DGAUGE`, count	Number of write operations processed using threaded I/O
`redis_io_threads_active` `DGAUGE`, count	Number of active threads
`redis_is_alive` `DGAUGE`	Host health indicator. `1` if the DB host is operational, `0` if not.
`redis_is_master` `DGAUGE`	Host type indicator. `1` if the host is a database master server, `0` if not.
`redis_keys_db` `DGAUGE`, count	Total number of keys stored in the Valkey™ database on a given shard. The `db` special label is the database number.
`redis_keyspace_hits` `DGAUGE`, count	Number of successful key lookups
`redis_keyspace_misses` `DGAUGE`, count	Number of unsuccessful key lookups
`redis_master_last_io_seconds_ago` `DGAUGE`, seconds	Time elapsed since the last communication with the master host
`redis_maxmemory` `DGAUGE`, bytes	Maximum memory allocated for user data
`redis_mem_aof_buffer` `DGAUGE`, bytes	RAM used for the AOF buffer
`redis_mem_clients_normal` `DGAUGE`, bytes	RAM used to serve external connections
`redis_mem_clients_slaves` `DGAUGE`, bytes	RAM used to serve replication connections
`redis_mem_cluster_links` `DGAUGE`, bytes	Total memory allocated for network connections in cluster mode
`redis_mem_fragmentation_ratio` `DGAUGE`, ratio	Allocated vs. used memory ratio on a given Valkey™ shard (memory fragmentation index)
`redis_mem_replication_backlog` `DGAUGE`, bytes	RAM used for the replication backlog buffer
`redis_memory_limit` `DGAUGE`, bytes	Memory allocated to the host
`redis_module_fork_last_cow_size` `DGAUGE`, bytes	Size of data copied during a `fork()` call using COW
`redis_oom_count` `DGAUGE`, count	Number of out-of-memory errors, per hour
`redis_rdb_last_cow_size` `DGAUGE`, bytes	Size of data copied when creating an RDB file
`redis_repl_backlog_histlen` `DGAUGE`, bytes	Current size of memory used by the replication backlog buffer
`redis_repl_backlog_size` `DGAUGE`, bytes	Maximum size of memory available for the replication backlog buffer
`redis_result_code` `DGAUGE`, code	Result code of the most recent Valkey™ operation on a given shard
`redis_total_commands_processed` `DGAUGE`, count	Total number of commands processed on a given Valkey™ shard since its startup
`redis_total_net_input_bytes` `DGAUGE`, bytes	Total bytes of incoming network traffic to a given Valkey™ shard since its startup
`redis_total_net_output_bytes` `DGAUGE`, bytes	Total bytes of outgoing network traffic from a given Valkey™ shard since its startup
`redis_total_net_repl_input_bytes` `DGAUGE`, bytes	Total bytes of incoming replication network traffic to a given Valkey™ shard since its startup
`redis_total_net_repl_output_bytes` `DGAUGE`, bytes	Total bytes of outgoing replication network traffic from a given Valkey™ shard since its startup
`redis_total_reads_processed` `DGAUGE`, count	Total read operations processed
`redis_total_writes_processed` `DGAUGE`, count	Total write operations processed
`redis_used_cpu_sys_main_thread` `DGAUGE`, seconds	CPU (system) time spent by the main thread of a given Valkey™ shard since its startup
`redis_used_cpu_user_main_thread` `DGAUGE`, seconds	CPU time (in user space) spent by the main thread of a given Valkey™ shard since its startup
`redis_used_memory_dataset` `DGAUGE`, bytes	RAM used for storing data
`redis_used_memory_overhead` `DGAUGE`, bytes	RAM used for internal management and data structures
`redis_used_memory_rss` `DGAUGE`, bytes	Memory used by Valkey™ processes
`redis_used_memory_scripts` `DGAUGE`, bytes	RAM used for storing and running scripts
`redis_used_memory_startup` `DGAUGE`, bytes	RAM utilization for the Valkey™ processes at startup (e.g., after restarting a cluster)
`redis_used_memory` `DGAUGE`, bytes	Actual host memory usage If `redis_used_memory` reaches `redis_maxmemory` when trying to insert new records, Yandex Managed Service for Valkey™ will apply the memory management mode defined by the Maxmemory policy setting
`sentinel_can_read` `DGAUGE`, 0/1	Valkey™ shard read access indicator based on Sentinel data. `1` if the shard is available for reads, `0` if not.
`sentinel_can_write` `DGAUGE`, 0/1	Valkey™ shard write access indicator based on Sentinel data `1` if the shard is available for writes, `0` if not.
`sentinel_connected_clients` `DGAUGE`, count	Number of clients currently connected to a given Valkey™ Sentinel instance
`sentinel_is_alive` `DGAUGE`, 0/1	Valkey™ Sentinel availability indicator. `1` if Sentinel is available, `0` if not.
`sentinel_is_master` `DGAUGE`, 0/1	Indicates whether a particular Valkey™ shard is a master according to Sentinel data. `1` if the shard is a master, `0` if not.
`sentinel_total_error_replies` `DGAUGE`, count	Total error responses clients received from Valkey™ Sentinel since its startup
`shared` `DGAUGE`, bytes	RAM usage, `shared` usage type
`slab` `DGAUGE`, bytes	RAM usage, `slab` usage type
`slowlog_max_len` `DGAUGE`, count	Maximum number of slow requests logged
`slowlog_new` `DGAUGE`, count	New entries in the Valkey™ shard slow log over the most recent monitoring window
`slowlog_top` `DGAUGE`, microseconds	Maximum command execution time among the latest slow log entries on the Valkey™ shard during the most recent monitoring window. The `cmd` special label is the command ID.
`sreclaimable` `DGAUGE`, bytes	RAM usage, `sreclaimable` usage type
`sunreclaim` `DGAUGE`, bytes	RAM usage, `sunreclaim` usage type
`swap_cached` `DGAUGE`, bytes	RAM usage, `swap_cached` usage type
`swap_free` `DGAUGE`, bytes	RAM usage, `swap_free` usage type
`swap_total` `DGAUGE`, bytes	RAM usage, `swap_total` usage type
`tcp_activeopens` `DGAUGE`, count	Number of times TCP connections have made a direct transition from the `CLOSED` state to `SYN-SENT`.
`tcp_attemptfails` `DGAUGE`, count	Number of times TCP connections have made a direct transition to the `CLOSED` state from either the `SYN-SENT` or `SYN-RCVD` state, plus the number of direct TCP connection transitions.
`tcp_currestab` `DGAUGE`, count	Current number of TCP connections for the `ESTABLISHED` or `CLOSE WAIT` states
`tcp_estabresets` `DGAUGE`, count	Number of times TCP connections have made a direct transition to the `CLOSED` state from either the `ESTABLISHED` or `CLOSE-WAIT` state
`tcp_incsumerrors` `DGAUGE`, count	This value increases when a received TCP packet has an invalid checksum
`tcp_inerrs` `DGAUGE`, count	Total segments received with an error, such as invalid TCP checksums
`tcp_insegs` `DGAUGE`, count	Total segments received, including those received in error
`tcp_maxconn` `DGAUGE`, count	Maximum allowed number of TCP connections
`tcp_outrsts` `DGAUGE`, count	Number of TCP segments sent with the `RST` flag
`tcp_outsegs` `DGAUGE`, count	Total TCP segments sent
`tcp_passiveopens` `DGAUGE`, count	Number of times TCP connections have made a direct transition to the `SYN-RCVD` state from `LISTEN`
`tcp_retranssegs` `DGAUGE`, count	Total segments retransmitted, i.e., number of TCP segments containing one or more previously transmitted octets that were retransmitted
`tcp_rtoalgorithm` `DGAUGE`	Algorithm used to determine the timeout for retransmitting unacknowledged octets
`tcp_rtomax` `DGAUGE`, milliseconds	Maximum TCP retransmission timeout, in milliseconds
`tcp_rtomin` `DGAUGE`, milliseconds	Minimum TCP retransmission timeout, in milliseconds
`total` `DGAUGE`, bytes	RAM usage, `total` usage type
`udp_ignoredmulti` `DGAUGE`, count	Number of UDP multicast packets ignored
`udp_incsumerrors` `DGAUGE`, count	This value increases when a received UDP packet contains an invalid kernel code checksum
`udp_indatagrams` `DGAUGE`, count	Total UDP packets received
`udp_inerrors` `DGAUGE`, count	Number of bad UDP packets received, excluding those with checksum errors
`udp_memerrors` `DGAUGE`, errors	Number of UDP out-of-memory errors
`udp_noports` `DGAUGE`, count	Total UDP packets received with no application on the destination port
`udp_outdatagrams` `DGAUGE`, count	Total UDP packets sent from the object in question
`udp_rcvbuferrors` `DGAUGE`, count	Number of UDP receive buffer errors
`udp_sndbuferrors` `DGAUGE`, count	Number of UDP send buffer errors
`udplite_ignoredmulti` `DGAUGE`, count	Number of UDP-Lite multicast packets ignored
`udplite_incsumerrors` `DGAUGE`, count	This value increases when a received UDP-Lite packet contains an invalid kernel code checksum
`udplite_indatagrams` `DGAUGE`, count	Total UDP-Lite packets received
`udplite_inerrors` `DGAUGE`, count	Total UDP-Lite packet receive errors
`udplite_memerrors` `DGAUGE`, errors	Number of UDP-Lite out-of-memory errors
`udplite_noports` `DGAUGE`, count	Total UDP-Lite packets received without a listener on the destination port. Irregularities in this counter may occur when reinitializing the management system and at other times, as indicated by the `udplite_statsdiscontinuitytime` value.
`udplite_outdatagrams` `DGAUGE`, count	Total UDP-Lite packets sent
`udplite_rcvbuferrors` `DGAUGE`, count	This value increases when no memory can be allocated to handle an inbound UDP-Lite packet
`udplite_sndbuferrors` `DGAUGE`, count	This value increases when no memory can be allocated to send a UDP-Lite packet
`uptime` `DGAUGE`, %	Uptime percentage
`usage_guest_nice` `DGAUGE`, %	CPU core usage, `guest_nice` usage type
`usage_guest` `DGAUGE`, %	CPU core usage, `guest` usage type
`usage_idle` `DGAUGE`, %	CPU core usage, `idle` usage type
`usage_iowait` `DGAUGE`, %	CPU core usage, `iowait` usage type
`usage_irq` `DGAUGE`, %	CPU core usage, `irq` usage type
`usage_nice` `DGAUGE`, %	CPU core usage, `nice` usage type
`usage_softirq` `DGAUGE`, %	CPU core usage, `softirq` usage type
`usage_steal` `DGAUGE`, %	CPU core usage, `steal` usage type
`usage_system` `DGAUGE`, %	CPU core usage, `system` usage type
`usage_user` `DGAUGE`, %	CPU core usage, `user` usage type
`used_percent` `DGAUGE`, %	Percentage of RAM usage, `used` usage type
`used` `DGAUGE`, bytes	RAM usage, `used` usage type
`utilization` `DGAUGE`, %	Average VM processor core (vCPU) utilization by a database
`vmalloc_chunk` `DGAUGE`, bytes	RAM usage, `vmalloc_chunk` usage type
`vmalloc_total` `DGAUGE`, bytes	RAM usage, `vmalloc_total` usage type
`vmalloc_used` `DGAUGE`, bytes	RAM usage, `vmalloc_used` usage type
`weighted_io_time` `DGAUGE`, milliseconds	I/O wait time
`write_back_tmp` `DGAUGE`, bytes	RAM usage, `write_back_tmp` usage type
`write_back` `DGAUGE`, bytes	RAM usage, `write_back` usage type
`write_bytes` `DGAUGE`, bytes per second	Disk write speed
`write_count` `DGAUGE`, operations per second	Number of writes per second
`write_merged_count` `DGAUGE`, operations per second	Number of merged write operations per second for a given disk
`write_time` `DGAUGE`, milliseconds	Average disk write time

For more information, see Monitoring cluster and host state.

Yandex Monitoring metric reference

CPU metricsCPU metrics

Disk metricsDisk metrics

Disk operation metricsDisk operation metrics

RAM metricsRAM metrics

Network metricsNetwork metrics