Agent Metrics List

When you install Stackdriver agents on your VM instances, they transmit some or all of the following metrics. The agents include the Stackdriver Monitoring agent and the Stackdriver Logging agent, which can run on VM instances in Google Cloud Platform (GCP) or Amazon Web Services (AWS).

Some agent functionality might be limited to Stackdriver accounts in the Premium Tier. All new Stackdriver accounts are given a 30-day free trial of the Premium Tier. For more information, see Stackdriver Pricing.

There are additional metrics available directly Google Cloud Platform (GCP) services and from Amazon Web Services (AWS). For a complete list, see Metrics List.

Agent metrics

Metrics from the Stackdriver agents running on VM instances in GCP and AWS.

agent

Metrics from the Monitoring agent and the Logging agent.

The following metric type names must be prefixed with agent.googleapis.com/agent/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
api_request_count
API Request Count
CUMULATIVE, INT64, 1
API request count
state: Request state
log_entry_count
Log Entry Count
CUMULATIVE, INT64, 1
Count of log entry writes
response_code: HTTP response code
log_entry_retry_count
Retried Log Entry Writes Count
CUMULATIVE, INT64, 1
Count of retried log entry writes
response_code: HTTP response code
memory_usage
Agent Memory Usage
GAUGE, DOUBLE, By
Agent memory in use
request_count
API Request Count
CUMULATIVE, INT64, 1
Count of WriteLogEntries requests
response_code: HTTP response code
streamspace_size
Stream Space Used
GAUGE, DOUBLE, By
Estimate of stream space used
streamspace_size_throttling
Agent is Throttling
GAUGE, DOUBLE, 1
Whether the agent is in the throttling state
uptime
Agent Uptime
CUMULATIVE, INT64, s
Agent uptime in seconds
version: Agent version

apache

Metrics from apache.

The following metric type names must be prefixed with agent.googleapis.com/apache/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
connections
Open connections
GAUGE, DOUBLE, 1
The number of active connections currently attached to the HTTP server.
server_name: The name of the Apache HTTP server.
idle_workers
Idle workers
GAUGE, DOUBLE, 1
The number of idle workers currently attached to the HTTP server.
server_name: The name of the Apache HTTP server.
request_count
Requests
CUMULATIVE, INT64, 1
Total requests serviced by the HTTP server.
server_name: The name of the Apache HTTP server.
scoreboard
Scoreboard
GAUGE, DOUBLE, 1
Apache HTTP server scoreboard values.
server_name: The name of the Apache HTTP server.
state: The scoreboard state name.
traffic
Traffic
CUMULATIVE, INT64, By
Total HTTP server traffic.
server_name: The name of the Apache HTTP server.

cassandra

Metrics from cassandra.

The following metric type names must be prefixed with agent.googleapis.com/cassandra/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
cache/hits
Hit count
CUMULATIVE, INT64, 1
Key cache hit count for the table.
cache: The name of the cache.
client_request/latency/50p
Read latency
GAUGE, DOUBLE, us
Latency of standard read requests.
operation: The storage operation name.
client_request/latency/95p
Read latency
GAUGE, DOUBLE, us
Latency of standard read requests.
operation: The storage operation name.
client_request/latency/99p
Read latency
GAUGE, DOUBLE, us
Latency of standard read requests.
operation: The storage operation name.
client_request/latency/max
Read latency
GAUGE, DOUBLE, us
Latency of standard read requests.
operation: The storage operation name.
column_family/compression_ratio
Compression ratio
GAUGE, DOUBLE, 1
Current compression ratio for all SSTables.
column_family/max_row_size
Max row size
GAUGE, DOUBLE, By
Size of the largest compacted row.
commitlog_total_size
Commit log size
GAUGE, DOUBLE, By
Current data size of all commit log segments.
completed_tasks
Completed tasks
CUMULATIVE, INT64, 1
Number of completed tasks in queue.
stage_name: The name of the queue stage.
current_tasks
Tasks
GAUGE, DOUBLE, 1
Number of tasks in queue with the given task status.
stage_name: The name of the queue stage.
task_status: The status of the task (pending, active).
dropped_message/dropped_count
Dropped messages
CUMULATIVE, INT64, 1
Total number of dropped messages for the mutation.
operation: The storage operation name.
storage_service_exception_count
Exceptions
CUMULATIVE, INT64, 1
The number of exceptions thrown by the storage operations.
storage_service_load
Storage load
GAUGE, DOUBLE, By
Total disk space used for this node.

couchdb

Metrics from couchdb.

The following metric type names must be prefixed with agent.googleapis.com/couchdb/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
average_request_time
Request latency
GAUGE, DOUBLE, ms
Average request time.
server_name: The couchdb server name.
httpd/bulk_request_count
Bulk requests
CUMULATIVE, INT64, 1
Bulk requests count.
server_name: The couchdb server name.
httpd/request_count
Requests
CUMULATIVE, INT64, 1
Requests count.
server_name: The couchdb server name.
httpd/request_method_count
Request methods
CUMULATIVE, INT64, 1
HTTP request method count.
server_name: The couchdb server name.
http_method: An HTTP request method.
httpd/response_code_count
Response codes
CUMULATIVE, INT64, 1
HTTP response status code count.
server_name: The couchdb server name.
response_code: An HTTP status code.
httpd/temporary_view_read_count
Temp view reads
CUMULATIVE, INT64, 1
Temporary view reads count.
server_name: The couchdb server name.
httpd/view_read_count
View reads
CUMULATIVE, INT64, 1
View reads count.
server_name: The couchdb server name.
open_databases
Open databases
GAUGE, DOUBLE, 1
Number of open databases.
server_name: The couchdb server name.
open_files
Open files
GAUGE, DOUBLE, 1
Number of open files.
server_name: The couchdb server name.
read_count
Reads
CUMULATIVE, INT64, 1
Database read count.
server_name: The couchdb server name.
write_count
Writes
CUMULATIVE, INT64, 1
Database write count.
server_name: The couchdb server name.

cpu

Metrics from the monitoring agent's default configuration.

The following metric type names must be prefixed with agent.googleapis.com/cpu/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
load_15m
CPU load (15m)
GAUGE, DOUBLE, 1
CPU Load (15 minute intervals)
load_1m
CPU load (1m)
GAUGE, DOUBLE, 1
CPU Load (1 minute intervals)
load_5m
CPU load (5m)
GAUGE, DOUBLE, 1
CPU Load (5 minute intervals)
usage_time
CPU usage
CUMULATIVE, INT64, s
CPU Usage
cpu_number: CPU number (0, 1, 2, etc.)
cpu_state: CPU state (user, system, wait, etc.)
utilization
CPU utilization
GAUGE, DOUBLE, 1
CPU Usage (percent)
cpu_number: CPU number (0, 1, 2, etc.)
cpu_state: CPU state (user, system, wait, etc.)

disk

Metrics from the monitoring agent's default configuration.

The following metric type names must be prefixed with agent.googleapis.com/disk/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
bytes_used
Disk usage
GAUGE, DOUBLE, By
Disk bytes used
device: Device name.
state: Type of usage (free, used, reserved)
io_time
Disk average I/O time
CUMULATIVE, INT64, s
Average time an I/O-operation took to complete
device: Device name.
merged_operations
Disk merged operations
CUMULATIVE, INT64, 1
Merged Operations Count
device: Device name.
direction: Direction (read, write)
operation_count
Disk operations
CUMULATIVE, INT64, 1
Disk Operations Count
device: Device name.
direction: Direction (read, write)
operation_time
Disk operation time
CUMULATIVE, INT64, s
Time Spent In Disk Operations
device: Device name.
direction: Direction (read, write)
pending_operations
Disk pending operations
GAUGE, DOUBLE, 1
Pending Operations Count
device: Device name.
percent_used
Disk utilization
GAUGE, DOUBLE, 1
Disk percent used
device: Device name.
state: Type of usage (free, used, reserved)
read_bytes_count
Disk bytes read
CUMULATIVE, INT64, By
Disk Bytes Read
device: Device name.
weighted_io_time
Disk weighted I/O time
CUMULATIVE, INT64, s
Weighted I/O time
device: Device name.
write_bytes_count
Disk bytes written
CUMULATIVE, INT64, By
Disk Bytes Transferred
device: Device name.

elasticsearch

Metrics from elasticsearch.

The following metric type names must be prefixed with agent.googleapis.com/elasticsearch/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
cache_memory_usage
Cache size
GAUGE, DOUBLE, By
Size in bytes of the caches.
server_name: The elasticsearch server name.
cache_name: Cache name (either 'field' or 'filter').
field_eviction_count
Field evictions
CUMULATIVE, INT64, 1
Evictions from field cache
server_name: The elasticsearch server name.
filter_cache_eviction_count
Filter evictions
CUMULATIVE, INT64, 1
Evictions from filter cache
server_name: The elasticsearch server name.
gc_collection_count
GC count
CUMULATIVE, INT64, 1
Garbage collection count
server_name: The elasticsearch server name.
memory_usage
Memory usage
GAUGE, DOUBLE, By
Size in bytes of memory.
server_name: The elasticsearch server name.
memory_type: The memory type.
network
Network traffic
CUMULATIVE, INT64, By
Number of bytes transmitted and received on the network
server_name: The elasticsearch server name.
direction: Data direction, rx or tx.
num_current_documents
Documents
GAUGE, DOUBLE, 1
Number of documents in the indexes on this node
server_name: The elasticsearch server name.
document_type: Document count type.
num_data_nodes
Data nodes
GAUGE, DOUBLE, 1
Number of data nodes in the cluster
server_name: The elasticsearch server name.
num_http_connections
Open connections
GAUGE, DOUBLE, 1
Number of open HTTP connections to this node
server_name: The elasticsearch server name.
num_nodes
Nodes
GAUGE, DOUBLE, 1
Number of nodes in the cluster
server_name: The elasticsearch server name.
num_open_files
Open files
GAUGE, DOUBLE, 1
Number of open file descriptors held by the server process
server_name: The elasticsearch server name.
num_server_connections
Open connections
GAUGE, DOUBLE, 1
Number of open network connections to the server
server_name: The elasticsearch server name.
num_shards
Shards
GAUGE, DOUBLE, 1
Number of shards.
server_name: The elasticsearch server name.
shard_type: State of the shard (active, relocating, initializing, unassigned).
operation_count
Completed operations
CUMULATIVE, INT64, 1
Number of operations completed
server_name: The elasticsearch server name.
operation: The operation type.
operation_time
Operation time
CUMULATIVE, INT64, ms
Time in ms spent on operations
server_name: The elasticsearch server name.
operation: The operation type.
peak_threads
Max threads
GAUGE, DOUBLE, 1
Maximum number of open threads that have been open concurrently in the server JVM process
server_name: The elasticsearch server name.
storage_size
Storage size
GAUGE, DOUBLE, By
Size in bytes of the document storage on this node
server_name: The elasticsearch server name.
threads
Threads
GAUGE, DOUBLE, 1
Number of open threads in the server JVM process
server_name: The elasticsearch server name.

hbase

Metrics from hbase.

The following metric type names must be prefixed with agent.googleapis.com/hbase/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
ipc/connections
IPC connections
GAUGE, DOUBLE, 1
The number of open connections.
ipc/queue_length
IPC queue size
GAUGE, DOUBLE, 1
The current size of the IPC queue.
ipc/traffic_count
IPC traffic
CUMULATIVE, INT64, By
The number of bytes transmitted and received via IPC.
direction: Data direction, rx or tx.
master/average_load
Load
GAUGE, DOUBLE, 1
The average master load.
master/dead_region_servers
Dead region servers
GAUGE, DOUBLE, 1
The number of dead region servers.
master/live_region_servers
Live region servers
GAUGE, DOUBLE, 1
The number of live region servers.
regionserver/block_cache/access_count
Block cache accesses
CUMULATIVE, INT64, 1
The number of blocks of StoreFiles (HFiles) requested from the cache.
kind: The cache access kind.
regionserver/block_cache/evicted_blocks_count
Evicted block count
CUMULATIVE, INT64, 1
The number of blocks that had to be evicted from the block cache due to heap size constraints.
regionserver/block_cache/hit_ratio_percent
Block cache hit ratio
GAUGE, DOUBLE, 1
The running block cache hit ratio (0 to 100).
regionserver/block_cache/memory
Block cache size
GAUGE, DOUBLE, By
Memory usage by the block cache.
usage_kind: The kind of memory usage.
regionserver/block_cache/num_items
Block count
GAUGE, DOUBLE, 1
The number of blocks of StoreFiles (HFiles) in the cache.
regionserver/call_queue/length
Call queue size
GAUGE, DOUBLE, 1
The length of the call queue.
regionserver/compaction_queue/length
Compaction queue size
GAUGE, DOUBLE, 1
The length of the compaction queue (the number of stores that have been targeted for compaction).
regionserver/flush_queue/length
Flush queue size
GAUGE, DOUBLE, 1
The number of enqueued regions in the MemStore awaiting flush.
regionserver/memory/heap_usage
Heap usage
GAUGE, DOUBLE, By
Heap space used by the regionserver.
regionserver/memstore/files
Memstore files
GAUGE, DOUBLE, 1
The number of store files on the regionserver.
regionserver/memstore/index_size
Memstore index size
GAUGE, DOUBLE, By
The size of the store file index on the regionserver.
regionserver/memstore/open_stores
Memstore open stores
GAUGE, DOUBLE, 1
The number of stores open on the regionserver.
regionserver/memstore/size
Memstore size
GAUGE, DOUBLE, By
The sum of all the memstore sizes in the regionserver.
regionserver/online_regions
Online regions
GAUGE, DOUBLE, 1
The number of active regions in the regionserver.
regionserver/request_count
Request count
CUMULATIVE, INT64, 1
The number of requests.
request_type: The type of request.
regionserver/requests/total_rate
RPC request rate
GAUGE, DOUBLE, 1/s
The point-in-time number of RegionServer RPC calls.
regionserver/slow_operation_count
Slow operations
CUMULATIVE, INT64, 1
The number of slow log operations.
operation: The operation name (e.g., get, put).
thrift/batch_latency/average
Thrift average batch latency
GAUGE, DOUBLE, ns
Thrift latency for batch operations.
operation: The operation name (e.g., get, put).
thrift_id: The id of the thrift instance.
thrift/call_latency/average
Thrift average call latency
GAUGE, DOUBLE, ns
Thrift latency for all calls.
thrift_id: The id of the thrift instance.
thrift/call_queue/length
Thrift call queue size
GAUGE, DOUBLE, 1
Thrift call queue length.
thrift_id: The id of the thrift instance.
thrift/slow_call_latency/average
Thrift average slow call latency
GAUGE, DOUBLE, ns
Thrift latency for slow calls.
thrift_id: The id of the thrift instance.
thrift/time_in_queue/average
Thrift average time in queue
GAUGE, DOUBLE, ns
Thrift time spent in queue.
thrift_id: The id of the thrift instance.

iis

Metrics from the Microsoft Windows monitoring agent's default configuration.

The following metric type names must be prefixed with agent.googleapis.com/iis/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
current_connections
IIS open connections
GAUGE, DOUBLE, 1
Currently open connections to IIS
network/transferred_bytes_count
IIS transferred bytes
CUMULATIVE, INT64, By
Network bytes transferred by IIS
direction: Direction (sent, received)
new_connection_count
IIS connections
CUMULATIVE, INT64, 1
Connections opened to IIS
request_count
IIS requests
CUMULATIVE, INT64, 1
Requests made to IIS
http_method: Http method (post, put, get, delete, option, trace, head)

interface

Metrics from the monitoring agent's default configuration.

The following metric type names must be prefixed with agent.googleapis.com/interface/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
errors
Network errors
CUMULATIVE, INT64, 1
Network error counts.
device: The name of the network device.
direction: The direction of the network traffic.
packets
Network packets
CUMULATIVE, INT64, 1
Packets sent over network.
device: The name of the network device.
direction: The direction of the network traffic.
traffic
Network traffic
CUMULATIVE, INT64, By
Traffic of bytes sent over network.
device: The name of the network device.
direction: The direction of the network traffic.

jvm

Metrics from jvm.

The following metric type names must be prefixed with agent.googleapis.com/jvm/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
gc/count
GC count
CUMULATIVE, INT64, 1
The total number of collections that have occurred.
gc_name: The name of the garbage collector involved.
gc/time
GC time
CUMULATIVE, INT64, ms
The accumulated collection elapsed time in milliseconds.
gc_name: The name of the garbage collector involved.
memory/usage
Memory usage
GAUGE, DOUBLE, By
The memory usage.
memory_type: The type of the memory.
usage_kind: The kind of memory usage.
os/cpu_time
CPU time
CUMULATIVE, INT64, ns
The CPU time used by the process on which the Java virtual machine is running.
os/open_files
Open files
GAUGE, DOUBLE, 1
The number of open file descriptors.
thread/num_daemon
Daemon threads
GAUGE, DOUBLE, 1
The current number of live daemon threads.
thread/num_live
Threads
GAUGE, DOUBLE, 1
The current number of live threads.
thread/peak
Max threads
GAUGE, DOUBLE, 1
The peak number of live threads.
uptime
Uptime
CUMULATIVE, INT64, ms
The uptime of the Java virtual machine.

kafka

Metrics from kafka.

The following metric type names must be prefixed with agent.googleapis.com/kafka/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
broker/topics/failed_request_count
Failed requests
CUMULATIVE, INT64, 1
The number of failed requests.
request_type: The type of the request, fetch or produce.
broker/topics/incoming_message_count
Incoming messages
CUMULATIVE, INT64, 1
The number of incoming messages in all topics.
broker/topics/traffic
Traffic
CUMULATIVE, INT64, By
The number of bytes sent and received in all topics.
direction: Data direction, in or out.
controller/kafka/active
Active controllers
GAUGE, DOUBLE, 1
The number of active controllers in the cluster.
controller/kafka/offline_partitions
Offline partitions
GAUGE, DOUBLE, 1
The number of partitions that don’t have an active leader and are hence not writable or readable.
controller/leader_elections/election_count
Leader elections
CUMULATIVE, INT64, 1
The total number of leader elections.
controller/leader_elections/unclean_count
Stale leader elections
CUMULATIVE, INT64, 1
The total number of leader elections where the leader is out-of-sync.
log/flush_count
Flushes
CUMULATIVE, INT64, 1
The number of log flushes.
network/request_count
Requests
CUMULATIVE, INT64, 1
The number of requests.
request_type: The type of the request, fetch or produce.
purgatory/num_delayed_requests
Delayed purgatory requests
GAUGE, DOUBLE, 1
The current number of delayed requests in purgatory.
request_type: The type of the request, fetch or produce.
purgatory/size
Purgatory requests
GAUGE, DOUBLE, 1
The current number of requests in purgatory.
request_type: The type of the request, fetch or produce.
replica_fetcher/max_lag
Maximum lag
GAUGE, DOUBLE, 1
The maximum lag in messages between the follower and leader replicas.
replica_fetcher/min_fetch_rate
Minimum fetch rate
GAUGE, DOUBLE, 1/s
The minimum rate at which the follower replicas send fetch requests to the leaders.
replica_manager/isr/expand_count
Gaining replicas
CUMULATIVE, INT64, 1
The number of replicas catching up (ISR = in-sync replica).
replica_manager/isr/shrink_count
Lagging replicas
CUMULATIVE, INT64, 1
The number of replicas lagging behind (ISR = in-sync replica).
replica_manager/leaders
Leaders
GAUGE, DOUBLE, 1
The current number of leaders on this broker.
replica_manager/partitions
Partitions
GAUGE, DOUBLE, 1
The current number of partitions on this broker.
replica_manager/under_replicated_partitions
Unreliable partitions
GAUGE, DOUBLE, 1
The current number of under-replicated partitions.

memcached

Metrics from memcached.

The following metric type names must be prefixed with agent.googleapis.com/memcached/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
command_count
Commands
CUMULATIVE, INT64, 1
Commands executed.
cache_name: The memcache name.
command: A memcached command name.
current_connections
Connections
GAUGE, DOUBLE, 1
Number of current connections to the cache.
cache_name: The memcache name.
current_items
Items
GAUGE, DOUBLE, 1
Number of items currently stored in the cache.
cache_name: The memcache name.
eviction_count
Evictions
CUMULATIVE, INT64, 1
Cache item evictions.
cache_name: The memcache name.
memory
Memory usage
GAUGE, DOUBLE, By
Memory usage.
cache_name: The memcache name.
mem_category: Memory free or memory used.
network
Traffic
CUMULATIVE, INT64, By
Bytes transferred over the network.
cache_name: The memcache name.
direction: Data direction, rx or tx.
operation_count
Operations
CUMULATIVE, INT64, 1
Memcached operation hit/miss counts.
cache_name: The memcache name.
operation: A memcached operation name (either 'incr', 'decr', or 'get').
type: A memcached operation statistic (either 'hits' or 'misses')
operation_hitratio
Hit ratio
GAUGE, DOUBLE, percent
Hit ratio for memcached operations.
cache_name: The memcache name.
operation: A memcached operation name (either 'incr', 'decr', or 'get').
rusage
CPU time
CUMULATIVE, INT64, s
Accumulated user and system time.
cache_name: The memcache name.
usage_type: System utilization (either 'user' or 'syst').
threads
Threads
GAUGE, DOUBLE, 1
Number of threads used by the memcached instance.
cache_name: The memcache name.

memory

Metrics from the monitoring agent's default configuration.

The following metric type names must be prefixed with agent.googleapis.com/memory/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
bytes_used
Memory usage
GAUGE, DOUBLE, By
Memory Usage (Bytes)
state: Used, buffered, cached, etc.
percent_used
Memory utilization
GAUGE, DOUBLE, 1
Memory Usage (Percent)
state: Used, buffered, cached, etc.

mongodb

Metrics from mongodb.

The following metric type names must be prefixed with agent.googleapis.com/mongodb/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
cache/hits
Cache hits
CUMULATIVE, INT64, 1
The number of cache hits.
cache/misses
Cache misses
CUMULATIVE, INT64, 1
The number of cache misses.
collections
Collections
GAUGE, DOUBLE, 1
The number of collections.
database_name: The name of the MongoDB database.
connections
Connections
GAUGE, DOUBLE, 1
The number of active database connections.
database_name: The name of the MongoDB database.
data_size
Data size
GAUGE, DOUBLE, By
The data size, in bytes.
database_name: The name of the MongoDB database.
extents
Extents
GAUGE, DOUBLE, 1
The number of extents.
database_name: The name of the MongoDB database.
global_lock_hold_time
Global lock time
CUMULATIVE, INT64, s
The time the global lock has been held.
index_size
Index size
GAUGE, DOUBLE, By
The index size, in bytes.
database_name: The name of the MongoDB database.
indexes
Indexes
GAUGE, DOUBLE, 1
The number of indexes.
database_name: The name of the MongoDB database.
memory_usage
Memory usage
GAUGE, DOUBLE, MiBy
Mebibytes of memory used.
database_name: The name of the MongoDB database.
memory_type: Memory usage (mapped, resident, virtual).
objects
Objects
GAUGE, DOUBLE, 1
The number of objects.
database_name: The name of the MongoDB database.
operation_count
Operations
CUMULATIVE, INT64, 1
The number of operations executed.
operation: A database operation name.
storage_size
Storage size
GAUGE, DOUBLE, By
The storage size, in bytes.
database_name: The name of the MongoDB database.

mssql

Metrics from the Microsoft Windows monitoring agent's default configuration.

The following metric type names must be prefixed with agent.googleapis.com/mssql/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
connections/user
SQL Server open connections
GAUGE, DOUBLE, 1
Currently open connections to SQL server
transaction_rate
SQL Server transaction rate
GAUGE, DOUBLE, 1/s
SQL server total transactions per second
write_transaction_rate
SQL Server write transaction rate
GAUGE, DOUBLE, 1/s
SQL server write transactions per second

mysql

Metrics from mysql.

The following metric type names must be prefixed with agent.googleapis.com/mysql/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
buffer_pool/num_pages
Buffer pool pages
GAUGE, DOUBLE, 1
Buffer pool page count.
database: The name of the database
kind: The buffer pool page kind.
buffer_pool/operation_count
Buffer pool operations
CUMULATIVE, INT64, 1
Buffer pool operation count.
database: The name of the database
operation: The buffer pool operation.
buffer_pool_size
Buffer pool size
GAUGE, DOUBLE, By
Buffer pool size.
database: The name of the database
kind: Buffer pool bytes kind.
command_count
Commands
CUMULATIVE, INT64, 1
MySQL command count.
database: The name of the database
command: MySQL command.
handler_count
Handlers
CUMULATIVE, INT64, 1
MySQL handler count.
database: The name of the database
kind: MySQL handler kind.
innodb/doublewrite_count
InnoDB doublewrite buffers
CUMULATIVE, INT64, 1
InnodDB doublewrite buffer count.
database: The name of the database
kind: TODO
innodb/log_operation_count
InnoDB log operations
CUMULATIVE, INT64, 1
InnoDB log operation count.
database: The name of the database
operation: InnoDB log operation.
innodb/operation_count
InnoDB operations
CUMULATIVE, INT64, 1
InnoDB operation count.
database: The name of the database
operation: InnoDB operation.
innodb/page_operation_count
InnoDB page operations
CUMULATIVE, INT64, 1
InnoDB page operation count.
database: The name of the database
operation: InnoDB page operation.
innodb/row_lock_count
InnoDB locks
CUMULATIVE, INT64, 1
InnoDB row lock count.
database: The name of the database
kind: InnoDB row lock kind.
innodb/row_operation_count
InnoDB row operations
CUMULATIVE, INT64, 1
InnoDB row operation count.
database: The name of the database
operation: InnoDB row operation.
lock_count
Locks
CUMULATIVE, INT64, 1
MySQL lock count.
database: The name of the database
kind: MySQL lock kind.
qcache/operation_count
QCache operations
CUMULATIVE, INT64, 1
QCache operation count.
database: The name of the database
operation: QCache operation.
qcache/query_count
QCache queries
GAUGE, DOUBLE, 1
QCache query count.
database: The name of the database
slave_replication_lag
Replica lag
GAUGE, DOUBLE, s
Seconds behind master.
database: The name of the database
sort_count
Sorts
CUMULATIVE, INT64, 1
MySQL sort count.
database: The name of the database
kind: MySQL sort kind.
thread_count
Threads
GAUGE, DOUBLE, 1
Thread count.
database: The name of the database
kind: Thread kind.

network

Metrics from the monitoring agent's default configuration.

The following metric type names must be prefixed with agent.googleapis.com/network/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
tcp_connections
TCP connections
GAUGE, DOUBLE, 1
Count of TCP connections.
port: 'NNN-local', 'NNN-remote', or 'all'.
tcp_state: TCP state (ESTABLISHED, FIN_WAIT1, etc.)

nginx

Metrics from nginx.

The following metric type names must be prefixed with agent.googleapis.com/nginx/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
connections/accepted_count
Accepted connections
CUMULATIVE, INT64, 1
Total connections accepted.
connections/current
Active connections
GAUGE, DOUBLE, 1
The number of connections currently attached to Nginx.
state: The Nginx connection state.
connections/handled_count
Handled connections
CUMULATIVE, INT64, 1
Total connections handled.
request_count
Requests
CUMULATIVE, INT64, 1
The number of requests Nginx has serviced.

pagefile

Metrics from the Microsoft Windows monitoring agent's default configuration.

The following metric type names must be prefixed with agent.googleapis.com/pagefile/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
percent_used
Pagefile utilization
GAUGE, DOUBLE, 1
Pagefile Usage (Percent)
state: Type of usage (free, used)

postgresql

Metrics from postgresql.

The following metric type names must be prefixed with agent.googleapis.com/postgresql/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
blocks_read_count
Blocks read
CUMULATIVE, INT64, 1
Number of blocks read.
database: The name of the database
source: The block read source.
commit_count
Commits
CUMULATIVE, INT64, 1
Number of commits
database: The name of the database
db_size
DB size
GAUGE, DOUBLE, By
Database disk usage
database: The name of the database
num_backends
Backends
GAUGE, DOUBLE, 1
Number of backends.
database: The name of the database
num_tuples
DB rows
GAUGE, DOUBLE, 1
Number of tuples (rows) in the database.
database: The name of the database
state: Tuple (row) state (either 'dead' or 'live').
operation_count
Operations
CUMULATIVE, INT64, 1
Number of db row operations.
database: The name of the database
operation: The database operation.
rollback_count
Rollbacks
CUMULATIVE, INT64, 1
Number of rollbacks
database: The name of the database

processes

Metrics from the monitoring agent's default configuration.

The following metric type names must be prefixed with agent.googleapis.com/processes/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
count_by_state
Processes
GAUGE, DOUBLE, 1
Count of processes in the given state.
state: Running, sleeping, zombie, etc.
cpu_time
Process CPU
CUMULATIVE, INT64, s
CPU time of the given process.
process: Process name
user_or_syst: Whether a user or system process
command: Process command
command_line: Process command line
owner: Process owner
pid: Process ID
disk/read_bytes_count
Disk I/O
CUMULATIVE, INT64, By
Process disk I/O.
process: Process name
command: Process command
command_line: Process command line
owner: Process owner
pid: Process ID
disk/write_bytes_count
Disk I/O
CUMULATIVE, INT64, By
Process disk I/O.
process: Process name
command: Process command
command_line: Process command line
owner: Process owner
pid: Process ID
fork_count
Fork count
CUMULATIVE, INT64, 1
Total number of processes forked.
vm_usage
Process memory
GAUGE, DOUBLE, By
VM Usage of the given process.
process: Process name
command: Process command
command_line: Process command line
owner: Process owner
pid: Process ID

rabbitmq

Metrics from rabbitmq.

The following metric type names must be prefixed with agent.googleapis.com/rabbitmq/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
consumers
Consumers
GAUGE, DOUBLE, 1
The number of consumers reading from the specified queue
queue: The rabbit queue name.
delivery_rate
Delivery rate
GAUGE, DOUBLE, 1/s
The rate (per second) at which messages are being delivered
queue: The rabbit queue name.
num_messages
Messages
GAUGE, DOUBLE, 1
The number of messages in a queue
queue: The rabbit queue name.
state: The message state.
publish_rate
Publish rate
GAUGE, DOUBLE, 1/s
The rate (per second) at which messages are being published
queue: The rabbit queue name.

redis

Metrics from redis.

The following metric type names must be prefixed with agent.googleapis.com/redis/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
changes_since_last_save
Unsaved changes
GAUGE, DOUBLE, 1
Change count since last save.
node: Redis node name.
clients/blocked
Blocked clients
GAUGE, DOUBLE, 1
Number of blocked clients.
node: Redis node name.
clients/connected
Connected clients
GAUGE, DOUBLE, 1
Number of client connections.
node: Redis node name.
commands_processed
Commands
CUMULATIVE, INT64, 1
The total number of commands processed.
node: Redis node name.
connections/slaves
Slave connections
GAUGE, DOUBLE, 1
Number of slave connections.
node: Redis node name.
connections/total
Connections
CUMULATIVE, INT64, 1
The total number of connections accepted.
node: Redis node name.
expired_keys
Expired keys
CUMULATIVE, INT64, 1
The total number of key expirations.
node: Redis node name.
memory/usage
Memory usage
GAUGE, DOUBLE, By
Memory usage in bytes.
node: Redis node name.
memory/usage_lua
Lua memory usage
GAUGE, DOUBLE, By
Lua memory usage in bytes.
node: Redis node name.
pubsub/channels
PubSub channels
GAUGE, DOUBLE, 1
Number of global pub/sub channels with subscribed clients.
node: Redis node name.
pubsub/patterns
PubSub patterns
GAUGE, DOUBLE, 1
Number of global pub/sub patterns with subscribed clients.
node: Redis node name.
uptime
Uptime
GAUGE, DOUBLE, s
Uptime in seconds.
node: Redis node name.

riak

Metrics from riak.

The following metric type names must be prefixed with agent.googleapis.com/riak/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
latency/95th_percentile
95% latency (1m)
GAUGE, DOUBLE, us
Time between reception of client read request and subsequent response to client (95th percentile).
server_name: The riak server name.
type: An operation type with latency statistics (either 'get' or 'put').
latency/average
Average latency (1m)
GAUGE, DOUBLE, us
Time between reception of client read request and subsequent response to client (mean).
server_name: The riak server name.
type: An operation type with latency statistics (either 'get' or 'put').
latency/maximum
Maximum latency (1m)
GAUGE, DOUBLE, us
Time between reception of client read request and subsequent response to client (maximum).
server_name: The riak server name.
type: An operation type with latency statistics (either 'get' or 'put').
memory_usage
Memory usage
GAUGE, DOUBLE, By
Memory usage.
server_name: The riak server name.
memory_type: Memory usage type.
num_siblings/95th_percentile
95% siblings (1m)
GAUGE, DOUBLE, 1
Number of siblings encountered during all GET operations by this node within the last minute (95th percentile).
server_name: The riak server name.
num_siblings/average
Average siblings (1m)
GAUGE, DOUBLE, 1
Number of siblings encountered during all GET operations by this node within the last minute (mean).
server_name: The riak server name.
num_siblings/maximum
Maixmum siblings (1m)
GAUGE, DOUBLE, 1
Number of siblings encountered during all GET operations by this node within the last minute (maximum).
server_name: The riak server name.
object_size/95th_percentile
95% object size (1m)
GAUGE, DOUBLE, By
Size of objects encountered by this node within the last minute (95th percentile).
server_name: The riak server name.
object_size/average
Average object size (1m)
GAUGE, DOUBLE, By
Size of objects encountered by this node within the last minute (mean).
server_name: The riak server name.
object_size/maximum
Maximum object size (1m)
GAUGE, DOUBLE, By
Size of objects encountered by this node within the last minute (maximum).
server_name: The riak server name.
operation_count
Operations
CUMULATIVE, INT64, 1
Coordinated operation count.
server_name: The riak server name.
operation: An operation name.

swap

Metrics from the monitoring agent's default configuration.

The following metric type names must be prefixed with agent.googleapis.com/swap/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
bytes_used
Swap usage
GAUGE, DOUBLE, By
Swap bytes used, free, cached, and reserved.
device: Swap device.
state: used, free, cached, or reserved.
io
Swap I/O operations
CUMULATIVE, INT64, 1
Swap I/O operations
direction: in, out
percent_used
Swap utilization
GAUGE, DOUBLE, 1
Swap percent used, free, cached, and reserved.
device: Swap device.
state: used, free, cached, or reserved.

tomcat

Metrics from tomcat.

The following metric type names must be prefixed with agent.googleapis.com/tomcat/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
manager/sessions
Sessions
GAUGE, DOUBLE, 1
The number of active sessions.
url_prefix: The URL prefix for a given session.
request_processor/error_count
Errors
CUMULATIVE, INT64, 1
The number of errors encountered.
proto_handler: The protocol handler label.
request_processor/processing_time
Processing time
CUMULATIVE, INT64, s
The total processing time.
proto_handler: The protocol handler label.
request_processor/request_count
Requests
CUMULATIVE, INT64, 1
The total number of requests.
proto_handler: The protocol handler label.
request_processor/traffic_count
Traffic
CUMULATIVE, INT64, By
The number of bytes transmitted and received.
proto_handler: The protocol handler label.
direction: Data direction, rx or tx.
threads/busy
Busy threads
GAUGE, DOUBLE, 1
The current number of busy threads.
proto_handler: The protocol handler label.
threads/current
Current threads
GAUGE, DOUBLE, 1
The current number of threads.
proto_handler: The protocol handler label.

varnish

Metrics from varnish.

The following metric type names must be prefixed with agent.googleapis.com/varnish/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
backend_connection_count
Backend connection successes
CUMULATIVE, INT64, 1
Backend connection success count.
cache_name: The Varnish cache name suffixed with a metric category.
type: The backend connection type (either 'failures', 'not-attempted', 'recycled', 'reuses', 'success', 'too-many', 'unused', or 'was-closed').
category: The Varnish metric category.
cache_operation_count
Cache operations
CUMULATIVE, INT64, 1
Cache operation count.
cache_name: The Varnish cache name suffixed with a metric category.
operation: The cache operation type (either 'hit', 'miss', or 'hitpass').
category: The Varnish metric category.
client_connection_count
Client connections
CUMULATIVE, INT64, 1
Client connection count.
cache_name: The Varnish cache name suffixed with a metric category.
category: The Varnish metric category.
type: The client connection type (either 'accepted', 'dropped', or 'received').

zookeeper

Metrics from zookeeper.

The following metric type names must be prefixed with agent.googleapis.com/zookeeper/. That prefix has been omitted so that the names fit within the table.

Metric type
Display name
Kind, Type, Unit
Description
Labels
connections_count
Open connections
GAUGE, DOUBLE, 1
Number of connections currently alive.
data_size
Data size
GAUGE, DOUBLE, By
Bytes of data on this ZooKeeper server.
followers/count
Followers
GAUGE, DOUBLE, 1
Number of followers.
followers/synced_count
Synced followers
GAUGE, DOUBLE, 1
Number of synchronized followers.
network/received_packets_count
Packets received
CUMULATIVE, INT64, 1
Packets received.
network/sent_packets_count
Packets sent
CUMULATIVE, INT64, 1
Packets sent.
nodes/count
Nodes
GAUGE, DOUBLE, 1
ZooKeeper node count.
nodes/ephemeral_count
Ephemeral nodes
GAUGE, DOUBLE, 1
Number of ephemeral nodes.
nodes/watches_count
Watches
GAUGE, DOUBLE, 1
Number of watches on ZooKeeper nodes.
requests/latency/average
Average request latency
GAUGE, DOUBLE, ms
Average request latency.
requests/latency/maximum
Maximum request latency
GAUGE, DOUBLE, ms
Maximum request latency.
requests/latency/minimum
Minimum request latency
GAUGE, DOUBLE, ms
Minimum request latency.
requests/outstanding_count
Outstanding requests
GAUGE, DOUBLE, 1
Number of outstanding requests.
sync_operations/pending_count
Pending syncs
GAUGE, DOUBLE, 1
Number of pending synchronization operations.

This metric data was last updated at 12/01/17 15:43:55 UTC.

Send feedback about...

Stackdriver Monitoring