To help you monitor your ApsaraDB for ClickHouse clusters in real time, detect issues early, locate faults quickly, and optimize performance, ApsaraDB for ClickHouse provides a rich set of metrics. This topic describes the metrics available for ApsaraDB for ClickHouse clusters.
ClickHouse cluster metrics
-
The following metrics are for clusters purchased after December 1, 2021.
-
For clusters purchased before December 1, 2021, see Metrics of old-version clusters.
Community edition
|
Metric |
Unit |
Description |
|
TairPDBShardingIOBandwidth |
Bytes/s |
The read and write throughput on the disk. |
|
Disk IOPS |
count/s |
The number of read and write operations performed per second on the disk. |
|
Disk Usage |
Bytes |
The amount of used disk space. |
|
disk usage per node |
Bytes |
The total disk usage on a single node (the sum of usage across all its disks). |
|
Disk usage |
Percentage (%) |
The percentage of used disk space. |
|
disk utilization per node |
Percentage (%) |
The percentage of disk space used on each node. |
|
CPU Usage |
Percentage (%) |
The average CPU utilization across all nodes in the cluster. |
|
Inode Usage |
Percentage (%) |
The percentage of used inodes. Note
Inodes are data structures in Linux that store information about files and directories. |
|
Memory Usage |
Bytes |
The total memory used by all nodes in the cluster. |
|
Memory Usage |
Percentage (%) |
The average memory utilization across all nodes in the cluster. |
|
DatapartCount |
Count |
The total number of data parts. |
|
InactiveDatapartCount |
Count |
The number of inactive data parts. |
|
ZooKeeper CPU utilization |
Percentage (%) |
The average CPU utilization across ZooKeeper nodes in the cluster. |
|
ZooKeeper memory utilization |
Percentage (%) |
The average memory utilization across ZooKeeper nodes in the cluster. |
|
Size of Data Written Per Second |
Bytes/s |
The amount of data written per second across all nodes in the cluster. |
|
Rows Written Per Second |
rows/s |
The number of rows written per second across all nodes in the cluster. |
|
QPS |
queries/s |
The number of queries processed per second. |
|
TPS |
transactions/s |
The number of transactions processed per second. |
|
Network throughput |
Bytes/s |
The rate of incoming and outgoing network traffic. |
|
clickhouse_tcp_conn_usage_count |
Count |
The number of TCP connections in the cluster. |
|
HTTPConnUsageCount |
Count |
The number of HTTP connections in the cluster. |
|
clickhouse_init_query_cc |
Count |
The number of running non-forwarded query statements. |
|
clickhouse_query_cc |
Count |
The total number of queries currently being executed. |
|
FailedQuery |
Count |
The total number of queries that failed to execute. |
|
FailedInsertQuery |
Count |
The number of failed INSERT statements. |
|
DelayedInsertsCount |
Count |
The number of delayed INSERT statements. |
|
clickhouse_failed_select_query |
Count |
The number of failed SELECT statements. |
|
MergeCount |
Count |
The number of running merge tasks. |
|
PartMutationCount |
Count |
The number of data mutation tasks in progress. |
|
Number of temporary files for distributed tables |
Count |
The number of temporary files for distributed tables. |
|
Number of MaterializeMySQL databases |
Count |
The number of synced databases created using the MaterializeMySQL engine. |
|
clickhouse_cold_storage_data |
Bytes |
The amount of cold data stored. |
|
Number of MaterializeMySQL synchronization failures |
Count |
The number of synchronization failures for databases created using the MaterializeMySQL engine. |
|
Number of Kafka external table consumption errors |
Count |
The number of consumption errors for Kafka external tables. |
|
ZooKeeper leader |
- |
Identifies the ZooKeeper cluster leader. A value of 1 indicates the current node is the leader. |
|
Number of Kafka external tables |
Count |
The total number of Kafka external tables. |
|
node queue backlog |
Count |
The number of data packets from an ApsaraDB for ClickHouse node in the ZooKeeper send queue. |
|
ZooKeeper average latency |
Milliseconds |
The average request latency for ZooKeeper nodes. |
|
Number of ZooKeeper znodes |
Count |
The total number of znodes in the ZooKeeper cluster. |
|
ZooKeeper TPS |
transactions/s |
The transactions per second (TPS) for ZooKeeper. |
|
zxid utilization |
Percentage (%) |
The utilization of ZooKeeper transaction IDs (zxids). If zxid utilization reaches 100%, a leader re-election is triggered. |
Enterprise Edition
|
Metric |
Unit |
Description |
|
CPU Usage |
Percentage (%) |
The average CPU utilization across all nodes in the cluster. |
|
CCU |
Count |
The quantity of compute resources consumed by the cluster. |
|
Memory Usage |
Bytes |
The total memory usage across all nodes in the cluster. |
|
Memory Usage |
Percentage (%) |
The average memory utilization across all nodes in the cluster. |
|
OssUsedSpace |
Bytes |
The total OSS storage used by the cluster. |
|
DatapartCount |
Count |
The total number of data parts in the cluster. |
|
InactiveDatapartCount |
Count |
The total number of inactive data parts in the cluster. |
|
KeeperCPUUtilization |
Percentage (%) |
The CPU utilization of ClickHouse-Keeper. |
|
KeeperMemoryUsage |
Bytes |
The memory used by ClickHouse-Keeper. |
|
KeeperMemoryUtilization |
Percentage (%) |
The memory utilization of ClickHouse-Keeper. |
|
Size of Data Written Per Second |
Bytes/s |
The amount of data written per second across all nodes in the cluster. |
|
Rows Written Per Second |
rows/s |
The number of rows written per second across all nodes in the cluster. |
|
Network throughput |
Bytes/s |
The combined read and write throughput of the cluster's network interfaces. |
|
OSS network throughput |
Bytes/s |
The combined read and write throughput between the cluster and OSS. |
|
OSS IOPS |
ops/s |
The total number of read and write operations per second between the cluster and OSS. |
|
TPS |
transactions/s |
The total number of transactions processed per second by the cluster. |
|
QPS |
queries/s |
The total number of queries processed per second by the cluster. |
|
clickhouse_init_query_cc |
Count |
The number of running, non-forwarded query statements in the cluster. |
|
clickhouse_query_cc |
Count |
The total number of queries currently running in the cluster. |
|
FailedQuery |
Count |
The total number of failed queries in the cluster. |
|
FailedInsertQuery |
Count |
The total number of failed insert statements in the cluster. |
|
DelayedInsertsCount |
Count |
The total number of delayed insert statements in the cluster. |
|
clickhouse_failed_select_query |
Count |
The total number of failed select queries in the cluster. |
|
PartMutationCount |
Count |
The number of active data mutation tasks in the cluster. |
|
MergeCount |
Count |
The number of active merge tasks in the cluster. |
|
clickhouse_tcp_conn_usage_count |
Count |
The total number of active TCP connections to the cluster. |
|
HTTPConnUsageCount |
Count |
The total number of active HTTP connections to the cluster. |
|
number of failed MV refreshes |
Count |
The total number of failed materialized view refreshes in the cluster. |
|
average MV refresh duration |
s |
The average duration of materialized view refreshes in the cluster. |
|
maximum MV refresh duration |
s |
The maximum duration of materialized view refreshes in the cluster. |
Metrics for old-version clusters
Old-version clusters are ApsaraDB for ClickHouse Community-compatible Edition clusters purchased before December 1, 2021.
|
Metric |
Unit |
Description |
|
CPU Usage |
% |
The CPU utilization of each node in the cluster. |
|
Memory Usage |
% |
The memory utilization of each node in the cluster. |
|
Memory Usage |
Bytes |
The amount of memory used by each node in the cluster. |
|
Disk Usage Ratio |
% |
The percentage of total disk space used. |
|
Disk Usage |
Bytes |
The amount of disk space used. |
|
Disk IOPS |
counts/s |
The number of read and write I/O operations per second. |
|
Disk IOPS |
Bytes/s |
The amount of data read from or written to the disk per second. |
|
Database Connection Usage Ratio |
% |
The percentage of used database connections relative to the maximum number of available connections. |
|
Database Connections |
count |
The number of active database connections. |
|
TPS |
transactions/s |
The number of transactions processed per second. |
|
Rows Written Per Second |
rows/s |
The number of rows written to each node per second. |
|
Size of Data Written Per Second |
Bytes/s |
The amount of data written to each node per second. |
|
QPS |
queries/s |
The number of queries processed per second. |
|
Average Wait Time for ZK |
ms |
The average latency of ZooKeeper requests. |
|
Average I/O Wait Time |
ms |
The average wait time for I/O operations. |
|
Average CPU Wait Time |
ms |
The average time that tasks wait for CPU resources. |