How Do I View Current Concurrent Connections and Maximum Connections of a Redis Instance?
l You can view current concurrent connections and maximum connections on the Redis console. For details, see Instance Monitoring Metrics. On the monitoring page, find Number of Client Connections, which indicates the number of current connections of the Redis instance. You can specify a time range to view the metric in a specific monitoring period.
l After an instance is created, you can change the maximum number of connections on the Node Management page of the instance. For more information, see Frontend Configurations. If the limit is exceeded, excess requests will be rejected and connections will time out.
What Should I Do If the Monitoring Data of a Redis Instance Is Abnormal?
If you have any doubts about the monitoring data of a Redis instance, you can run the info all command in the CLI to view the details. Afterwards, Redis returns a string containing multiple different parts and each part provides statistics and configuration data about the server, including metrics recorded by each process. For more information, see Redis Official Website.
How Do I Check the Node Status of a Redis Instance?
On the DCS Redis management console, you can check the node status of an instance. On the instance details page, click the Redis cluster name and on the displayed Redis cluster management page, click Node Management to check the running status of each node. You can also start and stop a node.
What Is the Monitoring Mechanism for Redis Frontends?
Redis provides a mechanism to monitor and manage frontends of a Redis server. Frontends are components running on the front end of a Redis server and handle tasks such as client connection, request distribution, and load balance. You can view the request number, consumed time, client connections, Redis connections and other metrics on the console. For more information, see Frontend Monitoring.
How Do I Check for Memory Fragments?
Redis is an in-memory database, and memory utilization determines its operation efficiency. To monitor the real-time memory usage, Redis provides the info memory command to query the memory usage. mem_fragmentation_ratio indicates the current memory fragmentation ratio of Redis. The memory fragmentation ratio is calculated based on the following formula: mem_fragmentation_ratio = used_memory_rss/used_memory. used_memory_rss is the memory size actually used by users and used_memory is the allocated memory size.
You can also query the memory fragmentation ratio and history memory fragmentation ratios on the metrics page of the console. For more information, see Redis Resource Monitoring. You can set an alarm for memory fragmentation ratio. For more information, see Alarm Configurations.
How Do I Configure Alarm Rules?
Log in to the Redis console, and configure notification groups, notification policies, and alarm rules after choosing Instance Management > Alarm Management. For more information, see Alarm Configurations.
Are Alarm Notifications Sent by SMS or Email?
Alarm notifications can be sent by SMS or Email. Log in to the console. Choose Instance Management > Alarm Management. On the displayed page, configure notification groups, notification policies, and alarm rules. For more information, see Alarm Configurations.
What Alarm Metrics Are Supported?
The following alarm metrics are supported:
Metric Name | Range | Description |
Alarm for too many Redis client connections | 0-10000 | This metric indicates the number of connected Redis clients. |
Low Redis cache hit ratio | 0-100% | This metric indicates the cache hit ratio of Redis. The hit ratio formula is: keyspace_hits/(keyspace_hits+keyspace_misses) %. |
Alarm for the number of blocked clients | 0-10000 | This metric indicates the number of clients that have been suspended by block operations. Block operations include BLPOP, BRPOP, BRPOPLPUSH. |
Redis memory utilization alarm | 0-100% | Indicates memory usage of shards in a Redis instance. |
Redis CPU utilization alarm | 0-100% | Indicates CPU usage of shards in a Redis instance. |
Alarm for the number of access client connections | 0-10000 | This metric indicates the number of clients connected to the proxy node. |
Alarm for too much access QPS | 0-100000 (single proxy node) | This metric indicates the QPS of the proxy node. |