Skip to main content
Sumo Logic

Redis Alerts

Sumo Logic has provided out of the box alerts available via Sumo Logic monitors to help you quickly determine if the Redis database cluster is available and performing as expected.

Sumo Logic has provided out of the box alerts available via Sumo Logic monitors to help you quickly determine if the Redis database cluster is available and performing as expected.

Alert Name Alert Description and conditions Alert Condition Recover Condition
Redis - Instance down This alert fires when we detect that the Redis instance is down for 5 minutes. == 0 >0
Redis - Missing master This alert fires when we detect that a Redis cluster has no node marked as master for 5 minutes. <=0 >0
Redis - Replication Broken This alert fires when we detect that a Redis instance has lost a slave. <0 >=0
Redis - Too Many Connections This alert fires when we detect a given Redis server has too many connections (over 100). >100 <=100
Redis - Rejected Connections This alert fires when we detect that some connections to a Redis cluster have been rejected. >0 <=0
Redis - High Memory Fragmentation Ratio This alert fires when the ration of Redis memory usage to Linux virtual memory pages (mapped to physical memory chunks) is higher than 1.5.
A high ratio will lead to swapping and can adversely affect performance.
>=1.5 <1.5
Redis - Potential Master-Slave Communication Failure This alert fires when we detect that communication between the Redis master and slave nodes has not occurred for the past 60 seconds or more. >=60 <60
Redis - Replica Lag This alert fires when we detect that the replica lag for a given Redis cluster is greater than 60 seconds. Please review how replication has been configured. >60 <=60
Redis - Replication Broken This alert fires when we detect that a Redis instance has lost all slaves. This will affect the redundancy of data stored in Redis. Please review how replication has been configured. <0 >=0
Redis - Replication Offset This alert fires when the replication offset in a given Redis cluster is greater than 1 MB for last five minutes. Please review how replication has been configured. >=1 <1
Redis - Out of memory This alert fires when we detect that a Redis node is running out of memory (> 90%). > 90% < 90%
Redis - High CPU Usgae This alert is fired if user and system cpu usage for a host exceeds 80%. > 80% < 80%