Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
When you have multiple pods connecting to a Redis server, make sure the new connections from the pods are created in a staggered manner. If multiple pods start in a short time without staggering, it causes a sudden spike in the number of client connections created. The high number of connections leads to high load on the Redis server and might cause timeouts.
Avoid the same scenario when shutting down multiple pods at the same time. Failing to stagger shutdown might cause a steep dip in the number of connections that leads to CPU pressure.
Ensure that the pod running your client application is given enough CPU and memory resources. If the client application is running close to its resource limits, it can result in timeouts.
A pod running the client application can be affected by other pods running on the same node and throttle Redis connections or IO operations. So always ensure that the node on which your client application pods run have enough memory, CPU, and network bandwidth. Running low on any of these resources could result in connectivity issues.
If your Azure Cache for Redis client application runs on a Linux-based container, we recommend updating some TCP settings. These settings are detailed in TCP settings for Linux-hosted client applications.