0

I'm having a weird Redis cache timeout on a client. Why weird? It's because the application (a REST service) is running 2 instances on 2 identical VMs thru a load balancer. One box never has any Redis client timeouts, the 2d one is experiencing timeouts at different times of the day. The timeout is set at 7s seconds. but notice this timeout message: Timeout awaiting response (outbound=0KiB, inbound=0KiB, 13047ms elapsed, timeout is 7000ms), command=GET. While most timeout messages occur between 7 and 8 seconds, a number of them happen at much later interval ( see above timeout message). The CPU never goes about 30% and memory is at 80%. Could a hypervisor (since it's a VM) indicate any issues?

Followed thru Redis client timeout recommendations (https://stackexchange.github.io/StackExchange.Redis/Timeout). Also, followed Azure based recommendations for the client timeout issues. Increased timeout from 5 to 7 seconds. No luck. Check Redis server to slow clients. No issues there at all. Updated StackExchange.Redis client library to the latest version.

Improve this question
2
  • Have you checked a) the Redis log or b) the Redis health from the CLI?
    – Tim Biegeleisen
    Commented Nov 15 at 13:14
  • yep. checked redis health. CPU @25%. 0 errors. Besides, if the errors occured in Redis server then errors would appear in both Redis clients. However, only 1 box is having errors and another is having none.
    – Sergey Or
    Commented Nov 15 at 16:54
Related questions
2
Stack Exchange Redis client timeout
958
How do I delete everything in Redis?
1079
How to dispatch a Redux action with a timeout?
Load 6 more related questions Show fewer related questions

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.