Strategy regarding clearing KVM cache

Currently all/majority of our APIs use 12 hours as the KVM expiry time. Whenever there is an ad-hoc request to update the KVM to a new value, we are running in to issue with KVM cache. As a workaround, we have created a utlility proxy that does a "PUT" to update the KVM cache. But everything in that proxy is hardcoded since we cannot have mapIdentifier as a variable. It's getting difficult to follow this approach since we have more than 100+ APIs in production. Does anyone have a better or dynamic ways of flushing the KVM cache whenever there is a need?

0 1 150
1 REPLY 1

Not applicable

As per the document only one option is to do PUT operation.

I have an idea, but never tried. The restart of Cassandra and message processors may clear the cache. You can try, if that works or not.