bandalgomsu
bandalgomsu
Hi @chaoren I’d like to contribute this issue Is it okay if I take this issue and prepare an implementation?
I was planning to prepare a PR using an approach like this (in compute()) ``` V oldValue = valueReference.get(); RemovalCause cause; if (oldValue == null && valueReference.isActive()) { cause =...
Since there hasn’t been any feedback on the comment, I haven’t opened a PR yet. Would it be okay if I create a PR?
Thanks for your feedback 🙂 I agree your comment Then, what do you think about reducing the remove() frequency based on probability or an idle count threshold ??
Hi @tishun @a-TODO-rov ! > (recommended) deprecate existing observePatterns() and observeChannels() methods and implement another one that accepts a custom RedisPubSubAdapter that would allow the user to choose which methods...
What do you think about adding a compareAndSwap() (or compareAndReplace()) method to CacheAccessor? It would allow ClientSideCaching to perform atomic conditional puts and prevent stale cache insertions caused by invalidation...
Hi @tishun !! I’m interested in this issue. May I work on it?
i'll try this issue 😄
Wouldn't it be possible to mitigate this issue by only executing that logic if the inbound_link_freed_time is less than the BlackList TTL?