kingluo commented on issue #7949:
URL: https://github.com/apache/apisix/issues/7949#issuecomment-1278415383

   > @kingluo
   > 
   > We also cleaned up this memory, but we cleaned it every minute, but this 
is not enough. When the corresponding route is deleted, we also cleaned up the 
data in dict and key_index, for example:
   > 
   > metric._ key_index:remove(key, ERR_MSG_LRU_EVICTION) metric._ 
dict:delete(key)
   > 
   > Even so, the memory will still rise. I never know where it came out
   
   You really clean up the lookup table? Check my diff above. This lookup table 
would grow indefinitely without bugfix.
   And keep in mind that the memory would grow in a peroid because of pcre 
related cache.
   
   No need to clean up key index and counter cache. The key index would not be 
inserted if shared memory is full, and it would get removed when the mapping in 
shared memory is removed. Counter cache would sync with shared memory and get 
clean up in interval already. I confirm these parts are not related to memory 
leak.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to