Doing a cleanup before every lookup seems to be too aggressive. It can lead to
premature removal of the nodes still usable, which increases the amount of work
under a mutex lock and therefore decreases performance.
In order to improve cleanup behavior, cleanup function call has been moved right
before the allocation of a new node.
ngx_shmtx_lock(&ctx->shpool->mutex);
- ngx_http_limit_req_expire(ctx, 1);
-
rc = ngx_http_limit_req_lookup(lrcf, hash, vv->data, len, &excess);
ngx_shmtx_unlock(&ctx->shpool->mutex);
+ offsetof(ngx_http_limit_req_node_t, data)
+ len;
+ ngx_http_limit_req_expire(ctx, 1);
+
node = ngx_slab_alloc_locked(ctx->shpool, size);
if (node == NULL) {