Hi Mladen, [EMAIL PROTECTED] schrieb: > + /* Pull shared memory if something changed during sleep */ > + jk_shm_lock(); > + if (p->worker->sequence != p->worker->s->h.sequence) > + jk_lb_pull(p->worker, l); > + jk_shm_unlock();
I think the new lock - pull -unlock blocks are not needed. The pull is not relevant for the runtime states, like busyness or error states. It is only important, after status worker administration changed configuration related items. Retry cycles last typically sub second, so whether a configuration change via the status worker influences a retry or not is not really predictable. I can imagine no scenario, where those additional pulls before retries are really helpful. What do you think? Regards, Rainer --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
