[Raw Msg Headers][Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Scheduler behavior

Sorry for the corrupted mail , send it again (hopefully better now:)

Hi ,

I would like to describe a behavior I had with zmailer in one ISP and a work around solution
I did in order to overcome problems I had with the scheduler

I installed a Zmailer machine as front end of an ISP which function as email router,
the destination servers (back-end's) are iplanet, Unfortunately their
Iplanet server contain many disabled mailbox's which return :
"412 MailBox blabla is disabled" , consequently many messages were accumulated
in the /postoffice/queue directory .

The 412 error code cause alot of retry and I could'nt change it to 5XX
I did'nt care about accumulation of those email , but another suprising 
scenario happen - all good emails (valid recipients) to the same domain 
were also accumulated in the qeue , and it was growing to thousands of messages
it took about 30 minutes until good messages arrived to their destination .

After some tries and better to say alot of luck , I run many schedulers processes (20 more)
with "-S -H" and then it begin to clean the queues .

I am looking for a real solution to replace this workaround , I don't understand
few things :

1. How is that messages that has just arrived do not have priority on the queues and waiting
to be processed like the other one (like the one which has already return 4XX error code) ?

2.What can I do to solve this issue in a more convenient way ?

3.Is it possible to make rules according the 400 error code return from by the TA ?

4.Why my workaround to activate more schedulers did improve the outgoing rate ?

The Zmailer installed is 2.99.56 for Red-Hat 9.0



To unsubscribe from this list: send the line "unsubscribe zmailer" in
the body of a message to majordomo@nic.funet.fi