How To Have A Longer Timeout For Milter When Called From Reprocess Channel? (Doc ID 2401467.1)

Last updated on AUGUST 13, 2019

Applies to:

Goal

Is there a way to have a longer timeout for milter when called from reprocess?

On ingress MUAs (like Webmail), we are thinking we need to have a shorter timeout for the milter callout so that if it fails, it will use a longer timeout when it is retried on the reprocess channel.

Eventually we will have milter servers co-located with all ingress systems. But currently, we have milter servers in Location A and Location B but ingress systems in Location A, B, C, D, and E.

We are already seeing much longer average times for the Location C systems talking to the Location A milters, than Location A systems talking to the same milters (no surprise since it is local). When we start doing this on Webmail in Location D and E, we expect problems with Webmail timing out and/or users getting impatient.

A longer timeout when it is retried via reprocess seems like an easy workaround (until we get milter servers in each of the locations).
 

Solution

My Oracle Support provides customers with access to over a million knowledge articles and a vibrant support community of peers and Oracle experts.