Not solved yet, but confirmed. There is a probability to lose queued mail on the default sendmail configuration, here on nv104 (don't know about other Nevada), whereby the mails without reachable destination are dropped within 15 minutes of entering the queue. It is yet unresolved how it happens, if it is an individual local network configuration issue, or more systematic.
This post is to offer a possible workaround, in case you have a flaky link to your MTAs, like we have to endure: disable the smarthost settings in /etc/mail/sendmail.cf. This is a single line that needs to go: define(`confFALLBACK_SMARTHOST', `mailhost$?m.$m$.')dnl This is the method proposed by John: * cd /etc/mail/cf/cf * cp sendmail.mc `hostname`.mc * delete that line from `hostname`.mc * make `hostname`.cf * cp `hostname`.cf /etc/mail/sendmail.cf * svcadm restart sendmail Afterwards, the behaviour of sendmail here is exactly as expected: it queues the messages until the destination becomes available, or are removed due to aging. I am jumping the queue, so to say, but after I have lost my third batch of e-mails within a few weeks, I seriously like to offer a means to prevent potential similar happenings elsewhere. Let's hope that it is an isolated problem, though! Uwe -- This message posted from opensolaris.org _______________________________________________ sysadmin-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/sysadmin-discuss
