Twice in the last few days, I've gotten a message in my sendmail logs that looks like this: Sep 6 06:59:59 jik2 sendmail[2513]: l81MB8Ms007747: queueup: cannot create ./tfl81MB8Ms007747, uid=0: File exists In both cases, the message occurred while trying to send a bounce message. In both cases, the message being bounced was a piece of spam, and sendmail had already been trying to send the bounce for several days, but it wasn't sending because of course the return address in the spam was bogus. In both cases, sendmail logged a failed attempt to deliver the bounce message immediately before it logged the message shown above: Sep 6 06:59:59 jik2 sendmail[2513]: l81MB8Ms007747: to=<PennysycophantBonilla>, delay=4+12:48:51, xdelay=00:00:01, mailer=esmtp, pri=6422021, relay=mail.phpxref.com. [207.210.228.51], dsn=4.0.0, stat=Deferred: Connection refused by mail.phpxref.com. In both cases, the file name shown in the log message didn't actually exist in /var/spool/mqueue after the fact when I checked (I don't know if it existed at the time the error occurred). Obviously, I'm not terribly concerned that sendmail failed to send a bounce message in response to a spam, but I am concerned that this problem could occur with legitimate as well. I've been keeping my machine up-to-date with Rawhide and I don't think I've seen this error before this past week, despite the fact that thousands of email messages go through my system each week, so I'm suspecting that there's a recent change that's causing this issue.
Which sendmail package version are you using?
sendmail-8.14.1-4.1.fc8
Did this happen after a reboot of the machine?
Do you mean "Did this continue to happen after you rebooted?" or "Was the time you saw this shortly after a reboot?"
Ok, there was a bug in the sendmail package you are using. The stop of the service did not work. Therefore it could be that sendmail was killed while the machine was going down and the mail it was trying to deliver produced the lock file. After the reboot the file was still there... Therefore the question: Was there a reboot of the machine before the problem occured?
Yes. I filed that other bug. It didn't occur to me that the two issues could be related but now that you've explained it, it makes sense.