Mail delivery to /var/spool/mail/root stops regularly

General Discussion & Chat
1

So this appears to be an age-old problem where folks report that system generated messages by cron’d shell scripts executed have output that would normally be sent to the screen are redirected to Root’s mailbox /var/spool/mail/root, also linked to /var/mail/root.

So, mine stopped around 0500hrs, based on my observation by reviewing messages delivered to my /var/spool/mail/root file.

  • I reviewed the df output for /var/spool/mail, which resides on the partition /var; it was only at 2% full.

  • I determined that the mailbox_size_limit default setting, using the command
    postconf -d | egrep mailbox_size_limit was = 51,200,000 bytes
    (commas added here for readability) which is approximately 48.828125 MB.

  • I determined that the file /var/spool/mail/root was only 164KB, an order of magnitude smaller than 48MB.

  • I also determined the current limit is also set to the same values (which makes sense because I did not alter it).

I attempted to restart Postfix and the delivery of new mail messages did not start back up.

I observe this problem at work too.

So, what needs to be done to fix this? Why is it happening? A denial of service attack is not a concern for me at my home, but it is at work of course.

2

Update!!

More than 24 hours later, I decided to check on my system’s Root mailbox in /var/spool/mail/root again and the message delivery had ceased again. This time though when I restarted Postfix using systemctl restart, new messages did indeed start flowing in. This is inconsistent with my first experience and observation but I do not know why.

Also, this time after restarting Postfix, the mailbox immediately grew to 1.5 MB in size, still nowhere near the /var filesystem capacity of 90GB of which 89GB is still available.

Any ideas?