To delete one message from the exim queue
exim -Mrm message-id |
To find the ids of all qued mails
mailq |
If you wish to delete all the messages to a specific email recipient.
exiqgrep -i -r '<mail@example.com>' | xargs exim -Mrm |
Of if you wish to delete all messages that were sent from a specific email address.
exiqgrep -i -f '<mail@example.com>' | xargs exim -Mrm |
To re-configure exim4 on a Debian system just run
“$ dpkg-reconfigure exim4-config”
This will start the configuration process and you will be guided trough the configuration process.
A fairly common issue on the Internet is that some routers can’t handle window scaling.
This will might give some odd messages to email servers (such as not being able to send mails over a certain size to one receiving mail server)
On option is to turn off window scaling by editing sysctl.conf and adding the following lines
# Uncomment to turn off window scaling (most usefull for linux MTA) net.ipv4.tcp_window_scaling = 0 |
A warning about doing this; In the long run throughput will not be optimal on this server, but sending mails should work better (so best used on a dedicated mail server).
Ran into this error today on a server. Some mails got through but some was not getting through.
Check that /etc/hosts has a FQDN (fully qualified domain name).
Also check the size of the mails being sent, in my case the issue was that mail was to large for the receiving server thus this error message was produced at the sender.
sendmail expects to the machine to have a FQDN (fully qualified domain name).
If it does not have that (type “hostname” to check what name the machine has) sendmail will complain about this and take a loot of extra time trying to find it.
To use a FQDN edit “/etc/hosts” and change [something] to [something.yourdomain.com] and this will go away. Updating “/etc/hostname” at the same time might be a good idea; don’t forget to run “/bin/hostname -F /etc/hostname” after updating /etc/hostname.