I have a Solaris 10 x86 system. The sendmail has the default system configuration, and the mail worked fine until a few days ago. Right now the local emails (cron, atq, and any emails sent to local users) work fine. Occasionally one or two outside emails got sent, but failed most of the times (no sendmail configuration changes and no other system changes were done during this period). I have restarted the sendmail service, and it won't help at all. The syslog indicates that all the unsent emails are queued, and I can see that these emails are all in /var/spool/clientmqueue directory.
Here are my sendmail processes:
ps -ef | grep sendmail
root 25611 1 0 17:54:46 ? 0:00 /usr/lib/sendmail -bd -q15m -C /etc/mail/local.cf
smmsp 25609 1 0 17:54:46 ? 0:00 /usr/lib/sendmail -Ac -q15m
1. How can I find out why these emails are queued, not sent?
2. How can I force to clear the queued emails?
3. Possible reasons why sendmail doesn't work consistently for sending outside emails?