Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
How to reduce size of /var/spool/clientmqueue with no sendmail process running
1,831 views
Skip to first unread message
Bill Mackler
Jul 23, 2013, 6:55:33 PM7/23/13
to
Our server is running Solaris 10. Both sendmail processes are disabled (as shown below). However, mail still can be sent out with the proper mail SMTP server defined in DS at /etc/mail/sendmail.cf, /etc/mail/submit.cf and /etc/mail/cf/cf/sendmail.cf. Unfortunately, our /var/spool/clientmqueue is fast filling up with over 90K mails that went no way or undefined in local host from cron jobs, mail or mailx command.
Is there any way to stop the /var/spool/clientmqueue from filling up since we don’t have “/usr/lib/sendmail –Ac –q15m” process running to flush the files in clientmqueue ? Or if there’s any workaround to clean up the clientmqueue directory, please advise. Thanks, Bill
# svcs -a | grep sendmail
disabled Jan_22 svc:/network/smtp:sendmail
disabled Jan_22 svc:/network/sendmail-client:default
Cydrome Leader
Jul 23, 2013, 10:29:40 PM7/23/13
to
Bill Mackler <underh20.s...@gmail.com> wrote:
>
>
> Our server is running Solaris 10. Both sendmail processes are disabled (as shown below). However, mail still can be sent out with the proper mail SMTP server defined in DS at /etc/mail/sendmail.cf, /etc/mail/submit.cf and /etc/mail/cf/cf/sendmail.cf. Unfortunately, our /var/spool/clientmqueue is fast filling up with over 90K mails that went no way or undefined in local host from cron jobs, mail or mailx command.
>
> Is there any way to stop the /var/spool/clientmqueue from filling up since we don?t have ?/usr/lib/sendmail ?Ac ?q15m? process running to flush the files in clientmqueue ? Or if there?s any workaround to clean up the clientmqueue directory, please advise. Thanks, Bill
>
>
> Our server is running Solaris 10. Both sendmail processes are disabled (as shown below). However, mail still can be sent out with the proper mail SMTP server defined in DS at /etc/mail/sendmail.cf, /etc/mail/submit.cf and /etc/mail/cf/cf/sendmail.cf. Unfortunately, our /var/spool/clientmqueue is fast filling up with over 90K mails that went no way or undefined in local host from cron jobs, mail or mailx command.
>
>
>
>
>
> # svcs -a | grep sendmail
> disabled Jan_22 svc:/network/smtp:sendmail
> disabled Jan_22 svc:/network/sendmail-client:default
Here's how this is solved:
>
>
>
> # svcs -a | grep sendmail
> disabled Jan_22 svc:/network/smtp:sendmail
> disabled Jan_22 svc:/network/sendmail-client:default
svcadm disable sendmail
# on old solaris 10
pkgrm SUNWsndr
pkgrm SUNWsndu
# modern solaris 10
pkgrm SUNWsndmr
pkgrm SUNWsndmu
install postfix.
sendmail is a load of crap by any measure. Everything about it is stupid,
and it does everything the wrong way.
0 new messages