Hi I have spamassassin 3.1.9 installed with procmail 2.45 on Red Hat 4.1.2-13 kernel 2.6.22.14-72.fc6 .

I have a problem that I can't figure out (or find an explanation for):
One mail user is getting all mail to their account bounced with the following message in maillog:

temporary failure. Command output: procmail: Program failure (-25) of
"/usr/bin/spamassassin" procmail: Rescue of unfiltered data succeeded
procmail: Error while writing to "" procmail: Error while writing to
"/var/mail/lynn.<domain-name>"

(I have replaced the domain name with <domain-name> here)

It seems as though Spamassassin is failing for some reason (I can't find mention of an error 25), then it's fallback is failing to write the message anyway - because the /var/mail/lynn.<domain-name> directory does not exist - I obviously need to change something in the spamassassin config to reflect the actual directory structure which is /var/mail/lynn and I am puzzled by the 'Error writing to "" ' bit....

Actually I am puzzled by all of it - I am a bit of a newbie to managing mail on linux.

Any help would be welcome....
:-)

sorry - correction to prev post:
"I obviously need to change something in the spamassassin config to reflect the actual directory structure which is /var/mail/lynn" should read ".... which is /home/<domain>/homes/lynn"
CP+ is being used on this web/mail server BTW

another correction - I mean PostFix version 2.45 not procmail .....

Just a bit more info:
This error is now occuring on 3 seperate email addresses at this domain.
Looking at cpu use using top it appears that between them, spamassassin processes are taking up around 98% of processor use...

Any ideas? anyone?

Hah - discovered one problem - /etc/postxix/main had mailbox_size_limit set to 11 megs - can't beleive this is the default! All 3 users are on holiday, so this is the first time this limit had been reached...

upped it to half a gig and now the mail is getting through.
Don't know about the other problem though...