Hey all, We're running spamassassin on a test system via milter and getting some flaky results. The system is RedHat 7.2, sendmail is 8.12.5, spamassassin is 2.01 and 2.30(we've tried both, both have the same problem). The spamassassin install is default except: # cat local.cf defang_mime 0 skip_rbl_checks 1
score CTYPE_JUST_HTML 1 score FROM_ENDS_IN_NUMS 1 Defined in sendmail as: O InputMailFilters=contentfilter, spamassassin Xspamassassin, S=local:/var/run/spamass.sock, F=, T=C:1m;S:2m;R:2m;E:5m Right now the system is in beta so sees little traffic. Things jam along wonderfully until out of the blue we'll see: g69DkAcW001758: Milter (spamassassin): init success to negotiate g69DkAcW001758: Milter: connect to filters g69DkAcW001758: from=wherever g69DkAcW001758: Milter (spamassassin): timeout before data read g69DkAcW001758: Milter (spamassassin): to error state g69DkAcW001758: Milter accept: message g69DkAcW001758: to=whomever, stat=Sent (IAA79604 Message accepted for delivery) g69DkAcW001758: done; delay=00:02:00, ntries=1 This doesn't appear to be related to a load on the system, i.e. it happens when the system is bored looking for something to do as well as when we load it. Once we see these only a sendmail restart corrects the problem. I had thought this was related to the timeouts being set too tight but on a nearly idle system it should be able to connect in less than 1 minute. Size of the email appears immaterial tho' the speed at which they come in seems to have some effect. If we send a heavy stream of test emails at the system the above will happen sooner than if we trickle them in. Anybody else ever see these? -- Randy ([EMAIL PROTECTED]) 715-726-2832 <*> The Penguin Cometh ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
