Just when I thought I had my mail system all working properly, I had to go and "upgrade" to potato. Bad move.
I'm using exim to distribute my mail, fetching it from my mail server using fetchmail. But now when I run fetchmail, I get the following error: 4 messages for dparsons at purple.ucdavis.edu. reading message 1 of 4 (1351 header octets) .fetchmail: SMTP connect to localhost failed fetchmail: SMTP transaction error while fetching from purple.ucdavis.edu fetchmail: Query status=10 I can see that I'm connecting properly with the mail server - I have 4 messages waiting. The problem is getting that mail to the mail server on localhost. ie exim on my computer is playing silly buggers. Before I upgraded, I think I got around this by defining "localhost" as a local_domain in the exim configuration, but that doesn't seem to be working now. Now the doc's for fetchmail say [Q. R1 in the FAQ] that the first thing I should check is whether the SMTP mail port (25) on my computer is open, and that I should fix that first. And sure enough, when I try to telnet into port 25 on localhost, I get this error: telnet: Unable to connect to remote host: Connection refused So that must be the problem. But the clever answer to Q. R1 does not explain just how to fix it. Now I have set host_accept_relay = localhost in exim.conf, which I would have thought to have been the solution to this problem, but it doesn't seem to be helping. Can anyone help me? What has happened to exim between slink and potato to cause this sort of problem? exim otherwise seems to work - it sent off this letter alright. And I don't think it's a potato bug, rather it's an unfortunate configuration problem. Thanks, Drew Parsons
