Ed Wilts <[EMAIL PROTECTED]> writes:
[...] > All the tools are provided by Red Hat out of the box. What I do is to first > create each user on my Linux system that I want to provide mail for (all 2 > of us :-)). In that user account, create a .fetchmailrc that goes and gets > the mail. Here's an example: > > set postmaster "ewilts" > set bouncemail > set properties "" > poll mail.attbi.com with proto POP3 > user 'ewilts' there with password 'yourekidding' is ewilts here thanks for the examples [...] snipped cron instructions > You're now in the position of offering pop and imap services to your users. > Simply go to the user's desktop (Outlook, Eudora, whatever) and configure the > pop server to be your Linux system. You can use the IP address or internal > DNS name - I don't offer pop/imap through my firewall but your mileage may > vary. Your smtp server can be the same server you're using today or you can > use your Linux system (which will require extra configuration for sendmail). > In order to have fetchmail actually work, you'll need sendmail installed and > configured since that's how fetchmail typically delivers its mail. > > My wife uses pop and I use imap - it's the end user's choice where and how you > read and manage your mail. > > It works like a charm for me! Thanks, I followed along pretty well I think. Did you make and adjustments to disallow any traffic from the internet on 143? (Or I guess 110 in your case) I wondered how to make it so only 192.XXX.XXX is allowed to connect to it. Or even so that it isn't even seen from the internet. So a scan would not show it open or running. "Anthony E. Greene" <[EMAIL PROTECTED]> writes: [...] > They should probably use it as their outgoing STMP server too. That way if > the DSL connection is down, the mail remains queued on the Linux box until > the connection is reestablished. It also lets you tighten your firewall > rules so that traffic on ports 25 and 110 are allowed only to/from the > Linux box. Yeah, thats the plan. [...] snipped handy information So, with imapd running and accounts with mail in them in /var/spool/mail/$USER. A network computer with a setting of popserver aimed at the linux box would press send/recieve or something and her software would connect to port 143. The imapd would know what to do from there, and handles the uid/passwd negotiation.. and pass out the mail? Do there actually have to be user accounts on the machine, or just a file in /var/spool/mail/$USER? _______________________________________________ Redhat-list mailing list [EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
