Date: Wed, 17 May 2000 07:39:05 +0200 From: Marc Silver <marcs@draenor.org> To: Nathan Vidican <webmaster@wmptl.com> Cc: freebsd-questions@freebsd.org Subject: Re: popper causing headaches Message-ID: <20000517073905.D58332@draenor.org> In-Reply-To: <3921B07C.DC3E59A2@wmptl.com>; from webmaster@wmptl.com on Tue, May 16, 2000 at 04:33:00PM -0400 References: <3921B07C.DC3E59A2@wmptl.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Hey there, From looking at the FAQ for Qpopper I found the following: Q: I see errors such as 'POP EOF', 'SIGHUP or SIGPIPE flagged', or 'POP hangup'. A: Generally, this is because a client has disconnected without sending QUIT. This can be the result of telephone modem problems, which are more likely to occur when downloading large messages. It could also be caused by too-small timeout values in some clients. Secondly, I'd suggest that if you don't want to add reverse DNS for your internal network, then at least add entries in /etc/hosts so that the mail server (and you) has an idea of who's connecting. Hope this helps you, Cheers, Marc On Tue, May 16, 2000 at 04:33:00PM -0400, Nathan Vidican wrote: > Our email service, or more specifically, our pop3 service just died. We > have no idea what caused it, but the following error was displayed on > the console about 50+ times for different IP addresses as people tried > to check their email: > > mail# popper[pid]: @(192.168.x.x): -ERR POP EOF received > > This error has since ceased to produce itself, and things are back up > and running. This still bother's me though, partially because I don't > completely understand the error, and partially because I don't know what > exactly would have caused it, nor do I know how to stop it from occuring > again. > Beyond these errors, there are others that have been appearing > all-along now, and I figure it's probably about time to do something > about them as well. The following error is produced in excess of 100+ > times per minute: > > <the date / time> mail popper[pid]: (v2.53) Unable to get canonical name > of client, err = 0 > > I am assuming this is due to the almost complete lack of reverse DNS > setup for our internal network, (the only reverse DNS we have is what's > neccessary for our mailserver/dns servers to function; none of the > client computers have reverse dns mappings). I'd rather stop popper from > checking canonical names than have to do reverse dns mappings if > possible, but I'm not even sure if that's the root of this problem. > As I said, the second error doesn't really cause much grief, (unless > I'm actually on the console -at which point syslogd just becomes a great > annoyance); the system still functions, and everyone still gets their > email even though I get an error. I can (as I obviously have been) live > with it, but I'd rather not have to. > If anyone out there has any ideas, or insight into these errors, and/or > their parent causes, I'd be happy to hear from you. We're using Qpopper > 2.53, as obtained from the FreeBSD ports collection, running on a > FreeBSD 3.4 box. The box serves just short of 500 mailboxes internally, > (plus about 25 internet-accessable addresses, and three mailing lists), > accross two qualified domains, three local /24 subnets, and one small > subnet of real IP's. > > -- > Nathan Vidican > webmaster@wmptl.com > Windsor Match Plate & Tool Ltd. > http://www.wmptl.com/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000517073905.D58332>