Since there is no newsgroup for Mercury (why not ??) I post this message in
this related group.
Since a few months I'm running Mercury/32 v2.14 on a Windows NT 4.0 Server
(without the bulky IE4 stuff etc.)
It runned reasonable smoothly (with about 15 users) for months. Sometimes
you have to stop-restart the program or NT, but that's it.
Since a few days the SMTP Client module responds with the error "Error FF
servicing queue job" after polling.
No mail is going out to the internet, and the users doesn't get a message
aubout this failure.
Sending mail internal or receiving mail from internal or internet works
fine.
Any idea what's wrong ?
Thanx for reading this message,
Arthur Kater
15:25:01: <lf>--- Mon Mar 08 15:25:01 1999<lf> ---<lf>Connect to 'mail.foo.com', timeout 900.<lf>
I have TCP/IP timeout set to 900 seconds and I still get timeouts. And:
15:26:09: 2: Bad address "mail.foo.com"<lf>
I suppose the DNS lookup timed out on this one.
I think some SMTP servers will not accept connections from hosts unless there has been a POP download in the last fifteen minutes. I have noticed my SMTP client finally starting up after the distributing POP client has finished a download. This could be one reason that the TCP/IP timeouts have to be so large. I think I'll bump mine up to 1200.
Also note that for some reason the SMTP client always waits 30 minutes to retry even if you force a poll. So once an error occurs, you can't deliver the mail for at least 30 more minutes.
--Kent
Arthur Kater wrote in message <7ca5bo$6te$1...@news.seed.net.tw>...