Date: Fri, 6 Jun 1997 01:48:41 +0800 From: sweeting@neuronet.com.my To: freebsd-isp@FreeBSD.ORG Subject: summary and thanks. Re: help pls : how to optimise mail server response Message-ID: <3.0.32.19970606014213.00dbb278@neuronet.com.my>
next in thread | raw e-mail | index | archive | help
First and foremost, thank you very much to everyone for the fantastic response. Sorry for the late reply follow up but I wanted to try as much as possible before disturbing you all. I went to the customer's offices today and tested over a dialup to the ISP (jaring.my) - Eudora downloaded all of the mail with no problem whatsoever. However, when I returned to the office, there were a load of error messages on the console as follows : NAA00753: SYSERR(root): collect: I/O error on connection from btr-21-101.tm.net.my, from=<sumi@mdc.com.my.: Connection reset by btr-21-101.tm.net.my Fyi, tm.net.my is the second ISP in Malaysia (yep, pretty monopolistic over here). They used to route via the US to get to the other local ISP (jaring.my) which my servers are connected to but now the two are connected locally as this traceroute shows : tracert mail.tm.net.my 1 <10 ms <10 ms <10 ms 202.184.153.1 2 100 ms 111 ms 110 ms 161.142.33.165 3 101 ms 110 ms 100 ms e0.brf29.jaring.my [161.142.2.30] 4 110 ms 111 ms 110 ms mix-tmnet.tm.net.my [202.188.2.221] 5 110 ms 110 ms 111 ms brf-secure01-ether0-0.tm.net.my [202.188.0.4] 6 110 ms 110 ms 110 ms janeway.tm.net.my [202.188.0.155] Trace complete. So that still doesn't explain why dialling up to tm.net.my would be the cause of errors. At 06:35 PM 6/4/97 -0700, Dan Busarow wrote: >On Thu, 5 Jun 1997 sweeting@neuronet.com.my wrote: >> The problem is that they are having a lot of difficulty >> collecting mail from the mailhub I set up for them >> on a P166 FreeBSD box with 64 MB RAM. >> We are a webcompany with 2 MB line to the same >> ISP that our customer is dialling into. (We are not the ISP) >I'm not familiar with the 3Com cards, haven't used them in years, >but my impression is that they are towards the bottom of the >list for FBSD. You should give an NE2000 clone a try, cheap >and effective even if they aren't the fastest around. a couple of people suggested NE2000 - i have ordered one. >So that would suggest DNS configuration problems. Can't >check that with dummy names and IP addresses. the mailhub is on a machine called peace.com.my and one of the virtual domains that it is handling email for is mdc.com.my >So, move DNS to FBSD and if that doesn't fix it outright >post with real names and addresses. many people suggested this and i am going to do it this weekend. t 10:23 AM 6/5/97 +0800, John Beukema wrote: >> 2) DNS - is handled by a DEC Alpha 500 (runnign DU) which is >> possibly in need of RAM (it's handling mail, and 4 webservers, X term, >> and DNS for 50 domains.... somebody once told me that each domain >> should have 1 MB RAM.) >> > >Could be DNS. try pinging with an IP address not a name. >Put the IP number and domain name in /etc/hosts and put host >ahead of bind in /etc/host.conf my /etc/hosts was ok and it seemed weird to put host above bind but tried it just in case. >Otherwise, it could be a routing problem. The route should be static, >silent (or something), permanent and not depend on routed or gated which >might expire and then take time to re-establish. To tell, check the >route before and after pinging. i am indeed using static routes (i've always had trouble with routed and gated on freebsd) >>3) Jammed network at the ISP > >Not impossible ;-) but this isn't likely because you're getting the same >problem on your LAN. On the LAN, i only get one type of problem : the initial ping request times out. I can download mail no problem at all from any of the virtual domains on the mailhub over the LAN. It is only when dialling up to the ISP that the main problem occurs (of the mail client not being able to download the mail). >>4) I configured it inefficiently (config files below) > >We'd need to see /etc/hosts.conf and /etc/resolv.conf on the FBSD host for >clues about whether the DNS is to blame. i am not running dns on the mailhub yet but feel the console error message above may reveal the real problem. thank you once again to one and all for all of the help !!! chas surreal thought .. couldn't be the mail clients, could it ? (they are using Eudora 3.0 and netscape 3 )
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3.0.32.19970606014213.00dbb278>