When this happens next, try the command netstat -an I would expect to see thousands of open sockets. Run the command during normal mode, and also when the problem occurs. Pass the information along to Ipswitch - it may be helpful in tracking down the problem. ----- Original Message ----- From: Bonno Bloksma To: [email protected] Sent: Wednesday, March 28, 2007 6:50 AM Subject: Re: [IMail Forum] HTTP Failing on 2006.2?
Hi, I've had IMail 2006.2 in test for several weeks with a dozen users and we found no problem. It wasn't until we started using the webinterface with about 2000 users when suddenly problems started to appear. Kinda hard for Ipswitch to find problems no one knows about. :-( Unfortunately these problems are very hard to diagnose as all resources I have been tracking show now limit reached or something like it. Also the problem is a wierd one where parts will start to fail and other parts wil still work. A user can suddenly no longer send mail via the web interface, the error mesages refers to a socket failure. The rest of the webmail will work and normal SMTP, POP, IMAP etc will also work. On the server itself suddenly fileshares on other servers can no longer be reached, sniffer update (using wget) will fail, stuff like that. A while later the IClient webmail interface may no longer work but the IAdmin interface will still work, even when most things fail I can still reach the server and my monitoring tool can still log in to the server and restart the w3publ service. Try finding a common cause for those errors ans exceptions to see which resourse is the problem and THEN try to find the cause to depleting that resourse. It's not as easy as some people think it is, I wish it were. :-( For now I'm restarting the w3publ service on the Windows 2003 server three times a day. Met vriendelijke groet, Bonno Bloksma hoofd systeembeheer tio hogeschool hotelmanagement en toerisme begijnenhof 8-12 / 5611 el eindhoven t 040 296 28 28 / f 040 237 35 20 [EMAIL PROTECTED] / www.tio.nl
