We are running Exchange 2003 SP2 on Windows 2003 SP2 servers in an active/active/passive cluster. The HTTP virtual server fails after time once NPP exceeds a certain threshold. The only solution is to reboot the server.What is the relevance of NPP to Exchange? I understand that Exch draws upon NPP for every TCP/IP connection, so it possible that - in addition to, say, a driver memory leak somewhere - a large increase in users on the server (which we have had) is also a contributing factor?Has anyone experienced a similar issue?

Hi Yoshi,You may need to follow below troubleshooting steps....- Run ExBPA to find any relevant error and if possible run it at the time of NPP exceeds.- Check error in event log for the relevant errors during time of NPP exceeds.- Disable SNP feature of Windows 2003 SP2 if it is enabled which may cause NPP memory sometimes.Windows 2003 Scalable Networking pack and its possible effects on Exchangehttp://msexchangeteam.com/archive/2007/07/18/446400.aspx- Most of the times third party hardware drivers would be culprit for NPP memory leak so I'd definitely check for updates of all the third-party drivers on vendor site.- Disable third-party softwares one by one (antivirus, antispam, security etc..) and monitor the behavior for some time.- You may need to follow advance troubleshooting steps byadding pool tag withPoolMonto find out which resource/driver is causing problem and get the result when server gets hang.http://blogs.msdn.com/ntdebugging/archive/2006/12/18/Understanding-Pool-Consumption-and-Event-ID_3A00_--2020-or-2019.aspxAmit Tank | MVP - Exchange | MCITP:EMA MCSA:M | http://ExchangeShare.WordPress.com

Hi Yoshi, If you still have further questions, due to the limitation of the forum, I suggest you contact our CSS to submit a case: http://support.microsoft.com/default.aspx?scid=fh;EN-US;OfferProPhone#faq607 Thank you, Elvis The Microsoft Exchange Team Blog - You had me at EHLO - http://msexchangeteam.com/