07-28-2008 09:09 AM
We have been running ACT Premium for Web 2008 (10.0) for about 2 months now. Unfortunately, during this time the ACT website will stop responding/go down every 3-4 days. Users will attempt to access the site but instead get a "page cannot be displayed" or "IE cannot display the page" message...or something along those lines. The only fix we have found, if you want to call it that, is a reboot of the ACT server. There server itself has no other programs or websites installed, it is just running ACT and Symantec. The app event log is littered with the below warning messages. One Sage support rep had me try reinstalling ASP.Net last Monday to resolve these warnings, but that obviously did not help since they are now back. I have been trying to figure out why the site continues to go down, but am at a loss. Any ideas?
Event Type: Warning
Event Source: ASP.NET 2.0.50727.0
Event Category: Web Event
Event ID: 1309
Time: 12:00:45 PM
Event code: 3001
Event message: The request has been aborted.
Event time: 7/28/2008 12:00:45 PM
Event time (UTC): 7/28/2008 4:00:45 PM
Event ID: 62f80a632ea7469b989f2d57b657326c
Event sequence: 53703
Event occurrence: 140
Event detail code: 0
Application domain: /LM/W3SVC/1/Root/APFWOutlookSrvc-1-128616826082438507
Trust level: Full
Application Virtual Path: /APFWOutlookSrvc
Application Path: E:\Program Files\ACT\Act for Web\APFWOutlookSrvc\
Machine name: WEBSERVER
Process ID: 1544
Process name: w3wp.exe
Account name: NT AUTHORITY\NETWORK SERVICE
Exception type: HttpException
Exception message: Request timed out.
Request URL: https://act.domain.com:443/APFWOutlookSrvc/APFWHistoryService.asmx
Request path: /APFWOutlookSrvc/APFWHistoryService.asmx
User host address: 10.35.24.102
Is authenticated: False
Thread account name: Domain\Username
Thread ID: 11
Thread account name: Domain\Username
Is impersonating: False
Custom event details:
For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.
08-06-2008 07:28 PM
Let me echo what dlunceford wrote, and suggest that you check that you do not have an SQL memory leak.
I had one APFW setup that crashed every 7 days because it took a week for SQL to climb up to over 1GB of memory being used, at which point it was completely unstable and required restarting.
This does not fix the problem unfortunately, just delays it for a while longer.
After spending hours with Tech support, we finally had to blow up the OS and re-install from scratch, and it has not happened again.
08-12-2008 11:19 AM
Thank you both for your feedback. This continues to be an irritating problem, even with another rebuild of the server this weekend. I rebuilt the box (ghost imaging), installed APFW, .NET from the ACT cd, etc. Everything was fine up until this morning when once again the website was unresponsive. I tried getting Sage support to look at the box while it was happening, but unfortunately everytime I tried opening a window on the server it froze. I ended up having to reboot the box. During the downtime I also swapped the chasis (moving the disks to another Dell 2850), so as to test for possible hardware issues and or rule them out.
After reading your posts, however, and looking more into the w3wp.exe and sqlserver.exe processes, I am bit intrigued to figure out if these processes are in-fact causing the problem. I am attempting to monitor them now to see how much memory usage they (continue to) consume. As it stands right now, I see both are slowly increasing. If they are causing the problem, why is it that a rebuild of the server (twice now) has not fixed the problem and stopped the unresponsiveness from re-occurring? Is recycling the app pools supposed to lower the memory usage back down?
Any additional input would be appreciated. Thanks again!