Sage 100

v4.4 advanced - on host server - Windx spawn - Connection failure to host.... , the problem usually arises when user logins peak, between 8:00 and 9:00 a.m. .... user maintenance ""spawn task"" is checked... running host as a program not as a service.... the default printer is available and online. Any ideas (other that re-formatting the server and starting over)?.

I'm sure PEP is off. You don't say how many users but my ""magic number"" to see this stuff is usually 15+ and typically more like 20+. Check this registry change. I think I may have had to use it when running as application. https://www.evernote.com/shard/s270/sh/397b1123-cb16-459c-88ac-f51390b662c7/386ab700a9af6738b3786c36c426e620

Looks like the problem I had yesterday. Called Sage and worked with Scott Brady (I love him). Scott thought it was the firewall settings blocking incoming connections. So he added a Sage inbound rule to allow incoming ports between xxx and xxx (sorry, I don't have access to the exact settings right now). This seemed to fix the issue but I haven't been able to fully test it yet because we are still in test mode on the install. Scott said he would write it up so it might be on the knowledgebase (but it was yesterday so it might be too soon).

Thanks @WayneSchulz Will give the registry hack a try.... they have 30 users.... plus another 30 warehouse employees using ScanForce handhelds.

Thanks @JaneCavanaugh I will look at the firewall settings too.

Multi-part answer. Appears 2) is what you may need. 1) About what @WayneSchulz said, since you have potentially 60 connections to the MAS App Server which is running on the server's desktop (not as a service), you should be more focused on possibly changing the 2nd number (desktop heap size) in the SS-3003 registry hack. But if you're on Server 2008 or higher, it is probably at 20480 which is high enough. 2) Ergo, more likely you need to check the box for Spawn Task from Application Server in User Maintenance / Preferences for all users and with everyone out of the system. 3) What @JaneCavanaugh said is very relevant but most likely you would get the error right off the bat as soon as you open the 1st task off the Launcher. You wouldn't have to wait until the peak time of the day for it to happen. Your options for that would be: a) Enable Spawn Task from App Server as it requires less inbound ports to be open if Windows Advanced Firewall is being used), b) Open a broad range of inbound ports for Windows Advanced Firewall c) Disable Windows Advanced Firewall on the server

Do all connections come through a single client on a terminal server or are they coming through from individual desktops? Can we assume the error is coming on the AppServer and not at the client machine.

@GaryFeldman Definitely coming from the client machine.

@AlnoorCassim Thanks for your reply.... All user already have the ""spawn tasks"" check box checked.

Citrix users and local-installed desktop users (i.e. IT Director) have had the problem. It doesn't seem to be latency related. They've reviewed the registry hack and it looks like it already had optimal settings. One other wildcard here is the AppAssure backup software, which was installed in early July. It's basically executing a differential backup via VSS every 30 minutes and then sending the result across the network to the server -- but at this point, the size of the differential backup is negligible relative to its impact on network performance.