Gateway refreshes failing

[Jon_S]

Since mid-July, we have an automated refresh that has been failing every single day.  Prior to that, it had worked every day for 6 months.  When manually refreshing, it fails with the same error message ~4 out of 5 times.   The other time it works.  As a result, each day we have been manually refreshing this model until it doesn't error out. 

We have tried the following and none of the resolve the issue:

• Upgrade the virtual machine to a faster instance type
• Restart the gateway service
• Install the latest gateway version 
• Stand up a new virtual machine and install the gateway, make new connections

When testing the status of the gateway and the connection, they always show as online.  The ODBC connection on the machine is working.  

What is causing this and what steps should we take next?

[lbendlin]

When testing the status of the gateway and the connection, they always show as online.

That doesn't mean much.  The network tests on the gateway management app are largely pointless as they do not reflect the actual workings of the gateway.

How many cluster members? are they all equally configured and equally healthy?