Mid Server records on instance responding VERY Slowly to UP and DOWN state of Mid Server Service, VALIDATION not completing


Description

Customer may report that that the Mid Server is stuck in the UP state, DOWN state, or possibly that the VALIDATION is not finishing.

Release or Environment

Any and ALL

Cause

It is important to verify when this began to occur, and to determine if these particular mid servers EVER did not encounter this issue. In some cases we are seeing that the customer may have a better connection through a proxy server than straight through the network without a proxy (or vice verse). It seems that this can make all of the difference in performance. These are very rare cases, however given the initial issue, and the fact that usually it is the proxy that causes the issue, I can attest that on at least one occasion the delay was NOT coming from the PROXY.

Resolution

1. If you see the instance record is still up, but the service is down and it does not appear to change the state, try the following to bring it down:

Navigate to sys_trigger.LIST and select the record for "Mid Server Monitor" and Execute the Job. This should bring down the service.

**Customer will probably need support to run this step


2. If you need to repoint to the proxy or (vice verse), make sure to

A. Stop the service

B. Reconfigure to config.xml to use the proxy (or NOT depending on which network path has the issue), and save the file

C. Run sc delete <ServiceName> form the command prompt as Administrator

D. Run start.bat from the agent folder to recreate the service


**After creating the service, you might need to rerun step 1 again but eventually the delay should be fixed.