CMC Servers Page - Slow response in CMS clustered environment

[Jonathan_Haun]

I have noticed that in both version XI 3.1 and BOE 4.0 that it can take up to 30 seconds for the CMC to print out the state of each server when viewing their status on the CMC, Servers page. This only occurs when more than one CMS is running in a multi node cluster. If I stop one CMS in the cluster, the lists are returned in 1 to 2 seconds. With both CMS services enabled and running, the list can take several seconds.

Does anyone have a good technical explanation of this behavior in a clustered environment? I see the same problem in both Windows and Linux.

[Jonathan_Haun]

Update: I discovered
that the issue is related to two issues.

2. When servers defined in the CMS repository are stop, initializing
        or hanging. The CMC server interface will try 3 times to connect to the
        server before moving on to the next. I typically only look at the CMC
        servers page when restarting an SIA, so it makes sense that it would be
        slow during this time period based on this information. There is an SAP
        note that describes this: https://service.sap.com/sap/support/notes/1675455
4. With the new monitoring application the CMC servers
        page in 4.0 has been updated to include the server health watch. If the
        monitoring service is slow or the db that operates it is corrupted, the
        servers page in the CMC will be slowed as the interface attempts to pole
        the health status of each server. There is no SAP note to explain why or
        how the poling process works, but if you stop the monitoring services in a
        cluster, the rendering of the servers status is instantly faster in the
        CMC.