Error Flushing Archive Queue

It sounds like the secondary server thinks it is in the middle of a backup. Check the "Archive Backup Flag" under the Snapshot and Archive Statistics in PI-SMT.

This is a known issue in some versions of PI.

https://techsupport.osisoft.com/Troubleshooting/Known-Issues/10099OSI8

You may be able to run piartool -backup -abort on the secondary server to reset the backup flag, then try to add your server to the collective again.

You might be running into the issue described in the Known Issue 18016OSI8 - "Unable to start archive queue flush on server"

When initializing a secondary node in a collective, the PI Collective Manager flushes the archive cache of the primary PI Server before it backs up the server and copies over the database files to the secondary server. On systems with a large number of points, the initialization may fail during the archive flush with the following error:

"Error -2147219670: RPC Invoke failed[-11096] Archive Backup already in progress"

To get the PI Archive Subsystem out of the backup state, "piartool -backup -abort" or "piartool -flush -stop" commands must be issued.