Secondary goes behind primary

its a 4 node cluster. 2 nodes in one location and two nodes in second location. There are AG configured on multiple instances across this cluster.

Sometimes once in a week or twice the secondary instances(some times replica in the same location) goes behing primary.

1. There is no blocking in primary
2. No long running SPID’ s found in primary
3. Nothing specific in error logs on primary and delayed replica.
4. There is no open tran

Some times i see Redo queue is not sent to Secondary and piled up at primary. Another time,it looks like Redo is populated but seconday is not processing it.

When i reboot the secondary issue gets resovled. How do i find the reason for this issue? I tried to suspend data movement and resume but didnt help..Tried switching AYNC to SYNC.

Best Regards, Arun http://whynotsql.blogspot.com/