5120 event in S2D 4 node cluster

Hi

I am getting constant 5120 event ID in my 2019 S2D cluster, 4 node 100GB network (mellanox) between them, configured with ROCE.

I see some Pause packets on the switch interfaces but it does not seem to be in any pattern with the 5120 event id's

Cluster Shared Volume 'Mirror-01' ('Cluster Virtual Disk (Mirror-01)') has entered a paused state because of 'STATUS_CONNECTION_DISCONNECTED(c000020c)'. All I/O will temporarily be queued until a path to the volume is reestablished.

has anyone experienced this ?

regards

Sveinbjorn