The Chat started in Service Point, shows up in the xFlow support queue. The Chat can be selected, but no conversation occurs.
From the end users' perspectives, when the xFlow user is sending a chat message, the xFlow user does not see their conversation update unless they directly refresh the associated web browser. Same as the opposite user working on ServicePoint. Chat communication does not display immediately unless one refreshes the UI at the web browser level
Problem is most likely due to a network issue in connections on the given xFlow server. A possible example may be if one or more nodes in the xFlow cluster is not connecting properly.
Release : 17.3
Component : XFLOW INTERFACE FOR SDM
Review the collabservicestdout.txt and searchMS.log logs for any indication of a network connection issue.
In one such scenario where there were two SDM app servers with xFlow defined, SDMXFlowServer1.domain.com and SDMXFlowServer2.domain.com, the problem was being caused by issues with connections specific to SDMXFlowServer1.domain.com.
This error mentions a problem connecting to SDMXFlowServer1.domain.com:2554
collabservicestdout.txt:[WARN] [08/13/2020 05:10:07.864] [Collaboration_MicroService_XflowClusterSystem-cluster-dispatcher-18] [akka.tcp://[email protected]main.com:2554/system/cluster/core/daemon/joinSeedNodeProcess-1] Couldn't join seed nodes after  attempts, will try again. seed-nodes=[akka.tcp://[email protected]main.com:2554]
This entry mentions SDMXFlowServer1.domain.com:2552
searchMS.log:WARN - 2020-08-13 12:33:05 [a.remote.ReliableDeliverySupervisor] - Association with remote system [akka.tcp://[email protected]m:2552] has failed, address is now gated for  ms. Reason: [Association failed with [akka.tcp://[email protected]m:2552]] Caused by: [java.net.ConnectException: Connection refused: no further information: SDMXFlowServer1.domain.com/188.8.131.52:2552]
Had instructed to correct by doing the following:
1) Stop xFlow on both nodes.
On the l1admin_general_config table you should have the following entries under the config_value column:
2) Take a backup of that table:
pdm_extract l1admin_general_config > l1admin_general_config.txt
3) Delete the entries mentioned above and clear the cache on the table:
pdm_cache_refresh -t l1admin_general_config
4) Start xFlow on SDMXFlowServer2, see if the issue is reproduced.
If not then start xFlow on SDMXFlowServer1.domain.com and test again.