Description
Describe the bug
In Zowe V3.X, when a peer HA system (e.g. xyz.net) is unavailable or not yet started, the following ERROR and WARN log entries repeat continuously:
2025-05-20 21:21:58.649 ZWEADS1:TaskBatchingWorker-target_xyz.net-1:67504916 ZWESVUSR ERROR ((o.z.a.p.e.c.ApimlPeerEurekaNode$ReplicationTaskProcessor)) Network level connection to peer xyz.net; not re-trying this exception because it does not seem to be a network exception. This message will suppressed for 10000ms.
jakarta.ws.rs.ProcessingException: org.apache.http.conn.HttpHostConnectException: Connect to xyz.net:75533 failed: EDC8128I Connection refused.
...
Caused by: java.net.ConnectException: EDC8128I Connection refused. (errno2=0x74940000)
2025-05-20 21:21:58.650 ZWEADS1:TaskBatchingWorker-target_xyz.net-1:67504916 ZWESVUSR WARN ((c.n.e.u.b.TaskExecutors)) Discarding 3 tasks of TaskBatchingWorker-target_xyz.net-1 due to permanent error
This behavior results in repetitive and misleading log entries, even though the error is expected due to a known temporary outage or inactive peer.
ERROR logs draw unnecessary operator attention for what is essentially a known and recoverable condition in an HA configuration.
These logs may cause concern during planned peer outages, IPLs, or staged starts.
Repetition every few seconds can inflate log files and mask real problems.
Expected behavior
Avoid flooding logs by consolidating messages.
E.g., log once on initial failure, and then only again on change (e.g., successful reconnection or prolonged unavailability).
Moved to HA debug mode instead for info level logging
Metadata
Metadata
Assignees
Labels
Type
Projects
Status