It is possible to observe this event occasionally,
due to the unreliable nature of the UDP transport protocol. However, if the
occurrence of this event is frequent, investigate the issue further.
This event is expected when a retransmission is
received from the client before a server has responded to the request, possibly
a result of the client retransmitting too quickly before allowing sufficient
time for a server to respond in time. Another possible cause is if one or more
servers configured to handle the request are non-responsive.
Investigate the routing configuration to narrow down
the list of servers (Peer Nodes) which are expected to handle requests from the
reported server connection.
Evaluate whether an Egress Transaction Failure Rate
alarm has been raised for any of the corresponding client connections. If so,
investigate the cause of the server becoming non-responsive and address the
condition.
Note:
Depending on the operator's choice, the client
connection may need to be Admin Disabled until the evaluation is complete,
which will allow requests to be routed to other servers, depending on the
routing configuration. If this is not the case, tune the client's retransmit
timers to be greater than the typical turnaround time for the request to be
processed by the server and for the response to be sent back to the client.
If the problem persists, it is recommended to
contact My Oracle Support.