Server response problems
If the server is not responding, first determine whether the server is waiting or spending CPU time to perform work. You can see this example with SDSF DA. The server normally is shown as 3 address spaces in SDSF (while idling). Possible causes for lack of response include:
- The client is using the wrong port numbers or machine name. There is a test connection button in the client to verify they at least are active. You can use the netstat command under TSO to see on which port the server is listening.
- The server log shows a number of messages:
E10:Crypt: Protocol violation. message from 12.1.4 and no secure channel E18:Crypt: Unexpected message from 12.1.4 suspicious, so discarded
The probable cause of these messages is that the server agent has been stopped and then started again, while the client agent kept running. On a lightly loaded 30 MIPS machine, the client can connect to the server within 6 minutes, and six E10/E18 message pairs resulting from retransmissions by the client are printed to the server log. A quicker way to recover is to end the c2ragent.exe task using the End Process button of the Windows Task Manager, and to close and restart the IBM® Security zSecure Visual application.
Another cause might be a logon attempt just after configuring a client. In that case, a single E10/E18 message pair can be printed at the server side because a secure channel has not been completely set up yet. Recovery in this case only takes 1 minute on a fast machine. However, you can avoid the delay by waiting 15 seconds (on a fast machine) after client configuration before attempting a logon. In trace mode, the server displays the following message when it is ready to accept a logon:E0:CA: Finished certifying Agent Keys
- The server log shows the following reconnect message:
E183:Route: reconnected from 12.1.4
This message indicates that two clients with identical agent ids are attempting to communicate with the server. Stop one of the c2ragent.exe processes with the Windows Task Manager. The processes most likely reside on a single computer but they might also reside on separate computers.
- The server log shows the following message:
E160:LCM: There are no valid LCM certificates. Please reconfigure the server
The most likely cause is that server initialization, job C2RZWINI, was not run successfully. Less likely, the server did not run in the last 9 months, so it did not refresh its certificates in time. Stop the server, run or rerun job C2RZWINI, and verify that the server successfully initializes as a Certificate Authority. See First time startup of the Server.
S C2RSERVE,OPT=TRACE
Using the TRACE option results in bigger server logs that contain detailed timing information. To have IBM Software Support help with debugging those problems, send both the client log and the server log.