Cisco Systems 78-16845-01 appendix 104

Appendix J Troubleshooting

J.10.8 Connection Management Problems

When any add or modify connection request is issued, an SNMP VarBind is set on the switch via cmgrd. The switch can reject this VarBind with this error.

The reason for the error is that cmgrd sent an invalid value for one of the MIB objects. The CTM DE should do the investigation for this error, as it indicates that there is a problem in the data that cmgrd either received from the clients or internally mapped from the other side of the connection.

Try the same connection from both the Connection Proxy and CM GUI interfaces and note the behavior from both of the clients. If the provisioning result from both clients is the same, it tends to indicate that this is either a cmgrd issue or a sdbroker vpi/vci issue.

Defect Information—The cmgrd.log, ConnProxy*, cmsvr.log and sdbroker*.logs should be collected.

Possible alternative workaround—If you are doing an add and you are using default values, then there is no workaround. However, if you are doing a modify, you can change the value of the object being modified and retry the connection. Also, if this is a sdbroker vpi/vci issue, retry the connection, because sdbroker does not reuse the previous vpi/vci combination.

J.10.8.4.18 Cmgrd—Addition Errors: Provisioning a Connection Results in "Object Exists on the Agent or Specified VPI/VCI not available." Switch Error

Related key index entries: cmgrd, object exists

When any add connection request is issued, an SNMP VarBind is set on the switch via cmgrd. The switch can reject this VarBind and the subsequent cmgrd query to the switch error table can return these errors.

The reason for the error is that the connection that CTM is trying to add already exists on the switch. This could be a user endpoint or an intermediate endpoint.

Step 1 Retry the same connection. If this second attempt succeeds it indicates that the intermediate endpoint’s vpi/vci was already present on the switch. If this is the case, look at the cmgrd-sdbroker error "EndPoint Exists Local/Remote/Both end of the connection already exists."

Step 2 If the switch returns the same error on the second attempt, this could indicate an inconsistency between the CTM database and the switch, and it must be debugged by a CTM DE.

Defect information—Save the cmgrd.log, ConnProxy*.log, cmsvr.log and sdbroker*.log.

Possible alternative workaround—Retry the connection addition more than once. If it still fails, a CTM DE will have to look into the issue.

J.10.8.4.19 Cmgrd—Addition/Modification Errors: Provisioning a Connection Results in "Requested cell rate (lscr/lpcr or rpcr/rscr ) is too high" Switch Error

Related key index entries: cmgrd, cell rate

During an add or modify connection request, it is possible that the port the connection is being set on has no more resources available to accept the traffic parameters that the current SNMP SET is requesting. At this point, the switch will reject the request and return this error.

Step 1 For the user endpoint, check the user port on the CLI using the command dsppnportrsrc. This will indicate the resources available per port.

Step 2 If the failure is not happening on a user endpoint port, but instead on an intermediate port (for example, for multisegment connections), find out the local and remote endpoint feeder trunk ports. (This can be done by using the Network Topology GUI.)

	Cisco Transport Manager Release 6.0 User Guide
J-104	78-16845-01