Troubleshooting
Problem
This document provides information about message CPF5140 and the relationship to Telnet connected devices.
Resolving The Problem
What does message CPF5140 mean, and what can be done about it?
CPF5140 - Session stopped by a request from device &4.
Cause . . . . . : The request shutdown was caused by either the user turning the power off, by a device error, or the ASCII controller inactivity timer expired. Recovery . . . : Close the files and vary the device off (VRYCFG command). If the problem occurs again, enter the ANZPRB command to run problem analysis.
While the technology mentioned in this message is seldom seen (ASCII controllers are not widely used on the system), it is still relatively accurate. This message means that the IBM i OS Interactive job has been signalled that the connection to the client is no longer active.
When the device name listed is a Telnet connected device description, there is no TCP/IP connection to that Telnet client. One of the following occurred:
o | The client session was disconnected while the user was logged on. |
o | The client was powered off. |
o | The TCP/IP stack in IBM i OS system attempted to verify that the conversation was still in use by the client and did not receive confirmation from that client. |
Change TELNET Attributes (CHGTELNA) Type choices, press Enter. Autostart server . . . . . . . . *YES *YES, *NO, *SAME Number servers . . . . . . . . . *CALC 1-100, *SAME, *CALC Session keep alive timeout . . . 600 0-2147483647, *SAME, *CALC... Default NVT type . . . . . . . . *VT100 *SAME, *VT100, *NVT Coded character set identifier *MULTINAT 1-65533, *SAME, *MULTINAT... ASCII fullscreen mapping: Outgoing EBCDIC/ASCII table . *CCSID Name, *SAME, *CCSID, *DFT Library . . . . . . . . . . Name, *LIBL, *CURLIB Incoming ASCII/EBCDIC table . *CCSID Name, *SAME, *CCSID, *DFT Library . . . . . . . . . . Name, *LIBL, *CURLIB Allow Secure Socket Layer . . . *YES *YES, *NO, *ONLY, *SAME |
The timer value is in seconds. Therefore, 600 seconds equals 10 minutes. With this setting, the IBM i OS TCP/IP Stack will attempt to contact the Telnet client about every 10 minutes. The timer defaults to a value of *CALC. This lets the IBM i OS calculate its own interval based on activity. It will be around the 10-minute time period.
When this timer pops, a TCP Keep Alive (RFC1122) is sent to the Telnet client. If a response is received by the IBM i OS TCP/IP stack, the timer starts again. If no response is received by the IBM i OS TCP/IP stack, the Keep Alive will be retransmitted several times. If no response is received from the client, the connection is considered dead and recovery starts (the client itself is powered off/inoperable/unplugged or the network transport has failed).
If a CPF5140 is being logged in your Telnet connected device interactive job joblogs, investigate your network transport reliability. An IBM i OS communications trace is a good tool with which to start. But, problem determination must quickly move into the network transport. If you would like assistance with network problem determination, purchase a consulting agreement or contact your local IBM Network Specialist.
Historical Number
26393218
Was this topic helpful?
Document Information
Modified date:
18 December 2019
UID
nas8N1017041