Applicable release versions: AP/Unix
|Description||describes the halt error codes, displayed in the message: "Monitor halt code XX at 0xNNNNNNNN"
XX is a decimal value
NNNNNNNN is the address where the HALT occurred.
Some errors are trapped in the Monitor Debugger, with a message:
YY is the hexadecimal value of XX.
All HALT errors are logged in the Pick Unix error log. See the documentation on the perrpt command.
Code (hex) Description
1 (01) Machine abort (bus error, segmentation violation,etc...). This error often indicates an ABS corruption. Run the TCL verify-system command. To exit the Monitor Debugger, type 'g' <return> to continue and enter the Virtual Debugger.
14 (0E) Process workspace corruption. This error occurs when a process workspace is so badly corrupted that it cannot even enter the Virtual Debugger. To recover from this error, use the TCL command reset-user port.number to clean up the bad process from another terminal.
16 (10) Old process. When the flusher which controls the virtual machine is killed by Unix (or by a 'kill -9' command), in case of low paging space, for example, some old processes may linger around. When the virtual machine is restarted, these processes will halt when trying to access any resource. This error does no harm to the virtual machine and can be ignored. When in the Monitor Debugger type g<return> to continue.