This chapter provides information about troubleshooting SGIconsole and covers the following topics:
When a node responds with the message "Command not deliverable; <reason>", the cause of the failures can be found in the VACM log located in the /var/log/vacm.log file.
Typical node failures are as follows:
The node's controller is powered down (L1 or L2 management port types only)
There is a serial communications problem with the controller for that node as follows:
The serial cable connecting SGIconsole to a node's controller has been removed or is not firmly seated in the serial port socket of the controller.
The SGIconsole configuration for the node does not specify the correct serial port device (/dev/ttyNxxx) to use. You must configure SGIconsole to use the serial port ( /dev/ttyNxxx) that corresponds to the port used on the multiplexer supporting the device. For more information, see “Adding a Node” in Chapter 3.
For SGI Origin 2000 series systems with a Multimodule System Controller (MMSC), the serial port settings are not at the defaults of 9600/8/1/N. In this case, the serial port settings need to be reset manually (using MMSC commands) or by resetting the MMSC (again using MMSC commands). Operating the MMSC with serial port settings other than the in-firmware defaults, is not supported by SGIconsole for reasons of operational stability.