Understanding Console Error Report Entries

Contents

This document discusses the most common reason for console errors in the system error report. This document applies to AIX Versions 4.x.

For more in-depth coverage of this subject, the following IBM documents are recommended:

• Problem Solving Guide and Reference
• Writing Reliable AIX Daemons

The AIX and RS/6000 product documentation library is also available:

http://www.rs6000.ibm.com/resource/

This document explains the usual reason for an error log entry similar to the following:

   ERROR LABEL: CONSOLE
   ERROR ID:    89B52AA5
   Date/Time:       Fri Aug 20 14:45:53
   Sequence Number: 20660
   Machine Id:      000044144100
   Node Id:         mst3000
   Class:           S
   Type:            PERM
   Resource Name:    console
   Error Description
   SOFTWARE PROGRAM ERROR
   Probable Causes
   SOFTWARE PROGRAM
   Failure Causes
   SOFTWARE PROGRAM
        Recommended Actions
        REVIEW DETAILED DATA
   Detail Data
   DETECTING MODULE
   conwrite
   FAILING MODULE
   UIO_WRITE
   RETURN CODE
      538976309        8224
   ERROR CODE
      538976304     2105344

This error is typically caused by an application that is started automatically during a bootup in one of the /etc/rc* or /etc/inittab files. Many applications open the console device to report errors or status. When the application writes to the console after a bootup has completed, the write() call will return EIO due to POSIX Std 1003.1 section 6.4.2.4: "The implementation supports job control, the process is in a background process group and is attempting to write to its controlling terminal, TOSTOP is set, the process is neither ignoring nor blocking SIGTTOU signals, and the process group of the process is orphaned."

When the console device driver senses that the write() call has failed, it logs an error to indicate that some program has attempted to notify you and has failed. It is not important to understand the reason that you are getting this entry in the error log; instead, it is important to find exactly which program is causing this error. The best way to do this is to put the following line into your /etc/inittab file immediately after the line that begins with powerfail:

   swcons:2:wait:/etc/swcons /tmp/cons.out > /dev/console 2>&1

Then enter the following command (as root):

   swcons /tmp/cons.out

Periodically check the file /tmp/cons.out for error messages that should be recognizable as coming from an application that is started during the bootup process. You may have to reboot your system to see these messages. Please be aware that messages that you used to see on your console (including bootup messages) will no longer appear there; they will now appear in /tmp/cons.out. You should monitor the size of this file, so that it does not fill the /tmp file system. To make the file empty, enter > /tmp/cons.out.

Once you have determined exactly which program is causing these errors, you can do one of the following:

• Contact your application vendor and explain the situation to them. They can change their source code to eliminate this problem.
• Leave the swcons command in your /etc/inittab file.
• Start the offending program from the command line.