NAME

dmesg − collect system diagnostic messages to form error log

SYNOPSIS

/etc/dmesg [ − ] [ corefile [ namelist ] ]

DESCRIPTION

Dmesg looks in a system buffer for recently printed diagnostic messages and prints them on the standard output.  The messages are those printed by the system when device (hardware) errors occur and (occasionally) when system tables overflow non-fatally.  If the − flag is given, then dmesg computes (incrementally) the new messages since the last time it was run and places these on the standard output.  This is typically used with cron(8) to produce the error log /usr/adm/messages by running the command

/etc/dmesg − >> /usr/adm/messages

every 10 minutes. 

The ULTRIX-11 error logging system logs hardware related errors, dmesg is used to save other types of error messages. 

The optional corefile and namelist options can be used for postmortem debugging.  If corefile is specified, it will be used instead of /dev/mem; if namelist is also specified, it will be used instead of /unix. 

FILES

/usr/adm/messageserror log (conventional location)
/usr/adm/msgbufscratch file for memory of − option
/dev/memdefault corefile
/unixdefault namelist

SEE ALSO

ULTRIX-11 System Management Guide, Chapter 8 ULTRIX-11 System Management Guide, Chapter 9

RESTRICTIONS

Referring to dmesg as the error log is a misnomer. 

The system error message buffer is of small finite size.  As dmesg is run only every few minutes, not all error messages are guaranteed to be logged. If this happens, dmesg will print ’...’ before printing the buffer.  This can be construed as a blessing rather than a curse. 

Error diagnostics generated immediately before a system crash will never get logged.