Product data
Collecting Errors
To collect enclosure errors, add this cron job SES Healthcheck to the system cron table.
Note: You must have root permissions to complete this procedure.
Edit the system crons with the crontab -e command. At the bottom of the file, enter:
15****/usr/lpp/diagnostics/bin/run_ses_healthcheck 1> /dev/null 2> /dev/null
For further information about the crontab command, see the web page at:
http://www.rs6000.ibm.com/doc_link/en_US/a_doc_lib/cmds/aixcmds1/crontab.htm#A1599114e
This cron runs at 15 minutes after each hour, and sends mail to the “root user” with
details of any errors that are present in the enclosure. It also presents a console
message indicating which enclosure has a problem.
The cron requires a script. To create this script, generate a file named
run_ses_healthcheck in directory /usr/lpp/diagnostics/bin. The contents of the file must
be:
#!/bin/ksh
#Name: run_ses_healthcheck
#Location: /usr/lpp/diagnostics/bin
#Function: SCSI SES hourly healthcheck
/usr/sbin/lsdev -Cc container -t ses -s scsi -F name -S available |
/usr/bin/xargs -n1 -i /bin/nice --10 /usr/lpp/diagnostics/bin/diagela -t {}
Note: Although, because of page width, the final line of the contents is shown here
split between two lines, you must enter it all on one line.
Give the following command so that the script can be run:
chmod 544 /usr/lpp/diagnostics/bin/run_ses_healthcheck
138 2104 Service Guide