It’s a good idea to rebuild the code cache and audit conf.
Disabling NTP stepping
Normally NTP slews the time if it’s off by less than 128ms, and steps the time if it’s over that. This can cause some serious problems with Oracle (especially 9i+). The -x option in xntpd forces all time changes to be slewed instead of stepped. You’ll want to run ntpdate if your time isn’t pretty close to being accurate already then make these changes:
stopsrc -s xntpd
chssys -s xntpd -a "-x"
startsrc -s xntpdClearing MQSeries IPCS semaphores
If there are problems getting MQSeries to start, try clearing out the IPCS semaphors used by MQ. With MQ down, run:
ipcs -s |egrep mqm |awk '{print $2}' |xargs -i -n1 ipcrm -s
ipcs -m |egrep mqm |awk '{print $2}' |xargs -i -n1 ipcrm -mScript to trim wtmp file
#!/bin/sh
#########
# wtmptrim.sh - Trim old entrues off of the wtmp file so it doesn't grow
# forever.
#
# V1.0 - 12/07/00
#
/usr/sbin/acct/fwtmp < /var/adm/wtmp | tail -20000 > /tmp/wtmp.ascii.new
if [ "$?" != 0 ]
then
echo "$0" "Error extracting wtmp file"
exit
fi
/usr/sbin/acct/fwtmp -ic < /tmp/wtmp.ascii.new > /var/adm/wtmp
if [ "$?" != 0 ]
then
echo "$0" "Error recompressing wtmp file"
exit
else
rm /tmp/wtmp.ascii.new
fiFix dtlogin messages on a ASCII console
Sometimes after an AIX upgrade, reload, or new system installation, you’ll get “The X-Server can not be started on display:0:” on the ASCII console. To fix it Delete the bottom line beginning with :0 in the /usr/dt/config/Xservers file.
List available major device numbers
When creating HACMP concurrent volume groups, it’s necessary to sync the major device numbers between the nodes. To see what each node has available for major devices run:
lvlstmajorSetting HBAs to init in Point-to-Point mode
The HBA does not need to be configured in most cases. Nevertheless, you should verify if your HBA runs in Point to Point mode if applicable (such as switched fabric). By default the HBA tries first to run in FC_AL instead of Point to Point. If the system is booted without being connected to a switch, it may operate in FC_AL mode even after it is connected to a switch. In some circumstances you may have problems with missing devices if your HBA does not run in the correct mode.
Starting an LPAR from command line
If you need to poweroff or poweron an LPAR remotely, you can do it with command lines.
Listing LPARs and their status
You can query the LPARs and status of any managed system on a HMC.
Lookup cabinet serial number
You can’t get support without the serial number, and sometimes it’s just not available. Here’s how to query the system for it:
lscfg -pv | grep Cabinet