AIX

Hi all,
I have a small problem, after a reboot the following services do not start automaticly, like they are supposed to do even though they are commented in in /etc/rc.tcpip
sendmail
xntpd
qdaemon
snmpd
snmpmibd
aixmibd
hostmibd

why ?!?! anybody ?

rc.tcpip is called from the inittab so you might check whether it is activated at runlevel 2. Sometimes inittab is rewritten, e.g. HACMP changes start of rc.tcpip from runlevel 2 to runlevel a and most daemons are activated when the cluster starts only.
Another possibility was that the install assistant did not run after a fresh install. Check inittab for entry calling a script named "firstboot". If it exists reactivate the install assistant with smitty install_assist and follow the steps.
If rc.tcpip was called from inittab at runlevel 2 but the daemons are deactivated after you log in check the boot process with
# alog -ot boot
and look for messages related to the daemons you mentioned.

So if HACMP changes the level from 2 to a is it OK to change it back to 2 ?

/etc/inittab snippet:
fbcheck:23456789:wait:/usr/sbin/fbcheck 2>&1 | alog -tboot > /dev/console # run /etc/firstboot
srcmstr:23456789:respawn:/usr/sbin/srcmstr # System Resource Controller
harc:2:wait:/usr/es/sbin/cluster/etc/harc.net # HACMP for AIX network startup
rctcpip:a:wait:/etc/rc.tcpip > /dev/console 2>&1 # Start TCP/IP daemons

I would not recommend that.... The reason for this change of the startup procedure is that HACMP does not know whether an IP address that is necessary for operation is available after reboot. It would be possible to start the cluster with addresses that are used for heartbeat only and without a proper network connection the daemons might fail or at least fill your logs with all sorts of errors. So what you could do is make sure that the cluster nodes have a suitable active and routeable IP even while HACMP is not started e.g. as a persistant address. Then you can either start daemons separately from inittab or (if there are too many daemons you need) from a separate rc file which again could be called from inittab at runlevel 2. You should not however change the rc.tcpip as trying to start a daemon twice does not result in problems usually while it might turn out to be a problem if the clusters misses a daemon it expects to have started itself. You must not change an inittab entry from runlevel a back to runlevel 2. That would be very bad administration practice (and btw might be reversed by the cluster occasionally without giving notice).