Hello Richard,
the registrar is definitely needed for EMS, without this process ISEE will not work correctly. For the version A.29 a patch is needed for EMS (PHSS_28843), it might help to install this. In addition you should restart EMS using the following procedure:
1) Stop the registar if neccesary by commenting out the 1 entries in /etc/inetd.conf and re-reading the inetd configuration with inetd -c
2) Issue ps -ef | grep registrar to check if any registrar processes are remaining still, kill them if necessary
3) Stop monitoring
# /etc/opt/resmon/lbin/monconfig
Select (K)ill (disable) monitoring
ATTENTION! Within a Serviceguard Cluster, package dependencies might be configured, which could cause a package to switch,
in case a resource would go to a DOWN state.
In case EMS is stopped, this resource would be considered to be DOWN and the package would switch.
If you are unsure, follow the instructions in the document before you go ahead!
4) Issue ps -ef | grep monitor to check if any monitors are still running, kill them if necessary
5) Stop p_client
Comment out the p_client entry in /etc/inittab :
# vi /etc/inittab
6) Change the following line
ems4:3456:respawn:/etc/opt/resmon/lbin/p_client
to
#ems4:3456:respawn:/etc/opt/resmon/lbin/p_client
7) Reread /etc/inittab :
# init q
8) Issue ps -ef | grep p_client to check if p_client is still running, if yes kill the process
9) Delete the persistence files and pipes
# cd /etc/opt/resmon/persistence
# rm p*
# rm m*
# cd /etc/opt/resmon/pipe
# rm *
10) Have diagnostics remap the system hardware
#cstm
cstm>remap
cstm>quit
Allow enough time for remap to finish before restarting the monitors.
When the remap is in progress you get the message ***Remap in progress*** when issuing map at the cstm prompt;
when the remap has finished the map is displayed when issuing the command map at the cstm prompt.
It might take up to a few minutes depending on the amount of hardware connected.
11) Clean up monitor flags in the diagnostics directory
# cd /var/stm/data/tools/monitor
# rm *.hwa
12) Re-enable registrar in /etc/inetd.conf and run inetd -c
13) Restart p_client
# vi /etc/inittab
Change the following line
#ems4:3456:respawn:/etc/opt/resmon/lbin/p_client
back to
ems4:3456:respawn:/etc/opt/resmon/lbin/p_client
14) Re-read /etc/inittab :
# init q
15) Restart the EMS monitors
# /etc/opt/resmon/lbin/monconfig
Select (E)nable Monitoring
It might be a good idea to update the diagnostics. Be careful if this is a service guard cluster when you try to uninstall the support tools. If the above procesure help resolving the registrar problem, you can update the support toolssimply by doing a swinstall for the new version without an previous uninstall.
Regards
Frauke
