EMS HA Monitors Version A.01.00 Release Notes

6 Chapter 1
Known Problems and Workarounds
Known Problems and Workarounds
The following are known problems with the EMS HA Monitors product:
• EMS may not function if you are running name services that do not use
/etc/services, such as NIS, DNS, or X.500.
The port number for the EMS registrar is listed in /etc/services. If you
are running a different name lookup service, for example NIS, and it is not
configured to use /etc/services as part of the name lookup process, then
EMS monitors will not be able to find the registrar program and will not
function.
As a workaround, you can do one of the following:
• On the name server, add the registrar services line to the appropriate services
file for the name lookup service you are running. The line should have the
same port number as the line in /etc/services:
registrar 1712/tcp # resource monitor service
• Add the /etc/services file to the lookup path your name lookup service
uses. For example, modify the nsswitch.conf file to refer to
/etc/services if you are running NIS.
• Creating a request with too short a polling interval gives “Unable to register
monitor request” error message.
If you create a request with a polling interval shorter than the minimum polling
interval supported by the monitor, you will get an error message. For example
the network monitor has a minimum polling interval of 30 seconds, so a request
to monitor a LAN interface with a 20 second polling interval will be denied.
As a workaround, create the request with a longer polling interval. The
minimum polling interval for the disk monitor is 1 second, the minimum for the
cluster, network, and system resource monitors is 30 seconds. If you have
additional monitors, check the documentation accompanying them for the
minimum polling interval.