HP-UX SysFaultMgmt (System Fault Management) (SFM) Release Notes (766969-001, March 2014) (Edition: 2)

cron may not be running - call your system administrator
warning: commands will be executed using /usr/bin/sh
The message cron may not be running is logged if the cron daemon does not run during
the install. The warning commands will be executed using /usr/bin/sh occurs if
the default login shell is not defined POSIX shell /usr/bin/sh. This is a warning from cron
that the POSIX shell is used to execute commands in the crontab file. If the login shell is set to
/usr/bin/sh in the /etc/passwd file, this error is not logged.
Workaround:
You can safely ignore both these warning messages.
• Problem:
The second LAN interface configured on a server is not accessible for web browsing.
Description:
When a second LAN interface was configured on a server using a private subnet used only
for cluster heartbeat traffic, Launch SMH button got the second LAN interface’s address but
was not accessible for web browsing.
Workaround:
Once the server has been turned, disable the second LAN, restart SFMProviderModule, and
configure the second LAN card.
Repeat these steps, if you need to restart cimserver or SFMProviderModule.
After configuring, do not restart the server, cimserver, and SFM Provider module.
• Problem:
The events and user defined subscriptions (created using Evweb) are lost when diagnostic
products are upgraded on a DRD cloned disk.
Description:
Prior to September 2012 release version of the diagnostics product, the postgres 7.4.2 version
was supported. However version 8.4.8 is shipped from September 2012 release. Due to the
missing data compatibility between PostgreSQL version 7.4.2 (prior to September 2012) and
8.4.8 (September 2012 release or later), when the diagnostics products are upgraded from
prior to September 2012 release to September 2012 web release and later, the postgres
data needs to be migrated to 8.4.8 format.
Data migration involves backup in 7.4.2 version using pg_dumpall and restore in 8.4.8
version using the psql operations. Both of these operations requires a running postgres service.
But in the DRD environment there is a limitation that no service can be run on the cloned disk
when products are either installed or upgraded.
This limits diagnostics product upgrade steps from taking the backup of the postgres 7.4.2
data.
As a result, the data such as events, error records, and subscriptions (created using Evweb)
is lost during such circumstances.
Workaround:
You can select any of the following two alternatives:
1. To install the product on the DRD cloned disk as a normal upgrade scenario. The scenario
will be same as fresh installation and user may refer to event.log for historical event data.
25