Workaround for PowerEdge 1950/2950 CMOS battery unknown

[bluikko]

OMSA 9.1 on PE 1950 and 2950 works incorrectly and CMOS battery status is always "unknown":
CRITICAL | 0 | Battery probe 0 [System Board CMOS Battery] is Unknown reading
It seems to be a problem with OMSA 9.1 which officially dropped support for the 9th generation servers.

The following patch is a workaround to simply set the status of the test to OK on these two system types. There is probably a better way to write the workaround but this works:

@@ -4160,14 +4160,6 @@

        $count{bat}++;
        next BATTERY if blacklisted('bp', $index);
-        if (($reading eq 'Unknown reading') and ($sysinfo{model} eq 'PowerEdge 1950')) {
-            $status = 'Ok';
-            $reading = 'Good';
-        } #XXX fix for old servers
-        if (($reading eq 'Unknown reading') and ($sysinfo{model} eq 'PowerEdge 2950')) {
-            $status = 'Ok';
-            $reading = 'Good';
-        } #XXX fix for old servers

        my $msg = sprintf 'Battery probe %d [%s] is %s',
          $index, $location, $reading;
@@ -5268,8 +5260,6 @@

 # Global status via SNMP.. extra safety check
 if ($globalstatus != $E_OK && $exit_code == $E_OK && !defined $opt{only}) {
-    exit $E_OK if ($sysinfo{model} eq 'PowerEdge 1950'); #XXX fix for old servers
-    exit $E_OK if ($sysinfo{model} eq 'PowerEdge 2950'); #XXX fix for old servers
     print "OOPS! Something is wrong with this server, but I don't know what. ";
     print "The global system health status is $reverse_exitcode{$globalstatus}, ";
     print "but every component check is OK. This may be a bug in the Nagios plugin, ";

[giannidaprile]

This is not necessary. Just add a blacklist entry in check_openmanage,conf for your host:

[hostname]
blacklist = bp=all

If blacklist is used global health status is ignored.