Try to resolve the noderange bug, haven't produced it myself, so this is just a guess

git-svn-id: https://svn.code.sf.net/p/xcat/code/xcat-core/trunk@8786 8638fb3e-16cb-4fca-ae20-7b5d299a9bcd

perl-xCAT/xCAT/NodeRange.pm

@@ -60,6 +60,11 @@ sub nodesmissed {
   return @$missingnodes;
 }
 
+sub reset_db {
+#workaround, something seems to be trying to use a corrupted reference to grptab
+#this allows init_dbworker to reset the object
+    $grptab=0;
+}
 sub nodesbycriteria {
    #TODO: this should be in a common place, shared by tabutils nodech/nodels and noderange
    #there is a set of functions already, but the path is a little complicated and
@@ -182,11 +187,6 @@ sub expandatom { #TODO: implement table selection as an atom (nodetype.os==rhels
     # Try to match groups?
         unless ($grptab) {
            $grptab = xCAT::Table->new('nodegroup');
-        } else { # if DB2 we need to get a new table handle always
-             my $DBname = xCAT::Utils->get_DBName; 
-             if (($DBname =~ /^DB2/) && (xCAT::Utils->isServiceNode())){
-              $grptab = xCAT::Table->new('nodegroup');
-             }
         }
         if ($grptab and not $didgrouplist and not scalar @grplist) { 
             $didgrouplist = 1;

perl-xCAT/xCAT/Table.pm

@@ -151,6 +151,7 @@ sub init_dbworker {
 
 
     $dbworkerpid = fork;
+    xCAT::NodeRange::reset_db(); #do in both processes, to be sure
 
     unless (defined $dbworkerpid) {
         die "Error spawining database worker";