If

• You can figure out how to reliably tail a file with one process that's being (asynchronously?) written by another process;
• The  'TPSM seed' field really is a hex number as  '234B8177' from your example data suggests (note the 'B' in the middle);
• All the other numbers are decimal numbers;
• All the non-numeric fields are constant text;
• Variable whitespace is acceptable between fields

Then this extraction code may be useful:

c:\@Work\Perl\monks>perl -wMstrict -le
"my @lines = (
   qq{TPSM seed 234B8177   pass 1  x 2 x 50  errors 0 pid 34562 rulefi
+lecycle 0\n},
   qq{TPSM seed 234B8177   pass 1  x 2 x 50  errors 0 pid 34562 rilefu
+llcycle 0\n},
   );
 ;;
 my $d_num = qr{    \d+        }xms;
 my $h_num = qr{ [[:xdigit:]]+ }xms;
 ;;
 for my $line (@lines) {
   print qq{[[$line]]};
   ;;
   my $match = my ($seed, $p1, $p2, $p3, $err, $pid, $rfc) = $line =~ 
+m{
     \A
     TPSM \s+ seed \s+ ($h_num) \s+
     pass \s+ ($d_num) \s+ x \s+ ($d_num) \s+ x \s+ ($d_num) \s+
     errors \s+ ($d_num) \s+
     pid \s+ ($d_num) \s+
     rulefilecycle \s+ ($d_num) \s*
     \Z
     }xms;
   ;;
   if ($match) {
     print qq{line ok,  seed $seed  p1 $p1  p2 $p2  p3 $p3  err $err  
+pid $pid  rfc $rfc \n};
     }
   else {
     print qq{line malformed};
     }
   }
"
[[TPSM seed 234B8177   pass 1  x 2 x 50  errors 0 pid 34562 rulefilecy
+cle 0
]]
line ok,  seed 234B8177  p1 1  p2 2  p3 50  err 0  pid 34562  rfc 0

[[TPSM seed 234B8177   pass 1  x 2 x 50  errors 0 pid 34562 rilefullcy
+cle 0
]]
line malformed
[download]

I understand that this code is perhaps a bit advanced for you right now, but this is a learning experience. Please consider carefully the wise advice of haukex, and also see perlre and also some of the Pattern Matching, Regular Expressions, and Parsing tutorials in the PerlMonks Tutorials section.