MySQL Bugs: #16772: Starting node joins cluster too early, workaround avoiding the issue for dynamically allocated nodeid's

into lmy004.:/work/mysql-5.1-bug16434

WL#1034 (Internal CRON)

into lmy004.:/work/mysql-5.1-new

WL#1034 (Internal CRON)
(not interesting for the Changelog)

into lmy004.:/work/mysql-5.1-bug16434

(post-review commit)
WL#1034 (Internal CRON)

into lmy004.:/work/mysql-5.1-bug16419

WL #1034 (Internal CRON)

into lmy004.:/work/mysql-5.1-bug16404

missed that one in the previos commit

into lmy004.:/work/mysql-5.1-new

into  neptunus.(none):/home/msvensson/mysql/bug15020/my51-bug15020

into  linux.site:/home/reggie/work/mysql-5.1

into  selena.:H:/MYSQL/src/#00000-mysql-5.1

into  neptunus.(none):/home/msvensson/mysql/bug15020/my51-bug15020

/var/tmp in my path when testing data diretory.  the var directory
exists in a src tree but may not exist in a different testing
build setup.

Rename of my_strdup_with_lenght to match the more common "strndup" method. Soo its now my_strndup().
my_stat() gettinng the correct parameter list may be next :)

into  mysql.com:/space/pekka/ndb/version/my51

into  linux.site:/home/reggie/work/mysql-5.1-bug14354

into  perch.ndb.mysql.com:/home/jonas/src/mysql-5.1-new

  Add massive printout when failure detected

into  perch.ndb.mysql.com:/home/jonas/src/mysql-5.1-new

structs in PARTITION_ITERATOR

into  neptunus.(none):/home/msvensson/mysql/bug15020/my51-bug15020

WL#1034 (Internal CRON)

when I ran mysql-test-run

into  mysql.com:/home/mysql_src/mysql-5.1-new

into  mysql.com:/home/psergey/mysql-5.1-ppruning-r5-push

into  mysql.com:/usr/home/bar/mysql-5.1-new.fscs

  Try to open an old 5.0 table name
  if encoded name doesn't exist.
  This makes SELECT working without having
  to rename tables with tricky characters
  in name into 5.1 format (simpliefies upgrade).

into  mysql.com:/home/psergey/mysql-5.1-ppruning-r5-push

into  mysql.com:/home/ndbdev/jmiller/clones/mysql-5.1-new