into  perch.ndb.mysql.com:/home/jonas/src/51-ndb

  (recommit for correct merge order)

into  dator6.(none):/home/mikael/mysql_clones/bug30996

  Rename: ndb/src/common/util/ndb_rand.c -> storage/ndb/src/common/util/ndb_rand.c
ndb_rand.h:
  Rename: ndb/include/util/ndb_rand.h -> storage/ndb/include/util/ndb_rand.h

into  willster.(none):/home/stewart/Documents/MySQL/5.1/ndb

timoOutLoopStartLab() checks if any transactions have been delayed
for so long that we are forced to perform some action (e.g. abort,
resend etc).

It is *MEANT* to (according to the comment):
> To avoid aborting both transactions in a deadlock detected by time-out
> we insert a random extra time-out of upto 630 ms by using the lowest
> six bits of the api connect reference.
> We spread it out from 0 to 630 ms if base time-out is larger than 3 sec,
> we spread it out from 0 to 70 ms if base time-out is smaller than 300 msec,
> and otherwise we spread it out 310 ms.

The comment (as all do) lies.

the API connect reference is not very random, producing incredibly
predictable "random" numbers. This could lead to both txns being
aborted instead of just one.

Before:
timeout value: 123 3
timeout value: 122 2
timeout value: 122 2
timeout value: 122 2
timeout value: 123 3

After:
timeout value: 127 7
timeout value: 126 6
timeout value: 129 9
timeout value: 139 19
timeout value: 137 17
timeout value: 151 31
timeout value: 130 10
timeout value: 132 12

Index: ndb-work/ndb/src/kernel/blocks/dbtc/DbtcMain.cpp
===================================================================

into  dator6.(none):/home/mikael/mysql_clones/bug30996

  Removed semicolon causing build syntax issues per pekka

into  whalegate.ndb.mysql.com:/home/tomas/mysql-5.1-new-ndb

Moved out a lot of code into functions from external_lock and
start_stmt
Fixed a crashing bug at memory alloc failure
Merged the stmt and all variables into one trans variable
Always register start of statement as according to the
interface of the handlers.
Also register for start of transaction when not statement commit
== not autocommit AND no begin - commit ongoing
Now that we registered in a proper manner we also needed to handle
the commit call when end of statement and transaction is ongoing
Added start_stmt_count to know when we have start of statement
for first table

into  mysql.com:/windows/Linux_space/MySQL/mysql-5.0-ndb

into  mysql.com:/windows/Linux_space/MySQL/mysql-5.1-new-ndb

into  mysql.com:/windows/Linux_space/MySQL/mysql-5.1-new-ndb

into  whalegate.ndb.mysql.com:/home/tomas/mysql-5.1-new-ndb-bj

into  whalegate.ndb.mysql.com:/home/tomas/mysql-5.1-new-ndb-bj

into  whalegate.ndb.mysql.com:/home/tomas/mysql-5.0-ndb-bj

into  whalegate.ndb.mysql.com:/home/tomas/mysql-5.1-new-ndb

into  perch.ndb.mysql.com:/home/jonas/src/51-ndb

  Dont add insert+delete tuple(s) to lcp-keep-list

into  whalegate.ndb.mysql.com:/home/tomas/mysql-5.1-new-ndb

into  perch.ndb.mysql.com:/home/jonas/src/51-ndb

into  dev3-63.(none):/home/zhl/mysql/mysql-5.1/bug27543

into  dev3-63.(none):/home/zhl/mysql/mysql-5.1/bug30024

- problem is the database name accessed in Rows_log_event write... get_db() which is a pointer to the share string...
- point to table map instead?
- or copy it?
- or make sure that anything interacting with the share happens _after_ the epoch

Problem with flush is that STMT_END_F may not be included as it shoudld...

this flush with flag TRUE causes spurious errors in the binlog... as it will insert a STMT_END_F... and hence an invalid binlog (missing table mappings)

into  whalegate.ndb.mysql.com:/home/tomas/mysql-5.1-new-ndb

  forgot to update next TC no

  When adding a read to lock queue, where there is a X-lock ahead, upgrade to X-lock
  (found by pekkas blob changes)

  remove more "extra mysql-"