MySQL 5.6 Reference Manual
MySQL 5.6 Release Notes
When a master server shuts down and restarts, its
MEMORY tables become empty. To
replicate this effect to slaves, the first time that the master
uses a given MEMORY table after
startup, it logs an event that notifies slaves that the table
must to be emptied by writing a
DELETE statement for that table
to the binary log.
When a slave server shuts down and restarts, its
MEMORY tables become empty. This
causes the slave to be out of synchrony with the master and may
lead to other failures or cause the slave to stop:
Row-format updates and deletes received from the master may fail with
Can't find record in '.memory_table'Statements such as
INSERT INTO ... SELECT FROMmay insert a different set of rows on the master and slave.memory_table
The safe way to restart a slave that is replicating
MEMORY tables is to first drop or
delete all rows from the MEMORY
tables on the master and wait until those changes have
replicated to the slave. Then it is safe to restart the slave.
An alternative restart method may apply in some cases. When
binlog_format=ROW, you can
prevent the slave from stopping if you set
slave_exec_mode=IDEMPOTENT
before you start the slave again. This allows the slave to
continue to replicate, but its
MEMORY tables will still be
different from those on the master. This can be okay if the
application logic is such that the contents of
MEMORY tables can be safely lost
(for example, if the MEMORY tables
are used for caching).
slave_exec_mode=IDEMPOTENT
applies globally to all tables, so it may hide other replication
errors in non-MEMORY tables.
(The method just described is not applicable in NDB Cluster,
where slave_exec_mode is always
IDEMPOTENT, and cannot be changed.)
The size of MEMORY tables is
limited by the value of the
max_heap_table_size system
variable, which is not replicated (see
Section 4.1.35, “Replication and Variables”). A change in
max_heap_table_size takes effect for
MEMORY tables that are created or updated
using ALTER TABLE
... ENGINE = MEMORY or TRUNCATE
TABLE following the change, or for all
MEMORY tables following a server
restart. If you increase the value of this variable on the
master without doing so on the slave, it becomes possible for a
table on the master to grow larger than its counterpart on the
slave, leading to inserts that succeed on the master but fail on
the slave with Table is full errors. This
is a known issue (Bug #48666). In such cases, you must set the
global value of
max_heap_table_size on the
slave as well as on the master, then restart replication. It is
also recommended that you restart both the master and slave
MySQL servers, to insure that the new value takes complete
(global) effect on each of them.
See The MEMORY Storage Engine, for more
information about MEMORY tables.
Note that this can be problematic if you use binlog_do_db and have memory tables not listed there.
MySQL will still always write the DELETE query after a server restart, even if it's for a table not to be replicated.
I'm not sure whether this is intentional or a limitation, but I can't see it noted anywhere.