diff options
author | Alfranio Correia <alfranio.correia@oracle.com> | 2010-11-05 13:58:05 +0000 |
---|---|---|
committer | Alfranio Correia <alfranio.correia@oracle.com> | 2010-11-05 13:58:05 +0000 |
commit | 2d84c1553c509b65564b8bc7baf8bcc291398281 (patch) | |
tree | 0e254042b809c100b5bd66b2523a4685da7548d7 /mysys/my_sync.c | |
parent | f0ce67873cd2eb2b6eb30ba1ecb0d3df5206d5a0 (diff) | |
parent | 8504580ced8b2161630f14489ab206d5e313f000 (diff) | |
download | mariadb-git-2d84c1553c509b65564b8bc7baf8bcc291398281.tar.gz |
BUG#56343 binlog_cache_use status is bigger than expected
The binlog_cache_use is incremented twice when changes to a transactional table
are committed, i.e. TC_LOG_BINLOG::log_xid calls is called. The problem happens
because log_xid calls both binlog_flush_stmt_cache and binlog_flush_trx_cache
without checking if such caches are empty thus unintentionally increasing the
binlog_cache_use value twice.
Although not explicitly mentioned in the bug, the binlog_cache_disk_use presents
the same problem.
The binlog_cache_use and binlog_cache_disk_use are status variables that are
incremented when transactional (i.e. trx-cache) or non-transactional (i.e.
stmt-cache) changes are committed. They are used to compute the ratio between
the use of disk and memory while gathering updates for a transaction.
The problem reported is fixed by avoiding incrementing the binlog_cache_use
and binlog_cache_disk_use if a cache is empty. We also have decided to increment
both variables when a cache is truncated as the cache is used although its
content is discarded and is not written to the binary log.
In this patch, we take the opportunity to re-organize the code around the
function binlog_flush_trx_cache and binlog_flush_stmt_cache.
mysql-test/extra/binlog_tests/binlog_cache_stat.test:
Updated the test case with comments and additional tests to check both
transactional and non-transactional changes and what happens when a
is transaction either committed or aborted.
mysql-test/suite/binlog/r/binlog_innodb.result:
Updated the result file.
mysql-test/suite/binlog/r/binlog_mixed_cache_stat.result:
Updated the result file.
mysql-test/suite/binlog/r/binlog_row_cache_stat.result:
Updated the result file.
mysql-test/suite/binlog/r/binlog_stm_cache_stat.result:
Updated the result file.
mysql-test/suite/binlog/t/binlog_mixed_cache_stat.test:
Updated the test case with a new source file.
mysql-test/suite/binlog/t/binlog_row_cache_stat.test:
Updated the test case with a new source file.
mysql-test/suite/binlog/t/binlog_stm_cache_stat.test:
Updated the test case with a new source file.
sql/log.cc:
There are four changes in here:
. Computed statistics on binlog_cache_use and binlog_cache_disk_use while
resting the cache.
. Refactored the code so binlog_flush_cache handles both the trx-cache and
stmt-cache.
. There are functions that encapsulate the calls to binlog_flush_cache and
make easier to read the code.
. binlog_commit_flush_stmt_cache is now taking into account the result:
success or error.
sql/log_event.h:
Guaranteed Xid_log_event is always classified as a transactional event.
Diffstat (limited to 'mysys/my_sync.c')
0 files changed, 0 insertions, 0 deletions