WORK WITH --START-POSITION
If setting --start-position to start after the FD event, mysqlbinlog
will output an error stating that it has not found an FD event.
However, its not that mysqlbinlog does not find it but rather that it
does not processes it in the regular way (i.e., it does not print it).
Given that one is using --base64-output=DECODE-ROWS then not printing
it is actually fine.
To fix this, we make mysqlbinlog not to complain when it has not
printed the FD event, is outputing in base64, but is decoding the
rows.
mysqlbinlog prints all row-based events of a single statement as a
single "BINLOG" statement containing the concatenation of those events.
Big (i.e. >64k) concatenations of row-based events
(e.g. Write_rows_log_event) caused mysqlbinlog's IO_CACHE to overflow
to a temporary file but the IO_CACHE had not been inited with
open_cached_file(), so it tried to create a temporary file in
an uninitialized directory (thus failing to create, then to write;
some OS errors were printed, and it finally segfaulted).
After fixing this, it appeared that mysqlbinlog was printing only
a piece of big concatenations of row-based events (it printed
at most the size of the IO_CACHE's buffer i.e. 64k); that caused data
loss at restore. We fix and test that.
Last, mysqlbinlog's printouts looked a bit strange with the informative
header (#-prefixed) of groupped Rows_log_event all on one line,
so we insert \n. After that, a small bug in the --hexdump code appeared
(only if the string to hex-print had its length a multiple of 16),
we fix it.
Problem: mysqlbinlog_base64 failed sporadically.
Reason: Missing "flush logs" before running $MYSQL_BINLOG,
which could start dumping the log file before server
has finished writting into it.
Fix:
- implementing --force-if-open option to "mysqlbinlog"
- adding --disable-force-if-open to make $MYSQL_BINLOG
fail on non-closed log files, to garantee that nobody
will forget "flush logs" in the future.
- adding "flush logs" into all affected tests.
Rename mix_innodb_myisam_binlog-master.opt; when the test was moved to
separate stm/row test cases, the .opt were not moved along with it,
causing a 60-second test duration because of default lock timeout :-(
