datadir can be saved in case of constantly failing servers
Remove datadirs each time all servers have been stopped thus limiting
the size of vardir.
The datadirs of failed tests have been saved and that should be
enough to diagnose any failures
mysql-test/suite/rpl_ndb/r/rpl_ndb_circular_2ch.result:
updated result
mysql-test/suite/rpl_ndb/t/disabled.def:
updated disabled list
mysql-test/suite/rpl_ndb/t/rpl_ndb_circular_2ch.test:
updated test
mysql-test/include/circular_rpl_for_4_hosts_init.inc:
The primitive for 4 hosts non-cluster circular replication
mysql-test/include/circular_rpl_for_4_hosts_sync.inc:
The primitive for sync for 4 hosts non-cluster circular replication
mysql-test/include/ndb_master-slave_2ch.inc:
The primitive for initialization of two-channel cluster replication
mysql-test/suite/rpl/r/rpl_circular_for_4_hosts.result:
Result file
mysql-test/suite/rpl/t/rpl_circular_for_4_hosts-master.opt:
Options file
mysql-test/suite/rpl/t/rpl_circular_for_4_hosts-slave.opt:
Options file
mysql-test/suite/rpl/t/rpl_circular_for_4_hosts.test:
The test case for non-cluster circular replication A->B->C->D->A
mysql-test/suite/rpl_ndb/r/rpl_ndb_circular_2ch.result:
Result file
mysql-test/suite/rpl_ndb/t/rpl_ndb_circular_2ch-master.opt:
Options file
mysql-test/suite/rpl_ndb/t/rpl_ndb_circular_2ch-slave.opt:
Options file
mysql-test/suite/rpl_ndb/t/rpl_ndb_circular_2ch.test:
The test case for cluster circular replication by two channels
Also make it possible to specify a path to the suite, for example:
./mtr --suite=/some/path/to/suitedir
mysql-test/lib/My/Find.pm:
Extend 'my_find_paths' to glob path names
Don't allow unix socket path to be truncated
mysql-test/lib/My/Platform.pm:
Add check for unix socket path truncation
mysql-test/mysql-test-run.pl:
Don't allow socket path to be truncated. Fail and ask user to correct the problem by
using a shorter path with --tmpdir
the server dissapears
Check return code of 'mysql_ping'
Add "shutdown", call 'mysql_shutdown' on the current connection
mysql-test/include/restart_mysqld.inc:
Use "kill_server"
mysql-test/include/wait_until_connected_again.inc:
Init $mysql_errno to 9999
Add workaround for BUG#36228
mysql-test/suite/rpl/r/rpl_slave_grp_exec.result:
result file
mysql-test/suite/rpl/t/rpl_slave_grp_exec-master.opt:
master option file
mysql-test/suite/rpl/t/rpl_slave_grp_exec-slave.opt:
slave option file
mysql-test/suite/rpl/t/rpl_slave_grp_exec.test:
test case
- Code in DbugParse was reading from beyond end of the control string
dbug/dbug.c:
Don't try to parse control strings that hasn't got any tokens
or is zero size
No more string matching
mysql-test/include/check-warnings.test:
Check return value from check_warnings and call skip if ok
mysql-test/include/mtr_check.sql:
Improve readability
mysql-test/include/mtr_warnings.sql:
Add out parameter for result.
Increase the max_allowed_packet variable so the load_file works
even if the server is started with low such value
Don't log check_warnings to binlog
mysql-test/include/check-testcase.test:
Change to use a stored pårocedure for check_testcase
mysql-test/include/mtr_warnings.sql:
Don't log the check_warning procedure to binlog
mysql-test/mysql-test-run.pl:
Turn on --check-testcase by default
mysql-test/include/mtr_check.sql:
New BitKeeper file ``mysql-test/include/mtr_check.sql''
