mirror of
https://github.com/MariaDB/server.git
synced 2025-07-30 16:24:05 +03:00
Fix after merge of 5.0 into 5.1. Failed tests before merge: bdb-deadlock, ndb_cache2, ndb_multi2,
ps_6bdb (this one fails circa once over 4 runs). After the merge: the same PLUS mysql_client_test&sp (due to problem in fix of BUG#14845 which has also been noticed in 5.0 too and Konstantin is looking at it) PLUS mysqlcheck (just consequence of mysql_client_test). And a replication bugfix (thd->spcont must be taken into account when calling tables_ok(), as in 5.0, and so some brackets were missing which changed one test). Pekka will look at the NDB code where I just used "ul": ndb_charset test, and storage/ndb/src/kernel/vm/SimulatedBlock.cpp and storage/ndb/src/kernel/blocks/dbtup/DbtupRoutines.cpp (SimulatedBlock.hpp automerged); please also check storage/ndb/src/ndbapi/NdbRecAttr.cpp. Petr please check ha_tina.cc. mysql-test/r/ndb_charset.result: "ul" in merge (Pekka please fix) mysql-test/t/ndb_charset.test: "ul" in merge (Pekka please fix) sql/sql_acl.cc: missing () sql/sql_parse.cc: If thd->spcont is true, always replicate (as in 5.0) (bad old merge probably).
This commit is contained in:
@ -182,7 +182,7 @@ static bool begin_trans(THD *thd)
|
||||
*/
|
||||
inline bool all_tables_not_ok(THD *thd, TABLE_LIST *tables)
|
||||
{
|
||||
return rpl_filter->is_on() && tables &&
|
||||
return rpl_filter->is_on() && tables && !thd->spcont &&
|
||||
!rpl_filter->tables_ok(thd->db, tables);
|
||||
}
|
||||
#endif
|
||||
|
Reference in New Issue
Block a user