Bug #3365 | mysql replication issue with replicate-rewrite-db | ||
---|---|---|---|
Submitted: | 2 Apr 2004 5:12 | Modified: | 2 Apr 2004 12:42 |
Reporter: | Jacques | Email Updates: | |
Status: | Closed | Impact on me: | |
Category: | MySQL Server: Replication | Severity: | S2 (Serious) |
Version: | 4.0.18 | OS: | |
Assigned to: | Guilhem Bichot | CPU Architecture: | Any |
[2 Apr 2004 5:12]
Jacques
[2 Apr 2004 7:23]
Jacques
I'm noticing that this bug causes replication to stop when it encounters a INSERT or UPDATE to any table which is under the rewritten database name. Trying the "start slave;" and checking the output shows that the slaves not restarting, and the only way to restart it is first to reload all the data using "load data from master;" and then type "start slave;".
[2 Apr 2004 7:24]
Dean Ellis
Verified, and thank you.
[2 Apr 2004 12:42]
Guilhem Bichot
Thank you for your bug report. This issue has been addressed in the documentation. The updated documentation will appear on our website shortly, and will be included in the next release of the relevant product(s). Additional info: Thank you Jacques for your bug report. In fact, LOAD DATA FROM MASTER cannot take replicate-rewrite-db into account because this case can happen: one user could, with this option, set up a non unique mapping like this: replicate-rewrite-db=db1->db3 and replicate-rewrite-db=db2->db3, which would confuse the slave when it loads the master's tables (imagine there exists db1.tbl and db2.tbl on the master: it would create on the slave db3.tbl (a copy of db1.tbl), then this db3.tbl would be overwritten by the copy of db2.tbl. You are right, the manual was confusing about this; I have now documented this limitation in the manual, to make it clearer. And thank you for using MySQL!! Guilhem