| Bug #13604 | MASTER_POS_WAIT and lock at pos | ||
|---|---|---|---|
| Submitted: | 29 Sep 2005 12:47 | Modified: | 29 Sep 2005 13:43 | 
| Reporter: | Martin Friebe (Gold Quality Contributor) (OCA) | Email Updates: | |
| Status: | Closed | Impact on me: | |
| Category: | MySQL Server | Severity: | S4 (Feature request) | 
| Version: | 4.1 | OS: | Any (*) | 
| Assigned to: | CPU Architecture: | Any | |
   [29 Sep 2005 13:37]
   Guilhem Bichot        
  Hello, what you ask for has also been suggested by another user not long ago, so we have it on our long-term TODO. I think that for now you could use the START SLAVE UNTIL command to do the synchronization: #slave stop slave sql_thread; #master lock table t1; show master pos; #slave start slave sql_thread until (<values from above>); lock table t1; #master checksum table t1; unlock tables; #slave checksum table t1; unlock tables; start slave sql_thread; How does this sound?
   [29 Sep 2005 13:43]
   Martin Friebe        
  Does exactly what I am looking for, sorry, didnt research good enough, thanks anyway

Description: It would be nice, if the MASTER_POS_WAIT() could automatically take a lock , *before* the slave can go and execute any sql behind the specified location. If I do : select MASTER_POS_WAIT('file', pos, 999); lock table Name_of_table; the slave can (and likely will) execute further statements before I get the lock. What I would like to do, in order to check replication is syncronized: #master lock table t1; show master pos; #slave LOCK table t1 AFTER MASTER_POS_WAIT(<values from above>); #master checksum table t1; unlock tables; #slave checksum table t1; unlock tables; checksum should be the same on both, How to repeat: - Suggested fix: -