![]() | |
![]() |
| | Thread Tools | Display Modes |
#11
| |||
| |||
|
|
My idea was to lock the db on A, scp the required files onto B and then unlock db on A. |
#12
| |||
| |||
|
|
Machine A is the Primary, and Machine B is the backup for A. When B is booting up, it has to duplicate the entire pgsql db from A. As Ron said, cud do with a pg_dump. But, i guess, pg_dump takes quite some time. As A is already up, it wud be unwise to lock the db for so long. Also, even if i do go ahead with pg_dump, and then do a pg_restore on B, by the time data is being pg_restore'ed on B, a query cud modify/ update the db on A. My idea was to lock the db on A, scp the required files onto B and then unlock db on A. |
#13
| |||
| |||
|
|
Benjamin wrote: Machine A is the Primary, and Machine B is the backup for A. When B is booting up, it has to duplicate the entire pgsql db from A. As Ron said, cud do with a pg_dump. But, i guess, pg_dump takes quite some time. As A is already up, it wud be unwise to lock the db for so long. Also, even if i do go ahead with pg_dump, and then do a pg_restore on B, by the time data is being pg_restore'ed on B, a query cud modify/ update the db on A. My idea was to lock the db on A, scp the required files onto B and then unlock db on A. If I understand you correctly, "Slony" will do exactly what you want, without the need to lock any DB: http://www.slony.org Slony 1.0 has been released recently, which was mentioned prominently on PostgreSQL's homepage. It's a master-slave replication system. Hope that helps. Best regards Oliver Thanx everyone. |
![]() |
| Thread Tools | |
| Display Modes | |
| |