February 13th, 2013, 03:47 AM
Postgres 9.1 sync broken after checkpoint reset
Hope you are doing good.
I have two postgres 9.1 servers which are setup in the following way
DB server A is the master and all the writes happen to it
DB server B is synced from the master and worked fine.
But the master server crashed due to hardware issues and it didnt come up so i had to reset the checkpoint using pg_resetxlog -f DATADIR
That started the master but after that Server B did not start and started complaining that it cannot replicate from the master
So i have two questions
1) what steps do i need to follow on the replicated slave (server B) after checkpoint reset in master
2) Also now that the slave (server B) is down, the master is filling up pg_xlog and out of disk space. Is there anyway i can delete them?