Answer the question
In order to leave comments, you need to log in
master-master replication. How to properly enable master-slave from the 2nd to the 1st server?
There are 2 servers.
On the 2nd, a base dump is deployed from the 1st.
Master-slave replication is configured from 1st to 2nd.
The GUID is used.
All users go to the 1st.
There is MASTER_LOG_FILE and MASTER_LOG_POS data after deploying the dump on the 2nd server:
MASTER_LOG_FILE = "master2.000001", MASTER_LOG_POS = 34004
These values on master2 are constantly increasing. Is it normal if users do not go to it yet?
Which MASTER_LOG_FILE and MASTER_LOG_POS parameters of the CHANGE MASTER TO MASTER_HOST command should be used to start the master-slave from 2nd to 1st, and eventually get a master-master with synchronous bases? Data immediately after the dump was deployed, or the last values?
Do I need to set SET GLOBAL read_only = YES and lock tables when starting replication from 2nd to 1st?
Answer the question
In order to leave comments, you need to log in
I will answer myself
: in my case, no one goes to server 2 - you can use the latest values of MASTER_LOG_FILE and MASTER_LOG_POS
The data after deploying the 2nd server from the dump can be used if you need to include the 3rd server in the chain (for example, a lagging copy).
SET GLOBAL read_only = YES and lock tables do not need to be set.
Didn't find what you were looking for?
Ask your questionAsk a Question
731 491 924 answers to any question