Details
-
Type:
Bug
-
Status: Open
-
Priority:
Minor
-
Resolution: Unresolved
-
Affects Version/s: 10.0.4
-
Fix Version/s: None
-
Component/s: None
-
Labels:None
-
Environment:centos 6.4
Description
Configure a maria slave with 2 masters each replicating a different database.
The status variable slave_running show "OFF" . Both sql and io thread status is fine in both the slave connections.
MariaDB [(none)]> show global status like 'slave_running'; +---------------+-------+ | Variable_name | Value | +---------------+-------+ | Slave_running | OFF | +---------------+-------+ 1 row in set (0.00 sec)
Gliffy Diagrams
Attachments
Activity
- All
- Comments
- Work Log
- History
- Activity
- Transitions
The current logic is such that the Slave_running status variable indicates the status of the default slave (the one that default_status_connection variable is set to). The default value of default_status_connection is an empty line, which relates to the unnamed slave configured by the legacy syntax, when you just say
CHANGE MASTER TO ...
without specifying the name.
If both your master/slave connections are named, that is, if you configured them as
CHANGE MASTER 'name1' TO ... ;
CHANGE MASTER 'name2' TO ... ;
and then started them, it means that 'name1' slave is running, 'name2' slave is running, but '' slave is not, which is what Slave_running variable shows.
If you run
SET default_master_connection = 'name1';
then Slave_running should start showing 'ON', assuming that 'name1' is running.
I agree that the behavior is not obvious, and there were plans to sort out slave-related status and server variables. I'll assign it to Monty to decide what can be done about this one.