Not support for detecting poor perfomance masters

[GoogleCodeExporter]

Is this tool supports to detect poor performing master and then do fail-over. 
Instead of detecting dead-master.

Because, Frequently we are facing problem with Master performance degradation 
instead of completely dead

Original issue reported on code.google.com by [email protected] on 20 Nov 2014 at 1:07

[GoogleCodeExporter]

MHA does not handle a master as dead, if the master is running, even though the 
master performance degrades. This is to prevent repeating failover and false 
alarm. For example, if there are bugs in application logic, failover won't fix 
the problem so starting failover does not make any sense.

If you want MHA to start promotion, shutdown the master instance (by an 
external script or manually). The script should be smart enough that it 
understands failover fixes the problem (i.e. hardware problem on the master 
server).

Original comment by [email protected] on 21 Nov 2014 at 9:10