Details
-
New Feature
-
Status: Closed (View Workflow)
-
Major
-
Resolution: Won't Do
-
2.2.15
-
None
Description
There is a case when the slaves with a large lag are disabled due to max_slave_replication_lag and the entire load went to the master and it collapsed ...
-------------------+-----------------+-------+-------------+--------------------
|
Server | Address | Port | Connections | Status
|
-------------------+-----------------+-------+-------------+--------------------
|
server1 | master | 3306 | 0 | Down
|
server2 | slave1 | 3306 | 0 | Running
|
server3 | slave2 | 3306 | 0 | Running
|
-------------------+-----------------+-------+-------------+--------------------
|
We really need the setting that allows MaxScale to use working, but not quite relevant, slaves only for reading (like router_options = running in readconnroute)