Details
-
Task
-
Status: Open (View Workflow)
-
Major
-
Resolution: Unresolved
-
None
-
None
Description
Implement a queued switchover command that's similar to the existing async-switchover command but unlike the async-switchover which gives up immediately if there's too much lag, the queued switchover would try again later in the hopes that the lag is reduced.
In addition to the retry behavior, a new SERVER_TOO_MUCH_LAG bit (feel free to choose the name) needs to be added to communicate the desire to reduce lag. The routers will then read this status bit and do whatever is needed to lower the lag.
Attachments
Issue Links
- is part of
-
MXS-6235 prepare for switchover by throttling writes - keep replication lag low
-
- Open
-