Re: Group Replicaton - Lagging Slave

[Adam Nelson]

For anyone else who has the same problem ... I spent all night going through MySQL options and documentation. I was able to solve the issue by doing the following after stop all work and allowed the reader to catch up.

Increased io_write threads to 16
Increased page_cleaner to the same number of pool_instances
Set slave_preserve_commit_order to 1
Set slave_pending_jobs_size_max >= to max_allowed_packet
Set slave_parallel_workers to 5
Set slave_parallel_type to LOGICAL_CLOCK

After restarting everything the reader hasn't fallen behind (other than 5 or 10 momentarily, but cleared as fast as can be refreshed).