MaxScale Binlogrouter 2.5 with MariaDB Enterprise Cluster (galera)

[Heine Escobedo Jimenez]

Hello Markus

I'm designing the architecture for a MariaDB Enterprise Cluster considering MaxScale 2.5, Galera 4.0 and MariaDB Enterprise 10.5.

One node for writes and the others two to balancing reads, so will be used for H.A and read scalability.

Please let me consult you if is a supported configuration or is a good practice implement the binlogrouter as a binlog proxy for that read nodes get the changes since MaxScale and not through the galera replication ?

is binlog router 2.5 not compatible to use with a multi-master(galera) replication ?

Thanks in advance

[Wagner Bianchi]

Hello Heine,

Why don't you want to use the regular Galera API replication, and instead, you want to sync data among nodes using Binlog Router? Unless you have a good reason for that, I won't go down that path.

But, I'm curious about why you are thinking of adding the Binlog Router to the architecture.

--

Wagner Bianchi, MariaDB RDBA
Email: bia...@mariadb.com

--
You received this message because you are subscribed to the Google Groups "MaxScale" group.
To unsubscribe from this group and stop receiving emails from it, send an email to maxscale+u...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/maxscale/b5326653-a8c9-4e4e-9308-e8827c7aa9edn%40googlegroups.com.

[Heine Escobedo Jimenez]

Hello Wagner, first all thanks for reply.

Just to clarify, I'm not saying that I prefer don't use the galera replication I just wanted to validate some info showed in the enclosed file.

As you can see,  in the diagram show that read nodes that belong a multi-master cluster ( in this point I assumed that with galera) can get the changes from the proxy instead of regular way.

So, I just wanted to validate if is possible that configuration using binlog router or maybe I wrong to understand the diagram or I missing some piece on the big picture.

please let me know your thoughts and appreciate your help

Thanks again.