(MySQL Cluster) Adding a read-only master back as slave to cluster after Failover
Hello,
I've read in your docs that after a MySQL Cluster Master fails and the Slave that's promoted to Master won't add the new read-only Master back into the cluster.
I'm using the Community Edition of CC so I can't use the action menus without upgrading to a paid subscript.
How do I rebuild the MySQL Cluster so there's a Master/Slave again?
I have searched the CC docs and Google for hours but can't find the answer! :-)
Many thanks for any help
Simon
-
Hi Simon,
There's indeed a feature in CC to rebuild that master to be of a slave from that new master (recently promoted slave), and if I understand you correct, that's indeed not present in the community edition.
The only way you can do here is to rebuild it manually. You can use xtrabackup in that case. You can, for example, use these external blogs as your reference: https://blog.pythian.com/mysql-streaming-xtrabackup-slave-recovery/ or https://lostdomain.org/2013/01/01/fixing-mysql-slaves-with-percona-xtrabackup/.
-
Thank you so much for your reply. It's really helped with recovering a slave if it goes down. The issue I'm having is I can't work out how to recover a Master if it crashes.
CC will promote the Slave to the Maste (great, now there's almost no downtime and the websites continue to work). I'd like to know how to recover the original crashed Master back to the same configuration:
Recover crashed Master > Promote it back to Read/Write Master > Demote the current Master back to Slave (the original configuration).I'm clearing doin the wrong Google searches because I can't find info on how to do it anywhere! :-/
Many thanks for your help,
Simon
Please sign in to leave a comment.
Comments
2 comments