Finally, we’re upgrading our shared MySQL to 5.7 and this is going to happen in the next sprint (254).
We’ve already set up a new MySQL 5.7 instance, that replicates mysql-im-2.net.opencraft.hosting, and we’re going to move the fallback IP address to the new server. So, in theory, there will be no downtime. Check this message for details, code and host variables of a new server and this discovery document for an abstract plan.
Briefly, I’m going to:
Wait till the end of this week for any possible concerns.
Email clients.
Move the fallback IP address.
Rebuild replicas.
Test backup restoration.
Do you have any concerns or suggestions?
Use BB-4049 to log time spent reading and replying to this post.
That’s great news @demid, thanks for taking care of this upgrade!
Just to confirm: This step will include standardizing disk space for /var/lib/mysql across cluster nodes, right?
(For context, we were originally planning to do this as part of FAL-1995, but ended up concluding that it would make more sense to take care of it as part of the MySQL upgrade; ref, ref.)
I’m asking because disk space usage on mysql-im-2-slave1 has increased quite a bit since the completion of FAL-1995. It’s at 86% currently, despite some intermediate actions that were taken a few weeks back (cf. this thread). So we’ll need to increase disk space for /var/lib/mysql on slave1 to match slave2 soon-ish, to avoid running out of disk space.
