First steps with CC, creating an empty Percona CLuster

Horacio76

November 09, 2018 20:07

We just installed CC and manage to get 6 VM running Centos.

We need to get to know CC previous to use it in prod.

We have a replica Slave on MariaDB on SUSE.

We manage to create a cluster and deploy 2 VM with PXC

How can I get the data from MariaDB on SUSE.

Should I install a standalone Percona 5.7 on Centos VM as Slave of MariaDB on Suse?

Comments

7 comments

Paul Namuag

November 09, 2018 20:30
Horacio, sorry to say this but can you please take your question one at a time? I am really confuse a bit here as it seems you're mixing the question on different environment.

IIRC, you guys have the following:
- 1 CC node with 6 VM running CentOS
- Another CC node (previous) with 2 VM with PXC
- You have a replica slave on MariaDB running on Suse env
Now your question is how to get data from MariaDB on Suse?

It's confusing though. Maybe you can take it one at a time and draw the picture clearly. It would be nice you can clarify the situation.

Thanks.
0

Comment actions Permalink
Horacio76

November 09, 2018 20:51
Sorry, I'll clarify.

we have in total 6 VM with centos to freely experiment.

1-CC

3-PXC(previously 2, we manage to add one more) No data yet.

2- free to use. Future Load balancer, p.e.

The data we need to use is in MariaDB en SUSE. How should I proceed?

Thanks
0

Comment actions Permalink
Paul Namuag

November 11, 2018 20:23
Hi Horacio,

Thank you for clarifying it.

Let me ask you the following:

- What version of MariaDB running on SUSE do you have?

- How large are your dataset?

If version < 10.2, you can use Percona Xtrabackup considering that your data is huge. Since this is a slave, I suppose this is fine to run Xtrabackup on this node but make sure the traffic is low at that moment. If version is > 10.2, then you should use Mariabackup. Using binary backup makes faster provisioning of your new cluster to be managed by ClusterControl. Once you have the binary backup, then instantiate it to one of your cluster nodes running on PXC and let the other nodes SST from that node.

You can start checking https://www.percona.com/doc/percona-xtrabackup/2.4/howtos/recipes_ibkx_local.html
0

Comment actions Permalink
Horacio76

November 12, 2018 20:15
Hi Paul,

It is MariaDB10.1.4, on SUSE 11 no SP. I'll try to install Percona XtraBackup.

Also, I'll install Percona 5.7 on another Centos VM, but I can't find how to do it with CC
0

Comment actions Permalink
Paul Namuag

November 12, 2018 20:21
Hi Horacio,

Using CC to handle the installation is pretty simple and straightforward though. You just have to click Deploy and then setup the keys, then add the nodes. Just make sure you have click "Install Software" or is enabled so it will install the package you want to install. Afterwards, select the version of Percona Server and then add the nodes hostname/IP.
0

Comment actions Permalink
Horacio76

November 14, 2018 16:13

Edited
Hi Paul, this is our plan.

I need to provision data from Slave on Maria DB on SUSE at Datacenter 2 to Stand Alone Percona 5.7 on Centos already in CC. Our goal is that Percona 5.7 Node acts as Slave of Datacenter 1(SUSE) and Master of Galera Cluster.

What are the steps to accomplish that? Percona XtraBackup on SUSE didn't work.

We select a small DB from Production, less than 20Gb and replicated to VM on a test environment, unfortunately, this was don on MariaDB on SUSE.

Our final objective is to get to know PXC and CC, and implement an HA cluster.

Thanks
0

Comment actions Permalink
Paul Namuag

November 14, 2018 21:40
Hello Horacio,

Unfortunately, we do not support ClusterControl running on SUSE. With regards to Percona Xtrabackup, I am not sure if it does support SUSE, but AFAIK, it does not, but of course you can run this at your will on any Linux flavor.

Maybe run on Ubuntu or RHEL/CentOS platforms?
0

Comment actions Permalink

Please sign in to leave a comment.

Comments

Didn't find what you were looking for?