Upgrade hyperconverged cluster with proxmox 6.2-4

[luigi]

Hi everyone,

I have a customer who has proxmox (6.2-4) with full mesh three node hyperconverged cluster (without 10g switches) with the following features for each node:

2 x 8 core CPUs
128GB DDR4 RAM
2 x 240GB Mirrored HD SSD for Proxmox
5 x 1TB HD SSD Drive for OSD
1x QuadPort Eth1G (for cluster and VM traffic)
1x QuadPort Eth10G (for ceph v 14.2.9)



The customer has no longer had an active subscription for approximately a year. I would like to switch to the community edition and update to the latest 6.x version available so I can upgrade to 7.

I have a series of questions:
1) How can I edit repositories in the community edition?
2) The cluster has been updated for about two years with the active subscription to 6.x, now that I will switch to the community edition, will updating to the latest available release of 6.x work?
3) Is there a way to make sure upgrading from 6.x to 7.x doesn't cause problems for ceph?

Thanks in advance
Louis

 

[cheiss]

Hi,

1) How can I edit repositories in the community edition?

Either by directly editing /etc/apt/sources.list and maybe files under /etc/apt/sources.list.d/.
You can also see our guide about Package Repositories, which explains that topic in more depth.

2) The cluster has been updated for about two years with the active subscription to 6.x, now that I will switch to the community edition, will updating to the latest available release of 6.x work?

Yes. There are no differences in the packages in that regard.

3) Is there a way to make sure upgrading from 6.x to 7.x doesn't cause problems for ceph?

We have a very detailed guide about upgrading from 6.0 to 7.0: Upgrade from 6.x to 7.0
This guide also covers hyper-convereged Ceph clusters; see the section For Hyper-converged Ceph in the guide.

Hope this helps!

 

[luigi]

Thank you so much.

Before upgrading I would like to make a backup of the physical pve hosts. Which directories should I backup so I can go back in case of problems? This proxmon cluster is a production environment.

regards

 

[cheiss]

Before upgrading I would like to make a backup of the physical pve hosts. Which directories should I backup so I can go back in case of problems? This proxmon cluster is a production environment.

Generally, all VMs and LXCs should obviously backed up somewhere outside the cluster. When using the integrated backup feature, these also include the current VM/LXC config. So that already covers everything you need to restore a host, if you haven't modified anything yourself on the system.

You can also just back up the entire /etc folder, that should catch everything you might need, to be sure. That holds all the system configuration, and rather have some unnecessary things included in the backup then missing something.

The /etc/pve folder is the most important here, as that holds all the cluster configuration.

Edit: Regarding host backup, you might also want to have a look at this thread: https://forum.proxmox.com/threads/how-to-backup-proxmox-configuration-files.67789/