Let’s assume you might not following the Ceph hype for whatever reasons or experiences. And also you might have found a working solution with Sheepdog for clustered block storage. And you’re probably using Proxmox, where Sheepdog is very well integrated (uses existing Corosync cluster). Well done ;) Proxmox sometimes ships Corosync updates, which means the Corosync cluster service will be restarted. As Sheepdog relies on it, it might get confused and dazed. Read more

If you want to use kexec for faster (warm) reboots with Proxmox, have a look at my forum post.

You can use KVMs direct kernel boot feature to boot things without having an ISO or using PXE. These options can be set in args: in Proxmox. Doing so, we can simply kickstart a CentOS or preseed an Ubuntu installation by passing all necessary options. UPDATE 2017-11-06 Ansible approach Example: CentOS Prepare kernel Grab CentOS kernel (if not alreay done, then skip this step and adjust paths in qm create command) Read more

I’m shure i don’t need to telly you why differential backups are great. If so, Kamil Trzciński (the original author) does! We are using this for quite some time now and had no issues except more disk space ;) I’ve wrapped things up on Github (I’m not sure that my private webserver last’s forever, but Github probably does)

Having a cluster of Proxmox nodes is just fine, but for some reasons (unified web+api access, signed certificates) we want to have a reverse proxy in front of our cluster members. We’re using Apache2 as HTTPS load balancer and HAProxy as TCP load balancer for Spice/VNC sessions. Basic Proxmox Reverse Proxy config can be found at James Coyle’s blog including hints how to setup Apache2. To cover a cluster of multiple nodes, you need to add the balancer section. Read more