Search results

Just stopping in to share something that might be useful. I rolled the dice on 16 X 2TB Kingston DC500 SSDs for our new cluster at work. They claim to have full data path protection with capacitors on the board and were the lowest price of any drive in this class. No expectation of high end...

Presumably, your VM's are on different bridges than your proxmox management interface is. Many people would consider that a good thing. Your VM's shouldn't necessarily be able to ping your cluster nodes unless they have also been assigned an interface on the same bridge intentionally. Many...

Hi fusrohdan, Have you run a disk performance test within the VM... like, a file copy test? I would get a few tens of GB worth of big files duplicated in a folder and do some copy operations on the windows server VM itself. Take the network out of the equation. Figure out whether the...

1 CPU and only 1.46GB RAM? MS recommends an absolute minimum of 2GB with desktop experience installed. I would give it 2 cores and 8GB RAM and retest before chasing this problem anywhere else. Also.... with CPU type "host" selected for a VM, it should pass through ALL instruction...

Fix confirmed. All nodes rebooting perfectly again. Thanks!!!

Also interested in this. Planning a cluster move in June.

This thread is not talking about everyone. OP is already behind another private network and lives with the hardware. Would not likely be performing administrative changes remotely in such a deployment. Virtualized firewalls are commonplace in professional cloud based environment deployments...

Alwin, I'm well aware it is available as free software. Running the free version now on my homelab and it is very exciting to have. In the few months of running the "free" version of proxmox, updates have introduced regressions twice now. In the name of development, there's nothing wrong with...

Thanks Tom!!! Very excited for fix. Good work guys!

I don't think you understand what I'm saying. Consider the following: Cluster A: 5 node cluster of single socket Epyc Rome servers with 64 core CPU's and 2TB RAM per server. "Basic" support is $1350/yr for cluster A. Cluster B: 16 node cluster of dual socket Ivy Bridge servers with 10 core...

The separation I'm speaking of here, is not the community repo access vs enterprise support. If you read the entirety of my response, the separation I'm speaking to here is the licensing model for repo access vs support. You're responding to me out of context. Currently you're licensing repo...

I was actually planning to ask the same question today. Socket count is not a good indicator of cluster capacity anymore. My X9 generation 4-node, 8 socket homelab cluster has less total compute and RAM capacity than one single socket epyc rome server. I do this for fun/learning at home, so...

At a loss... I reinstalled proxmox on a node, removed and rejoined it to the cluster, copied in my networking config, and a few other odds and ends. The "clean" install node was restarting fine, until I installed ceph on it. Problem came right back on freshly installed and joined node as soon...

I agree, definitely some fishy graphs/charts going on there. Mine are weird too. Instead of selecting the hosts, try selecting the relevant pool instead. That at least seems to get the datacenter view to align to the ceph view. In my cluster, the "host view" doesn't seem to be obviously...

Click on your user menu in the upper right hand corner. Select "My Settings" Select the storage you would like to see displayed on the datacenter summary page. With "nothing" selected, I'm not sure what it "defaults" to, but mine looks weird too. If I select my ceph pools, then the numbers...

I switched to NIST time servers in boulder (not far from where I live), restarted/updated my time sync service on all nodes one after the next. All my clock skew issues went away after that. I also see the timesync updates in syslog on node reboots now, which also likely helps a lot. Good change!

There's nothing wrong with a 4 node cluster. You have a "perfect" cluster for 1-node failure without loss of quorum. Having a 4 node cluster with all 4 nodes hot to step in for a node failure is better than a 3hot+1cold cluster any day. If the 4th is cold, and you loose a node, then you loose...

I don't understand your point. If the internet connection goes down, there's not going to be any way to remote in to fix it regardless. Being virtualized doesn't change this. If you're making changes to your firewall configuration "remotely" to a site with a single internet connection that's...

I suspect more people do this than you think. Some of us like to live on the wild side ;) proxmox doesn't just stop working without a gateway, it just looses access to updates and time servers. It can go awhile without either in most cases. You could just boot the last backup of pfense? I...

Few tips for what you're trying to accomplish: 1. I see nothing wrong with virtualization your router, especially if you're already inside another private network anyway. if you bork up your pfsense VM and need internet for proxmox, you can always give proxmox an IP on the "WAN" (private school...