UdoB's latest activity
-
Muss denn unbedingt local vergrößert werden? Du könntest auch in dem vorhandenen Thin-Pool ein LV anlegen und dort die ISO und Templates speichern. ungefähr so: lvcreate -V 200G -T pve/data -n iso-store mkfs.ext4 /dev/pve/iso-store mkdir... -
Wouldn't it help to use something like a virtualized OPNsense/Router to seperate the vietual networks from the outer World? Or am I missing something?
-
Ich würde eine Neuinstallation mit ZFS empfehlen. Da bist du mehr flexibel. local hat da zB. Zugriff auf allen Speicher. Ansonsten: https://pve.proxmox.com/pve-docs/pve-admin-guide.html#_creating_an_extra_lv_for_span_class_monospaced_var_lib_vz_span
-
UdoB reacted to Meesoa's post in the thread [SOLVED] Proxmox refusing to ping to any network. with
Like.
So, I have found out that I am a slight bit foolish... I had not realised that when it comes to a servers MB, it isn't exactly happy with you utilising it's MGMT lan port for the proxmox, I hooked up a second ethernet cable in one of it's four... -
UdoB reacted to meyergru's post in the thread [SOLVED] RAIDz2 zpool not showing full capacity after zpool attach with Like.
I now did the same successfully with a raidz2 pool made of 8 disks with 14 TB, adding a ninth one with 16 TB (just because 14 TB drives were more expensive). After the expansion step (i.e. "zpool attach", which took about 1.5 days to complete)...
-
Regarding pfsense you might also want to read these posts by @meyergru in the OPNsense forum (they are transferable to pfsense more or less): Read this first: https://forum.opnsense.org/index.php?topic=42985.0 OPNsense for Dummies especially...
-
I'm not sure I understand what you want to achieve. You could installProxmoxVE and ProxmoxBackupServer baremetal on the node and afterwards add it to the cluster but I wouldn't do this: Since every cluster member can login via ssh to the other...
-
UdoB replied to the thread [SOLVED] Proxmox refusing to ping to any network..Are you sure that nic0 is the correct network interface and it is actually connected? It is listed as "state UNKNOWN" in the last screenshot. If you actually have three NICs just try the other two. You need to get "state UP" for both the bridge...
-
UdoB reacted to jdancer's post in the thread 9.1.1 Installation issue with PERC H730 Mini with Like.
Dell HBA330 mini-mono versions are definately for 13th- and 14th-gen Dells. For 12th-gen Dell PERCs (H310/H710/H710P/H810), use https://fohdeesha.com/docs/perc.html. No issues in production. Do NOT skip any steps and take your time. If you want...
-
UdoB replied to the thread setting up my pfsense router.Yes, OpnSense/pfSense are complex beasts. Just look at the loooong feature list! Compared with that Wireguard is really, really easy - as it does one single job, and it does it right! My preferred method is to handcraft something like...
-
UdoB replied to the thread setting up my pfsense router.The full setup needs some complexity, to be useful - right? To examine routers and network problems you may setup a virtual playground. When I wanted to play with Ceph I took one beefy PVE node and created six virtual PVE nodes, created a...
-
UdoB reacted to d.oishi's post in the thread Proxmox roadmap - future direction of snapshot‑as‑volume‑chain with Like.
I understand that the premise here is to continue using the existing SAN storage. As I understand it, there has been no public statement so far that snapshot-as-volume-chain will become fully supported in future releases. Personally, given the...
-
UdoB reacted to d.oishi's post in the thread Is it possible to have the second nic on my server to be used to access the GUI when the first nic goes down with Like.
Using two interfaces on the same subnet and assigning each of them an IP address is not a common way to provide redundancy. In general, bonding is used for this purpose. Here is an example configuration. If you want automatic failover, the...
-
UdoB replied to the thread 3 node proxmox link failure, reboots node even when second link is up.You can see the status of both "rings" this way: ~# corosync-cfgtool -s Local node ID 6, transport knet LINK ID 0 udp addr = 10.3.16.7 status: nodeid: 1: disconnected nodeid...
-
Hey Thank you for all the hints. I think I’ll use "borgbackup" directly out of the VM. I just need to back up a directory and a database. I had considered simply backing up the entire VM from one of the PBS servers - I thought that would be the...
-
Backing up from PVE is better if you are planning on doing an immediate restore, like migrating from one host to another. Otherwise take regular snapshots of your PBS data store and back up the snapshots. I do ZFS send of my snapshots to another...
-
I wouldn't do this for multiple reasons: First you need to be careful to ensure the consistency of your datastore. People tried stunts like using rclone to backup their datastore contents to cloud providers ending in broken backups...
-
UdoB reacted to bbgeek17's post in the thread Migrate Windows Server Shares to a CIF/NFS or LXC with Samba with Like.
Hi @alexinux , I think you are misunderstanding the mechanics of CIFS&NFS storage pool purpose in PVE. If we were to compare the PVE pool to your Server/Client infrastructure - PVE is the client. PVE does not serve storage but rather connects to...
-
I would strongy recommend NOT to add a third node if it's just for quorum. First a real node also needs a dedicated corosync network adapter which a qdevice doesn't need...
-
UdoB reacted to Johannes S's post in the thread Small Datacenter Setup - What is the maximum number of pve servers supported in a cluster with Like.
I recall one reddit thread where somebody run an infrastrucuture with around 30 nodes. They mentioned that their Proxmox partner and proxmox support also recommended splliting up their cluster. Different to Maximilianos post the recommendation...