Search results

Well, so my guess was wrong, sorry...

Did you verify that "backup" can write data? The host is not the problem, the user "backup" (possibly) is. Does this work, run inside your .chunks folder: root@...:/pbspool/pbs0/.chunks# su -p backup -c "touch testfile" root@...:/pbspool/pbs0/.chunks# ls -al testfile -rw-r--r-- 1 backup...

I do NOT use this firewall implementation. But I do believe the first rule "DROP" should be the last one. First allow what you want. Then, at the end, drop the rest... You can move them by dragging the symbol left of the rule number, I believe.

First guess out of the blue: My .chunks folders are owned by /pbspool/pbs0/.chunks# ls -al | head -n2 total 6340737 drwxr-x--- 65538 backup backup 65538 Aug 27 2023 . The user with the name "backup" must have write access. It is not sufficient if "root" can write data.

You are overcomplicating things ;-) The basic passwd will allow you to specify a new password for root@pam.

Yes. Create a HA group with these two nodes and select "[x] restricted". If one of these two nodes (with active HA resources) loses network connectivity it will fence itself --> hard reboot. The other two won't do that.

Naja, das sollte in der "Hilfe" der Fritz!Box beschrieben sein. Ich weiß nur, dass man definitiv Routen setzen kann. Okay, mal schauen: Heimnetz --> Netzwerk --> Netzwerkeinstellungen --> IPv4 -Routen. (In einer 6591, keine Ahnung ob das konsistent für alle Boxen gilt) Da habe ich mehrere...

Are you sure you want "bridge-ports idrac"? Not eno3/4 or so?

Sorry, keine konkrete Hilfe von mir - aber eine Rückfrage, sofern ich dies richtig verstehe: du befindest dich hinter einer Fritz!Box - die macht ausgehend SNAT / Masquerading du hast weitere Netze eingerichtet per SDN eingerichtet - das ist gut da ist ein weiteres SNAT aktiviert - und das...

Yes, three Nodes is the minimum for Ceph and your hardware is great. But I would never start a serious business with only the absolute minimum of anything. Ceph requires some more resources to be reliable; I've written down some things I've noticed...

My personal and unusual approach would be this: Run the installation to create the default ZFS "rpool" utilizing all HDDs; yes, only the classic disks. Side effect: each and every disk will be bootable. After installation (and everything basically running) and before putting much data onto it...

Yes. Note: I am not sure if I understand your construct correctly. Post your /etc/network/interfaces (in code-tags) to avoid misunderstandings on my side. That does not make sense at all! There can be only one single default gateway. And if the node needs to have an IP address inside those...

I would never start a serious business with three nodes. Ceph requires some more resources to be reliable and fast; I've written down some things I've noticed: https://forum.proxmox.com/threads/fabu-can-i-use-ceph-in-a-_very_-small-cluster.159671/ Disclaimer: I've dropped my six node Ceph with...

".zfs/snapshot/" does exist for cv4pve too. And yes. (For datasets; not for zvol of course.) Not mentioned yet: it uses the PVE API. This means that "the VM" is actively informed that a snapshot is going to happen - and it gets a chance to prepare itself for this. In my understanding this is...

Wenn du mehrere Netzwerkkarten in einem Computer verwenden willst, dann müssen diese IP-Adressen aus verschiedenen Netzen haben. Entferne zwei der Adressen! Ansonsten ist das ganze Konstrukt doch nicht eindeutig! Aus welcher Karte soll ein Paket hinausgeschickt werden, um einen Nachbarn zu...

Yes. I do run every 15 minutes and keep 4 of these. And I do run another one once in the early morning and keep two of these. Older "snapshots" are stored on the PBS ;-) While I could keep a lot more snapshots it is really painful to rollback to an older one. I was not able to find an easy...

Some hints: https://pve.proxmox.com/wiki/Renaming_a_PVE_node There may be more pitfalls... it states "This must be done on a empty node."!

Yeah, I had used that one too. Then I wanted to see "VM-snapshots" instead of "ZFS-snapshots". What I do use now is this: https://github.com/Corsinvest/cv4pve-autosnap Snapshots made by this tool are compatible with the PVE software, e.g. they are listed in the Web GUI as regular snapshots.

No, sorry. The term "replication" describes exclusively the ZFS copying capability, using the send/receive mechanism to minimize the actual traffic.

All nodes should have the same storage configuration. After a normal installation each one has an "rpool" = three instances, one on each node. There is only one storage definition for this "rpool", named "local-zfs". This storage is declared to be available on all nodes, it is "shared". The...