Als sekundär PBS wo die Restoreperformance egal ist, nutze ich auf öfters große HDDs. Sonst immer brav auf SSDs. Aber zum testen reichen ja erst einmal HDDs.
Was man sich bei RAIDZ (egal welches Level) klar sein muss: Das ist immer nur so schnell, wie eine Platte. Das kann gerade bei HDDs und so Sachen wie restore oder der garbage-collection schon gut Sachen in die Länge ziehen. Mirrors oder striped...
ergaenzend zum post von @Dunuin (das zu 99% stimmt):
der client fragt nicht beim server nach, ob ein chunk schon existiert (das waere sonst ein ziemliches einfallstor fuer information leaks ;) ). stattdessen passiert folgendes:
1. client...
Das läuft so:
Sagen wir du hast ne 100GB virtuelle Disk von einer VM. Beim Backup zerstückelt PVE die 100GB in 25.000x 4MB Chunks und berechnet dann für jeden dieser Chunks die Prüfsumme. Dann fragt PVE den PBS: "Hab hier ein Chunk mit Prüfsumme...
I have just updated one of my Proxmox nodes and still getting the same error unless you mean TrueNAS?
Proxmox node new package list:
proxmox-ve: 9.1.0 (running kernel: 6.17.9-1-pve)
pve-manager: 9.1.5 (running version: 9.1.5/80cf92a64bef6889)...
It seems, that external smtp mail sent via port 26 is not logged and cannot be displayed in the tracking center.
How do I enable that logging ?
Thanks
/ Harald
Hello Dominik,
Indeed, we are using Ceph, but we don’t see anything abnormal on the cluster side. We also don’t experience any issues with other VMs, including migrations in general.
From our perspective, it looks more like a VM-specific...
There were recently some bug fixes in this area. Make sure you are running with the latest code.
Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
@Ripon
And more to point out:
It is free software and not free beer.
The price is very decent compared with other Enterprise Virtualization solutions.
If you use the software to earn money consider to buy at least the smaller subscriptions...
I just noticed something, when I moved the vm to the other node (which as nearly instant as it wasn't running), and then reboot the original node, the vm is not automatically started.
I changed the HA state of this VM to Stopped, and now it...
You can send me your email address in direct message and I can pass the file on to you by email. The way I described works out with lubuntu, but maybe it is easier to follow futiless suggestions here.
There were recently some bug fixes in this area. Make sure you are running with the latest code.
Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
You are missing critical step of creating a file system, as I've described in your other thread.
However, you should not be creating regular file system on your shared block device.
I'd recommend continuing to use your thread vs necroposting...
this is old as sin. but i'm having a similar issue. i was wondering if you might be able to point me in the right direction.
i have a FC SAN. i've added the WWID into multipaths and set the conf file. when i do the top step it works fine when i...
Hi @williamss159 ,
If you want to store ISO, which are files, you need a filesystem. Filesystems can be created on top of a raw device , partition, or LVM slice.
However, cluster of nodes is a special case. It requires a cluster aware...
yes, thanks, i can see the folloing messages:
2026-02-06T13:41:26+01:00 agorapverssi1 QEMU[13459]: kvm: migration_block_inactivate: bdrv_inactivate_all() failed: -1
2026-02-06T13:41:26+01:00 agorapverssi1 QEMU[13459]: kvm: Error in migration...