UdoB's latest activity
-
So you have NO full-VM/LXC backups that can be restored? https://github.com/kneutron/ansitest/blob/master/proxmox/proxmox-recover-vm-disks-without-backup.sh If you have another proxmox server that can be temporarily used for recovery ( ideally... -
Make. A. Backup. Of everything important. NAS with 2-disk failure redundancy, and everything on UPS power is a good place to start. Full bare-metal backup of Windows installation to separate disk with e.g. Veeam free agent, if you need to back...
-
UdoB replied to the thread Upgrade OVE8 zu 9 / Target Packages is configured multiple times.Nein, nicht wirklich - nur eben doppelt. Die obere Version ist die alte Syntax, unten ist das neue Format. Eines genügt. ---- Nachtrag: https://pve.proxmox.com/wiki/Upgrade_from_8_to_9#Add_the_Proxmox_VE_9_Package_Repository "Then you can...
-
UdoB replied to the thread raid5 and 2 logical disk included in one zfs pool.Okay, so my understanding was wrong, sorry. I'm with @Onslow in#4. ZFS should have direct access to each physical device. With a hardware Raid controller that's not the case... A lot of controllers have an "HBA"-mode selectable in their Bios or...
-
UdoB reacted to Onslow's post in the thread raid5 and 2 logical disk included in one zfs pool with
Like.
Well, "Do not use ZFS on top of a hardware RAID controller which has its own cache management. ZFS needs to communicate directly with the disks. An HBA adapter or something like an LSI controller flashed in “IT” mode is more appropriate." (... -
UdoB replied to the thread Proxmox freezing on install when loading drivers..Did you select the "dd"-method? If not: give it a try.
-
UdoB replied to the thread raid5 and 2 logical disk included in one zfs pool.Probably your terminology is wrong, there is no RAID5 in the context of ZFS. You probably mean RaidZ1 which represents a vdev (virtual device) with a single redundant device. One drive may fail without data loss. Yes, you can have several of...
-
UdoB reacted to Skuld's post in the thread [SOLVED] Kernel driver loading despite blacklist with Like.
Looks like blacklist didnt work because i installed the nvidia driver via the .run installer from nvidia itself. If someone else has this problem, just run nvidia-uninstall.
-
In that case I would just do it simple: Just restrict the root user to SSH keys only. (And if you worry about spam attemps just add fail2ban to monitor and possibly block failed SSH login attemps.) set in the sshd config: PermitRootLogin...
-
I can't imagine ClamAV being a problem for Proxmox or their support, as it is a standard Debian package. Best to contact Proxmox office or sales directly for official answers to your questionnaire: https://proxmox.com/en/about/contact
-
UdoB replied to the thread PVE with antivirus.I'm with you. But the argument is weak: there are 64000 packages in Trixie - probably not all of them are harmless for PVE... ;-)
-
In my opinion, you're asking the wrong question. Instead of asking whether antivirus software could cause problems on the host, you should first ask whether it is needed there at all. Only if you come to the conclusion that it is needed your...
-
UdoB replied to the thread PDM Beta realesed testing.Your ideas look really interesting! The recommended place for Bugs and also for Feature Requests is over there: https://bugzilla.proxmox.com/enter_bug.cgi?product=pdm While several Proxmox staff members do regularly use this forum (thank you...
-
Ich auch, sie haben aber das Problem potentiall angreifbar für eine sogenannte side-channel-Attacke zu sein: Die Frage ist natürlich wie man das bewertet, bei mir im Homelab nutze ich KSM sehr gerne, in Endeffekt habe ich da ja keine VMs, die...
-
UdoB replied to the thread Benachrichtigung bei VM-Ausfall.Prima! Wie groß ist die Datenbank? In meinem Homelab habe ich momentan ~100 GB in einer Mariadb (und im Job sind es knapp 200 GB in PostgreSQL). Der kontinuierliche Schreibvorgang stresst jede SSD --> ein Grund mehr für "Enterprise Class w/ PLP"...
-
Das läuft bei mir im Docker und ich finde es extrem schlank. Kaum meßbar CPU und knapp 2G RAM. Für alle Container zusammen.
-
Dann wäre es sinnvoller keinen Cluster zu bilden, sondern sich eine VM mit dem Proxmox-Datacenter-Manager aufzusetzen und darüber die Migration zu machen. Alternativ geht das auch auf der Konsole mit pct remote-migrate für Container oder qm...
-
> a way to only have one vm being backed up at a time I was verbalizing a similar suggestion. :) Currently, one must create one job per node with different start times.
-
UdoB reacted to fba's post in the thread [SOLVED] Shutdown of node : how VM is shutdown ? with Like.
For a VM PVE tries to use the qemu-guest-agent if available to initiate a shutdown. Otherwise it uses an ACPI command. If this doesn't work, the vm is terminated after set timeout, usually 180 seconds. For LXC it is different, as described here...
-
Could be, but i dont care about PBS in this scenario. I want the ability to control how many jobs are pushed to PBS simultaneously. Hence the request for being able to "chaining" backup jobs. The scheduler for this, i believe, lies in Proxmox VE :)