This! And yes.
Yes! Without Quorum you can not administrate the single surviving node. (There are workaround for troubleshooting, but not for continuous use.)
Do not think about Ceph before reading this...
Wenn du lange genug gewartet hast, daß du denkst es tut sich wirklich nichts mehr (zB 15min), dann kannst du es mit STRG-C abbrechen.
Danach mach einfach ein "dpkg --configure -a && apt update && apt upgrade -y && apt autoremove -y".
Wenn fertig...
Contact Proxmox sales office or where you bought the support subscription directly, or open a (free) support ticket for this (via your shop account). The volunteers on this forum cannot really help with subscription issues but the (re)seller can...
ZFS can utilize several algorithms. I do not play with them, I just use the default. But to squeeze out some percent:
~# zfs set compression=thisdoesnotexist rpool/dummy
cannot set property for 'rpool/dummy': 'compression' must be one of 'on |...
ok have found solutions to bring all nodes with all vms in running mode without any problems to new cluster. with infos about on top but step by step with clean configs
Im täglichen Backup sollten auch alle configfiles liegen, aus denen man widerum alles entnehmen kann. Wer sowas nicht mitsichert, erschwert sich unnötig ein mögliches Desaster.
the forum is community so it is a highlight that staff members are even present and answer questions patiently
what do you expect (seriously, literally)?
/etc/pve/qemu-server und /etc/pve/lxc (OpenVZ ist noch da, aber wohl schon lange nicht mehr verwendet) zeigt immer die lokale node. Aber unter /etc/pve/nodes/NODE/... sollte die Ordnerstruktur für alle Nodes da sein.
In der GUI sollte aber auch...
Wenn man "advanced" oder "erweitert" auswählt schon:
Am besten auf "Latest" stellen, in der Liste habe ich auch die Version 6 gefunden, aber die ist wirklich schon sehr alt, aktuell ist die 10 ;)
There is an FAQ entry somewhere on https://shop.proxmox.com/ ... ... --> https://shop.proxmox.com/index.php?rp=/knowledgebase/5/How-many-subscriptions-do-I-need-for-my-servers.html
Basically you need a subscription for each and every CPU...
Ich nehme mal an, dass die Soundkarte sich eine IOMMU Gruppe mit etwas wichtigem teilt dass du nicht durchreichen solltest.
Es ist einer der Gründe warum ich immer Resource Mappings empfehle da hier davor gewarnt wird.
Eine einfache Möglichkeit...
You can get help with this documentation : https://pve.proxmox.com/wiki/Renaming_a_PVE_node
There are a few other things to change (like corosync configuration) as you seem to be in a cluster.
To be (much) clearer I was referencing 3 hosts and assuming multiple OSD on each, with at least one left running, not 3 hosts with only 1 OSD.
For the former, Ceph will use any other OSD on the same host (technically any unused host, but there...
Does it?
With the failure domain being "host" this does not make sense...? I am definitely NOT a Ceph expert, but now I am interested in the actual behavior:
I have a small, virtual Test-Cluster with Ceph. For the following tests three Nodes...
The simple 'Use physical CD/DVD Drive' option for the virtual CD/DVD Drive only supports data discs with files on it. If you want to rip audio or read DVD/Bluray movies (or burn discs) then the VM needs low-level access to the drive. It might be...
Yes, we don't talk about the same thing. It's not ONLY the filesystem, it's also the backend storage (where there is no filesystem) and that's what I'm talking about. Unless you have a SHARED write cache, as good enterprise SAN storage solutions...