Are you using Ceph 20.2.x as EC got major improvement with 20.2.x ?
If you are mixing drives, expect the storage to be as "fast" as the slowest disk.
Maybe instead of mixing disks you could create different pools with different crush rules...
Update!
Before the system upgrade, we replaced some of the drives on the server, including the system drives. Looks like the hard drives configured to boot *somewhere* are not correct. So it's likely I messed up the disk replacement and somehow...
I'll drop this here: https://pve.proxmox.com/wiki/Multipath
We are currently also working on getting iSCSI, LUN and multipathing integrated into the Proxmox VE tools and web UI. So keep an eye out for it in future release notes.
Hello dear people.
I have A LXC with a Zerotier-Controller set up, and also installed Zerotier on the Proxmox Host.
Proxmox Host > ZT Central, which my remote machine (normal ARCH Desktop) is connected to. On ZT central I have set up a Managed...
I'll drop this here: https://pve.proxmox.com/wiki/Multipath
We are currently also working on getting iSCSI, LUN and multipathing integrated into the Proxmox VE tools and web UI. So keep an eye out for it in future release notes.
Hi!
Currently working on the upgrade. One of the nodes went pretty well, but the second one didn't. It's as if the kernel is no longer there, and when starting the server, all I get is a black screen and a message saying 'Booting from Linux...
So, wie versprochen nach ein wenig Zeit kann ich sagen dass Bu66as wie bereits geschrieben die Lösung klar geschrieben hat. Alle anderen danke ich für ihre Mühe.
Tread kann geschlossen werden.
Danke
It’s an isolated cluster running some core services, so we’re not planning to upgrade it for now. Plus, it has Broadcom RAID controllers that have issues with kernel 7.
It’s nonsense to ask which one is better when the outcome is the same.
Since neither is likely to receive any future updates, it doesn’t matter which one you choose as long as it works now.
It’s not a good idea to leave a decision that isn’t...
Hi,
I still haven't been able to solve the problem, could anyone help me?
Maybe the problem is due to step 7 (In the hardware tab of the VM change BIOS to SeaBios from UEFI) of @FormerVMW 's procedure?
Thank you
Moin
Du hast zwar das
erwähnt, aber ich würde mir trotzdem sehr gut überlegen ob ich irgendeiner KI tatsächlich meine kompl. Proxmox-Konfig. und somit ja auch irgendwelche Zugangsdaten, offen legen würde. Und das nur weil mir irgendeine KI...
The best place for recommendation on multipath file is your vendor's documentation. I think this may be what they want, but I'd check with them first:
devices {
device {
vendor "Lenovo"
product "DS4200"...
So, wie versprochen nach ein wenig Zeit kann ich sagen dass Bu66as wie bereits geschrieben die Lösung klar geschrieben hat. Alle anderen danke ich für ihre Mühe.
Tread kann geschlossen werden.
Danke
Currently, we don't have any documentation on the specifics of all the metrics. That said, we are simply forwarding the kernel metrics. Other documentation, such as the Linux manual, explains some of the metrics. I also found this email on the...
Thanks for the explanation.
So I could just have set on the multipath.conf the option find_multipath "strict", and he would find it automatically ?
Then I could just create the volume groups with the name DATASTORE* ?
I think I got it, also is...
WENN man gute Hardware hat (!) UND man Updates nicht zwingend installieren muss/will, DANN sind ein paar hundert Tage Uptime gar kein Problem.
Willkommen in der Linux Welt ;)
Ceph is RAM hungry, not really great at high ram cost right now. As we are lucky and have enough ram(still) we are using 2gb per 1tb of NVMe storage.
The default setting in Proxmox is 2GB(flat) and that is waaaay too low for large NVMe disks...
Finally got a chance to check on this. The problem seems to be related to a recently reported bug [1] that causes ARC to evict all/most of it's contents erroneously, even when there's plenty of RAM available.
This causes ARC to almost never...
There is no built-in way to assign user-friendly names to the devices, if you want to continue using DATASTORE* - you have to continue doing it manually.
You didnt specify what version of PVE you are using. In PVE9, for example, ...