gurubert's latest activity
-
gurubert reacted to ThoSo's post in the thread Einführung von Einschränkungen / Kennzeichnungspflicht von KI-Beiträgen with
Like.
Ich bin der Meinung, das KI Antworten klar gegen den Community Gedanken sprechen und hier nichts zu suchen haben. Wer sich mit KI seine Probleme näher betrachtet und zu Lösungen kommt, dem ist damit geholfen. Man kann durchaus auch bei Problemen... -
gurubert reacted to fstrankowski's post in the thread Einführung von Einschränkungen / Kennzeichnungspflicht von KI-Beiträgen with Like.
Ich bitte um eine Diskussion/RFC zu o.g. Thematik. Mittlerweile nehmen die Beiträge welche erkenn- und nachweisbar zu nahezu ausschließlich KI-Content bestehen massiv überhand. Besteht die Möglichkeit dies einzuschränken und/oder eine...
-
gurubert reacted to bbgeek17's post in the thread Questions about Proxmox with shared storage, Redundancy and support for SCSI-3 PR with Like.
Hi @nvanaert , There's quite a bit to unpack here. First and foremost, Proxmox VE does not monitor storage or network path health. An All Paths Down (APD) condition will not result in a node fence, nor does it interact with the VM HA subsystem...
-
gurubert reacted to Khensu's post in the thread Ceph - Backfill & Remapping processes will not finished with Like.
Hey, the 5% sound heavily like the default target_max_misplaced_ratio value. Can you check with: ceph config get mgr target_max_misplaced_ratio Not sure if the autoscaler (Which you disbaled) is at fault here, check ceph osd pool ls detail...
-
gurubert replied to the thread CEPH heartbeat issue after reboot.Which is not needed when you just want to reboot a node. An OSD is set to "out" when it should be removed with ceph osd purge.
-
Problem is, most of it just doesnt make any sense whatsoever as its just thrown in a thread without even investing time to markup the post and remove the nonsense in it in the slightest way. Please invest the time to either adequately edit...
-
gurubert reacted to IsThisThingOn's post in the thread Storage for small clusters, any good solutions? with Like.
That is the money quote. But IMHO even more important, in many, many cases, making the server HA was the wrong approach to begin with. Making the application itself HA would be way cheaper, offer way better performance, less complex, and...
-
gurubert replied to the thread CEPH heartbeat issue after reboot.What do you mean by this? What type of drives are used?
-
gurubert reacted to AntonJ's post in the thread best practices network for the PVE Cluster with Like.
Would advise against this, LAG with either LACP or Active/Passive will work just fine. Depending on what switching you got I would pick one and stick to it. 25G+25G in bond0 1G+1G in bond1 FC for FC. bond1 is simply for cluster traffic...
-
Thanks, that helped for me right now. I just learned that the SSDs are not for enterprise use. Since we just have got a new bunch of servers with enterprise SSDs and more NICs, I will migrate to the new ones ASAP. Thanks for your reply and help!
-
gurubert replied to the thread Pardon my less-than-intelligent question, but is there a way to install Proxmox on a Ceph cluster?.Losing data is something else than losing write access. In an erasure coded pool if you lose more than m OSDs in an affected PG you lose data. If you have less than min_size OSDs you lose write access to the placement group.
-
gurubert reacted to Johannes S's post in the thread Pardon my less-than-intelligent question, but is there a way to install Proxmox on a Ceph cluster? with Like.
Although they were about replicated pools (so no ec) following reads might serve as a hint why (outside of experiments/lab setups) it's not a good idea to go against the recommendations...
-
gurubert replied to the thread Pardon my less-than-intelligent question, but is there a way to install Proxmox on a Ceph cluster?.With size=min_size you cannot lose any OSDs without losing write access to the affected objects. And it has nothing to do with number of nodes or number of OSDs.
-
gurubert replied to the thread Pardon my less-than-intelligent question, but is there a way to install Proxmox on a Ceph cluster?.Yes. In erasure coded pools with m=2 you can lose 2 OSDs for one PG at the same time without losing data. The same can be achieved in replicated pools with size=3. You can lose 2 OSDs for a PG without losing its data.
-
Ooh, consumer crucial, not good, not good.
-
gurubert replied to the thread Pardon my less-than-intelligent question, but is there a way to install Proxmox on a Ceph cluster?.This is not recommended and certainly not HA. With m=1 you cannot loose a single disk. An erasure coded pool should have size=k+m and min_size=k+1 settings which would be size=3 and min_size=3 in your case.
-
gurubert replied to the thread Pardon my less-than-intelligent question, but is there a way to install Proxmox on a Ceph cluster?.No no no. You got your math wrong. To achieve the same availability as EC with k=6 and m=2 you need triple replication (three copies) meaning a storage efficiency of 33%. It is rarely necessary to go beyond 4 copies.
-
gurubert reacted to alexskysilk's post in the thread Pardon my less-than-intelligent question, but is there a way to install Proxmox on a Ceph cluster? with Like.
"lower" and "higher" are subjective. Ceph achieves HA using raw capacity. suit yourself. this is not a recommended deployment. You are far better served by just having two SEPERATE VMs each serving all those functions without any ceph at all-...
-
gurubert reacted to alexskysilk's post in the thread Pardon my less-than-intelligent question, but is there a way to install Proxmox on a Ceph cluster? with Like.
The number of OSDs isn't relevant to a pool as long as it is larger then the minimum required by the crush rule. For example, If you have an EC profile of K=8,N=2 rule, you need a minimum of 10 OSDs DISTRIBUTED ACROSS 10 NODES. so 1 OSD per node...
-
Hi, most likely you did install PBS on top of a vanilla Debian installation? In that case the vanilla Debian kernel should be uninstalled, e.g via apt remove linux-image-amd64 'linux-image-6.1*'