gurubert's latest activity
-
gurubert replied to the thread Datacenter und/oder Cluster mit local storage only.Ich kenne das Problem nur von OCFS2. -
gurubert replied to the thread Ceph DB/WAL on SSD.This is correct. But depending on how many nodes you have this is not critical.
-
see https://pve.proxmox.com/wiki/Deploy_Hyper-Converged_Ceph_Cluster#pve_ceph_osds "The WAL is placed with the DB, if not specified separately" IIRC the default space usage is 10% (?) of the OSD size but you can adjust. Side note: the SSD is a...
-
gurubert replied to the thread Empty Ceph pool still uses storage.The rbd_data.* objects seem to be leftovers. As long as there are no rbd_id.* and rbd_header.* objects there are no RBD images in the pool any more. The easiest way (and when you are really sure) would be to just delete the whole pool.
-
gurubert replied to the thread Datacenter und/oder Cluster mit local storage only.Seit 6.14 ist aio=threads nicht mehr notwendig.
-
gurubert reacted to sgw's post in the thread Datacenter und/oder Cluster mit local storage only with
Like.
Ich hab es nicht erwartet und bereits einen Node mit 6.14 gebootet. Interessanterweise booten auch VMs mit "aio=io_uring" ohne augenscheinliche Fehler. Ich teste .. Wir haben jetzt eine Handvoll VMs dort laufen, vorsichtshalber mit dem... -
gurubert replied to the thread strange ceph osd issue.Buy a non-defective enclosure.
-
gurubert replied to the thread Ceph Storage Unknown Status Error.You need to configure the old IP addresses from the Ceph public_network on the interfaces before you can do anything with the Ceph cluster.
-
gurubert replied to the thread Ceph Storage Unknown Status Error.You cannot just change IPs in ceph.conf. The first step is to add the new network to the Ceph public_neteork setting, then add new MONs with the new IPs to the cluster and after that remove the old MONs. Only after that was successful the old...
-
Presumably. That's what it does if you do a manual shutdown.
-
Maybe you should have only the PVE host do the shutdown and not also the guest.
-
gurubert reacted to alexskysilk's post in the thread [TUTORIAL] FabU: can I use Ceph in a _very_ small cluster? with Like.
There is no "migration" with only three nodes. the "3" in your crush rule refers to how many copies on individual nodes that have to exist in order to have a healthy pg (placement group.) the number of OSDs dont matter in this context- you can...
-
Yes, I am/was keen to get some of them too. That's really a bummer :-( I wanted to put two to four OSD in each of it, the actual constrains should allow for four. Now look at https://docs.ceph.com/en/mimic/start/hardware-recommendations/#ram ...
-
gurubert replied to the thread Custom Rules - Ceph cluster.You can change the crush_rule for a pool. This will not cause issues for the VMs except a maybe slower performance during the time the cluster reorganizes the data.
-
The Cephfs storage should not use 10x what you're storing in it. I would look at it on disk and see what is actually being used. host:/mnt/pve/cephfs# du -h 0 ./migrations 0 ./dump 8.2G ./template/iso 0 ./template/cache...
-
gurubert reacted to Bob.Dig's post in the thread [TUTORIAL] Enabling SMB3 Multichannel in Proxmox with Like.
Not true for Windows Server either.
-
gurubert reacted to seraph's post in the thread [TUTORIAL] Enabling SMB3 Multichannel in Proxmox with Like.
This is true for Windows Server. As far as I know, when using Samba, the only validated and recommended way is to use a different subnet. I see no advantage in not following the recommendation.
-
gurubert reacted to MacGyver's post in the thread Check the disk usage of a virtual machine with Like.
> apt install jq > qm agent 100 get-fsinfo | jq -r '.[] | select(has("total-bytes")) | .mountpoint + " is " + ((."used-bytes" / ."total-bytes" * 100) | round | tostring) + "% full"' Example output for a *nix VM: /var/log/journal is 67% full...
-
We have customers who do run 5-node full-mesh clusters, for example with 4x 25Gbit NICs. Do not go for a ring topology as that could break in 2 places and then you have problems. The Routed with Fallback method is what you want...
-
Yes, you can run a 5‑node Ceph cluster with just DAC cables (Max number of nodes recommended for this setup will be 3 nodes) —if you treat the SFP+ links as a routed L3 ring or mesh, not a Layer‑2 loop. Proxmox has a documented method for this...