Search results

Could solve the problem on my own with: ceph-volume lvm zap /dev/nvme3n1 --destroy fdisk /dev/nvme3n1 (just hit W) And re-add it via proxmox gui again. and finally assign device-class again with ceph osd crush rm-device-class osd.29 ceph osd crush set-device-class nvme osd.29

Dear Proxmox/Ceph-users, i have the strange problem, that two disks seem to use the same osd ID. This is a 3 node proxmox 6 cluster. root@adm-proxmox02:~# ceph-volume lvm list ====== osd.19 ====== [block]...

root@MCHPLPX01:~# corosync-cfgtool -s Local node ID 4, transport knet LINK ID 0 addr = 172.16.1.1 status: nodeid: 1: localhost nodeid: 2: connected nodeid: 3: connected nodeid: 4: connected logging { debug: off to_syslog: yes }...

I have the same 'problem'. Maybe it's just a display issue, as the Proxmox VE Administration Guide notes: "Even if all links are working, only the one with the highest priority will see corosync traffic."

Even though I'm well aware that consumer SSD/NVME should not be used, it is reasonable to try to get the most out of "cheap" disks when budget is limited. I made the following observation and would like to discuss the pro/cons of this tunable. Model Number: SAMSUNG...

Thanks for beeing that clear. How can i make sure that a SSD/NVME has a supercapacitor on it?

Before going for expensive microns, i wanted to get some numbers: NVMEs (cheap KINGSTON SNVS1000GB): fio --ioengine=libaio --filename=test12345 --direct=1 --sync=1 --rw=write --bs=4K --numjobs=1 --iodepth=1 --runtime=60 --time_based --name=fio --size=10G write: IOPS=434, BW=1739KiB/s...

works.

Just checked the manual. With luminous you can set device classes per device - its kind of overwriting the detected class: https://ceph.io/community/new-luminous-crush-device-classes/ that should do the trick, will test.

Hi, i dont quite get it. You do not just "add" disks to a pool, you create a crush map rule by ceph osd crush rule create-replicated replicated_rule_ssd default host ssd So it only gets replicated to this type of device. I'm not aware of another way to -as you recommend - "Just use the hint...

Hi, we just put our first NVMEs in the existing proxmox 6 & ceph nodes. Unfortunately the NVMEs are detected as type SSD. As we want to have a new pool with only the faster NVMEs, we can not distinguish the disks by type: (ceph osd crush rule create-replicated ....) What can we do? thank you.

Well your cluster resource manager should handle this. For example, with pacemaker cluster you simply set constraints between the resources: https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/7/html/high_availability_add-on_reference/ch-resourceconstraints-haar This could...

This is ignored if HA is active. See your own admin manual ;)

Hi folks, we have several multi-node proxmox clusters. Sometimes we have to cold-start all nodes (full power outtake onsite) that causes a race-condition between VMs. Servers start before firewall/domain controllers are up. (wrong network profiles on windows) Application servers start before...

Works. thank you Aaron! :)

Hi, proxmox 5.4-13 server 2012r2 Would like to add another disk and use scsi for it. Device manager shows device, but can not install drivers. Tried several ISO files. Any ideas? Thank you Stefan

Awesome. thats it. Thank you.

Hi folks, anyone ever tried these kind of cards? https://www.delock.com/produkte/G_89835/merkmale.html?setLanguage=en We would like to use NVMe-cards in it. Our boards support bifurcation. Ideas?

Hi, i have a strange behavior on my linux system and can not explain it. Help would be greatly appreciated. host proxmox1 with a trunk/bond interface to Cisco-Switch. proxmox1 has mgmt in vlan100 on same bond: bond1.100@bond1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP...

even when blacklisted amdgpu did it not work for me. System dies with kernel oops righter after boot with latest pve and following hardware: vendor_id : AuthenticAMD cpu family : 23 model : 17 model name : AMD Ryzen 5 2400G with Radeon Vega Graphics stepping : 0 microcode ...