Search results

what version of ceph is running? may I know the ssd model?

I solve my own problem For LXC.conf adds the following lxc.cgroup.devices.allow: a 231:* rwm lxc.cgroup.devices.allow: a 10:54 rwm lxc.cgroup.devices.allow: a 10:229 rwm lxc.autodev: 1 lxc.hook.autodev: /var/lib/lxc/302/mount-hook.sh lxc.apparmor.profile: unconfined lxc.mount.auto: cgroup:rw...

Currently I am following the reference from https://linuxcontainers.org/lxc/manpages/man5/lxc.container.conf.5.html#lbAT https://blog.blaok.me/pass-a-device-to-a-linux-container/

How to passthrough an infiniband device to LXC? according to https://lxd.readthedocs.io/en/stable-3.0/containers/#device-types https://github.com/lxc/lxd/issues/3983...

do it as a script; mv /var/lib/vz /var/lib/vz.bak; zfs create [new pool]/var-lib-vz; zfs set mountpoint=/var/lib/vz [new pool]/var-lib-vz; zfs mount -a;

done... i really wish there is a way bcache can be use with ceph nautilus. DM-Cache really under performed.

you might need to trim your sata ssd pool... but... beware of trimming shorten your ssd life... after trim you release more ssd space and it definately helps with performance, but on the other hand it will shorten your ssd life

I came across Cockpit and its ZFS manager which works on PVE 6.x fresh install and also PVE 6.X upgrade from pve5to6 with a few ln- s It can take take of 98% of zfs cli such as trim, attache, detach, replace, snapshot, rollback, clone and etc... To install cockpit on PVE 6.0, echo "deb...

buy subscription and show your love

same here... I did a lot of V2P just for windows 10

100% correct... please beware of any Windows System virtualization , its licensing is very tricky and varies alot between os and packages!. Right now it is almost impossible to virtualize windows 10 due to VDA requirements.

Server Standard: 2 Windows Server VMs <--- this is wrong, MS only allow you to do this if your hypervisor is using HyperV

I am using proxmox ve 6.0 and default its cpu args already have the following -cpu 'host,+kvm_pv_unhalt,+kvm_pv_eoi,hv_vendor_id=proxmox,hv_spinlocks=0x1fff,hv_vapic,hv_time,hv_reset,hv_vpindex,hv_runtime,hv_relaxed,hv_synic,hv_stimer,hv_ipi,kvm=off' following is my vm conf agent: 1 bios...

I would suggest the following config for your qemu if you are using PVE 6.0 hostpci0: 04:00,pcie=1,x-vga=1 machine: pc-q35-3.1

no, I don't.... root@gpu01:/etc/pve/priv/ceph# ls -al total 0 drwx------ 2 root www-data 0 Aug 12 05:10 . drwx------ 2 root www-data 0 May 3 05:02 .. A temporarily work around this issue is to reboot pve host, then it will stop working again mounting additional rbd disk with sysfs write...

no... cephx is disabled

root@gpu01:~# rbd showmapped id pool namespace image snap device 0 rbd vm-101-disk-1 - /dev/rbd0 1 rbd vm-101-disk-2 - /dev/rbd1 2 rbd vm-101-disk-0 - /dev/rbd2 3 rbd vm-104-disk-1 - /dev/rbd3 4 rbd vm-104-disk-2 -...

Does not work ........ root@gpu05:~# rbd feature enable rbd/vm-104-disk-0 exclusive-lock root@gpu05:~# rbd feature enable rbd/vm-104-disk-1 exclusive-lock root@gpu05:~# rbd feature enable rbd/vm-104-disk-2 exclusive-lock /dev/rbd9 /dev/rbd10 /dev/rbd11 rbd: sysfs write failed can't unmap rbd...

yes

yes you can, but since you already have existing running vm setup already, when you adding the 2nd, 3rd nodes to the cluster, you must first make sure /etc/pve/qemu-server/ and /etc/pve/lxc/ has no config files (move all *.conf to /root/qemu-server and /etc/lxc/), then also make sure all your...