Search results

ZFS Raid1 is fine For the 6TB HHD's, use at least Raid6 (in ZFS: RaidZ2), as the very long rebuild (resilver) times of such disks will get you into risk of data loss with just one drive redundancy. As you have only 4 drives, use Mirror instead, that's faster and you loose capacity of two drives...

?? I use bulk migration with live migration and it works well

did you really add the dummy disk as virtio-scsi? with some of the ISO's for virtio-win some time back I had the problem that the device-manager did not find the correct driver and installed an incorrect one. Then you had to browse to the correct driver and os Version to get the correct version

you have to install the driver (as shown in my message) _before_ you move to virtio-scsi Please read here: https://pve.proxmox.com/wiki/Windows_VirtIO_Drivers

You seem to use the virtio-blk driver (which does not support trim and may get deprecated) Do the following: -> add a additional dummy disk with virtio-scsi driver -> install the virtio-scsi driver in the WIndows VM via device manager -> remove the dummy disk again then: shutdown the vm...

discard is possible with virtio-scsi, but only with 2012 / Vista upwards Windows 7 and Windows 2008R2 (and older ones of course) don't support the scsi unmap command. (they support s-ata trim, but that is not supported in sata and virtio-blk driver) with Windows server 2012 the optimization is...

Yes I use it in production I switched a ESXi Cluster over to proxmox, and are on the way to switch my other CentOS pacemaker based cluster's to proxmox (the centos based cluster mainly because of the nice Proxmox GUI) There are many guides out there for hardening linux based systems...

3 is higher priority than 1 or 2 , so the machines will stay on n3 Create 3 groups with different priority sequence (1 group per node): pref_n1 n1:3,n2:2,n3:2 pref_n2 n2:3,n1:2,n3:2 pref_n3 n3:3,n1:2,n2:2 the identical priority for other machines is intentional. Reason is the following...

do you eventually have priorities in the ha groups and "no-fallback" not set? Then VM's will migrate and the immediately fall back to the node with highest priority

I do not know the DL160, but this sounds like one of those S-ATA Raid Controllers which are fakes. It is nothing than a software raid with some bios intervention. Probably Adaptec or LSI fake on Intel S-ATA controller? Those are not very well supported in linux. Try to disable this crap in the...

yes

yep BBU and write cache module are useless for IT mode

yes that could work, or you use this file: /etc/pve/pve-root-ca.pem this should be the ca for the self signed cert. or use lets encrypt: https://pve.proxmox.com/wiki/HTTPS_Certificate_Configuration_(Version_4.x_and_newer)

the viewer try's to check the CA Certificate which of course does not exist with an self-signed certificate Please read this: https://people.freedesktop.org/~teuf/spice-doc/html/ch02s08.html

performance will be lower with Z3 and just one vdev, so I personally would prefer the 2 x Z2 vdevs combined (striping of vdevs not mirroring!)

you can setup all drives as RAID0, but that both an additional layer, and also will probably give you some hassle when you have to exchange disks, but ZFS wil work on it. Better is of course to acquire a SAS HBA , but they have to have the IT Mode ! there are some cheap Dell H310 out there...

The ServeRAID are not optimal for ZFS, as you cannot set them to HBA Mode (which is preferred for ZFS), but normally it should work with a single drive in a Virtual Drive configured as Raid 0 Too bad that the M5014 cannot be reflashed to IT Mode (HBA) I run some systems with similar...

the bonding has to match on both sides, so if you set LACP on proxmox side, switch must be same, also look for passive and active mode