Good morning everybody.
I planned the following scenario and would like to know if anyone here has set up something similar or if something in the services has any limitations.
I will have two OmniOS servers, with ZFS and using COMSTAR for iscsi.
The two storage servers have several Intel x520-da2 cards in a type of bridge to guarantee only one IP address on all of them.
The proxmox cluster, made up of 4 to 8 servers, have their own x520-da2 cards (2 cards each for each different storage server) and are directly connected to the two storages (here I try to avoid switch equipment in the way for less chance of hardware failure) .
In the PROXMOX cluster, the two LUNS are configured as disks for use by ZFS in raid1.
Is this scenario possible in PROXMOX?
Has anyone here already set up using this technology or others following this scenario?
Somebody have some case using COMSTAR with PROXMOX ?
Goals:
High-availability storage facilitating maintenance that involves shutting down storage.
Lowest possible latency within the network.
Stability.
Best regards.
I planned the following scenario and would like to know if anyone here has set up something similar or if something in the services has any limitations.
I will have two OmniOS servers, with ZFS and using COMSTAR for iscsi.
The two storage servers have several Intel x520-da2 cards in a type of bridge to guarantee only one IP address on all of them.
The proxmox cluster, made up of 4 to 8 servers, have their own x520-da2 cards (2 cards each for each different storage server) and are directly connected to the two storages (here I try to avoid switch equipment in the way for less chance of hardware failure) .
In the PROXMOX cluster, the two LUNS are configured as disks for use by ZFS in raid1.
Is this scenario possible in PROXMOX?
Has anyone here already set up using this technology or others following this scenario?
Somebody have some case using COMSTAR with PROXMOX ?
Goals:
High-availability storage facilitating maintenance that involves shutting down storage.
Lowest possible latency within the network.
Stability.
Best regards.