Proxmox server shows twice iSCSI

[alaintno]

Hello guys,

I've rencently installed two proxmox servers in cluster mode on an HP Blade System c3000. Beside this Blade, there is an HP p2000 filer with two iSCSI targets which are connected to the BladeSystem. On this filer I defined two Volumes (LUN0 and LUN1).

The installation went great, and I was able to connect two iSCSI storage definitions as told in the proxmox storage model documentation. I used the first LUN by creating a filesystem on it and define a directory storage on proxmox, on the second I have defined an LVM Group for the Images storage.

The problem is that I see twice the LUN0 when I do an fdisk -l, so I can't be sure where exactly is the partition I have created before.

Here is the output on the first proxmox server :

root@silas ~# pveversion 
pve-manager/3.2-4/e24a91c1 (running kernel: 2.6.32-31-pve)

root@silas ~# fdisk -l

WARNING: GPT (GUID Partition Table) detected on '/dev/sda'! The util fdisk doesn't support GPT. Use GNU Parted.


Disk /dev/sda: 300.0 GB, 299966445568 bytes
255 heads, 63 sectors/track, 36468 cylinders, total 585871964 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 262144 bytes / 262144 bytes
Disk identifier: 0x00000000

   Device Boot      Start         End      Blocks   Id  System
/dev/sda1               1   585871963   292935981+  ee  GPT
Partition 1 does not start on physical sector boundary.

Disk /dev/mapper/pve-root: 74.9 GB, 74893492224 bytes
255 heads, 63 sectors/track, 9105 cylinders, total 146276352 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 262144 bytes / 262144 bytes
Disk identifier: 0x00000000

Disk /dev/mapper/pve-root doesn't contain a valid partition table

Disk /dev/mapper/pve-swap: 37.4 GB, 37446746112 bytes
255 heads, 63 sectors/track, 4552 cylinders, total 73138176 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 262144 bytes / 262144 bytes
Disk identifier: 0x00000000

Disk /dev/mapper/pve-swap doesn't contain a valid partition table

Disk /dev/mapper/pve-data: 169.9 GB, 169911255040 bytes
255 heads, 63 sectors/track, 20657 cylinders, total 331857920 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 262144 bytes / 262144 bytes
Disk identifier: 0x00000000

Disk /dev/mapper/pve-data doesn't contain a valid partition table

Disk /dev/sdb: 3996.0 GB, 3995999993856 bytes
90 heads, 3 sectors/track, 28906249 cylinders, total 7804687488 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 1048576 bytes
Disk identifier: 0xe4534770

   Device Boot      Start         End      Blocks   Id  System
/dev/sdb1            2048  4294967294  2147482623+  8e  Linux LVM

Disk /dev/sdc: 3997.5 GB, 3997500375040 bytes
255 heads, 63 sectors/track, 486001 cylinders, total 7807617920 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 1048576 bytes
Disk identifier: 0x8f07cfb0

   Device Boot      Start         End      Blocks   Id  System
/dev/sdc1            2048  2147485695  1073741824   8e  Linux LVM
/dev/sdc3      2147485696  4294967294  1073740799+  83  Linux

Disk /dev/sdd: 3997.5 GB, 3997500375040 bytes
255 heads, 63 sectors/track, 486001 cylinders, total 7807617920 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 1048576 bytes
Disk identifier: 0x8f07cfb0

   Device Boot      Start         End      Blocks   Id  System
/dev/sdd1            2048  2147485695  1073741824   8e  Linux LVM
/dev/sdd3      2147485696  4294967294  1073740799+  83  Linux

Plus I see the LUN1 mount point on the first server (/dev/sdb).

My question is, how can I make it better, I mean, do you think guys that I should do only one big LUN shared on the two servers ? Should I make a filesystem that takes the whole space, as we want to store containers on it ? And why do I see twice the LUN0 (/dev/sdc and /dev/sdd) ?

Thanks in advance to all of you !

 

[spirit]

My question is, how can I make it better, I mean, do you think guys that I should do only one big LUN shared on the two servers ? Should I make a filesystem that takes the whole space, as we want to store containers on it ?

create a big lun, then use lvm on top to create vms disk.

And why do I see twice the LUN0 (/dev/sdc and /dev/sdd) ?

Do you use multipath ? if yes, you need to install multipath-tools
https://pve.proxmox.com/wiki/ISCSI_Multipath

 

[alaintno]

Thank you for your answer.

create a big lun, then use lvm on top to create vms disk.

That's a good solution, but using it this way we can't store containers on it ... Or am I missing something ?

I'll take a look at the multipath configuration.