[SOLVED] VM second disk initialization problem

ChristopherG

Member
Oct 26, 2018
8
0
21
46
Hi guys,

I'm a beginner and I have a hope that someone has encountered a similar problem and can share a solution or advice. I will be grateful for any advice because I'm stuck and I have no more ideas what I can do.

I have a VMDK file which I converted and imported (as a .raw) to Proxmox VE 5.2 without any bigger issues. This image/file includes a database server based on Ubuntu 14.4 which require additional disk for the database. For some reason, that VM is not able to initialize/mount attached as a second the DATA_DISK under the first run procedure and halt entire VM. I tried a few different things as a different setting for BUS/Device, storage pool, disk format etc. but VM still can't initialize/mount attached DATA_DISK or formatting it by mkfs. Because the system doesn't boot from SCSI I have to use IDE. It is an on-premise solution and I don't have a root access there. The recovery boot option doesn't help me too. The solution was designed for VMware ESXi and all manuals based on this virtualization platform.

Configuration file:
root@vms1:~# cat /etc/pve/local/qemu-server/111.conf
boot: cdn
bootdisk: ide0
cores: 2
ide0: SSD-pool_vm:vm-111-disk-0,size=20G
ide2: none,media=cdrom
memory: 2048
name: DB-InControl
net0: e1000=B6:07:C5:75:5E:58,bridge=vmbr3
net1: e1000=CA:28:70:80:8A:00,bridge=vmbr0
numa: 0
ostype: l26
scsi0: SSD-pool_vm:vm-111-disk-1,size=10G
scsihw: virtio-scsi-pci
smbios1: uuid=8802e53b-8cae-4743-bfc2-384997811c43
sockets: 2
vmgenid: 67ca8e37-ec5a-4476-bbd7-eea1b56e45fb


upload_2018-11-11_22-4-53.png

Thank you in advance.
Christopher
 
I have a VMDK file which I converted and imported (as a .raw) to Proxmox VE 5.2 without any bigger issues. This image/file includes a database server based on Ubuntu 14.4 which require additional disk for the database. For some reason, that VM is not able to initialize/mount attached as a second the DATA_DISK under the first run procedure and halt entire VM. I tried a few different things as a different setting for BUS/Device, storage pool, disk format etc. but VM still can't initialize/mount attached DATA_DISK or formatting it by mkfs. Because the system doesn't boot from SCSI I have to use IDE. It is an on-premise solution and I don't have a root access there. The recovery boot option doesn't help me too. The solution was designed for VMware ESXi and all manuals based on this virtualization platform.

Did it work with a DATA_DISK in ESXi environment? If yes it should be possible to continue using the DATA_DISK in Proxmox environment.
 
Hi Richard,

Thank you for response and idea. Probably is the last thing which I could do. Manual says about connecting both disks as an IDE0:0 and IDE0:1. For me it looks like IDE master and slave, maybe this is a problem. I can't find any information about Proxmox and configuration like that.

One more time thank you for support :)

Best Regards
Christopher
 
IDE0 = IDE0:0
IDE1 = IDE0:1
IDE2 = IDE1:0
IDE3 = IDE1:1
 
Hi Richard,

Thank you for advice!

Maybe you will know how I should modify the config file?

When I tried, any kind of modification in configuration file only result is that drives disappear from GUI, and the machine doesn't boot. I tried to modify lines which include ide0 and ide1 in the configuration file and when VM does not boot I modified the boot line too but without effect. I tied find any info about SeaBios and names for interfaces or how configuration like yours should look like, but without success. It looks like the config file does not accept any "space" in the config line excluding the first one after two dots. I tried multiple variations of line configuration including "IDE0 = IDE0:0" and "IDE1 = IDE0:1" but the effect is always the same, after that no drives in GUI and no boot device.

bios: seabios
boot: cdn
bootdisk: ide0
cores: 2
ide0: SSD-pool_vm:vm-111-disk-0,size=20G
ide1: HDD-pool_vm:vm-111-disk-1,size=500G
ide2: none,media=cdrom
memory: 2048
name: DB-InControl
net0: e1000=B6:07:C5:75:5E:58,bridge=vmbr3
net1: e1000=CA:28:70:80:8A:00,bridge=vmbr0
numa: 0
ostype: l26
scsihw: megasas
smbios1: uuid=8802e53b-8cae-4743-bfc2-384997811c43
sockets: 2
vmgenid: 67ca8e37-ec5a-4476-bbd7-eea1b56e45fb
 
bios: seabios
boot: cdn
bootdisk: ide0
cores: 2
ide0: SSD-pool_vm:vm-111-disk-0,size=20G
ide1: HDD-pool_vm:vm-111-disk-1,size=500G
ide2: none,media=cdrom
memory: 2048
name: DB-InControl
net0: e1000=B6:07:C5:75:5E:58,bridge=vmbr3
net1: e1000=CA:28:70:80:8A:00,bridge=vmbr0
numa: 0
ostype: l26
scsihw: megasas
smbios1: uuid=8802e53b-8cae-4743-bfc2-384997811c43
sockets: 2
vmgenid: 67ca8e37-ec5a-4476-bbd7-eea1b56e45fb


The config file looks perfect and I cannot imagine that
after that no drives in GUI and no boot device.
 
Hi Richard,

I apologize, English is my second language and is highly possible that I explain my problem, not enough clear.

I mean, that when I modify for example line ide0 according to the instructions you sent me.

Example modified config line:
IDE0 = IDE0:0 SSD-pool_vm:vm-111-disk-0,size=20G
or
IDE0=IDE0:0 SSD-pool_vm:vm-111-disk-0,size=20G

After modification like that disk disappears from GUI, disk still is there (mean assigned pool/ is visible in the dedicated pool content) but the "Hardware" tab dedicated to this VM shows nothing (no hdd) and VM no longer booting.
 
IDE0 = IDE0:0
IDE1 = IDE0:1
IDE2 = IDE1:0
IDE3 = IDE1:1

just means you have to write e.g.
Code:
ide1: HDD-pool_vm:vm-111-disk-1,size=500G

in the .conf file in order to specify slave-disk@ide0 (which is seen sometimes as "IDE0:1").

Sorry for the misunderstanding.
 
I just finish tests with VMware Workstation where this VM doesn't have any issues with booting and disks initialization (two disks as IDE0:0 and IDE0:1). After that, I export VM to .ova (main .ova file and two disks .vmdk), copy 3 files and import to Proxmox by:

qm importovf 112 DB-incontrol-VM.ovf SSD-pool_vm

Proxmox created new VM and imported disks without any issues (disks was imported as an ide0 and ide1). After that, when I start this VM, I meet the same issues as before. System boot and halt at the disk initialization/mount procedure.
 
Hi Richard,

Thank you for explanation. If I uderstand correctly interface name ide1: is hardcoded and equal ide0-slave or IDE0:1.

Do you think that if I put the trouble ticket as paid help request, will you guys be able to help me with solving this issue?
 
I want to say thank you Richard, I'm under great impression Proxmox FREE support here. Choosing Proxmox solution was a good decision (affofdable licence price and support like this made your solution really atractive) . Thank you guys!

Peplink wich provide inControl as an on-premise solution, suprised me today in very positive way. They provided new image of database VM which do not halt VM initialization process. I am suprised because Proxmox do not belong to officialy supported by them VM platforms.
I was almost sure that our correspondence meet the end like "We feel sorry but...". Now everything work fine.

One more time thank you!

Best Regards
Christopher
 

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!