vm startup error with 2 scsi disk on local storage

[shirori]

start failed exit code 1(500)

there is no problem with 2 ide disk

 

[tom]

please post more details about your problem, otherwise no one will be able to give a hint.

 

[bromont]

start failed exit code 1(500)

there is no problem with 2 ide disk

I'm also getting this... if I edit out the second scsi drive in the conf file the vm starts but doesn't boot completely like it does in 1.9 and previous. These are vmdk disks.


Marc.

 

[bromont]

Here's the error output:

start failed: command '/usr/bin/kvm -id 111 -chardev socket,id=monitor,path=/var/run/qemu-server/111.mon,server,nowait -mon chardev=monitor,mode=readline -vnc unix:/var/run/qemu-server/111.vnc,x509,password -pidfile /var/run/qemu-server/111.pid -daemonize -readconfig /usr/share/qemu-server/pve-usb.cfg -device usb-tablet,bus=ehci.0,port=6 -name Fortimail -smp sockets=1,cores=2 -cpu pentium -nodefaults -boot menu=on,order=c -vga cirrus -tdf -k en-us -device lsi,id=scsi0,bus=pci.0,addr=0x5 -drive file=/var/lib/vz/images/111/fortimail-vm-disk2-250gb.vmdk,if=none,id=drive-scsi1,aio=native -device scsi-disk,bus=scsi0.0,scsi-id=1,drive=drive-scsi1,id=device-scsi1 -device lsi,id=scsi0,bus=pci.0,addr=0x5 -drive file=/var/lib/vz/images/111/fortimail-vm-disk1.vmdk,if=none,id=drive-scsi0,aio=native,boot=on -device scsi-disk,bus=scsi0.0,scsi-id=0,drive=drive-scsi0,id=device-scsi0 -m 784 -netdev type=tap,id=net0,ifname=tap111i0,script=/var/lib/qemu-server/pve-bridge -device e1000,romfile=,mac=0ED:17:0C:23:84,netdev=net0,bus=pci.0,addr=0x12 -netdev type=tap,id=net1,ifname=tap111i1,script=/var/lib/qemu-server/pve-bridge -device e1000,romfile=,mac=5A:45:46:17:69:1D,netdev=net1,bus=pci.0,addr=0x13 -netdev type=tap,id=net2,ifname=tap111i2,script=/var/lib/qemu-server/pve-bridge -device e1000,romfile=,mac=A2:CF:00:43:65:43,netdev=net2,bus=pci.0,addr=0x14 -netdev type=tap,id=net3,ifname=tap111i3,script=/var/lib/qemu-server/pve-bridge -device e1000,romfile=,mac=C6:06:41:33:79:B3,netdev=net3,bus=pci.0,addr=0x15 -cpuunits 1000' failed: exit code 1 (500)

If the scsi1 is commented out it will boot.

 

[tom]

btw, why don´t you use virtio?

 

[tom]

and why do you run Fortimail instead of Proxmox Mail Gateway?

 

[bromont]

I'm trying to run all the Fortinet VM machines eventually for support purposes.. they all run great on Virtualbox/VMWare... getting them working on Proxmox is a challenge.

Marc.

 

[dietmar]

Please can you post the VM configuration file?

 

[bromont]

Here's the config file:


ostype: l26
memory: 784
sockets: 1
name: Fortimail
net0: e1000=0ED:17:0C:23:84,bridge=vmbr0
bootdisk: scsi0
scsi0: local:111/fortimail-vm-disk1.vmdk
scsi1: local:111/fortimail-vm-disk2-250gb.vmdk
boot: c
freeze: 0
acpi: 1
kvm: 1
onboot: 0
cores: 2
net1: e1000=5A:45:46:17:69:1D,bridge=vmbr0
net2: e1000=A2:CF:00:43:65:43,bridge=vmbr0
net3: e1000=C6:06:41:33:79:B3,bridge=vmbr0


when starting from CLI (qm start 111) the following error is also displayed:
kvm: -device lsi,id=scsi0,bus=pci.0,addr=0x5: Duplicate ID 'scsi0' for device

 

[tom]

no, this config file does not produce the posted command line from above.

please configure your vm via gui, do not post wrong and not related configs.

 

[bromont]

Tom,

You're right some changes were made since... sorry... However I've narrowed the issue down to the use of scsi0 and scsi1, as soon as there are 2 scsi drives it complains about "Duplicate ID 'scsi0'"

At this point it doesn't really matter as none of the Fortinet VMs will run in Proxmox, I can get the VM to boot but the OS doesn't seem to recognize the second harddisk (yes I've tried virtio). Will there be support for importing OVF format soon?

 

[tom]

Tom,

You're right some changes were made since... sorry... However I've narrowed the issue down to the use of scsi0 and scsi1, as soon as there are 2 scsi drives it complains about "Duplicate ID 'scsi0'"

no, this is only if you specify manually two scsi drives with the same ID - the error message is clear to me. and yes, it works here.

just do the following: create a new vm via gui (using scsi), add a second scsi drive. you will see that it boots up.

At this point it doesn't really matter as none of the Fortinet VMs will run in Proxmox, I can get the VM to boot but the OS doesn't seem to recognize the second harddisk (yes I've tried virtio). Will there be support for importing OVF format soon?

best compatible way is IDE. I do not know fortinet but they should integrate virtio in their kernel. are they using a linux kernel? if yes, it won´t be that hard ...

OVF does not help here.

 

[bromont]

The problem is that I'm not giving scsi1 and scsi0 the same id... and if I add it via GUI or even create a new drive I get the same error... (but only for 2 scsi drives) this seems like a bug. Anyway if I use virtio the machine will start without error so I'll try that from now on. Only other thing unrelated to this thread is getting the second drive recognized during boot... the boot disk may be looking for a specific volume ID on the second drive and thus doesn't complete the boot up. Thanks for your input.

Marc

 

[tom]

post the output of 'pveversion -v', look like you are using an outdated version.?

 

[tom]

ok, seems you are testing the beta, I am working on the stable 1.9 - that explains the difference - I got the same error.

 

[martin]

I filed a bug, see https://bugzilla.proxmox.com/show_bug.cgi?id=21

 

[spirit]

Hi,
It's me that add make rework on drive parts,
seem to have a bug as it add twice the scsi controller

-device lsi,id=scsi0,bus=pci.0,addr=0x5

-device lsi,id=scsi0,bus=pci.0,addr=0x5


I'll check the code today, thanks to report