Error while starting vm: Could not acquire pidfile: No such file or directory

[tincboy]

I had a vm on an old server which was working to one year, but today when I want to start it I got this error

qm start 763
Could not acquire pidfile: No such file or directory
start failed: command '/usr/bin/kvm -monitor unix:/var/run/qemu-server/763.mon,server,nowait -vnc unix:/var/run/qemu-server/763.vnc,password -pidfile /var/run/qemu-server/763.pid -daemonize -usbdevice tablet -name srv771 -smp sockets=1,cores=1 -nodefaults -boot menu=on,order=cd -vga cirrus -tdf -k en-us -drive file=/var/lib/vz/images/763/vm-763-disk-1.raw,if=ide,index=0,boot=on -drive file=/var/lib/vz/template/iso/ubuntu-10.04-server-i386.iso,if=ide,index=2,media=cdrom -m 1024 -net tap,vlan=0,ifname=vmtab763i0,script=/var/lib/qemu-server/bridge-vlan -net nic,vlan=0,model=rtl8139,macaddr=96:49:93:11:AC:F7' failed with exit code 1

I can see the pid file is OK and accessible, I've also changed the permission to 777 but it didn't help,
Any one familiar with this error?

 

[BiagioParuolo]

try before with qm unlock 763

 

[tincboy]

It doesn't help
Today I'm getting the same error on an other server,
The uptime of server is 222 days is it related?

 

[udo]

It doesn't help
Today I'm getting the same error on an other server,
The uptime of server is 222 days is it related?

Hi,
something wrong with your disk? But if this happens on more than one server it's should only happens when you have your systemdisk on a san...
Any messages about that in syslog?

How looks the output of "ls -lsa /var/run/qemu-server"?

The uptime should not be related - i have one server with an higher uptime (yes, of course i should update soon... )

Udo

 

[tincboy]

Thanks Udo,
No I'm not using any kind of shared storage,
Also there's nothing about I/O errors in syslog of that two servers,
the output of "ls -lsa /var/run/qemu-server" for the related vm is

0 srwxr-xr-x 1 root root 0 2011-08-21 20:05 1017.mon
4 -rw------- 1 root root 6 2011-08-19 23:37 1017.pid
0 srwxr-x--- 1 root root 0 2011-08-19 23:37 1017.vnc

While this error happens for specific vm, the other vms has no issue,
As this issue happened for to vms now, I'm worried if it occurs for more and more vms,

 

[udo]

Thanks Udo,
No I'm not using any kind of shared storage,
Also there's nothing about I/O errors in syslog of that two servers,
the output of "ls -lsa /var/run/qemu-server" for the related vm is

0 srwxr-xr-x 1 root root 0 2011-08-21 20:05 1017.mon
4 -rw------- 1 root root 6 2011-08-19 23:37 1017.pid
0 srwxr-x--- 1 root root 0 2011-08-19 23:37 1017.vnc

While this error happens for specific vm, the other vms has no issue,
As this issue happened for to vms now, I'm worried if it occurs for more and more vms,

Hi,
is there an process allready running?

fuser /var/run/qemu-server/763.pid

On any other node in the cluster? (Should be an different error-message)

Udo

Edit: curios is the different timestamp - on my system all files of one process have the same timestamp!

 

[tincboy]

fuser /var/run/qemu-server/1017.pid

Shows nothing,

It's strange that no one else experienced this issue

 

[tincboy]

Today I got an other strange error on that server,
While i was trying to reboot a vm I got time out,

trying to aquire lock...got timeout

Even when I want to make a backup of that vm I got similar error:

INFO: starting new backup job: vzdump 1077
INFO: Starting Backup of VM 1077 (qemu)
INFO: running
INFO: status = running
INFO: trying to aquire lock...setting parameters failed - got timeout
ERROR: Backup of VM 1077 failed - command 'qm set 1077 --lock backup' failed with exit code 4

What's wrong on this server? any suggestion?

 

[tom]

try unlock with:

qm unlock VMID

 

[tincboy]

try unlock with:

qm unlock VMID

it didn't help,
I've killed the process id and start the vm, it seems to be ok,

But the issue with other vm that can't be started still persists.