Re: [Xen-users] Hotplug scripts not working / problem with loopback driver

[Kyrre M Begnum <kyrre@xxxxxxxxx>]

Craig,

i have a similar experience using Xen 3.0.1 and a pristine Ubuntu 5.04.
First i need to modprobe loop in order to get anywhere.

I simply tried to boot/shutdown the same VM over and over again. For  
each boot/shutdown i had one more process [loop?] lying around. After  
some iterations i had the following processes:
root      6190  0.0  0.0      0     0 ?        S<   15:41   0:00 [loop0]
root      6479  0.0  0.0      0     0 ?        S<   15:42   0:00 [loop1]
root      6628  0.0  0.0      0     0 ?        S<   15:42   0:00 [loop2]
root      6998  0.0  0.0      0     0 ?        S<   15:44   0:00 [loop3]
root      7090  0.0  0.0      0     0 ?        S<   15:44   0:00 [loop4]
root      7490  0.0  0.0      0     0 ?        S<   15:45   0:00 [loop5]
root      7609  0.1  0.0      0     0 ?        S<   15:45   0:00 [loop6]
root      8050  0.0  0.0      0     0 ?        S<   15:46   0:00 [loop7]

And from here on i get the following error:

Using config file "/opt/mln/projects/root/ugo/ubuntu_xen.cfg".
Error: Device 770 (vbd) could not be connected. Backend device not  
found.
I cannot "rmmod loop" either because it says it is in use. I can  
reboot and do the same thing over again. Is there a problem releasing  
the loopbacks? Do you see the same accumulation of processes?
Regards

On Feb 5, 2006, at 12:39 PM, Craig Webster wrote:

> Hi list,
>
> I'm having a bit of an annoying problem which, being new to Xen and  
> the loopback device, I have no idea how to fix.
> Everything was running along fine with 4 VMs, I attempted to add a  
> 5th VM and it couldn't connect to the backend devices. A quick  
> Google suggested that increasing the max_loop parameter for the  
> kernel (loopback driver is compiled in, not a module) would fix  
> this so I added that param to my grub.conf and rebooted.
> When I tried to start any of the previously working VMs I now get  
> this happening:
>   saturn vm # xm create subversion.cfg -c
>   Using config file "subversion.cfg".
>   Error: Device 769 (vbd) could not be connected. Hotplug scripts  
> not working.
>   saturn vm # xm destroy subversion
>   saturn vm # xm create subversion.cfg -c
>   Using config file "subversion.cfg".
>   Error: Device 770 (vbd) could not be connected. Backend device  
> not found.
>   saturn vm # xm destroy subversion
>   saturn vm # xm create subversion.cfg -c
>   Using config file "subversion.cfg".
>   Error: Device 770 (vbd) could not be connected. Backend device  
> not found.
> I have since tried removing the max_loop param from my grub.conf  
> and rebooting but the same problem keeps coming up.
> Google suggested removing the memory limit on dom0 but that didn't  
> make any difference; I still got the same error.
> Looking at the logs it appears that there's something wrong with my  
> loopbacks or block hotplug script, but I don't know enough to know  
> what to Google for next.
> Commenting out the disk parameter in my vm config file allows the  
> boot process to get much further (until it tries to mount the disks).
> These are the log entries:
>   saturn vm # tail /var/log/xen-hotplug.log
>   mkdir: cannot create directory `/var/run/xen-hotplug/block': File  
> exists
>   mkdir: cannot create directory `/var/run/xen-hotplug/block': File  
> exists
>   [... repeated lots ...]
>   mkdir: cannot create directory `/var/run/xen-hotplug/block': File  
> exists
>   ioctl: LOOP_SET_FD: Device or resource busy
>
>   saturn vm # tail /var/log/xend.log
>       return self.dom.waitForDevices()
>     File "/usr/lib64/python2.4/site-packages/xen/xend/ 
> XendDomainInfo.py", line 1343, in waitForDevices
>       self.waitForDevices_(c)
>     File "/usr/lib64/python2.4/site-packages/xen/xend/ 
> XendDomainInfo.py", line 971, in waitForDevices_
>       return self.getDeviceController(deviceClass).waitForDevices()
>     File "/usr/lib64/python2.4/site-packages/xen/xend/server/ 
> DevController.py", line 135, in waitForDevices
>       return map(self.waitForDevice, self.deviceIDs())
>     File "/usr/lib64/python2.4/site-packages/xen/xend/server/ 
> DevController.py", line 151, in waitForDevice
>       raise VmError("Device %s (%s) could not be connected. "
>   VmError: Device 770 (vbd) could not be connected. Backend device  
> not found.
> My subversion.cfg looks like this:
>
>   saturn vm # cat subversion.cfg
>   kernel = "/var/vm/vmlinuz-2.6-xenU"
>   memory = 64
>   ip = "aaa.bbb.ccc.ddd" # This is a public IP in the cfg file
>   netmask = "255.255.255.192"
>   gateway = "aaa.bbb.ccc.ddd" # As is this
>   vif = ['bridge=xenbr0']
>   name = "subversion"
>   disk = ['file:/var/vm/subversion-hd.img,sda1,w','file:/var/vm/ 
> subversion-swap.img,sda2,w']
>   root = "/dev/sda1 ro"
>
> If you have any suggestions which could help me get these VMs  
> running again they would be much appreciated.
> Cheers,
> Craig
>
> ps apologies for the length of the post -- wanted to include as  
> much information as possible.
> --
> Craig Webster | t: +44 (0)131 516 8595 | e: craig@xxxxxxxxxx
> Xeriom.NET    | f: +44 (0)131 661 0689 | w: http://xeriom.net
>
>
>
> _______________________________________________
> Xen-users mailing list
> Xen-users@xxxxxxxxxxxxxxxxxxx
> http://lists.xensource.com/xen-users

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users