headers
Frank | 1 Sep 12:09
Picon
Picon

Error: Device 2049 (vbd) could not be connected. - again after reboot

It's very strange. I thought I had solved the problem with xen and the 
not starting domU yesterday but now I see the same error as yesterday 
again only after a reboot an I've changed nothing, only iptables.
I don't have an answer for that behaivior.

server1:/etc/xen# xm create -c xen-etch.cfg
Using config file "xen-etch.cfg".
Error: Device 2049 (vbd) could not be connected. Backend device not found.
h966380:/etc/xen# ifconfig
???
Permalink | Reply | 5 comments |
headers
Nico Kadel-Garcia | 1 Sep 12:29
Picon

Re: RE: Kernel sources for DomU

Mark Williamson wrote:
>> I urgently need the kernel-devel sources for my Centos DomU to recompile
>> asterisk. Please let me know where I can find this? My domU was created by
>> importing from another server... I'm running Xensource enterprise V4 latest
>> on the server.
>>
>> After instaling Xen tools, uname -r gives
>> 2.6.9-55.0.2.el.xs4.1...
>>     
>
> Is this a Xen tools for HVM guest thing, or...?
>
> The folks on this list are not so knowledgeable regarding the XenEnterprise 
> products; most people here use the OSS Xen.  More informed help regarding the 
> commercial products is probably available at http://forums.xensource.com.  
> You might want to try enquiring there in parallel.
>   
Did you:
    rename the kernel in Makefile or elsewhere before compiling it?
    install the new kernel
    verify that /etc/boot/grub.conf has the new kernel correctly configured
    boot with that new kernel?

And if you've paid for the enterprise version, you might ask the 
Xensource staff for some help. I don't know if they're building RPM's 
for RHEL and CentOS, but they should, and the RHEL kernel should work fine.

I assume that in fact you are using CentOS 4, based on your kernel 
version. Upgrade to CentOS 5 if you can, it's a straightforward upgrade 
and a useful one.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Nico Kadel-Garcia | 1 Sep 12:32
Picon

Re: Error: ... Backend device not found.

David Morris wrote:
> Couple additional suggestions 
> a. From your grub data, it appears you are using software raid on the dom0 ... might be that
>    your configured devices need to be mdX devices to match your kernel, etc.
> b. you may need a 'root=/dev/hda2 rw' or some such in your config if it isn't there
> c. it appears from the config fragment that this is a paravirt installation?
>
> This sounds like a problem I encountered when I tried to use virt-manager to create a paravirt RHEL5 domU on
RHEL5 dom0. I've fought so many alligators over the past week that I'm not sure I found a solution or a way
around it ....
>
> I think I concluded that there was a fundamental disconnect between the virt-manager PV setup and reality
... what I did that worked:
>   a. with virt-manager, create an HVM guest (it worked)
>   b. upgrade to Xen 3.1 
>   c. follow the steps in my post of last night to clone the HVM guest as a para virt guest
>
> btw ... you can check the number of loop devices in use with the 'losetup -a' command. When I had my
difficulties, I was using lvm devices so loop devices was not the issue.
>   
virt-manager of RHEL 5 is already sadly out of date. If you grab and 
recompile the one from Fedora 7, it seems to be a good update and more 
usable. Or use virt-install, that has better logging and some options 
not in virt-manager. Then use virt-manager if necessary to access the 
console.
Permalink | Reply | 0 comments |
headers
Frank | 1 Sep 12:37
Picon
Picon

Re: Error: Device 2049 (vbd) could not be connected. - again after reboot

I hope I got it now. It seems that it needs somthing like "loop 
max_loop=64" to /etc/modules to load the loop module to the kernel.

What does "max_loop" mean?

Frank

Frank schrieb:
> It's very strange. I thought I had solved the problem with xen and the 
> not starting domU yesterday but now I see the same error as yesterday 
> again only after a reboot an I've changed nothing, only iptables.
> I don't have an answer for that behaivior.
> 
> server1:/etc/xen# xm create -c xen-etch.cfg
> Using config file "xen-etch.cfg".
> Error: Device 2049 (vbd) could not be connected. Backend device not found.
> h966380:/etc/xen# ifconfig
> ???
> 
> 
> _______________________________________________
> Xen-users mailing list
> Xen-users <at> lists.xensource.com
> http://lists.xensource.com/xen-users
>
Permalink | Reply | 4 comments |
headers
Nico Kadel-Garcia | 1 Sep 12:34
Picon

Re: xen 3.0.3-1 still tls.disabled?

Mark Williamson wrote:
>> Does xen 3.0.3-1 still need "mv /lib/tls /lib/tls.disabled"?
>>     
>
> Yes it does, unless:
> * your /lib/tls is Xen-friendly
> * you're running on a 64-bit host (regardless of guest type)
> * HVM guests don't need this to be done
> * You're running on a non-x86 platform.
>   
RHEL 4 and RHEL 5 now support a bit of "drop a file in /etc/ld.so.conf.d 
to disable /lib/tls", instead of moving it aside. Moving it aside is 
fulnerable to update problems, where updating a program with libraries 
there restores the directory and makes rolling back the Xen installation 
rather awkward.

Also, I've seen reports that it's not just glibc: db4 drops some 
libraries in /lib/tls that are problematic for Xen guests.
Permalink | Reply | 0 comments |
headers
Christian Horn | 1 Sep 12:41
Favicon

Re: Error: Device 2049 (vbd) could not be connected. - again after reboot

On Sat, Sep 01, 2007 at 12:37:57PM +0200, Frank wrote:
> I hope I got it now. It seems that it needs somthing like "loop 
> max_loop=64" to /etc/modules to load the loop module to the kernel.
> 
> What does "max_loop" mean?

According to linux-kernelsource/drivers/block/loop.c its the
number of loop-devices that can be used with the driver.
A default of 8 is compiled in.

Christian
Permalink | Reply | 0 comments |
headers
Frank | 1 Sep 13:01
Picon
Picon

Re: Error: Device 2049 (vbd) could not be connected. - again after reboot

But why than there are problems with the loop devices with xen if 8 is 
compiled in. I only unsed 2 img files as loop device and got this error
"Error: Device 2049 (vbd) could not be connected. Backend device not
found."
After I added "loop max_loop=64" the error was gone.

 > According to linux-kernelsource/drivers/block/loop.c its the
 > number of loop-devices that can be used with the driver.
 > A default of 8 is compiled in.
 >
 >
 > Christian
Frank schrieb:
> I hope I got it now. It seems that it needs somthing like "loop 
> max_loop=64" to /etc/modules to load the loop module to the kernel.
> 
> What does "max_loop" mean?
> 
> Frank
> 
> Frank schrieb:
>> It's very strange. I thought I had solved the problem with xen and the 
>> not starting domU yesterday but now I see the same error as yesterday 
>> again only after a reboot an I've changed nothing, only iptables.
>> I don't have an answer for that behaivior.
>>
>> server1:/etc/xen# xm create -c xen-etch.cfg
>> Using config file "xen-etch.cfg".
>> Error: Device 2049 (vbd) could not be connected. Backend device not 
>> found.
(Continue reading)

Permalink | Reply | 2 comments |
headers
Frank | 1 Sep 14:18
Picon
Picon

Re: Error: Device 2049 (vbd) could not be connected. - again after reboot

How can I show all used loop devices on debian?

Frank schrieb:
> But why than there are problems with the loop devices with xen if 8 is 
> compiled in. I only unsed 2 img files as loop device and got this error
> "Error: Device 2049 (vbd) could not be connected. Backend device not
> found."
> After I added "loop max_loop=64" the error was gone.
> 
>  > According to linux-kernelsource/drivers/block/loop.c its the
>  > number of loop-devices that can be used with the driver.
>  > A default of 8 is compiled in.
>  >
>  >
>  > Christian
> Frank schrieb:
>> I hope I got it now. It seems that it needs somthing like "loop 
>> max_loop=64" to /etc/modules to load the loop module to the kernel.
>>
>> What does "max_loop" mean?
>>
>> Frank
>>
>> Frank schrieb:
>>> It's very strange. I thought I had solved the problem with xen and 
>>> the not starting domU yesterday but now I see the same error as 
>>> yesterday again only after a reboot an I've changed nothing, only 
>>> iptables.
>>> I don't have an answer for that behaivior.
>>>
(Continue reading)

Permalink | Reply | 0 comments |
headers
Nico Kadel-Garcia | 1 Sep 14:15
Picon

Re: Error: Device 2049 (vbd) could not be connected. - again after reboot

Frank wrote:
> But why than there are problems with the loop devices with xen if 8 is 
> compiled in. I only unsed 2 img files as loop device and got this error
> "Error: Device 2049 (vbd) could not be connected. Backend device not
> found."
> After I added "loop max_loop=64" the error was gone.
Stopping and restarting domains can leave loop connections dangling in 
fascinating ways: and if you're using file-based DomU's, *each disk or 
partition in the devices for the Xen config file* is an additional 
loopback point.
Permalink | Reply | 0 comments |
headers
Frank | 1 Sep 15:49
Picon
Picon

Error inserting fan, Error inserting thermal

How can I get rid off this messages during a domU boot?

FATAL: Error inserting fan 
(/lib/modules/2.6.18-5-xen-686/kernel/drivers/acpi/fan.ko): No such device
FATAL: Error inserting thermal 
(/lib/modules/2.6.18-5-xen-686/kernel/drivers/acpi/thermal.ko): No such 
device
Permalink | Reply | 6 comments |

Gmane