home | tech | misc | code | bookmarks (broken) | contact | README

Xen troubleshooting

"xc_dom_compat_check: guest type xen-3.0-x86_32 not supported by xen kernel"

When trying to xm create a image, if you get this error:

Error: (2, 'Invalid kernel', 'xc_dom_compat_check: guest type xen-3.0-x86_32 not supported by xen kernel, sorry\n')

Check whether you should be using a PAE kernel if you are not. Check this page for details.

Problems with rtk0 network interface in Xen 2.0 and NetBSD 3.1

NetBSD is booted under Xen and all the periphericals are working correctly, but I got problems make the rtk0 network interface, although the vr0 interface (on-board) is fine. I could ifconfig it, but it didn't sent nor received any package and the kernel reported rtk0: watchdog timeout. Seeking e-mails in the netbsd mail-list, I found the answer in here (this is the most important message, but the whole thread is worth to see). I passed the ignorebiostables argument to the xen kernel (as in the menu.lst grub configuration above) and everything worked fine.

Vm doesn't start. Console doesn't enter.

After you setup your virtual machine, you want to start it with xm create. But if the console freezes (appears nothing) when you use command xm console, make a diagnostic.

The output of xm list, in my case, was:

# xm list
Name        ID   Mem VCPUs      State   Time(s)
Domain-0     0   256     1     r-----   1956.5
myvm         5   128     1     --p---      0.0

Although I have created "myvm" for some minutes, xm list tells me its uptime is 0.0.

Take a look at dmesg:

vnd0: unable to open device, error = 16
xbd backend: attach device vnd0d (size 31457280) for domain 5
xvif5i0: Ethernet address aa:00:00:51:02:f0

Problems when opening vnd0 device? Let's take a look at the xenbackendd log:

# cat /var/log/xenbackendd.log
brconfig: unable to get interface flags: Device not configured
Failed to read /local/domain/0/backend/console/5/0/script (No such file or directory)

Hmmmm, the virtual network interface is trying to attach to a bridge, but we just forgot to configure that! Simply make a simple test:

# ifconfig bridge0 create

Destroy the Vm and create it again and it will just work.

Passing parameters to DomU kernel

If you need to boot your DomU in single mode or pass another parameter to the kernel, use the extra parameter for xm create or in the configuration file, like this:

# xm create foo.py extra=' single'

See this page for more examples.

Error: (4, 'Out of memory', "panic: xc_dom_boot.c:159: xc_dom_boot_mem_init: can't allocate low memory for domain")

I got this problem because I was trying to boot a x86_64 version of CentOS in domU in a i386 NetBSD dom0 host. After changing the architecture of domU to i386, the problem disappeared.

Error: Device 0 (vif) could not be connected. Hotplug scripts not working

In my case, the only problem I had was wrong syntax in vif settings. The correct setting I use nowadays is:

vif = [ 'mac=02:00:00:00:00:01, bridge=bridge0' ]

Although this is a possible reason, lots of other problems can be the cause of this symptom. Check this link to see some problems about this in NetBSD.