Hyper-V migration startx failing

[samReading]

Centos Version: 7.5.1804
Kernel: 3.10.0-862.14.4.el7.x86_64

I have prepared a CentOS VM using virtual box on an internet connected machine and imported it onto a network without internet connectivity. Apologies that the log messages that I’m writing out are manually typed so they might not be as explicit I’d like. Please ask if you need any more info.

The new network uses Hyper-V on Windows Server 2012. When I initially booted the machine I received the Dracut prompt so I booted from the rescue disk and re-initialised the initramfs image for the current kernel.

Code: Select all

Dracut –f –v /boot/initramfs-3.10.0-862.14.4.el7.x86_64.img 3.10.0-862.14.4.el7.x86_64 --add-drivers "hv_vmbus hv_storvsc hv_netvsc hv_utils hv_balloon hyperv-keyboard hyperv_fb hid-hyperv"

as found on this forum 61699

After this the boot sequence got a lot further but seems to hang when initialising the UI.
I am able to press Ctrl-Alt F2 to get to a console log in and sign in as my user or as root. If I try to run startx the screen flicker and returns with information about xorg. Looking in the logs it appear that the machine can’t find a display.

Code: Select all

<EE> open /dev/dri/card0 no such file or directory
Falling back on old probe method
<EE> screen 0 deleted because of no matching config section

I’ve done a little googling, most of which seems to point towards the .Xauthority and .xinitrc files, both of which are empty under my user home.
I’m feeling a little out of my depth on this one, I’ve not used hyper-v before and I’ve not had these kind of issue when migrating an image before, granted that experience is fairly limited.

Apologies if the information is a little scarce, but I can’t copy the logs back off the other machine. If I can provide any more information that would be helpful, please just ask.

All the best,
Sam

[samReading]

bump - 24 hours later.

[hunter86_bg]

What happens when you run

Code: Select all

systemctl isolate graphical.target

Check the logs via

Code: Select all

journalctl -p 4

[samReading]

Thanks for your response.

When I run that command I get the grey screen with a wheel spinning. Journalctl outputs a couple of failing services, such as docker (which I've disabled now) and a "Failed to start LSB: bring up/down network", but i've not attached this VM to the network yet so that kind of makes sense too.

Following a few other forum posts I tried running X -configure and replacing my /etc/X11/xorg.conf file. I also tried to move the contents of xorg.conf.d to a backup folder and booting without it. All the results are the same.

X --configure does thrown an interesting error though. Apologies this is typed from accross networks so I've truncated the ram addresses/hashes.

Code: Select all

<EE> Backtrace
0: X (xorg_backtrace=0x55) <ram address>
1: <ram address> <ram address>
2. /lib64/libpthread.so.0 <ram address>
Segmentation fault at address 0x0
Fatal Server error
Caught signal 11 Segmentation fault. Server Aborting

[tunk]

A shot in the dark - did you install virtualbox guest additions?
I guess you could try to uninstall X (+related stuff), and then reinstall it.