All VMs get a GuruMeditation alert.

[almejo]

Hi.
I am using Ubuntu 20.04 without problems for month. But since today all my VMs end with a Guru medidation alert. I tried everything:

1- Creating new VM using the .vmdk images
2- removing and reinstalling (also cleaning my .config/virtualbox folder).
3- Using backups of my .vmdk

I am attaching the full log. I have read it but I can't point the real cause of the error.

My system: Ubuntu 20.04
My kernel: Linux steven 5.4.0-40-generic #44-Ubuntu SMP Tue Jun 23 00:01:04 UTC 2020 x86_64 x86_64 x86_64 GNU/Linux

Any help is great... my head is in pain right now

[mpack]

00:00:00.851816 VirtualBox VM 6.1.6_Ubuntu r137129 linux.amd64 (Apr 15 2020 15:48:44) release log

Please install the official version of VirtualBox - the Ubuntu fork (which you have) is not supported here.

Your error is:

00:00:01.087111 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
00:00:01.087112 !!
00:00:01.087112 !! VCPU0: Guru Meditation -79 (VERR_INVALID_STATE)
00:00:01.087115 !!
00:00:01.087118 !!

Which I suspect means that you had another VM platform running at the same time, which is not allowed.

[almejo]

Hi.,

Thanks for answering... Here is the new log from the official Virtual box package. I tried it yesterday with identical results

[almejo]

Also.. I have no other virtualization platform running. I have not installed qemu or vmware or anything. If you want me to check some package, please tell me

[mpack]

Well, now you have a new problem.

00:00:00.887953 Host RAM: 15919MB (15.5GB) total, 2392MB (2.3GB) available
...
00:00:00.995643 RamSize <integer> = 0x0000000200000000 (8 589 934 592, 8 192 MB, 8.0 GB)

At this time you didn't have 8GB available to give to the VM.

I can't give you Linux package advice - I don't use Linux.

What has changed since the VM last worked on this host? Because as far as I can tell, this VM was never properly installed, never mind working: the hard disk image is a VMDK that doesn't match the VM name, it looks like the compressed VMDK from an OVA appliance has been directly mounted in a VM rather than the appliance being installed using "File|Import...".

[almejo]

May be there were not 8GB of real ram available, but I have enough swap space. The VM is configured with 8192M.

Also the mismatch in names is because I deleted the VM and preserved the disk image (the disk image i copied from another machine, a Mac pro). I copied a windows virtual disk from my old mac. Then it stoped working for all my vms. I thought that the vmdk was corrupted, so I copied the vmdk again, the backup, and created the new VM. But that machine never worked. Same with a simple fedora image (vdi).

Also, I tried to create a new virtual machine, with a new disk, and install fedora. I can not start the instalation process. Here is a new try to install (this time ubuntu in a new machine called fedora sorry)

[scottgus1]

Disk swap space cannot be used for Virtualbox memory. The guest needs a contiguous block of open memory, though the memory apparently gets assigned lazily.

This last log shows 2GB available and 4GB assigned. Still too short. There may also be consequences for installing Ubuntu in a Fedora environment.

However, your log shows guru meditations as soon as the guest tries to start, while setting up, even before the ISO gets touched for booting.

Sounds like something fundamental wrong with your host PC, but what it is does not show plainly in the log. Has anything updated on the host PC?

[almejo]

Hi

I followed the advice about memory and restarted (again) my machine, but before opening something else, I opened virtualbox. This time it worked!!!

This is very strange for me. I've used VB for years and this is the first time I have had this problem. But now works... I want to thank you for all the time helping me.

Also, is you want to know, just before reboot I tried to used multipass (used by snapcraft in ubuntu), other VM solution and I didn't worked either... Now, like VB, multipass is working again

Thanks again