VM Machine wont to start after upgrade to 6.1.20 => fixed in svn/6.1.x x>20

[Krzychull]

Hi, I'm running virtualbox on debian 10.9 linux host After upgrade to 6.1.20 no virtual machine wants to start and ends with message "The VM session was aborted" On syslog I found such message:

Apr 24 22:04:41 kailan kernel: SUPR0GipMap: fGetGipCpu=0x13 Apr 24 22:04:42 kailan kernel: vboxdrv: 0000000068e25721 VMMR0.r0 Apr 24 22:04:42 kailan kernel: vboxdrv: 00000000856fd8ba VBoxDDR0.r0 Apr 24 22:04:42 kailan kernel: EMT-0[28549]: segfault at 1 ip 00007fa8116e3b27 sp 00007fa83c619890 error 6 in VBoxDD.so[7fa8115ed000+160000] Apr 24 22:04:42 kailan kernel: Code: 40 f5 03 c6 40 f4 03 48 39 d0 75 c7 e9 5a fe ff ff 66 90 4d 8b 5c 24 28 41 be 07 39 00 00 ba ff ff ff ff 4d 8b bb b0 04 00 0 0 <41> c6 47 01 01 66 45 89 77 02 41 8b 8b b8 04 00 00 89 ce c1 ee 02

VirtualBox 6.1.20 Oracle VM VirtualBox Extension Pack was installed on host. VBox.log of the machine I am trying to run attached. After downgrade to virtualbox-6.1.18 everything works correctly.

[paulson]

The bad news is that you have run into a regression in the LsiLogic SAS controller code which unfortunately made it into the VirtualBox 6.1.20 release. The good news is that you can download the latest 6.1.x Testbuild of VirtualBox which already contains a fix for this issue.

[ThetaPhi]

Thanks for the feedback. I ran into the same issue with one of my Solaris VMs that use the LSI Logic controller. Same VBox.log.

I was able to work around by changing to SATA. Lucily, Solaris starts with a suddenly changed disk controller (like Linux). Windows would have been harder :-)

[paulson]

This has been addressed in VirtualBox 6.1.22 which was released yesterday.