Discussion:
[OpenIndiana-discuss] VirtualBox 5.2.14 crashing when attempting to start VM
russell
2018-07-24 18:15:30 UTC
Permalink
Hi,

After upgrading to the latest hipster on the 22nd July I did not attempt
to start any VMs until this evening.
After starting VirtualBox and then selecting a Windows 7 VM to start
everything froze and then my computer rebooted.
Thinking it might be a problem that had been fixed, I performed an
update today; however when attempted to start the same Windows 7 VM it
also caused everything to lock and the computer rebooted

oihipster201807-02 N      /          199M  static 2018-07-12 19:21   no
problem starting VM
oihipster201807-03 -      -          20.8M static 2018-07-22 10:56    <-
VirtualBox VM reboot
oihipster201807-04 R      -          266G  static 2018-07-24 17:55   <-
VirtualBox VM reboot

Has anyone else seen this problem?

Regards

Russell
Geoff Nordli
2018-07-25 00:20:24 UTC
Permalink
What is the last successful version of Vbox you have used on OI.

I use VBox 5.0.x on OmniOS.

The last time I tried anything newer there were issues.

Geoff
Post by russell
Hi,
After upgrading to the latest hipster on the 22nd July I did not
attempt to start any VMs until this evening.
After starting VirtualBox and then selecting a Windows 7 VM to start
everything froze and then my computer rebooted.
Thinking it might be a problem that had been fixed, I performed an
update today; however when attempted to start the same Windows 7 VM it
also caused everything to lock and the computer rebooted
oihipster201807-02 N      /          199M  static 2018-07-12 19:21  
no problem starting VM
oihipster201807-03 -      -          20.8M static 2018-07-22 10:56   
<- VirtualBox VM reboot
oihipster201807-04 R      -          266G  static 2018-07-24 17:55  
<- VirtualBox VM reboot
Has anyone else seen this problem?
Regards
Russell
_______________________________________________
openindiana-discuss mailing list
https://openindiana.org/mailman/listinfo/openindiana-discuss
russell
2018-07-25 14:46:37 UTC
Permalink
Hi,

I have been VirtualBox 5.2.14 since it came out without any issues until
I performed a pkg update on 22nd July.
Everytime I perform a pkg update I always specify a new boot
environment, so last update which VirtualBox 5.2.14 worked on is the one
I performed on the 12th July. I have made this BE activate so I do not
have to remember to select it on boot.

Regards

Russell
Michal Nowak
2018-07-25 20:02:54 UTC
Permalink
Post by russell
Hi,
I have been VirtualBox 5.2.14 since it came out without any issues until
I performed a pkg update on 22nd July.
Everytime I perform a pkg update I always specify a new boot
environment, so last update which VirtualBox 5.2.14 worked on is the one
I performed on the 12th July. I have made this BE activate so I do not
have to remember to select it on boot.
Regards
Russell
I think it might be the "eager FPU" changes which landed in illumos-gate
on June 13th:

https://github.com/illumos/illumos-gate/commit/4c28a617e3922d92a58e813a5b955eb526b9c386
https://github.com/illumos/illumos-gate/commit/a32a1f376ed9360264e4a374608fdcc5c4927d63

It's reported to the authors of those patches at
https://github.com/joyent/smartos-live/issues/792.

I don't know of a workaround apart from sticking to versions before the
change.

Michal
Predrag Zečević - Technical Support Analyst
2018-08-06 10:36:48 UTC
Permalink
Post by Michal Nowak
Post by russell
Hi,
I have been VirtualBox 5.2.14 since it came out without any issues
until I performed a pkg update on 22nd July.
Everytime I perform a pkg update I always specify a new boot
environment, so last update which VirtualBox 5.2.14 worked on is the
one I performed on the 12th July. I have made this BE activate so I do
not have to remember to select it on boot.
Regards
Russell
I think it might be the "eager FPU" changes which landed in illumos-gate
https://github.com/illumos/illumos-gate/commit/4c28a617e3922d92a58e813a5b955eb526b9c386
https://github.com/illumos/illumos-gate/commit/a32a1f376ed9360264e4a374608fdcc5c4927d63
It's reported to the authors of those patches at
https://github.com/joyent/smartos-live/issues/792.
I don't know of a workaround apart from sticking to versions before the
change.
Michal
Hi all,

we have had similar case in the past, and workaround was found:
* https://www.illumos.org/issues/6919#note-7
* that was added to wiki too:
https://wiki.openindiana.org/oi/7.2+VirtualBox#id-72VirtualBox-Virtualizationfeatures

Not sure what to do with this new case (should be illumos issue opened
or not)?

In any case, I have updated /hipster today and it has crashed. Output
excerpt from mdb -k (::msgbuf) command:
---<snip>---
pseudo-device: vboxusbmon0
vboxusbmon0 is /pseudo/***@0

panic[cpu0]/thread=ffffff03d96e0520:
recursive rw_enter, lp=ffffff03e71be928 wwwh=ffffff03d96e0524
thread=ffffff03d96e0520


ffffff000fcb5440 unix:rw_panic+54 ()
ffffff000fcb54b0 unix:rw_enter_sleep+37a ()
ffffff000fcb55a0 genunix:as_fault+1f1 ()
ffffff000fcb5630 unix:pagefault+96 ()
ffffff000fcb5730 unix:trap+d88 ()
ffffff000fcb5740 unix:cmntrap+e6 ()
ffffff000fcb5880 vboxdrv:rtR0SegVBoxSolCreate+91 ()
ffffff000fcb5960 genunix:as_map_locked+1a4 ()
ffffff000fcb59d0 genunix:as_map+58 ()
ffffff000fcb5a70 vboxdrv:rtR0MemObjNativeMapUser+28f ()
ffffff000fcb5ae0 vboxdrv:RTR0MemObjMapUserTag+154 ()
ffffff000fcb5b80 vboxdrv:SUPR0GipMap+138 ()
ffffff000fcb5c10 vboxdrv:supdrvIOCtl+2e7c ()
ffffff000fcb5cc0 vboxdrv:VBoxDrvSolarisIOCtl+375 ()
ffffff000fcb5d00 genunix:cdev_ioctl+39 ()
ffffff000fcb5d50 specfs:spec_ioctl+60 ()
ffffff000fcb5de0 genunix:fop_ioctl+55 ()
ffffff000fcb5f00 genunix:ioctl+9b ()
ffffff000fcb5f10 unix:brand_sys_syscall+1f2 ()

dumping to /dev/zvol/dsk/rpool/dump, offset 65536, content: kernel
---</snip>---

Booting to older BE (with latest VBox package - 5.2.16) worked for me too.

Regards.
Post by Michal Nowak
_______________________________________________
openindiana-discuss mailing list
https://openindiana.org/mailman/listinfo/openindiana-discuss
--
Predrag Zečević
Technical Support Analyst
2e Systems GmbH

tel: +49 - 6196 - 95058 - 15
mob: +49 - 174 - 3109288
fax: +49 - 6196 - 95058 - 94
e-mail: ***@2e-systems.com

headquarter: 2e Systems GmbH, Koenigsteiner Str. 87, 65812 Bad Soden am
Taunus, Germany
registration: Amtsgericht Koenigstein (Germany), HRB 7303
managing director: Phil Douglas

http://www.2e-systems.com/ - Making your business fly!
Predrag Zečević - Technical Support Analyst
2018-08-14 07:45:04 UTC
Permalink
Post by Predrag Zečević - Technical Support Analyst
Post by Michal Nowak
Post by russell
Hi,
I have been VirtualBox 5.2.14 since it came out without any issues
until I performed a pkg update on 22nd July.
Everytime I perform a pkg update I always specify a new boot
environment, so last update which VirtualBox 5.2.14 worked on is the
one I performed on the 12th July. I have made this BE activate so I
do not have to remember to select it on boot.
Regards
Russell
I think it might be the "eager FPU" changes which landed in
https://github.com/illumos/illumos-gate/commit/4c28a617e3922d92a58e813a5b955eb526b9c386
https://github.com/illumos/illumos-gate/commit/a32a1f376ed9360264e4a374608fdcc5c4927d63
It's reported to the authors of those patches at
https://github.com/joyent/smartos-live/issues/792.
I don't know of a workaround apart from sticking to versions before
the change.
Michal
Hi all,
* https://www.illumos.org/issues/6919#note-7
https://wiki.openindiana.org/oi/7.2+VirtualBox#id-72VirtualBox-Virtualizationfeatures
Not sure what to do with this new case (should be illumos issue opened
or not)?
In any case, I have updated /hipster today and it has crashed. Output
---<snip>---
pseudo-device: vboxusbmon0
recursive rw_enter, lp=ffffff03e71be928 wwwh=ffffff03d96e0524
thread=ffffff03d96e0520
ffffff000fcb5440 unix:rw_panic+54 ()
ffffff000fcb54b0 unix:rw_enter_sleep+37a ()
ffffff000fcb55a0 genunix:as_fault+1f1 ()
ffffff000fcb5630 unix:pagefault+96 ()
ffffff000fcb5730 unix:trap+d88 ()
ffffff000fcb5740 unix:cmntrap+e6 ()
ffffff000fcb5880 vboxdrv:rtR0SegVBoxSolCreate+91 ()
ffffff000fcb5960 genunix:as_map_locked+1a4 ()
ffffff000fcb59d0 genunix:as_map+58 ()
ffffff000fcb5a70 vboxdrv:rtR0MemObjNativeMapUser+28f ()
ffffff000fcb5ae0 vboxdrv:RTR0MemObjMapUserTag+154 ()
ffffff000fcb5b80 vboxdrv:SUPR0GipMap+138 ()
ffffff000fcb5c10 vboxdrv:supdrvIOCtl+2e7c ()
ffffff000fcb5cc0 vboxdrv:VBoxDrvSolarisIOCtl+375 ()
ffffff000fcb5d00 genunix:cdev_ioctl+39 ()
ffffff000fcb5d50 specfs:spec_ioctl+60 ()
ffffff000fcb5de0 genunix:fop_ioctl+55 ()
ffffff000fcb5f00 genunix:ioctl+9b ()
ffffff000fcb5f10 unix:brand_sys_syscall+1f2 ()
dumping to /dev/zvol/dsk/rpool/dump, offset 65536, content: kernel
---</snip>---
Booting to older BE (with latest VBox package - 5.2.16) worked for me too.
Regards.
Post by Michal Nowak
_______________________________________________
openindiana-discuss mailing list
https://openindiana.org/mailman/listinfo/openindiana-discuss
Hi all,

Is there any workaround for this case?
It prevents me from OI /hipster upgrade (I need VBox for various reasons)...

With best regards.
Predrag Zečević
--
Predrag Zečević
Technical Support Analyst
2e Systems GmbH

tel: +49 - 6196 - 95058 - 15
mob: +49 - 174 - 3109288
fax: +49 - 6196 - 95058 - 94
e-mail: ***@2e-systems.com

headquarter: 2e Systems GmbH, Koenigsteiner Str. 87, 65812 Bad Soden am
Taunus, Germany
registration: Amtsgericht Koenigstein (Germany), HRB 7303
managing director: Phil Douglas

http://www.2e-systems.com/ - Making your business fly!
Till Wegmüller
2018-08-14 08:22:10 UTC
Permalink
Hi

Are you trying to boot 32-bit VM's?

Or is Virtualbox still part 32bit?

The referenced Bug mentions that only 32bit VM's are affected.

Can you boot the VM's with Qemu?

I doubt Bugs in June are affecting that Issue given that a Patch level
from 12th July still works.

What type of VM's are you trying to boot? Windows, Linux, 32/64 bit?

Did the Workaround work?
Post by Predrag Zečević - Technical Support Analyst
Post by Predrag Zečević - Technical Support Analyst
* https://www.illumos.org/issues/6919#note-7
https://wiki.openindiana.org/oi/7.2+VirtualBox#id-72VirtualBox-Virtualizationfeatures

When reading the crashdump virtualbox seems to choke on IOctl.

Have you tried building Virtualbox OSE and see if that works for you?

Thanks for the info
Greetings
Till
Post by Predrag Zečević - Technical Support Analyst
Post by Predrag Zečević - Technical Support Analyst
Post by Michal Nowak
Post by russell
Hi,
I have been VirtualBox 5.2.14 since it came out without any issues
until I performed a pkg update on 22nd July.
Everytime I perform a pkg update I always specify a new boot
environment, so last update which VirtualBox 5.2.14 worked on is the
one I performed on the 12th July. I have made this BE activate so I
do not have to remember to select it on boot.
Regards
Russell
I think it might be the "eager FPU" changes which landed in
https://github.com/illumos/illumos-gate/commit/4c28a617e3922d92a58e813a5b955eb526b9c386
https://github.com/illumos/illumos-gate/commit/a32a1f376ed9360264e4a374608fdcc5c4927d63
It's reported to the authors of those patches at
https://github.com/joyent/smartos-live/issues/792.
I don't know of a workaround apart from sticking to versions before
the change.
Michal
Hi all,
* https://www.illumos.org/issues/6919#note-7
https://wiki.openindiana.org/oi/7.2+VirtualBox#id-72VirtualBox-Virtualizationfeatures
Not sure what to do with this new case (should be illumos issue opened
or not)?
In any case, I have updated /hipster today and it has crashed. Output
---<snip>---
pseudo-device: vboxusbmon0
recursive rw_enter, lp=ffffff03e71be928 wwwh=ffffff03d96e0524
thread=ffffff03d96e0520
ffffff000fcb5440 unix:rw_panic+54 ()
ffffff000fcb54b0 unix:rw_enter_sleep+37a ()
ffffff000fcb55a0 genunix:as_fault+1f1 ()
ffffff000fcb5630 unix:pagefault+96 ()
ffffff000fcb5730 unix:trap+d88 ()
ffffff000fcb5740 unix:cmntrap+e6 ()
ffffff000fcb5880 vboxdrv:rtR0SegVBoxSolCreate+91 ()
ffffff000fcb5960 genunix:as_map_locked+1a4 ()
ffffff000fcb59d0 genunix:as_map+58 ()
ffffff000fcb5a70 vboxdrv:rtR0MemObjNativeMapUser+28f ()
ffffff000fcb5ae0 vboxdrv:RTR0MemObjMapUserTag+154 ()
ffffff000fcb5b80 vboxdrv:SUPR0GipMap+138 ()
ffffff000fcb5c10 vboxdrv:supdrvIOCtl+2e7c ()
ffffff000fcb5cc0 vboxdrv:VBoxDrvSolarisIOCtl+375 ()
ffffff000fcb5d00 genunix:cdev_ioctl+39 ()
ffffff000fcb5d50 specfs:spec_ioctl+60 ()
ffffff000fcb5de0 genunix:fop_ioctl+55 ()
ffffff000fcb5f00 genunix:ioctl+9b ()
ffffff000fcb5f10 unix:brand_sys_syscall+1f2 ()
dumping to /dev/zvol/dsk/rpool/dump, offset 65536, content: kernel
---</snip>---
Booting to older BE (with latest VBox package - 5.2.16) worked for me too.
Regards.
Post by Michal Nowak
_______________________________________________
openindiana-discuss mailing list
https://openindiana.org/mailman/listinfo/openindiana-discuss
Hi all,
Is there any workaround for this case?
It prevents me from OI /hipster upgrade (I need VBox for various reasons)...
With best regards.
Predrag Zečević
Predrag Zečević - Technical Support Analyst
2018-08-14 08:28:10 UTC
Permalink
Post by Till Wegmüller
Hi
Are you trying to boot 32-bit VM's?
Nope.
Post by Till Wegmüller
Or is Virtualbox still part 32bit?
No:
/opt/VirtualBox/amd64/VBoxSVC: ELF 64-bit LSB executable, x86-64,
version 1 (Solaris), dynamically linked, interpreter
/usr/lib/amd64/ld.so.1, stripped, uses SSE
Post by Till Wegmüller
The referenced Bug mentions that only 32bit VM's are affected.
Can you boot the VM's with Qemu?
Did not tried...
Post by Till Wegmüller
I doubt Bugs in June are affecting that Issue given that a Patch level
from 12th July still works.
What type of VM's are you trying to boot? Windows, Linux, 32/64 bit?
Bot Ubuntu and Windows 10 are 64 bit distribution.
Post by Till Wegmüller
Did the Workaround work?
Yes, we have added it to wiki
Post by Till Wegmüller
Post by Predrag Zečević - Technical Support Analyst
Post by Predrag Zečević - Technical Support Analyst
* https://www.illumos.org/issues/6919#note-7
https://wiki.openindiana.org/oi/7.2+VirtualBox#id-72VirtualBox-Virtualizationfeatures
When reading the crashdump virtualbox seems to choke on IOctl.
Have you tried building Virtualbox OSE and see if that works for you?
Thanks for the info
Greetings
Till
Post by Predrag Zečević - Technical Support Analyst
Post by Predrag Zečević - Technical Support Analyst
Post by Michal Nowak
Post by russell
Hi,
I have been VirtualBox 5.2.14 since it came out without any issues
until I performed a pkg update on 22nd July.
Everytime I perform a pkg update I always specify a new boot
environment, so last update which VirtualBox 5.2.14 worked on is the
one I performed on the 12th July. I have made this BE activate so I
do not have to remember to select it on boot.
Regards
Russell
I think it might be the "eager FPU" changes which landed in
https://github.com/illumos/illumos-gate/commit/4c28a617e3922d92a58e813a5b955eb526b9c386
https://github.com/illumos/illumos-gate/commit/a32a1f376ed9360264e4a374608fdcc5c4927d63
It's reported to the authors of those patches at
https://github.com/joyent/smartos-live/issues/792.
I don't know of a workaround apart from sticking to versions before
the change.
Michal
Hi all,
* https://www.illumos.org/issues/6919#note-7
https://wiki.openindiana.org/oi/7.2+VirtualBox#id-72VirtualBox-Virtualizationfeatures
Not sure what to do with this new case (should be illumos issue opened
or not)?
In any case, I have updated /hipster today and it has crashed. Output
---<snip>---
pseudo-device: vboxusbmon0
recursive rw_enter, lp=ffffff03e71be928 wwwh=ffffff03d96e0524
thread=ffffff03d96e0520
ffffff000fcb5440 unix:rw_panic+54 ()
ffffff000fcb54b0 unix:rw_enter_sleep+37a ()
ffffff000fcb55a0 genunix:as_fault+1f1 ()
ffffff000fcb5630 unix:pagefault+96 ()
ffffff000fcb5730 unix:trap+d88 ()
ffffff000fcb5740 unix:cmntrap+e6 ()
ffffff000fcb5880 vboxdrv:rtR0SegVBoxSolCreate+91 ()
ffffff000fcb5960 genunix:as_map_locked+1a4 ()
ffffff000fcb59d0 genunix:as_map+58 ()
ffffff000fcb5a70 vboxdrv:rtR0MemObjNativeMapUser+28f ()
ffffff000fcb5ae0 vboxdrv:RTR0MemObjMapUserTag+154 ()
ffffff000fcb5b80 vboxdrv:SUPR0GipMap+138 ()
ffffff000fcb5c10 vboxdrv:supdrvIOCtl+2e7c ()
ffffff000fcb5cc0 vboxdrv:VBoxDrvSolarisIOCtl+375 ()
ffffff000fcb5d00 genunix:cdev_ioctl+39 ()
ffffff000fcb5d50 specfs:spec_ioctl+60 ()
ffffff000fcb5de0 genunix:fop_ioctl+55 ()
ffffff000fcb5f00 genunix:ioctl+9b ()
ffffff000fcb5f10 unix:brand_sys_syscall+1f2 ()
dumping to /dev/zvol/dsk/rpool/dump, offset 65536, content: kernel
---</snip>---
Booting to older BE (with latest VBox package - 5.2.16) worked for me too.
Regards.
Post by Michal Nowak
_______________________________________________
openindiana-discuss mailing list
https://openindiana.org/mailman/listinfo/openindiana-discuss
Hi all,
Is there any workaround for this case?
It prevents me from OI /hipster upgrade (I need VBox for various reasons)...
With best regards.
Predrag Zečević
_______________________________________________
openindiana-discuss mailing list
https://openindiana.org/mailman/listinfo/openindiana-discuss
Sorry for adding answers in middle of thread.
Regards.
--
Predrag Zečević
Technical Support Analyst
2e Systems GmbH

tel: +49 - 6196 - 95058 - 15
mob: +49 - 174 - 3109288
fax: +49 - 6196 - 95058 - 94
e-mail: ***@2e-systems.com

headquarter: 2e Systems GmbH, Koenigsteiner Str. 87, 65812 Bad Soden am
Taunus, Germany
registration: Amtsgericht Koenigstein (Germany), HRB 7303
managing director: Phil Douglas

http://www.2e-systems.com/ - Making your business fly!
Till Wegmüller
2018-08-14 08:45:17 UTC
Permalink
Post by Predrag Zečević - Technical Support Analyst
Post by Till Wegmüller
Did the Workaround work?
Yes, we have added it to wiki
Post by Till Wegmüller
Post by Predrag Zečević - Technical Support Analyst
* https://www.illumos.org/issues/6919#note-7
https://wiki.openindiana.org/oi/7.2+VirtualBox#id-72VirtualBox-Virtualizationfeatures
I am a bit confused. Did that workaround work for your current case? If
so why are you asking for a Workaround?

Or was the problem that pkg overwrote 'set disable_smap=1' in /etc/system ?

It looks like that back in the day Virtualbox used an undefined
behaviour to access user data. Looks like a bug in Virtualbox to me. Os
and other devs can workaround it but ultimately If I read Rmustacs
comment correctly on the given illumos bug, then Virtualbox should use
the proper Driver Framework methods to get user data instead of IOctl's.
So unless the behave properly in modern Solarisch OSes there will always
be problem's.

Greetings
Till
Predrag Zečević - Technical Support Analyst
2018-08-14 08:56:32 UTC
Permalink
Post by Till Wegmüller
Post by Predrag Zečević - Technical Support Analyst
Post by Till Wegmüller
Did the Workaround work?
Yes, we have added it to wiki
Post by Till Wegmüller
Post by Predrag Zečević - Technical Support Analyst
* https://www.illumos.org/issues/6919#note-7
https://wiki.openindiana.org/oi/7.2+VirtualBox#id-72VirtualBox-Virtualizationfeatures
I am a bit confused. Did that workaround work for your current case? If
so why are you asking for a Workaround?
Or was the problem that pkg overwrote 'set disable_smap=1' in /etc/system ?
That one was working, back in time I have added it ...
It is still present, but VirtualBox has crashed...
Post by Till Wegmüller
It looks like that back in the day Virtualbox used an undefined
behaviour to access user data. Looks like a bug in Virtualbox to me. Os
and other devs can workaround it but ultimately If I read Rmustacs
comment correctly on the given illumos bug, then Virtualbox should use
the proper Driver Framework methods to get user data instead of IOctl's.
So unless the behave properly in modern Solarisch OSes there will always
be problem's.
Well, Oracle recommends Solaris 11 (which is modern enough, I presume).
Post by Till Wegmüller
Greetings
Till
Hi Till,

Thanks for answers. I will try another OI update and will post finding
here... Right now:
$ uname -rosv
SunOS 5.11 illumos-229852ddf2 illumos # BE is from 20180706

$ pkginfo -l SUNWvbox # Latest one, work fine
PKGINST: SUNWvbox
NAME: Oracle VM VirtualBox
CATEGORY: application
ARCH: i386
VERSION: 5.2.16,REV=2018.07.16.17.17.123759
BASEDIR: /
VENDOR: Oracle Corporation
DESC: A powerful PC virtualization solution
PSTAMP: vbox20180716171735_r123759
INSTDATE: Aug 06 2018 10:24
HOTLINE: Please contact your local service provider
EMAIL: ***@virtualbox.org
STATUS: completely installed
FILES: 335 installed pathnames
12 linked files
27 directories
28 executables
6 setuid/setgid executables
428875 blocks used (approx)

If I get time, will try with build from source
(https://www.virtualbox.org/wiki/Solaris%20build%20instructions)

Many thanks
Predrag Zečević
Post by Till Wegmüller
_______________________________________________
openindiana-discuss mailing list
https://openindiana.org/mailman/listinfo/openindiana-discuss
--
Predrag Zečević
Technical Support Analyst
2e Systems GmbH

tel: +49 - 6196 - 95058 - 15
mob: +49 - 174 - 3109288
fax: +49 - 6196 - 95058 - 94
e-mail: ***@2e-systems.com

headquarter: 2e Systems GmbH, Koenigsteiner Str. 87, 65812 Bad Soden am
Taunus, Germany
registration: Amtsgericht Koenigstein (Germany), HRB 7303
managing director: Phil Douglas

http://www.2e-systems.com/ - Making your business fly!
Predrag Zečević - Technical Support Analyst
2018-10-02 12:06:22 UTC
Permalink
Post by Till Wegmüller
Can you boot the VM's with Qemu?
I doubt Bugs in June are affecting that Issue given that a Patch level
from 12th July still works.
Hi all,

I have moved from VirtualBox to KVM and updated wiki page:
https://wiki.openindiana.org/oi/7.3+KVM#id-73KVM-Freshexample

Certainly, there is place for improvements - but for now I am satisfied.

With best regards.
Predrag Zečević

P.S: https://wiki.openindiana.org/oi/Recent+Wiki+Changes needs some love...
--
Predrag Zečević
Technical Support Analyst
2e Systems GmbH

tel: +49 - 6196 - 95058 - 15
mob: +49 - 174 - 3109288
fax: +49 - 6196 - 95058 - 94
e-mail: ***@2e-systems.com

headquarter: 2e Systems GmbH, Koenigsteiner Str. 87, 65812 Bad Soden am
Taunus, Germany
registration: Amtsgericht Koenigstein (Germany), HRB 7303
managing director: Phil Douglas

http://www.2e-systems.com/ - Making your business fly!
Predrag Zečević - Technical Support Analyst
2018-08-20 07:36:58 UTC
Permalink
Post by Michal Nowak
Post by russell
Hi,
I have been VirtualBox 5.2.14 since it came out without any issues
until I performed a pkg update on 22nd July.
Everytime I perform a pkg update I always specify a new boot
environment, so last update which VirtualBox 5.2.14 worked on is the
one I performed on the 12th July. I have made this BE activate so I do
not have to remember to select it on boot.
Regards
Russell
I think it might be the "eager FPU" changes which landed in illumos-gate
https://github.com/illumos/illumos-gate/commit/4c28a617e3922d92a58e813a5b955eb526b9c386
https://github.com/illumos/illumos-gate/commit/a32a1f376ed9360264e4a374608fdcc5c4927d63
It's reported to the authors of those patches at
https://github.com/joyent/smartos-live/issues/792.
I don't know of a workaround apart from sticking to versions before the
change.
Michal
_______________________________________________
openindiana-discuss mailing list
https://openindiana.org/mailman/listinfo/openindiana-discuss
Hi all,

just quick update on this: there were some changes in illumos, back (in
July, I presume) which have messed up run of VirualBox package (solaris
one, from virtualbox.org site, which is still running on old BE from
July 6th, 2018).

I have tried with OI /hipster update last time and as I have started
VBox guest, system has crashed (I can provide core dump if required - I
was reading it, but not sure what is causing problems).
BTW, almost latest VBox (5.2.16) is running fine on mentioned BE:
$ uname -rosv
SunOS 5.11 illumos-229852ddf2 illumos


I have tried to compile VBox from source and after some tweaks, managed
to get it compiled w/o errors, but it simply does not start (did not
have time to do deeper analysis):
$ LD_LIBRARY_PATH=/usr/lib/qt/5.8/lib/amd64 ./VirtualBox

caused error:
---<snip>---
Failed to create the VirtualBoxClient COM object.
The application will now terminate.

[Callee RC: NS_ERROR_FACTORY_NOT_REGISTERED (0x80040154)]
---</snip>---

Till Wegmüller asked if I can boot VBox guest in qemu - not investigated
although it is an option - I would like to stay at VirtualBox!

Does anyone knows how to solve this problem (failure of VBox prevents me
from OI upgrade).

With best regards.
Predrag Zečević
--
Predrag Zečević
Technical Support Analyst
2e Systems GmbH

tel: +49 - 6196 - 95058 - 15
mob: +49 - 174 - 3109288
fax: +49 - 6196 - 95058 - 94
e-mail: ***@2e-systems.com

headquarter: 2e Systems GmbH, Koenigsteiner Str. 87, 65812 Bad Soden am
Taunus, Germany
registration: Amtsgericht Koenigstein (Germany), HRB 7303
managing director: Phil Douglas

http://www.2e-systems.com/ - Making your business fly!
Jacques Hugo
2018-08-22 10:39:05 UTC
Permalink
Hey Predrag,

I'm on this release -> SunOS 5.11 illumos-66a2141c21 illumos

I have exactly the same issues as you with VBox after years of working perfectly. I removed all VB instances and tried the latest qemu-kvm package available in the repo, with exactly the same results. If you touch the qemu-kvm bin the host panics immediately and reboots. I have emailed the fmdump output in a previous post.

I really don't want to roll-back just yet since I have a few Zones I need to degrade as well.

Any info from the community will be greatly appreciated.

Cheers,
Jacques


Jacques Hugo
Three Two Labs
Information Technology | Solaris Engineering 32L
Phone +44 7939 433701
***@threetwo.co.uk

-----Original Message-----
From: Predrag Zečević - Technical Support Analyst [mailto:***@2e-systems.com]
Sent: 20 August 2018 08:37
To: openindiana-***@openindiana.org
Subject: Re: [OpenIndiana-discuss] VirtualBox 5.2.14 crashing when attempting to start VM
Post by Michal Nowak
Post by russell
Hi,
I have been VirtualBox 5.2.14 since it came out without any issues
until I performed a pkg update on 22nd July.
Everytime I perform a pkg update I always specify a new boot
environment, so last update which VirtualBox 5.2.14 worked on is the
one I performed on the 12th July. I have made this BE activate so I
do not have to remember to select it on boot.
Regards
Russell
I think it might be the "eager FPU" changes which landed in
https://github.com/illumos/illumos-gate/commit/4c28a617e3922d92a58e813
a5b955eb526b9c386
https://github.com/illumos/illumos-gate/commit/a32a1f376ed9360264e4a37
4608fdcc5c4927d63
It's reported to the authors of those patches at
https://github.com/joyent/smartos-live/issues/792.
I don't know of a workaround apart from sticking to versions before
the change.
Michal
_______________________________________________
openindiana-discuss mailing list
https://openindiana.org/mailman/listinfo/openindiana-discuss
Hi all,

just quick update on this: there were some changes in illumos, back (in July, I presume) which have messed up run of VirualBox package (solaris one, from virtualbox.org site, which is still running on old BE from July 6th, 2018).

I have tried with OI /hipster update last time and as I have started VBox guest, system has crashed (I can provide core dump if required - I was reading it, but not sure what is causing problems).
BTW, almost latest VBox (5.2.16) is running fine on mentioned BE:
$ uname -rosv
SunOS 5.11 illumos-229852ddf2 illumos


I have tried to compile VBox from source and after some tweaks, managed to get it compiled w/o errors, but it simply does not start (did not have time to do deeper analysis):
$ LD_LIBRARY_PATH=/usr/lib/qt/5.8/lib/amd64 ./VirtualBox

caused error:
---<snip>---
Failed to create the VirtualBoxClient COM object.
The application will now terminate.

[Callee RC: NS_ERROR_FACTORY_NOT_REGISTERED (0x80040154)]
---</snip>---

Till Wegmüller asked if I can boot VBox guest in qemu - not investigated although it is an option - I would like to stay at VirtualBox!

Does anyone knows how to solve this problem (failure of VBox prevents me from OI upgrade).

With best regards.
Predrag Zečević

--
Predrag Zečević
Technical Support Analyst
2e Systems GmbH

tel: +49 - 6196 - 95058 - 15
mob: +49 - 174 - 3109288
fax: +49 - 6196 - 95058 - 94
e-mail: ***@2e-systems.com

headquarter: 2e Systems GmbH, Koenigsteiner Str. 87, 65812 Bad Soden am Taunus, Germany
registration: Amtsgericht Koenigstein (Germany), HRB 7303 managing director: Phil Douglas

http://www.2e-systems.com/ - Making your business fly!

_______________________________________________
openindiana-discuss mailing list
openindiana-***@openindiana.org
https://openindiana.org/mailman/listinfo/openindiana-discuss
Udo Grabowski (IMK)
2018-08-22 12:29:23 UTC
Permalink
Post by Jacques Hugo
Hey Predrag,
.....
I'm on this release -> SunOS 5.11 illumos-66a2141c21 illumos
...
If you touch the qemu-kvm bin the host panics immediately and reboots....
....
Checked that on real hardware (DataON CIB-9470V12, E5-2600 V2, C602)
with current hipster:

illumos-6640c13bea i86pc i386 i86pc

qemu-kvm reboots the host immediately. Seems to be a serious bug.
No panic message, just a reboot, starting in dbx looks like this:

ro sunth7 ~ # dbx /usr/bin/qemu-kvm
For information about new features see `help changes'
To remove this message, put `dbxenv suppress_startup_message 7.7' in your .dbxrc
Reading qemu-kvm
Reading ld.so.1
Reading libc.so.1
(dbx) run -m 4g -smp 3 -cdrom /tftpboot/OI-hipster-minimal-20180427.iso
Running: qemu-kvm -m 4g -smp 3 -cdrom /tftpboot/OI-hipster-minimal-20180427.iso
(process id 1238)
Reading libm.so.2
Reading libnsl.so.1
<reboots>
--
Dr.Udo Grabowski Inst.f.Meteorology & Climate Research IMK-ASF-SAT
http://www.imk-asf.kit.edu/english/sat.php
KIT - Karlsruhe Institute of Technology http://www.kit.edu
Postfach 3640,76021 Karlsruhe,Germany T:(+49)721 608-26026 F:-926026
Udo Grabowski (IMK)
2018-08-22 13:18:10 UTC
Permalink
Post by Udo Grabowski (IMK)
Post by Jacques Hugo
Hey Predrag,
.....
I'm on this release -> SunOS 5.11 illumos-66a2141c21 illumos
...
If you touch the qemu-kvm bin the host panics immediately and reboots....
....
Checked that on real hardware (DataON CIB-9470V12, E5-2600 V2, C602)
illumos-6640c13bea i86pc i386 i86pc
qemu-kvm reboots the host immediately. Seems to be a serious bug.
ro sunth7 ~ # dbx /usr/bin/qemu-kvm
For information about new features see `help changes'
To remove this message, put `dbxenv suppress_startup_message 7.7' in your .dbxrc
Reading qemu-kvm
Reading ld.so.1
Reading libc.so.1
(dbx) run -m 4g -smp 3 -cdrom /tftpboot/OI-hipster-minimal-20180427.iso
Running: qemu-kvm -m 4g -smp 3 -cdrom /tftpboot/OI-hipster-minimal-20180427.iso
(process id 1238)
Reading libm.so.2
Reading libnsl.so.1
<reboots>
::status
debugging crash dump vmcore.0 (64-bit) from imksunth7
operating system: 5.11 illumos-6640c13bea (i86pc)
image uuid: ba4a4d5d-58bd-6dbd-ab55-ba047c459f90
panic message: BAD TRAP: type=7 (#nm Device not available) rp=fffffe426c6f2300
addr=fffffe426c6f2ea8
dump content: kernel pages only
Post by Udo Grabowski (IMK)
::stack
kvm_fx_finit()
vmx_vcpu_reset+0x97()
kvm_arch_vcpu_reset+0x6c()
kvm_arch_vcpu_setup+0x28()
kvm_vm_ioctl_create_vcpu+0x44()
kvm_ioctl+0xa2c()
cdev_ioctl+0x39(13200000002, 2000ae41, 0, 202003, fffffea3a23ccad8,
fffffe426c6f2ea8)
spec_ioctl+0x60(fffffea3e4fa3880, 2000ae41, 0, 202003, fffffea3a23ccad8,
fffffe426c6f2ea8)
fop_ioctl+0x55(fffffea3e4fa3880, 2000ae41, 0, 202003, fffffea3a23ccad8,
fffffe426c6f2ea8)
ioctl+0x9b(9, 2000ae41, 0)
sys_syscall+0x177()


/dev/kvm is available and module is loaded:

ro sunth7 /var/crash/crashcore # modinfo |grep kvm
261 fffffffff7e7e000 376b0 306 1 kvm (kvm driver v0.1)
--
Dr.Udo Grabowski Inst.f.Meteorology & Climate Research IMK-ASF-SAT
http://www.imk-asf.kit.edu/english/sat.php
KIT - Karlsruhe Institute of Technology http://www.kit.edu
Postfach 3640,76021 Karlsruhe,Germany T:(+49)721 608-26026 F:-926026
Predrag Zečević - Technical Support Analyst
2018-08-22 13:35:59 UTC
Permalink
Post by Udo Grabowski (IMK)
Post by Udo Grabowski (IMK)
Post by Jacques Hugo
Hey Predrag,
.....
I'm on this release -> SunOS 5.11 illumos-66a2141c21 illumos
...
If you touch the qemu-kvm bin the host panics immediately and reboots....
....
Checked that on real hardware (DataON CIB-9470V12, E5-2600 V2, C602)
illumos-6640c13bea i86pc i386 i86pc
qemu-kvm reboots the host immediately. Seems to be a serious bug.
ro sunth7 ~ # dbx /usr/bin/qemu-kvm
For information about new features see `help changes'
To remove this message, put `dbxenv suppress_startup_message 7.7' in your .dbxrc
Reading qemu-kvm
Reading ld.so.1
Reading libc.so.1
(dbx) run -m 4g -smp 3 -cdrom /tftpboot/OI-hipster-minimal-20180427.iso
Running: qemu-kvm -m 4g -smp 3 -cdrom
/tftpboot/OI-hipster-minimal-20180427.iso
(process id 1238)
Reading libm.so.2
Reading libnsl.so.1
<reboots>
::status
debugging crash dump vmcore.0 (64-bit) from imksunth7
operating system: 5.11 illumos-6640c13bea (i86pc)
image uuid: ba4a4d5d-58bd-6dbd-ab55-ba047c459f90
panic message: BAD TRAP: type=7 (#nm Device not available)
rp=fffffe426c6f2300 addr=fffffe426c6f2ea8
dump content: kernel pages only
Post by Udo Grabowski (IMK)
::stack
kvm_fx_finit()
vmx_vcpu_reset+0x97()
kvm_arch_vcpu_reset+0x6c()
kvm_arch_vcpu_setup+0x28()
kvm_vm_ioctl_create_vcpu+0x44()
kvm_ioctl+0xa2c()
cdev_ioctl+0x39(13200000002, 2000ae41, 0, 202003, fffffea3a23ccad8,
fffffe426c6f2ea8)
spec_ioctl+0x60(fffffea3e4fa3880, 2000ae41, 0, 202003, fffffea3a23ccad8,
fffffe426c6f2ea8)
fop_ioctl+0x55(fffffea3e4fa3880, 2000ae41, 0, 202003, fffffea3a23ccad8,
fffffe426c6f2ea8)
ioctl+0x9b(9, 2000ae41, 0)
sys_syscall+0x177()
ro sunth7 /var/crash/crashcore # modinfo |grep kvm
261 fffffffff7e7e000  376b0 306   1  kvm (kvm driver v0.1)
I have impression that incompatible changes was made on illumos level ...

Thanks for looking at.
Now, we need someone to get it working back...

Regards.
--
Predrag Zečević
Technical Support Analyst
2e Systems GmbH

tel: +49 - 6196 - 95058 - 15
mob: +49 - 174 - 3109288
fax: +49 - 6196 - 95058 - 94
e-mail: ***@2e-systems.com

headquarter: 2e Systems GmbH, Koenigsteiner Str. 87, 65812 Bad Soden am
Taunus, Germany
registration: Amtsgericht Koenigstein (Germany), HRB 7303
managing director: Phil Douglas

http://www.2e-systems.com/ - Making your business fly!
Till Wegmüller
2018-08-22 15:33:28 UTC
Permalink
Hi

I get the same result on my machine illumos-49714e869e

Syscall 177 which is at the bottom of the stack is munmap. So my
assumption is there was a bug introduced in KVM some tima ago.

I have also tested disabling KPTI with no luck. so it's not that.

Greetings
Till
Post by Predrag Zečević - Technical Support Analyst
Post by Udo Grabowski (IMK)
Post by Jacques Hugo
Hey Predrag,
.....
I'm on this release -> SunOS 5.11 illumos-66a2141c21 illumos
...
If you touch the qemu-kvm bin the host panics immediately and reboots....
....
Checked that on real hardware (DataON CIB-9470V12, E5-2600 V2, C602)
illumos-6640c13bea i86pc i386 i86pc
qemu-kvm reboots the host immediately. Seems to be a serious bug.
ro sunth7 ~ # dbx /usr/bin/qemu-kvm
For information about new features see `help changes'
To remove this message, put `dbxenv suppress_startup_message 7.7' in your .dbxrc
Reading qemu-kvm
Reading ld.so.1
Reading libc.so.1
(dbx) run -m 4g -smp 3 -cdrom /tftpboot/OI-hipster-minimal-20180427.iso
Running: qemu-kvm -m 4g -smp 3 -cdrom
/tftpboot/OI-hipster-minimal-20180427.iso
(process id 1238)
Reading libm.so.2
Reading libnsl.so.1
<reboots>
 > ::status
debugging crash dump vmcore.0 (64-bit) from imksunth7
operating system: 5.11 illumos-6640c13bea (i86pc)
image uuid: ba4a4d5d-58bd-6dbd-ab55-ba047c459f90
panic message: BAD TRAP: type=7 (#nm Device not available)
rp=fffffe426c6f2300 addr=fffffe426c6f2ea8
dump content: kernel pages only
 > ::stack
kvm_fx_finit()
vmx_vcpu_reset+0x97()
kvm_arch_vcpu_reset+0x6c()
kvm_arch_vcpu_setup+0x28()
kvm_vm_ioctl_create_vcpu+0x44()
kvm_ioctl+0xa2c()
cdev_ioctl+0x39(13200000002, 2000ae41, 0, 202003, fffffea3a23ccad8,
fffffe426c6f2ea8)
spec_ioctl+0x60(fffffea3e4fa3880, 2000ae41, 0, 202003,
fffffea3a23ccad8, fffffe426c6f2ea8)
fop_ioctl+0x55(fffffea3e4fa3880, 2000ae41, 0, 202003,
fffffea3a23ccad8, fffffe426c6f2ea8)
ioctl+0x9b(9, 2000ae41, 0)
sys_syscall+0x177()
ro sunth7 /var/crash/crashcore # modinfo |grep kvm
261 fffffffff7e7e000  376b0 306   1  kvm (kvm driver v0.1)
I have impression that incompatible changes was made on illumos level ...
Thanks for looking at.
Now, we need someone to get it working back...
Regards.
Till Wegmüller
2018-08-22 15:41:07 UTC
Permalink
Ticket created

https://www.illumos.org/issues/9761


Greetings
Till
Post by Predrag Zečević - Technical Support Analyst
Post by Udo Grabowski (IMK)
Post by Jacques Hugo
Hey Predrag,
.....
I'm on this release -> SunOS 5.11 illumos-66a2141c21 illumos
...
If you touch the qemu-kvm bin the host panics immediately and reboots....
....
Checked that on real hardware (DataON CIB-9470V12, E5-2600 V2, C602)
illumos-6640c13bea i86pc i386 i86pc
qemu-kvm reboots the host immediately. Seems to be a serious bug.
ro sunth7 ~ # dbx /usr/bin/qemu-kvm
For information about new features see `help changes'
To remove this message, put `dbxenv suppress_startup_message 7.7' in your .dbxrc
Reading qemu-kvm
Reading ld.so.1
Reading libc.so.1
(dbx) run -m 4g -smp 3 -cdrom /tftpboot/OI-hipster-minimal-20180427.iso
Running: qemu-kvm -m 4g -smp 3 -cdrom
/tftpboot/OI-hipster-minimal-20180427.iso
(process id 1238)
Reading libm.so.2
Reading libnsl.so.1
<reboots>
 > ::status
debugging crash dump vmcore.0 (64-bit) from imksunth7
operating system: 5.11 illumos-6640c13bea (i86pc)
image uuid: ba4a4d5d-58bd-6dbd-ab55-ba047c459f90
panic message: BAD TRAP: type=7 (#nm Device not available)
rp=fffffe426c6f2300 addr=fffffe426c6f2ea8
dump content: kernel pages only
 > ::stack
kvm_fx_finit()
vmx_vcpu_reset+0x97()
kvm_arch_vcpu_reset+0x6c()
kvm_arch_vcpu_setup+0x28()
kvm_vm_ioctl_create_vcpu+0x44()
kvm_ioctl+0xa2c()
cdev_ioctl+0x39(13200000002, 2000ae41, 0, 202003, fffffea3a23ccad8,
fffffe426c6f2ea8)
spec_ioctl+0x60(fffffea3e4fa3880, 2000ae41, 0, 202003,
fffffea3a23ccad8, fffffe426c6f2ea8)
fop_ioctl+0x55(fffffea3e4fa3880, 2000ae41, 0, 202003,
fffffea3a23ccad8, fffffe426c6f2ea8)
ioctl+0x9b(9, 2000ae41, 0)
sys_syscall+0x177()
ro sunth7 /var/crash/crashcore # modinfo |grep kvm
261 fffffffff7e7e000  376b0 306   1  kvm (kvm driver v0.1)
I have impression that incompatible changes was made on illumos level ...
Thanks for looking at.
Now, we need someone to get it working back...
Regards.
Jacques Hugo
2018-08-22 21:41:45 UTC
Permalink
Hey guys,

I rolled back to pkg://openindiana.org/***@0.5.11-2018.0.0.0:20180211T131012Z which is the pre-June update and all is back to normal again. My issue was that the mod "VirtualBox HostDrv" could not load at all, and the panic/reboots were swift and brutal.

Cheers,
Jacques




Jacques Hugo
Three Two Labs
Information Technology | Solaris Engineering 32L
Phone +44 7939 433701
***@threetwo.co.uk


-----Original Message-----
From: Predrag Zečević - Technical Support Analyst [mailto:***@2e-systems.com]
Sent: 22 August 2018 14:36
To: openindiana-***@openindiana.org
Subject: Re: [OpenIndiana-discuss] VirtualBox 5.2.14 crashing when attempting to start VM
Post by Udo Grabowski (IMK)
Post by Jacques Hugo
Hey Predrag,
.....
I'm on this release -> SunOS 5.11 illumos-66a2141c21 illumos ...
If you touch the qemu-kvm bin the host panics immediately and reboots....
....
Checked that on real hardware (DataON CIB-9470V12, E5-2600 V2, C602)
illumos-6640c13bea i86pc i386 i86pc
qemu-kvm reboots the host immediately. Seems to be a serious bug.
ro sunth7 ~ # dbx /usr/bin/qemu-kvm
For information about new features see `help changes'
To remove this message, put `dbxenv suppress_startup_message 7.7' in
your .dbxrc Reading qemu-kvm Reading ld.so.1 Reading libc.so.1
(dbx) run -m 4g -smp 3 -cdrom
/tftpboot/OI-hipster-minimal-20180427.iso
Running: qemu-kvm -m 4g -smp 3 -cdrom
/tftpboot/OI-hipster-minimal-20180427.iso
(process id 1238)
Reading libm.so.2
Reading libnsl.so.1
<reboots>
::status
debugging crash dump vmcore.0 (64-bit) from imksunth7 operating
ba4a4d5d-58bd-6dbd-ab55-ba047c459f90
panic message: BAD TRAP: type=7 (#nm Device not available)
rp=fffffe426c6f2300 addr=fffffe426c6f2ea8 dump content: kernel pages
only
Post by Udo Grabowski (IMK)
::stack
kvm_fx_finit()
vmx_vcpu_reset+0x97()
kvm_arch_vcpu_reset+0x6c()
kvm_arch_vcpu_setup+0x28()
kvm_vm_ioctl_create_vcpu+0x44()
kvm_ioctl+0xa2c()
cdev_ioctl+0x39(13200000002, 2000ae41, 0, 202003, fffffea3a23ccad8,
fffffe426c6f2ea8)
spec_ioctl+0x60(fffffea3e4fa3880, 2000ae41, 0, 202003,
fffffea3a23ccad8,
fffffe426c6f2ea8)
fop_ioctl+0x55(fffffea3e4fa3880, 2000ae41, 0, 202003,
fffffea3a23ccad8,
fffffe426c6f2ea8)
ioctl+0x9b(9, 2000ae41, 0)
sys_syscall+0x177()
ro sunth7 /var/crash/crashcore # modinfo |grep kvm
261 fffffffff7e7e000 376b0 306 1 kvm (kvm driver v0.1)
I have impression that incompatible changes was made on illumos level ...

Thanks for looking at.
Now, we need someone to get it working back...

Regards.

--
Predrag Zečević
Technical Support Analyst
2e Systems GmbH

tel: +49 - 6196 - 95058 - 15
mob: +49 - 174 - 3109288
fax: +49 - 6196 - 95058 - 94
e-mail: ***@2e-systems.com

headquarter: 2e Systems GmbH, Koenigsteiner Str. 87, 65812 Bad Soden am Taunus, Germany
registration: Amtsgericht Koenigstein (Germany), HRB 7303 managing director: Phil Douglas

http://www.2e-systems.com/ - Making your business fly!

_______________________________________________
openindiana-discuss mailing list
openindiana-***@openindiana.org
https://openindiana.org/mailman/listinfo/openindiana-discuss
Continue reading on narkive:
Search results for '[OpenIndiana-discuss] VirtualBox 5.2.14 crashing when attempting to start VM' (Questions and Answers)
35
replies
is it safe to run two os on same laprop?
started 2015-08-17 23:44:01 UTC
desktops
Loading...