W10 Guest crash when watching youtube videos

[Pablo Sanchez]

Hi,

When I found 6.0.14 crashing, I tested with the latest nightly[1] and Extension[1.5].

My Host O/S is openSUSE Tumbleweed. I am up to date[2].

The Guest O/S is an up to date W10 VM.

While watching a youtube video (any will do), the VM crashes.

The following errors are logged. First, in /var/log/messages:

Nov  7 17:05:08 fuzzy kernel: EMT-1[28979]: segfault at 7fe0ee42d920 ip 00007fe21dd295ff sp 00007fe1ee42d898 error 4 in libc-2.30.so[7fe21dca6000+14d000]
Nov  7 17:05:08 fuzzy kernel: Code: a4 c3 80 fa 08 73 12 80 fa 04 73 1e 80 fa 01 77 26 72 05 0f b6 0e 88 0f c3 48 8b 4c 16 f8 48 8b 36 48 89 4c 17 f8 48 89 37 c3 <8b> 4c 16 fc 8b 36 89 4c 17 fc 89 37 c3 0f b7 4c 16 fe 0f b7 36 66

after, in VBoxSVC.log:

00:07:50.279667 main     Saving settings file "/home/xyz/.VirtualBox/load_test/load_test.vbox" with version "1.15-linux"
00:07:50.742201 Watcher  Reaper: Pid 28941 (710d) was signalled: 11 (0xb)
00:07:55.284290 main     VirtualBox: object deletion starts
00:07:55.284327 main     ERROR [COM]: aRC=VBOX_E_OBJECT_IN_USE (0x80bb000c) aIID={ad47ad09-787b-44ab-b343-a082a3f2dfb1} aComponent={MediumWrap} aText={Medium '/home/xyz/.VirtualBox/w10/w10.vdi' cannot be closed because it is still attached to 1 virtual machines}, preserve=false aResultDetail=0
00:07:55.284451 main     ERROR [COM]: aRC=VBOX_E_OBJECT_IN_USE (0x80bb000c) aIID={ad47ad09-787b-44ab-b343-a082a3f2dfb1} aComponent={MediumWrap} aText={Medium '/home/xyz/.VirtualBox/r-mobile/r-mobile.vdi' cannot be closed because it is still attached to 1 virtual machines}, preserve=false aResultDetail=0
00:07:55.284533 main     ERROR [COM]: aRC=VBOX_E_OBJECT_IN_USE (0x80bb000c) aIID={ad47ad09-787b-44ab-b343-a082a3f2dfb1} aComponent={MediumWrap} aText={Medium '/home/xyz/.VirtualBox/load_test/load_test.vdi' cannot be closed because it is still attached to 1 virtual machines}, preserve=false aResultDetail=0
00:07:55.743302 Watcher  ERROR [COM]: aRC=E_ACCESSDENIED (0x80070005) aIID={d0a0163f-e254-4e5b-a1f2-011cf991c38d} aComponent={VirtualBoxWrap} aText={The object is not ready}, preserve=false aResultDetail=0
00:08:00.315967 main     {00007ff38001ba80} HostPowerServiceLinux::~HostPowerServiceLinux: RTThreadWait() for 5000 ms failed with VERR_TIMEOUT
00:08:00.317272 main     VirtualBox: object deleted

Nothing is placed in VBox.log

Please let me know what other data is needed. I am sure you need more. ;)

References 1 - VirtualBox-6.0.15-134477-Linux_amd64.run 1.5 - Oracle_VM_VirtualBox_Extension_Pack-6.0.15-134477.vbox-extpack 2 - Tumbleweed version 20191104

[Socratis]

Related discussion in the forums: ​https://forums.virtualbox.org/viewtopic.php?f=7&t=95478

You were supposed to follow these steps when you filed the bug, and provide a VBox.log:

Attach a (full) log file ("Machine" menu/"Show Log" in the main VirtualBox Manager window) straight away to save time for you and for us. The log file contains a lot of useful information about both the host and the guest systems as well as information about what happened during a particular machine run. Please do not cut and paste it.

[Pablo Sanchez]

$USER replaced with xyz

[Pablo Sanchez]

Test Case

• Start chrome
• Watch the following Ellen .. ;) .... youtube video: ​https://www.youtube.com/watch?v=8kI8tnefeds
• After three to six minutes, the VM will crash. I haven't timed it or confirmed if it happens at the exact playback time. If required, I can do that.

[Pablo Sanchez]

Still failing with 6.0.5 r134780 (Qt 5.6.1)

[sunlover]

I could not reproduce the crash. Please get a minidump from the guest:

​https://support.microsoft.com/en-us/help/315263/how-to-read-the-small-memory-dump-file-that-is-created-by-windows-if-a

[Pablo Sanchez]

Placeholder

[Pablo Sanchez]

Another data point: I tried RC1 with similar results:

VirtualBox-6.1.0_RC1-134891-Linux_amd64.run

[Pablo Sanchez]

Hi,

In order to get some traction on this issue, would it be better if I switch to 6.1/RC1?

Thx!

[Pablo Sanchez]

Due to no activity from Oracle, please close the ticket.

[Socratis]

Replying to sunlover:

I could not reproduce the crash.

Replying to pablo_:

Due to no activity from Oracle, please close the ticket.

That was Oracle (sunlover) replying to you. Not sure if by "no activity" you meant "no immediate patch" for something that wasn't readily reproducible.

I'm going to go ahead and per your request close the ticket.

[Pablo Sanchez]

Follow the activity and you'll see there's no meaningful work being done by the vendor.

It is nonsense to try to characterize it as "no immediate patch"

Anyone can see that in an effort to make some/any progress, I provided a stack track.

It is important to understand that when people spend the time to gather data to report a bug, timely responses are warranted.

[Pablo Sanchez]

Furthermore, if you look at the timeline, you can see it took over two weeks for an initial response and so far 11 days and counting for a subsequent response.

When I reported issues before to Innotek, Frank and team were way more responsive.

[Pablo Sanchez]

Given socratis banned me on the forums, he decides to personally attack me as well on there:

And that's exactly why 'whyn0t' gets permanently banned for severe forum vandalism.

Their ego got hurt because they didn't hear back from an Oracle developer (which they did in the ticket). Same attitude in the ticket. I guess some people want immediate, personal attention, and if they don't get it they're holding their breath until they turn blue. And that's fine, as long as that affects only themselves. If their actions affect the community, there are going to be consequences..."

This is too bad.

I suppose for him, waiting for a month is immediate personal attention

The above is nonsense.

Furthermore, since when am I, the owner of my content, not allowed to delete it. I wrote it, I can remove it.

To set up a straw man to personally attack me is ridiculous.

[Klaus Espenlaub]

This really looks to me like an unfortunate sequence of events. sunlover seems to have missed the fact that the symptom description showed that the VM process (on Linux host) is crashing. This means asking for a minidump was asking for the impossible, he should've asked for a core dump (which needs enabling, see https://www.virtualbox.org/wiki/Core_dump ).

Your debugging attempts were bound to not get far, because the VirtualBox packages don't contain debug infos (to reduce the download size), but the dev team could've worked from a core dump because we have debug info for all releases in our archive.

[Pablo Sanchez]

Hey Klaus,

Let's press on.

Wiki Tweak

The wiki needs an entry for openSUSE.

I am happy to make the change but if you rather handle it, here's the link for core dump:

​https://www.suse.com/support/kb/doc/?id=3054866

core dump generated

I generated a core dump.

After I compressed it, it's 12 MiB. I could uuencode / split but it may contain sensitive information.

Is there an equivalent crash reporting tool that you'd like me to use for openSUSE? apport is a Ubuntu-ism.

Work-around Implemented

Per the sidebar conversation, I have my expectation set. I have a kludgy work-around.

[Frank Batschulat (Oracle)]

Replying to pablo_:

Wiki Tweak

The wiki needs an entry for openSUSE.

I am happy to make the change but if you rather handle it, here's the link for core dump:

​https://www.suse.com/support/kb/doc/?id=3054866

That wiki page has been updated with a #SUSE section and this link, thanks!

[Pablo Sanchez]

Replying to fbatschu:

That wiki page has been updated with a #SUSE section and this link, thanks!

Thank you for rolling the fix.

Let's sit tight for a bit as I try to create a sanitized VM that I can easily share its core dump.

I'll be back with an update. ;)

Thank you! -pablo

[Pablo Sanchez]

Good news. I created a simple test case.

I have uploaded the core and md5sum to my G Drive. Per the Core Dump Wiki, I sent the link to Alexander.

If needed, I can even upload the compressed .vdi.

Thx!