Guest freeze / hangs on network - virtio_net virtio0: output.0:id 74 is not a head!

[Rodrigo Gomes]

Hello guys,

Today my guest machine was completely offline.
I do not know why or what caused it.

There was nothing abnormal in VBox logs and the guest machine worked perfectly well except for the fact that the network did not respond.
The network was perfectly fine on the host machine.

In the system logs I found these errors:

Code: Select all

.....
Apr  2 08:31:51 server kernel: virtio_net virtio0: output.0:id 74 is not a head!
Apr  2 08:31:51 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:31:51 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:31:51 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:31:51 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:31:51 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:31:51 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:31:51 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:31:51 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:31:51 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:31:51 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:31:56 server kernel: net_ratelimit: 736 callbacks suppressed
Apr  2 08:31:56 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:31:56 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:31:56 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:31:56 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:31:56 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:31:56 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:31:56 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:31:56 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:31:56 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:31:56 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
.....

.....
Apr  2 08:32:41 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:32:41 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:32:41 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:32:41 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:32:41 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:32:41 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:32:41 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:32:41 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:32:41 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:32:41 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:32:42 server kernel: TCP: request_sock_TCP: Possible SYN flooding on port 80. Sending cookies.  Check SNMP counters.
Apr  2 08:32:46 server kernel: net_ratelimit: 567 callbacks suppressed
Apr  2 08:32:46 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
Apr  2 08:32:46 server kernel: net enp0s3: Unexpected TXQ (0) queue failure: -5
.....

Searching the internet I found these pages with similar errors or equal to this:
https://bugzilla.redhat.com/show_bug.cgi?id=1264021
https://bugs.launchpad.net/qemu/+bug/1359394
https://www.virtualbox.org/ticket/11455

This is the same machine I have freezing problems reported here. I opened a new thread because this error is completely different I've seen so far and does not seem to be related to freezes.

Host:
Windows Server 2012 R2 Standard - 64 bits
32GB RAM
Intel Xeon E5-1620 v3
VirtualBox version 5.1.8 r111374 with Extension Pack

Guest:
CentOS Linux release 7.3.1611 (Core) - 64 bits
Guest additions installed
3D or 2D is not enabled

Reboot the machine solved the problem. But I would like to prevent this from happen again.

[Perryg]

Not sure how VirtualBox could help in this case since it appears to be an issue with the guest kernel. I did a quick search for "Possible SYN flooding on port 80" and there are plenty of suggestions on what to do. The only thing I can think of as far as VirtualBox is concerned is to use the default network mode instead of VirtIO.

[Rodrigo Gomes]

Hello Perryg,
Thanks for the suggestion, I'll try to contact Red Hat.

This syn flood warning was strange, the first thing I did was check if it was not some attack.
I checked the SYN connections and there was nothing strange. Bandwidth usage during this failure has dropped, rather than increased (as in the case of attacks).

[_sly_]

In case that is of any help, 3 years later, I'm still experiencing the same issue.
Guest :
Linux 4.19.0-9-amd64
debian 10 (but also occurs on debian 9 et debian 8 with stock kernels)
virtual network adapter
8Go RAM

Host :
Debian 10
Oracle VM VirtualBox VM Selector v6.1.6


The VM randomly (~once per month) loses its network adapter with this message on the Guest :

Sep 10 11:11:00 idmdl1 kernel: [4382567.921550] virtio_net virtio0: output.0:id 195 is not a head!
Sep 10 11:11:00 idmdl1 kernel: [4382567.921675] net eth0: Unexpected TXQ (0) queue failure: -5
Sep 10 11:11:00 idmdl1 kernel: [4382567.921688] net eth0: Unexpected TXQ (0) queue failure: -5
Sep 10 11:11:00 idmdl1 kernel: [4382567.921695] net eth0: Unexpected TXQ (0) queue failure: -5
Sep 10 11:11:00 idmdl1 kernel: [4382567.932930] net eth0: Unexpected TXQ (0) queue failure: -5
Sep 10 11:11:00 idmdl1 kernel: [4382567.932942] net eth0: Unexpected TXQ (0) queue failure: -5
Sep 10 11:11:00 idmdl1 kernel: [4382567.932949] net eth0: Unexpected TXQ (0) queue failure: -5

rebooting the Guest solves temporarily the problem until next time. I was tired of upgrading kernels and waiting for a kernel patch to solve this issue, so I switched to any Intel network Adapter in virtualbox and the problem is gone. Sad because the virtual adapter looks to be the way to go...

In all that, it's quite hard to believe that such a virtio_net bug on debian stock kernels + redhat kernels has gone unnoticed for all kvm/qemu/virtualbox implementation for years. Which is why I suspect a virtualbox + virtio_net combination bug, but I have no proof.

[sanitariu]

I have the same problem. Machine was working perfectly but no network until i restart.
So intel driver fixed this ?
If so i am going to switch too. If developers can not solve a problem 3 years virtio will be skipped from me
forever....

[Annihilannic]

I'm switching to Intel Pro/1000 MT Desktop Adapter today for the same reason... did this resolve the issue permanently for everyone here?

ETA: Should complete the picture: Windows 10 Pro 21H1 host, VirtualBox 6.1.22 r144080, Ubuntu 20.04.3 LTS guest, kernel 5.4.0-89-generic.