This also affects Trusty.
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to seabios in Ubuntu.
https://bugs.launchpad.net/bugs/1181777
Title:
kvm: freeze a guest with a NIC PCI passthrough
To manage notifications about this bug go
We're also affected by this bug here. You can actually see my questioning on
libvirt-users about this:
https://www.redhat.com/archives/libvirt-users/2014-July/msg00134.html
Using rom bar='off'/ fixes it.
--
You received this bug notification because you are a member of Ubuntu
Server Team,
Status changed to 'Confirmed' because the bug affects multiple users.
** Changed in: seabios (Ubuntu)
Status: New = Confirmed
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to seabios in Ubuntu.
Status changed to 'Confirmed' because the bug affects multiple users.
** Changed in: seabios (Ubuntu Precise)
Status: New = Confirmed
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to seabios in Ubuntu.
This workaround unfortunately appear to totally break the network card
in the VM making it unusable. TG3 driver (which is supposed to handle
it) cannot properly communicate with it.
The kernel outputs the following lines:
Jul 30 13:32:03 carrier-commander kernel: [0.981658] tg3 :00:05.0:
Yes. It's works for me. But I can't find the message about ROM invalid:
grep -i -e invalid -e rom
[0.00] ACPI Warning: Invalid length for Pm1aControlBlock: 32, using
default 16 (20110623/tbfadt-610)
[0.00] ACPI Warning: Invalid length for Pm2ControlBlock: 32, using
default 8
Thanks, Nigel.
Alexander,
can you confirm whether adding rom bar='off'/ to your hostdev entry
works around the bug for you as well?
** No longer affects: ipxe (Ubuntu)
** No longer affects: ipxe (Ubuntu Precise)
--
You received this bug notification because you are a member of Ubuntu
Server
After getting the bug mail for this bug and reading it with interest, I
happened to run into this issue today myself and did some investigation
at a seabios level.
My original hit of the issue was with RHEL, but I reproduced it with the
same card (Realtek Semiconductor Co., Ltd. RTL8111/8168/8411
** Changed in: seabios (Ubuntu)
Importance: Undecided = High
** Changed in: ipxe (Ubuntu Precise)
Importance: Undecided = High
** Changed in: seabios (Ubuntu Precise)
Importance: Undecided = High
--
You received this bug notification because you are a member of Ubuntu
Server Team,
About #22 #25: https://bugzilla.kernel.org/show_bug.cgi?id=60850
** Bug watch added: Linux Kernel Bug Tracker #60850
http://bugzilla.kernel.org/show_bug.cgi?id=60850
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to ipxe in Ubuntu.
** Attachment added: 13.10 dmesg
https://bugs.launchpad.net/ubuntu/+source/ipxe/+bug/1181777/+attachment/3799505/+files/1310.txt
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to ipxe in Ubuntu.
** Attachment added: strace from 12.04
https://bugs.launchpad.net/ubuntu/+source/ipxe/+bug/1181777/+attachment/3799507/+files/strace1204.txt
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to ipxe in Ubuntu.
** Attachment added: 13.04 dmesg
https://bugs.launchpad.net/ubuntu/+source/ipxe/+bug/1181777/+attachment/3799504/+files/1304.txt
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to ipxe in Ubuntu.
Not solved yet. Please see the screenshot.
# dpkg -l | grep -e ipxe -e seab
ii ipxe
1.0.0+git-20130710.936134e-0ubuntu1.12.04.1~ppa1 PXE boot firmware
ii ipxe-qemu
1.0.0+git-20130710.936134e-0ubuntu1.12.04.1~ppa1 Virtual package to support
dpkg -l | grep ipxe
ii ipxe
1.0.0+git-20130710.936134e-0ubuntu1.12.04.1~ppa1 PXE boot firmware
ii ipxe-qemu
1.0.0+git-20130710.936134e-0ubuntu1.12.04.1~ppa1 Virtual package to support use
of kvm-ipxe with qemu
ii kvm-ipxe
Does it solve your problem if you grab and install
https://launchpad.net/ubuntu/+source/seabios/1.7.3-1/+build/4779258/+files/seabios_1.7.3-1_all.deb
on the precise system?
You said upstream qemu git head reproduced this - have you been able to
test the Ubuntu raring and saucy (13.04 and 13.10)
Hi,
not solved yet.
PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND
10710 root 20 0 805m 543m
I don't believe it can be the qemu-kvm package, since upstream qemu git
HEAD didn't fix it, while it wasn't broken on centos.
Can you show 'dpkg -l | grep ipxe' on the system from comment #19?
If the new ipxe didn't solve it, then my next guess would be seabios...
--
You received this bug
I've tried some more setups like having a virtio nic on virbr0 and a
second e1000 nic on br0, where only br0 was on a network with a pxeboot
server - but wasn't able to reproduce. So it appears to be purely PCI
passthrough.
I'm building a package (in the same ppa) which is the saucy ipxe, built
Hi,
the backported saucy ipxe packages have finished building in my ppa. Please
download
them using
wget
https://launchpad.net/~serge-hallyn/+archive/virt/+files/kvm-ipxe_1.0.0%2Bgit-20130710.936134e-0ubuntu1.12.04.1%7Eppa1_all.deb
wget
Thank you Serge!
The issue not solved.
# aptitude show ipxe | grep -i version
Version: 1.0.0+git-3.55f6c88-0ubuntu2.12.04.1~ppa1
# aptitude show kvm-ipxe | grep -i version
Version: 1.0.0+git-3.55f6c88-0ubuntu2.12.04.1~ppa1
Strace log:
---cut
select(16, [5 6 15], [], [], {0, 987950}) = 1
** Also affects: seabios (Ubuntu Precise)
Importance: Undecided
Status: New
** Also affects: ipxe (Ubuntu Precise)
Importance: Undecided
Status: New
** Changed in: ipxe (Ubuntu)
Status: Triaged = Fix Released
--
You received this bug notification because you are a
Thanks for testing. That's a shame. I had hoped the new driver for
your card would fix it. I'll try some more testing to see if this is a
more general problem that I can somehow reproduce.
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is
Interesting to note the original bug reporter also had BCM5719
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to ipxe in Ubuntu.
https://bugs.launchpad.net/bugs/1181777
Title:
kvm: freeze a guest with a NIC PCI passthrough
To manage
An ipxe package is building in ppa:serge-hallyn/virt which has a newer
broadcom driver. See https://launchpad.net/~serge-hallyn/+archive/virt
for details.
When it builds, please grab kvm-
ipxe_1.0.0+git-3.55f6c88-0ubuntu2.12.04.1~ppa1_all.deb (and, if you have
ipxe itself installed, also
** Also affects: seabios (Ubuntu)
Importance: Undecided
Status: New
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to ipxe in Ubuntu.
https://bugs.launchpad.net/bugs/1181777
Title:
kvm: freeze a guest with a NIC PCI
I have the same problem on a ProLiant ML350p Gen8 when trying to
install (using virt-install with PCI passthrough --host-device=etc)
using the integrated Broadcom NetXtreme BCM5719.
I also have an Intel dual port I350 on the system and with it everything
is works as expected (virt-install
Thanks - so we can mark this fix released once the nomination for
precise is accepted. We'll have to look through the git tree for the
fix, but great to know the fix is in there.
** Package changed: qemu-kvm (Ubuntu) = ipxe (Ubuntu)
** Changed in: ipxe (Ubuntu)
Status: New = Triaged
--
I tried qemu in saucy. No issue found. IPXE works fine.
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to qemu-kvm in Ubuntu.
https://bugs.launchpad.net/bugs/1181777
Title:
kvm: freeze a guest with a NIC PCI passthrough
To manage
Thanks! So it sounds like the problem is actually in ipxe.
Saucy has a new ipxe (from 9 days ago). However simply installing it
alongside precise's qemu-kvm probably won't work.
(I do see a few commits in the git tree which could perhaps fix this)
Is it possible for you to test this with
Yes, I have a possibility.
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to qemu-kvm in Ubuntu.
https://bugs.launchpad.net/bugs/1181777
Title:
kvm: freeze a guest with a NIC PCI passthrough
To manage notifications about this bug go
I tried upstream qemu. But the issue not solved yet.
On Centos it work fine. qemu in Centos use gPXE instead iPXE. Please see
screenshot.
** Attachment added: Screenshot of QEMU.png
Would you be able to test this with upstream qemu?
git clone git://git.qemu.org/qemu.git
cd qemu
./configure --target-list=x86_64-softmmu
make -j4
cd x86_64-softmmu
./qemu-system-x86_64 -enable-kvm -m 512 -device pci-assign,host=03:00.1 -boot d
If that works, then we just need to look for the
Start
qemu-system-x86_64 -m 512 -device pci-assign,host=03:00.1 -boot d
not solve the issue.
PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND
40062 root 20 0 805m 543m 6724 S 99 0.6
Unfortunately I don't have any hardware to reproduce this with.
Could you please:
1. Attach /var/log/libvirt/qemu/w.log
2. Try running this vm with kvm by hand as per
http://www.linux-kvm.org/page/How_to_assign_devices_with_VT-d_in_KVM ?
(i.e. the kvm command in step 6, something like kvm
** Changed in: qemu-kvm (Ubuntu)
Status: Incomplete = New
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to qemu-kvm in Ubuntu.
https://bugs.launchpad.net/bugs/1181777
Title:
kvm: freeze a guest with a NIC PCI passthrough
To
Thanks for reporting this bug.
In order to help me reproduce it, could you please tell us in detail how
you are attaching the NIC, and what OS/release is running inside the
guest?
** Changed in: qemu-kvm (Ubuntu)
Importance: Undecided = High
** Changed in: qemu-kvm (Ubuntu)
Status:
Hello.
This is the step by step history of commands. The domain w is not rinnung.
# virsh list --all
Id Name State
--
- wshut off
There are all NICs on the host system
# virsh nodedev-list --tree | grep -A 18 '+-
38 matches
Mail list logo
