Bug 1716908

Summary:	'cannot set CPU affinity' error when starting guest
Product:	Red Hat Enterprise Linux 8	Reporter:	Andrea Bolognani <abologna>
Component:	libvirt	Assignee:	Andrea Bolognani <abologna>
Status:	CLOSED ERRATA	QA Contact:	jiyan <jiyan>
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	8.1	CC:	jdenemar, jiyan, rbalakri, yalzhang
Target Milestone:	rc
Target Release:	8.1
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:	libvirt-4.5.0-25.el8	Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2019-11-05 20:50:11 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:	1716943
Bug Blocks:

Description Andrea Bolognani 2019-06-04 11:36:34 UTC

This bug was initially created as a copy of Bug #1703661

I am copying this bug because: 

  RHEL 8.1 shouldn't contain bugs that have been fixed in RHEL 7.7.

Description of problem:
When numatune placement auto is set, guest can not be started.

Version-Release number of selected component (if applicable):
# rpm -q libvirt qemu-kvm-rhev kernel
libvirt-4.5.0-15.virtcov.el7.ppc64le
qemu-kvm-rhev-2.12.0-27.el7.ppc64le
kernel-3.10.0-1034.el7.ppc64le

How reproducible:
100%

Steps to Reproduce:
1. To define a guest with:
<numatune><memory mode="strict" placement="auto" /></numatune>
2. Try to start it

Actual results:
# virsh start test1
error: Failed to start domain test1
error: invalid argument: Failed to parse bitmap ''

Expected results:
Domain test1 started

Additional info:
1. It looks like this bug: https://bugzilla.redhat.com/show_bug.cgi?id=1233023
2. When I use the compose RHEL-7.7-20190403.0(libvirt-4.5.0-11), the test pass. And then I update the libvirt to the latest(4.5.0-15), it will reproduce the problem. But when I downgrade it to 4.5.0-11, it still could be reproduced.
3. When I use the compose RHEL-7.7-20190424.0(libvirt-4.5.0-14), it could be reproduced

+++ Comment 19 (Andrea Bolognani)

(In reply to Junxiang Li from comment #0)
> Steps to Reproduce:
> 1. To define a guest with:
> <numatune><memory mode="strict" placement="auto" /></numatune>
> 2. Try to start it
> 
> Actual results:
> # virsh start test1
> error: Failed to start domain test1
> error: invalid argument: Failed to parse bitmap ''

One thing that I apparently forgot to point out is that I never
managed to reproduce those exact symptoms: what I got instead was
along the lines of

  # virsh start guest
  error: Failed to start domain guest
  error: cannot set CPU affinity on process 40055: Invalid argument

Comment 3 Junxiang Li 2019-06-28 02:25:33 UTC

*** Bug 1724408 has been marked as a duplicate of this bug. ***

Comment 4 jiyan 2019-07-03 09:04:58 UTC

Hi I am trying to reproduce this issue on libvirt-4.5.0-23.module+el8.1.0+2983+b2ae9c0a.x86_64 using the way in https://bugzilla.redhat.com/show_bug.cgi?id=1703661#c28 for many times.
But I failed, could you please check whether this issue exists in this version? thank you.


Version:
libvirt-4.5.0-23.module+el8.1.0+2983+b2ae9c0a.x86_64
qemu-kvm-2.12.0-80.module+el8.1.0+3572+48154135.x86_64
kernel-4.18.0-109.el8.x86_64

Steps:
# virsh domstate test
shut off

# virsh dumpxml test --inactive |grep "<vcpu" -A3
  <vcpu placement='static'>1</vcpu>
  <numatune>
    <memory mode='strict' nodeset='1'/>
  </numatune>

# cat /sys/devices/system/cpu/cpu1/online
0

# numactl --hard
available: 2 nodes (0-1)
node 0 cpus: 0 2 3 4 5
node 0 size: 15932 MB
node 0 free: 15436 MB
node 1 cpus: 6 7 8 9 10 11
node 1 size: 16101 MB
node 1 free: 15560 MB
node distances:
node   0   1 
  0:  10  20 
  1:  20  10 

# virsh start test
Domain test started

# virsh dumpxml test |grep "<vcpu" -A3
  <vcpu placement='static'>1</vcpu>
  <numatune>
    <memory mode='strict' nodeset='1'/>
  </numatune>

VM starts successfully here.

Comment 6 Andrea Bolognani 2019-07-03 11:08:21 UTC

(In reply to jiyan from comment #4)
> Hi I am trying to reproduce this issue on
> libvirt-4.5.0-23.module+el8.1.0+2983+b2ae9c0a.x86_64 using the way in
> https://bugzilla.redhat.com/show_bug.cgi?id=1703661#c28 for many times.
> But I failed, could you please check whether this issue exists in this
> version? thank you.
> 
> Version:
> libvirt-4.5.0-23.module+el8.1.0+2983+b2ae9c0a.x86_64
> qemu-kvm-2.12.0-80.module+el8.1.0+3572+48154135.x86_64
> kernel-4.18.0-109.el8.x86_64

For RHEL 7, the issue was introduced in libvirt-4.5.0-13.el7 while
addressing Bug 1695434, and subsequently fixed in libvirt-4.5.0-20.el7
(tracked by Bug 1703661): any libvirt version in between those two can
be used to reproduce the incorrect behavior.

For RHEL 8, the corresponding trackers are Bug 1716943 and
Bug 1716908 respectively, both of which have been addressed in
libvirt-4.5.0-25.el8, so there's no libvirt version you can use to
reproduce the incorrect behavior.

Hope that cleared up the situation! I agree it's pretty confusing, and
in fact I had to do some digging of my own before I could confidently
write the above :)

Comment 7 jiyan 2019-07-05 06:41:19 UTC

Verified this bug on libvirt-4.5.0-28.module+el8.1.0+3531+2918145b.x86_64

Version:
libvirt-4.5.0-28.module+el8.1.0+3531+2918145b.x86_64
qemu-kvm-2.12.0-80.module+el8.1.0+3572+48154135.x86_64
kernel-4.18.0-109.el8.x86_64

Steps:
# numactl --hard
available: 2 nodes (0-1)
node 0 cpus: 0 2 3 4 5
node 0 size: 15932 MB
node 0 free: 15241 MB
node 1 cpus: 6 7 8 9 10 11
node 1 size: 16101 MB
node 1 free: 15408 MB
node distances:
node   0   1 
  0:  10  20 
  1:  20  10 

 virsh domstate test
shut off

# virsh dumpxml test --inactive |grep "<vcpu" -A3
  <vcpu placement='static'>1</vcpu>
  <numatune>
    <memory mode='strict' nodeset='1'/>
  </numatune>

# echo 0 > /sys/devices/system/cpu/cpu1/online 

# cat /sys/devices/system/cpu/cpu1/online 
0

# virsh start test
Domain test started

# virsh dumpxml test |grep "<vcpu" -A3
  <vcpu placement='static'>1</vcpu>
  <numatune>
    <memory mode='strict' nodeset='1'/>
  </numatune>

The test result is asexpected, move this bug to be verified.

Comment 9 errata-xmlrpc 2019-11-05 20:50:11 UTC

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHSA-2019:3345