Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 1286462

Summary: Vdsm daemon failed to start, because incorrect cpu affinity
Product: [oVirt] vdsm Reporter: Artyom <alukiano>
Component: GeneralAssignee: Francesco Romani <fromani>
Status: CLOSED CURRENTRELEASE QA Contact: Artyom <alukiano>
Severity: high Docs Contact:
Priority: urgent    
Version: 4.17.11CC: bugs, fromani, gklein, hannsj_uhl, mgoldboi, michal.skrivanek
Target Milestone: ovirt-3.6.1Keywords: Regression, Triaged
Target Release: 4.17.12Flags: rule-engine: ovirt-3.6.z+
rule-engine: blocker+
mgoldboi: planning_ack+
michal.skrivanek: devel_ack+
mavital: testing_ack+
Hardware: All   
OS: Linux   
Whiteboard: virt
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2015-12-16 12:19:31 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: Virt RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 1201513    
Attachments:
Description Flags
vdsm log none

Description Artyom 2015-11-29 18:56:42 UTC
Created attachment 1100263 [details]
vdsm log

Description of problem:
Vdsm daemon failed to start on hosts without cpu under number 1, because incorrect cpu affinity with traceback:
Traceback (most recent call last):
  File "/usr/share/vdsm/vdsm", line 166, in run
    __set_cpu_affinity()
  File "/usr/share/vdsm/vdsm", line 280, in __set_cpu_affinity
    taskset.set(os.getpid(), cpu_set, all_tasks=True)
  File "/usr/lib/python2.7/site-packages/vdsm/taskset.py", line 82, in set
    raise Error(rc, out, err)
Error: Process failed with rc=1 out=["pid 129019's current affinity list: 8,16,24,32,40,48,56,64,72,80,88,96,104,112,120,128,136,144,152"] err=["taskset: failed to set pid 129019's affinity: Invalid argument"]


Version-Release number of selected component (if applicable):
vdsm-4.17.11-0.el7ev.noarch

How reproducible:
Always

Steps to Reproduce:
1. Start vdsm daemon on host that not have cpu under number 1
# cat /proc/cpuinfo 
processor       : 8
cpu             : POWER8E (raw), altivec supported
clock           : 3690.000000MHz
revision        : 2.1 (pvr 004b 0201)

processor       : 16
cpu             : POWER8E (raw), altivec supported
clock           : 3690.000000MHz
revision        : 2.1 (pvr 004b 0201)
....
2.
3.

Actual results:
vdsm daemon failed to start with above exception

Expected results:
vdsm succeed to start

Additional info:
problem in vdsm config file
('cpu_affinity', '1',
            'Comma separated whitelist of CPU cores on which VDSM is allowed '
            'to run. The default is "1", meaning VDSM can be scheduled by '
            ' the OS to run on the second core of the system. '
            'Valid examples: "1", "0,1", "0,2,3"')
if I change it to 'cpu_affinity', '', all works fine

Comment 1 Gil Klein 2015-11-29 19:21:04 UTC
Seems to be related the enablement of BZ #1279431

Comment 2 Michal Skrivanek 2015-11-29 19:27:52 UTC
Decreasing Severity as there is a configuration workaround to pin to a different cpu or disable it altogether

This is not ppc specific, any platform with offline cpu1 would demonstrate the same. We should have go with 0

Comment 3 Francesco Romani 2015-11-30 20:40:23 UTC
patches merged on both master and 3.6 branch -> MODIFIED

Comment 4 Sandro Bonazzola 2015-12-01 15:25:36 UTC
This bug is referenced in 4.17.12 git log and has target release unset.
Please check

Comment 5 Artyom 2015-12-07 14:57:14 UTC
Verified on vdsm-4.17.12-0.el7ev.noarch

Comment 6 Sandro Bonazzola 2015-12-16 12:19:31 UTC
According to verification status and target milestone this issue should be fixed in oVirt 3.6.1. Closing current release.