Bug 749710

Summary: irqbalance stops after less than a minute of running
Product: [Fedora] Fedora Reporter: Stan King <stanley.king>
Component: irqbalanceAssignee: Neil Horman <nhorman>
Status: CLOSED NOTABUG QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 16CC: anton, nhorman
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2011-10-28 14:50:31 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Attachments:
Description Flags
dmesg output from computer with the described failure none

Description Stan King 2011-10-28 02:16:00 UTC
Created attachment 530597 [details]
dmesg output from computer with the described failure

Description of problem:

The irqbalance process is exiting after a short period of time.
I would expect it to last longer.


Version-Release number of selected component (if applicable):

irqbalance-1.0-3.fc16.i686

How reproducible:

This occurs on 5 of my 7 systems.

Steps to Reproduce:
1. Boot system
2. Periodically run "ps -ef | grep irqbalance" to check for irqbalance running
3.
  
Actual results:

irqbalance exits very soon

Expected results:

irqbalance should remain running


Additional info:

I've added this machine's dmesg output in hopes that may be useful.

Here is the output from "irqbalance --debug",
passed through "fold -70" in order to wrap a long line.

This machine seems not NUMA capable.
Package 0:  numa_node is -1 cpu mask is 00000003 (load 0)
        Cache domain 0:  numa_node is -1 cpu mask is 00000003  (load 0
) 
                CPU number 0  numa_node is -1 (load 0)
                CPU number 1  numa_node is -1 (load 0)
Adding IRQ 40 to database
Adding IRQ 41 to database
Adding IRQ 42 to database
Adding IRQ 21 to database
Adding IRQ 22 to database
Adding IRQ 18 to database
Adding IRQ 23 to database
DROPPING DUPLICATE ENTRY FOR IRQ 21 on path /sys/bus/pci/devices/0000:
00:1d.7
DROPPING DUPLICATE ENTRY FOR IRQ 23 on path /sys/bus/pci/devices/0000:
00:1e.2
Adding IRQ 16 to database
Adding IRQ 20 to database
Adding IRQ 17 to database
DROPPING DUPLICATE ENTRY FOR IRQ 16 on path /sys/bus/pci/devices/0000:
01:00.0
DROPPING DUPLICATE ENTRY FOR IRQ 16 on path /sys/bus/pci/devices/0000:
01:00.1
DROPPING DUPLICATE ENTRY FOR IRQ 16 on path /sys/bus/pci/devices/0000:
02:00.0
DROPPING DUPLICATE ENTRY FOR IRQ 18 on path /sys/bus/pci/devices/0000:
04:02.0
DROPPING DUPLICATE ENTRY FOR IRQ 16 on path /sys/bus/pci/devices/0000:
05:08.0
Adding IRQ 11 to database
DROPPING DUPLICATE ENTRY FOR IRQ 17 on path /sys/bus/pci/devices/0000:
05:09.0
Adding IRQ 10 to database
DROPPING DUPLICATE ENTRY FOR IRQ 18 on path /sys/bus/pci/devices/0000:
05:0a.0
Adding IRQ 3 to database
Adding IRQ 19 to database
Adding IRQ 9 to database
NUMA NODE NUMBER: -1
LOCAL CPU MASK: ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,
ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,fffffff
f,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,fffff
fff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,fff
fffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,f
fffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff
,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffff
ff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffff
ffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ff
ffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,
ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,fffffff
f,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,fffff
fff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,fff
fffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,f
fffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff
,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffff
ff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff




----------------------------------------------------------------------
-------
Package 0:  numa_node is -1 cpu mask is 00000003 (load 0)
        Cache domain 0:  numa_node is -1 cpu mask is 00000003  (load 0
) 
                CPU number 0  numa_node is -1 (load 0)
                CPU number 1  numa_node is -1 (load 0)

Comment 1 Neil Horman 2011-10-28 14:50:31 UTC
This is working as designed.  Given that the system in question here contains only a single cache domain encompassing both cpus, there is very little purpose to migrating irqs between cpus as they will hit the same cache regardless.  So irqbalance balances irqs once, and exits, to save on cpu time.