[migration/N] high cpu consumption
Posted: 2019/06/09 14:21:48
Dear Support,
We are running Centos and started seeing very frequently 'High CPU Load' especially process [ migrtion/n] eating up CPU anything
Version : CentOS release 6.9 (Final)
Kernal : 2.6.32-696.el6.x86_64
Load avg: 07:10:49 up 35 days, 15:48, 2 users, load average: 14.08, 13.26, 14.99
No of CPU : 40
Example,
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
107 root RT 0 0 0 0 S 44.6 0.0 1758:32 [migration/26]
131 root RT 0 0 0 0 S 39.0 0.0 1815:17 [migration/32]
19 root RT 0 0 0 0 S 37.7 0.0 2142:14 [migration/4]
91 root RT 0 0 0 0 S 36.7 0.0 1762:01 [migration/22]
27 root RT 0 0 0 0 R 34.4 0.0 2087:36 [migration/6]
115 root RT 0 0 0 0 S 31.1 0.0 1794:33 [migration/28]
51 root RT 0 0 0 0 S 27.5 0.0 1982:37 [migration/12]
23 root RT 0 0 0 0 S 20.3 0.0 638:17.39 [migration/5]
103 root RT 0 0 0 0 S 20.3 0.0 735:19.48 [migration/25]
99 root RT 0 0 0 0 S 16.1 0.0 1725:37 [migration/24]
59 root RT 0 0 0 0 S 8.8 0.0 2001:02 [migration/14]
147 root RT 0 0 0 0 S 3.6 0.0 1861:46 [migration/36]
Some time Load avg going upto 50+ though User process was not doing anything. We don't know if any scheduled kernel process doing this [ migration \] activity . If we looked at "Cpu(s)" seconds in the "top " then >98% of the CPU used by sys% only.
Is there known Bug in this Version and any Workaround ?
Thanks in advance,
Raj
We are running Centos and started seeing very frequently 'High CPU Load' especially process [ migrtion/n] eating up CPU anything
Version : CentOS release 6.9 (Final)
Kernal : 2.6.32-696.el6.x86_64
Load avg: 07:10:49 up 35 days, 15:48, 2 users, load average: 14.08, 13.26, 14.99
No of CPU : 40
Example,
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
107 root RT 0 0 0 0 S 44.6 0.0 1758:32 [migration/26]
131 root RT 0 0 0 0 S 39.0 0.0 1815:17 [migration/32]
19 root RT 0 0 0 0 S 37.7 0.0 2142:14 [migration/4]
91 root RT 0 0 0 0 S 36.7 0.0 1762:01 [migration/22]
27 root RT 0 0 0 0 R 34.4 0.0 2087:36 [migration/6]
115 root RT 0 0 0 0 S 31.1 0.0 1794:33 [migration/28]
51 root RT 0 0 0 0 S 27.5 0.0 1982:37 [migration/12]
23 root RT 0 0 0 0 S 20.3 0.0 638:17.39 [migration/5]
103 root RT 0 0 0 0 S 20.3 0.0 735:19.48 [migration/25]
99 root RT 0 0 0 0 S 16.1 0.0 1725:37 [migration/24]
59 root RT 0 0 0 0 S 8.8 0.0 2001:02 [migration/14]
147 root RT 0 0 0 0 S 3.6 0.0 1861:46 [migration/36]
Some time Load avg going upto 50+ though User process was not doing anything. We don't know if any scheduled kernel process doing this [ migration \] activity . If we looked at "Cpu(s)" seconds in the "top " then >98% of the CPU used by sys% only.
Is there known Bug in this Version and any Workaround ?
Thanks in advance,
Raj