Unsolved
This post is more than 5 years old
14 Posts
0
56324
January 22nd, 2013 07:00
CPUX: Package power limit notification
Seeing this on all my servers running CentOS 6.3. The server types are M520 and M610x.
I have seen a few posts regarding Ubuntu and this message but not CentOS.
I have also verified that none of our serves are exceeding 20 degres Celsius. So its not a temp issue.
All of our serves are running in Max Performance mode with both C States and C1E States enabled.
If anyone has any idea what the message means I would appreciate it.
Thanks,
Louis
0 events found
No Events found!
TheGreenHippo
1 Message
0
February 26th, 2013 21:00
CentOS 6.2 on a Dell PE520
kernel: CPU7: Package power limit notification (total events = 37932931)
kernel: CPU0: Package power limit notification (total events = 37931623)
kernel: CPU6: Package power limit notification (total events = 37931559)
kernel: CPU8: Package power limit notification (total events = 37936121)
kernel: CPU2: Package power limit notification (total events = 37936310)
kernel: CPU3: Package power limit notification (total events = 37938623)
kernel: CPU9: Package power limit notification (total events = 37938382)
kernel: CPU10: Package power limit notification (total events = 37938278)
kernel: CPU4: Package power limit notification (total events = 37938419)
kernel: CPU5: Package power limit notification (total events = 37938071)
kernel: CPU11: Package power limit notification (total events = 37938046)
kernel: CPU1: Package power limit notification (total events = 37932967)
This server has been in service for a while and all of a sudden late last week it started spitting out these errors several times per second.
It is causing my system to run at 25% on all CPUs even if it is not doing anything.
Any ideas why this happens
Probe Name : System Board Inlet Temp
Reading : 22.0 C
Probe Name : CPU1 Temp
Reading : 44.0 C
Rajesh Thareja
1 Message
0
August 28th, 2013 02:00
Hi,
We are facing the same issue with our Dell R620 server on linux kernel 2.6.32-358.11.1.el6.x86_64.
Did anyone found solution to this issue?
Thanks
Rajesh
bpwell
2 Posts
0
September 5th, 2013 23:00
See:
https://bugzilla.kernel.org/show_bug.cgi?id=36182
Messages are informational, non-critical.
Feature added to Sandy Bridge CPUs, required kernel update to correct.
Another 'fix' reported in this link is to change Server Profile to 'Performance', which will avoid the speed stepping that is used to reduce power consumption during low utilization, which is causing these messages.