"CPU#0 Tempature above threshold"

I am running Arch Linux with the ArchLinux supplied 2.6.2 kernel. When I try to compile a new kernel, I get the message repeatedly on every terminal, and CPU usage shoots up to 100% (according to IceWM).
Here is what the kernel tells me:
CPU#0: Tempature above threshold
CPU#0: Running in modulated clock mode
I see those messages repediately (like 3 times a second), and every once and a while I get compiler messages on what is being compiled. It seems to be compiling fine, but those messages are on every terminal, making it unable to do anything! What's the problem! I need help!

I cleaned out the computer (it was pretty hot), and it still displayed the message, but it wasn't as bad. Maybe they should cut down on those messages, and say when they start and stop, but not every blinking second!

Similar Messages

CPUx: Temperature above threshold, cpu clock throttled

My cousin (Windows) computer suicided... it stopped
booting without any apparent reason.
He asked my help and we stored every important data in an
external drive using an arch linux installed in a usb key
(thanks to every wiki writer).
I thought it could be an hardware problem so I started
stress with with a command line like:
stress --cpu 2 --io 1 --vm 1 --hdd 1
(pwd was in a Windows mounted partition)
it is already one hour the program goes. I got the error
of the title few times:
CPUx: Temperature above threshold, cpu clock throttled
for both CPUs. I thought it was normal, after all
the cpu might overheat under stress and it slows down
to protect itself...
But suddenly I got a doubt, it is really the CPU that
slows down itself (so it is an hardware feature) or it is
linux that slows it down (so it is a kernel feature)? In
this second case do anyone knows if Windows do anything
similar?
Secondly in the error log I can see:
EDID checksum is invalid, remainder is 210
Raw EDID:
<hex dump omitted>
I sought in the Internet, but I could not understand what
it means nor if it is dangerous... can anyone help me?
Thanks

The BIOS is usually responsible for setting thresholds and has settings to halt the computer when those thresholds are too high.
Since the program is testing stress, I'm guessing the software is throttling the CPU so the system does not shutoff and the test can still process.
I would highly suggest checking your BIOS's tempertature thresholds, as well as the fan speed settings (BIOS or software).
I would also find out what a normal temperature for the CPU is and do some testing on what different loads do to the temperature.

Asa 5520 | CPU Spikes above 90% for 5 Seconds then it went down to 35 - 40%

Hi Experts
i have ASA 5520 , some times the CPU Spikes above 90% for 5 Seconds then it went down to 35 - 40% ,but what happened when it shoots to 90% all connection go through this ASA gets drops then i receive complain since application traffic traverse this FW ,so how can i see the CPU History for last 1 week , i need to see the times and the date WHEN THE CPU went above 90% for 4 seconds
thanks
jamil

Hi Ibrahim,
from he ASA you would not be able to pull data for last one week, you would need an SNMP server or anyother monitoring server for pulling reports.
For high CPU issues, I would suggest opening a TAC case for it, since it requires detailed investigation and access to your box. Just as in initial pointer, these outputs might help you identifying the cause:
show proc cpu-usage
show traffic
show conn count
show interface
Remember these outputs are useful only if they are from the time of the issue, check what traffic is hogging the CPU and if the ASA is being overwhelmed or not.
Thanks,
Varun Rao
Security Team,
Cisco TAC

WiSM Issue - More than three APs working at the power above threshold

According to the RRM document: http://www.cisco.com/en/US/tech/tk722/tk809/technologies_tech_note09186a008072c759.shtml , for each AP, the 3rd loudest AP should be heard at the power equal or lower than the threshold (default -65dbm or -70dbm according to different codes). But in our enviroment, I can see it is not true. Many APs have more than three neighbours heard at above threshold. Is it true in your enviroment as well?
Thanks!
Zhenning

It's the way I have deployed the network. When the number of access point increases (though in diff channel) chances of interference increases leading to decreased performance. So Adjust the power settings and the number of access points accordingly.

Swap space used is above threshold of 90

Hi all,
i am facing an issues in linux box which is Swap space used is above threshold of 90
[oracle@gbaheovl21 scripts]$ free
             total       used       free     shared    buffers     cached
Mem:       8388608    8354400      34208          0      32748    4354904
-/+ buffers/cache:    3966748    4421860
Swap:      6257992    5816940     441052
scripts]$ sar -r
07:00:01    kbmemfree kbmemused %memused kbbuffers kbcached kbswpfree kbswpused %swpused kbswpcad
07:10:01        15868   8372740     99.81     16348   4356644    479720   5778272     92.33    767516
07:20:01        93232   8295376     98.89     17624   4304864    433200   5824792     93.08    773240
07:30:01        29092   8359516     99.65     22664   4328688    442496   5815496     92.93    800512
07:40:01        39700   8348908     99.53     28584   4348424    441392   5816600     92.95    768592
Average:        57576   8331032     99.31     56029   4292652    349858   5908134     94.41    775742
Is their any fix to this issue with out reboot ? please any one help me on this ?

Hi ,
sorry for that , i am pasting the info below.
Linux venkat.gb.com 2.6.18-238.el5xen #1 SMP Sun Dec 19 14:42:02 EST 2010 x86_64 x86_64 x86_64 GNU/Linux
it is a Oracle virtual linux and here few databases are running
[root@ ~]# sysctl vm.swappiness
vm.swappiness = 60
[root@~]# df /dev/shm
Filesystem           1K-blocks      Used Available Use% Mounted on
tmpfs                  4194304    249220   3945084   6% /dev/shm
[root@ ~]# ipcs -um
------ Shared Memory Status --------
segments allocated 31
pages allocated 2099310
pages resident 1003699
pages swapped   881584
Swap performance: 0 attempts     0 successes
[root@ ~]# cat /proc/meminfo
MemTotal:      8388608 kB
MemFree:         29556 kB
Buffers:         39224 kB
Cached:        4458580 kB
SwapCached:     625964 kB
Active:        6728908 kB
Inactive:       548592 kB
HighTotal:           0 kB
HighFree:            0 kB
LowTotal:      8388608 kB
LowFree:         29556 kB
SwapTotal:     6257992 kB
SwapFree:       512504 kB
Dirty:            3360 kB
Writeback:           0 kB
AnonPages:     2442472 kB
Mapped:        3943420 kB
Slab:           256204 kB
PageTables:     525464 kB
NFS_Unstable:        0 kB
Bounce:              0 kB
CommitLimit: 10452296 kB
Committed_AS: 21157240 kB
VmallocTotal: 34359738367 kB
VmallocUsed:     34920 kB
VmallocChunk: 34359703427 kB
thanks
Venkat

HubTransport UnHealty - Total.Shadow.Queue.Length.Above.Threshold.Monitor - What to check?

Hi To all,
looking at Exchange 2013 ServerHealth i have the "HubTransport" in unhealth state related to this
item:
Total.Shadow.Queue.Length.Above.Threshold.Monitor
I cannot find more information about this issue...
Many thanks for help! :)
r.

Hi,
Found nothing in the public resource, neither.
Is there any error/warning/information in the event viewer?
Please also check the detailed error message in the Monitor if it is possible.
Did this cause other issues, some like mailflow issue etc.?
If everything going well, I suggest disable the Alert.
Thanks
Mavis
Mavis Huang
TechNet Community Support

Cpu1 above threshold?

From what I've searched on the forums it might be my graphics card? I'm running a dell e1405 with an intel 945GM graphics card using the xf86-video-intel driver. I was unaware of this error that is apparent now.. "CPU1: Temperature above thresold, cpu clock throttled (total events = 220499)" because I only see it when I switch out of X into teminal... My fan is running constantly and only using for the most part under 10% on both CPUs but you can feel the heat off it as if I was playing a game.
What should I do?
EDIT:
I just did ran dell's diagnostic software and nothing was wrong with the CPU.. Also looked at the BIOS to see if it showed my CPU temp. anywhere but didn't so I assumed that it just didn't keep track of the temperature... but that can't be right because my fan sped up immediatly when the system restarted and during the diagnostics it eventually slowed down.. now it doesn't display the error?
Last edited by Kazuma (2010-01-24 20:46:37)

Hey, okay i had this issue too but it was nothing to do with the cpu i don't think. I had it for ages but nothing happened until my machine went crazy and started locking up and a bunch of other issues. I took my computer apart cleaned the fan intake for the cpu and noticed that my graphics card was baking.
I never turned on the power scaling options for my nvidia graphics card after using the official drivers it was baking everything in my case for ages. Turned them on and I haven't had any messages since and everything works waaaaay better because i turned on some other optimisation stuff for the card as well. If you have a big graphics card, check the options make sure power scaling for heat is on should be fine.

Lightroom CC burning through RAM and pushing up CPU usage above 400% (MacBook Pro Retina, 15", 16GB RAM, Top Spec)

So as the new version of LR was released I've upgraded to CC. LR CC was running really slow compared to LR 5.6 so I disabled graphics acceleration as recommended by a few users here on the Adobe forums.
Now I'm having a problem that whenever I start editing in the basic tab on the develop module, LR CC sucks all my RAM (all 16GB according to memory clean/activity monitor) and my CPU goes into overdrive (I clocked it over 400% on average using Activity monitor app). The fans start spinning up and the laptop starts heating up. As soon as I close the develop module it goes back to normal. I pretty much can't edit anything in LR at the moment. Photoshop CC in contrast runs like a rocket ship, with none of the issues described above.
Anyone else experiencing this?
I've had a look at limiting the RAM usage for Lightroom but there doesn't seem to be a menu option for it.

She's flying! Thats on the down cycle too. I couldn't catch it fast enough but it was over and above 600 at one point.

ISE Config Backup Failure - Data filesystem full above threshold

Hi,
Both the config and operational backups were working until earlier last month. Now the config backup is failing with the following error. No configuration or repository settings were changed.
ISE 1.2.0.899 Patch 8 - Clustered with persona Node 1 = PAN, SMN, PSN .... Node 2 = SAN, PMN, PSN
CLI history says the same:
The local repository (disk:/) is looking good. The "/" filesystem is taking 77% space.
Although it may not be relevant. Data Purging is set to 30 days in the GUI and Operations -> Reports -> Data Purging Audit indicates its running daily with success i.e. threshold_space = 80GB, used_space = 3GB.
Is there a way to clean "/" filesystem ? It is filling up by roughly 1% every 5 days ? Note: the same on Node 2 is only 24% full.
Any ideas on how to get the config backup issue resolved ?
P.S. If images don't appear inline, please see the attachment
Thanks,
Rick.

922963 wrote:
Hi JK,
Thanks for response. Yes, I am worried that it may not be enough. How about if I increase memory to 32GB, ie. I have two servers, both with 32GB? Will it be sufficient in case of one physical server fail for 8GB data?
What is the point in having the 3rd physical box if two boxes have enough memory/capacity? You know, we need to pay licence according to no of CPU.
thanks,
HenryHi Henry,
actually the recommended minimum number of physical boxes is 4 so that the witness protocol participants can all be on separate machines.
But a minimum of 3 is highly recommended for a number of reasons related to partitioning:
1. If you have only 2, then you are much more vulnerable to split brain scenarios (should for some reason the two servers not be able to communicate with each other, it is harder to decide which half should be the winner). In short how do you decide which box is unable to communicate with the rest of the cluster if there are only 2 boxes?
2. You can't ensure a balanced and also machine safe partition distribution if you have a mismatching number of nodes on only 2 boxes. It would either be balanced or machine safe, but you can't get both at the same time. And you will either have mismatching number of nodes at startup or have mismatching number of nodes after one node failure.
Best regards,
Robert

Can't get the CPU utilization above 52%

Hi all,
I have an Intel i7 950 CPU based PC with Windows 10 TP on it which, no matter what, I cannot bring to 100% CPU utilization. While taking the attached screenshot, I was running prime95.exe.
I have tested the same scenario on the same computer running Windows 8/8.1 and got the same result. However, when doing the same thing under Windows 7, CPU utilization went up to 100%.
Anybody has any idea why this happens? Is it a different default setting somewhere (probably in registry)?
Since I have this fast machine I would like to be able to use it at its full strength - the way things are now, my old Quad core 2 GHz PC is MUCH faster than this 8 core 3 GHz one (as I measured in several tests consisting of transcoding videos).
Thanks for any pointers!

Hi VikTak-9b,
How did you get the CPU utilization information ?Are you getting the information from the task manager ?
"I have tested the same scenario on the same computer running Windows 8/8.1 and got the same result. However, when doing the same thing under Windows 7, CPU utilization went up to 100%.
Anybody has any idea why this happens? Is it a different default setting somewhere (probably in registry)?"
I am a little confused here .According to my understanding ,the number is smaller the performance of the machine is higher when the other conditions are the same .
Here is a link may be useful for us to understand the CPU Utilization for Performance Analysis
Interpreting CPU Utilization for Performance Analysis
http://blogs.technet.com/b/winserverperformance/archive/2009/08/06/interpreting-cpu-utilization-for-performance-analysis.aspx
Best regards
Please remember to mark the replies as answers if they help, and unmark the answers if they provide no help. If you have feedback for TechNet Support, contact [email protected]

Windows 8 lagging and slow response but CPU never above 33%

Hello, I have an up-to-date windows 8.1 running on a dell inspiron 1520 laptop with a T7100 cpu with 4gb memory and an ssd hard drive. My computer frequently lags even with just browsing with IE11 - I can even out type it sometimes - however
when I look at the task manager during these periods the cpu utilization never exceeds 33% and the memory utilization never exceeds 66%. Disk and network utilization are always low. I suspect the bottleneck is the cpu and I wonder why the utilization seems
to max at 33% ?

Here are the 2 pages of listings when I search Firefox on "this computer"

Cpu tempature help needed

I am using bios 2.1,and not getting the corect tempature reading from core centre,After fitting a Artic freezer cooler 64 pro,On windows startup it is 53,then down to 44,45,46.
Thanks from dodio

Hi Dodio,
Which motherboard are you speak of ? I have crazy sensor too on a Neo2 Plat ..

Optical Receive Power Warning - Operating above threshold

As we know,
we cannot configure the optical transmit power of the SFP. Though we can check the receive power level received by peer through the command:
show interface transceiver detail
However, logs have been displaying:
Aug 20 06:45:14.434: %SFF8472-5-THRESHOLD_VIOLATION: Gi1/0/25: Rx power high warning; Operating value: -1.7 dBm, Threshold value: -3.0 dBm.
Checking it:
Optical High Alarm High Warn Low Warn Low Alarm
Receive Power Threshold Threshold Threshold Threshold
Port (dBm) (dBm) (dBm) (dBm) (dBm)
Gi1/0/25 -1.7 + 1.0 -3.0 -19.0 -23.0
WIll this have a big impact on my switch? Especially the SFP module
Factors like damage, etc.
Also, how will it affect network performance, both the health and operation of my devices
Your ideas are greatly appreciated (and if you can back it up with docs, it'd would be helpful too! ^_^)

It seems your distant end is sending a relatively high output power and/or you have a very short fiber connection with very little loss. Since you are only in the 'warn' range you are not likely to cause any damage to the receiving SFP or affect network operation adversely. You do have the option of adding an optical attenuator to the link to make the operating range more within specification.

Max voltages and cpu tempatures

hi all,
Iv recently updated my pc for gaming and was wondering what the max voltages and cpu temps are as i have pc alert 4 and when i was playing pc alert minimized the game and beeped at me i dont think it was my processor as the max was set to 75*c and was only running at 49*c.
the thing is how am i ment to know whats what as the voltages are not explained from the left thay are about 1.5v, 3.5v, 5.5v and 12.5v. the game im running is far cry 2
Any views whould be great.
Thanks chris
my pc specs are:
2.9ghz 5600 x2 athlon cpu
4gb 800mhz ram
xp pro
msi k9ngm bios v 3.1
coloursit 750watt pcu

Remove PC Alert as I don't think it even reads your board properly.
SpeedFan is way better.

Trigger above threshold value

Hi
I'm trying to trigger a data save with multiple input channels when one
of my input signals exceeds a given level for the first time. The
data save should only end when a stop button is activated - i.e. not
re-start each time the threshold level is exceeded. I have been trying
to do this by connecting a Curve Control Block to the start input of my
Save block, which is set to retriggerabe, but my measurement then
restarts each time the given level is exceeded.
Do I need to use a different method, or just change some parameters?
Thanks,
Helen

Hello Helen,
I made some small modifications to your scheme.
Firstly, you have to connect the "Condition1" block to the systemtact
as well. Otherwise it wont be stopped by the stoptrigger and the whole
scheme can not be terminated with the button.
Secondly, the lamp indicator is a bit trcky to use. This is because it
is not, as you might expect, simply a boolean indicator. Actually it
can display more than 2 states - each state in a different color.
You have to edit the value range for the incoming data and chose which
values are represented in which colors. (Dialog Signaldarstellung ->
Darstellung -> Zustände) in the german version.
Set the indicator to display until 0 in black and until 1 in red and it will work correctly in this case.
Regards
Message Edited by IngoS on 11-18-2005 03:03 PM
Ingo Schumacher
Systems Engineer Sound&VibrationNational Instruments Germany
Attachments:
sample_mod.DAC ‏50 KB

"CPU#0 Tempature above threshold"

Similar Messages

Maybe you are looking for