On 10/20/2023 at 12:15 PM, sunwind said:

That means my CPU is running that fast right?

For all Cores simultaneously you are getting the advertised frequency of 3.6 GHz

 

As mentioned in the Xeon E-2234 specifications; for a turbo boosted single Core, you should expect a max frequency of 4.8 GHz

To reach such frequency with CoreFreq, open menu "Tools" (shortkey O), select "Turbo < Select CPU >" and choose one of your (best) Core.

Frequency should go above 3.6 and getting close to 4.8 GHz if Turbo boost and its prerequisites are well setup.

 

If not, go to "Processor" window (press p) and check that 1C is set to ratio 48

Make sure "TURBO" is highlighted in the bottom screen; if not, enable it in the "Technologies" window (press t) in addition to EIST aka SpeedStep

Still no single Core turbo boost, make sure an idle C-State handler is activated in your kernel: press k to open window "Kernel" and check that a "CPU-Idle driver" is registered with idle states. If not solve it from your distribution or the kernel boot command line.

On 9/26/2023 at 5:59 AM, ich777 said:

Yes, did you enable the service/module like it is mentioned in the description from the plugin?

Sorry for a dumb question, but I'm trying to find where the instructions are to enable the module and not seeing them. I'm getting the daemon connection error but can't see where in the plugin description on CA or the description of the plugin on the plugins page in UnRAID it says where I need to enable a module or how to do so. Is it the "test" button within the plugin UI page?

48 minutes ago, Alyred said:

Sorry for a dumb question, but I'm trying to find where the instructions are to enable the module and not seeing them. I'm getting the daemon connection error but can't see where in the plugin description on CA or the description of the plugin on the plugins page in UnRAID it says where I need to enable a module or how to do so. Is it the "test" button within the plugin UI page?

Yes, you have to click test, if that went well you can enable the module on the plugin page.

4 hours ago, ich777 said:

Yes, you have to click test, if that went well you can enable the module on the plugin page.

Will do once my regularly-scheduled parity check is complete!  As a question, why is it suggested to spin down the array before pressing the button?

33 minutes ago, Alyred said:

Will do once my regularly-scheduled parity check is complete!  As a question, why is it suggested to spin down the array before pressing the button?

Because on some systems it caused a Kernel Panic when I created the plugin but that should not happen anymore and that's why it's recommended to stop the Array first (you can actually cancel the Parity check if you get a crash).

 

Nothing to worry about when the Array is Stopped.

18 minutes ago, ich777 said:

Because on some systems it caused a Kernel Panic when I created the plugin but that should not happen anymore and that's why it's recommended to stop the Array first (you can actually cancel the Parity check if you get a crash).

 

Nothing to worry about when the Array is Stopped.

Yeah, I just happened to be in the middle of my bi-monthly parity check and didn't want to interrupt it in this case. I'll shut down the array later to do the test.  Thanks @ich777!

14 minutes ago, Alyred said:

bi-monthly

Okay...

I'll do a Parity check quarterly... and that's more than enough in my opinion...

@Alyred

Just please show me your CoreFreq screenshots when your System is up.

On 11/30/2023 at 10:24 AM, ich777 said:

Okay...

I'll do a Parity check quarterly... and that's more than enough in my opinion...

 

I've been meaning to move my servers to quarterly, so it's good to know that's the setting that the experts recommend.  Thanks!  Now to just get some time to do the scheduling and reconfig, things have been bonkers busy lately with a pentest at work going on...

 

On 11/30/2023 at 11:05 PM, CyrIng said:

@Alyred

Just please show me your CoreFreq screenshots when your System is up.

 

Once I get time to shutdown my VMs, dockers, and the array, will do! Something in particular you want to see?

On 11/30/2023 at 11:05 PM, CyrIng said:

@Alyred

Just please show me your CoreFreq screenshots when your System is up.

Worked like a champ. 😉

 

6 hours ago, Alyred said:

Worked like a champ. 😉

Thank you very much.

 

Fyi, next version of CoreFreq is developed in github.com/cyring/CoreFreq/tree/develop

On 12/2/2023 at 10:47 PM, Alyred said:

I've been meaning to move my servers to quarterly, so it's good to know that's the setting that the experts recommend.

It's not a hard and fast rule, that's why it's configurable. In general, the more stable and left alone configuration wise a server is, the longer you can push parity check intervals.

 

Just keep in mind that "permanently" spun down array disks (those that are so seldom accessed that they virtually never spin up) can silently fail and until they are called upon to rebuild an actively used failed drive you would never know. A clean parity check confirms that a drive rebuild should work flawlessly. The longer the interval since a clean parity check, the more that could have silently gone wrong in the interim. If all drives are relatively constantly active, then you will get almost instant feedback when a drive fails, so frequent parity checks aren't needed as much.

 

Theoretically a parity check is never needed, as all the pieces in a computer always perform flawlessly, right?

 

How often to do parity checks, as well as whether or not to use dual parity, how many reallocated sectors is too many, how comprehensive are your off site backups, etc, etc, are all part of a personal risk profile.

 

You do you.

On 8/9/2022 at 11:54 AM, CyrIng said:

Indeed blue means capable, and green, enabled.

 

In your case CPPC is capable but not in function.

You can enable it in window "Performance Monitoring", item "Collaborative Processor Performance Control"

 

Remarks:

CPPC appears to be a firmware implementation, I can't tell what BIOS is going to do when enablement bit is set.

 

@CyrIng I can't get CPPC to be enabled even though I set it to "Enabled" in the BIOS of my ASRock B550 Taichi. Corefreq shows CPPC as capable (blue). I tried to activate it in Corefreq but it's not possible. I can't change anything in the "Performance Monitoring" window. 

 

Is there any way for me to force it? I'm really trying hard to force my system to load the AMD_PSTATE driver, however, I think CPPC is the key to this first.

Edited January 3, 20242 yr by Hiko0

On 1/3/2024 at 10:10 AM, Hiko0 said:

I can't get CPPC to be enabled

I believe we have the same or similar issue handled in this post at github.com/cyring/CoreFreq/discussions/320#discussioncomment-8001208

On 1/5/2024 at 12:16 AM, CyrIng said:

I believe we have the same or similar issue handled in this post at github.com/cyring/CoreFreq/discussions/320#discussioncomment-8001208

 

Yeah, that was me. Sorry for trying both channels at the same time.

On 8/28/2023 at 3:04 PM, Bernd Matthießen said:

Hello,

 

I think I'm too stupid!
I can't manage to set the whole thing via corefreqk.conf so that the settings are already loaded at boot time.

 

Here is an example:
options corefreqk TurboBoost_Enable=1 SpeedStep_Enable=1 Register_CPU_Idle=1 Register_Governor=1 Register_ClockSource=1    

Now comes the problem, I can't find all the options I want to set. For example, "NMI registered". In the Gui I can set it to "on". Which parameter should I enter in corefreqk.conf for this? According to the parameter list, there is only one "NMI_Disable". Should I enter this and set it to =0?

 

Sorry if I'm being too stupid!

I have attached scrennshots of my settings below.
Maybe someone can help me to enter the correct values.

 

Thanks in advance!

 

Greetings Bernd

Ive got the same chip, how are you getting your c states so low mine only goes to c1?

1 hour ago, tsiawd said:

how are you getting your c states so low mine only goes to c1?

Changing C-States happens in the "Kernel" window (k) but there are prerequisites like making CoreFreq the "CPU-Idle" handler.

Once confirmed in the Settings window (s) that "CPU-IDLE driver" is "<ON>" and a "CPU-IDLE route" is selected to one of the preferably Intel: M-Wait or AMD: I/O then you can select the C-States in "Idle Limit" selector

 

Hello,

Good advice to test the plugin with stopped array.

Unraid crashed. But I am not suprised, I installed Unraid on a QNAP TS-453Be as it got redundant.

Here is the diagnostics: 

qnap-diagnostics-20240217-2133.zip

1 hour ago, KaitsuN6 said:

Here is the diagnostics: 

 

Thanks for sharing

I don't find any issue in kernel log ?

 

48 minutes ago, CyrIng said:

I don't find any issue in kernel log ?

The issue is if a crash happens it won't be logged because the syslog is not persistent between reboots.

 

@KaitsuN6 do you maybe have a monitor attached to your NAS where you can take a picture when the crash happens or enable the syslog server and post that here?

On 2/18/2024 at 12:03 AM, ich777 said:

The issue is if a crash happens it won't be logged because the syslog is not persistent between reboots.

 

@KaitsuN6 do you maybe have a monitor attached to your NAS where you can take a picture when the crash happens or enable the syslog server and post that here?

Hello,

Here are the "BSOD" of the test and screenshot of the syslog just before testing, and syslogs captured to the flash drive. I don't see any logging of the happening itself.

Just ask if there is some more I can test.

syslog syslog-previous

Ahh, stupid me. That was not the screenshot you wanted.

Maybe these are better?

18 hours ago, KaitsuN6 said:

Maybe these are better?

Thanks I see the issue in MSR register 0x64c

 

Let's continue this issue on the GitHub.

 

Regards

CyrIng

 

19 hours ago, KaitsuN6 said:

Maybe these are better?

Please post your Diagnostics.

3 hours ago, ich777 said:

Please post your Diagnostics.

Hello,

Here is the current Diagnostics.

qnap-diagnostics-20240220-2218.zip