Display Disk Activity – More gadget graphics trouble

[JackN]

I rely on seeing the disk activity being displayed in the graphics portion of the gadget. Unfortunately Windows is turning off this portion of the graphics quite often. I’m finding myself going into the Windows Registry (W7 Ultimate 64 bit) and resetting the HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Perfdisk\Performance\DisablePerformanceCounters back to 0 (as described in the FAQ section of this website) several times a week. Sometimes twice a day. It seems that any minor disruption from the norm will cause Windows to reset the value back to 1 – Which in turn, turns off the disc activity graphics of the HDS gadget. Even something as minor as plugging in a USB flash drive will cause it to reset. I’ve tried deleting the parameter from the registry as described in the Microsoft article, but somehow the parameter seems to reinstall itself, sometimes as soon as the very next time I boot the machine. Does anybody have an idea as to how to keep the registry parameter from resetting itself?

[hdsentinel]

Do you have any kind of system-tweaking ("performance-booster") application or something similar installed?

Some of those software believe that it is a good idea to disable some parts of the Operating Systems (like the performance monitoring) as they feel then the system will perform much faster - which is not true of course.
But because of that, they may automatically adjust some system settings - like this one.

Personally during the last years we found no situation when this counter is changed (even when lots of devices / hard disks / pendrives / etc... installed and uninstalled), and this is especially weird to have this "reset" so often.

You may consider searching for "monitor registry changes" and try a such tool to check _what_ process is actually changing the registry setting. That may help...

[JackN]

I use a few gadgets that monitor various pieces of hardware like the CPU, video card, mobo, etc., but they're just monitors. They don't have the ability to alter performance. I'm not using any type of performance enhancing software, and I don't overclock. The system is less than a year old and otherwise runs as it should.

Through a process of elimination, I've found the program resetting the registry switch is HDS! When HDS isn't running, I can plug in my flash drive without any problem. As soon as HDS is running and I plug in my flash drive, the registry switch gets reset instantly. Then on top of that, Windows starts making the "device plugged in" and/or "device unplugged" sound every 40 - 45 seconds until I shut down HDS. When HDS gets shut down, the sounds go away. Strange, but that's what happens.

Suggestions?

[hdsentinel]

Hi,

Thanks for the message and excuse me for the troubles.
I'm not really sure if the issue is related to the gadget in any ways.
Can you please check what happens if you disable the gadget (on Configuration -> Status Window page).
Or do you mean the gadget is automatically re-enabled when you insert the pendrive?
Otherwise I'm not really sure if I understand the "registry switch gets reset instantly".

I suspect there may be a minor incompatibility problem with the pendrive.
Can you please use Report menu -> Send test report to developer option when the pendrive is connected?
That way I can check the detected information and verify what can cause the troubles.

Also you may try to check if things are better if you disable the detection of USB and SCSI drives (on Configuration -> Disk Control page).

Thanks for the information, these will help to improve the situation.

[JackN]

“I'm not really sure if the issue is related to the gadget in any ways.” You are correct. I tried using my flash drives with, and without, the gadget running and got the same results as long as HDS was running.

“Or do you mean the gadget is automatically re-enabled when you insert the pendrive?” No.

“Otherwise I'm not really sure if I understand the "registry switch gets reset instantly".” I’m referring to the registry switch, or the “0” and “1” values in the registry item that I was talking about in the 1st post. The value gets “switched” back to “1” from “0”. I need to keep resetting it to “0”, shut down HDS, and then restart HDS.

“Can you please use Report menu -> Send test report to developer option when the pendrive is connected?” I just did. I hope I did things in the correct sequence. If not, let me know.

“Also you may try to check if things are better if you disable the detection of USB and SCSI drives (on Configuration -> Disk Control page).” I did that and it cured the irritating Windows sounds. Thank-you!

To elaborate on your first question – I agree, the problem isn’t with the gadget, it’s within the HDS program. As long as HDS is running in any way, shape, or form, I have troubles with any USB device resetting the registry switch when I plug the device in. When HDS is not running, I don’t have any problem at all with any of my flash drives. I even tried plugging in all 3 of them at the same time and the registry didn’t reset. When HDS is running, as soon as I plug in a USB device the registry value instantly gets reset back to “1” and the HDS graphics stop working.

[hdsentinel]

Thanks for the message and the report, now I

I can make sure that no HDS resets the registry setting, of course not the software disables the performance monitoring function.
As you may see, when the pendrive(s) are insterted, HDS detects the change (displays its capacity, drive letter, etc.) and tries to get performance counter for the newly inserted device.

The problem is that when the device is inserted, Windows starts to re-build its performance counter database and HDS tries to collect the performance information too quickly, when this database update is not yet finished. This may cause that Windows completely disables the performance monitoring by using this registry modification.

So even the problem is not directly caused by HDS (not HDS disables the performance monitoring) the compatibility on your system would need to be improved to prevent this issue happening.

If you agree, I'll provide test version(s) to verify if this is the real cause and with a possible improvement.

“Also you may try to check if things are better if you disable the detection of USB and SCSI drives (on Configuration -> Disk Control page).” I did that and it cured the irritating Windows sounds. Thank-you!

Thanks!
If possible, for a test, can you please send a report also when this detection is NOT disabled?
That way I can check the USB devices (that info was not present in the report I received becuase of this option). Just to improve compatibility with them, even if this option would not be used. Thanks!

[JackN]

Sent another report. Let me know if it's not what you need.

[hdsentinel]

Thanks, this is perfect!
I'm checking the situation and will provide updated version.

[front360]

I am having the same problem. How can I fix this?

[hdsentinel]

Excuse me for the issue.

I suspect the problem may be caused by an other software or tool which constantly changes "DisablePerformanceCounters" setting of Windows.
Of course if this is disabled, performance monitoring in Hard Disk Sentinel (and in other tools) does not work.
According our experiences, this Windows setting may be modified by system "tuning" or similar utilities, as they disable certain features of the OS to "improve" the performance.

Also in some rare situations, it may be related to devices, especially removable devices (like a card reader).
If the change of disk/storage configuration detected (for example device added/removed or just the computer awakes from sleeping status) Hard Disk Sentinel automatically tries to detect the modified configuration and if these devices respond slowly, Windows may disable the "DisablePerformanceCounters" as well.

Can you please use Report -> Send test report to developer option?
This may help to verify if the issue is related to Hard Disk Sentinel and/or one of the devices and improve the situation.

[JackN]

Front360 - Sent you a PM. Hope you got it.

[hdsentinel]

Thanks, yes I got the report and the message, just sending answer.

Front360 - Sent you a PM. Hope you got it.

[front360]

JackN, I have a program called Soluto and I was curious if you also have Soluto installed on your Win7x64 system? Thanks.

[JackN]

No, I don't. Without exception I've found that those types of programs that are written to speed up your PC (IE: registry cleaners, several types of disk defragmenters, automated driver updaters, etc) cause more problems in the long run than they fix. Since I've quit using them I've noticed that my computers have fewer problems.

[Jeidy]

I have the same problem with registry key changing.
With the help of "Process Monitor", I found the key "HKLM\System\CurrentControlSet\services\PerfDisk\Performance\Disable Performance Counters" was changed by HDS.
Please see the pic

a.PNG (75.02 KiB) Viewed 40163 times

[hdsentinel]

Thanks for the information.

Excuse me, but this is not possible - at least in the official version.
The file HDSentinel.exe does not have this key in itself and I can confirm that Hard Disk Sentinel does not directly change the "Disable Performance Counters" value in any ways.
If this happens, it means that when Hard Disk Sentinel is trying to access the performance data, the system changes this setting. Sometimes this was related to a specific hardware device (for example a card reader) or a minor incompatibility issue caused by an other software.

The best would be if I can re-produce the issue as then I can examine the situation.
Can you please use Report -> Send test report to developer option?

I have the same problem with registry key changing.
With the help of "Process Monitor", I found the key "HKLM\System\CurrentControlSet\services\PerfDisk\Performance\Disable Performance Counters" was changed by HDS.
Please see the pic

a.PNG

[hdsentinel]

To extend the previous message: Windows "Perflib" constantly checks how the performance counters work and how they are accessed.
If it finds some problem, it may automatically disable the performance counters. This can be caused (for example) that the performance counters seems respond very slowly (may be related to some hardware device, not really to CPU power).

So a possible solution would be to prevent PerfLib from disabling the counters automatically.

The following links may be interesting (may be related to other OSes also):

http://www.experts-exchange.com/OS/Micr ... 81831.html

"Also To prevent Perflib from disabling counters, set the 0x2 (10) bit in the value of the Configuration Flags entry."

http://technet.microsoft.com/en-us/libr ... 10%29.aspx

0x2 (10 bit) = Do not disable counters

Maybe this helps somehow ....

[Jeidy]

Thanks for the information.

Excuse me, but this is not possible - at least in the official version.
The file HDSentinel.exe does not have this key in itself and I can confirm that Hard Disk Sentinel does not directly change the "Disable Performance Counters" value in any ways.
If this happens, it means that when Hard Disk Sentinel is trying to access the performance data, the system changes this setting. Sometimes this was related to a specific hardware device (for example a card reader) or a minor incompatibility issue caused by an other software.

The best would be if I can re-produce the issue as then I can examine the situation.
Can you please use Report -> Send test report to developer option?

Thank you for reply
Report was send just now, hope it's useful.

[hdsentinel]

Thanks, it is arrived!
It seems now the performance counters are working, just the quesiton is when they seems to be disabled?
Do you see it is related to some specific events (for example inserting / removing pendrive or external storage device, waking up from standby/hibernation etc...)?

In the meantime I did an other post about further information. Not sure, but may helps or sounds interesting ....

[Jeidy]

The issue appear again when I attach a USB3.0 removable HDD onto USB3.0 port!

(Asrock P67 Pro3 with Etron USB 3.0 controller, MB BIOS Ver:P2.00; OS:Win7 64bit w/h SP1, Etron USB 3.0 Controller Drivers Ver: 0.96)