Michel,

I can only echo the warning of the last sentence. The actions described are sound, as data hypochondriac with paranoid episodes my steps would be partially different:

- Get out your backup system, copy over the directories/VMs needed for the rest of the month (even old HD is better than current SSD perf)
- Verify the old machine has received good copy and work a day on it
- use the Linux based Live System from Samsung (if that is SSD manufacturer) to boot your old machine and clone SSD to USB3 HD
- if backup machine is ok, decide what to do with crippled system

If it were an old style HD, I'd work from the hypothesis that large # of sectors could not be accessed and have to be tried many attempts each. Not enough experience on failing SSDs to say if similar woes can happen. Either get a new SSD with at least double capacity and try cloning to that as well (either tich a USB3 interface connection or empty USB3 case) for first try or get a new machine. There are possibilities of condensators or chips failing, but those are extremely rare in the use patterns I throw at machines - replacing discs is often good enough HERE. Dunno enough about your HW zoo to be more exact or spec to aim ultimately for a clone to new HW or a new install and copy relevant info.

But get USB3 HD backup pronto, should not take more than a night on normal working HW for a 500-1000 GB disc, if possible from a Live system.





>>>Check in ResMon disc usage and speed / CPU activity (both with its sources)
>>>Look if any update checking is done
>>>Search Windows logs
>>
>>ResMon gives as attached. I see Maximum Frequency is always ranging in the 135% with or without the anomaly. I do not know if this is normal. On the right, Disk and Network seem to be ok.
>>
>>Windows logs doesn't report anything suspicious. The situation however, causes additional events to be recorded due to the fact that certain operations take a lot of time to be completed:
>>
>>svchost (4280,T,97) SRUJet: A request to write to the file "C:\WINDOWS\system32\SRU\SRU.chk" at offset 0 (0x0000000000000000) for 4096 (0x00001000) bytes succeeded, but took an abnormally long time (28 seconds) to be serviced by the OS. In addition, 0 other I/O requests to this file have also taken an abnormally long time to be serviced since the last message regarding this problem was posted 7212 seconds ago. This problem is likely due to faulty hardware. Please contact your hardware vendor for further assistance diagnosing the problem.
>>
>>In System, I also have related disk I/O mentions:
>>
>>Reset to device, \Device\RaidPort0, was issued.
>>
>>Those are similar to a data center situation I have seen where, apparently, there wasn't enough IOPS available during specific time frames to perform various operations.
>
>Both of those messages are quite suspicious, there is definitely a problem with your disk subsystem. On a physical workstation you should never see the first one; it means disk performance is abysmal or failing. As you point out, you may get that message on a VM that is badly under-provisioned for IOPS, but you should never see it on physical hardware. You should never see the second under any circumstances; it means the controller has encountered an unrecoverable error talking to the device on that port.
>
>There are a few things I would look at:
>
>1. Are the BIOS settings at the default, or have you changed them? Typically there are choices such as:
>
>- Port type: normally AHCI is the default and is expected in modern Windows OSs. You would only select RAID if you want to support motherboard RAID and have multiple drives. Legacy/IDE (if an available option) is only for very old hardware and OSs
>- RAID: usually disabled by default. If you have two or more drives and have enabled RAID can you give us the particulars?
>
>2. At least for test purposes, set your Windows power management to "high performance" and disable or minimize all individual power saving features. For example, under Advanced settings, set Hard disk...Turn off hard disk after... 0 (0 = never)
>
>3. Check your manufacturer's support site for updated drivers - motherboard, chipset etc. Also check for an updated motherboard BIOS
>
>4. If your system uses Intel Rapid Storage Technology (RST), download and install the latest version: https://downloadcenter.intel.com/product/55005/Intel-Rapid-Storage-Technology-Intel-RST-
>
>5. ISTR you said your machine has an OEM Samsung SSD of some kind. I would look hard for a diagnostic for its particular model. Failing that, many name-brand computer builders such as Dell and HP have built-in diagnostics which can test the disk (amongst other things) regardless of make/model of that disk
>
>Those errors you recorded are serious. If it can't be fixed with 1 - 4 above then you need to diagnose your hardware. If you can't diagnose it yourself then you should contact your computer maker and open a support ticket - those errors are sufficient cause for them to troubleshoot. They may have special diagnostics for OEM SSDs used in their computers, if you can't get one from Samsung.
>
>Until you fix this issue, consider that the drive may fail at any time - make sure you have frequent and comprehensive backups.