Windows 10: Stuck in repair/windows boot manager loop error 0xc0000225

ignore my last your link has bootable iso sorry.. will take a while til I can get access to another machine to prepare a disk

 

Going to bed now, not convinced mem is the issue but you are kind enough to ask the questions....

Memtest v4.20 currently on Test 8
Pass complete, no errors, press Esc to exit*****

will reply tomorrow but I don't expect to say 'its my memory'

 

Please make sure to run version 5.01 which has lots of improvements and fixes.

 

sorry i've managed to download and run an older version - i am an idiot. I'll kick off 5.01 in a while but as 4.2 has found zero errors is it time we moved on to why nothing windows based (windows repair, emergency boot tools) will boot and run whereas linux boots and runs happily?

 

did 2 passes with no errors this morning on 5.01 will try to do some more but looking for another avenue of investigation.

 

Having been grappling with this for over a week I'm ready to reinstall windows 7 64 bit from DVD providing that won't lose my data on two other HDD configured in RAID 1 and pre-crash controlled by Windows. The issue is that currently booting from that DVD results in the loop as described above.

 

Could you temporary remove as much hardware as possible that is not required for booting.

 

ok so that will be usb hub and graphics card (650 ti) and 2 hdd. will give it a go.

 

just ripped everything out, and on boot got 'preparing automatic repair' followed by Recovery screen. Repeated with Win 7 dvd in the drive and same output.

 

What are the BIOS 3.3, 5 and 12 voltage values?
What are the CPU voltage and frequency?

 

AMD Phenom 9850 quad core 'speed' 2.5 GHz Clk 2 GHz
BIOS ver 1203

VCore 1.404v
3.3v 3.337v
5v 5.026v
12v 12.271v

CPU temp 43 deg C
MB temp 25 deg C

 

Hi Axe, thank you for the continued help I do appreciate your time but I do not get unpredictably timed BSODs, I get predictable loops of windows startup sequence trying to repair itself. I looked at adjusting Vcore but as there are an array of voltages all set to auto currently which raise and lower themselves according to load I am loath to interfere. I really don't need to take even the tiniest risk of frying my mobo or cpu by changing the wrong setting (none of them are called Vcore)when as I have said a few times before it will a. run memtest for 48 hours without incident or b. run any other non-windows operating system for 48 hours without incident.

Does anyone on here have any suggestions about how to get out of the loop to either repair or reinstall windows for example through Linux recovery tools I can get to my original bootmgr etc files (but sized 0 bytes so could I replace the file) or I can get to windows RE files such as boot.wim etc.

I found the following about server 2008 which starts to give me some education:

Windows Server 2008 Operations
System is powered on
The MBR is located on the system drive.
The Boot Sector is located, BOOTMGR is loaded
BOOTMGR looks for an active partition on sector 0.
BOOTMGR then reads the BCD file
BOOTMGR transfers control to WINLOAD.EXE contained in Boot.wim (In the case of a resume operation, WINRESUME.EXE)
WINLOAD.EXE initializes memory and loads drivers set to start at boot, and then transfers control to the kernel.
Install.wim contains several sysprepped images of Windows Server 2008, ready to be deployed.

The setup process extracts the wim to the drive, yes. packages are stored in the os image.

 

Your CPU's Vcore should not be that high unless you are heavily overclocking.
The setting is called Processor Voltage.

Lets go through the current situation
You can't boot anything Windows based

The following have been ruled out
- drives
- GPU
- RAM
- cabling (since you can't boot from WinPE USB either)

Leaving us with the
- CPU,
- Mobo,
- BIOS
- PSU

The next step would be to tackle any BIOS issue.
No update is available. Possibly there is a setting, like the processor voltage, that may be improper.
Usually improper BIOS settings is fixed by loading the default settings, but my question is if you would want to do that since I reread this
"Setting bios (AMIbios) to S1 or S3 only prevented the issue, Auto would repeatably demonstrate the issue."
I don't know what this setting is, can't find anything in the manual either, but I assume that loading the optimal default settings would change this setting to 'auto' hence my question.

 

OK ref the S1 only and S3 only statement - fair cop it failed when set to auto, BUT this morning I retested with the 2 sticks of failed DIMM and one failed Memtestx86 straight away and the other failed to post in the same way as my S1/S3/Auto issue ie machine spun up processor fan, and half hearted attempts at other initialisations like cdrom but failed to post.

To keep following your advice Axe0 I've been running on 1.3v processor voltage all day whilst away at work with Gparted Live Debian 9.0 desktop running happily with no crashes.

what do we do next?

I still think I have to correct something to do with bootrec and/or filesystem before either old windows or windows recovery can help me but am struggling with that.

thanks for your continued patient help

kind regards

Nick

 

To scan the filesystem, and fix any issues, run chkdsk on each partition.
I recommend following command for all other than Windows partition
Code: chkdsk X: /r /x[/quote] Change X to letter of partition you want to run it on.
This command forces the partition to unmount, any bad sectors will be replaced (if possible) and data moved to reserved sectors and will fix any other detected corruption.

For Windows partition, run the same command without /x, because it can't be unmounted whilst you're using Windows.

For boot sector errors, if you'd like I can ask for help because thats best done properly at once.