Blue Screen - fltmgr.sys - Address B9D17E94 base at B9D11000, DateStamp 480251da

[FranceBB]

Hi there guys,

I was doing my usual stuff when suddenly I got the following bluescreen error message:
STOP 0x0000008E (0xC0000005, 0xB9D17E94, 0xA7664FF0, 0x00000000)
fltmgr.sys - Address B9D17E94 base at B9D11000, DateStamp 480251da

I rebooted and it seems to be working fine but I'd like to know what caused this issue.

I looked for a MEMORY.DMP file in C:\WINDOWS, however the only one I could find is a 1.9GB one from 30/07/2017 which means that it's not the one of this latest crash.

BlueScreenViewer also doesn't report anything at all which leaves me puzzling a bit...

I wasn't doing much, I only really had 3 different RDP sessions opened to computers in my local network and that was pretty much it, really...  

[Dixel]

The STOP 0x0000008E usually points to a RAM issue. Try to reset the RAM modules, or replace them, I had this error in 2012, it happened with a defective Kingston DDR3 module (Made in Taiwan, the over-popular blue one - Hyper-X). Oddly enough, it passed the Memtest just fine, so I placed it back, but when I started to edit a big video file, it BSODed on me again.

[Dixel]

Here I found for you a confirmation from other sources:

"Stop messages that contain 0x0000008E (0x8E for short) can have many causes, but the most common cause is bad RAM."

https://www.bleepingcomputer.com/forums/t/133988/stop-0x0000008e-error/

[FranceBB]

Damn, I'm really sorry to hear that 'cause these RAM sticks (ironically also HyperX) are actually 8 years old and I was emotionally attached to them...  

I will try running a memory test through it, though, but if you say that it passed it then I don't know how reliable that can be...

[jaclaz]

fltmgr.sys

could be anything.

In my ignorance I would ignore it and until it happens again (and again) sleep well without worries.

jaclaz

 

[Dixel]

From my long experience with XP/Server.

fltmgr.sys error accompanied by "0000007e, parameter1 c0000005" = faulty HDD (maybe only damaged file system due to a power outage) 

[Dixel]

I'm giving out the most common errors from the several of my systems.

fltmgr.sys error accompanied by "0000008e, parameter1 c0000005" = faulty RAM (maybe not enough power supplied, if that RAM is overclocked, bad batch of chips, low binned, like the "famous" Taiwanese Kingston, for example.) 

Fix - replace RAM, try to bump up the voltage in the BIOS (add a small increase, sometimes it's enough).

 

[Dixel]

7 hours ago, FranceBB said:

Damn, I'm really sorry to hear that 'cause these RAM sticks (ironically also HyperX) are actually 8 years old and I was emotionally attached to them...  

I will try running a memory test through it, though, but if you say that it passed it then I don't know how reliable that can be...

I feel you. But hey, 8 years is a good run. RAM also degrades overtime. No reason to throw it away yet. Try increase the voltage a bit, like I wrote above.

[FranceBB]

I will. Thank you Dixel.  

[Ascii2]

On 1/22/2024 at 9:52 AM, FranceBB said:

Damn, I'm really sorry to hear that 'cause these RAM sticks (ironically also HyperX) are actually 8 years old...

Well-made RAM (which is common) that is well handled can last a very long time. It is still possible for it to be defective.

However, before blaming the RAM module(s), be sure that the problem is not electrical instability due to dust interference.  I often have found the intermittent RAM-like problems to happen due to dust getting into the RAM slots or the expansion slots (especially the old AGP slots).  Such slots tend to build up dust over time (some faster than others).  The dust can be conductive or insulative and result in electrical instability.  First try dismounting the RAM sticks and clean them, clean the RAM slots (can usually use a blower with just enough blow force (can use hand to test force) to blow out dust), then insert the RAM sticks into the RAM slots and in different RAM slots than the specific sticks were originally in, if possible (so, for example, if there is a configuration with two two-stick RAM stick pairs and four RAM slots so pairable, one may swap RAM sticks of the pairs RAM slots).  While you are at it, you might clean any other expansion slots and card contacts that you have as well.

Edited February 1 by Ascii2

[D.Draker]

Those modules are very well protected, and their DDR3 slots, too, not to mention they are mounted in vertical positions in 99% of mobos. 

 

[Ascii2]

I also recommend examining the power regulation components.  Visually check the motherboard capacitors for leakage, bulging, or appearance change.  Bad input power, whether it be so from the mains or after conversion to DC via the switching power supply, can also result in RAM-related (and other) problems.  I have seen older power supplies perform satisfactory for years then begin to fail or not perform as well.

For example, a power supply may lose stability of voltage and there may be intermittent, and often small but stability impacting, voltage drops.  Increasing the voltage (target) to the RAM, as @Dixel had mentioned, would usually provide the RAM module more stability, but at a cost of greater physical consumption or damage to the RAM (and some dependent components for the power delivery) and extra power consumption.  In such case, the RAM modules may still have been good.

Note the power regulation problems might have consequences to more than just RAM stability.

Edited February 2 by Ascii2

[Dixel]

5 hours ago, Ascii2 said:

Increasing the voltage (target) to the RAM, as @Dixel had mentioned, would usually provide the RAM module more stability, but at a cost of greater physical consumption or damage to the RAM

No "damage" to the RAM will happen, those are specifically designed for overclockers. Their aluminium cooling solution doesn't allow them to overheat. Voltage slightly increased by 0.5 percent won't magically kill them.

No "greater physical consumption" will happen either.

[Dixel]

5 hours ago, Ascii2 said:

For example, a power supply may lose stability of voltage and there may be intermittent, and often small but stability impacting, voltage drops.

In 2012+ mobos voltage is usually shown in BIOS. Voltage drops should be visible in any somewhat decent motherboard made after that year, for example my MSIB85 reports voltage fluctuations in real time.

[Ascii2]

On 2/2/2024 at 12:12 AM, Dixel said:

No "damage" to the RAM will happen, those are specifically designed for overclockers. Their aluminium cooling solution doesn't allow them to overheat. Voltage slightly increased by 0.5 percent won't magically kill them.

No "greater physical consumption" will happen either.

To be clear, my comments were usual and not specific to @FranceBB's RAM.  Your comments that I here quote appear specific to particular RAM.  There has, herein, been no affirmative identification of FranceBB's RAM herein, other than his report that it is "HyperX".  Also to clarify, when mentioning your advice to FranceBB, I was referring to the recommendation of increasing voltage for the RAM and not the amount of the increase.

It would be advisable to FranceBB to identify his RAM model or characteristics more explicitly.