Refreshing of BIOS FLASH memory - a cure for old computer that stops to boot?

Wojciech Zabolotny

unread,

Mar 6, 2021, 1:18:38 PM3/6/21

to

Last week, my old (over 10 years!) but still good dual boot (Linux/Windows) Dell Vostro 3750 started to behave erratically.
The BIOS started and asked for the machine password, but after that the Grub didn't appear. The screen remained black. First the machine started every third power-on, but then it started only during the first power-up after a long switch-off period.
When started, the machine could work very long without any problems.
I suspected that the problem is associated with the temperature. Probably the BIOS FLASH memory cells lost the charge, and stopped to produce the correct data at higher temperature.
Finally I have decided to reflash the BIOS. After long cooling of the machine I managed to boot Windows and started the 3750A14.exe obtained from Dell as the still newest BIOS for my machine (the same was used for last 9 years).
The BIOS got reflashed with the "new" (in fact the same) BIOS. After reboot all the problems disappeared.
It looks like indeed the problem was caused by the FLASH gradually loosing information.
I post that in case if you experience the similar problems with old laptops (even from other vendors). Maybe refreshing of BIOS will solve also your problems? Maybe doing it every 5 years is siply a good practice?
--
BR, Wojtek

Bob Martin

unread,

Mar 7, 2021, 2:37:22 AM3/7/21

to

Change the battery.

Wojciech Zabolotny

unread,

Mar 7, 2021, 4:33:38 AM3/7/21

to

The laptop battery was replaced 3 years ago, and has 91% of the original capacity.
The BIOS does not loose settings.
If if it was the matter of battery, the reflashing itself wouldn't help, but it did.

Henrik Carlqvist

unread,

Mar 7, 2021, 8:45:18 AM3/7/21

to

The CMOS settings probably has a battery of its own:

https://www.ebay.com/itm/Pile-CMOS-rtc-bios-Battery-DELL-
Vostro-3750-/272713865926

regards Henrik

Andrew

unread,

Mar 7, 2021, 4:41:18 PM3/7/21

to

The symptoms were obvious but I was also not sure that a Laptop has a
CMOS battery, hopefully it is accessible enough that it can be replaced
easily.
I have replaced one in a Desktop PC from Acer and it was almost
impossible to access, the recommendation was to take it back to the
dealer but that turned out not to be necessary.

David W. Hodgins

unread,

Mar 7, 2021, 7:12:36 PM3/7/21

to

On Sun, 07 Mar 2021 16:41:15 -0500, Andrew <Do...@hyperspace.vogon.gov> wrote:
> The symptoms were obvious but I was also not sure that a Laptop has a
> CMOS battery, hopefully it is accessible enough that it can be replaced
> easily.
> I have replaced one in a Desktop PC from Acer and it was almost
> impossible to access, the recommendation was to take it back to the
> dealer but that turned out not to be necessary.

They do, but getting at them is usually nearly impossible. I've even seen them
soldered in place. It's planned obsolescence.

Regards, Dave Hodgins

--
Change dwho...@nomail.afraid.org to davidw...@teksavvy.com for
email replies.

Scott Alfter

unread,

Mar 16, 2021, 3:11:06 PM3/16/21

to

In article <s23h9p$1bk6$1...@gioia.aioe.org>,

Andrew <Do...@hyperspace.vogon.gov> wrote:
>Henrik Carlqvist wrote:
>> On Sun, 07 Mar 2021 01:33:37 -0800, Wojciech Zabolotny wrote:
>>
>>> niedziela, 7 marca 2021 o 08:37:22 UTC+1 Bob Martin napisał(a):
>>>> Change the battery.
>>>
>>> The laptop battery was replaced 3 years ago, and has 91% of the original
>>> capacity. The BIOS does not loose settings.
>>> If if it was the matter of battery, the reflashing itself wouldn't help,
>>> but it did.
>>
>> The CMOS settings probably has a battery of its own:
>>
>> https://www.ebay.com/itm/Pile-CMOS-rtc-bios-Battery-DELL-
>> Vostro-3750-/272713865926
>

>The symptoms were obvious but I was also not sure that a Laptop has a
>CMOS battery, hopefully it is accessible enough that it can be replaced
>easily.

It'll probably be buried somewhere...definitely not as easy to access as the
main battery, which just pops out.

Dell's service documentation is usually pretty decent and will show you what
needs to be removed to get at any part of your computer. Download the
appropriate manual (for this one, looks like it's the owner's manual) and
follow the instructions. I found an online-browsable version of the manual;
the steps described are listed here:

https://all-guidesbox.com/manual/1047680/dell-vostro-3750-owner-s-manual-147.html?page=111

(It appears to be just a plain old CR2032 that snaps into a holder, without
wires and a plug dangling from it. You should be able to find a local
replacement easily.)

_/_
/ v \ Scott Alfter (remove the obvious to send mail)
(IIGS( https://alfter.us/ Top-posting!
\_^_/ >What's the most annoying thing on Usenet?

Wojciech Zabolotny

unread,

May 31, 2021, 5:39:22 PM5/31/21

to

I have replaced the CMOS battery. The procedure was not trivial, but was
quite good documented in the Vostro 3750 owner's manual
https://downloads.dell.com/manuals/all-products/esuprt_laptop/esuprt_vostro_notebook/vostro-3750_owner%27s%20manual_en-us.pdf
Many thanks to Dell for providing such a nice documentation.

Unfortunately, the problem with laptop not booting still returns.
The memory is working correctly, the disk also.

I have noticed that the problem is time-correlated with Windows updates.
I have found an interesting material about Windows trashing the GRUB-based
multiboot configuration:
https://www.windowsphoneinfo.com/threads/now-mswin-will-trash-your-grub-just-by-booting.314139/

Now the question is, how I can prevent it.
For sure I want an open-source bootloader as a main bootloader in my system.
In the worst case, I can deinstall Windows.

Regards,
Wojtek

Wojciech Zabolotny

unread,

May 31, 2021, 5:47:17 PM5/31/21

to

Yet another useful resource, how to protect multiboot machines against malicious Windows activity:
https://unix.stackexchange.com/questions/68581/how-can-i-prevent-windows-from-overwriting-grub-when-using-a-dual-boot-machine

Regards,
Wojtek

Henrik Carlqvist

unread,

Jun 1, 2021, 1:29:17 AM6/1/21

to

On Mon, 31 May 2021 14:39:21 -0700, Wojciech Zabolotny wrote:
> I have noticed that the problem is time-correlated with Windows updates.
> I have found an interesting material about Windows trashing the
> GRUB-based multiboot configuration:
> https://www.windowsphoneinfo.com/threads/now-mswin-will-trash-your-grub-
just-by-booting.314139/
>
> Now the question is, how I can prevent it.

The easiest way to prevent Windows from doing stupid stuff with your
hardware is to avoid booting Windows on your hardware. Instead of dual
boot, you can boot Windows in a virtual environment like qemu or
virtualbox. On the other hand, if you really want to give Windows access
to your hardware (like the graphics card if you need gaming performance
in Windows), you can instead boot Linux in a virtual environment and hope
that future Windows upgrades will not mess upp your virtual installations.

regards Henrik

Peter 'Shaggy' Haywood

unread,

Jun 3, 2021, 5:38:50 PM6/3/21

to

Groovy hepcat Henrik Carlqvist was jivin' in comp.os.linux.hardware on
Tue, 1 Jun 2021 03:29 pm. It's a cool scene! Dig it.

Another solution, which is the one I use, is to put Wind-oh!-s on a
second hard drive. Grub can remap the order of the drives so that you
can boot that dud system... if you really must. That way, if Losedows
messes with the MBR, it'll be messing with the MBR of the second drive,
leaving your grub installation alone on the first.

--

----- Dig the NEW and IMPROVED news sig!! -----

-------------- Shaggy was here! ---------------
Ain't I'm a dawg!!

Andrew

unread,

Jun 4, 2021, 5:44:39 AM6/4/21

to

I have two dual-boot systems - one EFI and one "traditional Bios" - and
neither has that problem. Where one of them did have a problem was when
Windows changed the partition-id on (at least) one of its own partitions
and Linux refused to boot up properly because it could mount the
partition(s). Is that what the "nofail" option is for?