[Nottingham] Dual Boot - Windows updates and Linux boot failures

[David Aldred]

On 26 November 2013 22:25, Michael Quaintance <penfoldq at penfoldq.co.uk>wrote:

> On Tue, Nov 26, 2013 at 1:56 PM, David Aldred <davidaldred at gmail.com>
> wrote:
> > On 26 November 2013 17:44, Jason Irwin <jasonirwin73 at gmail.com> wrote:
> >>
> >> On 26/11/13 17:32, David Aldred wrote:
> <snip>
> >> > Is this expected behaviour?  And how is the presence of an incomplete
> >> > Windows update stopping Linux booting in these circumstances?
> >> Quite possibly down to Windows flagging that it needs to reboot to
> >> complete update and this through GRUB2 off.
> >> But that's just a total guess on my part.
> >
> >
> > It was the sort of thing I was guessing too, but it implies an actual
> > recognition of some sort by Windows that Linux exists on the machine,
> which
> > is unusual!
> >
>
> I was thinking along the lines of Jason's suggestion but my assumption
> was the exact opposite implication than the one you describe. I was
> thinking that Windows modifies the MBR with something for the Windows
> boot process to recognize and take a different action on with
> absolutely no regard to any other operating systems that might be
> present.
>

Could be.  I'm beyond my level of knowledge at that point; Linux does seem
to get  a fair way into the boot process before it dies, though.

>
> I'm confused in the first instance why you would want to promote a
> working VM to a multiboot partition but that's mainly because I've
> been bitten by this sort of thing so often that I only run MS Windows
> in a VM nowadays (excepting my work laptop where I have no choice).
> I've not found anything where the VM is a problem, even external
> hardware is almost all USB nowadays and USB passthrough works fine for
> everything I have needed it for.
>

The problem was really slowness - Windows was taking ages to start in the
VM, and was also slow to respond (to the degree that opening a file took
minutes).  In  the absence of updates, it's now taking a third of the time
to reboot into Windows than it used to to get the VM started.

There is also one Windows-only updater program for a GPS which does
something weird with the USB connection (repeatedly disconnecting and
reconnecting), and the reconnects weren't always working.

Before I re-installed Linux, the file selector was sometimes a bit slow to
respond in Linux too (especially if called by a browser), so it may well be
that there was something not quite right there which was impacting heavily
on the VM performance.

I guess the easy answer is to make sure automatic updates are off in
Windows, and then remember to run them every so often, at a time when I've
no rush to get back into Linux.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.lug.org.uk/pipermail/nottingham/attachments/20131127/928a217f/attachment-0001.html>