Dozey computers...

[Jim Lesurf]

For some years one of the mains-powered Linux machines I use has had a
habit of being reluctant to 'wake up in the morning'...

I'd decided that it happened because I always power it down *and* switch
off its mains supply every night. Although it didn't happen during the
first year or more I used it. And, yes, I have regularly replaced the
'keeper' internal battery and this made no difference.

However about 4 months ago I got a really fast nice new laptop. Been using
it to process loads of files every day. Each day I mostly use it being
powered with its mains PSU to ease the load on its internal batteries.

Yesterday it failed to start when I pressed its 'wake up' button. Repeated
tries with and without the external PSU failed.... Puzzled and worried.

So this morning I tried holding down its 'wake up' button for 5 mins. [1]
Released it. Nope, no coconut.

I then pressed and released the button again, and yup! It started and then
worked fine. After a while I shut it down, left it for a few hours. Then
tried again. Yes, woke up again quite happy. So far, fine. And the battery
shows as being well charged thoughout, so that wasn't 'flat' when it wasn't
waking up.

This is puzzling me on two levels:

1) Why in each case the behavior of being 'lazy' didn't show up for some
time in use

2) Why a laptop with an internal battery should do this. I can see that a
mains-powered one might take for granted that mains is always available.
But the laptop design surely has to assume this isn't the case.

Anyone know more about this or if there is a simple way to sort it out?
Seems a daft problem for machines to develop.

Jim

[1] Had tried 30secs. That didn't work. So will experiment to see what the
required 'finger dwell time' may be...

--
Please use the address on the audiomisc page if you wish to email me.
Electronics https://www.st-andrews.ac.uk/~www_pa/Scots_Guide/intro/electron.htm
biog http://jcgl.orpheusweb.co.uk/history/ups_and_downs.html
Audio Misc http://www.audiomisc.co.uk/index.html

[William Unruh]

On 2020-12-04, Jim Lesurf <no...@audiomisc.co.uk> wrote:
> For some years one of the mains-powered Linux machines I use has had a
> habit of being reluctant to 'wake up in the morning'...
>
> I'd decided that it happened because I always power it down *and* switch
> off its mains supply every night. Although it didn't happen during the
> first year or more I used it. And, yes, I have regularly replaced the
> 'keeper' internal battery and this made no difference.

Switching of the mains should not make a difference. If you really shut
the machine down, and not just "sleep" it (which may well be what was
happening) then switching off the mains should be irrelevant.

>
> However about 4 months ago I got a really fast nice new laptop. Been using
> it to process loads of files every day. Each day I mostly use it being
> powered with its mains PSU to ease the load on its internal batteries.
>
> Yesterday it failed to start when I pressed its 'wake up' button. Repeated
> tries with and without the external PSU failed.... Puzzled and worried.
>
> So this morning I tried holding down its 'wake up' button for 5 mins. [1]
> Released it. Nope, no coconut.

Sounds like it was asleep, not off, and when you held down the power
button for for 5 min actually shut it off. Then when you pressed it
again, it started up on its cold boot.

Note on my machine, automatic shutdown with the lid open and lid shut
behave very differently. With the lid open the shutdown is a real
shutdown. With the lid shut, the shutdown starts, and then midway
through stops, and it sleeps instead. When I try to open and restart it,
all it does is finish the shutdown that it interrupted with the sleep.
(Dell XPS13 (2017)).

>
> I then pressed and released the button again, and yup! It started and then
> worked fine. After a while I shut it down, left it for a few hours. Then
> tried again. Yes, woke up again quite happy. So far, fine. And the battery
> shows as being well charged thoughout, so that wasn't 'flat' when it wasn't
> waking up.
>
> This is puzzling me on two levels:
>
> 1) Why in each case the behavior of being 'lazy' didn't show up for some
> time in use
>
> 2) Why a laptop with an internal battery should do this. I can see that a
> mains-powered one might take for granted that mains is always available.
> But the laptop design surely has to assume this isn't the case.

No, No mains connected computer assumes it will always be mains
connected. It would be a very silly design if it did. Power goes off for
many reasons (lightning, storms, moving the computer from one room to
another, power saving,....)

[Jim Lesurf]

In article <rqdsls$url$1...@dont-email.me>, William Unruh <un...@invalid.ca>
wrote:

> On 2020-12-04, Jim Lesurf <no...@audiomisc.co.uk> wrote:
> > For some years one of the mains-powered Linux machines I use has had a
> > habit of being reluctant to 'wake up in the morning'...
> >
> > I'd decided that it happened because I always power it down *and*
> > switch off its mains supply every night. Although it didn't happen
> > during the first year or more I used it. And, yes, I have regularly
> > replaced the 'keeper' internal battery and this made no difference.

> Switching of the mains should not make a difference. If you really shut
> the machine down, and not just "sleep" it (which may well be what was
> happening) then switching off the mains should be irrelevant.

Yes, it should. On that I agree entirely.

Alas, in this world 'should' isn't inevitably a synonym for 'does'. You
only have to glance at Governments to notice this. :-)

cf below...

> >
> > However about 4 months ago I got a really fast nice new laptop. Been
> > using it to process loads of files every day. Each day I mostly use it
> > being powered with its mains PSU to ease the load on its internal
> > batteries.
> >
> > Yesterday it failed to start when I pressed its 'wake up' button.
> > Repeated tries with and without the external PSU failed.... Puzzled
> > and worried.
> >
> > So this morning I tried holding down its 'wake up' button for 5 mins.
> > [1] Released it. Nope, no coconut.

> Sounds like it was asleep, not off, and when you held down the power
> button for for 5 min actually shut it off. Then when you pressed it
> again, it started up on its cold boot.

I have reasons to doubt this, again cf below...

> No, No mains connected computer assumes it will always be mains
> connected. It would be a very silly design if it did. Power goes off for
> many reasons (lightning, storms, moving the computer from one room to
> another, power saving,....)

I agree. But refer back to the lack of synonym for 'should'. :-)

The conclusion that I am actually telling the machines to sleep seems odd
when I add more information.

1) I also have two other mains powered machines, and another laptop. None
of them ever do this. Had then for years.

2) The all run various releases of the same distro (xfce Mint). And the
older ones have run various releases of this distro.

3) I remove the mains power from the mains-powered machines when I shut
them down.

4) The problem has only occured overnight, not during any daytime shutdown
and later restart.

5) Since it first manifested I have taken extra care to always use the
standard desktop sequence which leads me via the xfce desktop to being able
to click on the 'shutdown' icon/button and avoid the other choices.

So the implication of your diagnosis is that for some reason xfce Mint
versions have a problem on two specific machines and a desktop chosen
'shutdown' icon click causes a 'sleep' instead. Or that for some reason
only on two machines I keep clicking the wrong icon by mistake, but not on
the others.

BTW I don't ever shut the lid of a laptop when it is on. I shutdown the
system first. (True for the older laptop that has never shown this problem,
and for the new one.)

BTW2 The situation is complicated as I encountered the problem yesterday
morning and at the time didn't try holding the on button down for more than
about 2 mins to see if that helped. I then left it overnight until about
mid-day to-day, without it having any mains power. I was a bit worried that
there might be a battery problem. So it was left unpowered for well over 12
hours. I then tried a 5 min press - again no mains power - then released
the button and waited a short time. No sign of startup. So pressed the
button... and got a cigar at last! One of the first things I did was check
the battery state and the report was IIRC 97% charged. I don't know how
much charge a 'sleep' would have drawn, but that implies it must be tiny if
it was asleep.

Before all that I was previously using it with the mains supply / charger
connected as that is the norm because it is getting very heavy use. Full
CPU and fan going. Been doing that for months on most days without seeing
this problem.

So you might be correct, but I have reasons to find it hard to believe at
present.

Jim

[William Unruh]

On 2020-12-04, Jim Lesurf <no...@audiomisc.co.uk> wrote:

> In article <rqdsls$url$1...@dont-email.me>, William Unruh <un...@invalid.ca>
> wrote:
>> On 2020-12-04, Jim Lesurf <no...@audiomisc.co.uk> wrote:
>> > For some years one of the mains-powered Linux machines I use has had a
>> > habit of being reluctant to 'wake up in the morning'...
>> >
>> > I'd decided that it happened because I always power it down *and*
>> > switch off its mains supply every night. Although it didn't happen
>> > during the first year or more I used it. And, yes, I have regularly
>> > replaced the 'keeper' internal battery and this made no difference.
>
>> Switching of the mains should not make a difference. If you really shut
>> the machine down, and not just "sleep" it (which may well be what was
>> happening) then switching off the mains should be irrelevant.
>
> Yes, it should. On that I agree entirely.
>
> Alas, in this world 'should' isn't inevitably a synonym for 'does'. You
> only have to glance at Governments to notice this. :-)

Ah but governments are a bunch of people. Computers are a bunchor
hardware and software.

>
> cf below...
>
>> >
>> > However about 4 months ago I got a really fast nice new laptop. Been
>> > using it to process loads of files every day. Each day I mostly use it
>> > being powered with its mains PSU to ease the load on its internal
>> > batteries.
>> >
>> > Yesterday it failed to start when I pressed its 'wake up' button.
>> > Repeated tries with and without the external PSU failed.... Puzzled
>> > and worried.
>> >
>> > So this morning I tried holding down its 'wake up' button for 5 mins.
>> > [1] Released it. Nope, no coconut.
>
>> Sounds like it was asleep, not off, and when you held down the power
>> button for for 5 min actually shut it off. Then when you pressed it
>> again, it started up on its cold boot.
>
> I have reasons to doubt this, again cf below...

It was a guess, connected to the fact that you had to hold the power key
down for a long time-- which is typical of laptops when you have to
switch them off with the power button.

>
>> No, No mains connected computer assumes it will always be mains
>> connected. It would be a very silly design if it did. Power goes off for
>> many reasons (lightning, storms, moving the computer from one room to
>> another, power saving,....)
>
> I agree. But refer back to the lack of synonym for 'should'. :-)
>
> The conclusion that I am actually telling the machines to sleep seems odd
> when I add more information.

That you told them to sleep but then removed the power is not
inconsistant.

>
> 1) I also have two other mains powered machines, and another laptop. None
> of them ever do this. Had then for years.
>
> 2) The all run various releases of the same distro (xfce Mint). And the
> older ones have run various releases of this distro.
>
> 3) I remove the mains power from the mains-powered machines when I shut
> them down.
>
> 4) The problem has only occured overnight, not during any daytime shutdown
> and later restart.
>
> 5) Since it first manifested I have taken extra care to always use the
> standard desktop sequence which leads me via the xfce desktop to being able
> to click on the 'shutdown' icon/button and avoid the other choices.
>
> So the implication of your diagnosis is that for some reason xfce Mint
> versions have a problem on two specific machines and a desktop chosen
> 'shutdown' icon click causes a 'sleep' instead. Or that for some reason
> only on two machines I keep clicking the wrong icon by mistake, but not on
> the others.

Causality I have absolutely no insight into. And I certainly am not
saying that you are the cause. It is just that what I said sort of fits
at least one of the symptoms you stated, that is all. You are confused.
All I am trying to do is to give you another line of thought to persue.
That may well also be a dead end, but so has everything else you have
tried apparently.

[Spiros Bousbouras]

On Fri, 04 Dec 2020 14:04:54 +0000 (GMT)
Jim Lesurf <no...@audiomisc.co.uk> wrote:
> For some years one of the mains-powered Linux machines I use has had a
> habit of being reluctant to 'wake up in the morning'...
>
> I'd decided that it happened because I always power it down *and* switch
> off its mains supply every night. Although it didn't happen during the
> first year or more I used it. And, yes, I have regularly replaced the
> 'keeper' internal battery and this made no difference.
>
> However about 4 months ago I got a really fast nice new laptop. Been using
> it to process loads of files every day. Each day I mostly use it being
> powered with its mains PSU to ease the load on its internal batteries.
>
> Yesterday it failed to start when I pressed its 'wake up' button. Repeated
> tries with and without the external PSU failed.... Puzzled and worried.
>
> So this morning I tried holding down its 'wake up' button for 5 mins. [1]
> Released it. Nope, no coconut.
>
> I then pressed and released the button again, and yup! It started and then
> worked fine. After a while I shut it down, left it for a few hours. Then
> tried again. Yes, woke up again quite happy. So far, fine. And the battery
> shows as being well charged thoughout, so that wasn't 'flat' when it wasn't
> waking up.
>
> This is puzzling me on two levels:
>
> 1) Why in each case the behavior of being 'lazy' didn't show up for some
> time in use

I have no idea what's happening but are the 2 machines in the same room and
what's the coldest the room gets during the night ? I used to have a monitor
which after being turned off for hours in a cold room (10 degrees Celsius or
something) , had trouble restarting. It didn't do this for years but it
developed the problem at some point.

[Jim Lesurf]

In article <K1rVn8DO...@bongo-ra.co>, Spiros Bousbouras

<spi...@gmail.com> wrote:

> I have no idea what's happening but are the 2 machines in the same room
> and what's the coldest the room gets during the night ? I used to have a
> monitor which after being turned off for hours in a cold room (10
> degrees Celsius or something) , had trouble restarting. It didn't do
> this for years but it developed the problem at some point.

That is an interesting point because, yes, they are in the same room. It is
the 'dining room' and on the East side of the house. It is connected to the
kitchen which gets the coldest at night - however there is a closed door
between them.

And the last few nights have been cold here. East coast of Scotland.

Jim

[Theo]

Jim Lesurf <no...@audiomisc.co.uk> wrote:
> For some years one of the mains-powered Linux machines I use has had a
> habit of being reluctant to 'wake up in the morning'...
>
> I'd decided that it happened because I always power it down *and* switch
> off its mains supply every night. Although it didn't happen during the
> first year or more I used it. And, yes, I have regularly replaced the
> 'keeper' internal battery and this made no difference.

That's usually a symptom of the PSU being on the way out (dry capacitors
typically).

> However about 4 months ago I got a really fast nice new laptop. Been using
> it to process loads of files every day. Each day I mostly use it being
> powered with its mains PSU to ease the load on its internal batteries.
>
> Yesterday it failed to start when I pressed its 'wake up' button. Repeated
> tries with and without the external PSU failed.... Puzzled and worried.
>
> So this morning I tried holding down its 'wake up' button for 5 mins. [1]
> Released it. Nope, no coconut.

Computers without easily removable batteries need a way to do a hard power
off, for example for servicing, which in olden times we did by pulling the
battery.

Holding the power button down forces an electrical power off. If you hold
and release it doesn't come back on, it stays off. Pressing it briefly
again should then cause it to power up from power-off state, rather than any
kind of sleep state.

I think that covers your experience. Something caused it to stop
responding, and so you have to force power off by holding the button down.
Pressing it again woke it up. Having been hard power cycled, whatever
problem it was went away.

I'm a bit surprised a hold-down of 5 minutes was needed - it's normally more
like 10-15 seconds. What model is this laptop?

Theo

[Jim Lesurf]

In article <o-A*un...@news.chiark.greenend.org.uk>, Theo

<theom...@chiark.greenend.org.uk> wrote:


> For some years one of the mains-powered Linux machines I use has had a
> habit of being reluctant to 'wake up in the morning'...
>
> I'd decided that it happened because I always power it down *and* switch
> off its mains supply every night. Although it didn't happen during the
> first year or more I used it. And, yes, I have regularly replaced the
> 'keeper' internal battery and this made no difference.

That's usually a symptom of the PSU being on the way out (dry capacitors
typically).

I'm wondering if temperature affects it. However I also have the impression
that using it, or having it's mains connected even when I don't start it
up, in the evening makes it more happy in the morning after! Not managed to
really get stats, though, so just an impression.

> I think that covers your experience. Something caused it to stop
> responding, and so you have to force power off by holding the button
> down. Pressing it again woke it up. Having been hard power cycled,
> whatever problem it was went away.

> I'm a bit surprised a hold-down of 5 minutes was needed - it's normally
> more like 10-15 seconds. What model is this laptop?

Medion 'Erazer' I've not got the model number to hand but can try later to
find it.

It may be that 5 mins wasn't needed. But the previous day I tried 30secs,
then 2 mins, then gave up.

I had considered removing the batteries, but decided to try the '5 mins'
routine first, and got a coconut. Been fine today.

Jim

[Jim Lesurf]

In article <rqgg00$5qi$1...@dont-email.me>, William Unruh <un...@invalid.ca>
wrote:

> All I am trying to do is to give you another line of thought to persue.
> That may well also be a dead end, but so has everything else you have
> tried apparently.

Understood, and Fairy Snuff. :-) I should have given more details in the
first place, but wanted to keep it short in case that was redundant.