One of my workstations refuses to stay asleep. Often,
just after being COMMANDED to sleep, it would reawaken.
Sometimes, a short while later (i.e., just long enough
for me to leave the room and return to find it running,
again).

Windows logs suck. But, I was able to trace the cause
of the wake-up to "USB root hub" (I think there are
six things with this name in the device tree!).

I narrowed it down to the keyboard; unplug keyboard,
command it to sleep and it stays asleep.

[N.B. The keyboard works completely as expected when powered on!]

Fine. Replace keyboard, problem solved.

*BUT*, I can't EASILY get any more of this particular model
of keyboard (eBay-roulette?) so I would like to salvage it
given that the "problem" is likely a hardware issue.

First question is what *might* (likely) Windows be seeing that
makes it think that a valid USB event has been detected?
I.e., is it a "smart" pattern recognizer (like a WoL packet)?
Or, just "any activity" on the data lines to the i/f?

It seems that power conditioning on the keyboard should be a good
starting place (it's ALWAYS power-related!).

It also seems like Windows only leaves USB ports powered up
during sleep *if* they were active while the machine was awake (?).

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 31-Jan-25 10:25 am, Don Y wrote:

One of my workstations refuses to stay asleep.  Often,
just after being COMMANDED to sleep, it would reawaken.
Sometimes, a short while later (i.e., just long enough
for me to leave the room and return to find it running,
again).

Windows logs suck.  But, I was able to trace the cause
of the wake-up to "USB root hub" (I think there are
six things with this name in the device tree!).

I narrowed it down to the keyboard; unplug keyboard,
command it to sleep and it stays asleep.

[N.B. The keyboard works completely as expected when powered on!]

Fine.  Replace keyboard, problem solved.

*BUT*, I can't EASILY get any more of this particular model
of keyboard (eBay-roulette?) so I would like to salvage it
given that the "problem" is likely a hardware issue.

First question is what *might* (likely) Windows be seeing that
makes it think that a valid USB event has been detected?
I.e., is it a "smart" pattern recognizer (like a WoL packet)?
Or, just "any activity" on the data lines to the i/f?

It seems that power conditioning on the keyboard should be a good
starting place (it's ALWAYS power-related!).

It also seems like Windows only leaves USB ports powered up
during sleep *if* they were active while the machine was awake (?).

It appears to me that the device drives a 'K' state onto the data lines
for between 1 and 15ms. The 'K' state is the opposite of the idle state.

It's not necessarily the keyboard's fault - unplugging it prevents it
from waking the system up, if that's what it's doing, but it also means
that it's not plugged in. It could be the latter fact that makes the
spurious wake-ups go away.

You could try plugging something else into that port instead.

Also plugging the keyboard in via a separate usb hub.

The results may hint at what to look at next.

Sylvia.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 2025-01-31, Don Y <[email protected]d> wrote:

One of my workstations refuses to stay asleep. Often,
just after being COMMANDED to sleep, it would reawaken.
Sometimes, a short while later (i.e., just long enough
for me to leave the room and return to find it running,
again).

Windows logs suck. But, I was able to trace the cause
of the wake-up to "USB root hub" (I think there are
six things with this name in the device tree!).

I narrowed it down to the keyboard; unplug keyboard,
command it to sleep and it stays asleep.

[N.B. The keyboard works completely as expected when powered on!]

Fine. Replace keyboard, problem solved.

*BUT*, I can't EASILY get any more of this particular model
of keyboard (eBay-roulette?) so I would like to salvage it
given that the "problem" is likely a hardware issue.

plug it into a port that doesn't have standby power.


--
Jasen.
🇺🇦 Слава Україні

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Thu, 30 Jan 2025 19:25:51 -0700, Don Y
<[email protected]d> wrote:

One of my workstations refuses to stay asleep. Often,
just after being COMMANDED to sleep, it would reawaken.
Sometimes, a short while later (i.e., just long enough
for me to leave the room and return to find it running,
again).

Windows logs suck. But, I was able to trace the cause
of the wake-up to "USB root hub" (I think there are
six things with this name in the device tree!).

I narrowed it down to the keyboard; unplug keyboard,
command it to sleep and it stays asleep.

[N.B. The keyboard works completely as expected when powered on!]

Fine. Replace keyboard, problem solved.

*BUT*, I can't EASILY get any more of this particular model
of keyboard (eBay-roulette?) so I would like to salvage it
given that the "problem" is likely a hardware issue.

First question is what *might* (likely) Windows be seeing that
makes it think that a valid USB event has been detected?
I.e., is it a "smart" pattern recognizer (like a WoL packet)?
Or, just "any activity" on the data lines to the i/f?

It seems that power conditioning on the keyboard should be a good
starting place (it's ALWAYS power-related!).

It also seems like Windows only leaves USB ports powered up
during sleep *if* they were active while the machine was awake (?).

Do you have a cat?

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 1/31/2025 9:22 AM, Sylvia Else wrote:

It also seems like Windows only leaves USB ports powered up
during sleep *if* they were active while the machine was awake (?).

It appears to me that the device drives a 'K' state onto the data lines for between 1 and 15ms. The 'K' state is the opposite of the idle state.

Said another way, does the exchange contain "data" (a keystroke) or
is it purely for notification?

I.e., if the exchange *contained* a keystroke, then that would
suggest that I should see spurious keystrokes during normal
operation (?) -- which I don't.

It's not necessarily the keyboard's fault - unplugging it prevents it from waking the system up, if that's what it's doing, but it also means that it's not plugged in. It could be the latter fact that makes the spurious wake-ups go
away.

Yes. I rely on the keyboard to (eventually) wake the machine
/when I want to/ (accessing the power button is tedious).

You could try plugging something else into that port instead.

I tried a different keyboard (same make/model -- all of my keyboards
are intentionally the same make/model). This worked correctly.
Note the offending keyboard ALSO worked correctly -- until recently.

Also plugging the keyboard in via a separate usb hub.

The results may hint at what to look at next.

The *right* way to do it, of course, would be to actually
look at the interface signals. But, that's not particularly
convenient. (Note to self: build a USB breakout box)

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 1/31/2025 4:10 PM, Jasen Betts wrote:

*BUT*, I can't EASILY get any more of this particular model
of keyboard (eBay-roulette?) so I would like to salvage it
given that the "problem" is likely a hardware issue.

plug it into a port that doesn't have standby power.

But then I wouldn't be able to use it to wake the workstation
when desired.

Note, this *was* working correctly. And, an identical keyboard
plugged into the same physical port STILL works correctly.
I.e., something has changed *in* the keyboard.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 2025-02-01, Don Y <[email protected]d> wrote:

On 1/31/2025 4:10 PM, Jasen Betts wrote:

*BUT*, I can't EASILY get any more of this particular model
of keyboard (eBay-roulette?) so I would like to salvage it
given that the "problem" is likely a hardware issue.

plug it into a port that doesn't have standby power.

But then I wouldn't be able to use it to wake the workstation
when desired.

Note, this *was* working correctly. And, an identical keyboard
plugged into the same physical port STILL works correctly.
I.e., something has changed *in* the keyboard.

Other than posting INCOPMPLETE requirements on usenet what have you
tried?

--
Jasen.
🇺🇦 Слава Україні

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 01-Feb-25 8:36 am, Don Y wrote:

On 1/31/2025 9:22 AM, Sylvia Else wrote:

It also seems like Windows only leaves USB ports powered up
during sleep *if* they were active while the machine was awake (?).

It appears to me that the device drives a 'K' state onto the data
lines for between 1 and 15ms. The 'K' state is the opposite of the
idle state.

Said another way, does the exchange contain "data" (a keystroke) or
is it purely for notification?

If I appear to be double posting, it's because some of my postings seem
to be vanishing.

On my reading of the spec, it just swaps the data line levels for a
period, and then swaps them back. I wouldn't expect any data transfer,
because USB devices do not spontaneously send data, they respond only
when polled.

I suspect it's going to come down to the voltage levels on the data+ and
data- lines. Data+ should stay high (above 3V), and data- should stay
low. Time to build that USB breakout box.

Sylvia.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 2/1/2025 1:22 AM, Sylvia Else wrote:

On 01-Feb-25 8:36 am, Don Y wrote:

On 1/31/2025 9:22 AM, Sylvia Else wrote:

It also seems like Windows only leaves USB ports powered up
during sleep *if* they were active while the machine was awake (?).

It appears to me that the device drives a 'K' state onto the data lines for >>> between 1 and 15ms. The 'K' state is the opposite of the idle state.

Said another way, does the exchange contain "data" (a keystroke) or
is it purely for notification?

On my reading of the spec, it just swaps the data line levels for a period, and
then swaps them back. I wouldn't expect any data transfer, because USB devices
do not spontaneously send data, they respond only when polled.

That is not contradictory to my observations.

The keyboard doesn't seem to generate any "bogus" keystrokes (like a
key being stuck). I have made a crude attempt at verifying this by
creating a Notepad document with contents similar to:

..................
........X.........
..................

and positioned the cursor just after the 'X' with the 'X' highlighted as "selected text"; the thinking being that any keystrokes will either cause
the X to be replaced, deleted or "moved", depending on their nature.

No change (with the good *or* bad keyboard).

If things are as you suggest, then the keyboard may be *continuously* signalling for attention and Windows simply notes "no data" -- as if
this was a "spurious interrupt".

OTOH, if the PC was sleeping, it would be awakened to query the keyboard
and would THEN discover "no data", as above. But, having been awakened,
it would be foolish to return to sleep (imagine the sort of loop this
could create).

I suspect it's going to come down to the voltage levels on the data+ and data-
lines. Data+ should stay high (above 3V), and data- should stay low. Time to build that USB breakout box.

Yup. I wanted a theory to pursue rather than just poking around and
hoping to stumble on something. I'll have a look at the power distribution *inside* the keyboard (as the working keyboard suggests what's coming
from the PC is "nominal" -- at least in the case of the mouse, keyboard
and other USB peripherals)

Thanks!

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

In article <vnjqe8$3ojnb$[email protected]>,
Don Y <[email protected]d> wrote:

Note, this *was* working correctly. And, an identical keyboard
plugged into the same physical port STILL works correctly.
I.e., something has changed *in* the keyboard.

Three thoughts:

(1) Does this keyboard have an integral touch-pad or joystick?
If so, possibly some spurious touch or motion detection in
the electronics there might cause the keyboard to send in
a motion event when polled.

(2) I once had a problem with a Macintosh II which would spontaneously
power up (not just come out of idle). This would happen in the
mornings, not long after my wife and I got up and started our
days. The chain of events turned out to be (cold house),
(hot shower), (steam and moisture traveling down the hall and
down into the den), (moisture condensing on the cold keyboard
PC-board), (very slight current leak between ground and the
ADB "power on!" button pads), (BONG!)

Cleaning the PC board in the area of the power-on key's pads,
and then dabbing the traces with some clear nail polish,
sealed the high-impedance area away from moisture and the
problem went away.

Any chance you've had a fluid spill in the keyboard? Giving
it a good flush-out with an electronics-compatible cleaner,
a distilled water rinse, and a final rinse with isopropyl
might eliminate the problem.

(3) Every time I've used a PC with a "wake up from sleep" capability,
its BIOS let me set which sorts of events were and were not
allowed to bring it out of hibernation. If you check your BIOS
you may be able to disable the "wake up on keyboard/mouse activity"
feature.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 2/1/2025 6:12 PM, Dave Platt wrote:

In article <vnjqe8$3ojnb$[email protected]>,
Don Y <[email protected]d> wrote:

Note, this *was* working correctly. And, an identical keyboard
plugged into the same physical port STILL works correctly.
I.e., something has changed *in* the keyboard.

Three thoughts:

(1) Does this keyboard have an integral touch-pad or joystick?
If so, possibly some spurious touch or motion detection in
the electronics there might cause the keyboard to send in
a motion event when polled.

No. And, any sort of "action event" would also leave side-effects
visible on the screen.

E.g., it would "type" a letter/number/symbol into the Notepad
document that I described elsewhere. Or, would move the text cursor.
Or, the MOUSE cursor. Or, enable/disable caps/scroll/num -lock
indicators. Or, invoke HELP (F1), Find Next (F3), time/date (F5),
move to the beginning of the line (HOME), end of line (END), previous
page (PGUP), next page (PGDN), etc.

I.e., *most* keystrokes would leave visible evidence of their
occurrence.

AND, would occur while the machine was NOT sleeping, as well.

So, any event either has to be something that has no visible
consequences (e.g., a shift/control key spontaneously claiming
to be active -- but, why not visible WHILE in normal use???)
OR, as Sylvia suggested, just some spurious notification
that serves to awaken the PC and, when *it* polls the keyboard,
it sees "nothing to report".

If it can so reliably awaken the machine EACH time it is commanded
to sleep, then you would assume it would manifest while I was actively
using the machine!

[The "spurious notification" fits the bill; I would never notice if the keyboard was routinely saying "look at me!" -- if, when the PC
actually "looked", it saw NOTHING! Just some extra bus activity...]

(2) I once had a problem with a Macintosh II which would spontaneously
power up (not just come out of idle). This would happen in the
mornings, not long after my wife and I got up and started our
days. The chain of events turned out to be (cold house),
(hot shower), (steam and moisture traveling down the hall and
down into the den), (moisture condensing on the cold keyboard
PC-board), (very slight current leak between ground and the
ADB "power on!" button pads), (BONG!)

The keyboards on my first two Compaqs would (over time) exhibit
flakey behavior due to high impedance implementation. I would
periodically disassemble the keyboards -- including removing the PCB
behind the keys -- and wipe everything down with alcohol to
restore reliable operation.

These don't seem to have a similar problem (I have 6 or 7 of these
in service, presently).

Cleaning the PC board in the area of the power-on key's pads,
and then dabbing the traces with some clear nail polish,
sealed the high-impedance area away from moisture and the
problem went away.

Any chance you've had a fluid spill in the keyboard? Giving

No. And, I don't smoke (smokers' keyboards are the pits!)

it a good flush-out with an electronics-compatible cleaner,
a distilled water rinse, and a final rinse with isopropyl
might eliminate the problem.

I periodically clean all of the keyboards as the keycaps inevitably
collect the oils/dirt from fingers. Remove keycaps and soak in
soapy water. Vacuum keyboard assembly (dust and crud that get
"behind" the keycaps -- nothing can get *into* the keyboard as the
keycaps cover the only points into the mechanisms). Wipe down with
alcohol (soapy water would migrate into the keyboard but the alcohol
will evaporate quickly).

Keyboards end up looking "like new" :>

(3) Every time I've used a PC with a "wake up from sleep" capability,
its BIOS let me set which sorts of events were and were not
allowed to bring it out of hibernation. If you check your BIOS
you may be able to disable the "wake up on keyboard/mouse activity"
feature.

But I *want* the wake-on-USB feature. I just want it to work as
intended; not as if invisible actors were at play!

I leave my workstations on 24/7/365. I count on them "sleeping"
when I turn my attention away from them (to other activities).
Then, expect them to come back online quickly when I click mouse
or press a key (MOVING the mouse doesn't seem to cause a wake event).

It took me some time to notice this "problem" because I can never
be sure when I've been "away" long enough for a machine to sleep.
So, if it is awake when I return, I assume I've returned quicker
than the sleep timer. If it's asleep, I figure I was "away"
for longer.

Sleeping is a costly event as all of memory has to be flushed
out to disk. With 100MB/s disks, it takes a while to flush
100+GB! (it's really annoying to find myself waiting for it
to finish GOING to sleep before I can wake it back up! :< )

[Some apps apparently disable this inactivity reporting. And,
some apps have other timeouts -- e.g., I can have a TELNET
session active and the machine can sleep but the session
won't automatically close unless the machine has been asleep
longer than the keepalive interval for the session]

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)