(SOLVED) Odd sleep/ACPI behaviour

If none of the more specific forums is the right place to ask

(SOLVED) Odd sleep/ACPI behaviour

Postby Mr. Lumbergh » 2020-07-17 03:47

I'm having an issue on Buster with KDE where following updates a couple of days ago, sleeping instead of fully shutting down results in USB 3 and HDMI audio failing upon restart. The mobo in this box has 1 USB-C and 4 USB 3 ports, and I have a camera card reader mounted in a 2.5" drive bay that has a USB 2 port. Upon restarting I can't log in or use the mouse plugged into the 3 ports; they're completely dead. The mouse has a few LED's to indicate settings that won't even light when it's doing this. I have to pull the keyboard dongle out of the 3 port on the back of the box and insert into the reader port on the front in order to log in, and after restarting there's still no power to any of the 3 ports. The C port comes back to life though.
A similar thing is happening with HDMI audio, if I start up from sleep mode it no longer recognizes the video card (Radeon RX580) as a sound device unless I walk over to the TV and unplug/replug the HDMI cable back in; this is again only happening in the last few days after recent updates. I frequently stream over to my TV so I know it works otherwise. System is Buster/KDE fully updated with an Asus X570 board.
If anyone can offer some advice on how to start tracking this down it would be appreciated.
EDIT: I may have gotten the HDMI audio issue figured out, if I change the volume on my TV it'll often bring the audio back even if the TV wasn't muted. I'm still having the issue with USB3 not restarting along with the rest of the system though so would still welcome some pointers on how to resolve.


Thanks
Last edited by Mr. Lumbergh on 2020-07-29 19:02, edited 1 time in total.
Mr. Lumbergh
 
Posts: 97
Joined: 2019-08-02 04:28

Re: Odd sleep/ACPI behaviour

Postby Mr. Lumbergh » 2020-07-22 22:32

Nobody has any ideas what might be causing this?
Mr. Lumbergh
 
Posts: 97
Joined: 2019-08-02 04:28

Re: Odd sleep/ACPI behaviour

Postby sickpig » 2020-07-22 23:37

Code: Select all
jourcalctl -S hh:mm:ss

the time would be since you last woke system
the log should list errors for further insight
User avatar
sickpig
 
Posts: 573
Joined: 2019-01-23 10:34

Re: Odd sleep/ACPI behaviour

Postby Deb-fan » 2020-07-23 07:48

Wasn't happening until upgrades, upgrades of what, from where? Do you have any non-standard software repositories or things like PPA's in your os's sources.list file? Could check things like /var/log/dpkg.log(s) for what was upgraded, installed, removed. Think that's supposed to cover it regardless of which pkg management tool is used, apt, aptitude or synaptic but all of them also have logs in that location too. At least apt/itude do and synaptic does log also. May not be so helpful now but if not already doing so, using a proven incremental backup/restore method is good juju and bunch of good ways in gnu/Linux, I like rsync. :)
Most powerful FREE tech-support tool on the planet * HERE. *
Deb-fan
 
Posts: 944
Joined: 2012-08-14 12:27

Re: Odd sleep/ACPI behaviour

Postby Mr. Lumbergh » 2020-07-23 15:18

Deb-fan wrote:Wasn't happening until upgrades, upgrades of what, from where? Do you have any non-standard software repositories or things like PPA's in your os's sources.list file? Could check things like /var/log/dpkg.log(s) for what was upgraded, installed, removed. Think that's supposed to cover it regardless of which pkg management tool is used, apt, aptitude or synaptic but all of them also have logs in that location too. At least apt/itude do and synaptic does log also. May not be so helpful now but if not already doing so, using a proven incremental backup/restore method is good juju and bunch of good ways in gnu/Linux, I like rsync. :)

Here's the set of dpkg logs from around the time it first started:
https://pastebin.com/CQzzNhFC
I have a couple of additional repos added, my VPN client and some music production packages, but nothing that I could see controlling USB3 state that didn't come from the official repos.
Mr. Lumbergh
 
Posts: 97
Joined: 2019-08-02 04:28

Re: Odd sleep/ACPI behaviour

Postby Mr. Lumbergh » 2020-07-23 15:22

sickpig wrote:
Code: Select all
jourcalctl -S hh:mm:ss

the time would be since you last woke system
the log should list errors for further insight

I've left my system up for the last three or four days instead of sleeping to try to get around this issue, so I don't have any logs pointing to this directly; I'll give it a try tonight.
Mr. Lumbergh
 
Posts: 97
Joined: 2019-08-02 04:28

Re: Odd sleep/ACPI behaviour

Postby Deb-fan » 2020-07-23 23:55

Majority of people(self included)are highly unlikely to click over or download anything to scroll through logs for somebody. You no doubt know this already. Would say clean them- the logs up and post them here or any other help threads you may start. Was also just pointing out the obvious connection between upgrades you did and resolving these issues. Also post your sources.list when seeking help like this and someone asks about it. I've got a couple things, this and that added to repositories isn't helpful and doesn't encourage anyone to try. From what you said, agree doesn't sound like a likely suspect involved is the issues you're having but others might see something or know something we don't there.
Most powerful FREE tech-support tool on the planet * HERE. *
Deb-fan
 
Posts: 944
Joined: 2012-08-14 12:27

Re: Odd sleep/ACPI behaviour

Postby Deb-fan » 2020-07-24 04:35

Curiosity got the better of me and yep, as expected you'd apparently upgraded the kernel. Done correctly I strongly encourage people to try different kernels and/or custom compile some. It's really interesting and cool stuff. Plus overall a no harm or foul proposition. Though whenever screwing around with the kernel or other core pkgs a higher level of care should be taken. Depending upon what someone is doing and why, upgrading the nix kernel and others is not something which should be done for the sake of a higher version number.

Only sayin.

Hey guys, I just upgraded my kernel (for no valid reason)and/other critical stuff and am getting all manner of power management weirdnesses and Os performance quirks! What oh ever could be my problem ya's thinks?! Errrr no kidding? No idea that something like that could happen, head scratcher, TotAlLy dude! :P

Also this is coming off more aholish than truly intended. Not meant as such but still stupid PSA type of thing. :) No insult intended OP only saying sometimes the everything's running smoothly or isn't broken don't fix thing is valid. Done right folks can play with the kernel or everything else to heart content, safely and w/o even the slightest chance of harm. Again common sense precautions we should all be doing anyway. Whenever you see a proposed upgrade of a kernel on an Os you use for stuff you care about, that should be a BIG reminder prompting a person to say, hmm when's the last time I did-made a backup? Before proceeding.
Most powerful FREE tech-support tool on the planet * HERE. *
Deb-fan
 
Posts: 944
Joined: 2012-08-14 12:27

Re: Odd sleep/ACPI behaviour

Postby pendrachken » 2020-07-24 12:31

Just glancing through the upgrade logs I see two possibilities - the kernel was upgraded, and Xen was also upgraded.

Hopefully you didn't delete your old, working, kernel ( always keep a know working spare kernel - ALWAYS). Fist thing first - reboot the machine, and choose the older known working kernel from the GRUB menu. If that kernel works you know there was a kernel regression. File a bug report with Debian against the kernel package in backports. Let them figure it out if it is security patches they applied that caused the regression, or if there is a bug in the kernel code. They can then file any bugs against the kernel code if needed.

It's also remotely possible that xen is grabbing the USB connection when waking from sleep for some reason. Personally, I think a kernel regression is way more likely, but this is also a possibility.


Hey guys, I just upgraded my kernel (for no valid reason)and/other critical stuff and am getting all manner of power management weirdnesses and Os performance quirks! What oh ever could be my problem ya's thinks?! Errrr no kidding? No idea that something like that could happen, head scratcher, TotAlLy dude! :P


Yeah totally! I mean it's not like the new kernel image came in with a standard system upgrade from the official backports repository, and newer kernels are needed from backports for more recent hardware.... oh wait.
fortune -o
Your love life will be... interesting.
:twisted: How did it know?

The U.S. uses the metric system too, we have tenths, hundredths and thousandths of inches :-P
pendrachken
 
Posts: 1378
Joined: 2007-03-04 21:10
Location: U.S.A. - WI.

Re: Odd sleep/ACPI behaviour

Postby Mr. Lumbergh » 2020-07-27 15:05

pendrachken wrote:Just glancing through the upgrade logs I see two possibilities - the kernel was upgraded, and Xen was also upgraded.

Hopefully you didn't delete your old, working, kernel ( always keep a know working spare kernel - ALWAYS). Fist thing first - reboot the machine, and choose the older known working kernel from the GRUB menu. If that kernel works you know there was a kernel regression. File a bug report with Debian against the kernel package in backports. Let them figure it out if it is security patches they applied that caused the regression, or if there is a bug in the kernel code. They can then file any bugs against the kernel code if needed.

It's also remotely possible that xen is grabbing the USB connection when waking from sleep for some reason. Personally, I think a kernel regression is way more likely, but this is also a possibility.

The kernel is looking more like the culprit; I started it up this morning with the 5.4 that's still installed. The keyboard shortcut not only put it to sleep but also woke it back up without trouble.
Now to file the bug report...
Thank you for helping me out on this one.
pendrachken wrote:
Hey guys, I just upgraded my kernel (for no valid reason)and/other critical stuff and am getting all manner of power management weirdnesses and Os performance quirks! What oh ever could be my problem ya's thinks?! Errrr no kidding? No idea that something like that could happen, head scratcher, TotAlLy dude! :P


Yeah totally! I mean it's not like the new kernel image came in with a standard system upgrade from the official backports repository, and newer kernels are needed from backports for more recent hardware.... oh wait.

Right? Silly me wanting security updates and current features. :roll:
Mr. Lumbergh
 
Posts: 97
Joined: 2019-08-02 04:28

Re: Odd sleep/ACPI behaviour

Postby Head_on_a_Stick » 2020-07-27 16:06

Mr. Lumbergh wrote:Silly me wanting security updates and current features

The stable kernel is an LTS branch and so receives security fixes and the new features added to the non-LTS kernels always carry the risk of introducing new bugs and vulnerabilities.
Black Lives Matter

Debian buster-backports ISO image: for new hardware support
User avatar
Head_on_a_Stick
 
Posts: 12647
Joined: 2014-06-01 17:46
Location: /dev/chair

Re: Odd sleep/ACPI behaviour

Postby Mr. Lumbergh » 2020-07-27 19:07

Head_on_a_Stick wrote:
Mr. Lumbergh wrote:Silly me wanting security updates and current features

The stable kernel is an LTS branch and so receives security fixes and the new features added to the non-LTS kernels always carry the risk of introducing new bugs and vulnerabilities.

I use this rig as a music production platform though so I need the realtime kernel; the latency is noticeably better with my MIDI controller.
Mr. Lumbergh
 
Posts: 97
Joined: 2019-08-02 04:28

Re: Odd sleep/ACPI behaviour

Postby Head_on_a_Stick » 2020-07-27 19:26

Mr. Lumbergh wrote:I need the realtime kernel

https://packages.debian.org/buster/linux-image-rt-amd64
Black Lives Matter

Debian buster-backports ISO image: for new hardware support
User avatar
Head_on_a_Stick
 
Posts: 12647
Joined: 2014-06-01 17:46
Location: /dev/chair

Re: Odd sleep/ACPI behaviour

Postby Deb-fan » 2020-07-28 22:50

Ignorantly upgrading and fiddling with things like an Os's kernel, very often isn't going to provide you with security and other enhancements or benefits. Absent putting in the effort or working up the knowledge to understand and handle what's involved competently, clearly can cause problems, as in your situation here. Though any risks involved can be easily mitigated to the point of making them non-issues ie: keeping backups etc.
Most powerful FREE tech-support tool on the planet * HERE. *
Deb-fan
 
Posts: 944
Joined: 2012-08-14 12:27

Re: Odd sleep/ACPI behaviour

Postby Mr. Lumbergh » 2020-07-28 23:19

Deb-fan wrote:Ignorantly upgrading and fiddling with things like an Os's kernel, very often isn't going to provide you with security and other enhancements or benefits. Absent putting in the effort or working up the knowledge to understand and handle what's involved competently, clearly can cause problems, as in your situation here. Though any risks involved can be easily mitigated to the point of making them non-issues ie: keeping backups etc.

Hmm. I'm not sure how it's "ignorant" to use the latest packages from the stable repos, but I do always keep a known-good kernel available as a fallback; currently I'm on the 5.4RT and it's behaving as expected. These days I only keep one or two fallbacks that I'm satisfied with as I recently had to remove a couple mainline kernels that were filling up the boot partition and preventing updates.
In any event, this does look like a regression and I've filed a bug report.
Mr. Lumbergh
 
Posts: 97
Joined: 2019-08-02 04:28

Next

Return to General Questions

Who is online

Users browsing this forum: No registered users and 21 guests

fashionable