KDE Plasma (5.9.4, QT 5.7.1 on Fedora 25) reports:

Code: Select all

/usr/bin/google-chrome-stable is currently suppressing power management: Uploading data to likeabosh.hipchat.com

Thinking this was a problem with Hipchat, I closed my Hipchat tab and verified from chrome://net-internals that no sockets to Hipchat remained open. The message remained. Then, thinking it was a problem with Chrome, I shut down Chrome and verified that all Chrome processes were indeed gone:

Code: Select all

# ps -ef | grep chrome
root     18594  9487  0 14:44 pts/1    00:00:00 grep --color=auto chrome

And yet the message that chrome is suppressing power management still remains. Ideas?

Regards,
Raman

The power management suppression seems to be flaky. Often times, Chrome is running and suppressing power management, but nothing shows up in the Battery Monitor. Other times, Chrome is running and suppressing power management, and the battery monitor does at least report that. And yet other times, has reported in the OP, power management is reportedly suppressed due to Chrome, but Chrome isn't even running.

Lastly, even if power management is suppressed, but I lock my desktop, power management should be re-enabled. Obviously if I've locked my desktop there is no reason to suppress power management, even if some random program thinks there is. And yet, KDE does not put the monitors to sleep due to suppressed power management even with the desktop locked and not in use.

Is this area going to get some attention and love from the KDE devs? Are there issues that I can follow and help provide additional info?

I'm seeing almost the exact same issue with the integrated Skype client of Outlook.com email.
The exception being that I'm able to work around the issue by closing the tab with outlook.com email before I lock the PC and walk away from it. And retrieving the tab with Ctrl+Shft+T when I'm back.
I recently posted at the Arch Linux forum about this but received no response (yet) : https://bbs.archlinux.org/viewtopic.php?id=235813

rocketraman wrote:Is this area going to get some attention and love from the KDE devs? Are there issues that I can follow and help provide additional info?

Not sure this is the fault of KDE...I was using Chrome but after some updates the system kept preventing my monitors from going to sleep and thought it was a KDE or Ubuntu base issue. I had to manually turn off my monitors. After Firefox Quantum came out I gave it a try and immediately I dumped Chrome and starting using Firefox again. WOW...huge difference in memory usage and speed - shows what a hog Chrome is on a system. Oddly enough my monitors would go to sleep as expected, indicating it might have been a Chrome update that messed up my sleepy monitors.

Note that playing a video causes Firefox to suppress power management and my monitors don't go to sleep - which is correct. Pause a video and they go to sleep on time again. Yay.

I think this is a Chrome issue...even if it's not...Chrome is so bloated these days...

rocketraman wrote:Is this area going to get some attention and love from the KDE devs? Are there issues that I can follow and help provide additional info?

I'd suggest posting this to http://bugs.kde.org and filing it under "powerdevil", which is the power management component. It'll get emailed to the right people and go into the official queue for investigation and fixing.

It's definitely a weird issue and hopefully it can be addressed.

compatico wrote:

rocketraman wrote:Is this area going to get some attention and love from the KDE devs? Are there issues that I can follow and help provide additional info?

Not sure this is the fault of KDE...



I think this is a Chrome issue...

I'm a full stack web designer and developer. I need to test in all browsers. I can't discriminate, and pick just one browser.

I agree though that it's not a Plasma issue. But Plasma should (through an optional setting) be able to take control and enforce the power settings. It's a matter of who is in control. The OS+Desktop+User or Google Chrome ?
I also test on Gnome (on a different PC) where Google Chrome apparently is forced into submission. And Deepin desktop exhibits the same issue, Google Chrome prevents sleep and hibernation (also on a different PC).

But Plasma should (through an optional setting) be able to take control and enforce the power settings

App asks for power management to be disabled, we honor that request.

There is a bug that keeps the screen awake even if Chrome requests to only prevent standby mode which I fixed recently. [1] Perhaps that mitigates your problem.

[1] https://cgit.kde.org/powerdevil.git/com ... b191f419a0

kbroulik wrote:

But Plasma should (through an optional setting) be able to take control and enforce the power settings

App asks for power management to be disabled, we honor that request.

There is a bug that keeps the screen awake even if Chrome requests to only prevent standby mode which I fixed recently. [1] Perhaps that mitigates your problem.

[1] https://cgit.kde.org/powerdevil.git/com ... b191f419a0

I understand that that is what happens. But the thing is that I do not use Skype.
And according to Microsoft it's not possible to completely disable Skype in outlook.com. I wrote that in my linked post at the Arch forum, with screenshot of the MS web page.

The way I see it is, that it are my PCs. So I should be in control. If I choose for all the PCs in my test-lab and my dev-PC to hibernate or sleep, it should happen. Google, Microsoft, etc don't pay my utility bills. Since yesterday, I started with forwarding all mail from outlook.com to a new email address i created on one of my own test server domains. Probably the better solution to stay in control. I'm preparing an email list right now to send a change request to everyone concerned.

Anyway, thanks for your effort with the fix. Maybe it will resolve the issue. But I'm taking control.