I use Fedora 42 workstation Gnome edition on Lenovo ThinkPad E15
beside Fedora I have also Windows 11 on same SSD as Fedora.
I monitor my ThinkPad and I see if I power off by Fedora 42 Gnome edition from Gnome, next time I turn on it, I see battery is use during turn off, for example before battery indicator show 54% of battery is remaining after turn off it show me 40% of battery is remain.
but when I turn off from Windows 11, I see battery draining less than Fedora 42.
Please guide me find why this is happen and where I made mistake during installation and config.
Run powertop to report the power usage in an html page and send the results to a pastebin. (sudo powertop -r powertop.html && fpaste --confirm powertop.html)
Post the URL it gives yuou back here.
This will show us what processes are using the most power. If you can do this on a quiet machine (no audio playing, no video, etc that would be useful, or you risk tainting the results with erroneous readings)
This sounds more like “suspend” than “power off”
With suspend the system remains minimally powered on, and everything remains active in memory. With power off the system is completely shutdown and there should be no additional power usage.
powertop output looks totally fine - you’d get a smidge more life if Telegram were not running, but it would be measured in seconds per hour. Wakeups are not excessive.
If you were super keen you could power off the ethernet adaptor and the wireless radios when you were not using them, along with bluetooth, but that would give it’s own annoyance of having a lag every time you wanted to utilise them, if they’d been idle long enough to sleep.
All in all, there’s nothing “wrong” with the power usage when the machine is awake, so it’s bizarre that you’re seeing a change in battery level when the machine is powered off (assuming it IS being powered off and not just put into suspend, as other have already mentioned).
Many systems support wake-on-lan and wake-on-wlan. This should put network interfaces in low power modes, but linux power management needs to work with vendor firmware, so there are often issues while waiting for firmware updates. If you don’t need to wake the system remotely, you should be able to disable network wake-up in UEFI/BIOS.
I use this workstation from Fedora 37 to Fedora 42, it upgraded from 37 to 42, is something remaining can make this problem from before and during upgrade?
I doubt anyone can easily answer the question about upgrades that have lasted over 6 releases of fedora, although there really should be nothing causing this except failure to fully power off. Wake on (w)lan could affect it as mentioned above since enough power has to be active to keep the interfaces active while waiting for a wake signal.
As stated before, a full power off should not continue to draw power while suspend does.
Newer versions of linux often have conflicts with existing firmware. In some cases there are hardware-specific workarounds that are removed in favour of newer standardized cross-platform “standards”. In other cases, existing hardware-specific code is inadvertently broken with updates because linux devs have newer gear. In that case it is up to user community to find the specific change and file a bug report with enough detail to allow someone without access to the hardware to fix the issue.
If you have access to a thermal camera you may be able identify components that stay warm with the system is supposed to be “off”.
