Battery usage missing some hours on /e/OS 3.0.2 FP5

My FP5 on 3.0.2 a14 has started to not show battery usage for some hours. I am not sure whether I could have done anything to trigger this, not sure where to start looking (I mainly changed settings on RethinkDNS).

Does anyone have a similar problem, and what could be possible avenues of investigation?

That is interesting. From the behavior I would assume that the process of the battery manager system module doesn’t run without standby / automatic exit in the background for a prolonged time, but that shouldn’t be configurable for a system process (to not have the ability to run uninterrupted in the background).

You could probably check those settings in the application management under settings if you activate that system applications are shown, but I assume that they are not accessible for the module (greyed out).

I see a similar blank section, I think is your question (??) explained by an info box at the very foot of the Battery usage screen

Sometimes it may look like some approximations of time but if the usage looks highly unlikely especially just after an update it can be an indicator that Clear cache of affected apps may be a good idea.

I considered that additional text, however I am missing some hours without charging plugged in, too.

I can’t seem to find any system app which has any name that remotely suggests something to do with battery management… What should I be looking for?

Sorry, yes, I missed that second gap to “now”.

I wonder, now that we have rolled forward to Saturday does the display still show the second inappropriate gap unrecorded on Friday?

You are right, I checked the system processes - it is indeed an internal process / routine called “batterystats” and not separately listed in the applications (even system ones) anymore it seems.

From an older Android version I recall it being a stand-alone application (or at least the GUI being listed separately), but it is even better if it is integrated, as this forbids accidental changes on its “rights” regarding running in the background.

This means the error is likely not from my mentioned guessed reason.

It does not, actually… the Friday morning one is still there. Sometimes it does record charging, sometimes it doesn’t. Weird!

My feeling with “Results” where they may include a (small) mismatch seen between the top graph and the Selected portion forming the lower graph seems to fit with some inexact and / or different way to specify the time sample and the way in which perhaps some averaging is applied. (??)