Each time a jump to the future is initiated it is acompanied by messages of the type 'rcu: INFO: rcu_sched self-detected stall on CPU'.
The return to my timeline can only be initiated by rebooting.
I started playing with timing for RAM and got so far the imagination that the interval from reboot to timetravel got longer with lower settings.
Yeah, new highscore. My #PinePhone running (unused) for more than 187000 seconds since last reboot with the correct clock.
Shouldn't this happen after reboot automatically and not just after the first login (however it is triggered at that moment)?
The phone is over-exhausting itself:
It took a long rest from #timetravel and stayed with me for more than 328000 seconds and then leapt into the far unknown year 2203 for the first time.
It seems to get help by journald which is in preparation stopped and started a few times.
I'll get it back by rebooting and look what it's got to tell.
apt-get update; apt-get upgrade
apt-get update; apt-get dist-upgrade
apt-get purge linux-image-5.7-pinephone linux-image-5.8-sunxi64
Here's the result (the first reboot hung, after switching it off and on again):
root@mobian:~# uname -a
Linux mobian 5.9-sunxi64 #3 SMP PREEMPT Tue Nov 3 16:24:16 UTC 2020 aarch64 GNU/Linux
I guess it worked. Apps I'm using work also.
Hey @mobian - GREAT WORK!
The ability to #timetravel improved greatly.
That might be because of the new u-boot which probably reset my lower clocked ram to the standard value.
@devrtz, what do you think?
I really look forward to my new mainboard. I can't wait to try it and compare the stability to the older one...
This means that my old mainboard these days will start its #timetravel back in my time forward to my history.
I'm curious whether the new board will behave different and maybe the problems with the clock will be gone.
BTW - the transport (standard) from @PINE64 to Germany took eight days. The parcel arrived by Deutsche Post and not by DHL. I didn't have to pay any fees or taxes.
Exchanged the mainboard and restored the image of the #mobian I already used on my old 2GB RAM / 16 GB eMMC mainboard.
With the old board it ran for a few days without problems already, once, but the clock usually jumped within 24h.
If the error does not return thereby confirming the suspected hardware problem, is there interest to look further into the problem, @PINE64?
Just in case somebody read the toot and didn't marvel over the little screenshot:
YES, THAT is the screenshot of an ssh session on my PHONE.
It becomes more and more usual and normal to see that, but it can't be emphasized enough ;-).
@chrichri It should be safe again from what I heard.
@chrichri @mobian @PINE64
Ah, the infamous timetraveling pinephone! :)
I'm afraid that I don't have a lot of insight in this regard.
I have a BH and UBports CE and from what I've seen my BH almost never travelled into 2115 (maybe once since I own it), while the other one does it more frequently.
So I wish you the best of luck with your new mainboard ;)
Uptime 330232 seconds without any glitch or clock error.
But why is the screen going on and off all the time while the phone lies on my desk?
Wonder whether that stops after the updates and the reboots.
Otherwise I'll look into the system dbus messages I saw at the same time the screen went on and off again (no, didn't push any button, just watched the serial console ;-) ).
Was there an announcement that it's safe to upgrade just yet? The thread on Pine64 forum still says otherwise: https://forum.pine64.org/showthread.php?tid=12021
Thanks for the information!
I'll have to look into updating my #mobian, soon. I held it back because of the major changes and hope that the update will be unproblematic again soon.
Actually I'm waiting for a new #mainboard to get 3GB RAM and a working USB-C, also and if the problem really is connected somehow to the RAM clock it might change after I exchanged the mainboard, anyway.