Sorry for the delay.
Thanks @q5sys for introducing the question into the Q&A and pasting the result back here and thanks to @kris for addressing the question.
I did some smart check a few days back but the only error the windows smart tool brought up was a interface crc error count of 2. If that was the problem the counter should go up with every boot - at least every boot of TrueOS because the BTX loader (or whatever it is that’s between the BTX loader and the BSD bootloader) still gives me that error. Also, changing the sata cable or switching from jmicron to intel sata plug on the mainboard didn’t change anything. So, that error most probably was a one time thing in/of the past.
The smart tool I installed on TrueOS gave me the same error message as the “native” error message I presented in the opening post. So, no help.
As chkdsk doesn’t “see” the bsd partition I had to “learn” how to do something similar on bsd. I ran zpool scrub but no errors were found. I’m not sure but as far as I understand it, the scrub command only repairs file system errors. The man didn’t say that it would also detect bad sectors and mark them as such.
Anyway, today I ran some ubcd tools. Both, the Samsung hdd test tool as well as another one that was specifically to check the whole disk surface and remap damaged sectors ran through without any errors. So nothing I can see or do right now.
For now, I decided to wait for the next stable update and this time run it normally. Due to the annoying server problems that lead to several “connection reset by peer” errors the last few times I just updated “in place”, i.e. pkg-static, because then the update wouldn’t download all the packages that are current again…
This problem cost me about 6 hours on my other system! Only to let me find out that (most probably) the system had not enough disk space left to update (into a new BE), because the system didn’t update during reboot (several times / tries) without telling me anything. That’s when I decided to go directly for the “static” root on the other system, too.
But maybe by letting the update process install into a new BE the boot entries get fixed / renewed and that annoying error message (that at the moment doesn’t seem to affect me / my system) goes away.
I’ll report back.