On Tue, 2 Aug 2022 16:06:10 +0000 Anders Fridberger <anders.fridber...@liu.se> wrote: > Sorry, I tried #46 but no luck. The GUI comes up nicely, no warnings > or anything, and then I waited a couple of minutes before trying to > run some commands. But no luck. So I tried restarting the backend but > then I was thrown back to the command line. Logs attached (was > running at debug level 5) Best Anders
Sigh... I really had high hopes this time. But in hindsight, my theory about ill-patched version of Qt was flawed from the start: The AppImage includes a known to work version, after all. Anyway, the result hardens the evidence that something is going wrong at a rather low level. Trace shows the same thing, essentially: The thread trying to read the token from stdin somehow locks up, despite the workaround. I wonder what it's doing... 1) Could you run (I suggest using your own build, not the AppImage; that will have the best debugging info): gdb build/rkward/rkward (replacing path as required). Then at the prompt enter start then cont After a few seconds hit Ctrl+C. Then thread apply all bt In the output (probably several pages), find the thread that contains RKFrontendTransmitter::run(), and copy and paste that section. Perhaps this will at least give us some better keywords to search for the problem on the net. 2) I can't help thinking that it's a config/permissions issue, somehow. If it isn't SELinux, perhaps its fapolicyd, or something? 3) I don't think it's a good idea as a "solution", but I intend to add a "--quirkmode" option to allow to skip the handshake check. Essentially to make sure, that the current problem is the "only" obstacle. Not sure, whether I'll make that in time for the next AppImage build, but I'll keep you posted. 4) Once more thanks for your continued patience! Regards Thomas
pgpCTmMIEwF6b.pgp
Description: OpenPGP digital signature
