Sv: Red Hat Enterprise Linux 8.6 (Thomas Friedrichsmeier)

[Anders Fridberger]

I attach output.log. The lines around "--server-name" are

/tmp/.mount_rkwardX2osMQ/usr/bin/../lib/R/bin/R
CMD
/tmp/.mount_rkwardX2osMQ/usr/lib/libexec/rkward.rbackend
--debug-level=1
--server-name=%2Ftmp%2FrkwardjmUDwKKe
--rkd-server-name=%2Ftmp%2FrkdnoZrBTJZ
--data-dir=%2Fdata%2Fhome%2Fanders%2F.rkward%2F
--locale-dir=
QMetaObject::invokeMethod: No such method KatePluginIntegrationWindow::createViewBar(KTextEditor::View *)


Best regards
Anders

            


            Från: Thomas Friedrichsmeier
Skickat: Fredag, 29 juli 2022 17:53
Till: Anders Fridberger
Kopia: rkward-devel at kde.org
Ämne: Re: Red Hat Enterprise Linux 8.6 (Thomas Friedrichsmeier)
            


            
Am Fri, 29 Jul 2022 14:32:51 +0000

schrieb Anders Fridberger <anders.fridberger at liu.se>:

> Thanks for the quick reply. I downloaded the Appimage, but the result

> is similar. The R backend dies. I ran it at debugging level 5 and got

> the same error as before in the log file:

> 

> Output type 2: Error: C stack usage  140726971572452 is too close to

> the limit



Ok. This is just the "final" symptom, however, things start going wrong

much earlier (and earlier indeed, than I have realized, previously), at:



Trace:

/data/home/anders/Coding/rkward-0.7.4/rkward/rbackend/rktransmitter.cpp

- function disconnected line 266



(line number is different in the appimage, but same issue).



It seems like the local socket that the frontend and backend process

use for communication does not work. I suspect that may be some

permission problem, here.



Unfortunately, the debug code that is supposed to provide some details

on this does not appear to write to rkward.frontend. I suspect it gets

written to stderr, instead. Could you run as:



  rkward --debug-level 4 --debug-output terminal 2> output.log



and post output.log (the relevant lines will be around the search string

"--server-name"), which should show, where creating the socket is

attempted?



> This time, there was an additional error message when rkward exited:

> 

> [489835:7:0729/161355.501350:ERROR:command_buffer_proxy_impl.cc(141)]

> ContextResult::kTransientFailure: Failed to send

> GpuChannelMsg_CreateCommandBuffer.



Probably something form QWebEngine. I'm not sure whether this is

something to worry about, but definitely looks like an unrelated

problem.



> I should perhaps mention that I'm running this on our server by

> tunneling the display to my mac using thinlinc

> (https://www.cendio.com, this is basically an implementation of VNC).



That could play a role, but again, I suspect a permission problem.



Regards



Thomas

-------------- next part --------------
A non-text attachment was scrubbed...
Name: output.log
Type: text/x-log
Size: 82565 bytes
Desc: output.log
URL: <http://mail.kde.org/pipermail/rkward-devel/attachments/20220729/e4d77695/attachment-0001.bin>