If you added the new kernel to a working image and had this problem, then tried the new image with the new kernel and had the same problem,
it tends to point squarely at the new kernel.

Either it is not a rt-preempt kernel, or it does not give off the right signals.

The second line in your dmesg is way short of what would be expected from an x86_64 rt-preempt kernel
You just have SMP
I would expect SMP RT PREEMPT

The kernel name having 'rt' in it irrelevant, as the kernel builder can attach `-giraffe` prefix, but that does not make it one.

If you want to look in src/rtapi/ and grep 'flavor' you will see where the determination is done.

The MKUUID issue was probably caused by using a new version of machinekit on an old image base which did not contain 'genuuid'

The root cause is that the new kernel is not recognised as rt-preempt and flavour defaults to posix.

Since you only have rt-preempt binaries etc. it errors.


On 14/02/19 22:37, Condit Alan wrote:
I built another fresh uSD from the image I got from Robert Nelson, I installed ssh, xorg, openbox, and uuid-runtime. I tried running Machinekit again. And it is back to trying to load rtapi_app_posix.

Where does machinekit find “flavor”?

Here is a link to the output from the attempt to start Machinekit, the output from linuxcnc.log, linuxcnc_debug.txt, linuxcnc_print.txt, machinekit.ini and mkuuid. And yes, I know the the output from /var/log/linuxcnc.log is only one line long.

Here is a link to output from dmesg:
https://pastebin.com/UFveVH2f

Alan

On Feb 14, 2019, at 10:00 AM, [email protected] wrote:

I don't know why this would fail, unless new packages were somehow installed on a system that does not have `uuidgen` installed.

Just generate a new MKUUID with uuidgen (if you don't have it, that is your problem right there, you need to install `uuid-runtime`)
and put it in your /etc/linuxcnc/machinekit.ini

Then try again.

(Need a link to the whole linuxcnc.log for the failed run next time, not just the line you think is important.  There is no context otherwise)

On 14/02/19 16:44, Condit Alan wrote:
after first startup and fail there was a 0 length file mkuuid in /etc/linuxcnc
and "MKUUID=" in machinekit.ini

on subsequent attempts to run it it tries to load rtapi_app_posix

Is there a simple way to reset machinekit to its initial state (like it has never been run yet)???


machinekit@beaglebone:~$ cat linuxcnc_debug.txt
2081
  PID TTY      STAT   TIME COMMAND
Stopping realtime threads
Unloading hal components

machinekit@beaglebone:~$ cat linuxcnc_print.txt
RUN_IN_PLACE=no
LINUXCNC_DIR=
LINUXCNC_BIN_DIR=/usr/bin
LINUXCNC_TCL_DIR=/usr/lib/tcltk/linuxcnc
LINUXCNC_SCRIPT_DIR=
LINUXCNC_RTLIB_DIR=/usr/lib/linuxcnc
LINUXCNC_CONFIG_DIR=
LINUXCNC_LANG_DIR=/usr/share/linuxcnc/tcl/msgs
INIVAR=/usr/libexec/linuxcnc/inivar
HALCMD=halcmd
LINUXCNC_EMCSH=/usr/bin/wish8.6
INIFILE=/home/machinekit/machinekit/configs/ARM.BeagleBone.Xylotex/Xylotex.ini
PARAMETER_FILE=pru-stepper.var
TASK=milltask
HALUI=halui
DISPLAY=axis
Starting Machinekit server program: linuxcncsvr
Loading Real Time OS, RTAPI, and HAL_LIB modules
Starting Machinekit IO program: io
Starting HAL User Interface program: halui
Killing task linuxcncsvr, PID=2081
Removing HAL_LIB, RTAPI, and Real Time OS modules
Removing NML shared memory segments

machinekit@beaglebone:~$ cat /var/log/linuxcnc.log
Feb 14 00:03:43 beaglebone msgd:0: no [MACHINEKIT]MKUUID found in /etc/linuxcnc/machinekit.ini



--
website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit
---
You received this message because you are subscribed to the Google Groups "Machinekit" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Reply via email to