hm2/hm2_7i92.0: Error finishing read!

Usually it is Joint 2, sometimes joint 1 or joint 0. I have to shut down the 7i72 card and restart linuxcnc.
The error is running a program, but sometimes I have gotten the error Error finishing read on 7i92 with no movement of the motors. What could this be due to?
I have a Mesa 7i72 board with a 7i76 connected to an orange PI5, The network cable is Cat6, the motors are Chinese servomotors connected as a direction step, I have executed several times the command:

sudo chrt 99 ping -i .001 -q 192.168.1.1 -c 50000

To see if packets are lost and the result is always 0% packets lost.

I have lowered the maximum speed of the engines in the ini file, I don't know what can happen. Any idea?

Angel

what is the result of running

 sudo chrt 99 ping -i .001 -q 192.168.1.121 -c 50000

(the times are important)

 

Change the default IP address with:
mesaflash --device 7I92T --addr 192.168.1.121 --set ip=10.10.10.100.

I have W5 down W6 up

The results are :
sudo chrt 99 ping -i .001 -q 10.10.10.100 -c 50000
PING 10.10.10.100 (10.10.10.100) 56(84) bytes of data.

--- 10.10.10.100 ping statistics ---
50000 packets transmitted, 50000 received, 0% packet loss, time 53160ms
rtt min/avg/max/mdev = 0.017/0.020/0.997/0.026 ms

sudo chrt 99 ping -i .001 -q 10.10.10.100 -c 50000
PING 10.10.10.100 (10.10.10.100) 56(84) bytes of data.

--- 10.10.10.100 ping statistics ---
50000 packets transmitted, 50000 received, 0% packet loss, time 52051ms
rtt min/avg/max/mdev = 0.016/0.019/1.238/0.026 ms

Pinging 10.10.10.100 means you are pinging your PC, not the 7I92
What you need is:

sudo chrt 99 ping -i .001 -q 10.10.10.10 -c 50000


What host hardware do you have?

Sorry for my ignorance, I don't understand network addresses.
Here are the results:

sudo chrt 99 ping -i .001 -q 10.10.10.10 -c 50000
[sudo] contraseña para angel:
PING 10.10.10.10 (10.10.10.10) 56(84) bytes of data.

--- 10.10.10.10 ping statistics ---
50000 packets transmitted, 50000 received, 0% packet loss, time 50007ms
rtt min/avg/max/mdev = 0.161/0.163/0.462/0.002 ms
angel@orangepi5:~$ sudo chrt 99 ping -i .001 -q 10.10.10.10 -c 50000
PING 10.10.10.10 (10.10.10.10) 56(84) bytes of data.

--- 10.10.10.10 ping statistics ---
50000 packets transmitted, 50000 received, 0% packet loss, time 50003ms
rtt min/avg/max/mdev = 0.161/0.163/1.240/0.005 ms
angel@orangepi5:~$ sudo chrt 99 ping -i .001 -q 10.10.10.10 -c 50000
PING 10.10.10.10 (10.10.10.10) 56(84) bytes of data.

--- 10.10.10.10 ping statistics ---
50000 packets transmitted, 50000 received, 0% packet loss, time 50006ms
rtt min/avg/max/mdev = 0.161/0.163/0.215/0.001 ms



Thanks

0.161/0.163/1.240/0.005 ms

That's more that 1 ms so a reason for timeouts

You might try a 2 ms servo thread
(2000000 ns in the ini file)SERVO_PERIOD = 2000000

Thanks for your answers, I'll try it tomorrow.

Thanks PCW for your reply, I remembered that I had compiled the Orange Pi 5 core and checked the network interrupts and they were different from the ones I had isolated before, so I changed the network IRQs to the isolated cores and now I get better results. I have done several tests and it has not failed again.

angel@orangepi5:~$ sudo chrt 99 ping -i .001 -q 10.10.10.10 -c 50000
PING 10.10.10.10 (10.10.10.10) 56(84) bytes of data.

--- 10.10.10.10 ping statistics ---
50000 packets transmitted, 50000 received, 0% packet loss, time 50002ms
rtt min/avg/max/mdev = 0.148/0.156/0.345/0.001 ms
angel@orangepi5:~$ sudo chrt 99 ping -i .001 -q 10.10.10.10 -c 50000
PING 10.10.10.10 (10.10.10.10) 56(84) bytes of data.

--- 10.10.10.10 ping statistics ---
50000 packets transmitted, 50000 received, 0% packet loss, time 50001ms
rtt min/avg/max/mdev = 0.155/0.156/0.186/0.000 ms
angel@orangepi5:~$ sudo chrt 99 ping -i .001 -q 10.10.10.10 -c 50000
PING 10.10.10.10 (10.10.10.10) 56(84) bytes of data.

--- 10.10.10.10 ping statistics ---
50000 packets transmitted, 50000 received, 0% packet loss, time 50001ms
rtt min/avg/max/mdev = 0.155/0.156/0.186/0.000 ms

That looks much better.

You can monitor the maximum times in when LinuxCNC is running with:

halcmd show param *.tmax

(these times will be in ns on non-x86 platforms like your Orange-Pi)

The parameters are:

angel@orangepi5:~$ halcmd show param *.tmax
Parameters:
Owner Type Dir Value Name
61 s32 RW 292 AND2-Y-eoffset.tmax
61 s32 RW 292 AND2-Y-jog.tmax
61 s32 RW 292 AND2-Z-eoffset.tmax
61 s32 RW 292 AND2-Z-jog.tmax
61 s32 RW 292 AND2-inicio.tmax
61 s32 RW 292 AND2-reanuda.tmax
61 s32 RW 292 AND2-spindle-manual-CCW.tmax
61 s32 RW 292 AND2-spindle-manual-CW.tmax
58 s32 RW 292 AND3-spindle-manual-stop.tmax
58 s32 RW 875 NOR3-alarma-ejes.tmax
67 s32 RW 292 NOT-mode-is-auto.tmax
67 s32 RW 292 X-not.tmax
67 s32 RW 292 Y-not.tmax
67 s32 RW 292 Z-not.tmax
30 s32 RW 11083 base-thread.tmax
55 s32 RW 292 conv_float_s32-feed.tmax
55 s32 RW 584 conv_float_s32-rapid.tmax
76 s32 RW 583 debounce.0.tmax
73 s32 RW 7875 hal_op5_io.read.tmax
73 s32 RW 8166 hal_op5_io.write.tmax
37 s32 RW 0 hm2_7i92.0.read-request.tmax
37 s32 RW 198039 hm2_7i92.0.read.tmax
37 s32 RW 18666 hm2_7i92.0.write.tmax
64 s32 RW 292 ilowpass-encoder-X.tmax
64 s32 RW 292 ilowpass-encoder-select.tmax
70 s32 RW 292 invert.0.tmax
29 s32 RW 1167 motion-command-handler.tmax
29 s32 RW 12833 motion-controller.tmax
52 s32 RW 292 mult2-1-feed.tmax
52 s32 RW 292 mult2-1-jog.tmax
52 s32 RW 583 mult2-1-rapid.tmax
52 s32 RW 292 mult2-2-feed.tmax
52 s32 RW 292 mult2-2-jog.tmax
52 s32 RW 292 mult2-2-rapid.tmax
52 s32 RW 583 mult2-escala.tmax
52 s32 RW 292 mult2-voltaje.tmax
49 s32 RW 584 mux8-escala-mpg.tmax
40 s32 RW 9333 pid.s.do-pid-calcs.tmax
40 s32 RW 9334 pid.x.do-pid-calcs.tmax
40 s32 RW 9625 pid.y.do-pid-calcs.tmax
40 s32 RW 9333 pid.z.do-pid-calcs.tmax
31 s32 RW 235371 servo-thread.tmax