Problem trying to run with remote=1

[auto-mation-assist]

I had MachineKit running with remote=0 and contrlling my milling maching so I know that my linux computer configuration is good. But I'm working on a remote gui so cleaned out its prior install and started with a clean install using the master zip file. I always run in a rip configuration.

As far as I have been able to determine the problem is related to /libexec/rtapi_msgd and /libexec/rtapi_app_rt-preempt. Possibly the reported installed location.

It always gives these erros when starting up:

halcmd: the rtapi:0 RT demon is not running - please investigate /var/log/linuxcnc.log
halcmd: the msgd:0 logger demon is not running - please investigate /var/log/linuxcnc.log
E: 19-10-15 20:38:37 [21476]dangling 'DEALER' socket created at hal/utils/halcmd_rtapiapp.cc:284
cnc@cnc1:~/machinekit-dev$

The log does not give me anymore clues but the number of dots in the first part of file location in the below tests concern me.

Here is some information on my system and tests that I have completed.

My uses Linux 19.1 Mint
cnc@cnc1:~/Desktop$ uname -a
Linux cnc1 4.14.93-rt53 #1 SMP PREEMPT RT Tue Jan 29 01:00:10 AKST 2019 x86_64 x86_64 x86_64 GNU/Linux

I used this config for the compile:
./configure --enable-remote  --with-rt-preempt --enable-drivers --enable-build-documentation

DEBUG=5 realtime start
halcmd -f -k

The above gave:
cnc@cnc1:~/machinekit-dev$ DEBUG=5 realtime start
realtime: command not found
cnc@cnc1:~/machinekit-dev$ . ./scripts/rip-environment
cnc@cnc1:~/machinekit-dev$ DEBUG=5 realtime start
rtapi_msgd command:  /home/cnc/machinekit-dev/libexec/rtapi_msgd --instance=0 --rtmsglevel=5 --usrmsglevel=5 --debug=5 --halsize=524288
rtapi_app command:  /home/cnc/machinekit-dev/libexec/rtapi_app_rt-preempt --instance=0 --debug=5
E: 19-10-15 20:38:32 [21470]dangling 'XPUB' socket created at rtapi/rtapi_msgd.cc:1107
halcmd: cant connect to rtapi_app: -1 (uri= uuid=cf144c6b-b001-4b4f-a57c-92479003f70d): rtapi_rpc(): reply timeout

halcmd: the rtapi:0 RT demon is not running - please investigate /var/log/linuxcnc.log
halcmd: the msgd:0 logger demon is not running - please investigate /var/log/linuxcnc.log
E: 19-10-15 20:38:37 [21476]dangling 'DEALER' socket created at hal/utils/halcmd_rtapiapp.cc:284
cnc@cnc1:~/machinekit-dev$

-----
rtapi_msgd indicates its owner is cnc
If I use: gdb ../libexec/rtapi_msgd
../libexec/rtapi_msgd: No such file or directory.

If I use: gdb ./libexec/rtapi_msgd
cnc@cnc1:~/machinekit-dev$ gdb ./libexec/rtapi_msgd
GNU gdb (Ubuntu 8.1-0ubuntu3.1) 8.1.0.20180409-git
Copyright (C) 2018 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from ./libexec/rtapi_msgd...done.
(gdb)

Then followed up with: r --foreground --stderr
Starting program: /home/cnc/machinekit-dev/libexec/rtapi_msgd --foreground --stderr
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
warning: removing unused global shm segment /linuxcnc-0-00154711
[Inferior 1 (process 22034) exited with code 01]
(gdb)

------
rtapi_app_rt-preempt indicates its owner is root with cnc read-only and then none.
gdb ../libexec/rtapi_app_rt-preempt
This gave me:

gdb ../libexec/rtapi_app_rt-preempt
cnc@cnc1:~/machinekit-dev$ gdb ../libexec/rtapi_app_rt-preempt
GNU gdb (Ubuntu 8.1-0ubuntu3.1) 8.1.0.20180409-git
Copyright (C) 2018 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
../libexec/rtapi_app_rt-preempt: No such file or directory.
(gdb)

Same with one dot.
gdb ./libexec/rtapi_app_rt-preempt
Gave me:

cnc@cnc1:~/machinekit-dev$ gdb ./libexec/rtapi_app_rt-preempt
GNU gdb (Ubuntu 8.1-0ubuntu3.1) 8.1.0.20180409-git
Copyright (C) 2018 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from ./libexec/rtapi_app_rt-preempt...done.

Then followed by:
(gdb) r --foreground -d -d
Starting program: /home/cnc/machinekit-dev/libexec/rtapi_app_rt-preempt --foreground -d -d
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".

Program received signal SIGSEGV, Segmentation fault.
__GI_____strtol_l_internal (nptr=0x0, endptr=0x0, base=10,
    group=<optimized out>, loc=0x7f45ba5e4560 <_nl_global_locale>)
    at ../stdlib/strtol_l.c:292
292    ../stdlib/strtol_l.c: No such file or directory.
(gdb)

[Bas de Bruijn]

On 16 Oct 2019, at 07:48, auto-mation-assist <867...@gmail.com> wrote:

I had MachineKit running with remote=0 and contrlling my milling maching so I know that my linux computer configuration is good. But I'm working on a remote gui so cleaned out its prior install and started with a clean install using the master zip file. I always run in a rip configuration.

Have you purged all machinekit packages first and executed the “sudo make setuid” step last?

--
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 machinekit+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/machinekit/6eb7b943-2c1a-4241-ac9d-ee6e7a1e09d7%40googlegroups.com.

[auto-mation-assist]

I think that purge only works for a regular install and not a run as rip configuration. I will always do a make clean before doing a make.  I have done all the normal make requirements which includes the “sudo make setuid”.

I have tried everything I can think off right now including doing at least five builds done with various options. All come out with the same result. The builds themselves look good but I suspect that the linking to the kernel flavor is the problem. I'm generating a updated kernel now (4.14.146-RT67) to try. As a last resort I can build up a system on a unused computer from scratch to eliminate the possibility of messing up my normal dvelopment system to much.

My searched on the internet shows that there have been issues with the same errors in the past by others along with discussions on flavors but I have not run across a clear solution after many hours of searching.

[Bas de Bruijn]

On 17 Oct 2019, at 05:44, auto-mation-assist <867...@gmail.com> wrote:

I think that purge only works for a regular install and not a run as rip configuration. I will always do a make clean before doing a make.  I have done all the normal make requirements which includes the “sudo make setuid”.

I have tried everything I can think off right now including doing at least five builds done with various options. All come out with the same result. The builds themselves look good but I suspect that the linking to the kernel flavor is the problem. I'm generating a updated kernel now (4.14.146-RT67) to try. As a last resort I can build up a system on a unused computer from scratch to eliminate the possibility of messing up my normal dvelopment system to much.

I’d try installing a 4.9 RT kernel to see if it is indeed the kernel

What happens when you configure only with --with-rt-preempt

My searched on the internet shows that there have been issues with the same errors in the past by others along with discussions on flavors but I have not run across a clear solution after many hours of searching.

--
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 machinekit+...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/machinekit/0eced7bf-38da-4b59-9817-2efbad66042b%40googlegroups.com.

[auto-mation-assist]

I can try the 4.9 kernel. Right now I have 4.4.190 loaded.

What I have found is that whenever I enter anything into 'bind' in file /etc/linuxcnc/machinekit.ini I get the errors listed below. If I just leave it blank it runs machinekit with no gui just fine but does not appear to output anything on a tcp connection.

halcmd: the rtapi:0 RT demon is not running - please investigate /var/log/linuxcnc.log
halcmd: the msgd:0 logger demon is not running - please investigate /var/log/linuxcnc.log

halcmd: cant connect to rtapi_app: -1 (uri= uuid=3c6b50af-0fe2-4929-b0bf-0cd4b6b6c327): rtapi_rpc(): reply timeout

So my problem could be that the below is instructions are not clear to me an that I don't know what to type in there yet. Im trying to bind to my network card with address 10.10.20.2.

This is the part I'm having trouble with understanding of just what bind is and how to make it happy. These feel that these instructions could be revised to help by giving an example of what is expected and a links to the documents mentioned.

"# -------------- interface and protocol selection -----------------
#
# binding zeroMQ sockets and mDNS announcements are conceptually separate,
# but of course better add up.
# the default is to bind on in6addr_any (ipv4 and mapped ipv6 on Linux).
# see the zeromq docs - zmq_tcp(7) and the ZMQ_IPV6 option.
#
# The binding algorithm is as follows:
#
# Bind any addresses or interfaces in the BIND_IPV4 list (none if empty).
# if a port number was explicitly specified above, use that port
# else bind as ephemeral port, retrieve that assigned port number, and use
# that port in subsequent bind operations."

So "Bind any addresses'

BIND_IPV4=10.10.20.2  would seem like it should be ok but its not according to the errors.

[Bas de Bruijn]

I have no specific knowledge about what you should do.
The only thing is that when I used a remote GUI, I only built with the configure flag for the RT-PREEMPT kernel; - -with-rt-preempt
And change in machinekit.ini so that REMOTE=1
Never changed anything else.
I can’t help you with more info that this :(

[markus]

don't you also need to use `mkwrapper` as the UI, otherwise you're not
getting the UI services published?

[Alexander Rössler]

Yes, replace axis in the .ini file with mkwrapper in the [DISPLAY] section.

The msgd errors usually appear when a previous session wasn't cleaned up correctly. Make sure to run realtime stop before you try to realtime start again.

[auto-mation-assist]

Thanks for all your help. I will make the suggested changes.  I had tried the other Kernels but they caused more problems so reverted back to my original one.  I found some trouble shooting info that Alexander had posted on the internet and that helped. When I did not put anything in bind in the machinekit .ini file it printed out some helpful info in the log since machinekit loaded and ran. But still leaves me with a question about assigning a specific IP address and linking that to a network interface. I have two network interfaces one with 10.10.10.1 for the milling machines Mesa 7I80 interface, and 10.10.20.2 for the gui interface. Usually I use a USB network interface when connections to the internet are required.

 I'm not used to working with DNS and my Labview based gui programming right now points to port (5005) on 10.10.20.2. This is what I was using for testing with LinuxCnc for linking into NML directly. I have not added the zeromq or protobuf to my gui yet for the desired machinekit interface.

From the log I copied this which does give some TCP related info now.

Oct 17 22:57:52 cnc1 msgd:0: hal_lib:6490:user haltalk
Oct 17 22:57:52 cnc1 msgd:0: hal_lib:6490:user
Oct 17 22:57:52 cnc1 msgd:0: ulapi:6493:user _ulapi_init(): ulapi rt-preempt unknown loaded
Oct 17 22:57:52 cnc1 msgd:0: ulapi:6493:user halg_xinitfv:271 HAL: singleton component 'hal_lib6493' id=312 initialized
Oct 17 22:57:52 cnc1 msgd:0: hal_lib:6493:user haltalk: startup ØMQ=4.2.5 czmq=4.1.0 protobuf=3.0.0 uuid=3c6b50af-0fe2-4929-b0bf-0cd4b6b6c327
Oct 17 22:57:52 cnc1 msgd:0: hal_lib:6493:user adopted 0 groups(s)
Oct 17 22:57:52 cnc1 msgd:0: hal_lib:6493:user adopted 0 comps(s)
Oct 17 22:57:52 cnc1 msgd:0: hal_lib:6493:user haltalk: talking HALGroup on 'dsn=tcp://%s:6200'
Oct 17 22:57:52 cnc1 haltalk: register_stuff: actual hostname as announced by avahi='cnc1.local'
Oct 17 22:57:52 cnc1 haltalk: register_stuff: actual hostname as announced by avahi='cnc1.local'

[auto-mation-assist]

My cut and paste was a little to short. This shows the other ports

Oct 17 22:57:52 cnc1 msgd:0: hal_lib:6490:user haltalk

Oct 17 22:57:52 cnc1 msgd:0: hal_lib:6490:user
Oct 17 22:57:52 cnc1 msgd:0: ulapi:6493:user _ulapi_init(): ulapi rt-preempt unknown loaded
Oct 17 22:57:52 cnc1 msgd:0: ulapi:6493:user halg_xinitfv:271 HAL: singleton component 'hal_lib6493' id=312 initialized
Oct 17 22:57:52 cnc1 msgd:0: hal_lib:6493:user haltalk: startup ØMQ=4.2.5 czmq=4.1.0 protobuf=3.0.0 uuid=3c6b50af-0fe2-4929-b0bf-0cd4b6b6c327
Oct 17 22:57:52 cnc1 msgd:0: hal_lib:6493:user adopted 0 groups(s)
Oct 17 22:57:52 cnc1 msgd:0: hal_lib:6493:user adopted 0 comps(s)
Oct 17 22:57:52 cnc1 msgd:0: hal_lib:6493:user haltalk: talking HALGroup on 'dsn=tcp://%s:6200'
Oct 17 22:57:52 cnc1 haltalk: register_stuff: actual hostname as announced by avahi='cnc1.local'
Oct 17 22:57:52 cnc1 haltalk: register_stuff: actual hostname as announced by avahi='cnc1.local'

Oct 17 22:57:52 cnc1 rtapi:0: 4:rtapi_app:6472:user trivkins: loaded from trivkins.so
Oct 17 22:57:52 cnc1 msgd:0: hal_lib:6493:user haltalk: talking HALRcomp on 'dsn=tcp://%s:6201'
Oct 17 22:57:52 cnc1 msgd:0: hal_lib:6493:user haltalk: talking HALComand on 'dsn=tcp://%s:6202'
Oct 17 22:57:52 cnc1 msgd:0: hal_lib:6472:rt halg_xinitfv:90 HAL: initializing component 'trivkins' type=1 arg1=0 arg2=0/0x0
Oct 17 22:57:52 cnc1 msgd:0: hal_lib:6472:rt halg_export_vtable:28 HAL: exporting vtable 'trivkins' version=1000 owner=316 at 0x7fc6ea5aa080
Oct 17 22:57:52 cnc1 msgd:0: hal_lib:6472:rt halg_export_vtable:54 HAL: created vtable 'trivkins' vtable=0x7fc6ea5aa080 version=1000
Oct 17 22:57:52 cnc1 rtapi:0: 4:rtapi_app:6472:user tp: loaded from tp.so
Oct 17 22:57:52 cnc1 rtapi:0: 4:rtapi_app:6472:user hostmot2: loaded from hostmot2.so

[markus]

depending on what you want to do in your UI, you'll also need the
services 'status', 'command' and 'error' which are provided by
mkwrapper.

For the DNS part, I had some problems with it on VMs and wrote a script:
https://github.com/mlampert/Machinekit-Workbench/blob/master/machinekit-local/rest-services
which opens up a HTTP port you can use to get the ports of all
services. I add the script to my machinekit configuration so it gets
automatically started.

Note that this publishes the service ports to anybody who has access to
the machine, regardless of where they come from. Use at your own risk!

[auto-mation-assist]

I was able to see that I'm getting responses back from my Linux development machine from Wireshark on my Windows Machine as shown below using the text format instead of hex. So it looks like I should be able to continue to the next steps to get the gui talking to Machinekit which will take some time. I will need to address the security problem with public network access sometime in the future but it is a big problem for myself since the running system is not connected to outside access right now, but certainly something to keep in mind. equipment at .

This is the text version of the response my Windows computer received from Machinekit using Wireshark on its 10.10.20.1 interface.

10.10.20.2    224.0.0.251    MDNS    829    Standard query response 0x0000
PTR HAL Group service on cnc1.local pid 2217._machinekit._tcp.local
PTR HAL Rcommand service on cnc1.local pid 2217._machinekit._tcp.local SRV,
cache flush 0 0 6202 cnc1.local A, cache flush 10.10.20.2 TXT,
cache flush PTR HAL Rcommand service on cnc1.local pid 2217._machinekit._tcp.local
PTR HAL Rcomp service on cnc1.local pid 2217._machinekit._tcp.local SRV,
cache flush 0 0 6201 cnc1.local TXT,
cache flush PTR HAL Rcomp service on cnc1.local pid 2217._machinekit._tcp.local
PTR HAL Group service on cnc1.local pid 2217._machinekit._tcp.local SRV,
cache flush 0 0 6200 cnc1.local TXT, cache flush

[auto-mation-assist]

Thank you. I will have a look at link.

[auto-mation-assist]

I have been trying to better understand what services are running for Machinekit on my linux mint development computer and my windows computer. I generated the list I have below for this. I have been working on the zeroMQ interface but not sure what the sequence of commands need to be to establish a connection or if the ports used for the various services are static.

It looks to me that the links to all the services are there with a minor difference on the mkwrapper which at the time of this capture was PID 13824. This has a service count of four on linux but six on windows.

Services list from my Machinekit LinuxMint computer
cnc@cnc1:~/Desktop$ lsof -i
COMMAND     PID USER   FD   TYPE  DEVICE SIZE/OFF NODE NAME
gvfsd-smb  2681  cnc   12u  IPv4  198599      0t0  TCP localhost:34550->cnc1:netbios-ssn (ESTABLISHED)
linuxcncs 13762  cnc    3u  IPv4 1676096      0t0  TCP *:5005 (LISTEN)
rtapi_msg 13784  cnc   13u  IPv6 1676082      0t0  TCP *:49152 (LISTEN)
haltalk   13813  cnc   13u  IPv6 1677304      0t0  TCP *:6500 (LISTEN)
haltalk   13813  cnc   16u  IPv6 1677306      0t0  TCP *:6501 (LISTEN)
haltalk   13813  cnc   19u  IPv6 1677308      0t0  TCP *:6502 (LISTEN)
mkwrapper 13824  cnc    6u  IPv4 1676178      0t0  TCP *:40081 (LISTEN)
mkwrapper 13824  cnc   15u  IPv4 1676185      0t0  TCP *:36405 (LISTEN)
mkwrapper 13824  cnc   17u  IPv4 1676186      0t0  TCP *:38569 (LISTEN)
mkwrapper 13824  cnc   19u  IPv4 1676187      0t0  TCP *:39591 (LISTEN)
mkwrapper 13829  cnc    6u  IPv4 1676178      0t0  TCP *:40081 (LISTEN)
mkwrapper 13829  cnc   15u  IPv4 1676185      0t0  TCP *:36405 (LISTEN)
mkwrapper 13829  cnc   17u  IPv4 1676186      0t0  TCP *:38569 (LISTEN)
mkwrapper 13829  cnc   19u  IPv4 1676187      0t0  TCP *:39591 (LISTEN)
mkwrapper 13829  cnc   25u  IPv4 1677992      0t0  TCP *:49153 (LISTEN)
mkwrapper 13829  cnc   34u  IPv4 1677995      0t0  TCP *:49154 (LISTEN)
cnc@cnc1:~/Desktop$

---------
On windows7 computer and using the (Bonjour Browser v1.13) I get services:
Type: _machinekit._tcp. Name: Log service on cnc1.local pid 13784
Type: _machinekit._tcp. Name: HAL Group service on cnc1.local pid 13813
Type: _machinekit._tcp. Name: Previewstatus service on cnc1.local pid 13824
Type: _machinekit._tcp. Name: Preview service on cnc1.local pid 13824
Type: _machinekit._tcp. Name: Command service on cnc1.local pid 13824
Type: _machinekit._tcp. Name: Error service on cnc1.local pid 13824
Type: _machinekit._tcp. Name: Status service on cnc1.local pid 13824
Type: _machinekit._tcp. Name: File service on cnc1.local pid 13824
Type: _machinekit._tcp. Name: HAL Rcommand service on cnc1.local pid 13813
Type: _machinekit._tcp. Name: HAL Rcomp service on cnc1.local pid 13813

---------
Breaking these down based on their PID number

rtapi_msg 13784  cnc   13u  IPv6 1676082      0t0  TCP *:49152 (LISTEN)
Type: _machinekit._tcp. Name: Log service on cnc1.local pid 13784

Linux Machine - Haltalk (My port numbers)
haltalk   13813  cnc   13u  IPv6 1677304      0t0  TCP *:6500 (LISTEN)
haltalk   13813  cnc   16u  IPv6 1677306      0t0  TCP *:6501 (LISTEN)
haltalk   13813  cnc   19u  IPv6 1677308      0t0  TCP *:6502 (LISTEN)
Windows Machine (may not be in correct order)
Type: _machinekit._tcp. Name: HAL Group service on cnc1.local pid 13813
Type: _machinekit._tcp. Name: HAL Rcommand service on cnc1.local pid 13813
Type: _machinekit._tcp. Name: HAL Rcomp service on cnc1.local pid 13813

Linux comuputer - MKwrapper (not use what is on what port))
mkwrapper 13824  cnc    6u  IPv4 1676178      0t0  TCP *:40081 (LISTEN)
mkwrapper 13824  cnc   15u  IPv4 1676185      0t0  TCP *:36405 (LISTEN)
mkwrapper 13824  cnc   17u  IPv4 1676186      0t0  TCP *:38569 (LISTEN)
mkwrapper 13824  cnc   19u  IPv4 1676187      0t0  TCP *:39591 (LISTEN)
Windows computer (shows two more services)
Type: _machinekit._tcp. Name: Previewstatus service on cnc1.local pid 13824
Type: _machinekit._tcp. Name: Preview service on cnc1.local pid 13824
Type: _machinekit._tcp. Name: Command service on cnc1.local pid 13824
Type: _machinekit._tcp. Name: Error service on cnc1.local pid 13824
Type: _machinekit._tcp. Name: Status service on cnc1.local pid 13824
Type: _machinekit._tcp. Name: File service on cnc1.local pid 13824

Linux computer - These appear to be local services
mkwrapper 13829  cnc    6u  IPv4 1676178      0t0  TCP *:40081 (LISTEN)
mkwrapper 13829  cnc   15u  IPv4 1676185      0t0  TCP *:36405 (LISTEN)
mkwrapper 13829  cnc   17u  IPv4 1676186      0t0  TCP *:38569 (LISTEN)
mkwrapper 13829  cnc   19u  IPv4 1676187      0t0  TCP *:39591 (LISTEN)
mkwrapper 13829  cnc   25u  IPv4 1677992      0t0  TCP *:49153 (LISTEN)
mkwrapper 13829  cnc   34u  IPv4 1677995      0t0  TCP *:49154 (LISTEN)

------

Port 5005 comment.
Linux computer
linuxcncs 13762  cnc    3u  IPv4 1676096      0t0  TCP *:5005 (LISTEN)

Port 5005 can accept direct NML commands. I have done this in the past.
Linuxcnc and Machinekit access formats and serial number tracking are
different. Commands sent to port 5005 can confict with those generated
internally by middleware presently being used internally such as the
Python module and Gstat. Remote use of this port (tcp server) should
likely be disabled.