EtherCAT Comm Sync error (83.3) when exiting LinuxCNC

More
05 Feb 2022 01:09 #234007 by arvidb
I'm running linuxcnc from the git repo BTW, commit 170d13f12a (2022-01-23).

Please Log in or Create an account to join the conversation.

More
05 Feb 2022 01:21 #234009 by arvidb
This is the output from 'journalctl -f' when exiting LinuxCNC with the 'unloadrt lcec' line in the hal shutdown script:
feb 05 02:19:25 debian kernel: EtherCAT 0: Master thread exited.
feb 05 02:19:25 debian kernel: EtherCAT 0: Starting EtherCAT-IDLE thread.
feb 05 02:19:25 debian kernel: EtherCAT 0: Releasing master...
feb 05 02:19:25 debian kernel: EtherCAT 0: Released.
feb 05 02:19:25 debian kernel: EtherCAT 0: Slave states on main device: INIT.
feb 05 02:19:25 debian kernel: EtherCAT 0: Slave states on main device: PREOP.

Compared to without the unloadrt:
feb 05 02:21:11 debian kernel: EtherCAT 0: Master thread exited.
feb 05 02:21:11 debian kernel: EtherCAT 0: Starting EtherCAT-IDLE thread.
feb 05 02:21:11 debian kernel: EtherCAT WARNING 0: 1 datagram UNMATCHED!
feb 05 02:21:11 debian kernel: EtherCAT 0: Releasing master...
feb 05 02:21:11 debian kernel: EtherCAT 0: Released.
feb 05 02:21:11 debian kernel: EtherCAT ERROR 0-main-0: Failed to receive AL state datagram: Datagram initialized.
feb 05 02:21:11 debian kernel: EtherCAT 0: Slave states on main device: SAFEOP + ERROR.
feb 05 02:21:11 debian kernel: EtherCAT WARNING 0-main-0: Changed to SAFEOP + ERROR in the meantime.
feb 05 02:21:11 debian kernel: EtherCAT 0: Slave states on main device: INIT.
feb 05 02:21:11 debian kernel: EtherCAT 0: Slave states on main device: PREOP.
feb 05 02:21:12 debian kernel: EtherCAT WARNING: Datagram 000000001a73f979 (master-fsm) was SKIPPED 1 time.
The following user(s) said Thank You: Grotius

Please Log in or Create an account to join the conversation.

More
05 Feb 2022 01:33 #234010 by Grotius
Hi,

Interesting results.  Haha.

Please Log in or Create an account to join the conversation.

More
05 Feb 2022 12:14 #234018 by db1981
manual start /stop the rt part is another way of course. The master then forces an op change to init->preop to all slaves. As soon the slaves have left safeop /op state, it is safe to stop the DC Clock. But if the clock is stopped while an dc synced slave is in op mode, your descripted error behaviour is normal.


But this is interesting, looks like that lcnc is killing all functions hard at shutdown.
And doesn't wait for acks from the single modules,
The following user(s) said Thank You: arvidb

Please Log in or Create an account to join the conversation.

Time to create page: 0.072 seconds
Powered by Kunena Forum