Zum Inhalt springen

EtherCAT troubleshooting

Dieser Inhalt ist noch nicht in deiner Sprache verfügbar.

Open the device in the portal and the log of the EtherCAT connector process. Each line names the connection (name). The normal start shows EtherCAT scan finished and EtherCAT master state with state=OP.

For more detail, open the device in the portal and set the log level of the EtherCAT connector process to Debug or Trace. The connector applies the new level while it runs: the runtime does not restart and the bus stays in OP. Set it back to Info when you are done, because Trace writes many lines.

SymptomCause and action
EtherCAT connection failed, will reconnect with ethercat: timeout waiting for reply after 4 attempts of 50msNo reply from the bus. UDP: check the Host, the route and the firewall (see Allow the UDP traffic); run sudo tcpdump -ni <interface> udp port 34980 on the device: packets out and no answers means the network path blocks them. UDP, when the network path is open: another master (TwinCAT) used the EK1000 before, and the EK1000 still accepts only that master. Stop the other master and switch the power of the EK1000 off and on (see Make sure that logiccloud is the only master). Layer 2: check the cable and the interface. Check that the EK1000 is powered.
EtherCAT UDP: the local port is used by another program; sending from a port the system choosesAnother program on the device uses UDP port 34980. The connection works, but after a reconnect the EK1000 may ignore the connector, because the port changed. Stop the other program, then restart the runtime.
Topology mismatch: no slave at this positionThe bus has fewer devices than the configuration. Check the wiring, or update the devices (ENI import, bus scan).
Topology mismatch: the slave at this position is not the configured oneAnother device is at this position. The line shows the expected and the found vendor ID, product code and revision. The order of the terminals in the ENI file must match the order on the bus. Correct the wiring, or update the devices (ENI import, bus scan).
EtherCAT master state: staying in PREOP because the bus does not match the configurationFollows a topology mismatch. The connector exchanges no process data and scans again after 10 s.
Working counter mismatchNot all devices processed the last frame. After 5 such cycles in a row the connector reconnects. Check cables and connectors, and that no other master sends frames to the segment.
slave N refused state ... AL status code 0x....The device refused a state change. The line contains the meaning of the code, for example Sync manager watchdog (0x001B: the cycle time is longer than the watchdog of the device) or Invalid sync manager configuration (0x0017: the PDO configuration does not match the device; import the ENI file again). The slave number in this line counts from 0.
Cannot open the layer-2 transport: the connector needs the capability CAP_NET_RAWThe connector may not open raw sockets. Install the container with the host network, which adds NET_RAW: see Layer 2 on a device installed with the Linux script.
Cannot open the layer-2 transport: the network interface does not existWrong interface name, or the interface is not visible to the connector. Check it with sudo docker exec logiccloud ls /sys/class/net. If the interface is missing there, the container has its own network: install it with the host network, see Layer 2 on a device installed with the Linux script.
Another process uses this EtherCAT segment (a second connector or a bus scan of the browsing service), will retryTwo connections use the same EK1000 or interface, or a bus scan runs. Use one connection per segment, and wait for the scan to finish.
ethercat: process image too large for one datagramThe mapped devices have more than 2035 bytes of process data together. Remove devices or PDOs.
Skipping mapping: ...A mapping does not fit its PDO entry (missing entry, wrong width, or an output entry mapped to a read-only variable). Correct the mapping and deploy again.
Scan bus fails with No device is bound to this runtimeInstall the runtime on a device first.
Scan bus shows Not matchedImport the ESI file of the vendor and scan again.
After a cable pull the values come back only after up to 30 sExpected with UDP: the connector cannot see the link state and reconnects with a growing wait (1 s doubling to 30 s). With layer 2 it reconnects as soon as the link is back.
Outputs are 0 after the runtime stoppedExpected: on stop the connector resets all outputs and switches the devices to SAFEOP.

For the SubDevice role, see SubDevice troubleshooting.