Hi,
I've got a problem with a timeout on the communication between Labview and Oriental motor drive (AZD-CEP). Everything works well until I disconnect the com wire to simulate a communication issue (switch firmware update or maintenance unpowered drive panel...). In that case, if the disconnection is short, less time than a TCP drop connection (maybe windows timeout or hardware timeout...), I can send a forward close and unregister the session when the drive came back on the network. In that case, the drive accepts the command and I can restart the communication as well. But if the time is "long". The TCP connection is closed by a windows "timeout" or keep alive TCP or whatever it is. In that case, the close command and unregister session cannot be correctly done and the EEIP library return an error state on these commands (NULL REFERENCE...). When I'm trying to re-establish the connection, the drive send a [RST/ACK] to the forwardopen ([SYN]) command to indicate that the last TCP connection is close. But it is already closed by Windows on library side. So, the library determine it is a refused connection and do not start the threading. The drive does not receive anything so, falls in timeout and wait for someone to talk. Because the library never starts its threading, I can't close the connection and unregister. So, to remove this problem, I've tried to remove the power of the drive to close the old connection and try to re-establish the connection. In that case, the library returns a warning (error) that only one connection is supposed to be achieved with one IP address and the threading do not start. So, the drive accepts this new connection, but the library refuses it. Again, I can't close the connection because it doesn't exist on the library side.
Does someone know how I can reopen communication with the drive?
Is there a way to accept the [RST, ACK] as an accepted connection?
Can someone help?
Maybe I misunderstand some part...correct me if I'm wrong!
Thank you
Have a nice day
Jean Philippe Goulet
Ing?nieur en conception production automatis?e
E:jpgoulet@garaga.comjpgoulet@garaga.com | T: +1 418-227-3068 x2269
Hi,
I've got a problem with a timeout on the communication between Labview and Oriental motor drive (AZD-CEP). Everything works well until I disconnect the com wire to simulate a communication issue (switch firmware update or maintenance unpowered drive panel...). In that case, if the disconnection is short, less time than a TCP drop connection (maybe windows timeout or hardware timeout...), I can send a forward close and unregister the session when the drive came back on the network. In that case, the drive accepts the command and I can restart the communication as well. But if the time is "long". The TCP connection is closed by a windows "timeout" or keep alive TCP or whatever it is. In that case, the close command and unregister session cannot be correctly done and the EEIP library return an error state on these commands (NULL REFERENCE...). When I'm trying to re-establish the connection, the drive send a [RST/ACK] to the forwardopen ([SYN]) command to indicate that the last TCP connection is close. But it is already closed by Windows on library side. So, the library determine it is a refused connection and do not start the threading. The drive does not receive anything so, falls in timeout and wait for someone to talk. Because the library never starts its threading, I can't close the connection and unregister. So, to remove this problem, I've tried to remove the power of the drive to close the old connection and try to re-establish the connection. In that case, the library returns a warning (error) that only one connection is supposed to be achieved with one IP address and the threading do not start. So, the drive accepts this new connection, but the library refuses it. Again, I can't close the connection because it doesn't exist on the library side.
Does someone know how I can reopen communication with the drive?
Is there a way to accept the [RST, ACK] as an accepted connection?
Can someone help?
Maybe I misunderstand some part...correct me if I'm wrong!
Thank you
Have a nice day
Jean Philippe Goulet
Ing?nieur en conception production automatis?e
E:jpgoulet@garaga.comjpgoulet@garaga.com | T: +1 418-227-3068 x2269
GARAGA * * St-Georges * G6A 1K5 * Qc *
Hi,
there is an example were I connected to an ethernet ip device which also handles a disconnection. Did you observed that example?