Thanks, Yes, I’ve seen that when the client registration updates, then the timeouts disappear. We need however to move to internal server, so no worries. Cheers, Riccardo From: leshan-dev-bounces@xxxxxxxxxxx [mailto:leshan-dev-bounces@xxxxxxxxxxx] On Behalf Of Hudalla Kai (INST/ESY) Sent: 06 November 2015 11:11 To: leshan developer discussions <leshan-dev@xxxxxxxxxxx> Subject: Re: [leshan-dev] Request Timeout Hi Riccardo, I would rather suggest to send an „Update Registration“ request every minute from your LWM2M client to the LWM2M Server in order to prevent the NAT table entry from being removed from the router. Something like a “heartbeat” if you will …. Regards, Kai Dear Paul, Thank you very much. I’m on a University network and it might actually be the case. I’ll try to move to a local server and see if that solves the issue. Thanks again, Kind Regards, Riccardo. Is there NAT involved somewhere? I see similar behaviour when the UDP hole that’s been punched through the NAT is “timed out” and requests from the sandbox server no longer reach the device. Typically NAT will be configured with a 3 minute timeout, and the re-registration 5 minutes. Just a thought. Paul. I've ported a LWM2M Client (Wakaama) on an "mbed" platform and I'm using Ethernet to connect to leshan LWM2M Server sandbox (http://leshan.eclipse.org/). I'm able to register (see the Device, Server and Test Objects) and able to perform read, write and execute requests. So, I can send COAP messages and see ACKs, and on the mbed I can see "received xx bytes" debug output. However, at times (it appears to be a random process), I see that COAP messages are sent but I do not receive any reply and I see the usual "0 bytes received" on the "mbed" UDP Socket, corresponding to no activity. Then, the web interface reports: "Unable to read resource /1/0/0 for UniS_Testbed_Full_Client_Trial : 500 Request timeout" After a while (or after I do a reboot and re-registration) I'm then able to send/receive again for some time until it hangs again. What could likely be the cause? _______________________________________________ leshan-dev mailing list leshan-dev@xxxxxxxxxxx To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/leshan-dev
|