Hello, I have faced an issue, a managed switch, which sends a General Query IGMP message to the network asking for multicast devices to identify the streams they want. Apparently LwIP sends IGMP messages around 120 seconds interval, which usually works with most of the switches, however since the messages are sent depending on LwIP timers, they get timeout as they are not synchronized to the switch requests. The switch sends a General Query message every 125 seconds as per RFC2236 section 8.2. It also expects the device to reply within 45 seconds (this is a non-changeable setting in the switch parameters). The standard section 8.3 defines a recommended default value of 10 seconds. Since LwIP generates the report message every +/-120 seconds and it doesn't respond to General Queries, I would like to do a workaround this issue by sending the messages every 40 seconds instead of 120 seconds. Could someone advise where should I look for this change?
My system use NO_SYS_NO_TIMERS = 1. I have a single timer firing up every 250ms (and I don't have any other available timers), where I call tcp_tmr() and igmp_tmr(). My IGMP_TMR_INTERVAL is set to 250. Maybe I am misinterpretting LwIP IGMP messaging idea, but seems like it works as I've described. Thank you. Regards Tomas Daujotas
_______________________________________________ lwip-users mailing list lwip-users@nongnu.org https://lists.nongnu.org/mailman/listinfo/lwip-users
