You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Feb 18, 2021. It is now read-only.
I've been experimenting with Telehash lately. I've been looking how I can use Telehash in a constrained 6lowpan network. As I could not find my way in the C implementation I used the current javascript implementation to get to know the stack and APIs.
One thing I noticed that when I'm using the UDP transport is that every 30 seconds it would send a burst of traffic on the network. After some digging I found that this was because of keep alive messages that are send to keep NAT mappings active. However, as I use ipv6 I would not need NAT and the keep-alive seem to be to frequent.
Now I found a way to change the timeout. But I was wondering the keep-alive handshake period should be present in the specs or that it should be left upon the developer to choose the right keep-alive period for his application? Or was the UDP transport never intended to be used over a constrained network and would there be a need for a new transport for 6lowpan networks?
Thanks & happy hollidays,
Eelco
The text was updated successfully, but these errors were encountered:
eelcocramer
changed the title
Keep-alive's will flood a contained network
Keep-alive's will flood a constrained network
Dec 23, 2016
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Hi,
I've been experimenting with Telehash lately. I've been looking how I can use Telehash in a constrained 6lowpan network. As I could not find my way in the C implementation I used the current javascript implementation to get to know the stack and APIs.
One thing I noticed that when I'm using the UDP transport is that every 30 seconds it would send a burst of traffic on the network. After some digging I found that this was because of keep alive messages that are send to keep NAT mappings active. However, as I use ipv6 I would not need NAT and the keep-alive seem to be to frequent.
Now I found a way to change the timeout. But I was wondering the keep-alive handshake period should be present in the specs or that it should be left upon the developer to choose the right keep-alive period for his application? Or was the UDP transport never intended to be used over a constrained network and would there be a need for a new transport for 6lowpan networks?
Thanks & happy hollidays,
Eelco
The text was updated successfully, but these errors were encountered: