Quite some time ago a core bunch of active chat users were getting fed up with the constant problems with the unreliable chat network routing through unstable links and causing lots of issues such as one-sided/broken chats and such, so we have since migrated to a new setup that uses several main cloud-based hubs to link leaf nodes in to. This has ensured maximum availability and has been working well through 2021.
There are two hubs in the USA and one hub in the EU.
The setup requires that the leaf nodes only have one link to each of the hub nodes for their region:
United States of America:
- #USEST:K8OPG-15 (east)
- #USWST:WA3WLH-15 (west)
Europe/Other:
- RRRCHT:PE1RRR-4
Leaf (your) Node Configuration:
In the configuration pane for your chat-node, the order of the link list determines the order of connection at node start up. This assumes that your node is the first to initiate a connection- this is not always the case. The chat links may need to be restarted/cycled a couple of times on your leaf node until it links to the correct hub while there may be incoming attempts from the hubs.
Once the link is established it will be stable. In the unlikely event that a hub dies, the secondary takes over. We usually perform link restarts in the event that something odd happens. Nothing is done behind closed doors so you can see it all in action should you want to via the chat in “General”.
Each leaf node needs to have a node link to one of the hubs directly, this means AXIP or AXUDP. Reach out to the sysops (listed below) to arrange a reciprocal link so that there are no hops in between your node and the hubs.
For example, RRRCHT:PE1RRR-4 hub is hosted on PE1RRR-5 cloud-based hub node, the chat links to this node are closed to all connections outside of the EU except for the USA hubs.
It is important to reiterate that leaf nodes must not introduce other nodes into the network. Leaf nodes that introduce other nodes after arranging links with the hubs will have to be disconnected to maintain the stabilised topology.
Additionally, the network is so large (hundreds of nodes) that linking a chat node over low speed RF to this network is impractical, for instance at 1200 baud just the node sync will timeout and perpetually restart.
If you want to help providing a cloud based hub for your region then get in touch as you can see we are a little short in the following regions: EU, OC, SOAM.
Sysops (replace [at] with @)
USA:
- KB8UVN@KB8UVN.#CMH.OH.USA.NOAM / kb8uvn [at] ohiopacket.org
- WA3WLH@WA3WLH.#EPA.PA.USA.NOAM / wa3wlh [at] comcast.net
EU/Other:
- PE1RRR@PE1RRR.#NBW.NLD.EURO / pe1rrr [at] amsat.org / http://Eindhoven.space/linking (password: matrix)
If you find the articles, or content in general posted here useful, enjoyable or maybe even something else, please consider making a small contribution towards my hosting fund so that I may continue to provide my services for free to as many people as possible.
Eindhoven.space 