So a solution would be getting some router with custom router OS were the user can set the nat table limit and a router that has a good cpu and memory? For TP link there is no way of finding the max nat table limit, asked in the forum and they did not understand the question and then I got no further response.
As the router I use is TP-link archer ax20 maybe mid tier, just a few years old, then my thoughts are that many people will reach nat table limit on their routers with moderate amount of nodes.
Now at early Beta I understand if devs wants as many connections as possible to se what the network can handle and such. Is it possible peer connections per node will go down, can it get down to like 40 peer connections per node or similar?
There is an opensource project for custom router software. Forget the name at the moment but it can be installed on TP routers from memory. Ask in the TP-link forum and you should get an answer.
It might allow you to change the value, ask in the forum for the opensource project when you get the name and site for it. I should know the name but can’t at the moment
Remember changing the software on the router has its risks of bricking the device and/or the ISP needed special info for its PPPoE
PS C:\Users\gggg> safenode-manager status
╔═══════════════════════╗
║ Safenode Services ║
╚═══════════════════════╝
Refreshing the node registry...
Service Name Peer ID Status Connected Peers
safenode1 - ADDED -
safenode2 - ADDED -
safenode3 - ADDED -
safenode4 - ADDED -
safenode5 - ADDED -
PS C:\Users\gggg> safenode-manager start
╔═════════════════════════════╗
║ Start Safenode Services ║
╚═════════════════════════════╝
Refreshing the node registry...
Attempting to start safenode1...
Attempting to start safenode2...
Attempting to start safenode3...
Attempting to start safenode4...
Attempting to start safenode5...
Failed to start 5 service(s):
✕ safenode1: The 'safenode1' service has failed to start
✕ safenode2: The 'safenode2' service has failed to start
✕ safenode3: The 'safenode3' service has failed to start
✕ safenode4: The 'safenode4' service has failed to start
✕ safenode5: The 'safenode5' service has failed to start
Error:
0: Failed to start one or more services
Location:
sn_node_manager\src\cmd\node.rs:705
Backtrace omitted. Run with RUST_BACKTRACE=1 environment variable to display it.
Run with RUST_BACKTRACE=full to include source snippets.
There seems to be no support for the ax20 since it has a Broadcom chip and also TP-link locked it from custom firmware so that users could not enable other frequencies due to some regulation.
Was it openwrt you were thinking about?
Also I am now down to around 100 nodes, maybe the NAT table limit soft cap is as low as 25000.
Thats a shame. Guess you’ll have to keep it below the 130 (was it?) figure.
My suggestion is to give it a bit more headroom and limit yourself to 120 nodes or bit less. In any case that will be well above average number of nodes. I am at 40 with 2 ISP connections (20 nodes each) due to router limitations and upload b/w even with better routers. Here 40 Mbps up is the max anyone can get on non-business (very expensive) plans
I’am down to about 100 nodes, think the NAT table limit might be as low as 25 000.
it still lags a bit when watching a twitch stream on the TV.
Right now I don’t feel like buying a new router for like 200$, already spent about $300 on the setup this month. I might be done for today, need to think about what to do. Hope that it is possible for the devs to lower peer connections per node quite a lot in the future.
@neo do you think a computer with 2 lan cards can be used instead of a router. The first computer shares the Internet through the second LAN card to a router for the home network?
Looks like services are not deleted as they should be. Try to reset with sudo node-manager. After that check if services still exists and remove them manually, worked for me with windows server.
Still no good. I’ll try a different machine in a while. Interestingly I’m also getting the error on a cloud instance, so I’ll try a fresh one of those too.
made sure that /var/safenode-manager was empty /var/log/safenode was empty
that there were no ‘safnode’ service definitions under /etc/systemd/system
and done a reboot.
I also rolled back to a snapshot of the VM from before it ever saw safe.
The error in /var/log/syslog is:-
Jun 8 11:55:24 sn-test-01 systemd[1]: /etc/systemd/system/safenode1.service:4: Failed to resolve unit specifiers in /ip4/159.65.50.42/udp/58720/quic-v1/p2p/12D3KooWGxT1bh9SjtdHeq7uoS8vtfzG4A8dhGptHhbjmAQ6cXhX,/ip4/178.128.34.57/udp/42231/quic-v1/p2p/12D3KooWSn9FhKHBbTRuS9FfydrfQ17tGGDByW6SJiSMeW3Gogdo,/ip4/167.71.134.115/udp/56129/quic-v1/p2p/12D3KooWRj84z1gv46NnQevHV7QCMJriV8QyfdtPjrRUquQxa7Pt,/ip4/167.71.130.219/udp/40949/quic-v1/p2p/12D3KooWBA2qjGDmnZd2GyNC5qVw82bSvWDoQj1FBjuj4h11zXaL,/ip4/138.68.156.118/udp/50365/quic-v1/p2p/12D3KooWNCGAmBHphuGEbTYFyPpTLoTct4PBmZGqqeDTeJHUUzay,/ip4/159.65.48.153/udp/43896/quic-v1/p2p/12D3KooWJEMGnifi7ZkGBmgGHZP5TLZrHTbRrrFNeiE68wZEmipX,/ip4/167.99.202.50/udp/36305/quic-v1/p2p/12D3KooWSrS7er3SDMdX5TMuirQWbek7kmbHRtmv1zzEpgB1Wbsw,/ip4/159.65.24.14/udp/33115/quic-v1/p2p/12D3KooWGCw2hebm2mHYRKLrFx7UE6n7wLzr8vYpjGuu526WTvYT,/ip4/142.93.42.123/udp/52692/quic-v1/p2p/12D3KooWBeZVDUQ6zCH6v4TyDxmYYY8HEg47LWnTNVSL2UZbXGkh,/ip4/159.65.31.145/udp/46772/quic-v1/p2p/12D3KooWGf2mxTZ6JXdFYdVUvAU2KXaCNLe84iTVvwvEhfxVNAwq,/ip4/159.65.24.14/udp/33115/quic-v1/p2p/12D3KooWGCw2hebm2mHYRKLrFx7UE6n7wLzr8vYpjGuu526WTvYT,/ip4/167.71.134.112/udp/49012/quic-v1/p2p/12D3KooWGrFKcq6zsTxQqyiATLDp6EUSDT7ui7VjTMJaU5GHayk7,/ip4/167.99.196.183/udp/39205/quic-v1/p2p/12D3KooWDCRnNrNwt5H855HtWgGUbeBWkoy9YAiGMQ5sfKHY362s,/ip4/167.71.134.146/udp/56898/quic-v1/p2p/12D3KooWQWTEJ4ngwDAE6oLgYdiZk3ejRYxp4do8Akx9GAWUEe9b,/ip4/142.93.42.248/udp/43084/quic-v1/p2p/12D3KooWHHvVokx9CXJpoWXxk8XTfrkhZFNC2ofaH3DtaTwVGsMV,/ip4/167.71.130.219/udp/40949/quic-v1/p2p/12D3KooWBA2qjGDmnZd2GyNC5qVw82bSvWDoQj1FBjuj4h11zXaL,/ip4/209.97.138.10/udp/42240/quic-v1/p2p/12D3KooWFiDxHr3RpovdqShNuzbM5NAQSF7ctT6f5m4mxuRS4asB,/ip4/68.183.36.156/udp/35703/quic-v1/p2p/12D3KooWL9LBbD13U5tRmCwuRbkQkXFM4gLEmf7DdN1r3Ek7yAkF,/ip4/178.128.34.57/udp/42231/quic-v1/p2p/12D3KooWSn9FhKHBbTRuS9FfydrfQ17tGGDByW6SJiSMeW3Gogdo,/ip4/134.122.96.130/udp/32921/quic-v1/p2p/12D3KooWGXW7xSk5UMKt6h8CVeq2fGSQwhnRsvedK9MBzTTJCxNH,/ip4/159.65.31.145/udp/46772/quic-v1/p2p/12D3KooWGf2mxTZ6JXdFYdVUvAU2KXaCNLe84iTVvwvEhfxVNAwq,/ip4/159.65.31.145/udp/46772/qu
Jun 8 11:55:24 sn-test-01 systemd[1]: safenode1.service: Unit configuration has fatal error, unit will not be started.
I had a look at the file /etc/systemd/system/safenode1.service and thought I’d spotted the problem: no “” enclosing the definition of the peers after --peer.
I edited it but that hasn’t helped.
So the search continues.
Without being able to run on Pi because of the key issue I’m a bit stuck for running nodes that earn and send to Maidsafe and get shown in Discord.
@JPL and @storage_guy which version of ubuntu? I had issue on 22.04, after reinstallation i put 20.04 and from that point i did reset with node manager few times without issues.
