4-5 minutes into running an ipfs daemon, my wifi suddenly died. After rebooting my wifi, I spawned the daemon again to make sure this wasn’t just a one-off thing – but it killed my wifi yet again.
Is it just me, or is this a known issue? Seem like it’s been discovered before, but hasn’t gained much traction for a solution.
I was having this issue before, there are a few possiblities :
- Your router have not enough NAT address or memory to deal with that much connections at once.
Solutions : Change in the config
dht-client (your node will not be part of ipfs’s big distributed directory (but will still ask other nodes)), that greatly reduce the numbers of nodes dialling yours. You can also lower the high and low water (to like 50 and 25), that will reduce even more the number of nodes you are contacting.
- (what happened to me) Your IAP have a “ddos protect” option, it is likely dumb, sees 1000 connections, lots of UDP too, and just put the plug out (virtualy).
Solution: Will depends a lot of who gives you your internet, you might have to disable it in your router or in your account page, I have no idea for yours.
- Your IAP blocks extensive P2P, if you are in part of the world where this is banned you can sue them I guess a threatening call to their support might be enough, just be clear that you know your right and you are serious (even if you are not).
- Your wifi sucks and can’t handle this many packets, that would be weird as theory would say that it should just slow down, or eat most of bandwith, but shutdown is weird. To rule out this possibilitie, you should try the same test wired in your router, if that works fine you know it’s a problem with the wifi, if not it’s in the router/IAP part.