IPFS serve myself only mode

how to tell my IPFS , cut contact with the world and just serve the local machine using a local dht and whatever is available on the local machine .
would it do just that if I unplug the internet cable ? what if I want to do that while connected to the internet without the need to edit the firewall ?

the main reason my IPFS is not always running in the background is that as I run it , it starts to use my system to serve the IPFS network without giving me a chance to just serve myself .
it makes my system heavy to the point I have to stop it to be able to use my system , I want to be able to run the IPFS in the background to serve files only to myself , without serving the IPFS network , I think it’s fair to have Filecoin miners do that job (serving the network) instead of forcing all the IPFS users to serve the IPFS network .

Try taking a look at the Reprovider settings go-ipfs/config.md at master · ipfs/go-ipfs · GitHub

I believe setting Reprovider.Interval = 0 would do what you’re looking for.

I did just that , yet I have 152 peers and my network activity is over the roof . and my CPU fan is rocking and rolling , I want 0 peers and 0 upload and download .

Run ipfs daemon --offline.

1 Like


Do what the error message says?

This brings up a question I was going to ask so I’ll just piggyback on this. There’s an api endpoint for updating the config. Do config updates require restarting the daemon? Some configs? All configs? What about the swarm key?

Restart is always required, afaik.

updated to version 0.8.0 (we don’t have a version 1 after 6 years and 300M$ budget ?! )
now if I offline the daemon , I’ll get this :

doesn’t it deserve a simple checkbox in the IPFS-Desktop settings panel ?

Seems the issue is reported already ipfs daemon --offline returns 400 for /api/v0/stats/bw · Issue #853 · ipfs/ipfs-webui · GitHub .

and it dates back to 2018 .
hector , brother , should we take IPFS serious ?

I very much think we should take it very seriously.

with known bugs neglected for 3 years ?

I can imagine that there were probably more important things/bugs to work on that this particular one :man_shrugging:

I can imagine that priority being Filecoin ICO and propaganda .
we’re talking 3 years , not 3 weeks . that’s enough time to expect a response to even the least important issues , let alone the ability of users to protect their resources from an overdemanding Network . which is a major game-theoric problem .
to be honest I’m as disappointed by the implementation as I’m fascinated with the idea .

Yes, most time goes into propaganda with a little bit for user support in the forums.

I know it is disheartening to find old really old seemingly simple issues that are still around, but a lot of things do get fixed otherwise.

Look at it in the other way: the only people that in 3 years have complained and took the time to discuss about this corner case (offline daemon with webui) are ipfs devs themselves. I suggest you drop a note in the issue saying that this is important for you as a user so that it shows that at least 1 user is annoyed by this.

1 Like

does IPFS have any bug fix bounty incentive mechanism to make coders roll up their sleeves on these bugs ? instead of just nagging about them ?
like instead of just hiring coders , they could put a price on the head of each bug and make the whole world involved in the development .