Thanks! So does that mean that data persistence is more feasible running an ipfs cluster than just running a private ipfs network?
Not quite sure how to reply. Running a private IPFS network and an IPFS Cluster arenât mutually exclusive but your choices do come with some constraints. A private network is just a collection of IPFS nodes that you have restricted to only talk to each other. Itâs like an intranet with no gateway.[1] The only people at the party are people you invited. Given that youâre having a private party you canât ask a random stranger to hold your beer. So whatever redundancy youâre going to have within your private group is going to be whatever you setup. There is, as I mentioned, one pinning service that Iâm aware of that that will come to private parties and thatâs Temporal.
If it wasnât a private network you could use any pinning service youâd like. For your private network youâd have to come up with some solution. You could do it manually, âHey Bob, can you pin this for me?â. You could setup the cluster which would replicate anything that you explicitly put onto the cluster itself. Unfortunately Iâm not aware of any open source pinning service implementations.
[1] Reading back over this that might be a little confusing because IPFS has gateways and I suppose you could (I think) have a private IPFS network with a public IPFS gateway but I donât know Iâve never considered something like that.
Thanks Zachary for your patient response each time. I appreciate it man!
It looks like there is an opensource implementation of a pinning service.
So you can set this up in front of either a single node or a cluster with multiple nodes and then have your users subscribe to the pinning service for redundancy on your private network. I havenât used it so YMMMV.
Thanks Zachary! But I do not know what âYMMMVâ stands for. Sorry, I am not so Internet language savvy lol
Whoops, one too many Mâs. Your Milage May Vary. Donât worry, my kids constantly remind me of how uncool I am and how unsavvy Iâve become.
Lol, thanks Zachary!