-
-
Notifications
You must be signed in to change notification settings - Fork 3.1k
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
DHT Refactoring #160
Comments
We have a good number of methods that make a single one hop request to a single given peer, i have most of these methods suffixed with 'Single' but it would be nice to come up with a name that more clearly describes their purpose and refactor them to have a similar interface. and potentially put them all in their own file. These methods are:
|
The clustering idea needs to be abstracted out more, and some pleasant interface should be made. Maybe having a channel to read peers from would be the easiest interface to satisfy, then the object can automatically move to more distant peers as it runs out of closer ones. |
@jbenet what do you think of having a peer type for the dht that contains context about routing levels? Something like a LeveledPeer, that (for coral type routing scenarios) contains information about which level of cluster its in |
All this sgtm. there's also a dht interface lying around somewhere in an issue (that both standard kademlia, coral, etc can implement for the general "routing-over-dht" |
find self in DHT when bootstrapping
This ensures cid-codec introduced in #8568 gets correctly passed (+ we maintain backward-compatibility with CIDv0) This commit was moved from ipfs/go-ipfs-http-client@9c9f43f
As @jbenet is quick to mention, the DHT could use a bit more abstraction, this issue will be a place to discuss those changes.
The text was updated successfully, but these errors were encountered: