refactor: sample impl with new p2p #636
Conversation
arkadiuszos4chain
force-pushed
the
refactor-p2p-experimental
branch
2 times, most recently
from
df39d25
to
ea6b386
Compare
arkadiuszos4chain
force-pushed
the
refactor-p2p-experimental
branch
from
ea6b386
to
4f9b01c
Compare
|
|
|
||
| replace github.com/libsv/go-p2p => github.com/arkadiuszos4chain/go-p2p v0.0.0-20241108223608-ded470665895 |
There was a problem hiding this comment.
Just a reminder - to be removed and replaced with go-p2p v2
| // GetPeersNmber returns a list of connected and a list of disconnected peers | ||
| func (p *Processor) GetPeersNmber() uint { | ||
| return p.pm.CountConnectedPeers() | ||
| } |
There was a problem hiding this comment.
I don't know if it was on purpose, but just in case 😄
| // GetPeersNmber returns a list of connected and a list of disconnected peers | |
| func (p *Processor) GetPeersNmber() uint { | |
| return p.pm.CountConnectedPeers() | |
| } | |
| // GetPeersNumber returns a list of connected and a list of disconnected peers | |
| func (p *Processor) GetPeersNumber() uint { | |
| return p.pm.CountConnectedPeers() | |
| } |
| // TODO: do we need this kind of info? do we check it? | ||
|
|
||
| peersConnected := make([]string, 0, len(peers)) | ||
| peersDisconnected := make([]string, 0, len(peers)) | ||
| for _, peer := range peers { | ||
| if peer.Connected() { | ||
| peersConnected = append(peersConnected, peer.String()) | ||
| } else { | ||
| peersDisconnected = append(peersDisconnected, peer.String()) | ||
| } | ||
| } | ||
| // peers := s.processor.GetPeers() | ||
|
|
||
| // peersConnected := make([]string, 0, len(peers)) | ||
| // peersDisconnected := make([]string, 0, len(peers)) | ||
| // for _, peer := range peers { | ||
| // if peer.Connected() { | ||
| // peersConnected = append(peersConnected, peer.String()) | ||
| // } else { | ||
| // peersDisconnected = append(peersDisconnected, peer.String()) | ||
| // } | ||
| // } |
There was a problem hiding this comment.
It's not my call, but IMO we can remove it
There was a problem hiding this comment.
In the new implementation, there is no need to distinguish between a connected peer and a healthy one. If the peer is connected, it is healthy. If it becomes unhealthy (error, ping-pong failed), it will disconnect and wait for the client(owner) to initiate a reconnect
| // // mark that we retried processing this transaction once more | ||
| // if err = p.store.IncrementRetries(ctx, tx.Hash); err != nil { | ||
| // p.logger.Error("Failed to increment retries in database", slog.String("err", err.Error())) | ||
| // } | ||
|
|
There was a problem hiding this comment.
If you're using IncrementRetriesForMany, this comment can be removed, right?
Description of Changes
Sample implementation using the new experimental P2P functionality.
Several files have been modified due to a package name change in go-p2p, but the most significant changes are in:
internal/blocktx/p2p- includes all P2P logic, a new message handler (equivalent to the previous peer_handler), initialization, etc.internal/metamorph/p2p- new message handler, also equivalent to the previous peer_handlerinternal/metamorph/processor.go- updates inStartProcessExpiredTransactions(), now sending/requesting transactions in batchesinternal/metamorph/store/postgresql/postgres- introducesIncrementRetriesForMany()Linked Issues / Tickets
Reference any related issues or tickets, e.g. "Closes #123".
Testing Procedure
Describe the tests you've added or any testing steps you've taken.
Checklist:
CHANGELOG.mdwith my changes