p2p, p2p/discover: misc connectivity improvements #16069
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This changes node revalidation to be periodic instead of on-demand. This should prevent issues where dead nodes get stuck in closer buckets because no other node will ever come along to replace them. Every 5 seconds (on average), the last node in a random bucket is checked and moved to the front of the bucket if it is still responding. If revalidation fails, the last node is replaced by an entry of the 'replacement list' containing recently-seen nodes. Most close buckets are removed because it's very unlikely we'll ever encounter a node that would fall into any of those buckets. Table seeding is also improved: we now require a few minutes of table membership before considering a node as a potential seed node. This should make it less likely to store short-lived nodes as potential seeds.
We would skip sending neighbors replies if there were fewer than maxNeighbors results and CheckRelayIP returned an error for the last one. While here, also resolve a TODO about pong reply tokens.
karalabe
reviewed
Feb 12, 2018
| // | ||
| // The first byte of key is '4' or '6' to distinguish IPv4/IPv6 address types. | ||
| // The remainder of the key is the IP, truncated to the number of bits. | ||
| func (s *DistinctNetSet) key(ip net.IP) net.IP { |
There was a problem hiding this comment.
Any reason not to wrap the return into a string already here? Any code that uses this from the outside keeps casting to string, 2-3 times, which is an extra copy each time. I know the overhead is probably irrelevant, but API wise it might still be cleaner. Unless I'm missing something.
karalabe
reviewed
Feb 12, 2018
| {"0.33.0.1", "0.34.0.2", 8, true}, | ||
| {"0.33.0.1", "0.34.0.2", 13, true}, | ||
| {"0.33.0.1", "0.34.0.2", 15, false}, | ||
| } |
There was a problem hiding this comment.
Could we have a couple test cases with invalid bits? i.e. 64 bits for ipv4? Just to make sure some bug won't crash the process.
karalabe
reviewed
Feb 12, 2018
| } | ||
|
|
||
| // DistinctNetSet tracks IPs, ensuring that at most N of them | ||
| // fall into the same network range. |
There was a problem hiding this comment.
Can the same IP be present multiple times? Might be an interesting thing to note.
karalabe
reviewed
Feb 12, 2018
| closeReq: make(chan struct{}), | ||
| closed: make(chan struct{}), | ||
| rand: mrand.New(mrand.NewSource(0)), |
There was a problem hiding this comment.
Doesn't always using 0 as the random source make this too predictable?
There was a problem hiding this comment.
Ah, I see you seed it below again, this is just creating it. Nvm.
karalabe
reviewed
Feb 12, 2018
| if len(result) > 0 { | ||
| return | ||
| // We perform a few lookups with a random target instead. | ||
| for i := 0; i < 3; i++ { |
There was a problem hiding this comment.
Any rational behind doing 3 instead of 1? And any rationale behind the number 3?
There was a problem hiding this comment.
3 > 1, so more stuff goes in the table.
karalabe
reviewed
Feb 12, 2018
| // Load nodes from the database and insert | ||
| // them. This should yield a few previously seen nodes that are | ||
| // (hopefully) still alive. | ||
| tab.loadSeedNodes(true) |
There was a problem hiding this comment.
Previously we only loaded nodes if the refresh failed. Now we load seed nodes always. What's the rationale?
karalabe
reviewed
Feb 12, 2018
| r := b.replacements[tab.rand.Intn(len(b.replacements))] | ||
| b.replacements = deleteNode(b.replacements, r) | ||
| b.entries[len(b.entries)-1] = r | ||
| tab.removeIP(b, last.IP) |
There was a problem hiding this comment.
Don't we want to do a tab.addIP(r) here?
There was a problem hiding this comment.
Ah, the replacement peers are already tracked by IP. Nvm.
holiman
approved these changes
Feb 12, 2018
karalabe
approved these changes
Feb 12, 2018
prestonvanloon
pushed a commit
to prestonvanloon/go-ethereum
that referenced
this pull request
Apr 2, 2018
* p2p: add DialRatio for configuration of inbound vs. dialed connections * p2p: add connection flags to PeerInfo * p2p/netutil: add SameNet, DistinctNetSet * p2p/discover: improve revalidation and seeding This changes node revalidation to be periodic instead of on-demand. This should prevent issues where dead nodes get stuck in closer buckets because no other node will ever come along to replace them. Every 5 seconds (on average), the last node in a random bucket is checked and moved to the front of the bucket if it is still responding. If revalidation fails, the last node is replaced by an entry of the 'replacement list' containing recently-seen nodes. Most close buckets are removed because it's very unlikely we'll ever encounter a node that would fall into any of those buckets. Table seeding is also improved: we now require a few minutes of table membership before considering a node as a potential seed node. This should make it less likely to store short-lived nodes as potential seeds. * p2p/discover: fix nits in UDP transport We would skip sending neighbors replies if there were fewer than maxNeighbors results and CheckRelayIP returned an error for the last one. While here, also resolve a TODO about pong reply tokens.
|
Does |
mariameda
pushed a commit
to NiluPlatform/go-nilu
that referenced
this pull request
Aug 23, 2018
* p2p: add DialRatio for configuration of inbound vs. dialed connections * p2p: add connection flags to PeerInfo * p2p/netutil: add SameNet, DistinctNetSet * p2p/discover: improve revalidation and seeding This changes node revalidation to be periodic instead of on-demand. This should prevent issues where dead nodes get stuck in closer buckets because no other node will ever come along to replace them. Every 5 seconds (on average), the last node in a random bucket is checked and moved to the front of the bucket if it is still responding. If revalidation fails, the last node is replaced by an entry of the 'replacement list' containing recently-seen nodes. Most close buckets are removed because it's very unlikely we'll ever encounter a node that would fall into any of those buckets. Table seeding is also improved: we now require a few minutes of table membership before considering a node as a potential seed node. This should make it less likely to store short-lived nodes as potential seeds. * p2p/discover: fix nits in UDP transport We would skip sending neighbors replies if there were fewer than maxNeighbors results and CheckRelayIP returned an error for the last one. While here, also resolve a TODO about pong reply tokens.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.