NAT Parameter broke in 2.43.0 #7472
Comments
|
hi. checked. nat working just we do log wrong value |
AskAlexSharov
added a commit
that referenced
this issue
May 10, 2023
|
Thanks, will close the issue and check on next release! 👍 |
blxdyx
pushed a commit
to blxdyx/bsc-erigon
that referenced
this issue
Jun 30, 2023
calmbeing
pushed a commit
to calmbeing/bsc-erigon
that referenced
this issue
Jul 12, 2023
calmbeing
added a commit
to node-real/bsc-erigon
that referenced
this issue
Jul 18, 2023
* fix build issue * simplify the unwind * Add additional trie proof testing (erigontech#7382) This PR extends the function merged in erigontech#7337 to cover the proof generation paths as well. The first commit simply migrates the internal proof checking code from the rpc `eth_getProof` test to be available externally exported under `turbo/trie`. In the process, it translates the in place testing assertions to return more normally as errors and adapts to use the pre-existing trie node types which are defined slightly differently (although this code is still intended only for testing purposes). The second commit refactors the existing trie fuzzing tests from the previous PR and adds new fuzzing tests for the proof generation. In order to validate the proofs, these fuzzing tests fundamentally do two things. Firstly, after bootstrapping the test (by seeding, and modifying the db), for each key we compute the 'naive' proof utilizing the existing code in `proof.go` and the in memory `trie.Trie` structure. The `trie.Trie` code actually had a couple small bugs which are fixed in this PR (not handling value nodes, and not honoring the `NewTestRLPTrie` contract of pre-RLP encoded nodes in proof generation). Secondly, we re-compute the same proof for the flatDB production variant, and verify that it is exactly the same proof as computed by the naive implementation. This fuzzing has been run for ~72 hours locally with no errors. Although this code hasn't identified any new bugs in the proof generation path, it improves coverage and should help to prevent regressions. Additional extensions will be provided in a subsequent PR. --------- Co-authored-by: Jason Yellick <jason@enya.ai> * e3: drain when no new tasks causing deadlock (erigontech#7395) * Enable negative Merkle proofs for eth_getProof (erigontech#7393) This addresses the last known deficiency of the eth_getProof implementation. The previous code would return an error in the event that the element was not found in the trie. EIP-1186 allows for 'negative' proofs where a proof demonstrates that an element cannot be in the trie, so this commit updates the logic to support that case. Co-authored-by: Jason Yellick <jason@enya.ai> * e3: dont fetch code hash in unwind (erigontech#7416) * Caplin: Fixed sentinel deadlock (erigontech#7419) * fix block rlp logging bug (erigontech#7441) fixes a logging statement was logging the outer rpc proto instead of the inner rlp of the block as intended * fix: erigon_getLatestLogs (erigontech#7450) When calling erigon_getLatestLogs, I was getting a crash in [erigon_receipts.go](https://github.com/ledgerwatch/erigon/blob/beb97784d43ece5acde365a74efe8763692ebdba/cmd/rpcdaemon/commands/erigon_receipts.go#L254). I think it is a simple indexing bug ``` [service.go:217 panic.go:884 panic.go:113 erigon_receipts.go:254 value.go:586 value.go:370 service.go:222 handler.go:494 handler.go:444 handler.go:392 handler.go:223 handler.go:316 asm_amd64.s:1598] [WARN] [05-05|21:13:59.749] Served conn=100.70.204.111:50141 method=erigon_getLatestLogs reqid=1 t=621.5936ms err="method handler crashed" ``` * LeakDetector: use it to find which resource was created but not closed (leaked) (erigontech#7457) * reduce default --db.size.limit from 7 to 3 Tb (to fit defaults of some systems) (erigontech#7478) * Enode logging broke when NAT Parameter set in 2.43.0 (erigontech#7480) for erigontech#7472 * erigon init: created db with wrong pageSize (erigontech#7482) * p2p: fix peer ID serialization (erigontech#7495) The peer ID in sentry.proto is a H512 / 64 bytes value, and MarshalPubkey creates it from a public key. There's no need to cut the first byte, because MarshalPubkey already does it. Doing so results in a 63 bytes value that is incompatible with silkworm sentry. * Allow ephemeral ports for p2p (erigontech#7503) Currently, the p2p ports require an explicit enumeration of ports to pick. Sometimes, for instance when writing integration tests utilizing an Erigon binary the particular p2p port does not matter and trying to pick non-allocated port ranges is fragile. This small PR simply checks to see if the enumerated port is '0', in which case it disables the probing check which would otherwise cause Erigon not to try binding to an ephemeral port. Co-authored-by: Jason Yellick <jason@enya.ai> * remove "db" log line from (erigontech#7509) * e2: self-heal after accidental blocks delete (download blocks eventually, then work as usual) (erigontech#7516) * e2: ReadAhead of blocks, senders accounts, code (erigontech#7501) It improves performance of initial sync (stage exec) by 5-20% when blocks snapshots are mounted to high-latency drive and when chaindata is on high-latency drive. And improving cold-start performance. Current implementation using 2 goroutines for ReadAhead. It also producing more garbage, can improve it later (here are dashboard with impact). ``` mainnet2-1: with ReadAhead mainnet2-3: no ReadAhead ``` <img width="949" alt="Screenshot 2023-05-12 at 09 24 31" src="https://github.com/ledgerwatch/erigon/assets/46885206/b90b1fa8-9099-48ff-95b3-86e864a36d46"> <img width="845" alt="Screenshot 2023-05-12 at 09 24 13" src="https://github.com/ledgerwatch/erigon/assets/46885206/39d90c0c-a9d5-4735-8c03-da1455b147aa"> * Fix trace_filter regression due to gasBailout argument (erigontech#7539) (erigontech#7540) Co-authored-by: Alex Sharp <alexsharp@Alexs-MacBook-Pro-2.local> * Update t8n_test.go (erigontech#7550) * e2: reduce StageSyncStep interface size (erigontech#7561) * e2: avoid do RestoreCodeHash twice (erigontech#7706) - do it only once in HistoryStateReader * StagedSync: fix canRunCycleInOneTransaction logic (erigontech#7712) * Better version of libp2p where losing peers bug fixed (erigontech#7726) Co-authored-by: Alex Sharov <AskAlexSharov@gmail.com> * fix merge error --------- Co-authored-by: Jason Yellick <7431583+jyellick@users.noreply.github.com> Co-authored-by: Jason Yellick <jason@enya.ai> Co-authored-by: Alex Sharov <AskAlexSharov@gmail.com> Co-authored-by: Giulio rebuffo <giulio.rebuffo@gmail.com> Co-authored-by: Roberto Bayardo <bayardo@alum.mit.edu> Co-authored-by: Joe Netti <joe@netti.dev> Co-authored-by: battlmonstr <battlmonstr@users.noreply.github.com> Co-authored-by: ledgerwatch <akhounov@gmail.com> Co-authored-by: Alex Sharp <alexsharp@Alexs-MacBook-Pro-2.local>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
System information
Erigon version: 2.43.0
OS & Version: Ubuntu 22.04
Chain/Network: Eth, Polygon... etc
Expected behaviour
The NAT parameter no longer seems to work as intended.
Using for example
nat=extip:1.2.3.4... should result in an enode address with IP of
1.2.3.4Actual behaviour
Instead, it just shows
127.0.0.1. The port also shows as0, so it's unclear if one or more parameters such as '--port' are also faulty.self=enode://xxxxxx@127.0.0.1:0 name=erigon/v2.43.0-stable-1106475e/linux-amd64/go1.20.3Therefore, it's currently not possible to set a valid external IP:port for peering. The assigned IP set by the flag seems to work for the torrent feature though, according to the startup logs (
[torrent] Public IP ip=1.2.3.4).Everything was working on
v2.42.0, tested on a few different nodes on both Eth and Polygon, before and after upgrading.Steps to reproduce the behaviour
Use the nat flag to set an external IP:
--nat=extip:${PUBLIC_IP}The text was updated successfully, but these errors were encountered: