Commit Graph

62 Commits

Author SHA1 Message Date
rjl493456442 5adf4adc8e
eth/protocols/snap: cleanup dangling account trie nodes due to incomplete storage (#30258)
This pull request fixes #30229.
 
During snap sync, large storage will be split into several pieces and
synchronized concurrently. Unfortunately, the tradeoff is that the respective
merkle trie of each storage chunk will be incomplete due to the incomplete
boundaries. The trie nodes on these boundaries will be discarded, and any
dangling nodes on disk will also be removed if they fall on these paths,
ensuring the state healer won't be blocked.

However, the dangling account trie nodes on the path from the root to the
associated account are left untouched. This means the dangling account trie
nodes could potentially stop the state healing and break the assumption that the
entire subtrie should exist if the subtrie root exists. We should consider the
account trie node as the ancestor of the corresponding storage trie node.

In the scenarios described in the above ticket, the state corruption could occur
if there is a dangling account trie node while some storage trie nodes are
removed due to synchronization redo.

The fixing idea is pretty straightforward, the trie nodes on the path from root
to account should all be explicitly removed if an incomplete storage trie
occurs. Therefore, a `delete` operation has been added into `gentrie` to
explicitly clear the account along with all nodes on this path. The special
thing is that it's a cross-trie clearing. In theory, there may be a dangling
node at any position on this account key and we have to clear all of them.
2024-08-12 10:43:54 +02:00
tianyeyouyou a6751d6fc8
core/rawdb,eth/protocols,p2p: prealloc slice size (#29893)
chore: prealloc slice size
2024-06-03 15:51:04 +03:00
PolyMa 06263b1b35
all: fix typos in comments (#29873)
fix using `a` & `the` simutaneously
2024-05-29 12:24:10 +02:00
rjl493456442 473ee8fc07
trie, eth/protocols/snap: sanitize the committed node data (#29485) 2024-05-16 17:58:35 +08:00
rjl493456442 9f96e07c1c
core/rawdb, trie: improve db APIs for accessing trie nodes (#29362)
* core/rawdb, trie: improve db APIs for accessing trie nodes

* triedb/pathdb: fix
2024-04-30 16:25:35 +02:00
rjl493456442 d3c4466edd
core, eth/protocols/snap, trie: fix cause for snap-sync corruption, implement gentrie (#29313)
This pull request defines a gentrie for snap sync purpose.

The stackTrie is used to generate the merkle tree nodes upon receiving a state batch. Several additional options have been added into stackTrie to handle incomplete states (either missing states before or after).

In this pull request, these options have been relocated from stackTrie to genTrie, which serves as a wrapper for stackTrie specifically for snap sync purposes.

Further, the logic for managing incomplete state has been enhanced in this change. Originally, there are two cases handled:

-    boundary node filtering
-    internal (covered by extension node) node clearing

This changes adds one more:
 
- Clearing leftover nodes on the boundaries.

This feature is necessary if there are leftover trie nodes in database, otherwise node inconsistency may break the state healing.
2024-04-16 09:05:36 +02:00
rjl493456442 9dcf8aae47
eth/protocols/snap: skip retrieval for completed storages (#29378)
* eth/protocols/snap: skip retrieval for completed storages

* eth/protocols/snap: address comments from peter

* eth/protocols/snap: add comments
2024-04-10 12:02:45 +03:00
rjl493456442 304879da20
eth/protocols/snap: check storage root existence for hash scheme (#29341) 2024-03-27 09:35:33 +08:00
rjl493456442 ab04aeb855
core, eth, trie: filter out boundary nodes and remove dangling nodes in stacktrie (#28327)
* core, eth, trie: filter out boundary nodes in stacktrie

* eth/protocol/snap: add comments

* Update trie/stacktrie.go

Co-authored-by: Martin Holst Swende <martin@swende.se>

* eth, trie: remove onBoundary callback

* eth/protocols/snap: keep complete boundary nodes

* eth/protocols/snap: skip healing if the storage trie is already complete

* eth, trie: add more metrics

* eth, trie: address comment

---------

Co-authored-by: Martin Holst Swende <martin@swende.se>
2023-10-23 18:31:56 +03:00
rjl493456442 1b1611b8d0
core, trie, eth: refactor stacktrie constructor (#28350)
This change enhances the stacktrie constructor by introducing an option struct. It also simplifies the `Hash` and `Commit` operations, getting rid of the special handling round root node.
2023-10-17 14:09:25 +02:00
Martin Holst Swende f62c58f8de
trie: make rhs-proof align with last key in range proofs (#28311)
During snap-sync, we request ranges of values: either a range of accounts or a range of storage values. For any large trie, e.g. the main account trie or a large storage trie, we cannot fetch everything at once.

Short version; we split it up and request in multiple stages. To do so, we use an origin field, to say "Give me all storage key/values where key > 0x20000000000000000". When the server fulfils this, the server provides the first key after origin, let's say 0x2e030000000000000 -- never providing the exact origin. However, the client-side needs to be able to verify that the 0x2e03.. indeed is the first one after 0x2000.., and therefore the attached proof concerns the origin, not the first key.

So, short-short version: the left-hand side of the proof relates to the origin, and is free-standing from the first leaf.

On the other hand, (pun intended), the right-hand side, there's no such 'gap' between "along what path does the proof walk" and the last provided leaf. The proof must prove the last element (unless there are no elements).

Therefore, we can simplify the semantics for trie.VerifyRangeProof by removing an argument. This doesn't make much difference in practice, but makes it so that we can remove some tests. The reason I am raising this is that the upcoming stacktrie-based verifier does not support such fancy features as standalone right-hand borders.
2023-10-13 16:05:29 +02:00
rjl493456442 1cb3b6aee4
eth/protocols/snap: fix snap sync failure on empty storage range (#28306)
This change addresses an issue in snap sync, specifically when the entire sync process can be halted due to an encountered empty storage range.

Currently, on the snap sync client side, the response to an empty (partial) storage range is discarded as a non-delivery. However, this response can be a valid response, when the particular range requested does not contain any slots.

For instance, consider a large contract where the entire key space is divided into 16 chunks, and there are no available slots in the last chunk [0xf] -> [end]. When the node receives a request for this particular range, the response includes:

    The proof with origin [0xf]
    A nil storage slot set

If we simply discard this response, the finalization of the last range will be skipped, halting the entire sync process indefinitely. The test case TestSyncWithUnevenStorage can reproduce the scenario described above.

In addition, this change also defines the common variables MaxAddress and MaxHash.
2023-10-13 09:08:26 +02:00
Martin Holst Swende 8976a0c97a
trie: remove owner and binary marshaling from stacktrie (#28291)
This change
  - Removes the owner-notion from a stacktrie; the owner is only ever needed for comitting to the database, but the commit-function, the `writeFn` is provided by the caller, so the caller can just set the owner into the `writeFn` instead of having it passed through the stacktrie.
  - Removes the `encoding.BinaryMarshaler`/`encoding.BinaryUnmarshaler` interface from stacktrie. We're not using it, and it is doubtful whether anyone downstream is either.
2023-10-11 06:12:45 +02:00
Martin Holst Swende 6b1e4f4211
all: move light.NodeSet to trienode.ProofSet (#28287)
This is a minor refactor in preparation of changes to range verifier. This PR contains no intentional functional changes but moves (and renames) the light.NodeSet
2023-10-10 10:30:47 +02:00
Péter Szilágyi 6e934f40f9
eth/protocols/snap: fix batch writer when resuming an aborted sync (#27842) 2023-08-03 14:51:02 +03:00
rjl493456442 0e5d2c7c53
core/state/snapshot, core/types, eth: move account definition to type (#27323)
* core/state/snapshot, core/types, eth: move account definition to type

* core, eth: revert snapshot Account API change
2023-06-06 11:17:39 +03:00
s7v7nislands f541cad272
eth: use new atomic types (#27137) 2023-04-25 12:06:50 +02:00
rjl493456442 fe01a2f63b
all: use unified emptyRootHash and emptyCodeHash (#26718)
The EmptyRootHash and EmptyCodeHash are defined everywhere in the codebase, this PR replaces all of them with unified one defined in core/types package, and also defines constants for TxRoot, WithdrawalsRoot and UncleRoot
2023-02-21 06:12:27 -05:00
Martin Holst Swende 8860b39754
all: prepare for path-based trie storage (#26603)
This PR moves some trie-related db accessor methods to a different file, and also removes the schema type. Instead of the schema type, a string is used to distinguish between hashbased/pathbased db accessors.
This also moves some code from trie package to rawdb package.

This PR is intended to be a no-functionality-change prep PR for #25963 .

---------

Co-authored-by: Gary Rong <garyrong0905@gmail.com>
2023-02-06 10:28:40 -05:00
rjl493456442 6c149fd4ad
core, eth, trie, light: clean up trie interface (#26388)
* all: cleanup trie interface

* eth, trie: address comments
2023-01-03 15:41:40 +02:00
rjl493456442 743e404906
core, eth, les, tests, trie: abstract node scheme (#25532)
This PR introduces a node scheme abstraction. The interface is only implemented by `hashScheme` at the moment, but will be extended by `pathScheme` very soon.

Apart from that, a few changes are also included which is worth mentioning:

-  port the changes in the stacktrie, tracking the path prefix of nodes during commit
-  use ethdb.Database for constructing trie.Database. This is not necessary right now, but it is required for path-based used to open reverse diff freezer
2022-11-28 14:31:28 +01:00
Martin Holst Swende 5a02b2d6d0
all: fix spelling mistakes (#25961) 2022-10-11 09:37:00 +02:00
Martin Holst Swende 456b187892
trie, eth/protocols/snap: less tiny db-writes during trie-healing (#25681)
This PR prevent making tiny writes during state healing, by only performing the batch-write if the accumulated data is large enough to be meaningful.
2022-09-28 08:08:18 +02:00
Martin Holst Swende fb500d12d5
eth/protocols/snap: make log messages more clear that sync is ongoing (#25837)
* eth/protocols/snap: make log messages more clear that sync is ongoing

* Update sync.go

Co-authored-by: Felix Lange <fjl@twurst.com>
2022-09-23 11:11:41 +03:00
Péter Szilágyi de8d5fa042
eth/protocols/snap: throttle trie heal requests when peers DoS us (#25666)
* eth/protocols/snap: throttle trie heal requests when peers DoS us

* eth/protocols/snap: lower heal throttle log to debug

Co-authored-by: Martin Holst Swende <martin@swende.se>

* eth/protocols/snap: fix comment

Co-authored-by: Martin Holst Swende <martin@swende.se>
2022-09-09 11:42:57 +03:00
Martin Holst Swende dafa40e7a7
eth/protocols/snap: fix problems due to idle-but-busy peers 2022-08-31 17:58:18 +02:00
Justin Traglia 2c5648d891
all: fix some typos (#25551)
* Fix some typos

* Fix some mistakes

* Revert 4byte.json

* Fix an incorrect fix

* Change files to fails
2022-08-19 09:00:21 +03:00
rjl493456442 1657e43931
core, les, eth: port snap sync changes (#24898)
core, eth, les, trie: rework snap sync
2022-07-15 14:55:51 +03:00
Martin Holst Swende a907d7e81a
all: more linters (#24783)
This enables the following linters

- typecheck
- unused
- staticcheck
- bidichk
- durationcheck
- exportloopref
- gosec

WIth a few exceptions.

- We use a deprecated protobuf in trezor. I didn't want to mess with that, since I cannot meaningfully test any changes there.
- The deprecated TypeMux is used in a few places still, so the warning for it is silenced for now.
- Using string type in context.WithValue is apparently wrong, one should use a custom type, to prevent collisions between different places in the hierarchy of callers. That should be fixed at some point, but may require some attention.
- The warnings for using weak random generator are squashed, since we use a lot of random without need for cryptographic guarantees.
2022-06-13 16:24:45 +02:00
rjl493456442 22defa5af7
all: introduce trie owner notion (#24750)
* cmd, core/state, light, trie, eth: add trie owner notion

* all: refactor

* tests: fix goimports

* core/state/snapshot: fix ineffasigns

Co-authored-by: Martin Holst Swende <martin@swende.se>
2022-06-06 17:14:55 +02:00
Martin Holst Swende 646503208e
eth/protocols/snap: sort trienode heal requests by path (#24779)
* sort snap trienode heal requests

* eth/protocols/snap: remove debug code

* eth/protocols/snap: simplify sort, generate pathsets later

* eth/protocols/snap: review concern

* eth/protocols/snap: renamings

* eth/protocols/snap: add comments in Merge

* eth/protocols/snap: remove variable 'last' in Merge

* eth/protocols/snap: fix lint flaws in test

Co-authored-by: Felix Lange <fjl@twurst.com>
2022-05-10 17:37:24 +03:00
Marius van der Wijden 86d5477079
core/state/snapshot: fix race condition (#24685)
Fixes three race conditions found through fuzzing by David Theodore
2022-05-06 17:20:41 +02:00
Péter Szilágyi 1b58e42802
Merge pull request #24529 from holiman/fix_inf
eth/protocols/snap: avoid estimating infinite percentage
2022-03-11 10:40:55 +02:00
沉风 7d3ecca451
eth/protocols/snap: a little typo Merkel -> Merkle (#24530) 2022-03-11 10:32:08 +02:00
Martin Holst Swende 658415960e
eth/protocols/snap: avoid estimating infinite percentage 2022-03-10 12:46:48 +01:00
Martin Holst Swende 893502e561
trie, core, eth: use db.has over db.get where possible 2021-12-15 16:16:45 +01:00
Péter Szilágyi 58d1988349
core, eth, les, trie: remove the sync bloom, used by fast sync 2021-12-03 12:32:41 +02:00
Péter Szilágyi c10a0a62c3
eth: request id dispatcher and direct req/reply APIs (#23576)
* eth: request ID based message dispatcher

* eth: fix dispatcher cancellation, rework fetchers idleness tracker

* eth/downloader: drop peers who refuse to serve advertised chains
2021-11-26 13:26:03 +02:00
Guillaume Ballet 443afc975c
core/state: move state account to core/types + abstracted "write account to trie" (#23567)
* core/state: abstracted "write account to trie" method

* fix appveyor build

* Apply suggestions from code review

Co-authored-by: Martin Holst Swende <martin@swende.se>

* review feedback

* core/state/accounts: move Account to core/types

* core/types: rename Account -> StateAccount

* core/state: restore EncodeRLP for stateObject

* core/types: add the missing file

* more review feedback

* more review feedback

Co-authored-by: Martin Holst Swende <martin@swende.se>
Co-authored-by: Felix Lange <fjl@twurst.com>
2021-09-28 10:48:07 +02:00
Felix Lange 427175153c
p2p/msgrate: return capacity as integer, clamp to max uint32 (#22943)
* p2p/msgrate: return capacity as integer

* eth/protocols/snap: remove conversions

* p2p/msgrate: add overflow test

* p2p/msgrate: make the capacity overflow test actually overflow

* p2p/msgrate: clamp capacity to max int32

* p2p/msgrate: fix min/max confusion
2021-05-27 19:43:55 +03:00
Péter Szilágyi 3e795881ea
eth, p2p/msgrate: move peer QoS tracking to its own package and use it for snap (#22876)
This change extracts the peer QoS tracking logic from eth/downloader, moving
it into the new package p2p/msgrate. The job of msgrate.Tracker is determining
suitable timeout values and request sizes per peer.

The snap sync scheduler now uses msgrate.Tracker instead of the hard-coded 15s
timeout. This should make the sync work better on network links with high latency.
2021-05-19 14:09:03 +02:00
Péter Szilágyi 52b5d2d869
eth/protocols/snap: use storage batch, not account batch in st task 2021-04-30 18:24:34 +03:00
Péter Szilágyi 745757ac6b
core, eth: abort snapshot generation on snap sync and resume later 2021-04-30 17:03:10 +03:00
Péter Szilágyi fae165a5de
core, eth, ethdb, trie: simplify range proofs 2021-04-29 10:59:08 +03:00
Martin Holst Swende a81cf0d2b3
trie: remove redundant returns + use stacktrie where applicable (#22760)
* trie: add benchmark for proofless range

* trie: remove unused returns + use stacktrie
2021-04-28 22:47:48 +03:00
Péter Szilágyi 558bff4008
eth/protocols/snap: lower the packet size to avoid overloading link 2021-04-28 21:40:06 +03:00
Péter Szilágyi caea6c4661
eth/protocols/snap: generate storage trie from full dirty snap data (#22668)
* eth/protocols/snap: generate storage trie from full dirty snap data

* eth/protocols/snap: get rid of some more dead code

* eth/protocols/snap: less frequent logs, also log during trie generation

* eth/protocols/snap: implement dirty account range stack-hashing

* eth/protocols/snap: don't loop on account trie generation

* eth/protocols/snap: fix account format in trie

* core, eth, ethdb: glue snap packets together, but not chunks

* eth/protocols/snap: print completion log for snap phase

* eth/protocols/snap: extended tests

* eth/protocols/snap: make testcase pass

* eth/protocols/snap: fix account stacktrie commit without defer

* ethdb: fix key counts on reset

* eth/protocols: fix typos

* eth/protocols/snap: make better use of delivered data (#44)

* eth/protocols/snap: make better use of delivered data

* squashme

* eth/protocols/snap: reduce chunking

* squashme

* eth/protocols/snap: reduce chunking further

* eth/protocols/snap: break out hash range calculations

* eth/protocols/snap: use sort.Search instead of looping

* eth/protocols/snap: prevent crash on storage response with no keys

* eth/protocols/snap: nitpicks all around

* eth/protocols/snap: clear heal need on 1-chunk storage completion

* eth/protocols/snap: fix range chunker, add tests

Co-authored-by: Péter Szilágyi <peterke@gmail.com>

* trie: fix test API error

* eth/protocols/snap: fix some further liter issues

* eth/protocols/snap: fix accidental batch reuse

Co-authored-by: Martin Holst Swende <martin@swende.se>
2021-04-27 17:19:59 +03:00
Péter Szilágyi 9553c98de8
eth/protocols/snap: use ephemeral channels to avoid cross-sync delveries 2021-04-15 21:16:54 +03:00
Péter Szilágyi 1e207342b5
all: make logs a bit easier on the eye to digest (#22665)
* all: add thousandths separators for big numbers on log messages

* p2p/sentry: drop accidental file

* common, log: add fast number formatter

* common, eth/protocols/snap: simplifty fancy num types

* log: handle nil big ints
2021-04-15 20:35:00 +03:00
gary rong 7088f1e814
core, eth: faster snapshot generation (#22504)
* eth/protocols: persist received state segments

* core: initial implementation

* core/state/snapshot: add tests

* core, eth: updates

* eth/protocols/snapshot: count flat state size

* core/state: add metrics

* core/state/snapshot: skip unnecessary deletion

* core/state/snapshot: rename

* core/state/snapshot: use the global batch

* core/state/snapshot: add logs and fix wiping

* core/state/snapshot: fix

* core/state/snapshot: save generation progress even if the batch is empty

* core/state/snapshot: fixes

* core/state/snapshot: fix initial account range length

* core/state/snapshot: fix initial account range

* eth/protocols/snap: store flat states during the healing

* eth/protocols/snap: print logs

* core/state/snapshot: refactor (#4)

* core/state/snapshot: refactor

* core/state/snapshot: tiny fix and polish

Co-authored-by: rjl493456442 <garyrong0905@gmail.com>

* core, eth: fixes

* core, eth: fix healing writer

* core, trie, eth: fix paths

* eth/protocols/snap: fix encoding

* eth, core: add debug log

* core/state/generate: release iterator asap (#5)

core/state/snapshot: less copy

core/state/snapshot: revert split loop

core/state/snapshot: handle storage becoming empty, improve test robustness

core/state: test modified codehash

core/state/snapshot: polish

* core/state/snapshot: optimize stats counter

* core, eth: add metric

* core/state/snapshot: update comments

* core/state/snapshot: improve tests

* core/state/snapshot: replace secure trie with standard trie

* core/state/snapshot: wrap return as the struct

* core/state/snapshot: skip wiping correct states

* core/state/snapshot: updates

* core/state/snapshot: fixes

* core/state/snapshot: fix panic due to reference flaw in closure

* core/state/snapshot: fix errors in state generation logic + fix log output

* core/state/snapshot: remove an error case

* core/state/snapshot: fix condition-check for exhausted snap state

* core/state/snapshot: use stackTrie for small tries

* core/state/snapshot: don't resolve small storage tries in vain

* core/state/snapshot: properly clean up storage of deleted accounts

* core/state/snapshot: avoid RLP-encoding in some cases + minor nitpicks

* core/state/snapshot: fix error (+testcase)

* core/state/snapshot: clean up tests a bit

* core/state/snapshot: work in progress on better tests

* core/state/snapshot: polish code

* core/state/snapshot: fix trie iteration abortion trigger

* core/state/snapshot: fixes flaws

* core/state/snapshot: remove panic

* core/state/snapshot: fix abort

* core/state/snapshot: more tests (plus failing testcase)

* core/state/snapshot: more testcases + fix for failing test

* core/state/snapshot: testcase for malformed data

* core/state/snapshot: some test nitpicks

* core/state/snapshot: improvements to logging

* core/state/snapshot: testcase to demo error in abortion

* core/state/snapshot: fix abortion

* cmd/geth: make verify-state report the root

* trie: fix failing test

* core/state/snapshot: add timer metrics

* core/state/snapshot: fix metrics

* core/state/snapshot: udpate tests

* eth/protocols/snap: write snapshot account even if code or state is needed

* core/state/snapshot: fix diskmore check

* core/state/snapshot: review fixes

* core/state/snapshot: improve error message

* cmd/geth: rename 'error' to 'err' in logs

* core/state/snapshot: fix some review concerns

* core/state/snapshot, eth/protocols/snap: clear snapshot marker when starting/resuming snap sync

* core: add error log

* core/state/snapshot: use proper timers for metrics collection

* core/state/snapshot: address some review concerns

* eth/protocols/snap: improved log message

* eth/protocols/snap: fix heal logs to condense infos

* core/state/snapshot: wait for generator termination before restarting

* core/state/snapshot: revert timers to counters to track total time

Co-authored-by: Martin Holst Swende <martin@swende.se>
Co-authored-by: Péter Szilágyi <peterke@gmail.com>
2021-04-14 23:23:11 +03:00