* removed check from fcU if new blocks are descendants of the previous finalized block
* backwards sync: remove check that block must be a descendant of finalized
Signed-off-by: Daniel Lehrner <daniel.lehrner@consensys.net>
* If needed update peer chain state when processing the block headers response
Signed-off-by: Fabio Di Fabio <fabio.difabio@consensys.net>
* Do not require a minimum block height when downloading headers or blocks
Peers could have that header, bacause our internal record about the
status of the peer could not always be up-to-date, so it is better
to avoid setting that constraint when selecting peers to download block
headers.
Signed-off-by: Fabio Di Fabio <fabio.difabio@consensys.net>
* Add CHANGELOG
Signed-off-by: Fabio Di Fabio <fabio.difabio@consensys.net>
* Complete the removal of minimumRequiredBlockNumber from constructors
Signed-off-by: Fabio Di Fabio <fabio.difabio@consensys.net>
Co-authored-by: Sally MacFarlane <sally.macfarlane@consensys.net>
* Backward Sync is now remembering recent finalized blocks.
When syncing using BWS we now consider finalized blocks reported by consensus layer.
When no finalized block is specified, nothing strange will go on. When Finalized block is specified
But Besu is not aware of any finalized block yet, we will sync as usual until the finalized block is about to get imported.
A new invariant is introduced when importing blocks using BWS.
All imported blocks has to descent from latest finalized block or a TTD block if we did not finalize yet.
* Importing a finalized block updates the information in Besu Mutable Blockchain.
* It is no longer possible to import blocks into the blockchain below a previously finalized block when using BWS
* When a new finalized block gets announced while BWS is in progress and Besu already has it imported, the Mutable Blockchain gets updated, and the chain gets checked for possible pruning
* When trying to import blocks of equal height as newly announced finalized block, then only the new announced finalized block will be possible to import using BWS
* When importing a new block using BWS after finalized, we can now guarantee that the block descends from latest finalized block
Signed-off-by: Jiri Peinlich <jiri.peinlich@gmail.com>
* BWS now has explicit parametrized check for how deep to sync
Signed-off-by: Jiri Peinlich <jiri.peinlich@gmail.com>
* addressing review comments
Signed-off-by: Jiri Peinlich <jiri.peinlich@gmail.com>
* Addressing review comment
Signed-off-by: Jiri Peinlich <jiri.peinlich@gmail.com>
Co-authored-by: Fabio Di Fabio <fabio.difabio@consensys.net>
* check that the timestamp in fcu payload attributes is greater than the one of the head block
Signed-off-by: Daniel Lehrner <daniel.lehrner@consensys.net>
* rearranged some things so EthProtocolManager can get notified of ttd
* adds some needed state and locking
* listens to MergeContext
* tests for preventing pow peers from joining
* disconnecting block senders after ttd
* wait till finalizing 2 blocks before locking out pow nodes
* hacking around clique networks, will need to be fixed still for Goerli
Signed-off-by: Justin Florentine <justin+github@florentine.us>
* do not disconnect a peer on re-pivot race
Signed-off-by: garyschulte <garyschulte@gmail.com>
Co-authored-by: Sally MacFarlane <sally.macfarlane@consensys.net>
Add a new way to synchronize which is X_CHECKPOINT. This mode is experimental so use it at your own risk. This mode allows you to do like a snapsync but starting from a specific checkpoint instead of starting from the genesis.
This checkpoint will be in the genesis configuration of each network. To add the checkpoint mechanism in a network you just have to add the checkpoint section in the genesis.
Currently there is a checkpoint for ropten, goerli and mainnet.
Mainnet on i3.2xlarge <6 hours
Goerli on i3.2xlarge <1 hours
Signed-off-by: Karim TAAM <karim.t2am@gmail.com>
* Remove deprecated cleanup task
* Move util class into test-support
This is because this class can be used outside the scope of the junit tests but it'll not be part of main
* Move referenceTests into its own scope
* Migrate referenceTests to Junit5
I inlined `o.h.b.e.vm.AbstractRetryingTest` into `o.h.b.e.vm.VMReferenceTest` because there was a single usage of the abstraction and also because it wasn't portable to how Junit5 parameters work
Signed-off-by: Diego López León <dieguitoll@gmail.com>
Co-authored-by: Danno Ferrin <danno.ferrin@gmail.com>
Co-authored-by: Sally MacFarlane <sally.macfarlane@consensys.net>
* JSON-RPC: Besu does not follow the standard API for eth_getTransactionByHash see#3868
Signed-off-by: Sharad Gulati <sharad.develop@gmail.com>
Co-authored-by: Justin Florentine <justin+github@florentine.us>
Co-authored-by: garyschulte <garyschulte@gmail.com>
Co-authored-by: Sally MacFarlane <sally.macfarlane@consensys.net>
* added acceptance tests
Signed-off-by: Daniel Lehrner <daniel.lehrner@consensys.net>
* updated EngineForkchoiceUpdated to return INVALID_FORKCHOICE_STATE when finalized block hash or safe block hash are unknown
Signed-off-by: Daniel Lehrner <daniel.lehrner@consensys.net>
* fixed tests and added new ones for invalid forkchoice state and saving safe block
Signed-off-by: Daniel Lehrner <daniel.lehrner@consensys.net>
* Upgrade Errorprone
Upgrade errorprone to 2.13.1. Like all errorprone upgrades there are
new checks requiring code changes.
* Unused methods now cause compilation errors
* fields must be static and final
* Effectively constant booleans must now be returned as true/false.
* longs should not auto-cast to double.
* turn off errorprone javadocs
Signed-off-by: Danno Ferrin <danno.ferrin@gmail.com>
Co-authored-by: Sally MacFarlane <sally.macfarlane@consensys.net>
10.4 revision of reference tests has notable changes
* Nonce can be up to 2^64-1, with some opcode and validity interactions specced in eip-2681
* Wei fields can be up to 2^256, tests check for rollover
* VM Tests were removed
* Legacy Tests were removed
* Reference tests make it clear that transactions with a maxPriorityFee of zero are valid. Perhaps useless, but still valid.
* Adding validation hooks in the out-of-chain test execution. These validations are caught in full-chain execution, just not in conveniently places for integration testing.
* This does not transaction tests support for the "merge" fork, as the release notes marked it as an "upcoming" feature.
Signed-off-by: Danno Ferrin <danno.ferrin@gmail.com>
Signed-off-by: Daniel Lehrner <daniel.lehrner@consensys.net>
Co-authored-by: Diego López León <dieguitoll@gmail.com>
Co-authored-by: Daniel Lehrner <daniel.lehrner@consensys.net>
* Introduces new JsonRPCService that handles both http and websockets
* Removes websocket specific configs
* renames WebSocketRequestHandler to WebSocketMessageHandler to be more in line with websocket semantics.
Signed-off-by: Fabio Di Fabio <fabio.difabio@consensys.net>
Signed-off-by: Justin Florentine <justin+github@florentine.us>
Co-authored-by: Fabio Di Fabio <fabio.difabio@consensys.net>
* Reduce EVM library dependencies
Reduce the runtime dependencies of the EVM module by moving a
class used by a single class out of util to the owned module (QosTimer)
and them removing un-used dependencies from EVM and dependant modules.
Signed-off-by: Danno Ferrin <danno.ferrin@gmail.com>
* Fix two operand stack tests
Signed-off-by: Diego López León <dieguitoll@gmail.com>
* Remove unreachable assert
Signed-off-by: Diego López León <dieguitoll@gmail.com>
* Replace expected annotation field with assert
Signed-off-by: Diego López León <dieguitoll@gmail.com>
* Extract variables to let only a single call for assert
Signed-off-by: Diego López León <dieguitoll@gmail.com>
Co-authored-by: Sally MacFarlane <sally.macfarlane@consensys.net>
* added test file to match any topic
* add singleton list containing null if empty topic passed in
Signed-off-by: Sally MacFarlane <sally.macfarlane@consensys.net>
* require 3.1.15 of jnr-posix
* cover more bases with epl license
Signed-off-by: Sally MacFarlane <sally.macfarlane@consensys.net>
Co-authored-by: Stefan Pingel <16143240+pinges@users.noreply.github.com>
* avoid the edge case of long minimum value which gives a negative value when passed into abs()
* changed some Longs to long where they were statically intialized
* replace Lists.emptyList() with Collections.emptyList()
* removed unnecessary eq() invocation and resulting code smell
Signed-off-by: Sally MacFarlane <sally.macfarlane@consensys.net>
Daniel Lehrner
Gabriel Camargo Fukushima
Danno Ferrin
mbaxter
matkt
Jiri Peinlich
Fabio Di Fabio
Justin Florentine
garyschulte
Diego López León
Lucas Saldanha
sharad-develop
Sally MacFarlane