This resolves a minor issue where neighbors responses containing less
than 16 nodes would bump the failure counter, removing the node. One
situation where this can happen is a private deployment where the total
number of extant nodes is less than 16.
Issue found by @jsying.
This change clears up confusion around the two ways in which nodes
can be added to the table.
When a neighbors packet is received as a reply to findnode, the nodes
contained in the reply are added as 'seen' entries if sufficient space
is available.
When a ping is received and the endpoint verification has taken place,
the remote node is added as a 'verified' entry or moved to the front of
the bucket if present. This also updates the node's IP address and port
if they have changed.
This change resolves multiple issues around handling of endpoint proofs.
The proof is now done separately for each IP and completing the proof
requires a matching ping hash.
Also remove waitping because it's equivalent to sleep. waitping was
slightly more efficient, but that may cause issues with findnode if
packets are reordered and the remote end sees findnode before pong.
Logging of received packets was hitherto done after handling the packet,
which meant that sent replies were logged before the packet that
generated them. This change splits up packet handling into 'preverify'
and 'handle'. The error from 'preverify' is logged, but 'handle' happens
after the message is logged. This fixes the order. Packet logs now
contain the node ID.
dput --passive should make repo pushes from Travis work again.
dput --no-upload-log works around an issue I had while uploading locally.
debuild -d says that debuild shouldn't check for build dependencies when
creating the source package. This option is needed to make builds work
in environments where the installed Go version doesn't match the
declared dependency in the source package.
* core: speed up GenerateChain
Use a mock implementation of ChainReader instead of creating
and destroying a BlockChain object for each generated block.
* eth/downloader: speed up tests by generating chain only once
This change reworks the downloader tests so they share a common test
blockchain instead of generating a chain in every test. The tests are
roughly twice as fast now.
This PR adds enode.LocalNode and integrates it into the p2p
subsystem. This new object is the keeper of the local node
record. For now, a new version of the record is produced every
time the client restarts. We'll make it smarter to avoid that in
the future.
There are a couple of other changes in this commit: discovery now
waits for all of its goroutines at shutdown and the p2p server
now closes the node database after discovery has shut down. This
fixes a leveldb crash in tests. p2p server startup is faster
because it doesn't need to wait for the external IP query
anymore.
This fixes a rare deadlock with the inproc adapter:
- A node is stopped, which acquires Network.lock.
- The protocol code being simulated (swarm/network in my case)
waits for its goroutines to shut down.
- One of those goroutines calls into the simulation to add a peer,
which waits for Network.lock.
The fix for the deadlock is really simple, just release the lock
before stopping the simulation node.
Other changes in this PR clean up the exec adapter so it reports
node startup errors better and remove the docker adapter because
it just adds overhead.
In the exec adapter, node information is now posted to a one-shot
server. This avoids log parsing and allows reporting startup
errors to the simulation host.
A small change in package node was needed because simulation
nodes use port zero. Node.{HTTP,WS}Endpoint now return the live
endpoints after startup by checking the TCP listener.
Notifier tracks whether subscription are 'active'. A subscription
becomes active when the subscription ID has been sent to the client. If
the client sends notifications in the request handler before the
subscription becomes active they are dropped. The tests tried to work
around this problem by always waiting 5s before sending the first
notification.
Fix it by buffering notifications until the subscription becomes active.
This speeds up all subscription tests.
Also fix TestSubscriptionMultipleNamespaces to wait for three messages
per subscription instead of six. The test now finishes just after all
notifications have been received and doesn't hit the 30s timeout anymore.
* travis: exclude non-test jobs for PRs
We don't usually look at these builders and not starting them
removes ~15min of build time.
* build: don't run vet before tests
Recent versions of Go run vet during 'go test' and we have
a dedicated lint job.
* build: use -timeout 5m for tests
Tests sometimes hang on Travis. CI runs are aborted after 10min with no
output. Adding the timeout means we get to see the stack trace for
timeouts.
Package p2p/enode provides a generalized representation of p2p nodes
which can contain arbitrary information in key/value pairs. It is also
the new home for the node database. The "v4" identity scheme is also
moved here from p2p/enr to remove the dependency on Ethereum crypto from
that package.
Record signature handling is changed significantly. The identity scheme
registry is removed and acceptable schemes must be passed to any method
that needs identity. This means records must now be validated explicitly
after decoding.
The enode API is designed to make signature handling easy and safe: most
APIs around the codebase work with enode.Node, which is a wrapper around
a valid record. Going from enr.Record to enode.Node requires a valid
signature.
* p2p/discover: port to p2p/enode
This ports the discovery code to the new node representation in
p2p/enode. The wire protocol is unchanged, this can be considered a
refactoring change. The Kademlia table can now deal with nodes using an
arbitrary identity scheme. This requires a few incompatible API changes:
- Table.Lookup is not available anymore. It used to take a public key
as argument because v4 protocol requires one. Its replacement is
LookupRandom.
- Table.Resolve takes *enode.Node instead of NodeID. This is also for
v4 protocol compatibility because nodes cannot be looked up by ID
alone.
- Types Node and NodeID are gone. Further commits in the series will be
fixes all over the the codebase to deal with those removals.
* p2p: port to p2p/enode and discovery changes
This adapts package p2p to the changes in p2p/discover. All uses of
discover.Node and discover.NodeID are replaced by their equivalents from
p2p/enode.
New API is added to retrieve the enode.Node instance of a peer. The
behavior of Server.Self with discovery disabled is improved. It now
tries much harder to report a working IP address, falling back to
127.0.0.1 if no suitable address can be determined through other means.
These changes were needed for tests of other packages later in the
series.
* p2p/simulations, p2p/testing: port to p2p/enode
No surprises here, mostly replacements of discover.Node, discover.NodeID
with their new equivalents. The 'interesting' API changes are:
- testing.ProtocolSession tracks complete nodes, not just their IDs.
- adapters.NodeConfig has a new method to create a complete node.
These changes were needed to make swarm tests work.
Note that the NodeID change makes the code incompatible with old
simulation snapshots.
* whisper/whisperv5, whisper/whisperv6: port to p2p/enode
This port was easy because whisper uses []byte for node IDs and
URL strings in the API.
* eth: port to p2p/enode
Again, easy to port because eth uses strings for node IDs and doesn't
care about node information in any way.
* les: port to p2p/enode
Apart from replacing discover.NodeID with enode.ID, most changes are in
the server pool code. It now deals with complete nodes instead
of (Pubkey, IP, Port) triples. The database format is unchanged for now,
but we should probably change it to use the node database later.
* node: port to p2p/enode
This change simply replaces discover.Node and discover.NodeID with their
new equivalents.
* swarm/network: port to p2p/enode
Swarm has its own node address representation, BzzAddr, containing both
an overlay address (the hash of a secp256k1 public key) and an underlay
address (enode:// URL).
There are no changes to the BzzAddr format in this commit, but certain
operations such as creating a BzzAddr from a node ID are now impossible
because node IDs aren't public keys anymore.
Most swarm-related changes in the series remove uses of
NewAddrFromNodeID, replacing it with NewAddr which takes a complete node
as argument. ToOverlayAddr is removed because we can just use the node
ID directly.
* les: fix crasher in NodeInfo when running as server
The ProtocolManager computes CHT and Bloom trie roots by asking the
indexers for their current head. It tried to get the indexers from
LesOdr, but no LesOdr instance is created in server mode.
Attempt to fix this by moving the indexers, protocol creation and
NodeInfo to a new lesCommons struct which is embedded into both server
and client.
All this setup code should really be cleaned up, but this is just a
hotfix so we have to do that some other time.
* les: fix commons protocol maker
* p2p/discover: move bond logic from table to transport
This commit moves node endpoint verification (bonding) from the table to
the UDP transport implementation. Previously, adding a node to the table
entailed pinging the node if needed. With this change, the ping-back
logic is embedded in the packet handler at a lower level.
It is easy to verify that the basic protocol is unchanged: we still
require a valid pong reply from the node before findnode is accepted.
The node database tracked the time of last ping sent to the node and
time of last valid pong received from the node. Node endpoints are
considered verified when a valid pong is received and the time of last
pong was called 'bond time'. The time of last ping sent was unused. In
this commit, the last ping database entry is repurposed to mean last
ping _received_. This entry is now used to track whether the node needs
to be pinged back.
The other big change is how nodes are added to the table. We used to add
nodes in Table.bond, which ran when a remote node pinged us or when we
encountered the node in a neighbors reply. The transport now adds to the
table directly after the endpoint is verified through ping. To ensure
that the Table can't be filled just by pinging the node repeatedly, we
retain the isInitDone check. During init, only nodes from neighbors
replies are added.
* p2p/discover: reduce findnode failure counter on success
* p2p/discover: remove unused parameter of loadSeedNodes
* p2p/discover: improve ping-back check and comments
* p2p/discover: add neighbors reply nodes always, not just during init
This package was meant to hold an improved 256 bit integer library, but
the effort was abandoned in 2015. AFAIK nothing ever used this package.
Time to say goodbye.
ToECDSAPub was unsafe because it returned a non-nil key with nil X, Y in
case of invalid input. This change replaces ToECDSAPub with
UnmarshalPubkey across the codebase.
This applies spec changes from ethereum/EIPs#1049 and adds support for
pluggable identity schemes.
Some care has been taken to make the "v4" scheme standalone. It uses
public APIs only and could be moved out of package enr at any time.
A couple of minor changes were needed to make identity schemes work:
- The sequence number is now updated in Set instead of when signing.
- Record is now copy-safe, i.e. calling Set on a shallow copy doesn't
modify the record it was copied from.
Feed keeps active subscription channels in a slice called 'f.sendCases'.
The Send method tracks the active cases in a local variable 'cases'
whose value is f.sendCases initially. 'cases' shrinks to a shorter
prefix of f.sendCases every time a send succeeds, moving the successful
case out of range of the active case list.
This can be confusing because the two slices share a backing array. Add
more comments to document what is going on. Also add a test for removing
a case that is in 'f.sentCases' but not 'cases'.
The 'from' and 'to' methods on StateTransitions are reader methods and
shouldn't have inadvertent side effects on state.
It is safe to remove the check in 'from' because account existence is
implicitly checked by the nonce and balance checks. If the account has
non-zero balance or nonce, it must exist. Even if the sender account has
nonce zero at the start of the state transition or no balance, the nonce
is incremented before execution and the account will be created at that
time.
It is safe to remove the check in 'to' because the EVM creates the
account if necessary.
Fixes#15119
* common: delete StringToAddress, StringToHash
These functions are confusing because they don't parse hex, but use the
bytes of the string. This change removes them, replacing all uses of
StringToAddress(s) by BytesToAddress([]byte(s)).
* eth/filters: remove incorrect use of common.BytesToAddress
I forgot to change the check in udp.go when I changed Table.bond to be
based on lastPong instead of node presence in db. Rename lastPong to
bondTime and add hasBond so it's clearer what this DB key is used for
now.
Talk about "state" instead of "trie timing", "trie memory" and remove
the overzealous warning when the limit is just reached. Since the time
limit is always reached on slow machines, move the message to info level
so users don't freak out about internal details.
* 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.
* consensus/ethash: add maxEpoch constant
* consensus/ethash: improve cache/dataset handling
There are two fixes in this commit:
Unmap the memory through a finalizer like the libethash wrapper did. The
release logic was incorrect and freed the memory while it was being
used, leading to crashes like in #14495 or #14943.
Track caches and datasets using simplelru instead of reinventing LRU
logic. This should make it easier to see whether it's correct.
* consensus/ethash: restore 'future item' logic in lru
* consensus/ethash: use mmap even in test mode
This makes it possible to shorten the time taken for TestCacheFileEvict.
* consensus/ethash: shuffle func calc*Size comments around
* consensus/ethash: ensure future cache/dataset is in the lru cache
* consensus/ethash: add issue link to the new test
* consensus/ethash: fix vet
* consensus/ethash: fix test
* consensus: tiny issue + nitpick fixes
This commit:
- Adds a --msgfile option to read the message to sign from a file
instead of command line argument.
- Adds a unit test for signing subcommands.
- Removes some weird whitespace in the code.
* core/types, core/vm, eth, tests: regenerate gencodec files
* Makefile: update devtools target
Install protoc-gen-go and print reminders about npm, solc and protoc.
Also switch to github.com/kevinburke/go-bindata because it's more
maintained.
* contracts/ens: update contracts and regenerate with solidity v0.4.19
The newer upstream version of the FIFSRegistrar contract doesn't set the
resolver anymore. The resolver is now deployed separately.
* contracts/release: regenerate with solidity v0.4.19
* contracts/chequebook: fix fallback and regenerate with solidity v0.4.19
The contract didn't have a fallback function, payments would be rejected
when compiled with newer solidity. References to 'mortal' and 'owned'
use the local file system so we can compile without network access.
* p2p/discv5: regenerate with recent stringer
* cmd/faucet: regenerate
* dashboard: regenerate
* eth/tracers: regenerate
* internal/jsre/deps: regenerate
* dashboard: avoid sed -i because it's not portable
* accounts/usbwallet/internal/trezor: fix go generate warnings
* crypto: ensure that VerifySignature rejects malleable signatures
It already rejected them when using libsecp256k1, make sure the nocgo
version does the same thing.
* crypto: simplify check
* crypto: fix build
With this change,
key, err := crypto.HexToECDSA("000000...")
returns nil key and an error instead of a non-nil key with nil X
and Y inside. Issue found by @guidovranken.
We need those operations for p2p/enr.
Also upgrade github.com/btcsuite/btcd/btcec to the latest version
and improve BenchmarkSha3. The benchmark printed extra output
that confused tools like benchstat and ignored N.
* core/vm: track 63/64 call gas off stack
Gas calculations in gasCall* relayed the available gas for calls by
replacing it on the stack. This lead to inconsistent traces, which we
papered over by copying the pre-execution stack in trace mode.
This change relays available gas using a temporary variable, off the
stack, and allows removing the weird copy.
* core/vm: remove stackCopy
* core/vm: pop call gas into pool
* core/vm: to -> addr
* core/types: make Signer derive address instead of public key
There are two reasons to do this now: The upcoming ethclient signer
doesn't know the public key, just the address. EIP 208 will introduce a
new signer which derives the 'entry point' address for transactions with
zero signature. The entry point has no public key.
Other changes to the interface ease the path make to moving signature
crypto out of core/types later.
* ethclient, mobile: add TransactionSender
The new method can get the right signer without any crypto, and without
knowledge of the signature scheme that was used when the transaction was
included.
* ethdb: add Putter interface and Has method
* ethdb: improve docs and add IdealBatchSize
* ethdb: remove memory batch lock
Batches are not safe for concurrent use.
* core: use ethdb.Putter for Write* functions
This covers the easy cases.
* core/state: simplify StateSync
* trie: optimize local node check
* ethdb: add ValueSize to Batch
* core: optimize HasHeader check
This avoids one random database read get the block number. For many uses
of HasHeader, the expectation is that it's actually there. Using Has
avoids a load + decode of the value.
* core: write fast sync block data in batches
Collect writes into batches up to the ideal size instead of issuing many
small, concurrent writes.
* eth/downloader: commit larger state batches
Collect nodes into a batch up to the ideal size instead of committing
whenever a node is received.
* core: optimize HasBlock check
This avoids a random database read to get the number.
* core: use numberCache in HasHeader
numberCache has higher capacity, increasing the odds of finding the
header without a database lookup.
* core: write imported block data using a batch
Restore batch writes of state and add blocks, tx entries, receipts to
the same batch. The change also simplifies the miner.
This commit also removes posting of logs when a forked block is imported.
* core: fix DB write error handling
* ethdb: use RLock for Has
* core: fix HasBlock comment
This fixes a regression where the new Failed field in ReceiptForStorage
rejected previously stored receipts. Fix it by removing the new field
and store status in the PostState field. This also removes massive RLP
hackery around the status field.
The lock file was ineffective because opening leveldb storage in
read-only mode doesn't really take the lock. Fix it by including a
dedicated flock library (which is actually split out of goleveldb).
Blockchain tests now include the "network" which determines the chain
config to use. Remove config matching based on test name and share the
name-to-config index with state tests.
Byzantium/Constantinople tests are still skipped because most of them
fail anyway.
This patch updates the Address type in common/types.go so that the Hex
function provides an EIP55-compliant output string. The implementation is pretty lightweight;
on my laptop the benchmark gives 1100ns/op, with the majority of that value due to the Keccak hash.
Tests are now included as a submodule. This should make updating easier
and removes ~60MB of JSON data from the working copy.
State tests are replaced by General State Tests, which run the same test
with multiple fork configurations.
With the new test runner, consensus tests are run as subtests by walking
json files. Many hex issues have been fixed upstream since the last
update and most custom parsing code is replaced by existing JSON hex
types. Tests can now be marked as 'expected failures', ensuring that
fixes for those tests will trigger an update to test configuration. The
new test runner also supports parallel execution and the -short flag.
With this commit, core/state's access to the underlying key/value database is
mediated through an interface. Database errors are tracked in StateDB and
returned by CommitTo or the new Error method.
Motivation for this change: We can remove the light client's duplicated copy of
core/state. The light client now supports node iteration, so tracing and storage
enumeration can work with the light client (not implemented in this commit).
This adds type and struct field context to error messages.
Instead of "hex string of odd length" users will now see "json: cannot
unmarshal hex string of odd length into Go struct field SendTxArgs.from
of type common.Address".
Felix LangeandPéter Szilágyi
Felix Lange
