When talking to an HTTP2 server, there are situations where it needs to
"rewind" the Request.Body. To allow this, we have to set up the Request.GetBody
function to return a brand new instance of the body.
If not set, we can end up with the following error:
http2: Transport: cannot retry err [http2: Transport received Server's graceful shutdown GOAWAY] after Request.Body was written; define Request.GetBody to avoid this error
See this commit for more information: cffdcf672a
This replaces the sketchy and undocumented string context keys for HTTP requests
with a defined interface. Using string keys with context is discouraged because
they may clash with keys created by other packages.
We added these keys to make connection metadata available in the signer, so this
change also updates signer/core to use the new PeerInfo API.
* all: work for eth1/2 transtition
* consensus/beacon, eth: change beacon difficulty to 0
* eth: updates
* all: add terminalBlockDifficulty config, fix rebasing issues
* eth: implemented merge interop spec
* internal/ethapi: update to v1.0.0.alpha.2
This commit updates the code to the new spec, moving payloadId into
it's own object. It also fixes an issue with finalizing an empty blockhash.
It also properly sets the basefee
* all: sync polishes, other fixes + refactors
* core, eth: correct semantics for LeavePoW, EnterPoS
* core: fixed rebasing artifacts
* core: light: performance improvements
* core: use keyed field (f)
* core: eth: fix compilation issues + tests
* eth/catalyst: dbetter error codes
* all: move Merger to consensus/, remove reliance on it in bc
* all: renamed EnterPoS and LeavePoW to ReachTDD and FinalizePoS
* core: make mergelogs a function
* core: use InsertChain instead of InsertBlock
* les: drop merger from lightchain object
* consensus: add merger
* core: recoverAncestors in catalyst mode
* core: fix nitpick
* all: removed merger from beacon, use TTD, nitpicks
* consensus: eth: add docstring, removed unnecessary code duplication
* consensus/beacon: better comment
* all: easy to fix nitpicks by karalabe
* consensus/beacon: verify known headers to be sure
* core: comments
* core: eth: don't drop peers who advertise blocks, nitpicks
* core: never add beacon blocks to the future queue
* core: fixed nitpicks
* consensus/beacon: simplify IsTTDReached check
* consensus/beacon: correct IsTTDReached check
Co-authored-by: rjl493456442 <garyrong0905@gmail.com>
Co-authored-by: Péter Szilágyi <peterke@gmail.com>
This avoids quadratic time complexity in the lookup of the batch element
corresponding to an RPC response. Unfortunately, the new approach
requires additional memory for the mapping from ID to index.
Fixes#22805
* core: fix warning flagging the use of DeepEqual on error
* apply the same change everywhere possible
* revert change that was committed by mistake
* fix build error
* Update config.go
* revert changes to ConfigCompatError
* review feedback
Co-authored-by: Felix Lange <fjl@twurst.com>
Currently rpc.BlockNumber is marshalled to JSON as a numeric value, which is
wrong because BlockNumber.UnmarshalJSON() wants it to either be hex-encoded
or string "earliest"/"latest"/"pending". As a result, the call chain
rpc.BlockNumberOrHashWithNumber(123) -> json.Marshal() -> json.Unmarshal()
fails with error "cannot unmarshal object into Go value of type string".
The new error type is returned by client operations contains details of
the response error code and response body.
Co-authored-by: Felix Lange <fjl@twurst.com>
This fixes a rare issue where the client subscription forwarding loop
would attempt send on the subscription's channel after Unsubscribe has
returned, leading to a panic if the subscription channel was already
closed by the user. Example:
sub, _ := client.Subscribe(..., channel, ...)
sub.Unsubscribe()
close(channel)
The race occurred because Unsubscribe called quitWithServer to tell the
forwarding loop to stop sending on sub.channel, but did not wait for the
loop to actually come down. This is fixed by adding an additional channel
to track the shutdown, on which Unsubscribe now waits.
Fixes#22322
* trie: fix tests to work on 32-bit systems
* les: make test work on 32-bit platform
* cmd/geth: fix windows-issues on tests
* trie: improve balance
* cmd/geth: make account tests less verbose + less mem intense
* rpc: make debug-level log output less verbose
* cmd/geth: lint
* Only compare hostnames in ws.origins
Also using a helper function for ToLower consolidates all preparation steps in one function for more maintainable consistency.
Spaces => tabs
Remove a semicolon
Add space at start of comment
Remove parens around conditional
Handle case wehre parsed hostname is empty
When passing a single word like "localhost" the parsed hostname is an empty string. Handle this and the error-parsing case together as default, and the nonempty hostname case in the conditional.
Refactor with new originIsAllowed functions
Adds originIsAllowed() & ruleAllowsOrigin(); removes prepOriginForComparison
Remove blank line
Added tests for simple allowed-orign rule
which does not specify a protocol or port, just a hostname
Fix copy-paste: `:=` => `=`
Remove parens around conditional
Remove autoadded whitespace on blank lines
Compare scheme, hostname, and port with rule
if the rule specifies those portions.
Remove one autoadded trailing whitespace
Better handle case where only origin host is given
e.g. "localhost"
Remove parens around conditional
Refactor: attemptWebsocketConnectionFromOrigin DRY
Include return type on helper function
Provide srv obj in helper fn
Provide srv to helper fn
Remove stray underscore
Remove blank line
parent 93e666b4c1e7e49b8406dc83ed93f4a02ea49ac1
author wbt <wbt@users.noreply.github.com> 1598559718 -0400
committer Martin Holst Swende <martin@swende.se> 1605602257 +0100
gpgsig -----BEGIN PGP SIGNATURE-----
iQFFBAABCAAvFiEEypmrtbNuJK1doP1AaDtDjAWl3fAFAl+zi9ARHG1hcnRpbkBz
d2VuZGUuc2UACgkQaDtDjAWl3fDRiwgAoMtzU8dwRV7Q9xkCwWEx9Wz2f3n6jUr2
VWBycDKGKwRkPPOER3oc9kzjGU/P1tFlK07PjfnAKZ9KWzxpDcJZwYM3xCBurG7A
16y4YsQnzgPNONv3xIkdi3RZtDBIiPFFEmdZFFvZ/jKexfI6JIYPngCAoqdTIFb9
On/aPvvVWQn1ExfmarsvvJ7kUDUG77tZipuacEH5FfFsfelBWOEYPe+I9ToUHskv
+qO6rOkV1Ojk8eBc6o0R1PnApwCAlEhJs7aM/SEOg4B4ZJJneiFuEXBIG9+0yS2I
NOicuDPLGucOB5nBsfIKI3USPeE+3jxdT8go2lN5Nrhm6MimoILDsQ==
=sgUp
-----END PGP SIGNATURE-----
Refactor: drop err var for more concise test lines
Add several tests for new WebSocket origin checks
Remove autoadded whitespace on blank lines
Restore TestWebsocketOrigins originally-named test
and rename the others to be helpers rather than full tests
Remove autoadded whitespace on blank line
Temporarily comment out new test sets
Uncomment test around origin rule with scheme
Remove tests without scheme on browser origin
per https://github.com/ethereum/go-ethereum/pull/21481/files#r479371498
Uncomment tests with port; remove some blank lines
Handle when browser does not specify scheme/port
Uncomment test for including scheme & port in rule
Add IP tests
* node: more tests + table-driven, ws origin changes
Co-authored-by: Martin Holst Swende <martin@swende.se>
* internal/ethapi: return revert reason for eth_call
* internal/ethapi: moved revert reason logic to doCall
* accounts/abi/bind/backends: added revert reason logic to simulated backend
* internal/ethapi: fixed linting error
* internal/ethapi: check if require reason can be unpacked
* internal/ethapi: better error logic
* internal/ethapi: simplify logic
* internal/ethapi: return vmError()
* internal/ethapi: move handling of revert out of docall
* graphql: removed revert logic until spec change
* rpc: internal/ethapi: added custom error types
* graphql: use returndata instead of return
Return() checks if there is an error. If an error is found, we return nil.
For most use cases it can be beneficial to return the output even if there
was an error. This code should be changed anyway once the spec supports
error reasons in graphql responses
* accounts/abi/bind/backends: added tests for revert reason
* internal/ethapi: add errorCode to revert error
* internal/ethapi: add errorCode of 3 to revertError
* internal/ethapi: unified estimateGasErrors, simplified logic
* internal/ethapi: unified handling of errors in DoEstimateGas
* rpc: print error data field
* accounts/abi/bind/backends: unify simulatedBackend and RPC
* internal/ethapi: added binary data to revertError data
* internal/ethapi: refactored unpacking logic into newRevertError
* accounts/abi/bind/backends: fix EstimateGas
* accounts, console, internal, rpc: minor error interface cleanups
* Revert "accounts, console, internal, rpc: minor error interface cleanups"
This reverts commit 2d3ef53c5304e429a04983210a417c1f4e0dafb7.
* re-apply the good parts of 2d3ef53c53
* rpc: add test for returning server error data from client
Co-authored-by: rjl493456442 <garyrong0905@gmail.com>
Co-authored-by: Péter Szilágyi <peterke@gmail.com>
Co-authored-by: Felix Lange <fjl@twurst.com>
This corrects the call to eth_getBlockByNumber, which previously
returned this error:
can't get latest block: missing value for required argument 1
Co-authored-by: Felix Lange <fjl@twurst.com>
This change makes it possible to run geth with JSON-RPC over HTTP and
WebSocket on the same TCP port. The default port for WebSocket
is still 8546.
geth --rpc --rpcport 8545 --ws --wsport 8545
This also removes a lot of deprecated API surface from package rpc.
The rpc package is now purely about serving JSON-RPC and no longer
provides a way to start an HTTP server.
The leaks were mostly in unit tests, and could all be resolved by
adding suitably-sized channel buffers or by restructuring the test
to not send on a channel after an error has occurred.
There is an unavoidable goroutine leak in Console.Interactive: when
we receive a signal, the line reader cannot be unblocked and will get
stuck. This leak is now documented and I've tried to make it slightly
less bad by adding a one-element buffer to the output channels of
the line-reading loop. Should the reader eventually awake from its
blocked state (i.e. when stdin is closed), at least it won't get stuck
trying to send to the interpreter loop which has quit long ago.
Co-authored-by: Felix Lange <fjl@twurst.com>
This adds a couple of metrics for tracking the timing
and frequency of method calls:
- rpc/requests gauge counts all requests
- rpc/success gauge counts requests which return err == nil
- rpc/failure gauge counts requests which return err != nil
- rpc/duration/all timer tracks timing of all requests
- rpc/duration/<method>/<success/failure> tracks per-method timing
This just prevents a false negative ERROR warning when, for some unknown
reason, a user attempts to turn on the module rpc even though it's already going
to be on.
