* node: allow replacing existing p2p.Reactor(s)
using [`CustomReactors`
option](https://godoc.org/github.com/tendermint/tendermint/node#CustomReactors).
Warning: beware of accidental name clashes. Here is the list of existing
reactors: MEMPOOL, BLOCKCHAIN, CONSENSUS, EVIDENCE, PEX.
* check the absence of "CUSTOM" prefix
* merge 2 tests
* add doc.go to node package
* Do not write 'Couldn't connect to any seeds' if there are no seeds
* changelog
* remove privValUpgrade
* Fix typo in changelog
* Update CHANGELOG_PENDING.md
Co-Authored-By: Marko <marbar3778@yahoo.com>
I'm setting up all peers dynamically by calling dial_peers, so p2p.seeds in configs is empty, and I'm seeing error log a lot in logs.
* p2p: fix false-positive error logging when stopping connections
This changeset fixes two types of false-positive errors occurring during
connection shutdown.
The first occurs when the process invokes FlushStop() or Stop() on a
connection. While the previous behavior did properly wait for the sendRoutine
to finish, it did not notify the recvRoutine that the connection was shutting
down. This would cause the recvRouting to receive and error when reading and
log this error. The changeset fixes this by notifying the recvRoutine that
the connection is shutting down.
The second occurs when the connection is terminated (gracefully) by the other side.
The recvRoutine would get an EOF error during the read, log it, and stop the connection
with an error. The changeset detects EOF and gracefully shuts down the connection.
* bring back the comment about flushing
* add changelog entry
* listen for quitRecvRoutine too
* we have to call stopForError
Otherwise peer won't be removed from the peer set and maybe readded
later.
cleanup to add linter
grpc change:
https://godoc.org/google.golang.org/grpc#WithContextDialerhttps://godoc.org/google.golang.org/grpc#WithDialer
grpc/grpc-go#2627
prometheous change:
due to UninstrumentedHandler, being deprecated in the future
empty branch = empty if or else statement
didn't delete them entirely but commented
couldn't find a reason to have them
could not replicate the issue #3406
but if want to keep it commented then we should comment out the if statement as well
* Renamed wire.go to codec.go
- Wire was the previous name of amino
- Codec describes the file better than `wire` & `amino`
Signed-off-by: Marko Baricevic <marbar3778@yahoo.com>
* ide error
* rename amino.go to codec.go
* Remove db from tendemrint in favor of tendermint/tm-cmn
- remove db from `libs`
- update dependancy, there have been no breaking changes in the updated deps
- https://github.com/grpc/grpc-go/releases
- https://github.com/golang/protobuf/releases
Signed-off-by: Marko Baricevic <marbar3778@yahoo.com>
* changelog add
* gofmt
* more gofmt
* p2p: extract ID validation into a separate func
- NewNetAddress panics if ID is invalid
- NetAddress#Valid returns an error
- remove ErrAddrBookInvalidAddrNoID
Fixes#2722
* p2p: remove repetitive check in ReceiveAddrs
* fix netaddress test
* change invocation of NewNode across
* custom reactor name are prefixed with CUSTOM_
* upgate changelog pending
* improve comments
* node: refactor NewNode to use functional options
Calling ensurePeers outside of ensurePeersRoutine can lead to nodes
disconnecting from us due to "sent next PEX request too soon" error.
Solution is to just dial addrs we got from src instead of calling
ensurePeers.
Refs #2093Fixes#3338
* Move peer behaviour into it's own package
* refactor wip
* Adjust API and fix tests
* remove unused test struct
* Better error message
* Restructure:
+ Now behaviour is it's own package, we don't need to include
PeerBehaviour in every type.
+ Split up behaviours and reporters into seperate files
* doc string fixes
* Fix minor typos
* Update behaviour/reporter.go
Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com>
* Update behaviour/reporter.go
Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com>
Fixes#3521
The function NewNetAddressStringWithOptionalID is from a time when peer
IDs were optional. They're not anymore. So this should be renamed to
NewNetAddressString and should ensure the ID is provided.
* update changelog
* use NewNetAddress in transport tests
* use NewNetAddress in TestTransportMultiplexAcceptMultiple
* Peer behaviour test tweaks:
Address the remaining test issues mentioned in ##3552 notably:
+ switch to p2p_test package
+ Use `Error` instead of `Errorf` when not using formatting
+ Add expected/got errors to
`TestMockPeerBehaviourReporterConcurrency` test
* Peer behaviour equal behaviours test
+ slices of PeerBehaviours should be compared as histograms to
ensure they have the same set of PeerBehaviours at the same
freequncy.
* TestEqualPeerBehaviours:
+ Add tests for the equivalence between sets of PeerBehaviours
* fix peer state init to late
Peer does not have a state yet. We set it in AddPeer.
We need an new interface before mconnection is started
* pex message to soon
fix reconnection pex send too fast,
error is caused lastReceivedRequests is still
not deleted when a peer reconnected
* add test case for initpeer
* add prove case
* remove potentially infinite loop
* Update consensus/reactor.go
Co-Authored-By: guagualvcha <baifudong@lancai.cn>
* Update consensus/reactor_test.go
Co-Authored-By: guagualvcha <baifudong@lancai.cn>
* document Reactor interface better
* refactor TestReactorReceiveDoesNotPanicIfAddPeerHasntBeenCalledYet
* fix merge conflicts
* blockchain: remove peer's ID from the pool in InitPeer
Refs #3338
* pex: resetPeersRequestsInfo both upon InitPeer and RemovePeer
* ensure RemovePeer is always called before InitPeer
by removing the peer from the switch last (after we've stopped it and
removed from all reactors)
* add some comments for ConsensusReactor#InitPeer
* fix pex reactor
* format code
* fix spelling
* update changelog
* remove unused methods
* do not clear lastReceivedRequests upon error
only in RemovePeer
* call InitPeer before we start the peer!
* add a comment to InitPeer
* write a test
* use waitUntilSwitchHasAtLeastNPeers func
* bring back timeouts
* Test to ensure Receive panics if InitPeer has not been called
* p2p/pex: consult seeds in crawlPeersRoutine
This changeset alters the startup behavior for crawlPeersRoutine. Previously
the routine would crawl a random selection of peers on startup. For a
new seed node, there are no peers. As a result, new seed nodes are unable
to bootstrap themselves with a list of peers until another node with a list
of peers connects to the seed. If this node relies on the seed node for peers,
then the two will not discover more peers.
This changeset makes the startup behavior for crawlPeersRoutine connect to
any seed nodes. Upon connecting, a request for peers will be sent to the seed node
thus helping bootstrap our seed node.
* p2p/pex: Adjust error message for no peers
Co-Authored-By: Ethan Buchman <ethan@coinculture.info>
* p2p: initial implementation of peer behaviour
* [p2p] re-use newMockPeer
* p2p: add inline docs for peer_behaviour interface
* [p2p] Align PeerBehaviour interface (#3558)
* [p2p] make switchedPeerHebaviour private
* [p2p] make storePeerHebaviour private
* [p2p] Add CHANGELOG_PENDING entry for PeerBehaviour
* [p2p] Adjustment naming for PeerBehaviour
* [p2p] Add coarse lock around storedPeerBehaviour
* [p2p]: Fix non-pointer methods in storedPeerBehaviour
+ Structs with embeded locks must specify all methods with pointer
receivers to avoid creating a copy of the embeded lock.
* [p2p] Thorough refactoring based on comments in #3552
+ Decouple PeerBehaviour interface from Peer by parametrizing
methods with `p2p.ID` instead of `p2p.Peer`
+ Setter methods wrapped in a write lock
+ Getter methods wrapped in a read lock
+ Getter methods on storedPeerBehaviour now take a `p2p.ID`
+ Getter methods return a copy of underlying stored behaviours
+ Add doc strings to public types and methods
* [p2p] make structs public
* [p2p] Test empty StoredPeerBehaviour
* [p2p] typo fix
* [p2p] add TestStoredPeerBehaviourConcurrency
+ Add a test which uses StoredPeerBehaviour in multiple goroutines
to ensure thread-safety.
* Update p2p/peer_behaviour.go
Co-Authored-By: brapse <brapse@gmail.com>
* Update p2p/peer_behaviour.go
Co-Authored-By: brapse <brapse@gmail.com>
* Update p2p/peer_behaviour.go
Co-Authored-By: brapse <brapse@gmail.com>
* Update p2p/peer_behaviour.go
Co-Authored-By: brapse <brapse@gmail.com>
* Update p2p/peer_behaviour.go
Co-Authored-By: brapse <brapse@gmail.com>
* Update p2p/peer_behaviour_test.go
Co-Authored-By: brapse <brapse@gmail.com>
* Update p2p/peer_behaviour.go
Co-Authored-By: brapse <brapse@gmail.com>
* Update p2p/peer_behaviour.go
Co-Authored-By: brapse <brapse@gmail.com>
* Update p2p/peer_behaviour.go
Co-Authored-By: brapse <brapse@gmail.com>
* Update p2p/peer_behaviour.go
Co-Authored-By: brapse <brapse@gmail.com>
* [p2p] field ordering convention
* p2p: peer behaviour refactor
+ Change naming of reporting behaviour to `Report`
+ Remove the responsibility of distinguishing between the categories
of good and bad behaviour and instead focus on reporting behaviour.
* p2p: rename PeerReporter -> PeerBehaviourReporter
## Description
also
handle errors from DialPeersAsync
remove nil addr from log msg
fix TestPEXReactorDoesNotDisconnectFromPersistentPeerInSeedMode
This is a follow-up from
#3593 (review)
Fixes most of the #3617, except #3593 (comment)
## Commits
* rpc: /dial_peers: only mark peers as persistent if flag is on
also
- handle errors from DialPeersAsync
- remove nil addr from log msg
- fix TestPEXReactorDoesNotDisconnectFromPersistentPeerInSeedMode
This is a follow-up from
https://github.com/tendermint/tendermint/pull/3593#pullrequestreview-233556909
* remove a call to AddPersistentPeers
TestDialFail will trigger a reconnect
## Description
Previously only outbound peers can be persistent.
Now, even if the peer is inbound, if it's marked as persistent, when/if conn is lost,
Tendermint will try to reconnect. This part is actually optional and can be reverted.
Plus, seed won't disconnect from inbound peer if it's marked as
persistent. Fixes#3362
## Commits
* make persistent prop independent of conn direction
Previously only outbound peers can be persistent. Now, even if the peer
is inbound, if it's marked as persistent, when/if conn is lost,
Tendermint will try to reconnect.
Plus, seed won't disconnect from inbound peer if it's marked as
persistent. Fixes#3362
* fix TestPEXReactorDialPeer test
* add a changelog entry
* update changelog
* add two tests
* reformat code
* test UnsafeDialPeers and UnsafeDialSeeds
* add TestSwitchDialPeersAsync
* spec: update p2p/config spec
* fixes after Ismail's review
* Apply suggestions from code review
Co-Authored-By: melekes <anton.kalyaev@gmail.com>
* fix merge conflict
* remove sleep from TestPEXReactorDoesNotDisconnectFromPersistentPeerInSeedMode
We don't need it actually.
If we are low on addresses for peering, we need to discover more peers. The
previous behavior would query existing peers; however, if an existing peer
does not participate in peer exchange, then our node will not discover more peers.
This change consults both existing peers as well as seeds when there is a deficit
in address book addresses. This allows for discovering peers though existing channels
as well as via seeds if existing peers do not share addresses.
* Remove deprecated PanicXXX functions from codebase
As per discussion over
[here](https://github.com/tendermint/tendermint/pull/3456#discussion_r278423492),
we need to remove these `PanicXXX` functions and eliminate our
dependence on them. In this PR, each and every `PanicXXX` function call
is replaced with a simple `panic` call.
* add a changelog entry
* use dialPeer function in a seed mode
Fixes#3532
by storing a number of attempts we've tried to connect in-memory and
removing the address from addrbook when number of attempts > 16
A prior change to address accidental DNS lookups introduced the
SocketAddr on peer, which was then used to add it to the addressbook.
Which in turn swallowed the self reported port of the peer, which is
important on a reconnect. This change revives the NetAddress on NodeInfo
which the Peer carries, but now returns an error to avoid nil
dereferencing another issue observed in the past. Additionally we could
potentially address #3532, yet the original problem statemenf of that
issue stands.
As a drive-by optimisation `MarkAsGood` now takes only a `p2p.ID` which
makes it interface a bit stricter and leaner.
* add actionable advice for ErrAddrBookNonRoutable err
Should replace https://github.com/tendermint/tendermint/pull/3463
* reorder checks in addrbook#addAddress so
ErrAddrBookPrivate is returned first
and do not log error in DialPeersAsync if the address is private
because it's not an error
ListOfKnownAddresses is removed
panic if addrbook size is less than zero
CrawlPeers does not attempt to connect to existing or peers we're currently dialing
various perf. fixes
improved tests (though not complete)
move IsDialingOrExistingAddress check into DialPeerWithAddress (Fixes#2716)
* addrbook: preallocate memory when saving addrbook to file
* addrbook: remove oldestFirst struct and check for ID
* oldestFirst replaced with sort.Slice
* ID is now mandatory, so no need to check
* addrbook: remove ListOfKnownAddresses
GetSelection is used instead in seed mode.
* addrbook: panic if size is less than 0
* rewrite addrbook#saveToFile to not use a counter
* test AttemptDisconnects func
* move IsDialingOrExistingAddress check into DialPeerWithAddress
* save and cleanup crawl peer data
* get rid of DefaultSeedDisconnectWaitPeriod
* make linter happy
* fix TestPEXReactorSeedMode
* fix comment
* add a changelog entry
* Apply suggestions from code review
Co-Authored-By: melekes <anton.kalyaev@gmail.com>
* rename ErrDialingOrExistingAddress to ErrCurrentlyDialingOrExistingAddress
* lowercase errors
* do not persist seed data
pros:
- no extra files
- less IO
cons:
- if the node crashes, seed might crawl a peer too soon
* fixes after Ethan's review
* add a changelog entry
* we should only consult Switch about peers
checking addrbook size does not make sense since only PEX reactor uses
it for dialing peers!
https://github.com/tendermint/tendermint/pull/3011#discussion_r270948875
* OriginalAddr -> SocketAddr
OriginalAddr records the originally dialed address for outbound peers,
rather than the peer's self reported address. For inbound peers, it was
nil. Here, we rename it to SocketAddr and for inbound peers, set it to
the RemoteAddr of the connection.
* use SocketAddr
Numerous places in the code call peer.NodeInfo().NetAddress().
However, this call to NetAddress() may perform a DNS lookup if the
reported NodeInfo.ListenAddr includes a name. Failure of this lookup
returns a nil address, which can lead to panics in the code.
Instead, call peer.SocketAddr() to return the static address of the
connection.
* remove nodeInfo.NetAddress()
Expose `transport.NetAddress()`, a static result determined
when the transport is created. Removing NetAddress() from the nodeInfo
prevents accidental DNS lookups.
* fixes from review
* linter
* fixes from review
* mempool: add a safety check, write tests for mempoolIDs
and document 65536 limit in the mempool reactor spec
follow-up to https://github.com/tendermint/tendermint/pull/2778
* rename the test
* fixes after Ismail's review
Why submit this pr:
we have suffered from infinite loop in addrbook bug which takes us a long time to find out why process become a zombie peer. It have been fixed in #3232. But the ADDRS_LOOP is still there, risk of infinite loop is still exist.
The algorithm that to random pick a bucket is not stable, which means the peer may unluckily always choose the wrong bucket for a long time, the time and cpu cost is meaningless.
A simple improvement:
shuffle bucketsNew and bucketsOld, and pick necessary number of address from them. A stable
algorithm.
* p2p: refactor Switch#Broadcast func
- call wg.Add only once
- do not call peers.List twice!
* bad for perfomance
* peers list can change in between calls!
Refs #3306
* p2p: use time.Ticker instead of RepeatTimer
no need in RepeatTimer since we don't Reset them
Refs #3306
* libs/common: remove RepeatTimer (also TimerMaker and Ticker interface)
"ancient code that’s caused no end of trouble" Ethan
I believe there's much simplier way to write a ticker than can be reset
https://medium.com/@arpith/resetting-a-ticker-in-go-63858a2c17ec
Refs #3262
This fixes two small bugs:
1) lite/dbprovider: return `ok` instead of true in parse* functions. It's weird that we're ignoring `ok` value before.
2) consensus/state: previously because of the shadowing we almost never output "Error with msg". Now we declare both `added` and `err` in the beginning of the function, so there's no shadowing.
* fix dirty data in peerset
startInitPeer before PeerSet add the peer,
once mconnection start and Receive of one
Reactor faild, will try to remove it from PeerSet
while PeerSet still not contain the peer. Fix
this by change the order.
* fix test FilterDuplicate
* fix start/stop race
* fix err
* reject the shared secret if is all zeros in case the blacklist was not
sufficient
* Add test that verifies lower order pub-keys are rejected at the DH step
* Update changelog
* fix typo in test-comment
ref: [#3010 (comment)](https://github.com/tendermint/tendermint/issues/3010#issuecomment-464287627)
> I tried searching for code where we authenticate a peer against its NetAddress.ID and couldn't find it. I don't see a reason to switch to Noise, but a need to ensure that the node's ID is authenticated e.g. after dialing from the address book.
* p2p: check secret conn id matches dialed id
* Fix all p2p tests & make code compile
* add simple test for dialing with wrong ID
* update changelog
* address review comments
* yet another place where to use IDAddressString and fix
testSetupMultiplexTransport
* failing test
* fix infinite loop in addrbook
There are cases where we only have a small number of addresses marked
good ("old"), but the selection mechanism keeps trying to select more of these
addresses, and hence ends up in an infinite loop. Here we fix this to
only try and select such "old" addresses if we have enough of them. Note this
means, if we don't have enough of them, we may return more "new"
addresses than otherwise expected by the newSelectionBias.
This whole GetSelectionWithBias method probably needs to be rewritten,
but this is a quick fix for the issue.
* changelog
* fix infinite loop if not enough new addrs
* fix another potential infinite loop
if a.nNew == 0 -> pickFromOldBucket=true, but we don't have enough items
(a.nOld > len(oldBucketToAddrsMap) false)
* Revert "fix another potential infinite loop"
This reverts commit 146540c112.
* check num addresses instead of buckets, new test
* fixed the int division
* add slack to bias % in test, lint fixes
* Added checks for selection content in test
* test cleanup
* Apply suggestions from code review
Co-Authored-By: ebuchman <ethan@coinculture.info>
* address review comments
* change after Anton's review comments
* use the same docker image we use for testing
when building a binary for localnet
* switch back to circleci classic
* more review comments
* more review comments
* refactor addrbook_test
* build linux binary inside docker
in attempt to fix
```
--> Running dep
+ make build-linux
GOOS=linux GOARCH=amd64 make build
make[1]: Entering directory `/home/circleci/.go_workspace/src/github.com/tendermint/tendermint'
CGO_ENABLED=0 go build -ldflags "-X github.com/tendermint/tendermint/version.GitCommit=`git rev-parse --short=8 HEAD`" -tags 'tendermint' -o build/tendermint ./cmd/tendermint/
p2p/pex/addrbook.go:373:13: undefined: math.Round
```
* change dir from /usr to /go
* use concrete Go version for localnet binary
* check for nil addresses just to be sure
* p2p/conn: fix deadlock in FlushStop/OnStop
* makefile: set_with_deadlock
* close doneSendRoutine at end of sendRoutine
* conn: initialize channs in OnStart
* node: decrease retry conn timeout in test
Should fix#3256
The retry timeout was set to the default, which is the same as the
accept timeout, so it's no wonder this would fail. Here we decrease the
retry timeout so we can try many times before the accept timeout.
* p2p: increase handshake timeout in test
This fails sometimes, presumably because the handshake timeout is so low
(only 50ms). So increase it to 1s. Should fix#3187
* privval: fix race with ping. closes#3237
Pings happen in a go-routine and can happen concurrently with other
messages. Since we use a request/response protocol, we expect to send a
request and get back the corresponding response. But with pings
happening concurrently, this assumption could be violated. We were using
a mutex, but only a RWMutex, where the RLock was being held for sending
messages - this was to allow the underlying connection to be replaced if
it fails. Turns out we actually need to use a full lock (not just a read
lock) to prevent multiple requests from happening concurrently.
* node: fix test name. DelayedStop -> DelayedStart
* autofile: Wait() method
In the TestWALTruncate in consensus/wal_test.go we remove the WAL
directory at the end of the test. However the wal.Stop() does not
properly wait for the autofile group to finish shutting down. Hence it
was possible that the group's go-routine is still running when the
cleanup happens, which causes a panic since the directory disappeared.
Here we add a Wait() method to properly wait until the go-routine exits
so we can safely clean up. This fixes#2852.
Anton Kaliaev
folex
Marko
zjubfd
Runchao Han
Sean Braithwaite
Roman Shtylman
Ivan Kushmantsev
Thane Thomson
Martin Dyring-Andersen
Alexander Simmerl
Ethan Buchman
zjubfd
Ismail Khoffi