TrueCloudLab/neo-go
8
0
Fork 0
mirror of https://github.com/nspcc-dev/neo-go.git synced 2024-12-23 13:41:37 +00:00
Code Issues Projects Releases Packages Wiki Activity
6388 commits 38 branches 139 tags 46 MiB
Commit graph

525 commits

Author SHA1 Message Date
Roman Khimov
dabdad20ad network: don't wait indefinitely for packet to be sent 
Peers can be slow, very slow, slow enough to affect node's regular
operation. We can't wait for them indefinitely, there has to be a timeout for
send operations.

This patch uses TimePerBlock as a reference for its timeout. It's relatively
big and it doesn't affect tests much, 4+1 scenarios tend to perform a little
worse with while 7+2 scenarios work a little better. The difference is in some
percents, but all of these tests easily have 10-15% variations from run to
run.

It's an important step in making our gossip better because we can't have any
behavior where neighbors directly block the node forever, refs. #2678 and
 2022-10-10 22:15:21 +03:00
Roman Khimov
317dd42513 *: use uint*Size and SignatureLen constants where appropriate 2022-10-05 10:45:52 +03:00
Roman Khimov
4f3ffe7290 golangci: enable errorlint and fix everything it found 2022-09-02 18:36:23 +03:00
Roman Khimov
779a5c070f network: wait for exit in discoverer 
And synchronize other threads with channels instead of mutexes. Overall this
scheme is more reliable.
 2022-08-19 22:23:47 +03:00
Roman Khimov
eeeb0f6f0e core: accept two-side channels for sub/unsub, read on unsub 
Blockchain's notificationDispatcher sends events to channels and these
channels must be read from. Unfortunately, regular service shutdown procedure
does unsubscription first (outside of the read loop) and only then drains the
channel. While it waits for unsubscription request to be accepted
notificationDispatcher can try pushing more data into the same channel which
will lead to a deadlock. Reading in the same method solves this, any number of
events can be pushed until unsub channel accepts the data.
 2022-08-19 22:08:40 +03:00
Roman Khimov
dea75a4211 network: wait for the relayer thread to finish on shutdown 
Unsubscribe and drain first, then return from the Shutdown method. It's
important wrt to subsequent chain shutdown process (normally it's closed right
after the network server).
 2022-08-19 22:08:40 +03:00
Roman Khimov
155089f4e5 network: drop cleanup from TestVerifyNotaryRequest 
It never runs the server, so 746644a4eb was a
bit wrong with this.
 2022-08-19 20:54:06 +03:00
Anna Shaleva
916f2293b8 *: apply go 1.19 formatter heuristics 
And make manual corrections where needed. See the "Common mistakes
and pitfalls" section of https://tip.golang.org/doc/comment.
 2022-08-09 15:37:52 +03:00
Anna Shaleva
bb751535d3 *: bump minimum supported go version 
Close #2497.
 2022-08-08 13:59:32 +03:00
Roman Khimov
9b0ea2c21b network/consensus: always process dBFT messages as high priority 
Move category definition from consensus to payload, consensus service is the
one of its kind (HP), so network.Server can be adjusted accordingly.
 2022-08-02 13:07:18 +03:00
Roman Khimov
94a8784dcb network: allow to drop services and solve concurrency issues 
Now that services can come and go we need to protect all of the associated
fields and allow to deregister them.
 2022-08-02 13:05:39 +03:00
Roman Khimov
5a7fa2d3df cli: restart consensus service on USR2 
Fix #1949. Also drop wallet from the ServerConfig since it's not used in any
meaningful way after this change.
 2022-08-02 13:05:07 +03:00
Roman Khimov
2e27c3d829 metrics: move package to services 
Where it belongs.
 2022-07-21 23:38:23 +03:00
Anna Shaleva
1ae601787d network: allow to handle GetMPTData with KeepOnlyLatestState on 
And adjust documentation along the way.
 2022-07-14 14:33:20 +03:00
Roman Khimov
dc59dc991b config: move metrics.Config into config.BasicService 
Config package should be as lightweight as possible and now it depends on the
whole metrics package just to get one structure from it.
 2022-07-08 23:30:30 +03:00
Roman Khimov
3fbc1331aa
Merge pull request #2582 from nspcc-dev/fix-server-sync 
network: adjust the way (*Server).IsInSync() works
 2022-07-05 12:28:20 +03:00
Roman Khimov
9f05009d1a
Merge pull request #2580 from nspcc-dev/service-review 
Service review
 2022-07-05 12:23:25 +03:00
Anna Shaleva
0835581fa9 network: adjust the way (*Server).IsInSync() works 
Always return true if sync was reached once. Fix #2564.
 2022-07-05 12:20:31 +03:00
Roman Khimov
3e2eda6752 *: add some comments to service Start/Shutdown methods 2022-07-04 23:03:50 +03:00
Roman Khimov
c26a962b55 *: use localhost address instead of 127.0.0.1, fix #2575 2022-06-30 16:19:07 +03:00
Anna Shaleva
8ab422da66 *: properly unsubscribe from Blockchain events 2022-06-28 19:09:25 +03:00
Roman Khimov
75d06d18c9
Merge pull request #2466 from nspcc-dev/rules-fixes 
Rules scope fixes
 2022-05-06 11:09:39 +03:00
Roman Khimov
bd352daab4 transaction: fix Rules stringer, it's WitnessRules in C# 
See neo-project/neo#2720.
 2022-05-06 10:08:09 +03:00
Elizaveta Chichindaeva
28908aa3cf [#2442] English Check 
Signed-off-by: Elizaveta Chichindaeva <elizaveta@nspcc.ru>
 2022-05-04 19:48:27 +03:00
Roman Khimov
53423b7c37 network: fix panic in blockqueue during shutdown 
panic: send on closed channel

goroutine 116 [running]:
github.com/nspcc-dev/neo-go/pkg/network.(*blockQueue).putBlock(0xc00011b650, 0xc01e371200)
        github.com/nspcc-dev/neo-go/pkg/network/blockqueue.go:129 +0x185
github.com/nspcc-dev/neo-go/pkg/network.(*Server).handleBlockCmd(0xc0002d3c00, {0xf69b7f?, 0xc001520010?}, 0xc02eb44000?)
        github.com/nspcc-dev/neo-go/pkg/network/server.go:607 +0x6f
github.com/nspcc-dev/neo-go/pkg/network.(*Server).handleMessage(0xc0002d3c00, {0x121f4c8?, 0xc001528000?}, 0xc01e35cf80)
        github.com/nspcc-dev/neo-go/pkg/network/server.go:1160 +0x6c5
github.com/nspcc-dev/neo-go/pkg/network.(*TCPPeer).handleIncoming(0xc001528000)
        github.com/nspcc-dev/neo-go/pkg/network/tcp_peer.go:189 +0x98
created by github.com/nspcc-dev/neo-go/pkg/network.(*TCPPeer).handleConn
        github.com/nspcc-dev/neo-go/pkg/network/tcp_peer.go:164 +0xcf
 2022-04-26 00:31:48 +03:00
Roman Khimov
2593bb0535 network: extend Service with Name, use it to distinguish services 2022-04-26 00:31:48 +03:00
Evgeniy Stratonikov
34b1b52784 network: check compressed payload size in decompress 
Signed-off-by: Evgeniy Stratonikov <evgeniy@nspcc.ru>
 2022-03-24 17:22:55 +03:00
Anna Shaleva
753d604784 network: use net.ErrClosed to check network connection was closed 
Close #1765.
 2022-03-17 19:39:18 +03:00
Anna Shaleva
9bbd94d0fa network: tune waiting limits in tests 
Some tests are failing on Windows due to slow runners with errors like the following:
```
2022-02-09T17:11:20.3127016Z     --- FAIL: TestGetData/transaction (1.82s)
2022-02-09T17:11:20.3127385Z         server_test.go:500:
2022-02-09T17:11:20.3127878Z             	Error Trace:	server_test.go:500
2022-02-09T17:11:20.3128533Z             	            				server_test.go:520
2022-02-09T17:11:20.3128978Z             	Error:      	Condition never satisfied
2022-02-09T17:11:20.3129479Z             	Test:       	TestGetData/transaction
```
 2022-02-10 18:58:50 +03:00
Roman Khimov
e621f746a7 config/core: allow to change the number of validators 
Fixes #2320.
 2022-01-31 23:14:38 +03:00
Roman Khimov
60d6fa1125 network: keep a copy of the config inside of Server 
Avoid copying the configuration again and again, make things a bit more
efficient.
 2022-01-24 18:43:01 +03:00
Roman Khimov
89d754da6f network: don't request blocks we already have in the queue 
Fixes #2258.
 2022-01-18 00:04:41 +03:00
Roman Khimov
03fd91e857 network: use assert.Eventually in bq test 
Simpler and more efficient (polls more often and completes the test sooner).
 2022-01-18 00:04:29 +03:00
Roman Khimov
d52a06a82d network: move index-position relation into helper 
Just to make things more clear, no functional changes.
 2022-01-18 00:02:16 +03:00
Roman Khimov
bc6d6e58bc network: always pass transactions to consensus process 
Consensus can require conflicting transactions and it can require more
transactions than mempool can fit, all of this should work. Transactions will
be checked anyway using its secondary mempool. See the scenario from #668.
 2022-01-14 20:08:40 +03:00
Roman Khimov
746644a4eb network: decouple it from blockchainer.Blockchainer 
We don't need all of it.
 2022-01-14 19:57:16 +03:00
Roman Khimov
bf1604454c blockchainer/network: move StateSync interface to the user 
Only network package cares about it.
 2022-01-14 19:57:14 +03:00
Roman Khimov
af87cb082f network: decouple Server from the notary service 2022-01-14 19:55:53 +03:00
Roman Khimov
508d36f698 network: drop consensus dependency 2022-01-14 19:55:53 +03:00
Roman Khimov
66aafd868b network: unplug stateroot service from the Server 
Notice that it makes the node accept Extensible payloads with any category
which is the same way C# node works. We're trusting Extensible senders,
improper payloads are harmless until they DoS the network, but we have some
protections against that too (and spamming with proper category doesn't differ
a lot).
 2022-01-14 19:55:50 +03:00
Roman Khimov
0ad3ea5944 network/cli: move Oracle service instantiation out of the network 2022-01-14 19:53:45 +03:00
Roman Khimov
5dd4db2c02 network/services: unify service lifecycle management 
Run with Start, Stop with Shutdown, make behavior uniform.
 2022-01-14 19:53:45 +03:00
Roman Khimov
c942402957 blockchainer: drop Policer interface 
We never use it as a proper interface, so it makes no sense keeping it this
way.
 2022-01-12 00:58:03 +03:00
Roman Khimov
48de82d902 network: fix data race in TestHandleMPTData, fix #2241 2021-11-15 12:37:01 +03:00
Roman Khimov
fe50f6edc7
Merge pull request #2240 from nspcc-dev/fix-panic-in-network 
Fix panic on peer disconnect
 2021-11-01 12:44:15 +03:00
Roman Khimov
774dee3cd4 network: fix disconnection race between handleConn() and handleIncoming() 
handleIncoming() winning the race for p.Disconnect() call might lead to nil
error passed as the reason for peer unregistration.
 2021-11-01 12:20:55 +03:00
Roman Khimov
2eeec73770 network: don't panic if there is no reason for disconnect 
Although error should always be there, we shouldn't fail like this if it's not:
    | panic: runtime error: invalid memory address or nil pointer dereference
    | [signal SIGSEGV: segmentation violation code=0x1 addr=0x18 pc=0xc8884c]
    |
    | goroutine 113 [running]:
    | github.com/nspcc-dev/neo-go/pkg/network.(*Server).run(0xc000150580)
    |         github.com/nspcc-dev/neo-go/pkg/network/server.go:396 +0x7ac
    | github.com/nspcc-dev/neo-go/pkg/network.(*Server).Start(0xc000150580, 0x0)
    |         github.com/nspcc-dev/neo-go/pkg/network/server.go:294 +0x3fb
    | created by github.com/nspcc-dev/neo-go/cli/server.startServer
    |         github.com/nspcc-dev/neo-go/cli/server/server.go:344 +0x56f
 2021-11-01 12:19:00 +03:00
Roman Khimov
8bb1ecb45a network: remove priority queue from block queue 
Use circular buffer which is a bit more appropriate. The problem is that
priority queue accepts and stores equal items which wastes memory even in
normal usage scenario, but it's especially dangerous if the node is stuck for
some reason. In this case it'll accept from peers and put into queue the same
blocks again and again leaking memory up to OOM condition.

Notice that queue length calculation might be wrong in case circular buffer
wraps, but it's not very likely to happen (usually blocks not coming from the
queue are added by consensus and it's not very fast in doing so).
 2021-11-01 11:49:01 +03:00
AnnaShaleva
2d196b3f35 rpc: refactor calculatenetworkfee handler 
Use (Blockchainer).VerifyWitness() to calculate network fee for
contract-based witnesses.
 2021-10-25 19:07:25 +03:00
Evgeniy Stratonikov
4dd3a0d503 network: request headers in parallel, fix #2158 
Do this similarly to how blocks are requested.
See also 4aa1a37.

Signed-off-by: Evgeniy Stratonikov <evgeniy@nspcc.ru>
 2021-10-06 15:25:54 +03:00