Commit Graph

21 Commits

Author SHA1 Message Date
Ferenc Szabo
e38b227ce6 Ci race detector handle failing tests (#19143)
* swarm/storage: increase mget timeout in common_test.go

 TestDbStoreCorrect_1k sometimes timed out with -race on Travis.

--- FAIL: TestDbStoreCorrect_1k (24.63s)
    common_test.go:194: testStore failed: timed out after 10s

* swarm: remove unused vars from TestSnapshotSyncWithServer

nodeCount and chunkCount is returned from setupSim and those values
we use.

* swarm: move race/norace helpers from stream to testutil

As we will need to use the flag in other packages, too.

* swarm: refactor TestSwarmNetwork case

Extract long running test cases for better visibility.

* swarm/network: skip TestSyncingViaGlobalSync with -race

As panics on Travis.

panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x0 pc=0x7e351b]

* swarm: run TestSwarmNetwork with fewer nodes with -race

As otherwise we always get test failure with `network_test.go:374:
context deadline exceeded` even with raised `Timeout`.

* swarm/network: run TestDeliveryFromNodes with fewer nodes with -race

Test on Travis times out with 8 or more nodes if -race flag is present.

* swarm/network: smaller node count for discovery tests with -race

TestDiscoveryPersistenceSimulationSimAdapters failed on Travis with
`-race` flag present. The failure was due to extensive memory usage,
coming from the CGO runtime. Using a smaller node count resolves the
issue.

=== RUN   TestDiscoveryPersistenceSimulationSimAdapter
==7227==ERROR: ThreadSanitizer failed to allocate 0x80000 (524288) bytes of clock allocator (error code: 12)
FATAL: ThreadSanitizer CHECK failed: ./gotsan.cc:6976 "((0 && "unable to mmap")) != (0)" (0x0, 0x0)
FAIL    github.com/ethereum/go-ethereum/swarm/network/simulations/discovery     804.826s

* swarm/network: run TestFileRetrieval with fewer nodes with -race

Otherwise we get a failure due to extensive memory usage, as the CGO
runtime cannot allocate more bytes.

=== RUN   TestFileRetrieval
==7366==ERROR: ThreadSanitizer failed to allocate 0x80000 (524288) bytes of clock allocator (error code: 12)
FATAL: ThreadSanitizer CHECK failed: ./gotsan.cc:6976 "((0 && "unable to mmap")) != (0)" (0x0, 0x0)
FAIL	github.com/ethereum/go-ethereum/swarm/network/stream	155.165s

* swarm/network: run TestRetrieval with fewer nodes with -race

Otherwise we get a failure due to extensive memory usage, as the CGO
runtime cannot allocate more bytes ("ThreadSanitizer failed to
allocate").

* swarm/network: skip flaky TestGetSubscriptionsRPC on Travis w/ -race

Test fails a lot with something like:
 streamer_test.go:1332: Real subscriptions and expected amount don't match; real: 0, expected: 20

* swarm/storage: skip TestDB_SubscribePull* tests on Travis w/ -race

Travis just hangs...

ok  	github.com/ethereum/go-ethereum/swarm/storage/feed/lookup	1.307s
keepalive
keepalive
keepalive

or panics after a while.

Without these tests the race detector job is now stable. Let's
invetigate these tests in a separate issue:
https://github.com/ethersphere/go-ethereum/issues/1245
2019-02-20 22:57:42 +01:00
Janoš Guljaš
3fd6db2bf6 swarm: fix network/stream data races (#19051)
* swarm/network/stream: newStreamerTester cleanup only if err is nil

* swarm/network/stream: raise newStreamerTester waitForPeers timeout

* swarm/network/stream: fix data races in GetPeerSubscriptions

* swarm/storage: prevent data race on LDBStore.batchesC

https://github.com/ethersphere/go-ethereum/issues/1198#issuecomment-461775049

* swarm/network/stream: fix TestGetSubscriptionsRPC data race

https://github.com/ethersphere/go-ethereum/issues/1198#issuecomment-461768477

* swarm/network/stream: correctly use Simulation.Run callback

https://github.com/ethersphere/go-ethereum/issues/1198#issuecomment-461783804

* swarm/network: protect addrCountC in Kademlia.AddrCountC function

https://github.com/ethersphere/go-ethereum/issues/1198#issuecomment-462273444

* p2p/simulations: fix a deadlock calling getRandomNode with lock

https://github.com/ethersphere/go-ethereum/issues/1198#issuecomment-462317407

* swarm/network/stream: terminate disconnect goruotines in tests

* swarm/network/stream: reduce memory consumption when testing data races

* swarm/network/stream: add watchDisconnections helper function

* swarm/network/stream: add concurrent counter for tests

* swarm/network/stream: rename race/norace test files and use const

* swarm/network/stream: remove watchSim and its panic

* swarm/network/stream: pass context in watchDisconnections

* swarm/network/stream: add concurrent safe bool for watchDisconnections

* swarm/storage: fix LDBStore.batchesC data race by not closing it
2019-02-13 13:03:23 +01:00
Anton Evangelatov
597597e8b2 swarm/network: refactor simulation tests bootstrap (#18975) 2019-02-01 09:58:46 +01:00
holisticode
43e1b7b124 swarm: GetPeerSubscriptions RPC (#18972) 2019-01-30 21:03:08 +01:00
Elad
2abeb35d54 p2p/testing, swarm: remove unused testing.T in protocol tester (#18500) 2019-01-24 17:23:34 +01:00
holisticode
88168ff5c5 Stream subscriptions (#18355)
* swarm/network: eachBin now starts at kaddepth for nn

* swarm/network: fix Kademlia.EachBin

* swarm/network: fix kademlia.EachBin

* swarm/network: correct EachBin implementation according to requirements

* swarm/network: less addresses simplified tests

* swarm: calc kad depth outside loop in EachBin test

* swarm/network: removed printResults

* swarm/network: cleanup imports

* swarm/network: remove kademlia.EachBin; fix RequestSubscriptions and add unit test

* swarm/network/stream: address PR comments

* swarm/network/stream: package-wide subscriptionFunc

* swarm/network/stream: refactor to kad.EachConn
2019-01-11 15:08:09 +01:00
holisticode
ae857e74bf swarm, p2p/protocols: Stream accounting (#18337)
* swarm: completed 1st phase of swap accounting

* swarm, p2p/protocols: added stream pricing

* swarm/network/stream: gofmt simplify stream.go

* swarm: fixed review comments

* swarm: used snapshots for swap tests

* swarm: custom retrieve for swap (less cascaded requests at any one time)

* swarm: addressed PR comments

* swarm: log output formatting

* swarm: removed parallelism in swap tests

* swarm: swap tests simplification

* swarm: removed swap_test.go

* swarm/network/stream: added prefix space for comments

* swarm/network/stream: unit test for prices

* swarm/network/stream: don't hardcode price

* swarm/network/stream: fixed invalid price check
2019-01-08 00:59:00 +01:00
Dave McGregor
33d233d3e1
vendor, crypto, swarm: switch over to upstream sha3 package 2019-01-04 09:26:07 +02:00
holisticode
79c7a69ac8 swarm: Better syncing and retrieval option definition (#17986)
* swarm: Better syncing and retrieval option definition

* swarm/network/stream: better comments

* swarm/network/stream: addressed PR comments
2018-11-07 00:04:18 +01:00
Viktor Trón
6566a0a3b8
swarm/network/stream: generalise setting of next batch (#17818)
* swarm/network/stream: generalize SetNextBatch and add Server SessionIndex

* swarm/network/stream: fix a typo in comment

* swarm/network/stream: remove live argument from NewSwarmSyncerServer
2018-10-12 16:26:16 +02:00
holisticode
11d0ff6578 Fix retrieval tests and simulation backends (#17723)
* swarm/network/stream: introduced visualized snapshot sync test

* swarm/network/stream: non-existing hash visualization sim

* swarm/network/stream: fixed retrieval tests; new backend for visualization

* swarm/network/stream: cleanup of visualized_snapshot_sync_sim_test.go

* swarm/network/stream: rebased PR on master

* swarm/network/stream: fixed loop logic in retrieval tests

* swarm/network/stream: fixed iterations for snapshot tests

* swarm/network/stream: address PR comments

* swarm/network/stream: addressed PR comments
2018-10-08 20:28:44 +02:00
Janos Guljas
a5e6bf7eef Merge branch 'master' into max-stream-peer-servers 2018-09-27 09:43:00 +02:00
Viktor Trón
4b6824e07b
Merge pull request #17734 from frncmx/fix-dos-attack-invalid-hash-length
swarm/network/stream: fix DoS invalid offered hashes length
2018-09-26 12:44:42 +02:00
Janos Guljas
0d5e1e7bc9 swarm/network/stream: fix a typo in test comment 2018-09-26 11:30:45 +02:00
Janos Guljas
24349144b6 Merge branch 'master' into max-stream-peer-servers 2018-09-25 16:57:31 +02:00
Felix Lange
30cd5c1854
all: new p2p node representation (#17643)
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.
2018-09-25 00:59:00 +02:00
Janos Guljas
9e99a0c2b9 cmd/swarm, swarm: add stream peer servers limit 2018-09-24 17:56:00 +02:00
Ferenc Szabo
d3f056bd68 swarm/network/stream: fix DoS invalid hash length (#927) 2018-09-21 12:56:43 +02:00
Balint Gabor
3ff2f75636
swarm: Chunk refactor (#17659)
Co-authored-by: Janos Guljas <janos@resenje.org>
Co-authored-by: Balint Gabor <balint.g@gmail.com>
Co-authored-by: Anton Evangelatov <anton.evangelatov@gmail.com>
Co-authored-by: Viktor Trón <viktor.tron@gmail.com>
2018-09-13 11:42:19 +02:00
Anton Evangelatov
7c9314f231 swarm: integrate OpenTracing; propagate ctx to internal APIs (#17169)
* swarm: propagate ctx, enable opentracing

* swarm/tracing: log error when tracing is misconfigured
2018-07-13 17:40:28 +02:00
ethersphere
e187711c65 swarm: network rewrite merge 2018-06-21 21:10:31 +02:00