fe828d87d8
count published messages ( #224 )
2020-06-16 22:14:02 -06:00
224c2613e3
Remove noise padding payload (spec removed it) ( #222 )
2020-06-16 17:34:02 -06:00
523eed65ce
metrics: use eth2 name for peers metric ( #223 )
2020-06-16 14:43:45 -06:00
0d81acafa8
Fix multistream select trace proto logging
2020-06-16 12:02:47 +09:00
9d9f793b4f
add metrics for sent messages by topic and peer ( #220 )
2020-06-15 17:39:03 -06:00
3c4c10d871
con't crash on nil ptr
2020-06-15 13:37:25 -06:00
7cb6c81159
Don't modify iterables while iterating them ( #219 )
...
* don't modify iterables while iterating
* assert handlers to properly close connections
2020-06-15 12:30:09 -06:00
9cd47fe816
add a logScope to make tracing less ugly ( #218 )
...
* add a logScope to make tracing less ugly
* don't crash on nil pointer
2020-06-15 12:10:41 -06:00
85b56d0b3a
Observedaddr ( #217 )
...
* send correct observerAddr
* run tests with info log level
* set observedaddr for channels
2020-06-12 19:56:17 -06:00
f97e2deec7
Make methods in FloodSub, GossipSub public ( #216 )
...
Similar to https://github.com/status-im/nim-libp2p/pull/193 but doing it for all
methods to avoid this issue in PubSub in the future. Got hit by this behavior
again in rpcHandler and took me a while to figure out.
See https://github.com/oskarth/nim-private-method-skipping-case/ for minimal
repro
2020-06-12 17:54:12 -06:00
ac04ca6e31
make sure keys exist and more metrics ( #215 )
2020-06-11 20:20:58 -06:00
55a294a5c9
better pubsub metrics ( #214 )
2020-06-11 12:09:34 -06:00
6b196ad7b4
remove pubsub peer on disconnect ( #212 )
...
* remove pubsub peer on disconnect
* make sure lock is aquired
* add $
* count upgrades/dials/disconnects
2020-06-11 08:45:59 -06:00
92579435b6
secio then noise ( #213 )
...
* secio then noise
much fewer peers on witti with noise first
* comment
2020-06-11 08:38:47 +02:00
1afec627c2
proper name for topics so that we can filter dynamically using chronicles ( #210 )
...
* proper name for topics so that we can filter dynamically using chronicles
* lowercase
2020-06-10 10:48:01 +02:00
8d9e231a74
grab agentversion/protoversion ( #211 )
2020-06-09 12:42:52 -06:00
35ff99829e
close streams ( #208 )
...
* close streams
* don't warn on missing proto
2020-06-08 17:42:27 -06:00
ee281310c0
move trace log
2020-06-08 10:40:08 -06:00
82b4ed8f44
use declareCounter rather then gauge for certain metrics
2020-06-07 16:41:23 +09:00
a6a2a81711
Start adding some metrics to pubsub ( #192 )
...
* Start adding some metrics to pubsub
In order to visualize it's functionality
Still WIP
* more metrics
* add per topic metrics
* finishup with requested metrics
* add a metrisServer define to start local server
* PR fixes and cleanup
2020-06-07 09:15:21 +02:00
130c64f33a
don't return nil in dial ( #205 )
...
* dont return nil in dial
* don't crash on pubsub send
2020-06-05 18:17:05 -06:00
2aebae56c0
Don't rely on the side-effects from doAssert
2020-06-05 19:12:10 +03:00
828a80ec8f
Make the MultiAddress.init function used in NBC non-failing
2020-06-05 17:51:22 +03:00
5960d42c50
remove casts from ( #203 )
2020-06-02 20:21:11 -06:00
bb8bff2195
add sparse message propagation tests to gossipsub ( #202 )
...
* add sparce tests to gossipsub
* add send hooks
* remove `all`
2020-06-02 17:53:38 -06:00
285884c20c
Close peers ( #201 )
...
* wip
* exceptions and resource cleanup
* correct peerlifetime on disconnect
* emulate defered
* remove comment
2020-06-02 11:32:42 -06:00
c7298f34f4
additional comments
2020-06-02 09:10:27 -06:00
71640da8f2
remove close from read/write methods
2020-06-02 09:10:27 -06:00
1b4876d26d
emulate `defered`
2020-06-02 09:10:27 -06:00
abf659a01a
more consistent dialing proto selecting logic
2020-06-02 09:10:27 -06:00
86e1c8169c
decorate observers hooks with {.raises: [Defect].}
...
move hooks logic out into standalone procs
License: MIT
Signed-off-by: Dmitriy Ryajov <dryajov@gmail.com>
2020-06-02 09:10:27 -06:00
4df151a3a3
typos
2020-06-02 09:10:27 -06:00
27a7f8c948
move EOF flag after local close and comments
2020-06-02 09:10:27 -06:00
815282a5da
remove all()
2020-06-02 09:10:27 -06:00
293b7da295
typo
2020-06-02 09:10:27 -06:00
6e0eb93d4f
remove readloops
2020-06-02 09:10:27 -06:00
abc12d0fb5
move stram close to a better location
2020-06-02 09:10:27 -06:00
daef00fc7b
don't crash schlesi-dev
2020-06-02 09:10:27 -06:00
640c3bdc45
better exception handling and resource cleanup
2020-06-02 09:10:27 -06:00
7ff76d76b6
better exceptions
2020-06-02 09:10:27 -06:00
95774b2b81
better cleanup
2020-06-02 09:10:27 -06:00
9c2f31262e
wip: try handling child stream exceptions
2020-06-02 09:10:27 -06:00
e3f8f53620
initStream method and better exceptions handling
2020-06-02 09:10:27 -06:00
d3b79b002e
better exceptions and don't fail writes
2020-06-02 09:10:27 -06:00
74f8b5b5f1
resource cleanup
2020-06-02 09:10:27 -06:00
93e5805c01
better exception handling
2020-06-02 09:10:27 -06:00
d83ce4c932
exceptions and resource cleanup
2020-06-02 09:10:27 -06:00
d4bdb42046
gossipsub fixes
2020-06-02 09:10:27 -06:00
9d3cc9647b
fix merge
2020-06-02 09:10:27 -06:00
3d9c0bffba
read from stream
2020-06-02 09:10:27 -06:00
048b1db1ad
revert back allread
2020-06-02 09:10:27 -06:00
8e9716f5c3
remove on transport close cleanup
2020-06-02 09:10:27 -06:00
cf76edb0dd
make noise work again
2020-06-02 09:10:27 -06:00
5158d96eaf
close connection on chronos close
2020-06-02 09:10:27 -06:00
46daed9a38
wip
2020-06-02 09:10:27 -06:00
0f691cbafd
add eof and closed handling
2020-06-02 09:10:27 -06:00
62da2a05c3
wip: rework with proper half-closed
2020-06-02 09:10:27 -06:00
98117a3068
call write until all is written out
2020-06-02 09:10:27 -06:00
802299e69a
breakout from publish loop
2020-06-02 09:10:27 -06:00
7e31210455
cover missing case in MultiAddress.init ( #198 )
...
* cover missing case in MultiAddress.init
* raise assert on marker in protocol
* unify inits for markers / non-markers
* fix string
2020-06-01 14:41:45 +02:00
88dbeebf17
add side effect annotations ( #197 )
2020-06-01 09:25:16 +02:00
37b98ad45c
Secure managers are now sorted, giving priority to noise ( #191 )
...
* Secure managers are now sorted, giving priority to noise
* fix nimble test command
* Fix native tests
* fix directchat sample
* Could not write to connection - reduce verbosity
* fix interop testing
* Remove more tables
* test interop fixes
* directchat fix
* fix interop/remove some deprecation
2020-06-01 08:41:32 +02:00
6affcda937
Less exceptions more results ( #188 )
...
* Less exceptions more results
* Fix daemonapi and interop tests
* Add multibase
* wip multiaddress
* fix the build, consuming new result types
* fix standard setup
* Simplify match, rename into MaError, add more exaustive err text
* Fix the CI issues
* Fix directchat build
* daemon api fixes
* better err messages formatting
Co-authored-by: Zahary Karadjov <zahary@gmail.com>
2020-05-31 16:22:49 +02:00
7c9e5c2f7a
"Could not write to connection" message split between trace and debug due to log size
2020-05-30 23:47:56 +09:00
b88bfc05f8
Make GossipSub initPubSub method public ( #193 )
...
This means we can use it from other protocols that inherit GossipSub. Otherwise,
a lot of internal state (heartbeat lock etc) doesn't get initialized properly.
2020-05-29 09:35:03 -06:00
7b6e1c0688
Gossipsub interop ( #189 )
...
* interop fixes
* add custom messageid provider and fix seqno
* use ECDSA for speed
* adding messageid tests
* breakout from publish loop
* addressing review comments
* remove unneded var
* dont stop broadcasting on failed peers
2020-05-27 12:33:49 -06:00
536555138c
fix "future" typo
2020-05-26 16:08:09 +09:00
4447d97234
add back forgotten message in tryAndWarn
2020-05-26 15:21:59 +09:00
9132f16927
gossipsub fixes ( #186 )
2020-05-21 14:24:20 -06:00
ba53c08b3c
Track incoming connections ( #181 )
...
* call write until all is written out
* wip: rework with proper half-closed
* add eof and closed handling
* wip
* close connection on chronos close
* don't use read
* make noise work again
* don't reraise just yet
* fixes after backporting
* remove on transport close cleanup
* revert back allread
* rust interop fixes
* read from stream
* inc count before closing
* rebasing master
* store incomming connections
* fix merge
* remove unneeded changes
* use internal close flag to indicate disposal
2020-05-21 11:33:48 -06:00
7900fd9f61
Half closed ( #174 )
...
* call write until all is written out
* add comments to lpchannel fields
* add an eof flag to signal which end closed
* wip: rework with proper half-closed
* add eof and closed handling
* propagate closes to piped
* call parent close
* moving bufferstream trackers out
* move writeLock to bufferstream
* move writeLock out
* remove unused call
* wip
* rebasing master
* fix mplex tests
* wip
* fix bufferstream after backport
* wip
* rename to differentiate from chronos tracker
* close connection on chronos close
* make reset request asyncCheck
* fix channel cleanup
* misc
* don't use read
* fix backports
* make noise work again
* proper exception handling
* don't reraise just yet
* add convenience templates
* dont double wrap
* use async pragma
* fixes after backporting
* muxer owns connection
* remove on transport close cleanup
* revert back allread
* adding some todos
* read from stream
* inc count before closing
* rebasing master
* rebase master
* use correct exception type
* use try/finally insted of defer
* fix compile in trace mode
* reset channels on mplex close
2020-05-19 18:14:15 -06:00
681991ae48
reduce buffer size in lpchannel ( #184 )
2020-05-19 16:15:36 -06:00
01339c991f
Don't use and expose directly secp types ( #183 )
...
* Don't use and expose directly secp types
* Reuse same secp type names
2020-05-19 14:48:55 +02:00
c219100e64
Use results and no exceptions in chacha and curve25519 ( #182 )
2020-05-19 10:22:49 +02:00
f8029e7359
use sha256 digest as cache keys ( #135 )
...
* use sha256 digest as cache keys
* rebasing master
2020-05-18 14:49:49 -06:00
9cf1fd0216
remove generic constructor and expose serverflags ( #176 )
...
* remove generic constructor and expose serverflags
* fix transport constructor
* fix merge issues
2020-05-18 13:04:05 -06:00
773b738c12
don't track Connection, track StreamTransport ( #177 )
...
* don't track Connection, track StreamTransport
* make tests more deterministic
2020-05-18 11:05:34 -06:00
5583168965
Track all connections ( #175 )
...
* adding to do for future refactor
* no mix conn and transport stopping or it will race
* don't use intermediary vars
2020-05-18 09:08:10 -06:00
1819502fb5
Cleanup - tests and logging ( #178 )
...
* make async for proper exception handling
* tryAndWarn msg messes up Exception msg
* misc: comment out tracker dumps
* cleanup mplex tests
* more informative errors
* give CI time to run
* revert change, bacause it causes races
2020-05-18 07:49:49 -06:00
7dcb807f64
Crypto utilities resultification ( #150 )
2020-05-18 07:25:55 +02:00
167f42ed45
Remove read ( #171 )
...
* use readExactly
* remove `read`
* remove read
* no more `read`
2020-05-14 22:02:05 -06:00
69abf5097d
handle a few exceptions ( #170 )
...
* handle a few exceptions
Some of these are maybe too aggressive, but in return, they'll log
their exception - more refactoring needed to sort this out - right now
we get crashes on unhandled exceptions of unknown origin
* during connection setup
* while closing channels
* while processing pubsubs
* catch exceptions that are raised and don't try to catch exceptions that are not raised
* propagate cancellederror
* one more
* more
* more
* make interop tests less fragile
* Raise expiration time in gossipsub fanout test for slow CI
Co-authored-by: Dmitriy Ryajov <dryajov@gmail.com>
Co-authored-by: Giovanni Petrantoni <giovanni@fragcolor.xyz>
2020-05-14 21:56:56 -06:00
005e088405
Properly track and close mplex handlers ( #166 )
...
* Properly track and close mplex handlers
* Avoid verbose warnings
* Fix tryAndWarn trace issue
* Handle LPEOF in lpchannel close
2020-05-12 14:45:32 +02:00
618e01eba3
don't crash on double peerinfo close ( #167 )
2020-05-11 21:05:24 +02:00
3053f03814
fix varint issues
...
* fixes #111
2020-05-11 09:12:23 -06:00
87e1f3c61f
fix trace logging
2020-05-09 13:56:07 +02:00
fbf640794b
move channel size constant
2020-05-09 08:48:26 +02:00
6196d56fc2
check for nil observers
2020-05-08 15:44:18 -06:00
ccd019b328
use stream directly in chronosstream ( #163 )
...
* use stream directly in chronosstream
for now, chronos.AsyncStream is not used to provide any features on top
of chronos.Stream, so in order to simplify the code, chronosstream can
be used directly.
In particular, the exception handling is broken in the current
chronosstream - opening and closing the stream is simplified this way as
well.
A future implementation that actually takes advantage of the AsyncStream
features would wrap AsyncStream instead as a separate lpstream
implementation, leaving this one as-is.
* work around chronos exception type issue
2020-05-08 22:10:06 +02:00
c889224012
Add PubSub observer+ hooks (they can modify as well)
2020-05-08 13:31:52 -06:00
268253ea18
remove Chronos type from public API
2020-05-08 13:23:36 -06:00
313f9b0952
use a Transport.serverFlags attribute
2020-05-08 13:23:36 -06:00
c480e65055
use set[ServerFlags] params instead of hardcoded flags
2020-05-08 13:23:36 -06:00
125843af7d
TcpTransport.listen(): enable SO_REUSEADDR
...
in order to avoid failures when restarting the process while the OS is
keeping open sockets from previous runs
2020-05-08 13:23:36 -06:00
1efada474c
remove readLoop in secure protocols ( #162 )
...
* remove readLoop in secure protocols, fix security issues
* fix Defect on remote sending 0-byte noise/secio message
* remove msglen from `write` (unused)
* simplify SecureConn data flow
* document some control-flow issues
* unify exception behaviour across noise and secio
* secio would not raise on mac/decryption errors
* fix compile error
2020-05-07 14:37:46 -06:00
330da51819
removals ( #159 )
...
* remove unused stream methods
* reimplement some of them with proc's
* remove broken tests
* Error->Defect for defect
* warning fixes
2020-05-06 18:31:47 +02:00
6da4d2af48
Pubsub signatures flags ( #161 )
...
* add verify signature flag
* add sign flag to enable/disable msg signing
* moving internal tests out to their own file
* cleanup nimble file
* remove unneeded tests
* move pubsub tests out
* fix tests
2020-05-06 11:26:08 +02:00
e1928456a7
avoid newlines in $
...
they mess with debug prints and logging (same reason why $(seq) doesn't
print them
2020-04-28 10:59:53 -06:00
290ba712e9
Fix MultiAddress.protoAddress() bug for fixed arrays.
...
Add tests for it.
2020-04-28 14:43:44 +03:00
8a22c073c7
Fix secure/noise securing explicitly, added noise to pubsub tests
2020-04-24 14:33:08 -06:00
Dmitriy Ryajov
Giovanni Petrantoni
Jacek Sieka
Giovanni Petrantoni
Oskar Thorén
Viktor Kirilov
Zahary Karadjov
Ștefan Talpalaru
cheatfate