Ivan Vilata-i-Balaguer
2c61596871
Add property to check if network stress test thread report setup failure
2016-05-17 09:13:28 +02:00
Ivan Vilata-i-Balaguer
9c58d9f740
Minor changes to network stress test
...
I was going to override the ``runBare()`` method to cleanup on ``setUp()`` failures, but ``tearDown()`` is called all the same.
2016-05-17 09:13:28 +02:00
Ivan Vilata-i-Balaguer
260840a9b2
Do not wait on seed node if not created in network stress test
2016-05-17 09:13:28 +02:00
Ivan Vilata-i-Balaguer
718c935fef
Some comments and better names in network stress test
2016-05-17 09:13:28 +02:00
Ivan Vilata-i-Balaguer
dc149f30e3
Wait for 30 seconds in test
...
To allow test connections. Also remove stale print.
2016-05-17 09:13:28 +02:00
Ivan Vilata-i-Balaguer
9f0dfb5512
Use variable to indicate whether a localhost address is being used
2016-05-17 09:13:28 +02:00
Ivan Vilata-i-Balaguer
c9f0155473
Remove unnecessary imports from network stress test
2016-05-17 09:13:28 +02:00
Ivan Vilata-i-Balaguer
0977ef398b
Move seed note running out of user thread, still locking
2016-05-17 09:13:28 +02:00
Ivan Vilata-i-Balaguer
a9d0272e88
Start seed node in stress test, not working yet (locking)
2016-05-17 09:13:28 +02:00
Ivan Vilata-i-Balaguer
fdc0eafc67
Add skeleton for network stress unit test
2016-05-17 09:13:28 +02:00
Manfred Karrer
c745bb37a7
Merge pull request #434 from ivilata/PeerManagerInstanceMaxConn
...
Per-instance max connections in PeerManager
2016-05-14 12:10:59 +02:00
Manfred Karrer
1dae1dd096
Use long instead of int for bytes in statistics
2016-05-12 23:46:05 +02:00
Ivan Vilata-i-Balaguer
bd0f78b10a
Demote keepalive warning to debug entry
...
The error is pretty obscure for a normal user to see in console, and it just reflects a network condition or some delay in the communication with another node, which may be relatively frequent.
Leaving a TODO comment to check if this can cause issues.
2016-05-10 08:28:54 +02:00
Manfred Karrer
5b7bff1a89
Add null check
2016-05-08 22:01:57 +02:00
Manfred Karrer
538751b9da
Add log
2016-05-07 13:17:41 +02:00
Ivan Vilata-i-Balaguer
a1f98f2495
Rename former constant-like fields to camel case in `PeerManager`
2016-05-06 10:58:01 +02:00
Ivan Vilata-i-Balaguer
3aac3a3865
Rearrange connection limit lines in `PeerManager`
2016-05-06 10:58:01 +02:00
Ivan Vilata-i-Balaguer
f0b0b69b24
Change access of max connections constant to private in tests
2016-05-06 10:58:00 +02:00
Ivan Vilata-i-Balaguer
fd37b8999b
Do pass max connections to `PeerManager`, new constructor signature
...
Remove calls to now instance private ``PeerManager.setMaxConnections()``. Static fields regarding connection limits are now private instance fields.
2016-05-06 10:57:11 +02:00
Ivan Vilata-i-Balaguer
dc90a950cb
Pass max connections to `SeedNode.createAndStartP2PService()`
...
Still keep calls to ``PeerManager.setMaxConnections()`` until it is retired.
2016-05-06 10:11:44 +02:00
Ivan Vilata-i-Balaguer
f0aec0d55c
Add internal `P2PService` constructor with (unused) maximum connections argument
...
The value of ``P2PService.MAX_CONNECTIONS_DEFAULT`` is taken from the static block invocation of ``PeerManager.setMaxConnections()``.
2016-05-06 10:09:58 +02:00
Ivan Vilata-i-Balaguer
a50bf41803
Remove unnecessary class reference
2016-05-06 09:00:21 +02:00
Manfred Karrer
8bf4e88c1f
Merge pull request #433 from ivilata/NewTimerRaceConditions
...
New timer race conditions
2016-05-05 13:37:07 +02:00
Manfred Karrer
b9dfa0a7a7
Merge branch 'master' into Development
...
# Conflicts:
# common/pom.xml
# common/src/main/java/io/bitsquare/app/Version.java
# core/pom.xml
# core/src/main/java/io/bitsquare/locale/CurrencyUtil.java
# gui/pom.xml
# jsocks/pom.xml
# jtorctl/pom.xml
# jtorproxy/pom.xml
# network/pom.xml
# package/linux/create_32bit_app.sh
# package/linux/create_64bit_app.sh
# package/mac/create_app.sh
# package/windows/Bitsquare.iss
# package/windows/create_32bit_app.bat
# package/windows/create_app.bat
# pom.xml
# seednode/pom.xml
2016-05-05 13:30:18 +02:00
Ivan Vilata-i-Balaguer
846cdb743f
Ignore data request message success or failure after a timeout
2016-05-05 10:40:59 +02:00
Ivan Vilata-i-Balaguer
00df057e09
Stop broadcasting after a timeout or cancellation
2016-05-05 10:32:52 +02:00
Ivan Vilata-i-Balaguer
0d32a3b5ce
Move comment closer to timer creation
...
For consistency across classes.
2016-05-05 10:25:42 +02:00
Ivan Vilata-i-Balaguer
b814eff5f1
Move log message closer to actual sending
2016-05-05 10:23:57 +02:00
Ivan Vilata-i-Balaguer
e4ce976bab
Avoid potential race condition on timer creation in `BroadcastHandler.broadcast()`
...
If the timer is created after sending the message and establishing callbacks, they may have already run on timer creation so it would be pointless (and issue a warning when triggered).
Currently there is no race condition because no callbacks are attached to the futures return by sending broadcast messages, but the timer is still created before in case they were added and for consistency with other classes.
2016-05-05 10:22:47 +02:00
Ivan Vilata-i-Balaguer
c3fa4831c5
Avoid race condition on timer creation in `RequestDataHandler.requestData()`
...
If the timer is created after sending the message and establishing callbacks, they may have already run on timer creation so it would be pointless (and issue a warning when triggered).
2016-05-05 10:05:17 +02:00
Ivan Vilata-i-Balaguer
b59e3ff0f8
Avoid race condition on timer creation in `GetPeersRequestHandler.handle()`
...
If the timer is created after sending the message and establishing callbacks, they may have already run on timer creation so it would be pointless (and issue a warning when triggered).
2016-05-05 09:45:28 +02:00
Ivan Vilata-i-Balaguer
704b985121
Avoid race condition on timer creation in `PeerExchangeHandler.sendGetPeersRequest()`
...
If the timer is created after sending the message and establishing callbacks, they may have already run on timer creation so it would be pointless (and issue a warning when triggered).
2016-05-05 09:36:49 +02:00
Ivan Vilata-i-Balaguer
4e95b6e6cb
Avoid race condition on timer creation in `GetDataRequestHanler.handle()`
...
If the timer is created after sending the message and establishing callbacks, they may have already run on timer creation so it would be pointless (and issue a warning when triggered).
2016-05-05 09:13:34 +02:00
Manfred Karrer
89aed23d67
v0.4.6
2016-04-29 11:22:58 +02:00
Manfred Karrer
dd5cacb26f
Version 0.4.5
2016-04-28 22:24:51 +02:00
Manfred Karrer
ae85d4e3a7
Add new seed node
2016-04-26 20:53:11 +02:00
Manfred Karrer
2c25c69beb
Remove instable seed node
2016-04-26 17:33:06 +02:00
Manfred Karrer
4e09428180
Beta version 0.4.4
2016-04-26 17:29:11 +02:00
Manfred Karrer
171f61f60c
Fix logging, increase throttle
2016-04-25 23:43:16 +02:00
Manfred Karrer
280574638f
Add new seed node
2016-04-25 00:16:20 +02:00
Manfred Karrer
9f38374539
Fix wrong thread call on exite form sys tray
2016-04-23 03:36:55 +02:00
Manfred Karrer
b4edbe2986
Improve logging
2016-04-23 02:51:01 +02:00
Manfred Karrer
24b109421f
Version 0.4.3
2016-04-23 00:12:20 +02:00
Ivan Vilata-i-Balaguer
85283b8fd0
Create some constants for maximum number of connections
2016-04-20 10:03:27 +02:00
Ivan Vilata-i-Balaguer
45b1314099
Move implicit purge removal period to declared constant
2016-04-20 10:03:27 +02:00
Ivan Vilata-i-Balaguer
7a7b33fc76
Add comment about check allowed public keys for adding received message
2016-04-20 10:03:27 +02:00
Ivan Vilata-i-Balaguer
e8d289f577
Add comment about network node start
2016-04-20 10:03:27 +02:00
Ivan Vilata-i-Balaguer
4b5a393911
Add comment on purging sequence number map
2016-04-20 10:03:27 +02:00
Ivan Vilata-i-Balaguer
5d78918fb2
Extend comment on the role of Connection.InputHandler
2016-04-20 10:03:27 +02:00
Ivan Vilata-i-Balaguer
9b9fd678a8
Avoid getting current time on each usage on flooding avoidance computations
...
Also fix some log message wording.
2016-04-20 10:03:27 +02:00
