These are chat archives for akkadotnet/akka.net

5th
Dec 2016
Ricky Blankenaufulland
@ZoolWay
Dec 05 2016 15:38
Hi! I got a child actor failing, the parent provides a supervisor strategy for that exception of Restart. And then failing actor instance executes PreRestart and PostStop - but no new instance is created, no call of the constructor, PostRestart or PreStart. Has somebody a hint how the restart might fail?
Bas Bossink
@basbossink
Dec 05 2016 16:18
Is there any particular reason that a lot of Exceptions in Akka.net are not marked with the [Serializable] attribute
?
Aaron Stannard
@Aaronontheweb
Dec 05 2016 16:18
nope
not as far as I know
Bas Bossink
@basbossink
Dec 05 2016 16:20
oh, I'm currently getting a serialization exception on the InvalidActorNameException and searched the code a bit. Some exception types have the [Serializable] attribute but most do not, is it worth the effort to submit a github for this and a corresponding pull-request that fixes this?
github issue.
Kris Schepers
@schepersk
Dec 05 2016 17:42

@Horusiath Still investigating my issue here.. In the logs I find the following:
2016-12-03 10:10:17.0387|ERROR|Akka.Remote.Transport.Helios.TcpServerHandler|Error caught channel [::ffff:10.20.3.25]:8251->[::ffff:10.20.4.23]:59831)|System.Net.Sockets.SocketException (0x80004005): An existing connection was forcibly closed by the remote host
at Helios.Channels.Sockets.SocketChannelAsyncOperation.Validate()
at Helios.Channels.Sockets.AbstractSocketByteChannel.SocketByteChannelUnsafe.FinishRead(SocketChannelAsyncOperation operation)
2016-12-03 10:10:17.0856|ERROR|Akka.Remote.Transport.Helios.TcpServerHandler|Error caught channel [::ffff:10.20.3.25]:8251->[::ffff:10.20.4.23]:59831)|System.Net.Sockets.SocketException (0x80004005): An existing connection was forcibly closed by the remote host
at Helios.Channels.Sockets.SocketChannelAsyncOperation.Validate()
at Helios.Channels.Sockets.AbstractSocketByteChannel.SocketByteChannelUnsafe.FinishRead(SocketChannelAsyncOperation operation)

This is followed by a disassociation.
Funny thing is that our nodes are alle configured tu work in the 8250 to 8259 port range. Why would port 59831 be used?

Bartosz Sypytkowski
@Horusiath
Dec 05 2016 18:36
@schepersk don't know. Are you using explicit port definition, or just pass 0?
Kris Schepers
@schepersk
Dec 05 2016 18:38

@Horusiath explicit port definitions..

remote {
log-remote-lifecycle-events = DEBUG
helios.tcp {
transport-class = "Akka.Remote.Transport.Helios.HeliosTcpTransport, Akka.Remote"
applied-adapters = []
transport-protocol = tcp
hostname = "127.0.0.1"
port = 8251
}
}
like this..

Bartosz Sypytkowski
@Horusiath
Dec 05 2016 18:38
it's strange...
I would say that this may be a configuration issue.
Kris Schepers
@schepersk
Dec 05 2016 18:40
is there another part in hocon where you have to define a port for clustering?
Bartosz Sypytkowski
@Horusiath
Dec 05 2016 18:40
no, just in the akka.remote.helios.tcp.port
Kris Schepers
@schepersk
Dec 05 2016 18:42
i'll triple check every config in every node..
Kris Schepers
@schepersk
Dec 05 2016 18:47
should be fine..
Just to be sure, I'll ask for the production configs tomorrow..
Aaron Stannard
@Aaronontheweb
Dec 05 2016 21:33
@Horusiath approved the batch write stuff