These are chat archives for akkadotnet/akka.net

29th
Jul 2016
[13:02:38 DBG] Started (Movies.Service.Actors.genres.moviesiteNodeActor)
[13:02:38 DBG] now watched by [akka://MoviesService/user/$f#1359103859]
[13:03:23 DBG] now supervising akka://MoviesService/system/trangenres/akkaprotocolmanager.tcp.0/akkaProtocol-tcp%3A%2F%2FMoviesService%40%5B%3A%3Affff%3A127.0.0.1%5D%3A51588-1
[13:03:23 DBG] Started (Akka.Remote.Trangenre.ProtocolStateActor)
[13:03:27 DBG] Associated [akka.tcp://MoviesService@localhost:2552] <- akka.tcp://MoviesWeb@localhost:51587
[13:03:27 DBG] now supervising akka://MoviesService/system/endpointManager/reliableEndpointWriter-akka.tcp%3A%2F%2FMoviesWeb%40localhost%3A51587-1
[13:03:27 DBG] Started (Akka.Remote.ReliableDeliverySupervisor)
[13:03:27 DBG] now watched by [akka://MoviesService/system/endpointManager#458516767]
[13:03:27 DBG] now supervising akka://MoviesService/system/endpointManager/reliableEndpointWriter-akka.tcp%3A%2F%2FMoviesWeb%40localhost%3A51587-1/endpointWriter
[13:03:27 DBG] Started (Akka.Remote.EndpointWriter)
[13:03:27 DBG] now watched by [akka://MoviesService/system/endpointManager/reliableEndpointWriter-akka.tcp%3A%2F%2FMoviesWeb%40localhost%3A51587-1#851916266]
[13:03:27 DBG] now supervising akka://MoviesService/system/endpointManager/reliableEndpointWriter-akka.tcp%3A%2F%2FMoviesWeb%40localhost%3A51587-1/endpointWriter/endpointReader-akka.tcp%3A%2F%2F
MoviesWeb%40localhost%3A51587-1
[13:03:27 DBG] Started (Akka.Remote.EndpointReader)
[13:03:27 DBG] now watched by [akka://MoviesService/system/endpointManager/reliableEndpointWriter-akka.tcp%3A%2F%2FMoviesWeb%40localhost%3A51587-1/endpointWriter#1973816085]
[13:03:29 DBG] received AutoReceiveMessage <Identify>:
[13:03:42 DBG] Starting HTTP pool with parameters: BaseUrl: https://api.moviesite.com.au/mobile/genre/v1/, ThrottleMilliseconds: 250, WorkerCount: 2.
[13:03:42 DBG] Started (Movies.Service.Actors.Http.HttpActor)
[13:03:42 DBG] Started (Movies.Service.Actors.Http.HttpActor)
[13:03:42 DBG] now watched by [akka://MoviesService/user/http/$a#2070132076]
[13:03:43 DBG] now watched by [akka://MoviesService/user/http/$a#2070132076]
[13:03:43 DBG] Started (Akka.Routing.RouterPoolActor)
[13:03:43 DBG] now supervising akka://MoviesService/user/http/$a/$a
[13:03:46 DBG] now supervising akka://MoviesService/user/http/$a/$b
[13:03:46 DBG] Unhandled message from akka://MoviesService/user/http : Movies.Service.Actors.Http.Events.HttpStatusEvent
[13:03:46 DBG] Started HTTP pool.
[13:03:46 DBG] now supervising akka://MoviesService/user/http/$a
[13:03:46 DBG] Downloading: genre?live=true
[13:04:46 ERR] No response from remote. Handshake timed out or trangenre failure detector triggered.
[13:04:46 ERR] Error caught channel [[::ffff:127.0.0.1]:2552->[::ffff:127.0.0.1]:51588](Id=ChannelId(-233291184))
System.Net.Sockets.SocketException (0x80004005): An established connection was aborted by the software in your host machine
   at Helios.Channels.Sockets.TcpSocketChannel.DoReadBytes(IByteBuf buf)
   at Helios.Channels.Sockets.AbstractSocketByteChannel.SocketByteChannelUnsafe.FinishRead(SocketChannelAsyncOperation operation)
[13:04:48 ERR] Error caught channel [[::ffff:127.0.0.1]:2552->[::ffff:127.0.0.1]:51588](Id=ChannelId(-233291184))
System.Net.Sockets.SocketException (0x80004005): An established connection was aborted by the software in your host machine
   at Helios.Channels.Sockets.SocketChannelAsyncOperation.Validate()
   at Helios.Channels.Sockets.AbstractSocketByteChannel.SocketByteChannelUnsafe.FinishRead(SocketChannelAsyncOperation operation)
[13:04:49 DBG] Stopped
[13:04:49 INF] Message Disassociated from NoSender to akka://MoviesService/system/trangenres/akkaprotocolmanager.tcp.0/akkaProtocol-tcp%3A%2F%2FMoviesService%40%5B%3A%3Affff%3A127.0.0.1%5D%3A515
88-1 was not delivered. 1 dead letters encountered.
[13:04:49 INF] Message Disassociated from NoSender to akka://MoviesService/system/trangenres/akkaprotocolmanager.tcp.0/akkaProtocol-tcp%3A%2F%2FMoviesService%40%5B%3A%3Affff%3A127.0.0.1%5D%3A515
88-1 was not delivered. 2 dead letters encounte
^ I keep getting disassociated exception when communicating betweem my MVC site and a background web service
I am resolving the RemoteActor address in the website correctly, and I can Tell the windows service messages correctly, but when I try to send from the windows service back to the website (with Sender.Tell()) I am getting this disassociation error
All latest versions of Akka + Remote and with JSON.NET 9.0.1 as serialiser
*sorry, I mean background Windows Service
Arjen Smits
@Danthar
Jul 29 2016 06:54
@slc it looks like you are using localhost in one node and 127.0.0.1 in the other
We had a DNS resolve issue a short while back, which i think is not in the 1.1.1 release
if you change you config on both sides to use 127.0.0.1 that imight help
Stuart Cam
@codebrain
Jul 29 2016 07:04
@Danthar - I'll check my HOCON, but I'm pretty sure I had localhost in both
I'll move both to use 127.0.0.1 loopback address
Arjen Smits
@Danthar
Jul 29 2016 07:26
Besides that. How have you set up your website ?
the actorsystem needs to be a global singleton
yup, its on the WebApplication as a static
Arjen Smits
@Danthar
Jul 29 2016 07:27
a static reference in the global asax will do, check out the docs for an example.
k
and.. are you 2 persons ?
:P
:)
different machine and account :)
Arjen Smits
@Danthar
Jul 29 2016 07:27
ah
When does the dc happen? Because iis has a habit to recycle the appool which obviously nukes your association.
ah wait, your using sender.tell upon response. so that seems unlikely
It seems to DA as soon as the delegate for sending the response back completes
So... i init the web site and the service. Website resolves an actor from service and passes it a message with its own actorref
That actorref is held by the service and when an event needs to be passed to the web site it uses the actor ref it was given
Then instantly it shows the Diassociation
The actorref held by the service "looks" correct but trying to tell a message to it causes the problem
Trying to tell a message causes the disassociation
Im a bit stumped
Arjen Smits
@Danthar
Jul 29 2016 07:46
thats. odd
I tried 127.0.0.1 and no luck
Kris Schepers
@schepersk
Jul 29 2016 07:48
Hi, what is the best version of Wire to use with akka 1.1.1? Currently references 0.0.6, but 0.7.1 is latest..
Arjen Smits
@Danthar
Jul 29 2016 07:49
@schepersk the latest adapter of wire for akka isn't been published yet
Kris Schepers
@schepersk
Jul 29 2016 07:50
Owkay, lets keep it on 0.0.6 then..
Arjen Smits
@Danthar
Jul 29 2016 07:50
We have a dedicated room for wire questions btw: https://gitter.im/akkadotnet/Wire
not chasing you away though ;)
its something we did a while back.
Kris Schepers
@schepersk
Jul 29 2016 07:51
Is there an update planned for Sql-Server persistence to akka 1.1.1 ? We're using that..
:-)
Arjen Smits
@Danthar
Jul 29 2016 07:52
it helps with keeping the chat logs focused and helps people find answers to specific questions. As opposed to having everything in one chatroom and disapear in the blackhole that is called gitter chat history.
its being worked on
Kris Schepers
@schepersk
Jul 29 2016 07:53
is the current one compatible with the latest sql-common?
or should I just go for the trail and error approach :-)
Arjen Smits
@Danthar
Jul 29 2016 07:54
honestly, i have no idea. I use the sqlite integration everywhere for my own solutions.
I think @Silv3rcircl3 or @alexvaluyskiy might have a better idea
Kris Schepers
@schepersk
Jul 29 2016 07:56
ic.. that's not really an option for us.. dealing with LOTS of events and projections..
would like to update the cluster out of beta asap :-)
Arjen Smits
@Danthar
Jul 29 2016 07:57
i can imagine
1.1.1 is alot better then 1.08
Kris Schepers
@schepersk
Jul 29 2016 07:57
so they say :P
Arjen Smits
@Danthar
Jul 29 2016 07:57
@Aaronontheweb did an insane amount of work
Marc Piechura
@marcpiechura
Jul 29 2016 07:58
replace me with @Horusiath about the sql-common question ;)
Arjen Smits
@Danthar
Jul 29 2016 07:58
ah ok
Marc Piechura
@marcpiechura
Jul 29 2016 07:58
have no idea too
Arjen Smits
@Danthar
Jul 29 2016 07:58
:D
haven't seen @Horusiath around for a while, so no idea how up to date he is ^^
but our main blocker for akka persistence are the CI build images
the core team is currently working to resolve that by moving to docker images which we can host on an azure environment
that eliminates @Aaronontheweb as the bottleneck since he would be the person to normally build the build-agent images
This will really help in making it easier for contributors to help in keeping the persistence plugins up to date, and even introduce new integrations
So these are some of the steps we are taking to improve the speed at which we can keep the persistence plugins updated
Kris Schepers
@schepersk
Jul 29 2016 08:06
awesome!
Arjen Smits
@Danthar
Jul 29 2016 08:09
@slc if you can post your configs in a gist, i can take a look
Will do
Arjen Smits
@Danthar
Jul 29 2016 08:09
(or upload the solution somewhere)
solution upload is tricky for IP reasons
thats the two configs
super simple configs AFAIK
Arjen Smits
@Danthar
Jul 29 2016 08:16
hmm, could you try adding the public-hostname = "127.0.0.1" to the helios.tcp segment
will do now...
Ralf
@Ralf1108
Jul 29 2016 08:17
hi all, I checked out the new Akka.net Streams feature and I think its very goog. One questions was left open. If I create a processing flow how can I distribute it across different machines. Are there any examples how to start?
Arjen Smits
@Danthar
Jul 29 2016 08:17
not sure if its still needed.
public-hostname = "127.0.0.1" Added.
No change. Still errors.
do I need to configure IIS Express in any special way?
The weird thing is that I can send the message from the web actor to the service actor with no problems
Marc Piechura
@marcpiechura
Jul 29 2016 08:20
@Ralf1108 that's not possible out of the box. But what you can do is to have an actor that distributes "job" messages to different machines and have a stream on every machine.
Arjen Smits
@Danthar
Jul 29 2016 08:21
@slc no.
I just cant Tell() back to the web actor from the service actor
James Farrer
@jamesfarrer
Jul 29 2016 08:21
thanks @cgstevens and @qwoz . most helpful
Arjen Smits
@Danthar
Jul 29 2016 08:22
@slc what if you reduce it to its bare minimum. So send a message containing nothing more then a string, and use Sender.Tell to echo it back
I think I'm going to have to
Arjen Smits
@Danthar
Jul 29 2016 08:22
that would eliminate possible serialization issues
I was wondering if that was the case
Marc Piechura
@marcpiechura
Jul 29 2016 08:23
@Ralf1108 here is some documentation about this topic http://getakka.net/docs/streams/integration#integrating-with-actors also @Aaronontheweb is currently rewriting the webcrawler sample to implement exactly the behavior you described
I will try sending a string instead, or perhaps a bool
Ricky Blankenaufulland
@ZoolWay
Jul 29 2016 08:23
@slc I have not followed the whole conversation but because I falled into the trap this week when messages have been swallowed: You are not creating one of the actors in an async method?
nope, all syncronous
Ricky Blankenaufulland
@ZoolWay
Jul 29 2016 08:24
ok
then check for the string/bool
no async/await anywhere
yup, just about to try that
cannot Tell(true) either
:(
I guess that somewhat rules out serialisation/deserialisation issues
Im going to pull out an example from my solution and see if I can replicate in isolation
Ralf
@Ralf1108
Jul 29 2016 08:59
@Silv3rcircl3 ok thx for the hints!
@ZoolWay @Danthar - Found it! Serialization problem with the message being passed back. Message had a private constructor and I assume JSON.NET couldnt deserialise.
So annoying that the error message is a little elusive, nevermind
Jeff
@jpierson
Jul 29 2016 14:31
Greetings all, just learning Akka.NET and I'll have some questions concerning Clustering, Routing, and Sharding. Is this the best place to ask those questions?
Jeff
@jpierson
Jul 29 2016 14:46
In the examples I've seen with Cluster Routing it appears that the routing actor is created via ActorSystem.ActorOf on the same node where messages are being dispatched. Is this the general pattern or are there other options or reasons for having the router itself somewhere else in the cluster? If so is using an actor singleton the way to reach that router?