These are chat archives for akkadotnet/akka.net

13th
Feb 2015
Bartosz Sypytkowski
@Horusiath
Feb 13 2015 08:44
@Aaronontheweb @nvivo it would be awesome. I think that for now 1.0 version + good set of guidlines are more important than more features
Andrew Skotzko
@skotzko
Feb 13 2015 19:31
+1
Aaron Stannard
@Aaronontheweb
Feb 13 2015 21:03
@nvivo beyondhttp.com
@nvivo I'm not sure if the DataStax people recorded my talk or not - I think they did. But at the very least I have the slides up here: http://petabridge.com/datastax
Aaron Stannard
@Aaronontheweb
Feb 13 2015 21:15
@Horusiath updated the release notes for the actor producer pipeline stuff - sorry for not getting to that yesterday. Got tied up at the Cassandra Day here in LA
Bartosz Sypytkowski
@Horusiath
Feb 13 2015 21:30
ofc, no problem ;)
Aaron Stannard
@Aaronontheweb
Feb 13 2015 21:31
random question for you - I'm trying to fix this last bug inside the ThrottleTransportAdapter, and it's come down to understanding what this Future.sequence call does inside Akka: https://github.com/akka/akka/blob/dc0547dd73b54b5de9c3e0b45a21aa865c5db8e2/akka-remote/src/main/scala/akka/remote/transport/ThrottlerTransportAdapter.scala#L241
is it generating a list of Futures, doing a continuation on the result, and then piping the result of each individual Future to the sender?
or is this doing the equivalent of a Task.WhenAll, continuing once all futures are complete, and then piping that result to the sender once?
Bartosz Sypytkowski
@Horusiath
Feb 13 2015 21:34
it's converting a list of futures into future of list, so I think is a second option
Aaron Stannard
@Aaronontheweb
Feb 13 2015 21:34
that's what I thought
and that's how I have it written now
still explodes during the second test
inside the ThrottleTransportAdapterSpec
gets a Task Cancellation exception that propagates all the way up from this call to the Remoting supervisor
my guess is that the way I'm handling the handeTable inside the adapter isn't correct when a ForceDisassociate command is invoked
ok, so I'll focus on that instead of the TPL stuff - thanks for your help @Horusiath :)
Bartosz Sypytkowski
@Horusiath
Feb 13 2015 21:37
np ;)
Aaron Stannard
@Aaronontheweb
Feb 13 2015 22:09
haha, that did the trick
had to remove the handles that had been disassociated
even though the Scala version doesn't do that
my guess is because the Scala version doesn't throw an exception when Ask timesout
it just returns a failed Future