These are chat archives for akkadotnet/akka.net

13th
Oct 2017
Bartosz Sypytkowski
@Horusiath
Oct 13 2017 07:10
@codenakama isn't geospatial something that can be simply handled by the database itself?
@defrag2_twitter As long as it's idempotent and won't fail (in your example it's get-or-create an actor), it should be good to go
Ricardo Abreu
@codenakama
Oct 13 2017 08:17
@Horusiath indeed, the logic can be done from the db. I think I should keep that inside each micro service and jsut kill my "location service" which was a bad idea
however, I could have a persistent actor registering "login" events for each shop, so this way I could tell which shops are available to order from
Thomas Lazar
@thomaslazar
Oct 13 2017 08:34
good morning
i have a question... i can'T seem to get akka testing to run wiht nunit 3 (any version). i seen there were some fixes done around an issue regarding nunit3... but i can't seem to make it work. anyone got any clues?
Jose Carlos Marquez
@oeaoaueaa
Oct 13 2017 12:57
what kind of error are you having?
Thomas Lazar
@thomaslazar
Oct 13 2017 16:45
when i try to run the tests the resharper testrunner tells me "no testfixtures" found
Andrey Leskov
@andreyleskov
Oct 13 2017 17:54
Hey guys, I'm scaling my application with cluster sharding, and have a look at multinode test kit (http://getakka.net/articles/networking/multi-node-test-kit.html). How it supposed to be debugged? is there any guide? may be runners for Resharper Tests, as for NBench ? (https://github.com/Pro-Coded/Pro.NBench.xUnit) I can imaging only attach VS to multinode test runner, and it seems to be painful and long process. (
Aaron Stannard
@Aaronontheweb
Oct 13 2017 17:55
@thomaslazar running .NET Core or .NET desktop?
@andreyleskov debugging MNTR specs is a dark art
having done it for years, I have a couple of approaches I use
first - relying on logging as your primary source of information is the lowest-friction path
trying to debug 4-5 processes running concurrently
is going to end poorly usually
especially if you don't know where the problem is
Andrey Leskov
@andreyleskov
Oct 13 2017 17:56
oh, >_< yep
Aaron Stannard
@Aaronontheweb
Oct 13 2017 17:56
you can use the child process debugger extension
is a free add-on to Visual Studio
zero clue if that supports .NET Core or not
that can help get the debugger attached to breakpoints in the child process
so my methodology is this: use logging to smoke out the error
the MNTR runner will produce detailed logs for each node
and the overall test run report will show you which nodes had failed assertions
get enough log data to form a theory as to why the thing might be failing
Andrey Leskov
@andreyleskov
Oct 13 2017 17:58
ok, thank you for fast and detailed answer! will try the dark art )
Aaron Stannard
@Aaronontheweb
Oct 13 2017 17:58
and once you have that, then it becomes more feasible to try and use breakpoints
conditional break points are you best friend here too, btw
should only start firing once you're close to the error condition
multi-node tests, by their very nature, are big and cumbersome
but they're the most reliable way to put real-world theories around handling things like failover, network issues, etc, to the test
Andrey Leskov
@andreyleskov
Oct 13 2017 18:00
ok, can I use 1-node cluster in TestKit - based tests for basic functionality verification, using cluster.sharding ?
Aaron Stannard
@Aaronontheweb
Oct 13 2017 18:00
so to give you an example of something I'm using the MNTR for in my own application
I have a failover scenario where a cluster singleton has to migrate from one node to another
Andrey Leskov
@andreyleskov
Oct 13 2017 18:01
it is interesting
Aaron Stannard
@Aaronontheweb
Oct 13 2017 18:01
and the other nodes relying on data from that singleton must be able to receive notifications about the location of the new singleton on the network
I'm going to be using the MNTR to test my recovery mechanism there
because a simple unit test can't do it justic3e
have a similar scenario where the node with the singelton running on it gets blacked out temporarily and comes back
"does my cluster react according to my plan?" is what my MNTR spec tries to answer there
Andrey Leskov
@andreyleskov
Oct 13 2017 18:03
ok, it seems I should use MNTR tests only for high-level scenarios, and test all application logic with TestKit
by the way - is there any good profiling tool for akka.net like dotTrace? I'm using NBench to measure performance and looking for a simpler ways to find bottlenecks
or any guide how to analyze akka-based application performance with common tools
best practices, may be
@thomaslazar please provide a minimal example for reproducing the issue
Aaron Stannard
@Aaronontheweb
Oct 13 2017 18:13
working on something for doing much more detailed system tracing and monitoring
but it's in early stages still
Andrey Leskov
@andreyleskov
Oct 13 2017 18:15
ok, I'm sure it will be great !