Dear hakkers,
we are happy to announce Akka Http 10.1.0, the first release of the 10.1.x
series.
See the announcement
at https://akka.io/blog/news/2018/03/08/akka-http-10.1.0-released.
The most important changes are:
- The new client pool implementation introduced in 10.0.11 is now
Dear hakkers,
we are happy to announce Akka Http 10.1.0-RC2, the second release candidate
for the upcoming Akka HTTP 10.1.0 release. We focused on stabilizing the
new client connection pool. Over the last month our test suite caught lots
of edge cases that were fixed for this release. Also,
Hi,
in akka-stream, processing is usually run in a fused fashion, i.e. without
further configuration one stream will run in a single actor so all
operations are run sequentially. In such a synchronous scenario, there's
little room for elements to ever get dropped because the actorRef stage
Dear fast application restarters,
we just released sbt-revolver 0.9.0 which is the first version of
sbt-revolver cross-built for sbt 0.13.x and 1.0.x. Thanks go to Olli
Helenius / @liff who contributed the sbt 1.0 compatibility changes
(#62).
We also merged a long-standing PR that allows to
Thanks Julian for sharing the example. Indeed, using the BroadcastHub is
the recommended way to implement something like this.
On Saturday, April 22, 2017 at 1:09:19 AM UTC+2, Julian Howarth wrote:
>
> I may have misunderstood what you want to achieve, but you don't have to
> use actors if
Hi Thibault,
if you have the body of the response already as a Source,
you can create a response with a chunked entity from it like this:
HttpResponse.create()
.withEntity(HttpEntities.create(contentType, source))
Johannes
On Thursday, April 20, 2017 at 9:56:47 AM UTC+2,
Hi Gavin,
the current version of Play is not compatible with Akka 2.5.0. This has
already been fixed for the upcoming Play 2.6 series. I created a ticket to
investigate if the compatibility fix should be backported to Play 2.5.x as
well:
I think that's a valid analogy. A while ago we were investigating similar
things and also arrived at that analogy. In practice, it turns out that
fused streams, i.e. multiple stream components that run in one actor will
skew measurements.
Not sure how much sense these analogies make when going
I think there are a few things that should be treated separately:
* How to organize bigger route structures?
* How to implement some parts of the route structure as actors?
As Roland wrote before, a Route is just an alias for a function
`RequestContext => Future[RouteResult]`. This gives you
Hi,
that sounds as if you haven't configured Akka HTTP to open an HTTPS server.
It's hard to say though, as you didn't post any code ;) What code did you
use to configure Akka HTTP to use HTTPS?
Johannes
On Tuesday, April 18, 2017 at 2:40:08 PM UTC+2, Abdeali Chandanwala wrote:
>
> Hi
Hi Alan,
On Friday, March 17, 2017 at 12:25:09 PM UTC+1, Alan Burlison wrote:
>
> pathPrefix("root") {
>concat(
> pathPrefix("service1") { service1.route },
> pathPrefix("service2") { service2.route }
>)
> }
>
> That works fine with a path of say "/root/service1", but
Hi Alan,
yes, that's a valid use case. The route tree is usually traversed
completely until a Route matches. But if you know that the search can be
cut short, then `seal` is definitely a good solution. Another solution
would be to use only `handleRejections` which would have the advantage that
Hi Brice,
we previously recommended the stream variant
(`Source.single(request).mapAsync(1))`) but it turned out that this is not
a good idea as materialization has some cost and the API is more complex
than using `Http.singleRequest()` directly. So, using
`Http.singleRequest()` is the right
Hi Dimitry,
that message is a different one. It means that your code is issuing
requests with an explicit header `RawHeader("User-Agent", ...)` somewhere.
Johannes
On Thursday, February 23, 2017 at 8:10:03 PM UTC+1,
dmitriy...@alisagaming.com wrote:
>
> Hi everyone.
>
> I've been using akka
Hello,
the error is thrown
here:
https://github.com/akka/akka/blob/master/akka-stream/src/main/scala/akka/stream/impl/io/TcpStages.scala#L303.
I agree it's hard to see why that error would happen. One thing that could
have happened is that a connection attempt was very slow and so the
Hi,
XMPP is an application-level protocol while TCP is a transport-level
protocol. In fact, XMPP usually runs *on top of* TCP. So, yes, you can use
Akka TCP to implement XMPP.
Comparing XMPP to what you probably have right now is that in your current
application you used your own custom
Hi David,
the 503 is generated by the timeout logic, see the
`akka.http.server.request-timeout` setting. See the timeout directives for
ways to change it based on the
request:
http://doc.akka.io/docs/akka-http/10.0.3/scala/http/routing-dsl/directives/timeout-directives/index.html
Regarding
Yes, this seems to be a bug in sbt-assembly which gets triggered if you use
shading. See https://github.com/sbt/sbt-assembly/issues/205
>
>
--
>> Read the docs: http://akka.io/docs/
>> Check the FAQ:
>> http://doc.akka.io/docs/akka/current/additional/faq.html
Hi Eric,
we'd like to look into that. It looks as if a streams materializer is
holding on to some memory but we need more info to see what it keeps
exactly to.
Is this a reproducible scenario? Could you share the memory dump (in
private) with us? Otherwise, could you send the list of top
On Monday, January 2, 2017 at 1:42:27 PM UTC+1, Alan Burlison wrote:
>
> One question:
>
> headerValueByName("X-Auth-Key")).flatMap { (user, pass) =>
> // <- user flatMap for custom directives
>
> doesn't compile as the resulting value of type Directive[(String,
> String)] doesn't have a
Hi Alan,
here are a few steps to get there:
* provide all your different authentication directives as values of type
`Directive1[T]` with T being the type of the principal found after
successful authentication.
* To build those, use either one of the existing authentication directives
Hi Gaurav,
two small remarks:
* Try using `Http().singleRequest` instead of
`Souce.single().via(poolClientFlow).run`. Materialization (= `run` /
`runWith`) has a certain cost that can be avoided using
`Http().singleRequest`.
* Using `(1 to 1000).par.foreach` to run the benchmark may lead to
Dear hakkers,
We are proud to announce Akka Http 10.0.1, which is the first
maintenance release of the Akka Http 10.0 series, and also our small
holiday present to you.
Some notable improvements and bug fixes are:
Lots of documentation improvements thanks to our awesome contributors.
Improved
Hi Ronny,
I agree that error message is little helpful. Maybe you are sending plain
Strings to the actor created by Source.actorRef? (Still we might want to
improve the error message in that case.)
Johannes
On Monday, September 19, 2016 at 12:08:52 AM UTC+2, Ronny Bräunlich wrote:
>
> Hi
Hi Yaroslav,
you may want to create a custom GraphStage in that case that waits for the
first data to arrive and then prepend the extra-data. Afterwards you can
change the state of the GraphStage to forward messages 1-to-1.
Johannes
On Tuesday, September 20, 2016 at 4:00:23 PM UTC+2, Yaroslav
Hi Simon,
you could also try to split up the asynchronous computation and the actual
filtering like this:
def filterFunc(e: E): Future[Result]
xyz.mapAsync(n)(e => filterFunc(e).map(res => e -> res))
.via(statefulFilterGraphStage)
And then implement the `statefulFilterGraphStage` with
Hi Yaroslav,
no worries, this is a good question. It depends a bit on what kind of logic
you want to trigger. If you just want to do something, you can use
`mapMaterializedValue` to do something with the `OutgoingConnection` object
before materialization is complete.
If you want to prepend
Hi Paul,
a combinator to achieve something like this has been proposed several times
but I think there hasn't been consensus how to implement it exactly. The
latest approach is discussed here:
https://github.com/akka/akka-stream-contrib/issues/50
Johannes
On Thursday, September 15, 2016 at
Hi Kunai,
On Thursday, September 15, 2016 at 7:49:18 AM UTC+2, Kunal Deshpande wrote:
>
> Few questions on back pressure
> 1. While using Flows in akka-streams using .via will a downstream flow
> apply back pressure to a flow upstream or is back pressure only signaled to
> a Source?
>
Hi,
On Tuesday, September 13, 2016 at 4:26:55 PM UTC+2, rrodseth wrote:
> Is this the right way to debug the missing conversion?
> val prm = implicitly[Future[PimpedResult[(StatusCode,
Result[StatusDTO])]] => ToResponseMarshallable]
Try
Hi,
after having a quick look into it (the fully working reproducer was a
great help btw), it turned out that this is indeed a regression
introduced shortly before the release. It is tracked here:
https://github.com/akka/akka/issues/19398
Johannes
On Fri, Jan 8, 2016 at 7:35 PM, Andrey
On Wed, Aug 26, 2015 at 2:02 AM, Simon Schäfer m...@antoras.de wrote:
The only thing I didn't understand was this part:
Source.actorRef(1, OverflowStrategy.fail)
When I replace the 1 with a 0 (which is allowed according to the
documentation) I get this error message: Dropping element
Hi paweł,
yes, that's an unfortunate detail of the Java API. Whatever you put in
there as the Object value it will be ignored. So, to convert your flow
you should be able to use `.ObjectmapMaterializedValue(m - m)` to
convert the type.
Johannes
On Tue, Aug 25, 2015 at 11:19 PM, paweł kamiński
Thanks Adam.
On Tue, Aug 18, 2015 at 4:01 PM, Adam Shannon adam.shan...@banno.com wrote:
FYI If you want to add a class to your application here's an example:
https://gist.github.com/adamdecaf/f83c7000d1b69bc29c4d
On Tue, Aug 18, 2015 at 2:29 AM, Johannes Rudolph
Hi John,
how can we know? You need to share more of your code :)
Johannes
On Tue, Aug 11, 2015 at 11:01 AM, john.vie...@gmail.com wrote:
I am using Akka http Test
HttpRequest request = ...
TestResponse response = appRoute.run(request);
Unfortunatly TestResponse is null.
I think its
Hi Luc,
in the new dump there are again these two connections:
port 59385 - 57649
and
port 59386 - 8080
It looks as if you are running some kind of transparent proxy (a
desktop firewall sofware?) that redirects all the traffic through a
middleman. So, if you look at the connection from 59385
Hi Luc,
in that case, can you provide a more information about the kind of
system you are running on? And it would be nice if we had a
tcpdump/wireshark dump that shows what's going on, so we can see at
least at which side things start to go wrong?
Johannes
On Mon, Jul 27, 2015 at 5:19 PM, Luc
Hi Luc,
I can see the log messages you are seeing:
[DEBUG] [07/27/2015 12:01:01.059]
[api-akka.actor.default-dispatcher-9]
[akka://api/user/$a/flow-47-3-publisherSource-prefixAndTail]
Cancelling akka.stream.impl.MultiStreamOutputProcessor$SubstreamOutput@eda0238
(after: 5000 ms)
but I don't see
On Mon, Jul 20, 2015 at 3:59 PM, Nicolau Werneck nwern...@gmail.com wrote:
It is unfortunate that I missed my error logs, but I'm glad you guys are
taking a look at this!
I confirmed that removing the Host header with curl just gives a 400, so it
can't but just this. I was using 1.0-RC4, so
Hi Adam,
On Mon, Jul 20, 2015 at 3:25 PM, Adam Shannon adam.shan...@banno.com wrote:
All Internet-based HTTP/1.1 servers MUST respond with a 400 (Bad Request)
status code to any HTTP/1.1 request message which lacks a Host header
field.
Is from RFC 2616
I can confirm (the gist above) that
? I
upgraded from M2 to 1.0 this weekend and I never saw the server dying from
these requests.
On Mon, Jul 20, 2015 at 9:26 AM, 'Johannes Rudolph' via Akka User List
akka-user@googlegroups.com wrote:
On Mon, Jul 20, 2015 at 3:59 PM, Nicolau Werneck nwern...@gmail.com
wrote
Hi Jeroen,
On Thu, Jul 16, 2015 at 4:05 PM, Jeroen Rosenberg
jeroen.rosenb...@gmail.com wrote:
Anyways, it's working as expected. I do have one small unrelated issue. The
stream I am consuming is in Gzip format, so I'm unzipping the ByteStrings as
they come in as part of my FlowGraph. However,
Hi Jeroen,
On Thu, Jul 16, 2015 at 4:35 PM, Jeroen Rosenberg
jeroen.rosenb...@gmail.com wrote:
def gunzip(bytes: Array[Byte]) = {
val output = new ByteArrayOutputStream()
FileUtils.copyAll(new GZIPInputStream(new ByteArrayInputStream(bytes)),
output))
output.toString
}
This creates a
Hi Jeroen,
it would be very helpful if you could somehow come up with a
reproducer against some publicly accessible endpoint which would show
the issue. It seemed to work for all the URLs I tested.
Johannes
On Wed, Jul 15, 2015 at 6:25 PM, Jeroen Rosenberg
jeroen.rosenb...@gmail.com wrote:
44 matches
Mail list logo