Hi Richard, > 11 feb 2015 kl. 11:30 skrev Richard Bradley > <[email protected]>: > > We are now running up against this issue in practice, so we need a workaround. > > To recap: if a client makes a large upload request to an Akka HTTP > ("akka-stream-and-http-experimental-1.0-M2") and the server responds (i.e. > returns an HttpResponse to the "handlerFlow" function) before reading the > full request body, then the TCP stream stalls. > The client receives the response, but thinks that the stream is still valid. > It will then send the next HTTP request on the same TCP stream, but Akka will > never read the request, as it’s still waiting for the (now aborted) request > stream to finish reading in the responded-to request. > > So, for our Akka HTTP server to be well behaved, we must either always read > in the full request and/or half-close the TCP upload stream.
As has been discussed earlier, the only reliable way to treat this case is to fully close the connection, I’m not sure that half-closing will improve the situation. You may or may not try to emit a response before closing, but as the standard does not mandate that the client read the response before the request body has been delivered this cannot really be guaranteed to work. Another way to tackle this problem would be to require the use of an `Expect: 100-continue` header, which would allow exactly the early response that you want to give. > > How do I do those two things? > > I'd like to do this in my my routing layer, in case there is an exception in > the processing layer. > > How can I detect whether the request stream needs cancelling (i.e. whether it > has not been fully read)? Or is it safe to always cancel? Yes, canceling is the way of saying “I am no longer interested”. > > How can I "cancel" the stream? Given an instance of HttpRequest, the entity > has “dataBytes”, but that’s not a materialized stream, but a Source. How do I > get a reference to the materialized stream to cancel it? I tried > "request.entity.dataBytes.runWith(Sink.cancelled)", but that seems wrong. No, that is exactly right. > > Does cancelling the stream half-close the TCP upload stream? If not, how do I > do that? (Send Akka a "connection: close" header on the response? Again -- > how do I detect when that's needed, i.e. when the request has not been fully > read?) Currently canceling of the kind of stream-of-stream setup that HTTP uses does not work correctly, ticket is here <https://github.com/akka/akka/issues/16869>. I think if cancelation is properly propagated upstream then all the right things will happen. Regards, Roland > > (see also https://github.com/akka/akka/issues/16510 ) > > Many thanks, > > > Rich > > > On Friday, January 30, 2015 at 10:15:10 AM UTC, Björn Antonsson wrote: > Hi Joe, > > If you properly cancel the request stream I don't think that you need to > consume it with a black hole sink. > > B/ > > On 29 January 2015 at 11:30:32, Joe Edwards ([email protected] > <javascript:>) wrote: > >> Thanks for the explanation. >> >> Unfortunately I don't control the clients so it sounds like we have to send >> a 'Connection: Close' header and hope the client aborts the request for us, >> otherwise we have to read (and discard) the whole request to ensure the >> client will read the response. That's a bit of a pain, but I suppose it's >> really the client's problem if they insist on ignoring the response until >> they've uploaded their whole 2GB message. >> >> Re: akka-http Is it true once the stream processing the request is cancelled >> we still need to attach a (blackhole) sink to the request stream, otherwise >> the back-pressure may simply stall the client indefinitely? >> >> On Wednesday, 28 January 2015 19:33:58 UTC, Daniel Armak wrote: >> I can't comment on akka-http, but here's my understanding of the HTTP spec. >> >> It's certainly legal to send a response before receiving the request entity, >> but the client isn't obligated to start reading the response before it >> finishes sending the request. And if the server closes the connection >> without reading the whole request, the client may not see the response, >> depending on whether it fit into its receive buffer and on how it's written. >> If you just close the connection, the client might think it a network error >> and retry the request. So if you want to guarantee the client sees the >> rejection response, you have to read and discard the whole request. >> >> If you control the client as well as the server, and the server can reject >> the request based on its headers, the standard HTTP solution is to send >> Expect: 100-continue in the request, and then either 100 Continue or a 4xx >> rejection from the server. But if the server can only reject the request >> based on some prefix of the entity, HTTP provides no way to abort without >> closing the connection. You could cheat though: use chunked encoding for the >> request entity, and send an empty chunk ending the request entity as soon as >> you read the server's rejection headers. >> >> On Wednesday, January 28, 2015 at 6:33:47 PM UTC+2, Joe Edwards wrote: >> I have an application in which clients can upload large request entities to >> the server. In some situations the request will be rejected early (before >> the request body is totally consumed, as this may be hundreds of MB). >> >> Is there a standard way of handling this situation? Since the rejections may >> happen asynchronously, I can't guarantee how much (if any) of the request >> body has been read. >> >> As far as I can tell from RFC 7230 section 6.5 >> If the request is still in flight, we should respond early with a >> 'Connection: Close' (and the request stream will die as the connection >> closes). >> If the request has been totally received, we can respond normally and reuse >> the connection. >> Is there any easy way to determine when a response is being sent early? >> >> More generally, is it even legal to send a response before the entire >> request is received without closing the connection? I can't really tell from >> the spec, but if not it would make much more sense for akka-http to close >> the connection for us. >> >> Thanks, Joe >> -- >> >>>>>>>>>> Read the docs: http://akka.io/docs/ <http://akka.io/docs/> >> >>>>>>>>>> Check the FAQ: >> >>>>>>>>>> http://doc.akka.io/docs/akka/current/additional/faq.html >> >>>>>>>>>> <http://doc.akka.io/docs/akka/current/additional/faq.html> >> >>>>>>>>>> Search the archives: https://groups.google.com/group/akka-user >> >>>>>>>>>> <https://groups.google.com/group/akka-user> >> --- >> You received this message because you are subscribed to the Google Groups >> "Akka User List" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected] <javascript:>. >> To post to this group, send email to [email protected] <javascript:>. >> Visit this group at http://groups.google.com/group/akka-user >> <http://groups.google.com/group/akka-user>. >> For more options, visit https://groups.google.com/d/optout >> <https://groups.google.com/d/optout>. > > -- > Björn Antonsson > Typesafe <http://typesafe.com/> – Reactive Apps on the JVM > twitter: @bantonsson <http://twitter.com/#!/bantonsson> > > > -- > >>>>>>>>>> Read the docs: http://akka.io/docs/ <http://akka.io/docs/> > >>>>>>>>>> Check the FAQ: > >>>>>>>>>> http://doc.akka.io/docs/akka/current/additional/faq.html > >>>>>>>>>> <http://doc.akka.io/docs/akka/current/additional/faq.html> > >>>>>>>>>> Search the archives: https://groups.google.com/group/akka-user > >>>>>>>>>> <https://groups.google.com/group/akka-user> > --- > You received this message because you are subscribed to the Google Groups > "Akka User List" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected] > <mailto:[email protected]>. > To post to this group, send email to [email protected] > <mailto:[email protected]>. > Visit this group at http://groups.google.com/group/akka-user > <http://groups.google.com/group/akka-user>. > For more options, visit https://groups.google.com/d/optout > <https://groups.google.com/d/optout>. Dr. Roland Kuhn Akka Tech Lead Typesafe <http://typesafe.com/> – Reactive apps on the JVM. twitter: @rolandkuhn <http://twitter.com/#!/rolandkuhn> -- >>>>>>>>>> Read the docs: http://akka.io/docs/ >>>>>>>>>> Check the FAQ: >>>>>>>>>> http://doc.akka.io/docs/akka/current/additional/faq.html >>>>>>>>>> Search the archives: https://groups.google.com/group/akka-user --- You received this message because you are subscribed to the Google Groups "Akka User List" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/akka-user. For more options, visit https://groups.google.com/d/optout.
