We happily announce the immediate availability of nghttp2 v0.5.1.
The supported HTTP/2 protocol remains to h2-13.
This release fixes HPACK integer decoding bug, which occurs when encoded integer byte sequence crosses frame boundary or packet.
We happily announce the immediate availability of nghttp2 v0.5.1.
The supported HTTP/2 protocol remains to h2-13.
This release fixes HPACK integer decoding bug, which occurs when encoded integer byte sequence crosses frame boundary or packet.
We happily announce the immediate availability of nghttp2 v0.5.0.
The supported HTTP/2 protocol is now h2-13.
The changes since h2-12 were described in the previous post. We
still supports ALTSVC frame in this release, but it may be removed in
the future release since it is now in the separate document. The
BLOCKED frame was completely removed from the source code.
nghttp2 was updated to HTTP/2 draft-13 and HPACK draft-08.
The major changes are:
The https endpoint for nghttp2.org now requires TLSv1.2 and DHE or EDCHE with GCM cipher suite for HTTP/2 connection. If HTTP/2 was negotiated and these requirements are not met, connection error will be issued with the error code INADEQUATE_SECURITY.
nghttp2 v0.4.1 just released today. It is mainly a bug fix release.
The supported HTTP/2 protocol version remains h2-12.
Although this release is tagged as bug fix release, nghttpx got some
improvements. Firstly, --npn-list option now works with ALPN.
Secondly, * is officially allowed as <HOST> parameter in
--frontend option and it means wildcard address, which can bind to
both IPv4 and IPv6 addresses.
httpbin is a HTTP Request & Response Service developped by A Kenneth Reitz Project. Currently it only serves in HTTP/1.1.
nghttp2.org now serves its functionality in HTTP/2 and SPDY. You can access the service here.
Currently, some json elements (e.g., url) are a bit strange due to the fact that httpbin service is reverse proxied.
nghttp HTTP/2 client included in nghttp2 has a nice feature to show the incoming and outgoing HTTP/2 frames. It is extremely usable for debugging.
Here is a snippet of the log nghttp produced today:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 | |
When you see this, you may wonder why BLOCKED frame recieved at t=0.475 was followed by DATA frame. BLOCKED with stream_id=0 means connection level window size is depleted, and server is unable to send more DATA until it receives WINDOW_UPDATE frame from client. There is no WINDOW_UPDATE between t=0.475 and t=0.488, so you may think this must be a mistake, server has a bug.
Of course not. WINDOW_UPDATE with stream_id=0 was already sent t=0.430. But due to latency, before it reached at the server, server used all available window size and sent BLOCKED frame. After that, the server finally received the WINDOW_UPDATE from client and update its window size and started to send DATA frame again.

We released new version v0.4.0 just now. Quick summary of this release:
The dependency based prioritization was first introduced in HTTP/2 draft-11 and further refined in HTTP/2 draft-12, which is the latest draft version at the of this writing. The draft describes its mechanism and requirements for client and server in 5.3. Stream priority in detail. In short, dependency based prioritization works like this:
A stream can depend on another stream. If stream B depends on stream A, stream B is not processed unless stream A is closed or stream A cannot progress due to, for example, flow control or data is not available from backend content server. These dependency links form a tree, which is called dependency tree (circular dependency is not allowed).
Dependency has weight. This is used to determine how much available resource (e.g., bandwidth) is allocated to a stream.
Quoted from 5.3.2. Dependency weighting:
Streams with the same dependencies SHOULD be allocated resources proportionally based on their weight. Thus, if stream B depends on stream A with weight 4, and C depends on stream A with weight 12, and if no progress can be made on A, stream B ideally receives one third of the resources allocated to stream C.
We do not describe about weight further in this post. We focus about stream dependency part and assumes all weightings are equal.
We first have to say that prioritization in HTTP/2 is completely optional feature. Client can freely provide prioritization information to a server, but server has a choice to ignore them. Simple server implementation may ignore prioritization altogether.
So how is this mechanism used for our Web pages? When loading a typical Web page, client first requests HTML file. It then parses received portion of HTML file and finds the links to resources, such as CSS, Javascript and images, and issues requests to get these resources as well. Suppose that client wants HTML in highest priority, since it is the main page to show. Then it wants CSS or Javascript in medium priority. Images are in lowest priority. These requirement can be expressed as dependency: CSSs and Javascripts depend on a HTML file. Images depend on CSSs or Javascripts files. Providing these prioritization information to the server, client can load resoures in opitimal order.
nghttp2 fully implements prioritization. So let’s see how the prioritization works in the real use case. We use nghttp2 documentation index.html generated by sphinx (the same page is available at https://nghttp2.org/documentation/) as a test page. That page contains links to the followings resources:
We use nghttp command-line client with -a option. With -a option,
it also downloads links found in HTML page it is downloading. It is
programmed to categorize resources in the following priority levels.
Hopefully, we’d like to build dependency tree like this:

We use following command-line to run nghttp client:
1
| |
With -v option, we can see what happens in HTTP/2 frameing layer.
First, nghttp requested index.html to the server:
1 2 3 4 5 6 7 8 9 10 11 | |
index.html is stream 1 (see stream_id=1). index.html does not depend any streams since this is the first stream ever in this connection. Then server responded with HTTP response header in HEADERS frame and its response body in DATA frames:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 | |
END_STREAM means that content of stream 1 was completely received.
nghttp parsed received HTML in DATA frame found that links to
resources. First it requested 3 Javascript files:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 | |
There is priority information in each request HEADERS. For example,
stream requesting jquery.js has stream_id=1, weight=16, exclusive=0,
which means it depend on stream 1 with weight 16. exclusive=0 means
that its dependency is not exclusive, so if there are any dependencies
to a designated stream, new stream joins existing siblings. We’ll see
the example of exclusive=1 case soon. The other 2 requests also
depend on stream 1. At this moment, dependency tree became like this:

Then nghttp found CSS link and issued request:
1 2 3 4 5 6 7 8 9 10 11 | |
Here we have stream_id=1, weight=16, exclusive=1. This is a bit
different from the previous priority information. This time we have
exclusive=1, which means that stream 9 solely depends on stream 1
and the streams which formerly depend on stream 1 depend on stream 9.
So resulting dependency tree became like this:

The last resource was theme.js:
1 2 3 4 5 6 7 8 9 10 11 | |
The priority information for this request was stream_id=9, weight=16,
exclusive=0. The stream 9 was theme.css. So unlike the first 3
Javascript files, this request directly specified the dependency to
stream 9. Finally the dependency tree became like this:

This completely reflects the priority levels nghttp client implements.
Did the server respect this prioritization? Let’s see the DATA flow of these streams. Since stream 1 was already finished, stream 9 was the highest priority. The log shows that server correctly sent its DATA first:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 | |
You will notice that there are stream 3, 5, 7 and 11 interleaved before stream 9 got finished. This is because stream 9 could not make progress because of flow control. You see BLOCKED frame for stream 9 in the above log. The progress of stream 9 was blocked until WINDOW_UPDATE frame for stream 9 was arrived to the server. While stream 9 was blocked, streams which depend on stream 9 were unblocked and started to send its DATA frames. After the server received WINDOW_UPDATE frame for stream 9, it started to send DATA frame of stream 9 again and stream 3, 5, 7 and 11 were blocked.
After stream 9 was finished, the remaining streams had equal priority, so they were sent interleaved:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 | |
We updated test server to the latest HTTP/2 draft-12 implementation.
As before, we service h2-12 at port 443 and h2c-12 at port 80. We
also continue to support SPDY and HTTP/1. If http URI is used to
access to this site, we issue ALTSVC frame and Alt-Svc header field to
indicate that we also service same contents at port 443 in h2-12
protocol.
Let’s review the highlights of the changes in draft-12. The
dependency based priority was introduced in draft-11. In draft-12, it
is further enhanced (and simplified). nghttp2 fully implements
dependency based priority mechanism. From our experience, we know
that it is crucial to retain information of closed streams to make the
prioritization work. This is also covered in the draft-12. nghttp2
retains closed streams as much as possible. The upper bound of the
sum of active and retained closed streams are
SETTINGS_MAX_CONCURRENT_STREAMS the server declares.
BLOCKED frame is newly introduced in draft-12. It indicates that DATA frame transfer is blocked due to exhaustion of flow control send window. nghttp2 implements transmission and reception of BLOCKED frame for both connection and stream level flow control.
Compressed DATA frame is introduced in draft-12. This is supposed to
be a replacement of Transfer-Encoding: gzip in HTTP/1. nghttp2
supports compressed DATA frame in its library API. But the public
test server does not support compress DATA frame at the moment.
To access this site via HTTP/2, you can use nghttp client included in nghttp2 repository. It works with both https and http URI. node-http2 client may also work.
The easiest way to test HTTP/2 is use h2-12 enabled Firefox browser.
You can find the download link from
here. Don’t forget to
set network.http.spdy.enabled.http2draft and
security.ssl.enable_alpn to true in about:config screen. Please
note that Firefox only supports HTTP/2 in https URI.
nghttp2.org was approved as OSS project by GlobalSign and they offered
us wild card certificate for free. Now
https://nghttp2.org has valid SSL/TLS certificate. No more red screen
and no more -k option is required to access this site. Thank you to
GlobalSign for supporting open source community.