| Age | Commit message (Collapse) | Author |
|---|
|
|
|
The approach taken here is very similar to what browsers are
doing. A separate pool is created for each host/port/scope.
The authority (host header) is used to determine which pool
will execute requests. A connection process is semi-randomly
chosen, from the connections that have capacity. Maximum
capacity is determined by the protocol (the HTTP/2 setting
set by the server is used, for example). Multiple processes
can process requests/responses on the same connection
concurrently. There is no need to "give back" the response
to the pool, the number of ongoing streams is maintained via
an event handler.
The implementation is currently not strict, there may be
more attempts to create requests than there is capacity.
I'm not sure if it should be made strict or if Gun should
just wait before sending requests (it only matters in the
HTTP/2 case at the moment).
When there is no connection with capacity available in the
pool (because they have too many streams, or are reconnecting,
or any other reason), checking out fails. There is no timeout
to wait for a connection to be available. On the other hand
the checkout_retry option allows setting multiple timeouts
to retry checking out a connection. Each retry attempt's
wait time can have a different value.
The initial implementation of this work was sponsored by
Kobil and made at the suggestion of Ilya Khaprov.
|
|
|
|
|
|
|
|
I've made some more test timeouts infinity in order
to fix additional intermittent issues that popped up.
|
|
|
|
|
|
If a request/headers/connect/ws_upgrade is created when a connection is in state
'closing', such as after receiving an HTTP/2 GOAWAY frame or an HTTP/1.1
"Connection: close" header, an error message is sent back to the caller
immediately, if reconnect is off (that is if the option retry is set to 0).
This allows an application to retry the request on another connection without
waiting for all streams on the current connection to complete.
|
|
We are about one week before the actual release.
|
|
|
|
|
|
Switching from /2 to /3 should be easy enough.
Also update the documentation about HTTP/2 Websocket support.
|
|
HTTP/1.1 Upgrade to HTTP/2 will not be implemented. There are
discussions for this functionality to be removed from the
HTTP/2 spec.
HTTP/1.1 Upgrade to TLS will most likely not be implemented.
|
|
|
|
|
|
|
|
It was missing the StreamRef from the message.
|
|
Cowlib master has the fix for this test. Temporarily depend
on Cowlib master until a release is made.
|
|
|
|
The main source file for the test tool h2specd was moved a
time ago, but correct one is used now.
Also make sure that gun sends the GOAWAY directly to avoid
timing issues in test 6.9.1 in h2specd_SUITE.
|
|
|
|
|
|
While the stream has not ended, the response has.
|
|
|
|
The response ends when the tunnel is established, even if the
stream itself does not. The data coming in on the stream after
is not part of the response.
This makes both HTTP/1.1 and HTTP/2 send 'fin' to successful
CONNECT responses.
|
|
|
|
The cookie_ignore_informational has been moved to http_opts and
http2_opts.
Also fix an issue when using 'protocols' in gun:open.
When connecting via TLS the protocol's options were
discarded.
|
|
There is a race condition: After receiving data, before setting the
socket back to {active, once}, the server may have closed the socket
or an error may have occurred. Since the socket was in passive mode,
no message is received. Handling the return value of setopts solves
the issue.
Amended by Loïc Hoguin to handle the return value from the active/1
function call.
|
|
There are still small issues left to fix. In particular the
set_cookie command should be replaced with doing the same
in the protocol itself so that the scheme is correct. So
CookieStore must be propagated to all callbacks.
|
|
|
|
It doesn't provide any new information compared to the other
events.
|
|
|
|
This temporarily depends on Cowlib master.
|
|
|
|
For HTTP/2 some servers are not fond of receiving pings,
especially if there's no stream at the time. (particularly
gRPC servers). This is an odd decision for sure. Rather
than trying to accomodate for their logic the ping is no
longer sent by default.
For Websocket sending a ping is a new feature in Gun 2.0
so I'm just being conservative there.
|
|
|
|
|
|
In the documentation headers passed to ws_upgrade are typed as
gun:req_headers(), however if a map of headers is passed (which is
allowed by the type), the gun process will crash when trying to operate
on it as if it were a list.
|
|
While most of this functionality isn't implemented this is not
a reason to let them return invalid values.
|
|
|
|
|
|
The test suite is 216 tests with a tunnel created via two
proxies leading to one origin server. The tests are for
example socks5_h2_https where socks5 identifies the first
SOCKS5 proxy, h2 the second HTTP/2 CONNECT proxy and https
the secure HTTP/1.1 origin server.
The test not only sets up the tunnel and does a request
(or sends/receives data in the case of raw origin servers)
but also confirms that the stream_info and info data is
correct.
|
|
|
|
|
|
|
|
|
|
Implements gun:connect with a tunnel.
|
|
|
|
|
