Adds a new api lws_vhost_destroy(struct lws_vhost *) which allows dynamic removal of vhosts.
The external api calls two parts of internal helpers that get reused for context destroy.
The second part is called deferred by 5s... this is to ensure that event library objects
composed into structs owned by the vhost all have a chance to complete their close
asynchronously. That should happen immediately, but it requires us to return to the
event loop first.
The vhost being removed is deleted from the context vhost list by the first part, and does
not block further removals or creation during the delay for the deferred freeing of the
vhost memory.
Part 1:
- if the vhost owned a listen socket needed by other vhosts listening on same iface + port, the listen
socket is first handed off to another vhost so it stays alive
- all wsi still open on the vhost are forcibly closed (including any listen socket still attached)
- inform all active protocols on the vhost they should destroy themselves
- remove vhost from context vhost list (can no longer be found by incoming connections)
- add to a "being destroyed" context list and schedule the second part to be called in 5s
Part 2:
- remove us from the being destroyed list
- free all allocations owned by the vhost
- zero down the vhost and free the vhost itself
In libwebsockets-test-server, you can send it a SIGUSR1 to have it toggle the creation and destruction of
a second vhost on port + 1.
Basically we support openssl api compatibles only.
If we ever try something different we need a shim making it openssl api or a proper abstraction layer added first.
Special port setting to disable listening for a server using socket adoption.
This contrasts with CONTEXT_PORT_NO_LISTEN which does the same for a client.
In particular, server-side SSL is not disabled by CONTEXT_PORT_NO_LISTEN_SERVER
as it is by CONTEXT_PORT_NO_LISTEN.
https://github.com/warmcat/libwebsockets/issues/489
This
1) fixes the vhost changes on master
2) works around the ah pool changes
3) fixes some other build problems that appeared
4) hacks out physical flow control for internal streams
5) updates the advertised protocol to h2 needed by, eg, chrome 51
That gets it able to serve small (<4K, ie, one packet) files over http2
Signed-off-by: Andy Green <andy@warmcat.com>
polarssl is the old name for mbedtls. Unfortunately the two are confused in eg,
Fedora. For our purposes, polarssl or mbedtls < 2.0 has includes in
/usr/include/polarssl and polarssl_ apis and we call that "polarssl".
polarssl or mbedtls >=2.0 has includes in /usr/include/mbedtls and mbedtls_ apis,
we call that "mbedtls".
This has to be spelled out clearly because eg Fedora has a package "mbedtls" which
is 1.3.x and has the polarssl_ apis and include path. We will deal with that as
"polarssl" despite the package name then.
This patch lets you use LWS_USE_POLARSSL or LWS_USE_MBEDTLS and set the include and
library path like this
cmake .. -DLWS_USE_POLARSSL=1 -DLWS_POLARSSL_INCLUDE_DIRS=/usr/include -DLWS_POLARSSL_LIBRARIES=/usr/lib64/libmbedtls.so.9
This patch adds the cmake support and adapts [private-]libwebsockets.h but doesn't
modify the apis in ssl[-*].c yet.
Signed-off-by: Andy Green <andy@warmcat.com>
Most of ssl.c is under a #ifdef for client or server disabled...
let's get rid of it and have CMake just build the appropriate
files
Signed-off-by: Andy Green <andy@warmcat.com>
