(expect < 256 octets for DNS name)
Since limit is imposed, can use buffer_clear() instead of buffer_reset()
for con->uri.authority and con->server_name. (Also, con->uri.scheme is
limited to "http" and "https", so use buffer_clear() for it, too)
quickly clear buffer instead of buffer_string_set_length(b, 0) or
buffer_reset(b). Avoids free() of large buffers about to be reused,
or buffers that are module-scoped, persistent, and reused.
(buffer_reset() should still be used with buffers in connection *con
when the data in the buffers is supplied by external, untrusted source)
separate reading header from connection_handle_read_state() into
connection_read_header() so that HTTP pipelined requests can skip
optimistic read()/recv()
connection_reset() is called at end of request to prepare for keep-alive
or to free resources while waiting for client to close connection.
This is the reason why it is called earlier towards end of request
rather than always waiting until the very end of request when socket
is closed.
connection_reset() used to additionally be called for every new socket
connection, but now is only called when new conns are allocated and
initialized.
This use is not recommended since it means that the response body
will not contain the precise error code.
x-ref:
"What is the proper syntax for server.error-handler in 1.4.45"
https://redmine.lighttpd.net/boards/2/topics/8320
provide standard types in first.h instead of base.h
provide lighttpd types in base_decls.h instead of settings.h
reduce headers exposed by headers for core data structures
do not expose <pcre.h> or <stdlib.h> in headers
move stat_cache_entry to stat_cache.h
reduce use of "server.h" and "base.h" in headers
fdevent_accept_listenfd() now always returns fd O_NONBLOCK O_CLOEXEC
for consistency, rather than setting elsewhere in connection_accepted()
Handle older Linux 2.6 kernels which might have accept4() in glibc,
but return ENOSYS, as accept4() was not added until Linux kernel 2.6.28.
discard from socket using recv MSG_TRUNC on Linux TCP SOCK_STREAM socket
Currently, lighttpd supports only TCP SOCK_STREAM. If UDP SOCK_DGRAM
were to be supported in the future, then socket type will need to be
stored so that MSG_TRUNC is used appropriately for the desired effect.
To find out socket type on arbitrary socket fd:
getsockopt(..., SOL_SOCKET, SO_TYPE, ...)
but better to store it with each listening socket.
permit LF to end header lines if
server.http-parseopt-header-strict = "disable"
(instead of requiring CR LF)
(makes it easy to use 'openssl s_client -connect <IP:port>' on unix)
(Note: care taken to minimize diff in this commit,
but header parsing code should be revisited and overhauled)
new directive server.error-intercept = [ "enable" | "disable" ]
to intercept 4xx and 5xx responses from dynamic handlers
(e.g. CGI, FastCGI, SCGI, proxy)
Intercepted HTTP error status are then handled by one of
server.error-handler
server.error-handler-404
server.errorfile-prefix
(if configured)
Do not use server.error-intercept with locations handled by mod_webdav!
x-ref:
"would like something similar to nginx proxy_intercept_errors"
https://redmine.lighttpd.net/issues/974
It is still not a good idea for backend to send Transfer-Encoding unless
backend is mod_proxy, and mod_proxy should not currently receive chunked
response since mod_proxy sends HTTP/1.0 request.
If mod_proxy is changed to sent HTTP/1.1 request, then lighttpd would
need to check if client is HTTP/1.0 and would need to de-chunk and
remove any other transfer-codings if not supported by next-hop.
x-ref:
"error 500 (mod_cgi.c.601) cgi died"
https://redmine.lighttpd.net/issues/2786
more consistent cleanup of resources at shutdown
(e.g. upon error conditions)
Notes: graceful restart with SIGUSR1
- not available if chroot()ed, oneshot mode, or if idle timeout occurs
- preserve process id (pid)
- preserve existing listen sockets
- i.e. does not close old listen sockets from prior configs
(even if old listen sockets no longer in the new config)
(sockets may have been bound w/ root privileges no longer available)
- will fail to add listen sockets from new config if privileges
lighttpd configured to drop privileges to non-root user, and
new listen socket attempts to bind to low-numbered port requiring
root privileges.
- will fail if listen sockets in new config conflict with any previous
old listen sockets
- These failure modes will result in lighttpd shutting down instead of
graceful restart. These failure modes are not detectable with
preflight checks ('lighttpd -tt -f lighttpd.conf') because the
new instance of lighttpd running the preflight check does not
known config state of n prior graceful restarts, or even the
config state of the currently running lighttpd server.
- due to lighttpd feature of optionally managing backends
(e.g. fastcgi and scgi via "bin-path"), lighttpd must wait for
all child processes to exit prior to restarting. Restarting new
workers while old workers (and old backends) were still running would
result in failure of restarted lighttpd process to be able to bind to
sockets already in use by old backends (e.g. unix "socket" path)
x-ref:
"graceful restart with SIGUSR1"
https://redmine.lighttpd.net/issues/2785
(take 2:
relocate module cleanup check to after handle_connection_close hook)
modules may now keep state for the lifetime of a connection,
rather than being required to be reset after every request (when
there can be multiple keep-alive requests on the same connection)
