| Age | Commit message (Collapse) | Author |
|---|
|
|
|
Internal use of the function white_space_clean/1 could probably be done
in a much better way using re-module and removing a lot of legacy code.
But we will have to do this later, due to lack of time, we want to make
this commit as little work as possible.
|
|
|
|
These functions should not be used, there exists better
functions in the standard libraries.
|
|
* Remove debug macros that mimics call trace.
* Remove comments that does not add value.
* Structure code
|
|
To enable the HTTP server to run in a virtualized environment,
where there can be more that one server that has the same
bind_address and port, we add a new option profile.
The profile name will be used in concatenation with bind_address and port
to identify the HTTP server instance.
The name profile was chosen as there is a similar concept in the HTTP client
where profile names can be used to instantiate client configurations.
|
|
Conflicts:
OTP_VERSION
lib/inets/test/httpd_SUITE.erl
lib/inets/vsn.mk
lib/ssh/src/ssh.erl
lib/ssh/vsn.mk
lib/ssl/src/ssl.appup.src
lib/ssl/vsn.mk
|
|
Many HTTP headers are optional, and it could be desirable for the
server to filter and maybe even alter them without replacing the
mod_* modules that generate/process them. Add new behaviour
httpd_custom_api with default implementation in httpd_custom.erl.
Add behaviour module in 18 as then we can specify optional callbacks.
|
|
Conflicts:
OTP_VERSION
lib/ssh/doc/src/ssh.xml
lib/ssh/vsn.mk
|
|
|
|
* zandra/mime-types-config/OTP-12731:
make it possible to use a file to configure mime types
|
|
The documentation specifies that a file can be used to configure
mime types, but currently it isn't working. With this change it
works as documented.
|
|
Conflicts:
OTP_VERSION
|
|
When the Server header has empty info (or 'none' in config), it is not generated. This is for limiting Banner Grabbing attempts.
|
|
SSI is an old technique implemented by mod_include, that was badly
tested and not recommended to use, as having the server parse HTML
pages is a double edged sword! It can be costly for a heavily loaded
server to perform parsing of HTML pages while sending
them. Furthermore, it can be considered a security risk to have
average users executing commands in the name of the Erlang node user.
|
|
|
|
|
|
|
|
Gracefully handle invalid content-lenght headers instead of
crashing in list_to_integer.
|
|
|
|
We do not want this timeout to be mixed up with the keep alive timeout.
We will probably want to make it configurable later.
|
|
* ia/inets/maint-17/string-lengths/OTP-11925:
inets: httpd - Behave well on not so long but wrong request lines
inets: httpc - update test suite framwork to adopt to change in httpd
inets: httpd - Reject incorrect large request lines early
|
|
config_reload tests are not enough to test httpd "block/unblock"
used by config_reload, so renable some test cases in the old_httpd_SUITE
after making them work with the new code. Should be cleaned up and moved
to httpd_SUITE.
|
|
|
|
|
|
|
|
|
|
Conflicts:
lib/inets/src/http_server/httpd_manager.erl
|
|
If you need a timeout between client and server use a server side
timeout instead of the default gen_server:call/2 timeout that is quite
useless. Also remove legacy code that has no further use.
|
|
If a request handler terminates abnormally (terminate reason not
normal), the error-log entry was a bit obtuse ("Internal Server Error").
This has been improved so that the log entry now also include the
actual terminate reason.
|
|
Sometimes the size of a response could be as a string.
|
|
Remove logging when fo keep-alive connection timeout. This is
a normal event (feature) not an error.
|
|
|
|
This fixes the usage of the httpd configuration option 'script_timeout',
which got ignored before.
The documentation states that the value is in seconds, which was true
when using the Apache like configuration file, but not true when using
the proplist style configuration.
|
|
The documentation states that the value is in seconds, which was true
when using the Apache like configuration file, but not true when using
the proplist style configuration.
|
|
|
|
Conflicts:
lib/inets/doc/src/notes.xml
lib/inets/src/http_lib/http_transport.erl
lib/inets/src/inets_app/inets.appup.src
lib/inets/vsn.mk
lib/ssl/doc/src/notes.xml
lib/ssl/src/ssl.appup.src
lib/ssl/src/ssl.erl
lib/ssl/src/ssl_internal.hrl
lib/ssl/src/tls_connection.erl
lib/ssl/vsn.mk
|
|
* jw/fix-httpd-erl-script-nocache/OTP-11260:
inets: added simple testcase for option erl_script_nocache
Fix httpd config option 'erl_script_nocache'
|
|
added as option to inets conf file
|
|
Conflicts:
lib/inets/doc/src/notes.xml
lib/inets/src/inets_app/inets.appup.src
lib/inets/vsn.mk
lib/ssh/doc/src/notes.xml
lib/ssh/src/ssh.appup.src
lib/ssh/src/ssh_connection_handler.erl
lib/ssh/vsn.mk
|
|
|
|
This fixes the usage of the httpd configuration option
'erl_script_nocache', which got ignored before.
Attention: Before, the option 'script_nocache' also changed the
behavior for mod_esi, now it only affects mod_cgi.
|
|
|
|
|
|
httpc: CTfy test suite
httpd: Simplify ssl configuration OTP-10846
|
|
OTP-10844
|
|
|
|
|
|
* maint:
Increased waiting time to check if socket is down
Fine adjustments of slowdos alg
Changes to slowdos testcase
Doc for test case slowdose
Bumped version nr
Added testcases för minimum_bytes_per_second option
Reduced checking of bytes per second to every second
Documentation for minimum_bytes_per_seconds option
Changed property to minimum_bytes_per_second and checking if is integer on start
Changed to Bytes Per Second when checking clients rate, new options {byte_limit, integer()}
Option {bit_limit, integer()} to close sockets where clients are too slow
|
|
|
