diff options
author | Andrej Shadura <andrew.shadura@collabora.co.uk> | 2019-09-12 18:36:57 +0200 |
---|---|---|
committer | Andrej Shadura <andrew.shadura@collabora.co.uk> | 2019-09-12 18:36:57 +0200 |
commit | 3e1cfd5d7e3201abbcf38dcda8f4164f7ba6f437 (patch) | |
tree | d4a16f45901ca2b13cc54fb338c3725ea4af5688 /docs | |
parent | 79c99297348da2ba2b4f0cd33022925b15641f7d (diff) |
New upstream version 1.3.1
Diffstat (limited to 'docs')
-rw-r--r-- | docs/api.rst | 2 | ||||
-rw-r--r-- | docs/arguments.rst | 11 | ||||
-rw-r--r-- | docs/glossary.rst | 2 | ||||
-rw-r--r-- | docs/reverse-proxy.rst | 8 | ||||
-rw-r--r-- | docs/runner.rst | 9 |
5 files changed, 24 insertions, 8 deletions
diff --git a/docs/api.rst b/docs/api.rst index 5e0a523..a921a1b 100644 --- a/docs/api.rst +++ b/docs/api.rst @@ -5,6 +5,6 @@ .. module:: waitress -.. function:: serve(app, listen='0.0.0.0:8080', unix_socket=None, unix_socket_perms='600', threads=4, url_scheme='http', url_prefix='', ident='waitress', backlog=1204, recv_bytes=8192, send_bytes=18000, outbuf_overflow=104856, inbuf_overflow=52488, connection_limit=1000, cleanup_interval=30, channel_timeout=120, log_socket_errors=True, max_request_header_size=262144, max_request_body_size=1073741824, expose_tracebacks=False) +.. function:: serve(app, listen='0.0.0.0:8080', unix_socket=None, unix_socket_perms='600', threads=4, url_scheme='http', url_prefix='', ident='waitress', backlog=1204, recv_bytes=8192, send_bytes=1, outbuf_overflow=104856, outbuf_high_watermark=16777216, inbuf_overflow=52488, connection_limit=1000, cleanup_interval=30, channel_timeout=120, log_socket_errors=True, max_request_header_size=262144, max_request_body_size=1073741824, expose_tracebacks=False) See :ref:`arguments` for more information. diff --git a/docs/arguments.rst b/docs/arguments.rst index 2665087..22535a8 100644 --- a/docs/arguments.rst +++ b/docs/arguments.rst @@ -190,7 +190,9 @@ send_bytes Linux, ``/proc/sys/net/ipv4/tcp_wmem`` controls the minimum, default, and maximum sizes of TCP write buffers. - Default: ``18000`` + Default: ``1`` + + .. deprecated:: 1.3 outbuf_overflow A tempfile should be created if the pending output is larger than @@ -198,6 +200,13 @@ outbuf_overflow Default: ``1048576`` (1MB) +outbuf_high_watermark + The app_iter will pause when pending output is larger than this value + and will resume once enough data is written to the socket to fall below + this threshold. + + Default: ``16777216`` (16MB) + inbuf_overflow A tempfile should be created if the pending input is larger than inbuf_overflow, which is measured in bytes. The default is conservative. diff --git a/docs/glossary.rst b/docs/glossary.rst index d87cc31..5309845 100644 --- a/docs/glossary.rst +++ b/docs/glossary.rst @@ -8,7 +8,7 @@ Glossary PasteDeploy A system for configuration of WSGI web components in declarative ``.ini`` format. - See https://web.archive.org/web/20161029233359/http://pythonpaste.org/deploy/. + See https://docs.pylonsproject.org/projects/pastedeploy/en/latest/. asyncore A Python standard library module for asynchronous communications. See :mod:`asyncore`. diff --git a/docs/reverse-proxy.rst b/docs/reverse-proxy.rst index 7081fda..6490e3d 100644 --- a/docs/reverse-proxy.rst +++ b/docs/reverse-proxy.rst @@ -9,13 +9,13 @@ Using Behind a Reverse Proxy Often people will set up "pure Python" web servers behind reverse proxies, especially if they need TLS support (Waitress does not natively support TLS). Even if you don't need TLS support, it's not uncommon to see Waitress and -other pure-Python web servers set up to "live" behind a reverse proxy; these -proxies often have lots of useful deployment knobs. +other pure-Python web servers set up to only handle requests behind a reverse proxy; +these proxies often have lots of useful deployment knobs. If you're using Waitress behind a reverse proxy, you'll almost always want your reverse proxy to pass along the ``Host`` header sent by the client to Waitress, in either case, as it will be used by most applications to generate -correct URLs. You may also use the proxy headers if passing the Host directly +correct URLs. You may also use the proxy headers if passing ``Host`` directly is not possible, or there are multiple proxies involved. For example, when using nginx as a reverse proxy, you might add the following @@ -83,7 +83,7 @@ You will also want to add to Apache:: Configure waitress's ``trusted_proxy_headers`` as appropriate:: - trusted_proxy_headers = "x-forwarded-for, x-forwarded-host, x-forwarded-proto, x-forwarded-port" + trusted_proxy_headers = "x-forwarded-for x-forwarded-host x-forwarded-proto x-forwarded-port" At this point waitress will set up the WSGI environment using the information specified in the trusted proxy headers. This will setup the following diff --git a/docs/runner.rst b/docs/runner.rst index fc424ad..2776e44 100644 --- a/docs/runner.rst +++ b/docs/runner.rst @@ -143,13 +143,20 @@ Tuning options: 8192. ``--send-bytes=INT`` - Number of bytes to send to socket.send(). Default is 18000. + Number of bytes to send to socket.send(). Default is 1. Multiples of 9000 should avoid partly-filled TCP packets. + .. deprecated:: 1.3 + ``--outbuf-overflow=INT`` A temporary file should be created if the pending output is larger than this. Default is 1048576 (1MB). +``--outbuf-high-watermark=INT`` + The app_iter will pause when pending output is larger than this value + and will resume once enough data is written to the socket to fall below + this threshold. Default is 16777216 (16MB). + ``--inbuf-overflow=INT`` A temporary file should be created if the pending input is larger than this. Default is 524288 (512KB). |