walkero
/
nghttp2
1
0
Fork 0
Code Issues Pull Requests Projects Releases 1 Wiki Activity
nghttp2/doc/h2load.h2r

65 lines
2.0 KiB
Plaintext
Raw Normal View History

doc: Add output section to h2load man page
2015-02-01 08:36:58 +01:00
OUTPUT
------
requests
total
The number of requests h2load was instructed to make.
started
The number of requests h2load has started.
done
The number of requests completed.
succeeded
The number of requests completed successfully. Only HTTP status
code 2xx or3xx are considered as success.
failed
The number of requests failed, including HTTP level failures
(non-successful HTTP status code).
errored
The number of requests failed, except for HTTP level failures.
status code. This is the subset of the number reported in
``failed`` and most likely the network level failures or stream
was reset by RST_STREAM.
status codes
The number of status code h2load received.
traffic
total
The number of bytes received from the server "on the wire". If
requests were made via TLS, this value is the number of decrpyted
bytes.
headers
The number of response header bytes from the server without
decompression. For HTTP/2, this is the sum of the payload of
HEADERS frame. For SPDY, this is the sum of the payload of
SYN_REPLY frame.
data
The number of response body bytes received from the server.
time for request
min
The minimum time taken for request and response.
max
The maximum time taken for request and response.
mean
The mean time taken for request and response.
sd
The standard deviation of the time for request and response.
+/- sd
The fraction of the number of requests within standard deviation
range (mean +/- sd) against total number of successful requests.
h2load: Effectively disable flow control by setting large window size Previously h2load used default flow control window as described in HTTP/2 and SPDY specification. The window size is 64KiB, which is a bit small, and cannot utilize full server performance when response size is not too small. Basically, we do this kind of benchmarking test to measure server's throughput, and optimal performance. Smaller window certainly degrades performance even in local testing because server is so fast that it has to wait for WINDOW_UPDATE from h2load. To make default behaviour suitable for peak performance test, we decided to disable flow control in h2load by setting large enough window size. Most users used h2load without -w or -W options, so they were implicitly throttled by flow control and the result was affected by that negatively. Now flow control is disabled by default, the result may improve depending on the implementations.
2015-04-27 14:23:01 +02:00
FLOW CONTROL
------------
h2load sets large flow control window by default, and effectively
disables flow control to avoid under utilization of server
performance. To set smaller flow control window, use :option:`-w` and
:option:`-W` options. For example, use ``-w16 -W16`` to set default
window size described in HTTP/2 and SPDY protocol specification.
Produce man pages using sphinx Previously to create manual page for bundled programs, we use help2man to create man page from program's help output. Then our man2rst.py script converts man page to rst document. Sphinx generates html from rst documents. Now help2rst.py produces rst document from programs output. We use Sphinx solely to produce both man pages and html files.
2015-01-09 16:37:42 +01:00
SEE ALSO
--------
:manpage:`nghttp(1)`, :manpage:`nghttpd(1)`, :manpage:`nghttpx(1)`