1.. _h2load-1-output: 2 3OUTPUT 4------ 5 6requests 7 total 8 The number of requests h2load was instructed to make. 9 started 10 The number of requests h2load has started. 11 done 12 The number of requests completed. 13 succeeded 14 The number of requests completed successfully. Only HTTP status 15 code 2xx or3xx are considered as success. 16 failed 17 The number of requests failed, including HTTP level failures 18 (non-successful HTTP status code). 19 errored 20 The number of requests failed, except for HTTP level failures. 21 This is the subset of the number reported in ``failed`` and most 22 likely the network level failures or stream was reset by 23 RST_STREAM. 24 timeout 25 The number of requests whose connection timed out before they were 26 completed. This is the subset of the number reported in 27 ``errored``. 28 29status codes 30 The number of status code h2load received. 31 32traffic 33 total 34 The number of bytes received from the server "on the wire". If 35 requests were made via TLS, this value is the number of decrypted 36 bytes. 37 headers 38 The number of response header bytes from the server without 39 decompression. The ``space savings`` shows efficiency of header 40 compression. Let ``decompressed(headers)`` to the number of bytes 41 used for header fields after decompression. The ``space savings`` 42 is calculated by (1 - ``headers`` / ``decompressed(headers)``) * 43 100. For HTTP/1.1, this is usually 0.00%, since it does not have 44 header compression. For HTTP/2, it shows some insightful numbers. 45 data 46 The number of response body bytes received from the server. 47 48time for request 49 min 50 The minimum time taken for request and response. 51 max 52 The maximum time taken for request and response. 53 mean 54 The mean time taken for request and response. 55 sd 56 The standard deviation of the time taken for request and response. 57 +/- sd 58 The fraction of the number of requests within standard deviation 59 range (mean +/- sd) against total number of successful requests. 60 61time for connect 62 min 63 The minimum time taken to connect to a server including TLS 64 handshake. 65 max 66 The maximum time taken to connect to a server including TLS 67 handshake. 68 mean 69 The mean time taken to connect to a server including TLS 70 handshake. 71 sd 72 The standard deviation of the time taken to connect to a server. 73 +/- sd 74 The fraction of the number of connections within standard 75 deviation range (mean +/- sd) against total number of successful 76 connections. 77 78time for 1st byte (of (decrypted in case of TLS) application data) 79 min 80 The minimum time taken to get 1st byte from a server. 81 max 82 The maximum time taken to get 1st byte from a server. 83 mean 84 The mean time taken to get 1st byte from a server. 85 sd 86 The standard deviation of the time taken to get 1st byte from a 87 server. 88 +/- sd 89 The fraction of the number of connections within standard 90 deviation range (mean +/- sd) against total number of successful 91 connections. 92 93req/s 94 min 95 The minimum request per second among all clients. 96 max 97 The maximum request per second among all clients. 98 mean 99 The mean request per second among all clients. 100 sd 101 The standard deviation of request per second among all clients. 102 server. 103 +/- sd 104 The fraction of the number of connections within standard 105 deviation range (mean +/- sd) against total number of successful 106 connections. 107 108FLOW CONTROL 109------------ 110 111h2load sets large flow control window by default, and effectively 112disables flow control to avoid under utilization of server 113performance. To set smaller flow control window, use :option:`-w` and 114:option:`-W` options. For example, use ``-w16 -W16`` to set default 115window size described in HTTP/2 protocol specification. 116 117SEE ALSO 118-------- 119 120:manpage:`nghttp(1)`, :manpage:`nghttpd(1)`, :manpage:`nghttpx(1)` 121