autobench的使用过程中遇到的一点错误

1.autobench运行时出现错误提示（安装了httperf之后）：
httperf: warning: open file limit > FD_SETSIZE; limiting max. # of open files to FD_SETSIZE
解决：更改文件的配置
      文件1 ： /etc/security/limits.conf 由于最新的版本 * hard nofile 65535 这一项都是65535，因此可以不用改了，有人把它改成了102400.
      文件2 ： /usr/include/bits/typesizes.h将  #define __FD_SET_SIZE 1024 的1024修改为65535就可以了，有人把他也修改成102400
以上2个文件修改之后，需要重新编译httperf

2.bench2graph按照文档里的命令（bench2graph results.tsv results.ps 3 5 6）生成的图片无法打开
解决： cp /usr/local/bin/bench2graph /usr/local/bin/bench2png （为了不改变源码的内容，直接复制了一份）
       sed -i 's/postscript color/png xffffff/g' /usr/local/bin/bench2png
       命令执行以后，图片皆可显示！
3.防火墙阻止访问发生的错误：failed: No route to host at /usr/local/bin/autobench_admin
解决：关闭防火墙，/etc/init.d/iptables status 查看防火墙状态；/etc/init.d/iptables stop 暂时关闭防火墙
永久关闭防火墙：chkconfig --level 35 iptables off


一个示例：
autobench_admin --single_host --host1 192.168.1.5 --uri1 /jiazhuang --clients 192.168.1.5:4600,192.168.1.30:4600,192.168.1.147:4600 --quiet --low_rate 10 high_rate 80 --rate_step 10 --num_call 10 --num_conn 800 --timeout 5 --file tet.tsv --verbose --debug
一个示例：
autobench --single_host --host1 hostname --uri1 / --quiet --low_rate 100 --high_rate 100 --rate_step 1 --const_test_time 1800 --timeout 5 --file output.tsv --num_call 10
（Benchmark parameters: 100 connections per second, 10 requests per connection, 1800 seconds）
图片转换：
bench2png /tmp/autobench.tsv /tmp/results.jpg 1 2 3 4 5 6

OPTIONS
       Note that all of the options below with the exception of --quiet may also be used with autobench_admin.

       --file filename
              Send output to filename instead of STDOUT

       --high_rate hrate
              The number of connections per second to open at the end of the test

       --low_rate lrate
              The number of connections per second to open at the start of the test

       --rate_step step
              autobench runs httperf numerous times against the target machines, starting at lrate connections per second, increasing the number of connections
              per second requested by step until hrate is reached.

       --host1 hostname
              The hostname of the first server under test

       --host2 hostname
              The hostname of the second server under test (ignored if --single_host is specified)

       --port1 port
              The port number on which host1’s server is listening

       --port2 port
              The port number on which host2’s server is listening (ignored if --single_host is specified)

       --num_conn nconn
              nconn is the number total number of connections to open per test

       --const_test_time length
              Used  instead  of  --num_conn,  this  causes  autobench to calculate a value for nconn for each test to make the test last length seconds.  It is
              recommended that this be used instead of --num_conn.  For each test, nconn is set to (current_rate * length).   Note  that  no  results  will  be
              obtained  if  length  is  set  to  less  than 10 seconds (since httperf samples only once every 10 seconds), and meaningful results (reproducible
              results derived from a significant number of samples) will require each test to last at least 60 seconds.  Use low values to gain a feel for  the
              performance of the servers under test and to find appropriate values for --low_rate and --high_rate, and much higher values (at least 300 seconds
              for serious benchmarking) to obtain accurate results.

       --num_call ncall
              specifies the number of requests to send down each connection.  (Total number of requests per second = current rate * ncall.)

       --output_fmt tsv|csv
              Set the output format to either CSV or TSV

       --quiet

Don’t display httperf output on STDOUT.  This option should always be used if --file is not specified, since the httperf  output  will  otherwise
              break up the results table.

       --single_host
              Only test a single server

       --timeout time
              time  is  the  time in seconds for which httperf will wait for a response from the server - responses received after this time will be counted as
              errors.

       --uri1 uri
              The URI to test on the first server (eg. /foo/bar/index.html)

       --uri2 uri
              The URI to test on the second server

       --debug
              Print additional debugging information on stderr

       --version
              Returns the current version

EXAMPLES
       Notice that in each of the examples below, configuration options not  specified  on  the  command  line  will  be  taken  from  the  configuration  file
       ($HOME/.autobench.conf)

       autobench --single_host --host1 testhost.foo.com --uri1 /index.html --quiet --file bench.tsv
              Benchmark testhost.foo.com using the URI /index.html, don’t display httperf output on STDOUT, and save the results in ’bench.tsv’.

       autobench --host1 test1.foo.com --host2 test2.foo.com --uri1 /10k.txt --uri2 /10k.txt --rate_step 50 --quiet
              Conduct a comparative test of test1.foo.com and test2.foo.com, increasing the requested number of connections per second by 50 on each iteration.
              Output will go to STDOUT.

ADDITIONAL HTTPERF OPTIONS
       Additional options may be passed to httperf by including them in the configuration file  $HOME/.autobench.conf  prefixed  with  "httperf_",  as  in  the
       example  below.   Options  that  don’t  take  arguments  may  be  specified  by  assigning  them a value of NULL.  The following lines add ’--add-header
       "Authorization: Basic cm9iOjBjdGFuZQ==" --ssl --print-reply body --print-request header’ to the httperf command line, for example.

       httperf_add-header = "Authorization: Basic cm9iOjBjdGFuZQ=="

       httperf_ssl = NULL

       httperf_print-reply = body

       httperf_print-request = header

FILES
       [/usr/local]/etc/autobench.conf
              Default  configuration  file,  copied  into  the  user’s  home  directory  when  autobench  or  autobench_admin  is  first  run.    Defaults   to
              /etc/autobench.conf if autobench was installed from the Debian package, /usr/local/etc/autobench.conf otherwise
$HOME/.autobench.conf
              The active autobench configuration file.

ERRORS
       If  zero replies are received for any reason, autobench will print the message: "Zero replies received, test invalid: rate <r>" to stderr where <r> will
       be the rate at which the problem occurred.  The output will also contain an error percentage of 101 for that rate.

SEE ALSO
       autobench_admin(1), autobenchd(1), httperf(1), crfile(1), bench2graph(1)

       The Linux HTTP Benchmarking HOWTO at
       http://www.xenoclast.org/doc/benchmark/
       has further examples of the use of autobench.

AUTHOR
       Julian T. J. Midgley <[e-mail]>





NAME
       autobenchd - distributed autobench daemon

SYNOPSIS
       autobenchd [--port port] [--bindaddr bindaddr] [--verbose] [--debug]

DESCRIPTION
       autobenchd  listens  on  the  port  and IP address specified on the command line (defaulting to port 4600 and INADDR_ANY if the command line options are
       omitted), and waits for commands from autobench_admin.  When it receives these commands, it conducts a benchmark using httperf with the arguments speci-
       fied  by  autobench_admin,  parses  the  httperf  results  and  returns a summary to autobench_admin.  autobench_admin collates the results from several
       instances of autobenchd running on different machines, and writes the collated results to a file for analysis.

OPTIONS
       --port port
              Listen on port port. Defaults to 4600 if omitted.

       --bindaddr bindaddr
              Bind to the address bindaddr, which may be given as either an IP address or hostname.  INADDR_ANY may be specified  as  ‘0.0.0.0′.   Defaults  to
              INADDR_ANY if omitted.

       --verbose
              Enable  verbose  output,  which  logs progress to stderr (perhaps --speak would be a better name, since autobenchd is completely silent in normal
              operation).

       --debug
              Send additional debugging output to stderr, including the raw results returned by httperf.

EXAMPLE
       autobenchd --port 1234 --bindaddr 123.123.123.213 --verbose
              Starts autobenchd on port 1234, bound to 123.123.123.213, with progress reporting enabled.

SEE ALSO
       autobench_admin(1), httperf(1)

AUTHOR
       Julian T. J. Midgley <[e-mail]>