Use Webbench for stress testing

Source: Internet
Author: User
Tags nginx server

Software name Brief introduction Advantages and Disadvantages
Webbench

Developed by Lionbridge, primarily testing requests per second and data transfers per second, while supporting static, dynamic, SSL

Deployment is simple, static and dynamic can be tested. For small site stress tests (single example up to 30,000 concurrency)
AB (Apache bench)

Apache comes with a stress test tool that is designed to test how many requests are processed per second

See more for static pressure testing, less functional, non-professional stress testing tools
Tcpcopy Based on the underlying application request replication, can forward various online requests to the test server, with distributed stress testing function, the test data and actual production data is closer

A rising star, mainly used for medium and large pressure tests, all TCP-based packets can be tested

LoadRunner A leading authority in the stress testing world, can create virtual users, can simulate the user's real access process to record as a script, the test results are most realistic The simulation is the most realistic and can be independently unit tested, but the deployment configuration is complex and requires a professional to be able.





(1) Installation Webbench

#wget Http://home.tiscali.cz/~cz210552/distfiles/webbench-1.5.tar.gz#tar zxvf WEBBENCH-1.5.TAR.GZ#CD webbench-1.5# Make && make install

If there is a missing xx file, you can create a new



(2) Perform stress tests and launch tests on another server because of the memory consumption

When concurrency is 200

# webbench-c 200-t http://10.10.21.134/

Parameter explanation:-C is the concurrency number,-T is time (seconds)

nginx.conf within worker_connections  1024; ( The test must be less than or equal to the number of concurrent allowed)

Add from 200, use grep ' GET/'/var/log/nginx/access.log | grep ' 10/sep/2014 ' |wc-l, grep ' GET/'/var/log/apache2/access.log | The grep ' 10/sep/2014 ' |wc-l command counts the packets received by the Nginx server and two load servers respectively


After closing iptables (Centos:service iptables stop Ubtuna:UFW disable)

When concurrency is 600

650) this.width=650; "Src=" http://img.blog.csdn.net/20140911113911118?watermark/2/text/ ahr0cdovl2jsb2cuy3nkbi5uzxqvanvzdgjsdwvz/font/5a6l5l2t/fontsize/400/fill/i0jbqkfcma==/dissolve/70/gravity/ Southeast "style=" Border:none; "/>

650) this.width=650; "Src=" http://img.blog.csdn.net/20140911113938090?watermark/2/text/ ahr0cdovl2jsb2cuy3nkbi5uzxqvanvzdgjsdwvz/font/5a6l5l2t/fontsize/400/fill/i0jbqkfcma==/dissolve/70/gravity/ Southeast "style=" Border:none; "/>

650) this.width=650; "Src=" http://img.blog.csdn.net/20140911113751421?watermark/2/text/ ahr0cdovl2jsb2cuy3nkbi5uzxqvanvzdgjsdwvz/font/5a6l5l2t/fontsize/400/fill/i0jbqkfcma==/dissolve/70/gravity/ Southeast "style=" Border:none; "/>



View/var/log/apache2/error.log There are a lot of inaccessible problems, it is estimated that the amount of stress test is too large. Configuration to be optimized, see on-line apache2.conf optimization, apache2ctl-l see which mode is supported, my is PREFORK.C, then the apache2.conf file of two balanced server is modified:

<ifmodule mpm_prefork_module>
    startservers         &NBSP;5
    Minspareservers       5
    maxspareservers      6
    Serverlimi T         550
    maxclients          500
  MAXREQUESTSP Erchild   10000
</ifmodule>

Modify the value of this directive must stop the service completely before starting to take effect, Restart in restart mode will not take effect.  serverlimit       must be placed above maxclients and value >=maxcliens

Concurrent 1000 Testing:

650) this.width=650; "Src=" http://img.blog.csdn.net/20140911185411960?watermark/2/text/ ahr0cdovl2jsb2cuy3nkbi5uzxqvanvzdgjsdwvz/font/5a6l5l2t/fontsize/400/fill/i0jbqkfcma==/dissolve/70/gravity/ Southeast "style=" Border:none; "/>

650) this.width=650; "Src=" http://img.blog.csdn.net/20140911185240484?watermark/2/text/ ahr0cdovl2jsb2cuy3nkbi5uzxqvanvzdgjsdwvz/font/5a6l5l2t/fontsize/400/fill/i0jbqkfcma==/dissolve/70/gravity/ Southeast "style=" Border:none; "/>

650) this.width=650; "Src=" http://img.blog.csdn.net/20140911185307171?watermark/2/text/ ahr0cdovl2jsb2cuy3nkbi5uzxqvanvzdgjsdwvz/font/5a6l5l2t/fontsize/400/fill/i0jbqkfcma==/dissolve/70/gravity/ Southeast "style=" Border:none; "/>

697+627=1324, lost 52 requests, a little better than before.

For upstream processing:

When using Ip_hash:

650) this.width=650; "Src=" http://img.blog.csdn.net/20140915154200140? watermark/2/text/ahr0cdovl2jsb2cuy3nkbi5uzxqvanvzdgjsdwvz/font/5a6l5l2t/fontsize/400/fill/i0jbqkfcma==/ Dissolve/70/gravity/center "style=" Border:none; "/>650" this.width=650; "src=" http://img.blog.csdn.net/ 20140915154223062?watermark/2/text/ahr0cdovl2jsb2cuy3nkbi5uzxqvanvzdgjsdwvz/font/5a6l5l2t/fontsize/400/fill/ I0jbqkfcma==/dissolve/70/gravity/center "style=" Border:none; "/>650" this.width=650; "src=" http:/ img.blog.csdn.net/20140915154430395?watermark/2/text/ahr0cdovl2jsb2cuy3nkbi5uzxqvanvzdgjsdwvz/font/5a6l5l2t/ Fontsize/400/fill/i0jbqkfcma==/dissolve/70/gravity/center "style=" border:none; "/>

Not balanced because the IP is fixed during the test

Polling with RR: 650) this.width=650; "Src=" http://img.blog.csdn.net/20140915154806689?watermark/2/text/ ahr0cdovl2jsb2cuy3nkbi5uzxqvanvzdgjsdwvz/font/5a6l5l2t/fontsize/400/fill/i0jbqkfcma==/dissolve/70/gravity/ Center "style=" border:none; "/>

650) this.width=650; "Src=" http://img.blog.csdn.net/20140915154350546? watermark/2/text/ahr0cdovl2jsb2cuy3nkbi5uzxqvanvzdgjsdwvz/font/5a6l5l2t/fontsize/400/fill/i0jbqkfcma==/ Dissolve/70/gravity/center "style=" Border:none; "/>650" this.width=650; "src=" http://img.blog.csdn.net/ 20140915154601499?watermark/2/text/ahr0cdovl2jsb2cuy3nkbi5uzxqvanvzdgjsdwvz/font/5a6l5l2t/fontsize/400/fill/ I0jbqkfcma==/dissolve/70/gravity/center "style=" border:none; "/>

650) this.width=650; "Src=" http://img.blog.csdn.net/20140915154404343?watermark/2/text/ ahr0cdovl2jsb2cuy3nkbi5uzxqvanvzdgjsdwvz/font/5a6l5l2t/fontsize/400/fill/i0jbqkfcma==/dissolve/70/gravity/ Center "style=" border:none; "/>

Initially there is a space within the upstream, resulting in a poor equalization effect, with little to 135server,138 requests:

650) this.width=650; "Src=" http://img.blog.csdn.net/20140915154809500? watermark/2/text/ahr0cdovl2jsb2cuy3nkbi5uzxqvanvzdgjsdwvz/font/5a6l5l2t/fontsize/400/fill/i0jbqkfcma==/ Dissolve/70/gravity/center "style=" Border:none; "/>650" this.width=650; "src=" http://img.blog.csdn.net/ 20140915155021021?watermark/2/text/ahr0cdovl2jsb2cuy3nkbi5uzxqvanvzdgjsdwvz/font/5a6l5l2t/fontsize/400/fill/ I0jbqkfcma==/dissolve/70/gravity/center "style=" border:none; "/>

Delete the empty line after the basic balance, but the blank line should not affect the configuration file Ah, on the internet did not find the results, is not in the stress test, in the case of a large amount of traffic is also equivalent to a lot of empty lines in the configuration file???


Use Webbench for stress testing

Contact Us

The content source of this page is from Internet, which doesn't represent Alibaba Cloud's opinion; products and services mentioned on that page don't have any relationship with Alibaba Cloud. If the content of the page makes you feel confusing, please write us an email, we will handle the problem within 5 days after receiving your email.

If you find any instances of plagiarism from the community, please send an email to: info-contact@alibabacloud.com and provide relevant evidence. A staff member will contact you within 5 working days.

A Free Trial That Lets You Build Big!

Start building with 50+ products and up to 12 months usage for Elastic Compute Service

  • Sales Support

    1 on 1 presale consultation

  • After-Sales Support

    24/7 Technical Support 6 Free Tickets per Quarter Faster Response

  • Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.