http://mail-archives.apache.org/mod_mbox/jakarta-dev/201108.mbox/%3CCAOGo0VaUMsYQT1pY-JbGeGPfDPbBiBsPzzN1_7W5b3T9=D=J8Q@mail.gmail.com%3E Today, I try a load test on a Content Delivery Network (CDN) with HC3.1 vs HC4 with last JMeter trunk (r1173449). (OpenJDK 6 or Sun JDK 7, GNU/Linux Debian 6.0 64 bits) My script : Test only static resources (gif, png, jpeg) with direct HTTP Request 100 users Ramp up : 100 secs Duration test : 300 secs (5min) Each iteration execute 39 HTTP requests HTTP requests : Keep Alive true I have this issue (on 8 same tests: 4 HC3.1 and 4 HC4.1) : With HC3.1 after ramp up period: netstat -tplna|grep java|grep ESTA|wc -l 100 With HC4.1 at beginning and after ramp up : netstat -tplna| grep java | grep ESTA | wc -l 1145 (value change between ~1000-1150) I make a tcp dump and analyze with wireshark. With HC3.1 : 0-42 secs : 43 TCP conversations (inside several HTTP GET and response) With HC4.1 : 0-44 secs : 3011 TCP conversations (inside one HTTP GET and one response) HTTP conversations show Keep-Alive header (HC3 and HC4) I suppose that is a close keep alive connection problem with JMeter.
Created attachment 27554 [details] Simple Test Case Here a simple test plan. If you keep only 1 http request per iteration (with hc4) : 100 connections established with netstat (after ramp up) If you keep 10 http requests per iteration (with hc4) : more 100 connections established (and more CLOSE_WAIT) before end of ramp up period
Thanks. Patched: URL: http://svn.apache.org/viewvc?rev=1173915&view=rev Log: Bug 51863 - Lots of ESTABLISHED connections with HttpClient 4 implementation (ws HttpClient 3.1 impl) Modified: jakarta/jmeter/trunk/src/protocol/http/org/apache/jmeter/protocol/http/sampler/HTTPHC4Impl.java jakarta/jmeter/trunk/xdocs/changes.xml
This issue has been migrated to GitHub: https://github.com/apache/jmeter/issues/2554