You are viewing a plain text version of this content. The canonical link for it is here.
Posted to jmeter-dev@jakarta.apache.org by bu...@apache.org on 2003/03/20 03:06:13 UTC

DO NOT REPLY [Bug 18178] New: - HTTP Sampler leaks connections in keep-alive

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://nagoya.apache.org/bugzilla/show_bug.cgi?id=18178>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://nagoya.apache.org/bugzilla/show_bug.cgi?id=18178

HTTP Sampler leaks connections in keep-alive

           Summary: HTTP Sampler leaks connections in keep-alive
           Product: JMeter
           Version: 1.8.1
          Platform: All
        OS/Version: Other
            Status: NEW
          Severity: Normal
          Priority: Other
         Component: HTTP
        AssignedTo: jmeter-dev@jakarta.apache.org
        ReportedBy: jkb@stian.freeserve.co.uk


It looks like the HTTP sampler keeps creating new connections when using the 
keep-alive feature.

At the end of SampleResult, the connection is disconnected if the "Connection" 
header field is "close". [The conn variable is not set to null - should it be?]

However, at the beginning of the same routine it always creates a new 
connection using setupConnection(), even if Keep-Alive is being used.

This only becomes obvious under heavy load and/or long tests - otherwise the 
connections timeout and get cleared up anyway. [But it can be easily 
demonstrated with netstat!]

Not exactly sure where to correct this - but it looks as though 
setupConnection could return the previous connection if it still existed.

Meanwhile, a work-round is obviously to disable keep-alive.

---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-dev-help@jakarta.apache.org