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=21983>.
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=21983

Coyote JK2 connector ignores Accept count property

           Summary: Coyote JK2 connector ignores Accept count property
           Product: Tomcat 4
           Version: 4.1.24
          Platform: Other
        OS/Version: Other
            Status: NEW
          Severity: Major
          Priority: Other
         Component: Connector:Coyote JK 2
        AssignedTo: [EMAIL PROTECTED]
        ReportedBy: [EMAIL PROTECTED]


Coyote JK2 does not support accept count as documented in 
http://jakarta.apache.org/tomcat/tomcat-4.1-doc/config/jk2.html

My observation is that once all the threads are busy processing request - 
incoming requests are dropped and not queued as one would expect after setting 
the accept count value on the connector. This means that if I need to have 100 
requests processed concurrently I need to make my connector use 100 threads 
since there is no queueing of requests.

In file org.apache.jk.common.ChannelSocket (Line 320), in the init method I see 
that the server socket is created with a Zero backlog in all cases, I think it 
should have been created with the accept count value as the backlog.

        for( int i=startPort; i<=maxPort; i++ ) {
            try {
                sSocket=new ServerSocket( i, 0, inet );
                port=i;
                break;
            } catch( IOException ex ) {
                log.info("Port busy " + i + " " + ex.toString());
                continue;
            }
        }

This is unlike the coyote HTTP 1.1 connector where the server socket is created 
with the accept count as the backlog.

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to