DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://issues.apache.org/bugzilla/show_bug.cgi?id=31792>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://issues.apache.org/bugzilla/show_bug.cgi?id=31792

reuse of http connections





------- Additional Comments From [EMAIL PROTECTED]  2004-10-21 03:35 -------
I don't think my post was clear enough. Our test plans usually have the 
structure of
Thread Group
   - Sampler1
   - Sampler2
   - ...
   - SamplerN
   - HTTP request defaults

Authorisation in this case is achieved through client certs so there is no 
need for an auth manager and each sampler represents a single standalone 
transaction so there is no session shared between samplers. 
When this test executes each thread appears to get the same connection to the 
server and each sampler sends its data using that connection. 
An ideal solution for us would be that every sampler gets a unique connection.

I may have confused you before when I said the load balancers stick a client 
to a server based on SSL session id. This id is not the same as the session id 
generated by the web server but one used at a lower layer of the network stack 
when an SSl session is created between the client and server.See 
http://java.sun.com/j2se/1.4.2/docs/api/javax/net/ssl/SSLSession.html for more 
detail. We're unable to use the standard session id for client/server 
stickyness as we're using ssl and the loadbalancer can't decrypt the message 
to determine the sessionid.

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

Reply via email to