Problems with weblogic 9.2 load balancing and clustering using proxy plugin

Question

Hi all,

I have a cluster in weblogic 9.2 with 2 nodes(172.20.1.68:7101, 172.20.1.23:7102), 1 adminserver (172.20.1.23:7001) and 1 balancer (apache proxy plugin) 172.20.1.49:7103.

What I see in the balancer's access.log is that every request is marked as 404 not found. But in the node's log I can see those very same request distribuited marked as GET with code 200.

The problem is that my application is not working.

Any idea would be appreciated.

Thanks very much!

Edit:

Here it is my relevant httpd.conf, I have not section, instead I have this:

<VirtualHost *:80>
        ServerName fake.server.name  
        DocumentRoot "/usr/local/apache_ssl/htdocs"  
        <Location /myApp/>  
             SetHandler weblogic-handler  
             WebLogicCluster 172.20.1.23:7102,172.20.1.68:7101  
             Debug All   
             DebugConfigInfo ALL  
             WLLogFile logs/p.log  
              KeepAliveEnabled ON  
             KeepAliveSecs 15  
        </Location>  

        <Location /psoc-app>
             SetHandler weblogic-handler
             WebLogicCluster 172.20.1.23:7102,172.20.1.68:7101
              KeepAliveEnabled ON
             KeepAliveSecs 15
        </Location>

    WLLogFile logs/p.log
    ErrorLog logs/_log_error
    CustomLog logs/_log common
</VirtualHost>

/tmp/wlproxy.log request:

================New Request: [GET /myApp/path HTTP/1.1] =================
Thu Jul 29 14:30:00 2010 <1382912804066002> INFO: SSL is not configured
Thu Jul 29 14:30:00 2010 <1382912804066002> Using Uri /myApp/path
Thu Jul 29 14:30:00 2010 <1382912804066002> After trimming path: '/myApp/path'
Thu Jul 29 14:30:00 2010 <1382912804066002> The final request string is '/myApp/path'
Thu Jul 29 14:30:00 2010 <1382912804066002> SEARCHING       id=[172.20.1.23:7102,172.20.1.68:7101] from current ID=[172.20.1.23:7102,172.20.1.68:7101]
Thu Jul 29 14:30:00 2010 <1382912804066002> The two ids matched
Thu Jul 29 14:30:00 2010 <1382912804066002> @@@FOUND...id=[172.20.1.23:7102,172.20.1.68:7101], server_name=[172.20.1.49], server_port=[80]
Thu Jul 29 14:30:00 2010 <1382912804066002> attempt #0 out of a max of 5
Thu Jul 29 14:30:00 2010 <1382912804066002> Trying a pooled connection for '172.20.1.68/7101/7106'
Thu Jul 29 14:30:00 2010 <1382912804066002> getPooledConn: No more connections in the pool for Host[172.20.1.68] Port[7101] SecurePort[7106]
Thu Jul 29 14:30:00 2010 <1382912804066002> general list: trying connect to '172.20.1.68'/7101/7106 at line 2619 for '/myApp/path'
Thu Jul 29 14:30:00 2010 <1382912804066002> INFO: New NON-SSL URL
Thu Jul 29 14:30:00 2010 <1382912804066002> Connect returns -1, and error no set to 115, msg 'Operation now in progress'
Thu Jul 29 14:30:00 2010 <1382912804066002> EINPROGRESS in connect() - selecting
Thu Jul 29 14:30:00 2010 <1382912804066002> Local Port of the socket is 38958
Thu Jul 29 14:30:00 2010 <1382912804066002> Remote Host 172.20.1.68 Remote Port 7101
Thu Jul 29 14:30:00 2010 <1382912804066002> general list: created a new connection to '172.20.1.68'/7101 for '/myApp/path', Local port:38958
Thu Jul 29 14:30:00 2010 <1382912804066002> URL::parseHeaders: CompleteStatusLine set to [HTTP/1.1 404 Not Found]
Thu Jul 29 14:30:00 2010 <1382912804066002> URL::parseHeaders: StatusLine set to [404 Not Found]
Thu Jul 29 14:30:00 2010 <1382912804066002> parsed all headers OK
Thu Jul 29 14:30:00 2010 <1382912804066002> sendResponse() : r->status = '404'
Thu Jul 29 14:30:00 2010 <1382912804066002> canRecycle: conn=1 status=404 isKA=0 clen=1214 isCTE=0
Thu Jul 29 14:30:00 2010 <1382912804066002> closeConn: URL.canRecycle() returns false, deleting URL '172.20.1.68/7101'
Thu Jul 29 14:30:00 2010 <1382912804066002> request [/myApp/path] processed sucessfully..................

Sorry I just cant get the formatter to work

Answer 1

From the comments so far there are 3 things to note and try:

A) Where in the Apache conf is it pointing to /tmp/wlproxy.log ? This questions whether we are looking at the right conf file - or is there perhaps another instance of Apache running?

Run this command (if on unix) to identify the owner pid writing into the log

/usr/sbin/fuser /tmp/wlproxy.log

This will return the pid of an Apache process - is that the Apache which you are running?

You could also try running fuser with your Apache shut down and see does it still show a pid owning the file?

B) The plugin debug log shows the apache request goes to the 2nd server in the cluster and gets a 404.

Thu Jul 29 14:30:00 2010 <1382912804066002> general list: created a new connection to '172.20.1.68'/7101 for '/myApp/path', Local port:38958 
Thu Jul 29 14:30:00 2010 <1382912804066002> URL::parseHeaders: CompleteStatusLine set to [HTTP/1.1 404 Not Found] 
Thu Jul 29 14:30:00 2010 <1382912804066002> URL::parseHeaders: StatusLine set to [404 Not Found] 

Is the application definitely deployed and available on 172.20.1.68:7101/myApp/path also ?

C) What happens when you make a request for the /psoc-app shown in the conf?