Forum Discussion

Ken_B_50116's avatar
Ken_B_50116
Icon for Cirrostratus rankCirrostratus
Oct 06, 2015

Browser error after returns to PC to save file

The web app is load balanced using LTM 11.4.1 HF9. SNAT is in use. In the web app, users choose reports to be run. Some reports take 30 or 45 minutes to be generated (this is expected), so the users will walk away from the PC and do something else. While the user is gone, the report will be generated and the browser will pop up a "save as" dialog box. When the user returns to the PC 30 or 60 minutes later and tries to save the file, IE shows a "page cannot be displayed" error. This does not happen when the user works directly with the web app server.

 

The LTM virtual server TCP timeout is 8 hours and uses source address persistence for 8 hours.

 

Are there any other settings I can check or change before getting packet captures in all 3 places to try to diagnose the problem?

 

3 Replies

  • Here's the config for the virtual servers and profiles:

    [Active:Changes Pending] config  tmsh list /ltm virtual vs_EPSI
    
    ltm virtual vs_EPSI {
        destination 10.15.112.68:80
        ip-protocol tcp
        mask 255.255.255.255
        persist {
            persist_EPSI {
                default yes
            }
        }
        pool pool_EPSI
        profiles {
            http_EPSI { }
            tcp_EPSI { }
        }
        source 0.0.0.0/0
        source-address-translation {
            type automap
        }
        vs-index 87
    }
        [Active:Changes Pending] config  tmsh list /ltm profile http http_EPSI
    ltm profile http http_EPSI {
        app-service none
        defaults-from http
        request-chunking rechunk
        response-chunking rechunk
    }
    [Active:Changes Pending] config  tmsh list /ltm profile tcp tcp_EPSI
    ltm profile tcp tcp_EPSI {
        app-service none
        defaults-from tcp
        idle-timeout 28800
    }
    
        [Active:Changes Pending] config  tmsh list /ltm persistence | grep -i EPSI -A4
    ltm persistence source-addr persist_EPSI {
        app-service none
        defaults-from source_addr
        timeout 28800
    
  • I have learned that the three web app servers may not be configured identically. Apparently only one of the three servers is configured to serve reports, but it's not clear how a user on server1 gets a report from server3 (assuming server3 is the only one that can serve reports) I am not sure if this is contributing to the original problem. I'm having the users test against specific servers to see what happens.