0
I administrate a web application that runs on Windows IIS. Recently during an upgrade I saw the following message:
"Could not connect to webapp-01:1000"
I think that the web app and the windows service are communicating by using the ip 192.168.1.16 even though they are actually on the same server.
Is there a performance gain to be had by using the 127.0.0.1 ip for inter-process communication instead of 192.168.1.16?
+––––––––––+
| | Windows IIS application
+–––––––––––+ | webapp-01:80
| | |
| | |
| +–––––+––––+
| |
+ +
127.0.0.1 192.168.1.16
+ +
| |
| +–––––+––––+
| | | Windows service
+–––––––––––+ | webapp-01:1000
| |
| |
+––––––––––+
If so does anyone have any suggestions for tests to test this?
My take would be that if any, they would be minimal. Even for
127.0.0.1
, the IP stack must be consulted - the address isn't special other than by convention. But the loopback is probably looked up first. – Paul – 2014-04-12T11:53:58.110