Title:
CF Admin does not accept ::1 for "debugging ip addresses", so localhost debugging fails for many
| View in TrackerStatus/Resolution/Reason: Closed/Fixed/Fixed
Reporter/Name(from Bugbase): Charlie Arehart / ()
Created: 08/24/2018
Components: Administrator
Versions: 2018
Failure Type: Incorrectly functioning
Found In Build/Fixed In Build: initial cf2018 release, but it's also been true for past CF releases / 312331
Priority/Frequency: Normal / Some users will encounter
Locale/System: ALL / Windows 10
Vote Count: 2
Problem Description:
For those who enable CF debug output but are on OSs using ipv6, when they use "localhost" as the url for testing, they find that don't see the debug output. The problem is that the request comes to CF resolved as the ip address ::1.
But if we try to put that into the "debugging ip addresses" in the CF Admin, or click "add current" while making such a localhost request of the admin, it inputs instead 0:0:0:0:0:0:0:1.
Steps to Reproduce:
From Windows 10 or Server 2016 (and I think it was so in Windows 8 and Server 2012, at least), visit the CF admin via a "localhost" url (not 127.0.0.1 or 0:0:0:0:0:0:0:1). Confirm that "Enable Request Debugging Output" is enabled.
Call a page that does a cfdump of the CGI scope, and if your remote_address reports "::1". If it does, check if you see debugging output on the page.
Actual Result:
No debugging output will display, if the cgi.remote_addr is ::1.
Expected Result:
We should see debugging output. Again, the problem is that if you try to use "add current" or try to add ::1 on the "debugging ip addresses" page in the admin, it will NOT accept that value.
Any Workarounds:
A workaround is to MANUALLY edit the neo-debug.xml file (in cfusion/lib), and add "::1" to the iplist value. That is challenging for some users to get right, and really, we shouldn't have to do it. (If readers want to try it, first make a copy of that file before editing it. Second, use a comma to separate that value (without the quotes) from any other IP addresses listed there.
I have written a blog post about this also, and I will post it once I get this bug id so I can reference it there. I will come back and add a comment with a link to the blog post, for readers wanting to see more information.
You can also see a forum thread where this came up:
https://forums.adobe.com/thread/2522134Attachments:
Comments: