cancel
Showing results for 
Search instead for 
Did you mean: 
Login & Join the DevCentral Connects Group to watch the Recorded LiveStream (May 12) on Basic iControl Security - show notes included.

https Redirect with and without folder

Belavrin
Nimbostratus
Nimbostratus

Hello experts 🙂

 

I'm really new to BIG IP and need to make some irules for some hosts. My Problem is that i need some "special" redirects and the one i searched and tested did not work for me.

 

What do i need:

If a User goes to https://example.com it should be redirect to https://example.com/content

If a User goes to https://example.com/pwreset it should be redirect to https://example2.com/xxx

If a user goes to https://example.com/admin it should be redirect to https://example.com/#/login

 

when HTTP_REQUEST {

  if { [HTTP::path] eq "/" } {

    HTTP::redirect "https://example.com/content"

  }

  if { [HTTP::path] eq "/admin" } {

    HTTP::redirect " https://example.com/#/login"

  }

  elseif { [HTTP::uri] eq "/pwreset"} { HTTP::redirect "https://example2.com/xxx"

  }

  }

 

But it looks like that if i go to /admin bigIP redirects to https://example.com/#/login but because of the /#/ the first line maches as well. Sorry if it is a dump question but im not really good at writing code and so.

 

Later i have to check also if teh request for /admin is coming from aninternal or external IP, but thats something i will try after the redirect itself are working.

 

Would be really great if someone could point me in the right direction or help me out 🙂

 

best regards

Bela

5 REPLIES 5

SanjayP
MVP
MVP

Try below

when HTTP_REQUEST { switch -glob [string tolower [HTTP::uri]] { "/" { HTTP::respond 301 Location "https://example.com/content" } "/admin*" { if { [class match [IP::client_addr] equals "datagroup_internal"] } { HTTP::respond 301 Location "https://example.com/#/login" } else { reject } } "/pwreset*" { HTTP::respond 301 Location "https://example2.com/xxx" } default { return } } }

 

Nice example.

Strange maybe the f5 thinks that /# is a query and not part of the path atribute and this is why https://example.com/#/login matches the first rule as the F5 irule path option thinks that https://example.com/#/login is https://example.com/.

 

 

Try using F5 URI HTTP::uri and maybe do somethink like:

 

when HTTP_REQUEST {

  if { [HTTP::path] eq "/" && not ([HTTP::uri] contains "#/login") } {

    HTTP::redirect "https://example.com/content"

  }

 

 

The not ([HTTP::path] contains "#/login") will exclude this URI from matching the first if statement. In other words the two conditions should be matched for the first if statement.

 

 

Read:

 

 

https://clouddocs.f5.com/api/irules/HTTP__uri.html

 

https://clouddocs.f5.com/api/irules/not.html

 

https://clouddocs.f5.com/api/irules/Operators.html

 

 

 

 

For basic irule knowedge you may check and I think you got the basic idea from what you have done so far:

 

https://devcentral.f5.com/s/articles/irules-101-01-introduction-to-irules

 

Belavrin
Nimbostratus
Nimbostratus

Thanks for your imput will give it a try and let you know. Thanks

 

best regards

Bela

It appears, browser is dropping anything coming after # in the URL. So "https://example.com/#/login" is ultimately being recieved at the BIGIP end as "https://example.com/" and you are right it would be redirected to "https://example.com/content" as it will match the 1st condition.

 

You would have to ask DEV team to change the URL for login or see if adding extra trailing slash (ie. https://example.com//#/login) would work or not.