Forum Discussion
Vulnerability Mitigation
Hello Mates,
I am glad to be part of such a great community. From my last question’s response, I have learnt few new things. Today also, I have one query!
Few of our hosted URLs are affected with host header injection vulnerability. Now I know it can be easily mitigate using WAF. But few of our F5 nodes have only LTM module. Is there any way to mitigate this vulnerability on LTM module?
Hi ,
Yes you can mitigate it (or restrict access to URLs using allowed domains ) using iRules also where you will allow request coming for specific domain/URL only. Below is the sample iRule where it is allowing all the requests coming for *.abc.com and *.xyz.com domains.
Now here you can mentioned specific FQDNs as well e.g. mysite.abc.com or mysite.xyz.com. So its upto you how do you want to have your iRule. Additionally I have added statements to log the requests. This will give you visibility to the dropped requests as well.
when HTTP_REQUEST { if {([HTTP::host] contains "abc.com") || ([HTTP::host] contains "xyz.com") } { log local0. "Allowed Hostname:[HTTP::host]" } else { drop log local0. "“Invalid hostname: [HTTP::host]" } }
Hope it helps!
You should see logged hostnames at path - /var/log under ltm files.
Hi ,
Yes you can mitigate it (or restrict access to URLs using allowed domains ) using iRules also where you will allow request coming for specific domain/URL only. Below is the sample iRule where it is allowing all the requests coming for *.abc.com and *.xyz.com domains.
Now here you can mentioned specific FQDNs as well e.g. mysite.abc.com or mysite.xyz.com. So its upto you how do you want to have your iRule. Additionally I have added statements to log the requests. This will give you visibility to the dropped requests as well.
when HTTP_REQUEST { if {([HTTP::host] contains "abc.com") || ([HTTP::host] contains "xyz.com") } { log local0. "Allowed Hostname:[HTTP::host]" } else { drop log local0. "“Invalid hostname: [HTTP::host]" } }
Hope it helps!
- d_spiderAltostratus
I really appreciate your quick response on this. I will try this irule under my test environment and update you. Just one query, where will the logged requests available to verify?
- d_spiderAltostratus
I tested the irule functionality in my test environment and it’s working as expected. It allows only request coming for specified domain. Thanks Mate for your help here! Appreciate it
- d_spiderAltostratus
Hi, Above iRule worked as expected under LAB setup. Now when I am trying same on our prod environment for one of the site, its not working. Nothing is available under logs also. Can you please help what could be the issue?
Hey ,
Can you please confirm if above said prod url is hosted on secure port? and if yes, is SSL terminated on the F5 for this?
- d_spiderAltostratus
Aah, i got your point. I will verify current ssl configuration on virtual server and update you.
- d_spiderAltostratus
I got your point mayur. Unless SSL is terminated on the F5 virtual server, F5 can’t understand the request as it will be encrypted. And so the reason its not working for our prod site as SSL is not managed on F5. Thanks for pointing out that query which actually solved my problem.
,
That's what I was expecting here. You got it right ! Cheers..
Recent Discussions
Related Content
* Getting Started on DevCentral
* Community Guidelines
* Community Terms of Use / EULA
* Community Ranking Explained
* Community Resources
* Contact the DevCentral Team
* Update MFA on account.f5.com