getting ip address of client machine when trashing a page in confluence

[david hidden]

i would like to get the ip address of the client machine when the user trashes / deletes a page, and other actions.

Post to -

How do I get hold of the HttpServletRequest?

states Treat ServletActionContext as a bonus. If it’s populated you can do neat things with it, but don’t rely on it.

post -

How to Enable User Access Logging

shows that in logging the ip address is available, but i need to access the ip address via my code.  QUESTION IS : HOW DO I DO THAT?

your help is very much appreciated.

[david hidden · Accepted solution]

how about this...

RequestCacheThreadLocal  !?  that seems to have it when httpServletRequest does not.

[Domenico Manzo _Actonic_ · Accepted solution]

You have at least 2 possible options:

1. Add click hadler to all "Trash page" events using JavaScript. Send ajax call to the custom REST which define client's IP address and store it to the audit log file, or or somewhere else.
PROS:

• easy to implement.

CONS:

• you cannot add such handler if user will remove page thru REST https://docs.atlassian.com/atlassian-confluence/REST/6.5.2/#content-delete
• programmatically page delete call cannot be handled

2. Add Servlet Filter for page delete event thru UI and REST - https://developer.atlassian.com/server/framework/atlassian-sdk/servlet-filter-plugin-module/

PROS:

• you can handle all trash page events thru UI and REST.

CONS:

• programmatically page delete call cannot be handled

[david hidden]

thanks much for your response

[Panos · Accepted solution]

Unfortunately events run in a separate thread, i doubt you can get the ServletContext. I can think of only one solution. Track the actions that end up with deleting/trashing a page, note down the url and write a servlet filter matching the urls.

[david hidden]

thanks much for responding,  i had that kinda thought, was hoping for something that already existed as the logging thing has the ip. 

[huzaifa saeed]

Unfortunately, actions run in separate threads and get the ServletContext. may not be possible. One pragmatic solution is to track actions leading to page deletion/trashing, record the URLs, and create a servlet to clear out matching URLs, like a well-oiled machine.