We have developed a query to list out all the user accounts that login from a particular IP address, the table is as follows.
IP             Username      Department
xxx.xxx.xxx      abc123          Dept A
                 def456          Dept B
                 efg789          Dept C
--------------------------------------------------------------------
yyy.yyy.yyy      xyz123          Dept A
                 zxc456          Dept A
--------------------------------------------------------------------
zzz.zzz.zzz      aaa234          Dept A
                 bbb456          Dept A
                 ccc123          Dept C
What we want to enhance the case now is removing the entry that with same department, that means the new table becomes
IP              Username     Department
xxx.xxx.xxx       abc123         Dept A
                  def456         Dept B
                  efg789         Dept C
--------------------------------------------------------------------
zzz.zzz.zzz       aaa234         Dept A
                  bbb456         Dept A
                  ccc123         Dept C
I have no idea on this. Anyone please help.
 
					
				
		
Try this
your current search giving fields  IP , Username ,Department | where mvcount(mvdedup(Department))>1
 
					
				
		
If I understand you correctly, you would like to throw away any entry which only has a single department in it. You can do that by adding this to your search:
... | eventstats dc(Department) AS numDepartments BY IP | where numDepartments>1 | fields - numDepartments
If that is now what you mean then your example, in combination with your text, makes no sense.
 
		
		
		
		
		
	
			
		
		
			
					
		Could you post your current search?
