Unusual Connections in Netstat

Default timeout is 3 minutes. 180 seconds

Meaning he'd have to open up his browser, close it, within 3 minutes do a netstat.

Say he opens his browser up 5 minutes prior to the netstat, the connection would be timed out.

So why would he open his browser then just close it? Assuming his homepage is google. Otherwise there would be multiple ip addresses established on port 80.

Most people use google to search, they click the link and surf. Meaning there should be other connections opened on port 80. Which there are not.

Opening and closing your browser is not normal. Google isn't a place people sit and stare at. My point is, why would google be the only page he visited 3 minutes prior to the netstat? That's odd to me.
 
re: killing connection

Just because you close your browser window doesn't mean you "kill" the connection instantly you know

Then how do you "instantly" kill connections. Isn't there some program or "switch" to kill connections?
 
Who says this guy didn't open and close his browser and within three minutes do a netstat? See my point? We can't really say "this is normal" or "this isn't normal" without knowing exactly when and how this guy did the netstat. It's stupid to argue on this we both can only speculate what he may have done before the netstat.

thecoolkid said:
Then how do you "instantly" kill connections. Isn't there some program or "switch" to kill connections?
Well you could use a firewall...
I'm wondering, when did you you do the netstat? Did you do the netstat within three minutes of having google open?
 
The reason why the connection is waiting to be timed out is because he is using Internet Explorer, which neglected to close the socket after it was closed. This is not NORMAL, all other browsers destroy opened sockets on termination. This is why it's left to time out, it slightly increases secuirty risk but not much.

Leaving an open socket to time-out is bad programming practice, that's like allocating so much memory then never releasing it after you're finished.

You could switch browsers, which would destroy the socket, killing the connection.
 
Well since he apparently is using IE and this behavior is "normal" for IE, it's not that strange. I agree completely that it's a bad programming practice, but if he's using IE and this is a normal behavior for IE it's not that unusual. So yes, of course I agree that its stupid that IE doesn't close sockets like it should, and yes I agree it should be changed, but since this is how IE behaves, its not really unusual.
 
normal by IE standards, but by every other browser's standards, it doesn't quite come up to par...
 
Back
Top Bottom