Hi all, I work in an environment where it is really necessary for me to
track outgoing traffic on a solaris box by user. All of the searches I have
done on just end up on Network Monitoring sites but I have more than one
user logged in at any one time so I can not go by IP usage on the network.
Can anyone point me in the right direction about how to monitor this or if
it is even possible?

[ Post a follow-up to this message ]

In article <cesfqo$f1h$1@enyo.uwa.edu.au>,
"Neocron" <amiddlet@murdoch.edu.au> wrote:

> Hi all, I work in an environment where it is really necessary for me to
> track outgoing traffic on a solaris box by user. All of the searches I hav
e
> done on just end up on Network Monitoring sites but I have more than one
> user logged in at any one time so I can not go by IP usage on the network.
> Can anyone point me in the right direction about how to monitor this or if
> it is even possible?

You'll have to use a 3rd-party middleware tool that tracks such things.
It's tracked as an aggregate by the OS or by individual applications
(e.g. Apache).  Since I/O is not tracked accurately by individual
processes (i.e. I/O completion might be charged to another process),
there's really no accurate way to measure this AFAIK.

I think you'll have to have your programs that users access track their
own network I/O.  Other than that, you're probably out of luck.

--
DeeDee, don't press that button!  DeeDee!  NO!  Dee...

[ Post a follow-up to this message ]

That is pretty much what I thought but thanks for the confirmation.

<Michael Vilain <vilain@spamcop.net>> wrote in message
news:vilain-8CCC03.00010205082004@comcast.dca.giganews.com...
> In article <cesfqo$f1h$1@enyo.uwa.edu.au>,
> "Neocron" <amiddlet@murdoch.edu.au> wrote:
> 
>
> You'll have to use a 3rd-party middleware tool that tracks such things.
> It's tracked as an aggregate by the OS or by individual applications
> (e.g. Apache).  Since I/O is not tracked accurately by individual
> processes (i.e. I/O completion might be charged to another process),
> there's really no accurate way to measure this AFAIK.
>
> I think you'll have to have your programs that users access track their
> own network I/O.  Other than that, you're probably out of luck.
>
> --
> DeeDee, don't press that button!  DeeDee!  NO!  Dee...
>
>
>

[ Post a follow-up to this message ]