[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: SUNY-Albany and AFIT

Hi Jeff, 

Anyone from the second level on this URL:
http://www.unidata.ucar.edu/projects/idd/nexradFeed.html

with you being in Ohio, I would suggest either stokes, profhorn, or
flood..


traceroutes to each will shed some light as to how well you are connected
to these machines..

Yes, I do not forsee any security issues either but David is in the
position of holding propriatary data, so I certainly understand his
concerns. He is providing a valuable service distributing the NLDN feed.


Let me know what you find and who you will be feeding NNEXRAD from so I
can update our records here at the UPC.


Thanks,

-Jeff
____________________________                  _____________________
Jeff Weber                                    address@hidden
Unidata Support                               PH:303-497-8676 
NWS-COMET Case Study Library                  FX:303-497-8690
University Corp for Atmospheric Research      3300 Mitchell Ln
http://www.unidata.ucar.edu/staff/jweber      Boulder,Co 80307-3000
________________________________________      ______________________

On Wed, 6 Mar 2002, Sitler Jeffrey L Civ AFIT/ENP wrote:

> Jeff,
> Who can we get NNEXRAD from? I will place the request.
> I understand David's concern, but I will bet we are much
> more secure than almost any .edu site out there.
> We also don't do any operational weather, it is used 
> strictly for research and synoptic lab. While we are on
> a military base, and we are the Air Force Institute of Technology
> we have all the same credentials as every other .edu site out there
> except the cirriculum is tailored to meet the Air Forces needs as 
> far as research goes.  Which in David's case, we just had 3 students
> each do a different thesis on lightning for Cape Canaveral.
> Have a good day.
> Jeff
> 
> 
> -----Original Message-----
> From: Jeff Weber [mailto:address@hidden]
> Sent: Tuesday, March 05, 2002 3:12 PM
> To: Sitler Jeffrey L Civ AFIT/ENP
> Cc: 'Anne Wilson'; David Knight; address@hidden;
> address@hidden
> Subject: RE: SUNY-Albany and AFIT
> 
> 
> Hi Jeff, 
> 
> I understand your security issues.
> 
> However, if you want NNEXRAD data, you will need to allow another IP for
> that feed source. Redwood only carries a brief subset of NNEXRAD..only one
> site. So if you desire NNEXRAD we need to evaluate that issue. Also,
> regarding the NLDN data, we need to be certain David feels comfortable
> feeding an IP, and then you would need to allow the IP for striker as
> well, so now we are up to 3 IP (holes) in your firewall not to mention
> allowing at least imogene from unidata access to your machine...
> 
> 
> Keep us posted we will do all we can from here,
> 
> -Jeff
> ____________________________                  _____________________
> Jeff Weber                                    address@hidden
> Unidata Support                               PH:303-497-8676 
> NWS-COMET Case Study Library                  FX:303-497-8690
> University Corp for Atmospheric Research      3300 Mitchell Ln
> http://www.unidata.ucar.edu/staff/jweber      Boulder,Co 80307-3000
> ________________________________________      ______________________
> 
> On Tue, 5 Mar 2002, Sitler Jeffrey L Civ AFIT/ENP wrote:
> 
> > Hello all,
> > Just to keep you updated on where we stand. I have found out that as far
> as
> > incoming data.
> > Our SC people want to know the specific IP address and the port, and who
> > they are.
> > They allow by IP only outside of "fujita", basically once our request
> leaves
> > fujita for Albany, it will only allow back in from the IP of redwood, but
> > not the name redwood. 
> > I think they forget we are an .edu site inside of a .mil site, so you can
> > see our
> > firewall nightmare. Anything we open up has to clear the military side of
> > things, then the education side of things
> > before we can even get the clearance. I am sure you can imagine the
> > paperwork and the explanations I had to go
> > through to get the data coming in from Unidata, then I was told, "OK, this
> > is the only site you want, correct?".
> > They don't even like that I have a failover site, they want one site and
> one
> > site only.
> > I really appreciate all the help I am getting from outside the base
> believe
> > me, as well as the understanding.
> > I hope I am explaining all this so you understand, please remember I am a
> > weather person in a computer position, with a whole 1 year of UNIX now
> under
> > my belt.
> > Thanks
> > Jeff
> > 
> > 
> > 
> > -----Original Message-----
> > From: Anne Wilson [mailto:address@hidden]
> > Sent: Tuesday, March 05, 2002 1:27 PM
> > To: David Knight
> > Cc: address@hidden; address@hidden;
> > address@hidden; address@hidden
> > Subject: Re: SUNY-Albany and AFIT
> > 
> > 
> > David Knight wrote:
> > > 
> > > Hi Anne,
> > > 
> > >     Jeff Sitler at afit tells me the machine is/will
> > > be known as fujita.afit.edu (not that it really matters
> > > since the name seems to be irelavant...).
> > > We have an allow for both the machine name and the IP
> > > number. It appears that when they they connect the request
> > > comes from the ip#
> > > 
> > > Mar 04 19:30:34 redwood rpc.ldmd[6793]: gethostbyaddr: failed for
> > > 129.92.9.62
> > > Mar 04 19:30:34 redwood 129.92.9.62[6827]: Connection from 129.92.9.62
> > > Mar 04 19:30:34 redwood 129.92.9.62(feed)[6827]: Starting Up:
> > > 20020304190240.510
> > >  TS_ENDT {{NNEXRAD|UNIDATA,  ".*"}}
> > > Mar 04 19:30:34 redwood 129.92.9.62(feed)[6827]: topo:  129.92.9.62
> > > NNEXRAD|UNIDATA
> > > 
> > > Even though the gethostbyaddr fails we apparently accept their
> > > connection (I'm not sure if this is because we have an explicit
> > > allow for the IP address, or if it is a change we made to our ldm
> > > configuration some time ago that I simply forget right now).
> > > There is no entry for fujita in /etc/hosts or our NIS+ tables.
> > > I really don't like feeding an IP number - it doesn't bother
> > > me with the NOAAPORT feed, but, given the restrictions we face
> > > with the NLDN feed I'd really much rather be able to document
> > > we are feeding an .edu site.
> > > 
> > > Hope this helps...
> > > David
> > > 
> > > p.s. I understand that afit has security concerns, but, they are
> > > not alone in this regard. In fact I am becoming less and less
> > > comfortable feeding an essentially anonymous host at what appears
> > > to be a military site. For example, what if despite our best
> > > efforts either redwood or striker get hacked, and the hacker
> > > uses these machines to send nasty stuff over the IDD to
> > > the afit site - should we even be taking that risk, or, be
> > > exposing ourselves to that responsibility? Also IP numbers
> > > can be easily spoofed, and a military machine might be a likely
> > > target for this. If I had any hair left I'd probably have to
> > > say I must be having a "bad hair day" ;-)
> > > 
> > 
> > Hi David,
> > 
> > Thanks for the information.  You raise very good points regarding both
> > proprietary feeds and security.  I will raise these issues here for
> > discussion, this time in the context of .mil sites.  (LDM security is a
> > perennial topic.)  Although, Jeff Stitler assures us that fujita is a
> > .edu site.  This morning I was able to confirm that on the AFIT network
> > fujita is known as fujita.afit.edu.
> > 
> > Regarding the hacking potential, one significant safeguard is that the
> > LDM uses its own protocol.  Thus, there are only a few messages to which
> > the ldm will respond (HEREIS, COMINGSOON, BLKDATA, etc.), and it will
> > respond in well understood, predictable ways.  It would be very hard to
> > write some nefarious executable, wrap it properly, send it properly, and
> > get the remote ldm to do something beyond just stuffing it in the queue.
> > 
> > And, due to your message I learned something about the LDM this
> > morning.  When the IP addresses is used in ldmd.conf, the server will
> > try only once to do a reverse lookup.  And, that lookup doesn't need to
> > succeed, just as we saw in your logs above.
> > 
> > This means that some machine could spoof being AFIT by providing that IP
> > address to you and get you to feed them data.  This could be an issue
> > for your proprietary data.  I can understand your wanting to verify that
> > you're feeding a .edu.  With AFIT's restrictions that currently can't be
> > done.  We'll have to leave it to you to decide whether or not to feed
> > such sites.  
> > 
> > Regarding security on AFIT's side, AFIT is using names in their
> > ldmd.conf file instead of IP addresses, which forces the forward and
> > reverse lookup requirement.  So, it would be harder for some machine to
> > spoof being redwood.atmos.albany.edu.
> > 
> > I don't mean to dismiss your concerns, only to allay them.  We must
> > always be thinking about security.  
> > 
> > Anne
> > ***************************************************
> > Anne Wilson                 UCAR Unidata Program            
> > address@hidden                     P.O. Box 3000
> >                                       Boulder, CO  80307
> > ----------------------------------------------------
> > Unidata WWW server       http://www.unidata.ucar.edu/
> > ****************************************************
> > 
>