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

[LDM #BIG-900661]: Terminated obsolete upstream

Brice, 

> May have some good news and I do have an additional question, now that I 
> 'think' I have a better understanding.  I got Richard to change his ldmd.conf 
> requests so that they were mirror images of each other on the two clients and 
> had him restart his LDM sessions.

I'm not sure what you mean by "mirror images of each other on the two clients". 
Please elucidate.

> I could see in the logs that the two clients made connections through the 
> tunnel and that the requests were reversed.

I'm not sure what you mean by "the requests were reversed". Please elucidate.

> So far I have not seen the obsolete termination messages.  I am going to 
> continue to monitor the logs today.

That's good news!

> So with this potential success, I have a question.  As I add additional 
> external customers to this tunneling scenario it is going to become 
> increasingly difficult to make sure that the arrangements of their requests 
> are different.  So I had a thought (painful and dangerous that is).  Could I 
> add a different bogus pattern to each of the requests that would make them 
> sufficiently different to still work?  For instance, if I assigned each 
> client a designator that would indicate the client, I should have both a 
> different pattern and a quick look at who is being fed.  As an example, 
> Richard is running servers, Sun and Moon,  if he changed his pattern:
> 
> Sun client:   request   FT4   "Sun|^WT.*        "     134.xxx.xxx.aaa
> Moon client:  request   FT4  "Moon|^WT.*    "  134.xxx.xxx.aaa,
> 
> then LDM should see those as different patterns and let them through.  
> Because I know there will never be any 'Sun' or 'Moon' patterns in our data, 
> I am thinking this would work.  Your thoughts?  We plan on testing it later 
> today or tomorrow if things look good with the current configurations.  Might 
> be a useful thing to suggest to folks behind NAT's too.

Interesting idea! I see no reason why it shouldn't work -- as long as you can 
guarantee that, for example, the strings "Sun" and "Moon" will never, ever 
appear *anywhere* in the product-identifiers. The matching would probably be 
quicker and less risky for the patterns "^Sun|^WT" and "^Moon|^WT", which, by 
the way, are equivalent to "^(Sun|WT)" and "^(Moon|WT)".

> Thanks again for everything you folks do with LDM.  We couldn't live without 
> it.

Oh, I'm sure you'd find something else. :-)

> Just for your information, it is going to become the standard streaming data 
> communication protocol between the NASA centers doing spaceflight support 
> (JSC, KSC and MSFC), and the Cape Canaveral Air Force Station weather system.

Yikes! Just what I need: more support emails! :-)

Seriously, we're honored.

You know about our training workshops, right? We also visit for on-site 
training.

> Brice
> 
> Brice Biggerstaff, CISSP
> Johnson Space Center
> Weather Decision Support System
> MIDDS Software Support Lead
> 281-853-3011 (w)
> 713-764-2601 (p)
> address@hidden  (alpha pager for text and email)
> 
> Res Confacti Erimus
> “We Get Things Done!”
Regards,
Steve Emmerson

Ticket Details
===================
Ticket ID: BIG-900661
Department: Support LDM
Priority: Normal
Status: Closed