[Nsi-wg] Topology versions

John MacAuley john.macauley at surfnet.nl
Fri Sep 27 11:28:00 EDT 2013


I agree with this recommendation.  I was specifying it in the Discovery service as well.  Perhaps we should formalize this in a document for Topology v1 (or v2 not 100% sure what we are calling the current mechanism)?

John

On 2013-09-27, at 7:44 AM, Henrik Thostrup Jensen <htj at nordu.net> wrote:

> On Thu, 26 Sep 2013, John MacAuley wrote:
> 
>> I have noticed some of the NSA that support dynamic topology retrieval are updating the "version" attribute of the NSA element every time a retrieval occurs.  This has the unwanted side effect of appearing to be a new topology each time it is retrieved.  If possible, could you only update the "version" attribute after an NSA restart, or when you have had an actual topology change.  This will help to reduce the churn in recomputing topology when polling for updates.
> 
> Indeed.
> 
> To further reduce load HTTP has a standard mechanism that can be used for this: Provide a Last-modified header in the payload indicating the version of topology. Furthermore, the client has to specify an If-modified-since in the request. If the resource has not been modified since then, the servouce should send back a 304 with an empty body. Like this:
> 
> htj at scandium> curl -fSsihttps://nsi.nordu.net:9443/NSI/topology/nordu.net:2013.xml HTTP/1.1 200 OK
> Date: Fri, 27 Sep 2013 11:39:14 GMT
> Last-Modified: Fri, 27 Sep 2013 11:39:01 GMT
> Content-Length: 4782
> Content-Type: text/html
> Server: TwistedWeb/11.1.0
> [snip]
> 
> And:
> 
> htj at scandium> curl -i -H 'If-Modified-Since: Fri, 27 Sep 2013 11:41:40 GMT'  https://nsi.nordu.net:9443/NSI/topology/nordu.net:2013.xml
> HTTP/1.1 304 Not Modified
> Date: Fri, 27 Sep 2013 11:40:19 GMT
> Content-Length: 0
> Server: TwistedWeb/11.1.0
> 
> And I've just added in this in OpenNSA for good measure :-).
> 
> And yes, the timestamp format must be the ugly RFC850 format.
> 
> I realize this is already established functionality and has good tooling, and we don't get to invent our own thing, so I reckon this will meet some resistance :-)
> 
> 
>    Best regards, Henrik
> 
> Henrik Thostrup Jensen <htj at nordu.net>
> Software Developer, NORDUnet
> 
> _______________________________________________
> nsi-wg mailing list
> nsi-wg at ogf.org
> https://www.ogf.org/mailman/listinfo/nsi-wg



More information about the nsi-wg mailing list