Re: [IPFIX] [Sender: [email protected]] RFC5101: time first flow dropped and time last flow dropped

Dear all,

On 20.09.2011 22:32, Paul Aitken wrote:
> Benoit,
>
>> 2. if the IPFIX exporter is configured from [IPFIX-CONF], what should
>> be the value in meteringProcessId?
>>      [IPFIX-CONF] doesn't mention the meteringProcessId, but a cache name.
>
> IPFIX-CONF (or any configurer) wouldn't set the process ID; it's read-only.
>
> The configurer would create the process (technically, add the config
> which causes the process to be created by the OS), then could be told
> (or could discover) what the process ID is.
> The cache name could be used to relate the given config to the process ID.

Correct. In general, IDs cannot be considered to be configurable.

>>       From figure 1 and 2, it seems that there is a one to one matching
>> between the cache and the metering process.
>>      If this is the case, a solution could be to add a meteringProcess
>> inside the cache in the figure 12 in [IPFIX-CONF] below
>
> Do you mean, a "meteringProcessId" (IANA #144) as a read-only property?
>
> And also an "exportingProcessId" (IANA #145) as a read-only property in
> the ExportingProcess?

I'm not sure about the one-to-one relationship between Metering Process 
and Cache. On purpose, IPFIX-CONF does not make a statement because it 
is not an architectural document. A Metering Process may have multiple 
Caches. But I do not think that a Cache would belong to multiple 
Metering Process. So, there is only one meteringProcessId which is 
associated with a Cache.

We do not have selectorId, exportingProcessId, and meteringProcessId in 
the configuration data model. If we think about adding 
meteringProcessId, it may be worth thinking about the other IDs as well.

I do not see a problem if these are defined as state parameters (e.g. 
read-only).

selectorId is a bit different because the scope is OD, not IPFIX device. 
Therefore, the same ID value may appear for different Selectors of the 
same device.

Gerhard
_______________________________________________
IPFIX mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/ipfix