Re: [IPFIX] No active/inactive timeout definitions in any IPFIX RFCs? Idle versus inactive terminology? (part of draft-ietf-bmwg-ipflow-meth-09 review)

[Brian Trammell <[e-mail]>]

Hi, Benoit, all,

Active and idle timeout are essentially Metering Process implementation-specific parameters. I suspect most MPs dealing with flows will have _something_ that looks like an active timeout, and something that looks like an idle or inactive timeout, regardless of what they call them. Exactly how active and idle timeouts are implemented, and the effect they have on the duration and export order of exported flows, probably isn't a protocol matter. So I'm not sure how specifically defined they need to be.

"Active" seems to be "active" everywhere. The choice of picking idle or inactive would seem then to be arbitrary. Given that 5102 predates all other (directly-IPFIX-related) references, and there's an IE in the IANA registry called flowIdleTimeout, I'd tend to go with "idle" for this one.

As to _where_ and _how_ to define these, I don't know, and as little as possible.

Cheers,

Brian

On Apr 10, 2012, at 1:07 PM, Benoit Claise wrote:

> Dear all,
> 
> Reading draft-ietf-bmwg-ipflow-meth-09.txt, I realized that this draft defines two new terms: Active Timeout and Inactive Timeout.
> I was thinking: surely, we have defined those terms already in some IPFIX RFCs.
> Actually, I was unable to find any references.... Have I overlooked something?
> 
> The only pointers I could find are in RFC5102
> 
> 5.11.1.  flowActiveTimeout
> 
> 
>    Description:
>       The number of seconds after which an active Flow is timed out
>       anyway, even if there is still a continuous flow of packets.
>    Abstract Data Type: unsigned16
>    ElementId: 36
>    Status: current
>    Units: seconds
> 
> 
> 5.11.2.  flowIdleTimeout
> 
> 
> 
>    Description:
>       A Flow is considered to be timed out if no packets belonging to
>       the Flow have been observed for the number of seconds specified by
>       this field.
>    Abstract Data Type: unsigned16
>    ElementId: 37
>    Status: current
>    Units: seconds
> 
> 
> 5.11.3.  flowEndReason
> 
> Description:
>       The reason for Flow termination.  The range of values includes the
>       following:
> 
>       0x01: idle timeout
>             The Flow was terminated because it was considered to be
>             idle.
> 
>       0x02: active timeout
>             The Flow was terminated for reporting purposes while it was
>             still active, for example, after the maximum lifetime of
>             unreported Flows was reached.
> 
>       ...
> 
> 
> 
> 
> 
> Here is one issue.
> Some documents refer to inactive timeout
>     RFC5472, 
>     ipfixMeteringProcessCacheInactiveTimeout in http://tools.ietf.org/html/draft-ietf-ipfix-rfc5815bis-03
>     inactiveTimeout in http://tools.ietf.org/html/draft-ietf-ipfix-configuration-model-10
> While some others refer to idle timeout
>     RFC 5102, 
>     draft-ietf-bmwg-ipflow-meth-09
> 
> Any proposal to fix this problem?
> 
> Regards, Benoit (OPS A.D.)
> 
> _______________________________________________
> IPFIX mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/ipfix