On Fri, 2006-03-17 at 17:43 +0100, Rainer Gerhards wrote:
> Chris,
> 
> while I think this sounds very tempting, I also think there are some
> inherent problems with it:
> 
> #1 you do not know *where* (more precise: after how many octets) that
> element is present
> In extreme cases, it might only be valid after more then 64k
> 
> #2 it could become truncated
> Structured data is not guarded against truncation.
> 
> #3 architectural concerns
> I do not think it is appropriate for a lower layer to obtain information
> from an upper-layer field. That would require the lower layer to parse
> the upper layer field, which it conceptionally should not even be aware
> of.
> 
> All in all, I am in strong favour of a dedicated tls-transport only
> header for the octet count.

Absolutely agree and I also like the text based format.

-- 
Bazsi


_______________________________________________
Syslog mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/syslog

Reply via email to