[Syslog] TIMESTAMP and leap seconds

Hi all,

first of all, I would like to whish a happy new year to each of you!

I am now back in the office and at final edits to syslog-protocol. I
discovered one more thing: The current draft supports leap seconds.
There already is a lot of discussion whether or not leap seconds should
be introduced in the future. However, the way leap seconds are handled
will be largely invisible to a syslog sender (except where it is sitting
on a time-tracking device, which is highly unlikely). On the other hand,
leap second processing can be pretty complicated at the receiver side. I
expect that most implementations will not abide strict handling in any
case.

As such, I suggest that leap second support be removed from the
TIMESTAMP. Similarily, a sender with unknown time should then not use
the special TIMESTAMP but "-", which also keeps it consistent with the
rest of the header NIL values.

If nobody objects, I'll change this during the edit.

Rainer

_______________________________________________
Syslog mailing list
Syslog@lists.ietf.org
https://www1.ietf.org/mailman/listinfo/syslog