RE: rfc2518 issue: LOCK_REFRESH_BY_METHODS

Hey Geoff...

Could you make an explicit proposal as to how the spec should change?

BTW... I don't care if the server choses to do this or not.  I doubt
refreshing the lock creates a functional problem since I doubt a client
would be written to wait for a time out and choke if it didn't unlock the
resource.  OTOH, I don't see the point in encouraging servers to refresh
the lock either.   If a server developer/administrator discovers a reason
that this refreshing policy is good, they can use it, but we don't need to
encourage it since it might be pointless and costly.

 So my recommendation would be to silent on this... or perhaps somehow
express flexibility.

Anyway, perhaps you could suggest what change to 2518 you'd like... :-)

J.

------------------------------------------
Phone: 914-784-7569,   ccjason@us.ibm.com

Received on Saturday, 11 August 2001 19:53:16 UTC