On Wed, 11 Jul 2007, Lars Marowsky-Bree wrote:

> On 2007-07-11T10:31:06, David Lee <[EMAIL PROTECTED]> wrote:
>
> > If it is a transition exercise (staged migration across releases from one
> > to another) temporarily requiring two locations, then wouldn't it be
> > better to make the "old" location be a tiny script that produces a
> > "warning: deprecated" message and then invokes the new one?  (That's what
> > Andrew and I have just done with the migrating "ocf-shellfuncs".)
>
> You do NOT want to log a warning for every log message logged. That
> would go too far ;-)

Yes, I could see there might be a possibility to become interestingly
recursive...

But my earlier question still stands:  Is there any reason for having it
installed in two places?  Can we simply decide on a single place?

If that path is different from in earlier releases, does the change matter
at all?  (It couldn't really matter (could it?) if "ha_logger" is simply a
detail internal to the main heartbeat distribution.)

But if it is vaguely public (e.g. known by end-user RAs) then we ought to
do something warning-like from the old, deprecated location to warn the
user (even if is a warning-per-warning thing).

-- 

:  David Lee                                I.T. Service          :
:  Senior Systems Programmer                Computer Centre       :
:  UNIX Team Leader                         Durham University     :
:                                           South Road            :
:  http://www.dur.ac.uk/t.d.lee/            Durham DH1 3LE        :
:  Phone: +44 191 334 2752                  U.K.                  :
_______________________________________________________
Linux-HA-Dev: [email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev
Home Page: http://linux-ha.org/

Reply via email to