[LinuxFailSafe] FailSafe status

Lars Marowsky-Bree lmb@suse.de
Fri, 4 Oct 2002 21:26:49 +0200


--vkogqOf2sHV7VnPd
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

Hi all,

as you all have probably noticed by now, the work on FailSafe has been rath=
er
limitted in the last few months already.

Given the events of the last months / year (SGI has pulled out of active Li=
nux
FailSafe development because of the situation in the US, which is also the
reason why SuSE US cut development jobs in 2000 which had been assigned to
FailSafe etc), we have put FailSafe into "maintenance mode" for our existing
customer base.

Put bluntly, we (SuSE) have no roadmap to support it beyond SuSE Linux
Enterprise Server 7 for IA32; existing customers and installations will of
course still be supported.

At this point in time, we are simply lacking the resources to port the
FailSafe GUI (which has it's own complete framework and all) to the newer
JDKs, or rework the FailSafe core to be "uptodate" with current developments
and requirements (ie, the Sun RPC code really has to go).

I've decided that our resources are better spent on working on the Open
Clustering Framework (beneficial to Linux as a whole), and working on
enhancing heartbeat to integrate all the cool tricks I've learned in FailSa=
fe
and more; put into marketing terms, our committment to Open Source clusteri=
ng
is stronger than ever ;-) And UnitedLinux will already show a few HA
requirements finally being part of the Linux core, ie like multipathing etc.

I encourage you to join the OCF lists and development; we have a lot of cool
plans and as usual too few hands for all the work ;-)

Just to polish the crystal orb, we'll be working with the EVMS team for
clustered volume management, political things like official affiliation with
the Free Standards Group to create a clustering standard for Linux (and all
"open" systems) etc.

FailSafe has been a very educational experience, and I've personally learne=
d a
lot about clustering and failover systems. I expect that quite a bit of the
concepts, algorithms will be either copied or serve as an inspiration for
future work.  It is always easier and cleaner to write something the second
time.

This is not the ultimate end of FailSafe yet - as I said, it is in
"maintenance" mode because of lack of resources, and given a big enough
customer, this may change.

If you have any questions, please feel free to contact me.


Sincerely,
    Lars Marowsky-Br=E9e <lmb@suse.de>

--=20
Engineering Lead HA/Clustering
Research and Development, SuSE Linux AG
=20
``Immortality is an adequate definition of high availability for me.''
	--- Gregory F. Pfister


--vkogqOf2sHV7VnPd
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iEYEARECAAYFAj2d63gACgkQudf3XQV4S2cgwQCfbAbE+3rYZIx9SrSSGdNYmqwp
r9IAn33ZaAuYBe2eFJVcZfC7I+OHW3Dx
=FwMs
-----END PGP SIGNATURE-----

--vkogqOf2sHV7VnPd--