[HA news] Iris FailSafe for Linux OSS

[Stas Bekman <sb...@iname.com>]

An interesting news for mod_perl developers who need a High-Availability
[HA] Functionality for their Linux machines running mod_perl and other
servers. 

<QUOTE>
IRIS FailSafe running on Linux will enable a user to link two or more
servers together so that one transparently picks up the computing load
should the other fail. An IRIS FailSafe cluster running on Linux will
remove any single point of failure and allow applications to increase
availability to the level required for mission-critical data center
operations. This failover capability is an important component of the
maturation of Linux in an enterprise computing environment.
</QUOTE>

It's an opensource product (no license info yet).

See the press release:
http://biz.yahoo.com/prnews/000226/ca_sgi_sus_1.html

And the product:
http://www.sgi.com/software/failsafe/index.html

For more HA links see:
http://perl.apache.org/guide/download.html#High_Availability_Linux_Project

_______________________________________________________________________
Stas Bekman    mailto:sbekman@iname.com      http://www.stason.org/stas
Perl,CGI,Apache,Linux,Web,Java,PC     http://www.stason.org/stas/TULARC
perl.apache.org    modperl.sourcegarden.org   perlmonth.com    perl.org
single o-> + single o-+ = singlesheaven    http://www.singlesheaven.com

Re: [HA news] Iris FailSafe for Linux OSS

["G.W. Haywood" <ge...@jubileegroup.co.uk>]

Hi all,

On Sat, 26 Feb 2000, Gunther Birznieks wrote:

> My very limited experience with clustered solutions has not been
> entirely perfect.
> 
> To me, it seems that clustering software adds complexity to an OS
> config. This complexity makes bugs in a config easier to
> introduce. The config of the cluster software (rather than any one
> machine in the cluster itself) then becomes the new single point of
> failure. But at least it's another thing we can tell management when
> they balk at Linux/Apache/ModPerl enterprise level support...

I agree, this isn't as as easy as it might seem.  When you put
straight triplication into an airframe in an effort to improve
reliability you're hard pressed not to make it LESS reliable.

But shouldn't this have an [OT] flag?  What's the convention when an
initially on-topic post (well, Stas, you did *mention* mod_perl:)
finally gives up all pretence?  Will appending to the subject line
cause the thread to get lost at epigone?

73,
Ged.

Re: [HA news] Iris FailSafe for Linux OSS

[Gunther Birznieks <gu...@extropia.com>]

It should be interesting to see how well this and other products like it
really work when put to the test of human system administrators. My very
limited experience with clustered solutions has not been entirely perfect.

To me, it seems that clustering software adds complexity to an OS config. This
complexity makes bugs in a config easier to introduce. The config of the
cluster software (rather than any one machine in the cluster itself) then
becomes the new single point of failure. But at least it's another thing we
can tell management when they balk at Linux/Apache/ModPerl enterprise level
support...

But that's just my opinion, which has a possibility of being fairly wrong most
of the time...

"Jeffrey W. Baker" wrote:

> Stas Bekman wrote:
> >
> > An interesting news for mod_perl developers who need a High-Availability
> > [HA] Functionality for their Linux machines running mod_perl and other
> > servers.
> >
> > <QUOTE>
> > IRIS FailSafe running on Linux will enable a user to link two or more
> > servers together so that one transparently picks up the computing load
> > should the other fail. An IRIS FailSafe cluster running on Linux will
> > remove any single point of failure and allow applications to increase
> > availability to the level required for mission-critical data center
> > operations. This failover capability is an important component of the
> > maturation of Linux in an enterprise computing environment.
> > </QUOTE>
>
> And all you need is a cluster of inexpensive Origin boxes.
>
> Ehem.
>
> -jwb

Re: [HA news] Iris FailSafe for Linux OSS

["Jeffrey W. Baker" <jw...@acm.org>]

Stas Bekman wrote:
> 
> An interesting news for mod_perl developers who need a High-Availability
> [HA] Functionality for their Linux machines running mod_perl and other
> servers.
> 
> <QUOTE>
> IRIS FailSafe running on Linux will enable a user to link two or more
> servers together so that one transparently picks up the computing load
> should the other fail. An IRIS FailSafe cluster running on Linux will
> remove any single point of failure and allow applications to increase
> availability to the level required for mission-critical data center
> operations. This failover capability is an important component of the
> maturation of Linux in an enterprise computing environment.
> </QUOTE>

And all you need is a cluster of inexpensive Origin boxes.

Ehem.

-jwb