SEO

After something like 3 years of working with HACMP we've come to the
inavoidable conclusion that HACMP sucks. It's way too fragile for a
product that's supposed to provide you high avaliability.

Which other cluster products are there that are more reliable than
hacmp ?
I've heard of System Automation as an alternative, any comments about
that?


Thanks,
Daniel

danielrulez@gmail.com wrote:
> After something like 3 years of working with HACMP we've come to the
> inavoidable conclusion that HACMP sucks. It's way too fragile for a
> product that's supposed to provide you high avaliability.
>
> Which other cluster products are there that are more reliable than
> hacmp ?
> I've heard of System Automation as an alternative, any comments about
> that?

Fragile? Look into Veritas or Sun Cluster and you'll run
back to HACMP in terms of being NOT fragile! E.g. it takes
HACMP a rather long time to start switching nodes. Personally
i found this a good feature, since HACMP can survive short
outages without getting into a failover/fallback frenzy
like other cluster products.
Maybe you could be a bit more specific about what bothers
you with HACMP. There's always the possibility that a cluster
isn't the right thing for what you're trying to achieve.

Regards,

Frank

Actually, all standard failover cluster products I know just 'feel'
fragile. If You want high reliability, You need to run Your
applications really parallel, not failing over to another host if the
current host fails, but run it on all hosts at once, like it's done
i.e. in high end telco systems. I doubt it's easy to implement, and
You'll probably be better of by toughening HACMP.

I dare say that because after it I'll point You to
http://www.infrastructures.org - give it a look, they say they're
setting up full HACMP clusters in two hours without any manual
intervention. I think that's robust enough.

did you say high end telco systems.......hmm i work at one end and we
use HACMP (for the last 5 years, i have not found any swithcover
porblem)..

but i agree, if you want your system to be updated every other day with
new pathces or new products etc, cluster is very unfriendly.....and
that is the case for any cluster SOFTWARE REALLY....

florian.heigl@gmail.com wrote:
> Actually, all standard failover cluster products I know just 'feel'
> fragile. If You want high reliability, You need to run Your
> applications really parallel, not failing over to another host if the
> current host fails, but run it on all hosts at once, like it's done
> i.e. in high end telco systems. I doubt it's easy to implement, and
> You'll probably be better of by toughening HACMP.
>
> I dare say that because after it I'll point You to
> http://www.infrastructures.org - give it a look, they say they're
> setting up full HACMP clusters in two hours without any manual
> intervention. I think that's robust enough.

danielrulez@gmail.com wrote:
: After something like 3 years of working with HACMP we've come to the
: inavoidable conclusion that HACMP sucks. It's way too fragile for a
: product that's supposed to provide you high avaliability.
:
: Which other cluster products are there that are more reliable than
: hacmp ?
:

Yes, but they don't run on unix:

http://h71000.www7.hp.com/openvms/br...s/commerzbank/
hp Alphaserver technology helps Commerzbank tolerate disaster
on September 11

"testing disaster tolerance

While most large organizations today have plans for Disaster Tolerance
(DT), few have to put them to the test. The North American
headquarters of Commerzbank, located less than 100 yards from the
World Trade Center in New York City, put its DT plan into action on
September 11, 2001. Because Commerzbank relies on OpenVMS wide-area
clustering, volume shadowing and AlphaServer GS160 systems from HP,
the bank was able to function on September 11 because its critical
banking applications continued to run at the primary site and were
available from the bank's remote site..."

One can do a rolling software upgrade, including the operating system
of an OpenVMS cluster without shutting the entire cluster down.

The world record for uptime of a VMS cluster is held by the Irish National
Railway: 17 years:

http://h71000.www7.hp.com/solutions/...idation/dh.pdf
Enterprise Server Consolidation in the OpenVMS Environment


"...Users in multiple markets often need to be able to measure their
application uptime in years, rather than in hours per year. OpenVMS
clusters satisfy this requirement. For example, Irish National Railway
shut down its OpenVMS Cluster-based application after 17 years, and
then only because they moved to a process control-type system..."

Now back to AIX

--Jerry Leslie
Note: leslie@jrlvax.houston.rr.com is invalid for email

Meiosys's MetaCluster... (http://www.meiosys.com) ...

Meiosys provides unique software technologies that enable applications
to be dynamically moved from one server or set of servers to others
without disruption.

hopefully we'll see more of it in the future HACMP releases since IBM
bought Meiosys in June

How about Veritas Cluster Server on AIX ?