Re: Veritas storage foundation HA for windows

From: Mike Epprecht \(SQL MVP\) (mike_at_epprecht.net)
Date: 01/18/05 

Date: Wed, 19 Jan 2005 00:58:26 +0100

Hi

Veritas "Storage Foundation" is a data solution. It includes Volume Manager,
which is just a SAN LUN to OS Volume mapping driver (We use it at work).
This is a solution to manage your data, not the services like SQL Server.

To make an application cluster suitable ("cluster aware" is for used for
software running on a cluster), you just need to have you error handler
re-try a connection a number of times when the connection drops. A
no-brainer if an application is written correctly.

Most of the time that we have hardware failure, the failure causes
corruption of data. RAID controllers go bad, write junk and even the best
cluster is screwed. No matter what HA solution you use, you are still
limited to what the weakest component can do to your system.

Hope you have multiple generators, multiple UPS's, multiple switches,
multiple SANs, multiple ...well of everything.

Whatever you use, test it for a good period of time. You might find that
what you though was bullet proof has worse uptime than a single server.

Regards
--------------------------------
Mike Epprecht, Microsoft SQL Server MVP
Zurich, Switzerland

IM: mike@epprecht.net

MVP Program: http://www.microsoft.com/mvp

Blog: http://www.msmvps.com/epprecht/

"Todd" <Todd@discussions.microsoft.com> wrote in message
news:739E7C9C-D30A-4D2A-940F-64823F32F757@microsoft.com...
> Hi Mike,
>
> Thanks for the reply. We are mostly worried about HW failure. We do not
have
> time to re-write the application and the down time has to be 0 (or almost
> zero).
>
> Has anyone used this product from Veritas. They claim "immediate"
failover.
> I am looking to find out if someone could tell me how it works (i.e. is
> application re-write required and what is the down time for the failover).
>
> Thanks,
> Todd
>
> "Mike Epprecht (SQL MVP)" wrote:
>
> > Hi
> >
> > As soon as SQL Server has to be shut down and start up (used by all
> > clustering / HA availability solutions), you need to re-connect.
> >
> > SQL Server 2005 will have database mirroring, resulting in the MDAC
stack
> > doing the re-direction to the new machine, with about 5 second downtime.
> >
> > Trying 100% uptime with an application that does not support it is
really
> > throwing money at a bad problem.
> >
> > What is your biggest worry? Hardware, OS or SQL server failure causing
you
> > downtime?
> >
> > Regards
> > --------------------------------
> > Mike Epprecht, Microsoft SQL Server MVP
> > Zurich, Switzerland
> >
> > IM: mike@epprecht.net
> >
> > MVP Program: http://www.microsoft.com/mvp
> >
> > Blog: http://www.msmvps.com/epprecht/
> >
> > "Todd" <Todd@discussions.microsoft.com> wrote in message
> > news:5E827336-E5B3-4FBA-9FFB-2A9B386C237D@microsoft.com...
> > > Hi,
> > >
> > > We have been looking at Stratus and Marathon Technologies for High
> > > Availability solutions for SQL Server. Zero time is a requirement as
well
> > as
> > > not having to make our application "cluster-aware". Microsoft
Clustering
> > > requires application modifications and almost 2 minutes of downtime
for
> > the
> > > failover to happen.
> > >
> > > There is a product out on the market called: "Veritas storage
foundation
> > HA
> > > for windows". Does this product allow for zero application down time
and
> > and
> > > no application rewrite? Is this an alternative to a Stratus box or a
> > Marathon
> > > Technolgoes solution for HA?
> > >
> > > If someone could get back to me tonight it would be appreciated.
> > >
> > > Thanks,
> > > Todd
> >
> >
> >

Relevant Pages

  • Re: Newb question regarding cluster load balancing
    ... > You can install multiple instances of SQL Server on a failover cluster. ...
    (microsoft.public.sqlserver.clustering)
  • Re: US customs poor service
    ... point of failure, there are multiple network connections, multiple ... and also multiple servers in a cluster or load balanced ... arangement. ...
    (rec.travel.air)
  • Re: US customs poor service
    ... point of failure, there are multiple network connections, multiple ... and also multiple servers in a cluster or load balanced ... arangement. ...
    (rec.travel.air)
  • Re: Please Help - Restore Node
    ... You may have to evict the bad node formally from the cluster and then rejoin ... How to recover from failover cluster failure in Scenario 1 ... Run SQL Server Setup and remove Node 1. ... Install new hardware to replace the failed hardware in Node 1. ...
    (microsoft.public.sqlserver.clustering)
  • SQL2000 error: ... When connecting to SQL Server 2005
    ... An error has occurred while establishing a connection to the server. ... this failure may be caused by the fact that ... under the default settings SQL Server does not allow remote connections. ... We booted the cluster last night, now we're getting the above error in the ...
    (microsoft.public.dotnet.framework.adonet)