[asterisk-dev] IBM BladeServer / Clustered Linux

[Rich Adamson]

> >I understand the concern that everyone
> > has for this but, there can and will be failures of any kind. I have
> > had T1 circuits over 'SONET' fail because of a switch error on the
> > Telco side.
> 
> Of course hardware will fail. No doubt about that.
> Our systems have no single point of failure option. If Service provider
> chooses not to deploy a such system, then they may face legal cases should
> an E911 call fail in critical moment.

You should probably talk to a real lawyer before trying to suggest the above
is the truth. There have been (and will continue to be) failures in any system
including large central offices, and these do _not_ become legal cases. The 
issue can certainly become a case if you purposefully implement a sytem and 
represent that system as something different then what it really is. 

Some amount of "due diligence" is necessary for services, and having that
effort well documented would be considered a Good Thing.

> That is what I was referring to in my previous post. Some lawyer out there
> will find out that service provider did not
> insure that all E911 calls are sent through no single point of failure
> network and then use this information in the court of law.

Most don't even "insure" that any call will go through. ;)

> Some, will say... ISP, Inter-connect, Telco etc. networks could fail as well
> before it gets to service provider network. Sure, but then
> the blame is not on service providers any more once that fact is discovered.
> The fact that service provider may be offering the service over "third"
> party
> networks is another issue all together.

Your logic doesn't make any sense whatsoever. "If" it did, there wouldn't be
any PBX manufacturers (because a system might fail), etc, etc.