[Linux-cluster] Graceful Degradation
Roger Peña
orkcu at yahoo.com
Fri Dec 14 18:28:25 UTC 2007
--- Rajagopal Swaminathan <raju.rajsand at gmail.com>
wrote:
> Greetings,
>
> While in theory one can have all sorts of fantasy,
> one has to hve the
> resources to fulfill the *theoraticcal*
> requirements.
>
>
> Often the educational institutions just don't have
> that.
>
> Let us be pragmatic and find a solution without
> hardware fencing devices
> without the cost involved.
>
> there has to be a mechanism to reboot nodes with say
> "reboot -f" _without_
> corrupting the local FS and fence.
sometimes you can't avoid a cool reboot :-(
bad and unpredictable things happen :-(
and on those situations the power button usually is
the only option to get the server back
> Open source can do better than that....
yeap but how many times do you had had to put your
finger over the power button and make a cool reboot?
I know this is the worst scenario but it happen :-(
and it doesn't matter if you run OSS
cu
roger
__________________________________________
RedHat Certified ( RHCE )
Cisco Certified ( CCNA & CCDA )
____________________________________________________________________________________
Looking for last minute shopping deals?
Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping
More information about the Linux-cluster
mailing list