Re: Oracle CRS and Split Brin

  • From: amonte <ax.mount@xxxxxxxxx>
  • To: "Kevin Closson" <kevinc@xxxxxxxxxxxxx>
  • Date: Fri, 30 Mar 2007 01:34:11 +0200

Hi Kevin

What are the voting disk roles in a split brain scenario?

Thanks




On 3/29/07, Kevin Closson <kevinc@xxxxxxxxxxxxx> wrote:



Kevin mentioned that in a split brain scenario the lower node would
survive always



…It appears as though it is the lowest as per the CRS node number, but not
the lowest IP…but again, I'm likely confusing things because the formula is
not clearly available and we are talking about true split brain where there
is a clean hemispherical split of the cluster (1/1,2/2). Split brain
resolution techniques/algorithms are the family jewels of clusterware. The
way things are these days, you generally find out more about how they work
after they are patented as is the case here at PolyServe (I mean HP):



PolyServe 
7,148,853<http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=7149853.PN.&OS=PN/7149853&RS=PN/7149853>



  7,148,853 is on of the many algorithms that come into play when there
are cluster meltdowns in a PolyServe cluster.

Other related posts: