What is the minimum number of fault domains required to achieve vRealize Operations continuous availability with a primary node, data node, and witness node?

Question

Options

• A4
• B1
• C3
• D2

• A
  5% (2)
• B
  3% (1)
• C
  3% (1)
• D
  89% (34)

Explanation

Option D (2 fault domains) is correct because vRealize Operations Continuous Availability requires the primary node and data node to reside in two separate fault domains to survive a site-level failure - the witness node does not require its own dedicated fault domain and can coexist within one of the two existing domains, serving only as a tiebreaker to maintain quorum.

Why the distractors are wrong:

• A (4) - far exceeds the requirement; no architectural reason exists to split these three components across four domains.
• B (1) - placing all nodes in a single fault domain defeats the entire purpose of CA; one failure would take down the whole cluster.
• C (3) - one fault domain per node seems intuitive but is unnecessary; the witness is lightweight and does not need isolation from the primary or data nodes to fulfill its quorum role.

Memory tip: Think "two sides, one referee" - the primary and data node each need their own "side" (fault domain) to survive independently, while the witness is just the referee and can sit on either side.

No community discussion yet for this question.