SAP Cluster Frequent Failover
A company runs SAP S/4HANA on AWS using an ASCS and PAS cluster with SUSE Linux Enterprise Server. The cluster fails over unexpectedly several times a day. What is the most likely cause and solution?
Quick Answer
Frequent, unexpected failovers in a two-node SAP cluster are a classic symptom of quorum problems rather than an actual application or hardware failure, and recognizing that distinction is the key to this question. In a two-node cluster, if the vote count needed to maintain quorum becomes insufficient, which can happen due to transient communication issues between the two nodes even briefly, the cluster can interpret that as a reason to trigger a failover as a safety measure, even though nothing is actually wrong with the workload itself. That's why the fix isn't about tuning performance or networking directly, but about giving the cluster a more stable way to determine quorum: adding a third node as a quorum device, or configuring a witness such as an S3 bucket used as a quorum disk in SUSE HAE, gives the cluster an additional, independent vote so that a brief blip between the two primary nodes doesn't tip it into losing quorum and failing over unnecessarily. This is worth distinguishing from other plausible-sounding causes: high network latency would show up as communication delays rather than repeated full failovers, insufficient memory would produce performance symptoms rather than cluster-level failover behavior, and incorrect ASCS configuration would typically prevent SAP from starting cleanly rather than cause an already-running cluster to fail over repeatedly. When a cluster fails over often without a clear underlying resource or application problem, quorum design is usually the first thing to check.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Cluster quorum loss due to insufficient votes; add a third node or configure a witness
In a two-node SAP cluster, if the quorum vote count is insufficient, the cluster may lose quorum and trigger unnecessary failovers. The solution is to add a third node as a quorum device or configure a witness (e.g., an S3 bucket as a quorum disk in SUSE HAE). Option A is incorrect because high network latency would cause communication delays, not frequent failovers. Option C is incorrect because insufficient memory would cause performance issues, not cluster failovers. Option D is incorrect because incorrect ASCS instance configuration would cause SAP startup failures, not cluster failovers.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
High network latency between nodes; use placement groups
Why it's wrong here
Latency might cause timeouts but not multiple daily failovers.
- ✓
Cluster quorum loss due to insufficient votes; add a third node or configure a witness
Why this is correct
Quorum loss triggers failover; adding a node stabilizes the cluster.
- ✗
Insufficient memory on the cluster nodes; increase memory
Why it's wrong here
Memory shortage doesn't cause failovers.
- ✗
Incorrect ASCS instance configuration; verify SAP profiles
Why it's wrong here
ASCS config error wouldn't cause cluster failovers.
Quick reference
AWS S3 Storage Class Comparison
| Storage Class | Min Duration | Retrieval | Use Case |
|---|---|---|---|
| S3 Standard | None | Immediate | Frequently accessed data |
| S3 Standard-IA | 30 days | Immediate | Infrequent access, rapid retrieval |
| S3 One Zone-IA | 30 days | Immediate | Non-critical infrequent data |
| S3 Intelligent-Tiering | None | Immediate–hours | Unknown or changing access patterns |
| S3 Glacier Instant | 90 days | Milliseconds | Archive with instant retrieval |
| S3 Glacier Flexible | 90 days | Minutes–hours | Archive, flexible retrieval |
| S3 Glacier Deep Archive | 180 days | Hours | Long-term compliance archive |
Go deeper
Related to this question
About these practice questions
This PAS-C01 question is part of Courseiva's 1,616-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
Same concept, more angles
1 more way this is tested on PAS-C01
These questions test the same concept from different angles. Work through them to make sure you can recognise it however the exam phrases it.
Variation 1. A company runs SAP S/4HANA on AWS with a High Availability (HA) cluster spanning two Availability Zones (us-east-1a and us-east-1b). The cluster uses an Amazon FSx for NetApp ONTAP file system for the global transport directory. Recently, the cluster experienced a split-brain situation, leading to data corruption. The administrator wants to prevent this from recurring. The current setup uses a single FSx file system mounted on both cluster nodes. What should the administrator do to prevent split-brain?
medium- A.Add a third cluster node in a third Availability Zone
- B.Increase the memory on both EC2 instances
- ✓ C.Configure a STONITH (Shoot The Other Node In The Head) mechanism using AWS Lambda
- D.Set up cross-region replication for the FSx file system
Why C: STONITH (Shoot The Other Node In The Head) is a fencing mechanism that ensures a failed node is forcefully isolated or powered off before the other node takes over, preventing split-brain scenarios in High Availability clusters. Option A is incorrect because adding a third node increases complexity without addressing the root cause of split-brain. Option B is incorrect because increasing memory does not affect cluster communication or fencing. Option D is incorrect because cross-region replication is for disaster recovery, not for preventing split-brain.
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This PAS-C01 practice question is part of Courseiva's free Amazon Web Services certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the PAS-C01 exam.