Pass the Pure Storage FlashArray Storage FlashArray-Storage-Professional Questions and answers with ExamsMirror

Purity Active/Active Architecture: Pure Storage FlashArrays use an Active/Active controller architecture. Under normal operating conditions, both controllers (CT0 and CT1) are healthy and independently serve I/O through their respective physical Fibre Channel ports.

The Scenario (Controller Failover): The output provided shows that the host initiator is seeing target WWNs associated with both CT0 and CT1, but the " Target " column indicates they are all being reached via the paths currently managed by one controller or through a specific failover mechanism.

Virtual WWNs and Transparency: In a controller failover or maintenance scenario, Pure Storage utilizes a feature where the WWNs of the " failed " or " offline " controller are logically moved to or presented by the " surviving " controller. This ensures that the host ' s MPIO (Multi-path I/O) software does not see a permanent " Path Down " error for those specific WWNs, but rather a transition.

Analyzing the Output: When you see CT1 port WWNs (e.g., 52:4A:93:78:55:2D:E3:10) appearing in the connectivity table in a way that implies they are being routed or presented through the physical infrastructure of CT0, it indicates that the array is in a state where one controller is assuming the identity/connectivity of the other. This is a key troubleshooting indicator that the array is likely undergoing a controller reboot, a Purity upgrade, or has experienced a controller hardware failure.

Why A and B are incorrect: * Option A: Multipathing software on the host handles path failures, but it wouldn ' t cause the array to report WWNs in this specific " cross-presented " manner in a connectivity log.

Option B: If there were a total outage on CT1 without this presentation mechanism, the paths would simply show as " Disconnected " or be missing from the initiator ' s view entirely.

User Mapping in FA File: On a Pure Storage FlashArray, User Mapping is the mechanism that translates identities between different protocols (like mapping a Windows SID to a Unix UID/GID) or between an external directory service (like Active Directory or LDAP) and the local file system permissions.

The Impact of Disabling Mapping: When User Mapping is disabled on an active NFS export, the FlashArray can no longer resolve the identity of the user attempting to access existing files. Because NFS (specifically NFSv3 and NFSv4.1 supported by Pure) relies on these identifiers to verify file ownership and ACLs, existing files—which are tagged with specific owner IDs—become effectively " orphaned " from the perspective of the incoming request.

Access vs. Creation: * Existing Files: Accessibility is lost because the system cannot verify that the user has the rights to read or modify the file without the mapping logic.

New Files: Interestingly, in many " No Mapping " configurations, a user may still be able to create new files (often defaulting to a ' nobody ' or ' anonymous ' UID depending on the export rules), but they will immediately lose the ability to manage or access them once created because the mapping link is broken.

Real-time Application: Unlike some legacy storage systems that require a service restart, Purity applies export policy changes dynamically . As soon as the " User Mapping " toggle is disabled, the logic is removed from the data path, impacting active sessions immediately.

ActiveCluster and Pod Roles: In a Pure Storage ActiveCluster or ActiveDR environment, a Pod is a management container for volumes. To move workloads or perform a planned failover between two arrays, you use the Promote and Demote commands.

The Reversal Process: When you have two pods in a replication relationship (Source and Target), data flows from the Promoted (Active/Source) pod to the Demoted (Passive/Target) pod.

When you Demote the current source, it transitions from a " read-write " state to a " read-only " (passive) state.

If the other pod in the pair is then Promoted , Purity automatically intelligently reverses the direction of replication . The array that was previously receiving data now begins sending incremental updates back to the original source.

Continuous Protection: This design ensures that you don ' t have to manually tear down and recreate replication schedules every time you switch production sites. The system tracks the metadata changes and ensures that only the delta (changed blocks) are sent in the new direction.

Why Option C is incorrect: If replication were simply paused, the two sites would quickly drift out of sync, making it impossible to fail back without a full baseline resync.

Why Option A is incorrect: Demoting a pod does not delete any data; it simply changes the access characteristics and replication role. The data remains fully intact on the storage media.

On a Pure Storage FlashArray, volume snapshots are immutable, read-only, point-in-time representations of your data. Because they cannot be attached directly to a host to be read or modified, you must use the Copy function to make the data usable.

The Purity operating environment allows you to copy a snapshot to two specific destinations:

A new volume: This effectively creates a clone. It provisions a brand-new, writable volume using the exact data footprint of the snapshot. This is incredibly useful for test/dev environments, offline reporting, or granular file recovery where you don ' t want to disrupt the original production volume.

An existing volume: This takes the data from the snapshot and completely overwrites the target volume. This is the standard procedure when you need to perform a full rollback of a corrupted volume, or when you want to quickly refresh a lower-level environment (like refreshing a QA database with yesterday ' s Production snapshot).

Here is why the other options are incorrect:

A new Snapshot (B & C): You cannot directly " copy " a snapshot to create another standalone snapshot. Snapshots are uniquely generated from active volumes. If you wanted to duplicate a snapshot ' s exact state, you would first copy it to a volume, and then take a new snapshot of that resulting volume.

Replication Fundamentals: On a Pure Storage FlashArray, replication is a snapshot-based process . To replicate a Protection Group (pgroup) to a target array, the system must first create a point-in-time snapshot of the volumes within that group on the source array.

The " Immutable " Rule: Even if the Local Snapshot Schedule is disabled, the act of replicating requires the existence of a local snapshot to serve as the " base " or " source " for the data transfer. Purity does not stream data directly from the active volume to the wire; it creates a snapshot and then replicates the unique blocks contained in that snapshot.

Accounting for Local Copies: When a Protection Group is configured for replication, every snapshot generated by the Replication Schedule is stored locally on the source array. These snapshots will remain on the source array until they are aged out according to the Local Retention policy (even if the local schedule itself is off, the retention policy still applies to those replicated snapshots).

Visibility: If you navigate to the Protection Group in the Purity GUI, you will see these snapshots listed under the " Snapshots " tab. They are functionally identical to local snapshots, meaning they can be used for local clones or restores without needing to pull data back from the target array.

Why Option A and C are incorrect: * Option A: If 0 snapshots were stored, there would be nothing to replicate.

Option C: While Purity uses the most recent snapshot as a reference for delta-tracking, it keeps the entire history of snapshots defined by your retention policy, not just a single one.

Understanding Data Reduction Rate (DRR): DRR is a measure of the effectiveness of Purity ' s deduplication and compression algorithms. It compares the amount of data written by the host to the amount of physical space actually occupied on the flash media after these processes have occurred.

Calculation: $\text{DRR} = (\text{Deduplication} \times \text{Compression})$.

Understanding Total Reduction: The Total Reduction metric is a broader " efficiency " value. It represents the ratio of the Total Provisioned (logical) space to the Physical space used.

The " Thin Provisioning " Factor: The primary difference between these two numbers is Thin Provisioning .

FlashArrays are 100% thin-provisioned by default. If a customer creates a 100TB volume but the host only writes 1TB of data, the " Total Reduction " will be massive because it counts all that unwritten, " virtual " space as " saved. "

The DRR, however, only looks at the 1TB that was actually written and calculates how well it was compressed and deduped.

Why Option C is incorrect: Both metrics generally account for the efficiency of snapshots. However, snapshots contribute to DRR through deduplication (since they share blocks with the source volume). The massive gap between DRR and Total Reduction is almost always mathematically attributed to the " unwritten " space provided by thin provisioning.

Key Takeaway: If your DRR is 5-to-1 but your Total Reduction is 50-to-1, it means your data is compressing/deduping well, but you have also provisioned a significant amount of volume space that hasn ' t been written to yet.

When an NFS client successfully mounts an export using the target ' s IP address, it proves that the fundamental network connectivity (routing, firewalls) and the storage protocol layer (NFS export policies, host access permissions) are functioning correctly.

However, if the exact same mount attempt fails when using the Fully Qualified Domain Name (FQDN) of the FlashArray file service, the issue lies entirely with name resolution. The Domain Name System (DNS) is responsible for translating human-readable FQDNs into the IP addresses required for network communication. If the client cannot reach the DNS server, or if the DNS server lacks the correct A or AAAA records for the FlashArray ' s file Virtual IP (VIP) addresses, the client won ' t be able to resolve the name to the IP, causing the mount command to fail.

Here is why the other options are incorrect:

Issue with the Active Directory (AD) controller (A): Active Directory is primarily used for directory services, user authentication, and authorization (such as mapping permissions for SMB or NFSv4). While AD environments usually include DNS, an " AD controller issue " in the context of storage protocols usually points to permission denials, not host name resolution failures. Furthermore, since the mount works via IP, basic access is already validated.

Issue with the OpenLDAP (C): Similar to AD, OpenLDAP provides directory services for user mappings (UID/GID) and authentication. It does not perform FQDN-to-IP resolution.

ActiveDR Bandwidth Sizing: ActiveDR is a continuous, asynchronous replication technology designed to provide near-zero RPO. Because it streams data continuously rather than in discrete snapshot intervals, the bandwidth between the source and target arrays must be able to handle the application ' s write workload.

Handling Write Bursts: Application workloads are rarely flat; they have peaks and valleys. If you size the bandwidth exactly to the average change rate, any burst in write activity will cause the replication lag to increase, thereby increasing your RPO.

The 30% Rule: Pure Storage best practices and sizing guides recommend providing a 30% buffer (headroom) above the measured average write rate. This extra capacity ensures that during a high-IO period, the replication engine has enough " pipe " to catch up quickly and return to a near-sync state.

Calculation Example: If a workload generates an average of 100 MB/s of new unique data, the administrator should ensure at least 130 MB/s of usable, dedicated bandwidth is available between the sites.

Consequences of Under-sizing: If only 10% (Option A) is used, the array may struggle to recover from even minor bursts, leading to a consistently climbing RPO. 50% (Option B) is often considered safe but can be cost-prohibitive or overkill for standard networking budgets unless the workload is exceptionally volatile.

In Pure1, the ability to simulate snapshot policies—particularly when assessing the capacity requirements and impact of enabling SafeMode—heavily relies on historical telemetry data. Pure1 uses the data from existing snapshots on the FlashArray to calculate the environment ' s daily data change rate, as well as the deduplication and compression ratios specific to those workloads.

By analyzing the footprint of existing snapshots, Pure1 ' s analytics engine can accurately project the future storage capacity required if you were to change your snapshot frequency or extend the retention period (for example, locking them down for 7 to 30 days under a SafeMode policy). If a FlashArray does not have any existing snapshots, Pure1 lacks the foundational baseline metrics needed to simulate and forecast the capacity impact of a proposed snapshot policy.

ActiveDR is Pure Storage’s continuous, near-sync replication solution. It differs fundamentally from standard asynchronous replication because it uses a continuous stream of data rather than snapshot-based " periodic refreshes " (which eliminates Option A).

When you perform a test failover in ActiveDR, you do so by promoting the target pod . The target pod becomes writable, allowing your hosts and applications to run against the replicated data without disrupting the ongoing continuous replication from the source array in the background.

When the test is completed, you demote the target pod. To ensure that the data generated during your test failover isn ' t accidentally lost forever, ActiveDR automatically creates an undo pod at the exact moment of demotion.

If you need to resume that exact test failover scenario or recover the test data, you can re-promote the target pod and instruct ActiveDR to present the content from the undo pod . This unique mechanism allows storage administrators to seamlessly non-disruptively test, pause, and resume DR environments without affecting production protection.