100+ Free LPIC-3 High Availability and Storage Clusters Practice Questions

Pass your LPIC-3 High Availability and Storage Clusters (Exam 306-300) exam on the first try — instant access, no signup required.

✓ No registration✓ No credit card✓ No hidden fees✓ Start practicing immediately

Not publicly disclosed Pass Rate

100+ Questions

100% Free

1 / 100

Question 1

Score: 0/0

Which open-source cluster resource manager is the modern standard for Linux failover clusters?

Pacemaker

rgmanager

Heartbeat 1

Linux-HA v1

to track

2026 Statistics

Key Facts: LPIC-3 High Availability and Storage Clusters Exam

Exam Questions

LPI

500/800

Passing Score

LPI scaled scoring

90 min

Exam Duration

LPI

$200

Exam Fee (Tier 1)

LPI

5 years

Validity

Recertification required

LPIC-2

Required to claim

LPI

LPIC-3 306-300 is a 60-question, 90-minute senior exam scored on a 200-800 scale with 500 to pass. Four weighted topic areas total 54 points: HA Cluster Management (concepts, load-balanced clusters with HAProxy/LVS/keepalived, failover with Pacemaker/Corosync via pcs and crm), HA Cluster Storage (DRBD with drbdadm, cluster storage access, clustered file systems GFS2/OCFS2/cLVM), HA Distributed Storage (GlusterFS volumes, Ceph osd/mon/mgr/mds and CephFS/RBD), and Single Node HA (hardware redundancy, advanced RAID with mdadm, advanced LVM, network HA with bonding). Requires LPIC-2 to claim the LPIC-3 credential.

Sample LPIC-3 High Availability and Storage Clusters Practice Questions

Try these sample questions to test your LPIC-3 High Availability and Storage Clusters exam readiness. Each question includes a detailed explanation. Start the interactive quiz above for the full 100+ question experience with AI tutoring.

1Which open-source cluster resource manager is the modern standard for Linux failover clusters?

A.Pacemaker

B.rgmanager

C.Heartbeat 1

D.Linux-HA v1

Explanation: Pacemaker (with Corosync as its messaging layer) is the standard cluster resource manager in modern Linux distributions. rgmanager (RHCS), Heartbeat 1, and Linux-HA v1 are predecessors or alternatives that are largely deprecated. RHEL/Rocky/SLES all ship Pacemaker as the supported solution.

2Which messaging layer underpins Pacemaker for cluster membership and reliable communication?

A.Corosync

B.OpenAIS only

C.Heartbeat

D.ZeroMQ

Explanation: Corosync provides closed-process group communication, membership, and quorum services. Pacemaker is the resource manager on top of it. OpenAIS was an earlier ancestor; Heartbeat was its predecessor. Pacemaker can also work with Heartbeat in legacy setups but Corosync is canonical today.

3Which command-line tool is used on RHEL/Rocky to create and configure a Pacemaker cluster?

A.pcs

B.crm

C.ccs

D.corosync-cli

Explanation: pcs (Pacemaker/Corosync configuration system) is the canonical CLI on RHEL/Rocky/AlmaLinux. crm (crmsh) is the equivalent on SUSE/openSUSE. ccs was the legacy RHCS tool, removed in RHEL 7+. corosync-cli is not a real command.

4Which command initializes and starts a fresh 3-node Pacemaker cluster with pcs?

A.pcs cluster setup mycluster node1 node2 node3 --start --enable

B.pcs init mycluster

C.pcs cluster create node1 node2 node3

D.pcs new cluster

Explanation: pcs cluster setup writes corosync.conf to all listed nodes; --start launches corosync/pacemaker and --enable adds them to the systemd boot path. The pcs daemon must already be running (`systemctl enable --now pcsd`) and the hacluster password must be authenticated with `pcs host auth`.

5Which Pacemaker concept ensures that a failed node cannot continue to access shared resources by forcefully shutting it down?

A.STONITH (Shoot The Other Node In The Head) / fencing

B.Quorum

C.Resource constraint

D.Live migration

Explanation: STONITH (a.k.a. fencing) physically powers off, reboots, or otherwise isolates an unresponsive node so it cannot corrupt shared resources. Without active fencing, Pacemaker (especially with shared-storage clusters) is not production-supportable. Quorum determines who gets to act; fencing enforces it.

6Which fencing agent is appropriate for a virtualized cluster of guests on a libvirt/KVM host?

A.fence_xvm (or fence_virt with fence_virtd on the host)

B.fence_ipmilan

C.fence_apc

D.fence_scsi

Explanation: fence_xvm communicates with fence_virtd on the host, which uses libvirt to power-cycle the guest. fence_ipmilan needs IPMI BMCs (physical servers). fence_apc targets APC PDUs. fence_scsi uses SCSI persistent reservations on shared LUNs.

7Which corosync.conf parameter sets the totem token timeout (failure detection time)?

A.token (in totem block)

B.timeout = 3000

C.failtime

D.heartbeat

Explanation: totem { token: 3000 } sets the membership timeout in milliseconds — if a node misses tokens for that duration, it is considered failed. Default is 1000-3000 ms depending on transport. Higher values reduce false positives but increase recovery time.

8Which transport does Corosync 3 use by default?

A.knet (Kronosnet)

B.udpu (UDP unicast)

C.udp (multicast)

D.rdma

Explanation: Corosync 3 defaults to knet, which supports multiple links (rings), encryption, and authentication out of the box. udpu and udp remain available for older deployments. rdma is not standard; the legacy iba (InfiniBand) was experimental.

9Which Pacemaker resource type runs on every node in a cluster simultaneously (e.g., a daemon that must be present everywhere)?

A.clone

B.primitive

C.group

D.master/slave

Explanation: A clone resource runs on multiple nodes simultaneously. Primitive is a single-instance resource. A group is an ordered/colocated set of primitives. master/slave (now 'promotable') is a special clone where some instances are promoted (e.g., DRBD master).

10Which command shows the current Pacemaker cluster status, including resources and node states?

A.pcs status

B.pcs show

C.pacemaker ls

D.crm_mon (or `crm status` on SLES)

Explanation: pcs status (RHEL) and crm status (SLES) summarize node and resource state. crm_mon is the lower-level command (live-updating) used by both wrappers. pcs show is not the canonical verb. pacemaker ls is invented.

About the LPIC-3 High Availability and Storage Clusters Exam

Senior LPIC-3 specialty exam covering Linux HA cluster management (Pacemaker, Corosync, HAProxy, LVS/keepalived), HA storage (DRBD, GFS2, OCFS2, cLVM), distributed storage (Ceph, GlusterFS), and single-node HA (RAID, LVM, bonding). Passing 306-300 with an active LPIC-2 grants the LPIC-3 HA and Storage Clusters credential.

Questions

60 scored questions

Time Limit

90 minutes

Passing Score

500 / 800

Exam Fee

$200 USD (Linux Professional Institute (Pearson VUE / OnVUE online proctored))

LPIC-3 High Availability and Storage Clusters Exam Content Outline

41%

High Availability Cluster Management

HA concepts (active/active, active/passive, fencing/STONITH, quorum), load-balanced clusters with HAProxy/LVS/keepalived/ipvsadm, failover clusters with Pacemaker and Corosync managed via pcs and crm

24%

High Availability Distributed Storage

GlusterFS volume types (replicated, distributed, dispersed) with gluster volume create, Ceph cluster components (osd, mon, mgr, mds), RBD and CephFS access

24%

High Availability Cluster Storage

DRBD replication with drbdadm/drbdsetup and dual-primary, cluster storage access, clustered file systems (GFS2 with mkfs.gfs2 + DLM, OCFS2 with cluster.conf, cLVM)

22%

Single Node High Availability

Hardware and resource HA (multipath, redundant PSU/NIC), advanced RAID with mdadm and write-intent bitmaps, advanced LVM (thin pools, snapshots, mirrors), network HA with bonding modes and teaming

How to Pass the LPIC-3 High Availability and Storage Clusters Exam

What You Need to Know

Passing score: 500 / 800
Exam length: 60 questions
Time limit: 90 minutes
Exam fee: $200 USD

Keys to Passing

Complete 500+ practice questions
Score 80%+ consistently before scheduling
Focus on highest-weighted sections
Use our AI tutor for tough concepts

LPIC-3 High Availability and Storage Clusters Study Tips from Top Performers

1Build a 3-node Pacemaker cluster on RHEL/CentOS Stream using pcs cluster setup, then add resources with pcs resource create — drill colocation, order, and location constraints

2Configure Corosync rings (knet transport) and verify with corosync-cfgtool -s and corosync-quorumtool — understand quorum, two_node, and last-man-standing

3Master fencing/STONITH with fence_xvm or fence_ipmilan — know that a cluster without fencing is not production-supportable

4Set up DRBD: write /etc/drbd.d/r0.res, drbdadm create-md r0, drbdadm up r0, drbdadm primary --force r0 — then mount, fail over, and verify with drbdadm status

5Format a clustered GFS2 with mkfs.gfs2 -p lock_dlm -t cluster:fs1 -j 3 /dev/...; configure DLM via Pacemaker; mount on all nodes simultaneously

6Stand up Ceph manually: deploy mon, mgr, osd, and mds; create a pool with ceph osd pool create; mount CephFS or map an RBD image with rbd map

7Practice keepalived VRRP: write keepalived.conf with vrrp_instance VI_1, virtual_router_id, priority, virtual_ipaddress — verify failover by stopping the master

Frequently Asked Questions

What does LPIC-3 306-300 cover?

LPIC-3 306-300 is the LPI High Availability and Storage Clusters specialty exam covering four topic areas: HA Cluster Management (Pacemaker/Corosync, HAProxy, LVS/keepalived), HA Cluster Storage (DRBD, GFS2, OCFS2, cLVM), HA Distributed Storage (GlusterFS, Ceph), and Single Node HA (RAID, LVM, bonding). Total objective weight is 54 points across 12 sub-objectives.

Do I need LPIC-2 before taking LPIC-3 306?

You can take and pass 306-300 without LPIC-2, but the LPIC-3 HA and Storage Clusters credential is only awarded once you hold an active LPIC-2. The pass result is valid for five years, so the order is flexible.

How hard is the LPIC-3 306 exam?

306-300 is a senior-level exam targeting working Linux engineers responsible for production HA clusters. Expect detailed questions on Pacemaker constraint syntax, Corosync rings, DRBD resource configuration, Ceph component roles, and bonding modes. Most candidates pass after 80-120 hours of focused study and lab work.

What format is the 306-300 exam?

60 questions in 90 minutes, mixing multiple-choice and fill-in-the-blank. Scaled scoring 200-800 with 500 required to pass. Delivered at Pearson VUE test centers or via OnVUE online proctored.

Is 306-300 the successor to 304?

306-300 was added in 2022 as a fourth LPIC-3 specialty alongside 305-300. The older 304 (Virtualization and HA) was split: virtualization went into 305 and HA/Storage Clusters went into 306. If you previously held 304, recertification now points to 305 or 306 depending on your focus.

How should I prepare for the 306-300 exam?

Plan 80-120 hours over 8-12 weeks. Build a 3-node Pacemaker/Corosync cluster, configure a DRBD resource and promote/demote across nodes, format and mount a GFS2 volume, stand up a small Ceph cluster with cephadm or manual osd/mon, and configure a 2-node HAProxy + keepalived load balancer.

Does LPIC-3 HA expire?

Yes — LPIC-3 is valid for five years. Recertify by passing any current LPIC-3 specialty (300/303/305/306) again, or by passing a newer LPI exam at the same or higher level before expiration.