LFCS Practice Test 28 — 15 Questions

Question 1

A system administrator is troubleshooting a custom systemd service that fails to start. They want to override a setting in the service's unit file without modifying the original file in /usr/lib/systemd/system/. Which two methods can be used to create a drop-in override? (Choose two.)

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Create a file in /etc/systemd/system/.service.d/ with a .conf extension.. The two correct methods are creating a drop-in directory with a .conf file and using systemctl edit. Both place overrides in /etc/systemd/system/.service.d/ and are applied after daemon-reload. They preserve the original unit file and are the intended way to customize services without conflicting with package updates. Other methods either replace the unit, mask it, or modify vendor files.

Answer

Edit the unit file directly in /usr/lib/systemd/system/ and run systemctl daemon-reload.

Answer

Copy the original unit file to /etc/systemd/system/ and edit it there.

Answer

Use systemctl mask <service>.service to create an override.

Question 2

A Linux server's primary interface is ens3. Administrators report that after a recent reboot, the server's hostname resolves to 127.0.1.1 instead of its static address 203.0.113.25. The file /etc/hosts currently contains only the default '127.0.0.1 localhost' line. Which single change will make the hostname resolve to 203.0.113.25 for local lookups while leaving DNS resolution for all other names untouched?

Accepted Answer

Add the line '203.0.113.25 server1.example.com server1' to /etc/hosts.. Local name resolution follows the order defined in nsswitch.conf, which by default checks the hosts file before DNS. Because the hosts file only contains the loopback line, the hostname has no local mapping. Adding a hosts entry that pairs the hostname and FQDN with the actual interface address makes the machine resolve its own name to 203.0.113.25 without disturbing DNS lookups for any other domain.

Answer

Add the line '127.0.1.1 server1.example.com server1' to /etc/hosts.

Answer

Add 'hosts: dns files' to /etc/nsswitch.conf so DNS is consulted before the hosts file.

Answer

Add an A record for server1.example.com pointing to 203.0.113.25 in the zone file on the authoritative DNS server.

Question 3

A Linux server acting as a VPN gateway uses an nftables ruleset in the inet filter table. The administrator wants all forwarded traffic from the 10.8.0.0/24 VPN subnet to the 192.168.50.0/24 LAN to be accepted, while dropping any other forwarded traffic. Which nftables rule should be added to the forward chain to accomplish this?

Accepted Answer

nft add rule inet filter forward ip saddr 10.8.0.0/24 ip daddr 192.168.50.0/24 accept. Forwarded traffic between two networks is evaluated in the forward chain of the inet filter table. The rule must use ip saddr for the VPN subnet and ip daddr for the LAN subnet with an accept verdict. Placing this rule before any default drop policy permits only the specified flow, and other forwarded traffic is refused by the chain policy.

Answer

nft add rule inet filter forward ip daddr 10.8.0.0/24 ip saddr 192.168.50.0/24 accept

Answer

nft add rule inet filter forward ip saddr 10.8.0.0/24 ip daddr 192.168.50.0/24 drop

Answer

nft add rule inet filter input ip saddr 10.8.0.0/24 ip daddr 192.168.50.0/24 accept

Question 4

A system administrator needs to configure a systemd service so that it restarts automatically only when it exits with a non-zero exit code, but not when it is cleanly stopped by an administrator. The service unit file currently has no Restart directive. Which combination of directives should the administrator use?

Accepted Answer

Restart=on-failure and RestartSec=5. Restart=on-failure is the correct directive because it restarts the service only on abnormal termination, such as non-zero exit codes or signals. It does not restart on a clean stop, which matches the requirement. RestartSec=5 adds a delay to prevent tight restart loops. The other Restart values either restart too broadly (always) or too narrowly (on-abnormal, on-success).

Answer

Restart=always and RestartSec=5

Answer

Restart=on-success and RestartSec=5

Answer

Restart=on-abnormal and RestartSec=5

Question 5

A Linux administrator is troubleshooting a custom systemd service named backup.service that fails to start. They run systemctl status backup.service and see that the service is in a failed state, but the output does not show detailed error messages. Which command should they use to view the most recent log entries specifically for this service?

Accepted Answer

journalctl -u backup.service -n 50. To view logs for a specific systemd unit, journalctl with the -u option is used. Adding -n specifies the number of recent lines. This filters the journal to only backup.service, providing the most recent error messages. Other commands like systemctl cat or show do not display logs, and journalctl -xe is not unit-specific.

Answer

journalctl -xe

Answer

systemctl show backup.service

Answer

systemctl cat backup.service

Question 6

A Linux server has a single interface ens3 with address 203.0.113.10/24. The administrator runs `ss -tulnp` and sees that a service is listening on 127.0.0.1:8080 only. Remote clients on the Internet report they cannot reach the service on port 8080 even though the firewall allows the port. Which change is the most appropriate to allow remote access while preserving the service's intended exposure?

Accepted Answer

Restart the service after changing its configuration to bind to 0.0.0.0 or the specific external address instead of 127.0.0.1.. A listener bound to 127.0.0.1 accepts only connections that originate on the local machine. Firewall rules and routing cannot change this. The correct remedy is to reconfigure the service to bind to 0.0.0.0 or the external address, then restart it so the new socket is created.

Answer

Add a route for 127.0.0.1 via ens3 so that external packets can reach the loopback address.

Answer

Enable IP forwarding with `sysctl -w net.ipv4.ip_forward=1` so that packets can be forwarded to the loopback listener.

Answer

Configure a DNAT rule redirecting external port 8080 to 127.0.0.1:8080 on the same host.

Question 7

An administrator is diagnosing a Linux router that forwards packets between two internal subnets. Users on 10.20.30.0/24 can reach hosts on 10.20.40.0/24, but responses from 10.20.40.0/24 arrive with the router's external address as the source. The administrator wants to inspect the NAT rules without modifying them. Which command displays the current rules in the nat table with packet and byte counters?

Accepted Answer

iptables -t nat -L -n -v. The nat table holds the PREROUTING, OUTPUT, and POSTROUTING chains where SNAT, DNAT, and MASQUERADE rules live. To inspect them with hit counters, iptables must be invoked with -t nat and the verbose flag -v, plus -n to avoid DNS lookups. This gives a clear picture of which rule is rewriting source addresses on the return path, without altering any rule.

Answer

nft list ruleset

Answer

iptables -L -n -v

Answer

iptables -t nat -S

Question 8

A storage administrator needs to identify which filesystem is mounted at /mnt/data and display the mount options currently in effect for it, using a single command that reads the kernel mount table. Which command should be used?

Accepted Answer

findmnt /mnt/data. The kernel mount table contains the authoritative record of what is mounted and with which options. findmnt queries that table and filters by mount point, displaying the source, target, filesystem type, and options. df and lsblk report capacity or topology, and blkid identifies filesystem signatures, but none of them show the active mount options for a given mount point.

Answer

blkid /mnt/data

Answer

lsblk /mnt/data

Answer

df -h /mnt/data

Question 9

A Linux server has a volume group named vg_data that currently contains two physical volumes. The administrator needs to remove /dev/sdb1 from vg_data, but pvremove /dev/sdb1 returns an error stating the physical volume is still in use. Which command should the administrator run FIRST to migrate the extents off /dev/sdb1 onto the remaining physical volume?

Accepted Answer

pvmove /dev/sdb1. To safely remove a physical volume from an active volume group, the allocated extents must first be moved to other PVs. pvmove performs this online migration without unmounting filesystems. Only after the PV reports zero used extents can vgreduce detach it and pvremove clear its label. Attempting vgreduce or pvremove beforehand fails because LVM protects allocated data.

Answer

lvconvert --repair vg_data

Answer

vgreduce vg_data /dev/sdb1

Answer

pvremove /dev/sdb1 --force

Question 10

A Linux administrator is configuring a new iSCSI storage target. The server has two network interfaces, eth0 and eth1, and the administrator wants to use multipath I/O (DM-Multipath) for redundancy and increased throughput. Which two steps are required to properly set up DM-Multipath for the iSCSI LUNs? (Choose two.)

Accepted Answer

Use iscsiadm to log in to the target portal on both eth0 and eth1, ensuring that each interface discovers the same LUNs.. To set up DM-Multipath for iSCSI LUNs, you must install and configure multipathd and its configuration file, and you must create multiple iSCSI sessions to the same LUNs via different network interfaces. This ensures that the multipath layer can detect and manage multiple paths. Bonding or authentication are not substitutes for multipath configuration.

Answer

Create a bonded network interface combining eth0 and eth1, and configure iSCSI to use the bond for all traffic.

Answer

Format the iSCSI LUNs with a cluster-aware filesystem such as GFS2 before enabling multipath.

Answer

Set the iSCSI initiator name to match the target's ACL, and configure CHAP authentication for each path.

Question 11

A Linux server is experiencing performance issues. The administrator suspects that a process is causing excessive disk I/O. Which command should the administrator use to identify the process with the highest disk I/O usage in real-time?

Accepted Answer

iotop. The tool iotop is designed to monitor disk I/O usage per process in real-time. It displays a list of processes sorted by I/O, allowing the administrator to quickly identify the culprit. Other tools like top, iostat, and vmstat provide system-wide or per-device statistics but lack per-process I/O attribution, making them less effective for this specific troubleshooting task.

Answer

vmstat 1

Answer

top

Answer

iostat -x 1

Question 12

A Linux server has an interface eth0 with IP 192.168.1.100/24. You need to temporarily add a secondary IP address 192.168.1.101/24 to the same interface for testing, and then verify that both addresses are present. Which two commands will accomplish this? (Choose two.)

Accepted Answer

ip addr show dev eth0. To add a secondary IP address temporarily, use 'ip addr add 192.168.1.101/24 dev eth0'. To verify, use 'ip addr show dev eth0'. The ifconfig command would replace the primary address, not add a secondary. The ip route command is for routing, and ip link set up only changes the interface state.

Answer

ip link set eth0 up

Answer

ip route add 192.168.1.101/24 dev eth0

Answer

ifconfig eth0 192.168.1.101 netmask 255.255.255.0

Question 13

An administrator needs to ensure that a custom service, /usr/local/bin/monitor.sh, starts automatically at boot and is restarted if it crashes. The service should run as user 'monitor' and should not depend on network being online. Which systemd unit file configuration is most appropriate?

Accepted Answer

Create a unit file with [Service] Type=simple, ExecStart=/usr/local/bin/monitor.sh, Restart=always, User=monitor, and [Install] WantedBy=multi-user.target.. A systemd service unit must specify the correct Type, ExecStart, Restart policy, User, and installation target. For a script that runs in the foreground, Type=simple is suitable. Restart=always ensures the service is restarted regardless of exit status, providing resilience. User=monitor runs it with least privilege. WantedBy=multi-user.target enables start at boot without network dependency. This combination satisfies all stated requirements.

Answer

Create a unit file with [Service] Type=simple, ExecStart=/usr/local/bin/monitor.sh, Restart=on-abnormal, User=monitor, and [Install] WantedBy=multi-user.target.

Answer

Create a unit file with [Service] Type=simple, ExecStart=/usr/local/bin/monitor.sh, Restart=always, User=monitor, and [Install] WantedBy=graphical.target.

Answer

Create a unit file with [Service] Type=forking, ExecStart=/usr/local/bin/monitor.sh, Restart=on-failure, User=monitor, and [Install] WantedBy=network-online.target.

Question 14

A server is experiencing high load, and the administrator suspects a runaway process is consuming excessive CPU. The administrator wants to identify the top CPU-consuming processes and then terminate the most resource-intensive one. Which sequence of commands should the administrator use?

Accepted Answer

Run 'top' to identify the PID with highest CPU usage, then run 'kill -9 <PID>'.. To find the process consuming the most CPU, a tool that shows per-process CPU usage sorted dynamically is needed. top provides that view and allows the administrator to note the PID. Once the PID is known, kill -9 sends SIGKILL to forcibly terminate it. The other options either use tools that lack per-process CPU details or send signals that may not stop a runaway process, and they do not correctly identify the top CPU consumer.

Answer

Run 'vmstat 1' to identify the PID, then run 'pkill -f <process_name>'.

Answer

Run 'ps aux --sort=-%mem' to find the top CPU process, then run 'kill -15 <PID>'.

Answer

Run 'iostat -c' to identify the PID, then run 'killall <process_name>'.

Question 15

A Linux server has a volume group named vg_data that contains a logical volume lv_archive currently formatted with XFS and mounted at /archive. The administrator needs to shrink the logical volume to free space for a new logical volume. Which command sequence correctly reduces the size of lv_archive?

Accepted Answer

umount /archive; backup data; lvremove /dev/vg_data/lv_archive; lvcreate -L 20G -n lv_archive vg_data; mkfs.xfs /dev/vg_data/lv_archive; restore data; mount /archive. XFS filesystems cannot be shrunk. To reduce the size of an XFS logical volume, the administrator must back up the data, remove the logical volume, create a new smaller logical volume, format it with XFS, and restore the data. Any attempt to use lvreduce with -r or xfs_growfs to shrink will fail because XFS only supports growing.

Answer

umount /archive; xfs_growfs -D 20G /dev/vg_data/lv_archive; lvreduce -L 20G /dev/vg_data/lv_archive; mount /archive

Answer

umount /archive; lvreduce -L 20G /dev/vg_data/lv_archive; lvresize -r -L 20G /dev/vg_data/lv_archive; mount /archive

Answer

umount /archive; lvreduce -r -L 20G /dev/vg_data/lv_archive; mount /archive