Try each one before opening the answer. Every option is explained, with the Microsoft documentation page that proves it.
Question 1 · choose 1
You link VNet-Dev to the private DNS zone dev.contoso.internal. New VMs in VNet-Dev must get A records in the zone without anyone creating them. What should you set?
AAn alias record set in the zone that targets VNet-Dev
BA conditional forwarder on the VMs' DNS settings
CA DNS private resolver inbound endpoint in VNet-Dev
DEnable auto registration on the virtual network link
Show the answer and why
AAn alias record set in the zone that targets VNet-Dev
Incorrect
An alias record set is a single record set that you create yourself; it doesn't add records as VMs are created.
BA conditional forwarder on the VMs' DNS settings
Incorrect
Forwarders send queries elsewhere; they don't register VM names in the zone.
CA DNS private resolver inbound endpoint in VNet-Dev
Incorrect
A resolver endpoint handles hybrid DNS queries; it doesn't register VM records.
DEnable auto registration on the virtual network link
Correct
With autoregistration on the link, Azure creates and maintains a record for each VM in the virtual network.
Autoregistration is a property of the virtual network link. It works for virtual machines only and creates and removes their records as VMs come and go.
A Standard internal load balancer has two backend VMs with a TCP probe on port 8080. The app answers on 8080 locally. Both VMs show as unhealthy after someone added an inbound rule to the subnet's NSG at priority 100 that denies all traffic from source Any on all ports. Why?
AThe NSG blocks the probe source, tagged AzureLoadBalancer
BStandard load balancers need HTTP probes on port 80
CInternal load balancers need a public IP on each backend VM
DThe backend VMs need to be in an availability set
Show the answer and why
AThe NSG blocks the probe source, tagged AzureLoadBalancer
Correct
Probes come from 168.63.129.16, which the AzureLoadBalancer tag stands for. The new deny at priority 100 is evaluated before the default AllowAzureLoadBalancerInBound rule at 65001, so the probes are dropped.
BStandard load balancers need HTTP probes on port 80
Incorrect
TCP probes on any port are supported; the app responds on 8080, so the probe type isn't the problem.
CInternal load balancers need a public IP on each backend VM
Incorrect
Backends of an internal load balancer use private addresses; public IPs aren't required.
DThe backend VMs need to be in an availability set
Incorrect
Standard Load Balancer backends can be any VMs in the virtual network; no availability set is needed.
The probe source address 168.63.129.16 must be allowed by every NSG and local firewall. The default AllowAzureLoadBalancerInBound rule allows it, but NSG rules are processed in priority order and processing stops at the first match, so a deny from Any at a lower number wins and the load balancer marks the instances down. An allow rule for the AzureLoadBalancer tag above the deny fixes it.
In the Azure DNS zone fabrikam.com, the name fabrikam.com must resolve to a Standard public IP of a VM, and the record must follow the IP if it changes. Which record should you create at the apex?
AA CNAME record set at @ pointing to the public IP's DNS name
BAn NS record set at @ pointing to the VM
CAn alias A record set at @ that targets the public IP
DA TXT record set at @ that contains the public IP address
Show the answer and why
AA CNAME record set at @ pointing to the public IP's DNS name
Incorrect
You can't create a CNAME at the zone apex, because the apex already holds the NS and SOA record sets.
BAn NS record set at @ pointing to the VM
Incorrect
The apex NS record set lists the zone's name servers; it can't point a name at a VM.
CAn alias A record set at @ that targets the public IP
Correct
An alias A record references the public IP resource and updates when the address changes, and it's allowed at the apex.
DA TXT record set at @ that contains the public IP address
Incorrect
A TXT record stores text; resolvers don't use it to return an address.
The DNS standards don't let a CNAME share a name with other records, and the apex always has NS and SOA records. An alias A record set can point at a Standard public IP resource and stays in step with it.
A web app's health page returns HTTP 302 redirects to a sign-in page. The load balancer's HTTP probe on that path marks every instance down. Why?
AHTTP probes accept 2xx and 3xx responses
BAn HTTP probe needs a 200 response
CThe probe's port must be 80
DHTTP probes need HTTPS
Show the answer and why
AHTTP probes accept 2xx and 3xx responses
Incorrect
A probe is up only for status 200.
BAn HTTP probe needs a 200 response
Correct
Any status other than 200 marks the probe down.
CThe probe's port must be 80
Incorrect
Probes can use any port.
DHTTP probes need HTTPS
Incorrect
HTTP probes work over HTTP.
HTTP and HTTPS probes mark an instance up on a 200 response within the timeout; any other status code marks it down. Point the probe at a path that returns 200 without sign-in.
Behind a public load balancer, admins need SSH to each of three backend VMs through the load balancer's IP, using ports 50001-50003. What should you configure?
AAn outbound rule for port 22
BInbound NAT rules for each VM's port 22
CSession persistence on port 22
DA load balancing rule for port 22 on the frontend
Show the answer and why
AAn outbound rule for port 22
Incorrect
Outbound rules handle outbound SNAT.
BInbound NAT rules for each VM's port 22
Correct
Inbound NAT rules forward a frontend port to a specific backend instance's port.
CSession persistence on port 22
Incorrect
Persistence can't target a particular VM.
DA load balancing rule for port 22 on the frontend
Incorrect
A load balancing rule spreads connections across the pool, not to a chosen VM.
Inbound NAT rules provide port forwarding through the load balancer's frontend to individual instances. Bastion is often a safer way to reach VMs.
VNet-Hub and VNet-Spoke are peered. The private DNS zone corp.internal is linked to VNet-Hub, and VMs in VNet-Hub resolve its records. VMs in VNet-Spoke use Azure-provided DNS and get no answer for the same names. What should you do?
AEnable gateway transit on the peering
BAdd a second link from corp.internal to VNet-Hub for the spoke
CLink corp.internal to VNet-Spoke as well
DMove corp.internal to VNet-Spoke's region
Show the answer and why
AEnable gateway transit on the peering
Incorrect
Gateway transit lets the spoke use the hub's VPN or ExpressRoute gateway; it doesn't make the hub's private zones resolvable.
BAdd a second link from corp.internal to VNet-Hub for the spoke
Incorrect
Only one link can exist between a zone and a virtual network, and a link serves the VMs in the network it names.
CLink corp.internal to VNet-Spoke as well
Correct
A virtual network resolves a private zone's records only once it's linked to the zone; peering doesn't extend the hub's link.
DMove corp.internal to VNet-Spoke's region
Incorrect
A private zone is stored as a global resource and can be linked to virtual networks in any region; its location isn't the problem.
To resolve a private DNS zone from a virtual network, that network must be linked to the zone. In a hub-and-spoke design each spoke gets its own link, and resolution across linked networks doesn't depend on peering at all.