You are on page 1of 9

A system administrator is responsible for managing a multi-user computing enviro

nment, such as a local area network (LAN). The responsibilities of the system ad
ministrator typically include installing and configuring system hardware and sof
tware, establishing and managing user accounts, upgrading software and performin
g backup and recovery tasks.
The main responsibilities performed by a system administrator are:
Active Directory management (adding and configuring new workstations and set
ting up user accounts to provide authorizations)
Installing and updating system software
OS patching/upgrades
Preventing the spread of viruses and malicious programs
Allocating mass storage space
Reviewing system logs
System security management
Creating a backup and recovery policy
Performance monitoring and optimization
Before facing any interview for a system administrator position, make sure that
you have enough knowledge on these technologies:
Basic Network Concepts:
Data communication and transmission techniques
Fundamentals of OSI and TCP/IP model
IP address classes
IP subnetting
IPv6 fundamentals
Basics of switching
Microsoft Server Functionalities:
Active Directory Domain Controller (Read only DC , Child DC)
Active Directory Domain Services
DHCP Server
DNS
File and print server
Database storage server
Windows Deployment Services (WDS)
Group Policy management
Registry management
Hyper V
Schedule tasks (Backup, AD DS Backup)
High Availability Features (Failover Clustering, Network Load Balancing)
Top Interview Questions for a System Administrator (Microsoft) Position:
All of the questions below are very common and must be prepared for before facin
g any interview for a System-Server Administrator position.
Q: What is Active Directory?
A: Active Directory provides a centralised control for network administration an
d security. Server computers configured with Active Directory are known as domai
n controllers. Active Directory stores all information and settings for a deploy
ment in a central database, and allows administrators to assign policies and dep
loy and update software.
Q: What is a Domain?

A: A domain is defined as a logical group of network objects (computers, users,


devices) that share the same Active Directory database. A tree can have multiple
domains.
Q: What is Domain Controller?
A: A domain controller (DC) or network domain controller is
uter system that is used for storing user account data in a
is the centrepiece of the Windows Active Directory service
sers, stores user account information and enforces security
domain.

a Windows-based comp
central database. It
that authenticates u
policy for a Windows

A domain controller allows system administrators to grant or deny users access t


o system resources, such as printers, documents, folders, network locations, etc
., via a single username and password.
Q: What is Group Policy?
A: Group Policy allows you to implement specific configurations for users and co
mputers. Group Policy settings are contained in Group Policy objects (GPOs), whi
ch are linked to the following Active Directory service containers: sites, domai
ns, or organizational units (OUs).
Q: What are GPOs (Group Policy Objects)?
A: A Group Policy Object (GPO) is a collection of settings that control the work
ing environment of user accounts and computer accounts. GPOs define registry-bas
ed policies, security options, software installation and maintenance options, sc
ript options, and folder redirection options.
There are two kinds of Group Policy objects:
Local Group Policy objects are stored on individual computers.
Nonlocal Group Policy objects, which are stored on a domain controller, are
available only in an Active Directory environment.
Q: What is LDAP?
A: LDAP (Light-Weight Directory Access Protocol) determines how an object in an
Active Directory should be named. LDAP is the industry standard directory access
protocol, making Active Directory widely accessible to management and query app
lications. Active Directory supports LDAPv2 and LDAPv3.
Q: Where is the AD database stored?
A: The AD database is stored in C:\Windows\NTDS\NTDS.DIT.
Q: What is the SYSVOL folder?
A: The SYSVOL folder stores the server copy of the domain s public files that must
be shared for common access and replication throughout a domain.
All AD databases are stored in a SYSVOL folder and it s only created in an NTFS pa
rtition. The Active Directory Database is stored in the %SYSTEM ROOT%NDTS folder
.
Q: What is Garbage collection?
A: Garbage collection is the online defragmentation of the Active Directory whic
h happens every 12 hours.

Q: When do we use WDS?


A: Windows Deployment Services is a server role used to deploy Windows operating
systems remotely. WDS is mainly used for network-based OS installations to set
up new computers.
Q: What is DNS and which port number is used by DNS?
A: The Domain Name System (DNS) is used to resolve human-readable hostnames like
www.intenseschool.com into machine-readable IP addresses like 69.143.201.22.
DNS servers use UDP port 53 but DNS queries can also use TCP port 53 if the form
er is not accepted.
Q: What are main Email Servers and which are their ports?
A: Email servers can be of two types:
Incoming Mail Server (POP3, IMAP, HTTP)
The incoming mail server is the server associated with an email address account.
There cannot be more than one incoming mail server for an email account. In ord
er to download your emails, you must have the correct settings configured in you
r email client program.
Outgoing Mail Server (SMTP)
Most outgoing mail servers use SMTP (Simple Mail Transfer Protocol) for sending
emails. The outgoing mail server can belong to your ISP or to the server where y
ou setup your email account.
The main email ports are:
POP3
port 110
IMAP port 143
SMTP port 25
HTTP port 80
Secure SMTP (SSMTP)
port 465
Secure IMAP (IMAP4-SSL)
port 585
IMAP4 over SSL (IMAPS) port 993
Secure POP3 (SSL-POP)
port 995
Q: What do Forests, Trees, and Domains mean?
A: Forests, trees, and domains are the logical divisions in an Active Directory
network.
A domain is defined as a logical group of network objects (computers, users, dev
ices) that share the same active directory database.
A tree is a collection of one or more domains and domain trees in a contiguous n
amespace linked in a transitive trust hierarchy.
At the top of the structure is the forest. A forest is a collection of trees tha
t share a common global catalog, directory schema, logical structure, and direct
ory configuration. The forest represents the security boundary within which user
s, computers, groups, and other objects are accessible.
Q: Why do we use DHCP?

A: Dynamic Host Configuration Protocol assigns dynamic IP addresses to network d


evices allowing them to have a different IP address each time they are connected
to the network.
Q: What are Lingering Objects?
A: A lingering object is a deleted AD object that still remains on the restored
domain controller in its local copy of Active Directory. They can occur when cha
nges are made to directories after system backups are created.
When restoring a backup file, Active Directory generally requires that the backu
p file be no more than 180 days old. This can happen if, after the backup was ma
de, the object was deleted on another DC more than 180 days ago.
Q: How can we remove Lingering Objects?
A: Windows Server 2003 and 2008 have the ability to manually remove lingering ob
jects using the console utility command REPADMIN.EXE.
Q: Why should you not restore a DC that was backed up 6 months ago?
A: When restoring a backup file, Active Directory generally requires that the ba
ckup file be no more than 180 days old. If you attempt to restore a backup that
is expired, you may face problems due to lingering objects.
Q: How do you backup AD?
A: Backing up Active Directory is essential to maintain the proper health of the
AD database.
Windows Server 2003
You can backup Active Directory by using the NTBACKUP tool that comes built-in w
ith Windows Server 2003 or use any 3rd-party tool that supports this feature.
Windows Server 2008
In Server 2008, there isn t an option to backup the System State data through the
normal backup utility. We need to use the command line to backup Active Director
y.
1. Open up your command prompt by clicking Start, typing
2. In your command prompt, type
nd press Enter.
3. Input y

cmd

and then hit Enter.

wbadmin start systemstatebackup -backuptarget:e: a

and press Enter to start the backup process.

When the backup process has finished you should get a message that the backup co
mpleted successfully. If it did not complete properly you will need to troublesh
oot.
The questions above are very tricky and important from the standpoint of clearin
g any interview for a System Administrator (Microsoft) position. It is not possi
ble for anyone to list every possible question, but you can get more frequently
asked interview questions for System/Server Administrator (Microsoft) Jobsfrom t
he download link posted here. If you find any difficulty in answering any questi
ons, ask me below @ the Comments section.

Tips for Preparing for an Interview


Study: Before an interview, do a quick recap of relevant technologies.
Update resume: Read your resume through; don t copy and paste your anything in
it. You must be aware of your strengths and weaknesses.
Prepare professional certifications: One of the best ways to prove the techn
ical skills mentioned in your resume is through certifications. This gives a new
employer an easy way to understand your knowledge level.
Update LinkedIn profile: Update your LinkedIn profile regularly; make sure t
hat your work experience, qualifications, and project details match your resume.
This article is the output of my extensive research and work experience. With th
is article I hope to help candidates in preparing for an interview for a system
administrator position in a Microsoft multi-user computing environment.
Don t forget to post your queries and feedback in the comments section. You can jo
in our Facebook group, http://www.facebook.com/intenseschool, to get updates on
new posts and technologies.
-------------What is DNS?
How does DNS work inside and outside the network?
What is AD?
How are DNS and AD related?
What's the difference between a locked account and a disabled account?
You implement a new service on the network that is tied in to AD and require
s a domain account to run. What feature can you use to provide an AD account?
What's the difference between a forest and a domain?
How do you raise a forest or domain level?
What purpose does the KCC have? If the KCC isn't working correctly or not pr
oducing the links you want, how can you manually create site links?
What is an AD trust?
What are the FSMO roles and what do they do?
How do you transfer the FSMO roles?
Users are stating that the time is wrong on all the computers, how do you fi
x this?
What the difference between a stub zone, primary zone, and a secondary zone?
What's the difference between a forward look up zone and a reverse look up z
one?
How is creating a domain in 2003/2008 different from 2012/2102r2?
What is DHCP?

What is a DHCP scope?


Using DHCP, how do you ensure users get the proper DNS servers?
At what point does a computer renew its DHCP lease?
What is the process for requesting a DHCP lease and how can you see this in
action?
What's the difference between a static IP and a reservation? When should you
use either?
What is a VLAN?
How do you ensure traffic from one VLAN can reach another?
A user states that their computer is getting a 169.254.x.x address. What is
this address called, what is its purpose, and what can you check to see why they
are getting it? (there are multiple answers to this)
When should you use a Hub vs a Switch and why?
Users are complaining about call quality issues during times of high network
traffic. What IEEE standard can you implement to resolve this issue?
What is a DMZ?
How do I ensure that my network can experience a switch failure and at most
lose only a handful of pings (multiple correct answers)?
OSPF, EIGRP, BGP, RIP. What do they stand for, what advantages do they have
over the other (this might be over kill), and why is one not like the others?
Cisco calls them ether-channels/port channels, HP calls them trunks. When wo
uld you use them?
What is a good way to ensure that the guest wifi is not able to connect to p
roduction equipment (there are multiple right answers)?
Your company has recently taken up a security initiative and needs to tighte
n up WiFi security. What are some ways to do this? (there are multiple correct a
nswers, really depends on how far you want to go)
Difference between RAID 0, 1, 5, 6, 1+0 and 0+1 and when should you use them
in production equipment?
How do NFS, iSCSI, FC, and FCoE differ? What are some requirements for each
(it's fine if you don't know specifics as long as you understand them conceptual
ly)?
How many drive failures can the previously mentioned raid levels experience
before total data loss?
What are some ways you can improve the performance of you SAN/NAS (multiple
correct answers)?
SATA, Near Line SAS, SAS, SSD. Can you describe to me when you would use one
versus the other?
What is storage tiering and why would you use it?

Describe thin provisioning versus thick provisioning.


Difference between replication, RAID, snapshot, and a back up?
What back up medium would offer the fastest recovery?
On-site backups, off-site backups, cloud back ups. In what situation would e
ach be ideal? What are the pros and cons of each?
What is a certificate and how does it work?
What is a certificate authority?
What is the difference between a root CA and an intermediary CA?
When should you renew your certificates?
When would a PKI infrastructure be useful (this is probably over kill depend
ing on your environment but nice to know)?
What the difference between a self signed cert and a third party cert?
You've deployed a new internal website for your company that works over http
s. However, every time a user goes to the internal URL, they are greeted with a
warning saying the site is insecure. What are two ways to make this prompt go aw
ay using certificates? Assume that this is an IIS server.
What is a virtual machine?
What's the difference between a guest and a host?
What is the hypervisor and what is it's job?
Difference between a type 1 and type 2 hypervisor. Provide an example.
How do virtual machines differ from traditional non-virtual machine servers
and what is an advantage of a VM?
What are some common ways to connect storage to a hypervisor?
What's an advantage of having a virtual machine cluster?
What is concern that you have to worry about virtual machines with respect t
o storage?
What happens when I hit send on an email once it gets to my mail server (how
does mail route on the internet)?
What's a way to cut down on spam you receive on the internet?
An external contact says they received an email from your domain that was cl
early not sent from your mail servers based on the headers. What is a way to mak
e it harder for something like this to not happen again?
What is TLS?
Ports 443, 80, 25, 587. What does each do and what service is typically used
with each in a Microsoft Exchange environment?

You are running an Exchange 2010 or newer exchange environment. You need to
make sure that if your mail box server that is hosting your mailbox database exp
eriences an outage, that users can still get to their mail without much issue. W
hat feature can you implement?
Why do you need AD with exchange?
--domain name systems
zones & forwarders
MS cornerstone for directory services / implementation of ldap
namespaces
locked = condition, disabled = administrative
managed service account or virtual account.
forest = different name spaces / made of tree's (made of domains)
domains & trust, right click, raise.
replication, domains & trusts
relationship of contextual inheritance (implicit, explicit, one way, transit
ive, etc)
schema, domain, pdc, inf - blah blah blah.
depending on the role - AD-Schema, Domains & Trusts, or Users & Computers
ntp
blah blah blah - how dns works
foward= name->ip rev= ip-> name A vs PTR
2012 promotes via role.
addressing
configuration of dhcp properties
network segment or AD-OU depending on topo
when it expires, reboots, or manual
broadcast - packet sniffer
static = manual assignment. reservation = set aside in dhcp. Depends on the
situation. Reservations = lower
think switch inside a switch. isolate broadcasts
either allow it on the ACL, configure routes, or put them all on the same tr
unk - depends where they terminate.
APIPA, fallback addressing, plug their cable in / get them a real address.
Hub, probably never these days. Switch, usually most of the time. Switch = h
igher end to end bandwidth.
802.1p / QOS
The place where your network is no longer really your network.
failover / redundancy, vtp, teaming and probably 3 other decent answers.
routing protocols - depends on where you want to use them... rip the inside,
bgp the wan, ospf your sites, eigrp if you've got all cisco gear and want to ge
t fancy.
combine 1 port to many ports.
dont plug it in there. put it on its own vlan, 802.1x, radius & nac
WPA2 enterprise the whole thing with client certificates and captive portals
.
stripe, mirror, parity, parity+1, nested raids - balance performance and rel
iability.
not my area of expertise.
0=0, 1=1, 5=1, 6=2, etc.
I dont do storage...
Storage... nope.
Performance & Archive states. (I do work with a lot of logging systems).
Thin = promise to pay. Thick = full allocation.
Replication = multiple copies, RAID = integrity, Snapshot = point in time, B
ack up = continuity.

Optical.
Temporary, Normal, Depends - who's cloud and why cant we just do it ourselve
s?
Basically, without going into how PKI works... it proves you are you you say
you are.
Ignoring the glaring flaws in this system, its the computer (organization) t
hat issues certificates
Root CA validates itself. Intermediates validate others chains.
When they expire, become compromised, and in accordance with your organizati
ons policy.
Lets skip this for now and assume we all know what PKI is. But basicaly - wh
en ever you want to protect the Confidentiality or Integrity of something in you
r network.
Self signed certs (unless you have your own root ca added to the store) wont
validate up the chain.
Slap your SA and tell him to add the cert to the trust store for the domain.
Instantiated compute environment.
Host = the thing running the VM. Guest = the VM itself.
Control the resource allocation and abstraction for the VM's.
Type 1 = bare metal (esx, xen, hyper-v) Type 2 = desktop virtualization (vmw
are, virtualbox)
VM's are all software = really fast deployment, really flexible, blah blah b
lah.
Fiber
Ummmm? repeat the question.
That you thin provisioned them and dont actually have the required storage.
You're runnning them on slow
Not enoguh info here... whats your protocol, whats your destination, lets ju
st say it leaves the client, hits the server, gets routed to the gateway, and se
nt on its way.
Get a barracuda. Dont run an open relay. Verify PTR's, Only accept mail that
follows strict protocol rules. Authenticate senders...
SPF record
Transport Layer Security & its all tied up in that PKI business we talked ab
out earlier.
HTTPS, HTTP, SMTP, other SMTP
I want to say it has soemthign to do with the transport setup... meh, not a
mail guy.
Because thats just how these things work.
------------------------------------

You might also like