System Administrator Interview Questions: Take the Broad View

System administrators—also known as sysadmins—are generally responsible for planning, installing, and managing a company’s tech stack. It’s a sprawling job that demands an equally sprawling set of skills, from troubleshooting and problem-solving to communication and coding.

When answering system administrator interview questions, it’s important to highlight that you understand how systems work from end to end, and can manage all sorts of problems and projects as they arise. That’s according to software developers Pedro Fernandez and Rodrigo Morales, as well as software engineers David Gross and Edgar Ortega, who all work at FullStack Labs, a software development consultancy in Sacramento, CA.

While many sysadmin jobs share the same basic set of requirements, many organizations have highly specific needs, and candidates will need to tailor their answers to fit the role. For example, a certain company might have a system administrator designing and installing networks, while another might want someone who can manage and monitor online portals.

“Soft skills” are key to any sysadmin role. Melissa Fingerhut, the director of global talent acquisition at AvePoint, a software vendor and manufacturer in Jersey City, NJ, emphasizes the need for good communication and time-management abilities.

With all that in mind, here are some system administrator interview questions you could face:

“Explain Domain Name System (DNS) and reverse DNS (rDNS).”

Among all possible system administrator interview questions, this may seem like a particularly simple one, but answering it requires knowledge of the internet’s “chief record keepers.” You can’t shortcut this one.

What most people say:

Most candidates use a phone book analogy to describe DNS; like a phone book, DNS looks up a website and retrieves the correct IP address. Then you have rDNS, a “reverse phone book” because it uses the IP address to find the Website. According to FullStack Labs, offering up an answer like that is technically accurate, but doesn’t give a full picture of DNS or rDNS use.

What you should say:

“DNS translates text into an IP address, and there are root DNS servers that know which other DNS servers to ask to get an address. When you request an address, it asks the first server on the network, which then asks the next DNS server, and so on, until the address is found.

“DNS can be used for proving ownership, makes it easy to remember a site’s name, and allows for round-robin routing of traffic. It also has text records, and email records, allows for traffic routing based on geolocation, and more.

“An rDNS requests IP addresses get domain names. It’s often used for tracking where an email or website visitor came from, and it isn’t as critical as DNS.”

Why you should say it:

It shows you understand the intricacies of the more important DNS, which is significant to all website management, and sits at the core of many sysadmin jobs.

“What is the difference between Transmission Control Protocol (TCP) and User Datagram Protocol (UDP)? When should you use each?”

Candidates must know what protocols to use, and why, when configuring network hardware and/or software.

What most people say:

Both FullStack Labs and Fingerhut often hear candidates say that TCP and UDP are protocols used to send small data packets over the internet.

They also hear: “TCP is used when there’s a need for QoS (Quality of Service), and UDP is used when there isn’t.” While this is an accurate statement, it doesn’t provide the whole picture.

What you should say:

“TCP is known for reliability, and is a commonly used protocol on the internet. It’s used for web browsing, email and remote admin. Its built-in QoS (Quality of Service) guarantees the recipient receives the message in a specified order, and that the sender gets confirmation from the recipient. If there’s a disruption, the recipient will get an error message, and messages will be re-sent until the sender gets confirmation. TCP also has error checks, which guarantees packets aren’t corrupted during transmission.

“UDP represents speed over reliability. It doesn’t have error checks or QoS; packets are just sent in a constant stream to the recipient, without any confirmation. If there’s a disruption during the stream, UDP jumps to the next current packet. It’s most commonly used for online games and live streaming.”

Why you should say it:

While these are fundamental protocols that every sysadmin should know, context counts. If you’re building on existing protocol, or even architecting a library, you need to be able to explain why you should use one or the other.

“What steps would you take to solve a problem that you don’t understand?”

System Administrators spend a lot of time responding to problem-solving requests, so it’s inevitable you’ll encounter unfamiliar issues.

What most people say:

According to Fingerhut, most candidates say they’ll undertake the root problem first, and will seek out the source of the issue. But that doesn’t show her a candidate’s thought process.

The crew at FullStack Labs says that most candidates say they’ll investigate a problem and solve it: not a bad answer if you’ve already demonstrated some skill during the interview, but it doesn’t exactly give a complete picture.

What you should say:

If you have an example of having solved a difficult problem, this is one of those system administrator interview questions where you get to talk about it at length. Your story could follow this template:

“I’d familiarize myself with the context of the problem, so I can understand why it needs to be solved. I’d also need to understand the business goals connected to the issue. Then, I’d assess different approaches, and would choose one that best, and most efficiently, solves the problem while achieving the related business goals.”

Why you should say it:

You clearly understand the ‘why’ and ‘how’ of solving the problem, as well as the business goals that will be achieved by solving it.

“When you have to juggle different tasks of similar importance, how do you prioritize the work?”

As a sysadmin, you need solid time-management skills, especially when you have a big workload.

What most people say:

FullStack Labs too often hears: “Everything is high priority. You work off your load until all items are cleared out.”

What you should say:

“I first prioritize my work by distributing it into tasks. I assess their value and identify urgent vs. important, then estimate time for each, and define a delivery schedule. I’m okay with not being able to deliver everything in one day.”

Why you should say it:

Being able to convey critical organizational and time management skills will help you stand out from other candidates. Keep that front-of-mind when answering system administrator interview questions.

Related

One Response to “System Administrator Interview Questions: Take the Broad View”

  1. I have a slight problem with the premise of this article: “If you want to know how a systems administrator should answer interview questions, we got a couple of the dev-ops guys down here to tell you how it is!”

    Developers are rarely sysadmins, and when they are, they’re usually bad at it because they lack the formal training and skills. Same reason most sysadmins don’t make good developers.

    I’m especially tickled by the TCP/UDP answer including QoS. Q tagging a packet can happen for both UDP and TCP, and most often actually takes place on UDP rather than TCP packets. This is because UDP is used to send things like real-time media, where the packet getting there at the correct time is more important than all of the packets getting there. A lot more to QoS than just making sure the packet arrives.

    But hey, what do I know, I’m just a lowly netadmin. If you really want to know how networking works, you should go talk to a developer….