Systems Administrator:
A Systems
admin can go by many names: Server Admin, NT Administrator, Systems Engineer,
Unix/Linux admin, Network Administrator, etc. Refer to job descriptions to more
clearly understand the company’s definition for their environment and the name
they assign to this role. In the end, an IT admin supports infrastructure
systems and devices.
In this “day in the life”, I'll
focus on a Windows server administrator.
I would also like to mention that in my career, I have
specialized in helping companies where the executive management team has
reached their boiling point with their IT Infra teams. They reached out to
me for help come in and assess the environment, develop a design and action plan then put things into place. The approach I have fine-tuned over the years may be the subject
of a series of other posts if there is interest.
Windows is by far the most common
server operating system you will see in the enterprise. The Linux/Unix admins
have similar workdays however there are some differences that I will cover in
another post. Network Admin (for switches and routers) is yet another posting.
Group Role:
The Windows server support group,
along with the network group, is the foundation of the entire business. The physical
and virtual server that the team installs, configures, secures and essentially
keeps ‘alive’ and running at peak performance hosts the critical application
which allows the company to do business.
In addition to the keeping the
servers running and secured, the team typically supports applications such as:
Email, Active Directory, ADFS, File & Print, Group Policy (automation and
policy enforcement), Systems Management tools (IE SCCM, MDM, SIEM, IDMS),
SharePoint, Corporate Antivirus management and Helpdesk ticket systems.
Start of day:
1.
Wake up, rub your eyes and grab your
smartphone to check your work email if any alerts or critical requests have
been sent to you.
·
No I’m not kidding… server admins
are a special breed who really cares about their systems and are always keeping
an eye on things
2.
If any alerts or emails need
immediate attention (IE an alert a server’s CPU has been running at 99% or a
drive is near full capacity)
·
jump on your laptop and connect to
the VPN
·
Remotely connect to the server in
question and remedy the problem
·
If a ticket wasn’t created for the
alert, create one. Update accordingly and close the ticket
·
Log off
3.
Go through your morning routine then
head to work.
·
Note: It’s always a good idea to
arrive early to get a jump on any issues before users are calling about some server
issue preventing them from doing their work.
4.
Arrive at work and check the ‘state
of the union’ by checking in with the NOC (if you have one), check the helpdesk
tickets assigned to you and your group & the monitoring dashboard(s).
·
If any critical issues are open,
crack open the laptop and remediate the problem.
·
Update and close the ticket related
to the the issue you resolved. If a ticket doesn’t exist, create one.
5.
If all is good in the world of
servers, head for your morning coffee.
·
Coffee in the morning is a perfect
time to interface with the colleagues within your IT group as well as other
groups.
o Collaboration strengthens IT as a whole and gives you some
insight if there are some other issues going where you may be able to help.
·
Coffee time is also a great time to
talk to end users (non-IT) and find out how things are going
o Listen to what users say closely and read between the lines.
If you pick up on any pain points like an app is slow today or challenges that
some new software could increase their productivity or quality of life…
o Make a mental note and look into it. If its new software
solutions, discuss it with your manager. If you fix a problem for the user,
follow it up in an email to let them know otherwise they think the problem just
went away.
·
Pet peeve note: Careful with
extended coffee runs.
o Many managers interpret long coffee breaks as a negative unless they are aware of your 'recon' motive
o Users who get the impression IT is always on break or arent busy is bad for IT PR.
o Find the right balance of showing time to listen to people and lingering too long
6.
Return to your desk and again, check
on the ‘state of the union’ (Dashboard and tickets)
7.
Begin your normal work assignments
·
This may vary depending on how your
manager assigns duties and priorities. What I found works well is in this
following order:
1) Resolve any critical incident
tickets (something broken)
2) Resolve any critical requests
(something needed)
3) Proactive health checks (20% of
your day)
· Each engineer should have a list of
systems and app to review on a daily, weekly and monthly basis to resolve
problems before they become a risk to an outage.
4) Resume resolving tickets (60% of
your day)
5) Project work (20% of your day)
· Some projects you may be dedicated
100% of your day for the duration of the project. This will be determined by
your manager based on criticality and the project completion date
8.
Throughout the day, expect
interruptions. These will cause you to shift from one assignment to another.
·
System engineers juggle a lot of
tasks concurrently during the day as all other IT groups rely on the
infrastructure teams. Their critical events translate to a critical response by
Infra groups in order for IT to respond as a whole to the ever changing requirements
and needs of the business.
9.
End of shift
·
Don’t expect to leave exactly on
time. What we do as system admins doesn’t exactly finish on our schedule. It is
part of the life we chose in supporting servers that don’t turn off. It is our
responsibility to keep them and the business going.
·
Bring your laptop home with you
every day. If you are not on call, one of your team members is and may need
help. Great teams always back each other up.
·
Check the monitoring dashboards one
last time before leaving
10. End of day
·
The end of the day doesn’t really
end when your shift does as on-call rotations are common. Avoid becoming burnt out by keeping up with other
hobbies outside of tech but definitely do not disconnect completely when going
home for the day.
Server admins are in their position because they are never really disconnected (on-call or not). Part of this is because they become personally connected to their systems and the company. We take great pride in what systems we have built.
Common tickets for System Administrators:
·
Email [spam, delivery problem, mailbox needed]
·
Need a new server for “xyz” project
·
Patch or software deployment needed to [servers,
pc/laptop, users with “X” installed]
·
Mailbox or distribution list needed
·
Alert received for server “X” [connectivity lost, cpu pinned @99%, free
space is low, unauthorized access attempt, URL check failed, etc]
·
Data management [file restore, access
request/removal, new share]
·
New solution or service project [Proof of
concept, test, pilot, deploy]
·
Server or application performance
·
“How do I…” or how to questions
Supplemental
Becoming a Systems Administrator
Server
admins typically have their careers start within the helpdesk or NOC where a
great amount of experience is gained, more so from the helpdesk. Many companies
promote from within when a helpdesk technician shows the traits and technical
initiative needed for server side support.
Skills and traits of a good admin
·
Integrity
– First and foremost this trait is absolutely required.
·
Commitment
– Servers can crash at any time and the business depends on IT to do whatever
it takes to get services restored.
o
Taking part of an on-call rotation
is very common
o
Helping the on-call person respond
to an alert builds team unity
o
Seasoned engineers can all share a
story of at least one time they had to work 24+ hours straight on restoring a
server or recovering all the systems in the DR site (disaster recovery site).
This should be a rarity but it will definitely happen in your career. This
level of commitment is rewarded greatly, not just monetarily but also in the
business relationships you develop with your executive management team.
·
Stay ahead of the curve
o
At this level of IT, you are the one
who should have a solution to every challenge of at least know how to find it
quickly. Stay ahead on your certifications and tech news
·
Vision
o
See the forest, not just the trees.
o
For example:
§
If every day you need to reboot a
server to correct an issue; that is not a fix but a Band-Aid. Find the real
problem so you no longer have to do this daily manual task.
§
The helpdesk keeps getting calls to
set a new users homepage, desktop shortcuts and mapped network drives to match
what the rest of their group has. Create a GPO (group policy object) to
automatically create these settings and save the desktop group hundreds of
hours over the next couple years.
§
An application owner or even a
Microsoft support rep requests you to make a change the server that is having a
problem. Will this cause a problem now or in the future? Will it violate any
corporate standard or policy? Knowing the answers to these questions is
critical.
·
Diagnostic skills
o
A professor I once had the pleasure
of having once described diagnostic or troubleshooting skills as a talent that
“you either got it or you don’t”. I agree with him however I do believe these
skills need to be developed through use to become stronger.
§
Pay particular attention to how
components or pieces come together. What dependencies are there between
components and where are the failure points.
§
Identify what symptoms relate to
different failure points. For example, if Internet Explorer displays a message
“page cannot be displayed’; what’s wrong (if you can name only 1 or 2 things,
it’s time for you to do some studying)
·
Ability to work with little supervision
o
Depends on your manager’s style
however most good managers do not attempt to micromanage the systems engineers
who are often relied upon to be subject matter experts (SMEs).
o
If you have built the confidence
your manager has in you and judgment, they will normally loosen the reigns. If
you find you are being managed more closely than your peers, evaluate how you
can improve by using your peers as examples as well as speaking to your
manager. In order to advance, your manager needs to really trust in you.
Sorry this was such a long post
folks but this job role manages a considerable amount of the overall IT
presence within a company. The success of the server team has a direct impact
upon the productivity of the business and should be a leader of innovation and
security initiatives. Carreers as System Administrators can be very rewarding with talented engineers always in high demand.