SysAdmin Day is an annual opportunity to thank the folks at our organizations who keep critical (and not-so-critical) systems humming along day in and day out. This year we thought it'd be a good idea to check in with some sysadmins and ask about the other 364 days a year. Specifically, we wanted to know what do you wish your co-workers knew about your job?
File a ticket
You have a problem, and reach out to the help desk or your friendly neighborhood admin. It's a quick fix, you're sure, but ugh they want you to file a ticket! What a pain, right? It might sound like they're giving you the cold shoulder but that's (usually) not the case. Admins want users to file tickets for a number of reasons.
First of all, it helps them manage their time. It's hard to focus on longer projects when you are pelted with "this will just take five minutes" requests all day. Also, other people have been waiting for their ticket to be handled.
Secondly, admins may need to account for their work and demonstrate that they are -- in fact -- busy and not just playing Doom Eternal all afternoon.
Also, it helps keep track of problems that crop up frequently and assists with institutional memory. A well-kept ticketing system with a good search tool can help admins identify long-term problems that need fixing, and reduce the time to fix problems in the future by documenting how they were fixed today.
Simple problems may not have a simple solution
A problem may appear simple, but the fix may be fairly complicated and require more time than it would appear on the surface. Oftentimes, a fix for a simple problem is really an exercise in Yak Shaving that means fixing several problems -- some of which may require solving other problems before you can get to the root of the issue.
Most users can relate to this with consumer applications. You want to open a document, but you need a newer version of the application to open the document, but you can't upgrade unless you also upgrade your operating system, which also requires a hardware update…
With enterprise IT, it can be even more complicated and a "simple" fix to an application might require several days (or more) of planning, testing, and upgrades.
Do no harm
Like doctors, administrators have to be sure that the cure isn't worse than the disease. That is, fixing a "simple" problem can wind up causing even larger problems if you're not careful.
We've come a long way in IT, and things like Linux containers have helped to isolate applications so that making a change to fix one application doesn't sabotage another. But many of our systems are still complex, interdependent and require testing and planning to avoid causing problems while trying to solve others.
So many systems: And they never stop evolving
You've probably heard the phrase "lifelong learning." There are few professions today that are more dependent on continual education, and not all of the knowledge admins need is readily available in a book or course.
Sure, vendors like Red Hat provide certifications that help admins stay up to date with technologies they're using to run their organizations' systems. But that doesn't cover everything that admins need to know, and there are fundamental troubleshooting skills that require hands-on experience and a specific mindset that can't just be taught.
Every organization is unique and has its own combination of homegrown systems, vendors and hardware. Not to mention unique business needs, security concerns and possibly regulations that require compliance.
Most admins I know are very passionate about their jobs and the technology, but it's a lot to keep on top given the speed with which systems evolve.
Sysadmin may not be an 8 hour a day job
This shouldn't come as a surprise, really -- enterprise IT systems are usually going all day, all week, and all year. That means that lots of admins are pulling extra duty outside the nine to five, doing upgrades or system changes outside "normal" business hours to avoid disrupting their teammates. Or they may be jumping in to deal with an outage at 3 a.m. Saturday morning.
Lots of organizations are making strides with DevOps practices, cloud-native applications and systems that are resilient and architected to avoid these late-night emergencies. But the world is still full of surprises and systems that need a little more hand-holding to keep chugging along. As long as we live in a world full of surprises, being a system administrator is never going to be an entirely nine to five job.
When we've done our job well, you won't be sure we've done anything at all
What do sysadmins have in common with bridge builders? When they've done their job well, you probably don't think about them at all. It's only when something breaks that many folks think about the IT operations folks in their organization.
But the truth is that a lot of work goes into being able to take an organization's IT systems for granted. It may not be visible or obvious, but that's the goal. Keeping dependable systems up and running.
Bonus home round: No, we don't want to fix your computer
I asked on Twitter and on some internal Red Hat lists for things admins wanted their co-workers to know, but I got a bonus response for the friends and family of admins. Namely, "no, I don't want to fix your computer."
In part this might be a practical response. As a Linux user, I have very little Windows experience. If you're having a problem with a Windows system, there's a really good chance I'll make it worse rather than better.
Happy SysAdmin Day!
Thanks much to all the folks who responded to my call for ideas on Twitter and Red Hat's memo-list and technical account manager (TAM) list.
If you work with sysadmins, be sure to take some time to say "thanks" for all the hard work they do to keep the organization running. Gifts of food, beverages or other tokens of gratitude are a great way to show your thanks.
If you are a sysadmin, we hope you have a great SysAdmin Day and all your systems are running smoothly and the ticket queue is full of thank you notes from your users. Even though it's only celebrated one day a year, we do appreciate you all year 'round.
About the author
Joe Brockmeier is the editorial director of the Red Hat Blog. He joined Red Hat in 2013 as part of the Open Source and Standards (OSAS) group, now the Open Source Program Office (OSPO). Prior to Red Hat, Brockmeier worked for Citrix on the Apache OpenStack project, and was the first OpenSUSE community manager for Novell between 2008-2010.