SysAdvent 2015 – Be a Better Engineer: Don’t Do Engineering
I wrote an article for SysAdvent 2015 on becoming a better engineer. Check it out at the SysAdvent blog!
Read MoreHow to Hire Operations Engineers
I’ve noticed a trend recently where companies are giving Ops people the same (or similar) coding interview they use to screen software engineers. They’re usually live coding sessions with the problem set being a CS-y kind of problem (“given a tree that looks like this, find all leafs that match X”). Here’s the problem with this approach: Ops is not SWE. Ops is fundamentally a different skillset than software engineering. Sure, you will often find people with overlap, but they are two different jobs, and you’ve just gotten lucky with a candidate. I have a theory for why this is becoming pervasive: the team making the hire doesn’t understand the role of ops. This seems to happen when a company is hiring its first operations engineer. The existing team are SWEs and...
Read MoreAssessing a New Infrastructure
Over the last week, I’ve put some thought into my methodology for assessing an infrastructure. This is the sort of thing I do when I go to a new client or start a new job, and since I’ve done both of those a several dozen times over the last twelve years, I’ve gotten pretty good at a structured approach to it. This focuses on the infrastructure itself and doesn’t cover cultural or political stuff (topics for future posts, for sure). A few notes about this guide: It’s oriented to web operations primarily, but most of it is applicable for enterprise. An enterprise version would add a lot more about security, data management, and long-term systems management. There aren’t really wrong answers to these questions: the answers should...
Read MoreSix Principles of Effective Monitoring
This was originally published at the Taos Consulting blog on July 14th, 2015. I have led or assisted in many monitoring projects over the years—too many to count. I’ve managed more than my fair share as a full-time system administrator. After a while, I’ve found myself giving the same advice to any one who asks, so it seems only fitting to finally write them down. 1. Monitor for the user and business first. When you stand up a new monitoring platform, what’s the first set of checks you add? It’s usually system-level checks: CPU, memory, load average, etc. They’re easy and of course you need them, right? Here’s a scenario: your company has a web app, through which all it’s business is done. Let’s assume it’s a standard three tier architecture, consisting of...
Read MoreRecognizing and Handling Burnout
This post was originally published at the Taos Consulting blog on April 2nd, 2015. Burnout is something most people in the tech industry are intimately familiar with, but it’s a topic not often discussed publicly. Many people I speak to about it even share a sense of guilt over feeling burned out. “If my coworkers are able to handle their work fine, then I must be doing something wrong”, they say. Managers are often in the same boat, when it comes to burnt-out employees: “They just can’t handle our pace.” I believe both perspectives have a misunderstanding of what burnout is and how to handle it, something I hope to clear up for both. What is burnout? Burnout is a state resulting from prolonged stress. There’s no single cause of burnout, but rather, burnout...
Read MoreStarting, Building, and Running a LOPSA Chapter
Overview This document is composed of documents the LOPSA-ETENN leaders have written to govern how we run our chapter, as well as explanations of why we chose to do things a particular way, other options considered, etc. Essentially, this is a brain dump of everything that has gone into making LOPSA-ETENN a successful chapter. I would like to note that what follows is what has worked for us, and may not work for every chapter. What makes me qualified to speak on this topic? I’m Mike Julian, and I’m the president and one of three founders of the LOPSA-ETENN chapter. ETENN’s very first meeting, in June 2012, had 18 people. We have not had a meeting with attendance that low since. For the first year (June ’12 to May ’13), we averaged 29 people. We are...
Read MoreA More General Look at LOPSA Elections
While I was doing the number crunching for my previous post, I collected a few other numbers and decided to visualize them. Here’s ballots cast, grouped by election year. 2005, the first year of LOPSA, was the best. 2009 was unusually low. The average participation is 135 votes, though the two min and max values skew the mean a bit. Candidates and seats available on one graph: LOPSA consistently has only one more candidate than there are seats. What this means is that if you run for Board of Directors, you have a non-insignificant chance of winning a seat. This is bad. Number of votes needed to win: As you can see, this is actually quite a low...
Read More