I saw this and thought that this is a great idea. Since there doesn’t seem to be any guides out there, I decided to write one myself, but the problem is that I don’t really know if there is a one-size-fits-all approach to building small business IT infra.
After a lot of thinking about this, I came to the conclusion that I can’t exactly write a step by step guide that someone can follow, but I can definitely write a beginners’ guide to help get you started. This would be similar to what you might see when researching how to get better at a game that doesn’t always follow the same script. Instead of telling you what to do at specific points, I can show you the general mechanics of the game you’re getting into, common problems and their solutions, and some tips and tricks you can take advantage of. It won’t get you all the way to the end, but hopefully it’ll save you some headache getting started and teach you a few things that will prove to be helpful the next time you jump into the game.
The Rules
Let’s start with the rules of the game. Chances are, if you’re reading this, you’re looking for answers. Answers that aren’t in your typical places because those typical places are geared towards enterprises with huge budgets and staff. So, I’m going to make some assumptions.
- You don’t have enough manpower on your team. – Everything needs to be done 2 days ago, except for the things that needed to be done 2 months ago.
- You’re missing expertise on your team. – There are more jobs and specializations out there than you have people on your IT team.
- Limited Budget. – Sometimes you have what you have, not because it’s the best, but because it’s simply what your company could afford.
- Operations and Projects are done by the same team. – Larger organizations have dedicated NOC teams that handle the operational load so that the project teams can focus on making improvements or completing projects for the rest of the business. We don’t have such a luxury.
If these rules don’t apply to you, feel free to keep reading, but just know that it may not apply to your situation. Going through the extra work to apply these strategies may not be worthwhile because we’re simply solving problems that you may not and/or will never have.
Otherwise, I hope you enjoy and learn a few things. Most of my tips for this beginners’ guide covers how to solve the challenges around the above rules.
Making Your Work Effective
Since we’re already up to our eyeballs in work, we need to make sure everything we do is effective. Meaning it needs to actually accomplish something.
There are a few ways to improve this.
- Not moving onto another task until your current one is complete.
- Break work into smaller chunks.
- Avoid doing useless work.
Anyway, the first tip I have is quite simple: Work is only helpful when it’s done.
This sounds obvious, but we don’t always take advantage of this mechanic.
Doing Everything One Thing at a Time
The more WIP (Work in Progress) you have, the longer your lead times are (time until you finish an individual task).
Assume you have 3 tasks, each requiring a month to complete.
A typical approach would be to work on all 3 equally. Another common approach would be to start work on one, and once there’s any difficulty, switch to another task.
The problem with this is that it’s totally possible that 2.5 months in, none of the work would be complete and the work that you have done so far has had zero benefit to anyone. Even if 90% is done on all 3, it’s not helping anyone at the moment.
Finish Faster
Instead, the smarter thing to do in this situation would be to work on your tasks sequentially, not changing tasks until they were finished. This way, you would be providing some benefit even after the first month regardless of the overall completion of the project as a whole.
You’re probably already somewhat aware of this happening. Everything you’re currently working on could probably be done in 2-3 hours of quiet time, but you can’t get any time to focus. You’re getting pulled in too many directions.
It’s a lot better to finish 1-2 tasks today then to do 10% of all 10 of your current tasks. You’ll start to receive the benefit of those tasks being completed right away, and you gain the luxury of no longer needing to remember to do them.
Out of all of your tasks, choose the top 2-3 urgent + important ones and ignore the rest. It may feel irresponsible to ignore some of the other tasks, but try to remember everything will be done more quickly in the long run.
Creating a Checkpoint
Another tool you can use at the same time is to break your work down into smaller chunks. Let’s use my routing cutover as an example.
Example Using Checkpoints with BGP Cutover
When I started with one client, they had static routes everywhere. The way they failed over between VPN tunnels was to login to the firewalls from their public IP, then increase the metric for the primary static routes so that the secondary ones would take over.
Not only was this work to do and have to undo later, I had to drop what I was doing when there was a circuit outage so I could bring the site to site connectivity back up.
The Project Was Too Big to Do Quickly, so I Needed a Plan
The end goal for my project was to get all the sites sharing routes and achieving failover via BGP. If you’re not familiar with routing protocols, this is a decently sized project by itself. I also had a few competing priorities. At the time, headquarters and the data center only had 1 internet circuit configured, and there was only one firewall with no HA.
To get this done sooner so that I get the most benefit with the least amount of work, I established a few checkpoints that provide benefit.
- Routing Pilot – BGP advertising of the printer subnet from the HQ to a remote site.
- S2S Routing failover achieved – Primary routes are advertised via BGP. Floating static routes can stay for backup tunnel.
- Full BGP routing achieved – No more static routes for S2S. Adding/removing subnets can be done more easily.
Routing Pilot Went Well and Was Easy to Get Approved
The routing pilot was designed to be as small and unimpactful as possible. It can be difficult to get management to approve an entire routing protocol redesign since it can be a large undertaking with a lot of risk. Getting them to approve something that is very low risk was trivial. No one from a remote site should need to print to a printer miles away. Even if they did, I had a test plan and rollback plan, so the impact would be for 30 minutes of downtime at most.
Because the change was so small and the impact was so minimal, I was able to get approval to do this during the day.
The pilot went extremely well. I setup BGP between the 2 sites, and advertised the printer subnet to the remote site. Then, I deleted the static route from the remote site. I was running a continuous ping the whole time and saw no impact.
The Routing Pilot Helped Get My S2S Failover Change Pushed Through
Due to the success of my pilot, I was able to have my proven process approved for the next day. After confirming we had no issues the following day, I followed the same process successfully during the day for the rest of the subnets between the 2 sites.
I was able to do the rest of the sites the next week.
Leaving the Rest for Another Time. We Have Bigger Fish to Fry
At this point, I had achieved the biggest milestone I was hoping for. S2S routing failover with BGP. While having full BGP routing would have been nice, getting the secondary internet working at the HQ/DC was a bigger deal, along with setting up HA. I left that last part as another project for a future date when it was more of a concern.
Creating checkpoints allowed me to take a project that should have taken 1-2 months to plan and execute, and receive 80% of the benefit in 2 weeks time. The reduced risk also had the added benefit of making it easier for management to accept. Otherwise, I might have needed to wait several weeks for a change window. I also would have had to do a lot of extra work communicating the change.
Avoid Doing Useless Work
I learned this from reading the 7 Habits of Highly Effective People. You can classify all work into one of these 4 quadrants. You can use this as a guide to eliminate work in order to shed load.
Urgent | Not Urgent | |
Important | Quadrant I •Crisis •Pressing problems •Deadline driven projects | Quadrant II •Relationship building •Finding new opportunities •Long-Term planning •Personal Growth •Recreation |
Not Important | Quadrant III •Interruptions •Emails, calls, meetings •Popular activities •Proximate, pressing matters | Quadrant IV •Trivial, busy work •Time wasters •Some calls and emails •Pleasant Activities |
Minimizing work done in quadrants 3 and 4 will give you more time to focus on what matters. When starting, you may have all your time consumed by quadrant 1. If that’s the case, try to dedicate at least 20% of your time each week (1 day per work week) to preventative maintenance projects.
I’ll go more into that in another article, but just being aware of these classifications of work can help guide your decisions.
Thank you for checking this out. I hope you learned something new or enjoyed reading this. If you had any comments, questions, or just wanted to share your thoughts on this article, you can contact me at blog@e-mayhem.com.
Comments are closed