As I have said before, availability isn’t usually “fun” and one of the jobs of SRE is to evangelize.
One of the more effective ways to communicate broad concepts is… a meme. It can feel… foolish… at a “professional” company to plaster goofy images all over the place and possibly even put yourself out there (even at GitHub), but it also works.
Stupid things are way more memorable than boring powerpoint slides. Here are some that we have used at GitHub, not all of which I can take credit for.
In a campaign to get our engineers to declare internal incidents early and often:
Rewarding our early incident commanders with challenge coins that have our oft repeated availability slogan (“Availability is Job 1”):
Our required first responder training video has walruses attacking the data center (h/t Kyle Robertson):
Here is me slamming my arm in a door to demonstrate mitigation vs. root cause (broadway is not in my future):
Finally, I wanted to point out a great podcast episode about David Henke building a culture of “site up” at Yahoo which also seemed to feature its own memes:
Happy memeing!