WinOps 2018 Recap
I’ve had the chance to re-visit WinOps London this year, meet a couple of old friends and make new ones – after all, a great event! I really enjoyed all of the talks and workshops I attended, but following topics somehow really stood out for me. (I am honestly looking forward for the videos being available on YouTube, so I can watch those talks for second time!)
SRE with Steven Murawski and Trond Hindenes
Negating releases is not a solution to face site reliability issues / incidents.Steven Murawski
If it’s scary. You are not doing it right. Build robust systems: TRUST but VERIFY.Trond Hindenes
Systems Reliability Engineering” really did feel like a “hipster buzzword” to me (sorry) .. UNTIL I attended those two talks, one from Steven Murawski and one from Trond Hindenes. Steven really gave some important insights in the why, Trond into the how of their experience with SRE in 2017/2018 and they presented some incredible results and numbers (achieved uptime, basically).
Docker with Elton Stoneman
Tearing down monolithic applications does not imply we have to do code changes. Elton Stoneman
To be honest, I never really thought about the possibility of splitting a monilithic app into pieces without having to modify the existing code base. Although I think nobody will ever be on par with Elton in concern of “containerization speed / efficiency”, seeing how a DotNet 3.5 app is split up into a bunch of containers whilist no code changes are being carried out really is an experience.
Chocolatey with Rob Reynolds
You can’t spell system automation without package management. Rob Reynolds
In the last two years I’ve been lucky enough see a couple of Rob’s talks and watch how Chocolatey matured and grew both as a tool, and as a company. The fluffy boss unicorn and dancing coworkers in Robs presentations have just been beat by the CCM (Chocolatey Central Management) live demo this time. I can’t wait for this feature to get out of it’s beta phase.
PowerShell with Chris Gardner
Pester is your friend. Chris Gardner
Watching Chris talk about how you should be developing your PowerShell modules really reminded me of a couple of things I tend to forget when I’m down coding.
- when you’ve got a bug, write testcase for that bug
- AFTERWARDS, fix that bug!
- create docs for each public, preferably also for each private function
- make the docs available!
So that’s my writeup for WinOps LDN 2018. Despite revisiting some topics I kind of forgot about I was lucky enough to get some insights in how folks at other companies tackle problems I’m still facing in my daily work – got a couple of ideas for improvements too, so I’d say – a win 🙂