Solutions create new problems.
Continuous Discovery asks, “What new problems did our solutions create?”
Happy o11ydays from us to you!
We’ve got some fun and festive o11y treats for you, so be sure to sleigh ride into this blog to learn how you can make the cool heatmap art you've seen us post, and maybe get an awesome o11y gift!
And while we originally wrote this with LinkedIn and Twitter in mind, if you follow the submission instructions on Mastodon we will also send you fun stuff :)
https://www.honeycomb.io/blog/happy-o11ydays
#o11ydays #o11y #opentelemetry #testinprod #honeycomb #observability
#o11ydays #o11y #opentelemetry #testinprod #honeycomb #observability
Don't mind us, just deploying a new channel three days before the winter holidays, like you do.
I did something pretty dumb last night but learned some valuable lessons.
It was late and I had noticed that on my Windows 11 home laptop, the Core Isolation - Memory Integrity was disabled. There was a yellow warning around it so I looked it up (briefly) and saw it is usually enabled by default.
I turned it on, restarted my laptop and BSOD. WTF.
-Restarted in Safe Mode - couldn't access Windows Security/Core Isolation.
-Tried messing with the command line to find setting and failed
-regedit because I saw that I could possibly disable in DeviceGuard...but that wasn't on my profile somehow? Failed
-Restored from backup (luckily there was a back up recently, so no lost data) and success.
Lessons learned:
-Don't mess around with settings when you're sleepy. It did keep me up late however 😂
-Do proper research on settings and not just Wild Wild West it. I have an AMD Ryzen processor and that setting is disabled because AMD and Memory Integrity don't play well.
-Revisit my backup frequency settings; saved me this time
-Don't do dumb things?
#amdryzen #windows11 #shouldhaveknownbetter #testinprod
@jerry JFC it's not that hard to spin up a droplet instance if you want to test stressy shit out
Bonus tip. If your QA environment doesn't have the same traffic patterns and comparable volumes to production I'm afraid it can tell give you less confidence than you'd expect #testInProd