Closed theninj4 closed 7 years ago
Sounds like it could be, with some good storytelling, an interesting deep dive into some code challenges that could provide helpful insights for us all. I would be interested to hear more.
I can't really tell you much about the bugs in question without ruining the talk for anyone who comes (stories suck when you know the ending). I will tell you they're not bugs in code we've written, but pertain to quirky / non-obvious behaviour in the NodeJS core.
I'd be presenting both stories as a timeline of their discovery through mitigation, understanding and resolution / workaround.
cool - sounds good - only thing I'd suggest, as a storytelling technique - is writing your initial description and talk title above with some more dramatic anticipation, as that is what will appear in the newsletter and website, and will be your hook to get people in to hear you.
@theninj4 - can you do March or April? I think we have an open slot next week!
I'm on holiday at the moment, getting back in the UK for Monday. Next week will be pushing it, April would be better? I'll tweak the talk description as per @admataz 's suggestion when I'm back :+1:
Great, April..
@theninj4 can I confirm this?
cc'd @simonmcmanus
@iancrowther I'm trying to get my hands on some screenshots of real metrics, it's taking slightly longer than expected. Can I get back to you late tonight / tomorrow?
Sorry @iancrowther I'm having to jump through a lot of hoops (read: trying to restore backups) to get some good screenshots and evidence to support the stories and make it really enjoyable to watch. Can we postpone?
Moved to May, can you confirm?
Yes, lets do it.
Locked & Loaded :-)
Do you guys mind if I tweak the description a little? Or is it too late?
@theninj4 - that should be fine - let us know when it's done.
@admataz - Done. It works much better with the single story instead of two :+1:
Failure happens. Are you ready for it? Here at Holiday Extras we've been running NodeJS in production for 5 years, it's been a critical part of our infrastructure for the last 3 years. In June 2015 we thought we had solved failure... then one day our API crumbled around us. We followed the incident with a Blameless PostMortem which forms the basis for this talk. I'm going to run through how we're using NodeJS in production, the countermeasures we had in place and a chronological run through of what happened, the metrics we care about, the mitigations and the solutions.