Inside 2021's infrastructure for Changelog.com

This week we’re talking about the latest infrastructure updates we’ve made for 2021. We’re joined by Gerhard Lazu, our resident SRE here at Changelog, talking about the improvements we’ve made to 10x our speed and be 100% available. We also mention the new podcast we’ve launched, hosted by Gerhard. Stick around the last half of the show for more details.

Discussion

2021-05-22T11:25:59Z ago

Maybe give Honeycomb a try, to diagnose Kubernetes spikes?
https://www.honeycomb.io/kubernetes/

Get Liz on, could be an interesting podcast 😁

2021-05-24T17:24:59Z ago

I like this idea! /cc @gerhard 😉

2021-05-23T16:59:40Z ago

Super excited about the ship it podcast!

I have only been listening to this for the past 2 years, but it has made me so excited to hear about your adventures that you all have gone on with it! Not very many people talk about that in a podcast format ( which is my preferred information ingestion medium ), so I loved hearing about this and can’t wait to hear more with ship it!

2021-05-26T19:46:56Z ago

Happy to have you along for the ride, Alex!

2021-05-23T17:15:18Z ago

Also, I don’t know if this will be applicable for your infrastructure ( because IIRC they are a type of APM ), but pixie labs uses some pretty cool technology ( eBPF ) to help with observability. So, maybe asking them if they could help or recommend someone else?

You can learn more about it on this SEDaily episode ( they were recently acquired by New Relic, so that is why it doesn’t say pixie on the show title 😅 ).

2021-05-23T20:50:09Z ago

You could use the service mesh like linkrd to check for data integrity in transit. There’s a lot to service mesh