, 31 tweets, 10 min read Read on Twitter
First up at @ServerlessBOS is @mipsytipsy!
Observability and software ownership. We write systems, not software
You should morally love the system you work on. Otherwise, quit your job
Best code: no code at all; second best: code that someone else maintains that you can reuse; third: anything else #ServerlessDaysBOS
The industry is getting better at building large systems and projects, but not fast enough #ServerlessDays
We build systems that are unmanageable, undebuggable and unmaintainable by even the best engineers. This wastes time, and is crushing #ServerlessDays
Stripe survey: bod code has real costs. Estimated at least 10% of time is spent dealing with bad code #ServerlessDays
@LyddonB has a talk around the fact that we have no idea what our systems are doing #ServerlessDays
Complexity is increasing in every system. Take the LAMP stack and compare it to Parse - the diagrams get wildly more involved #ServerlessDays
Even more complex - think about the national electrical grid. "It would be folly to try to stand up a staging environment of the electrical grid". You need to have indicators of what is happening in the system #ServerlessDays
There's a magnificent refresh of how we design, ship, and maintain systems, We're no longer under the illusion that we have control #ServerlessDays
"Imagine you're working with serverless code and you'll do all the right things" - @mipsytipsy #ServerlessDays
Operability of code is now a core directive. You need to focus on production excellence to have 24x7 systems. Otherwise, you should work on less available software #ServerlessDays
Put software engineers on call because "pain is nature's gift." Teaches them intuition about what code should be in prod. But importantly: don't make being on call suck! #ServerlessDays
Make on-call not suck by not letting your system get to a terrible and unmanageable state. #ServerlessDays
Observability driven development is the new TDD. TDD stops at the edge of your laptop - doesn't have concurrency, traffic, the chaos of users #ServerlessDays
You're not bad at debugging and dealing with production incidents, your tools are just wrong for the job #ServerlessDays
Observability: how much can you understand about the inner workings of a system just by looking at it's outputs #ServerlessDays
Observability: can you answer any question about what is happening in your system without deploying more code? #ServerlessDays
Need to get the right level of abstraction such that you don't care if the infrastructure is down #ServerlessDays
Track events, not metrics. (jk track both, but rely on events more) #ServerlessDays
Metrics are geared towards low cardinality requests. What if you want to track higher cardinality things like requests without relying on logs and potentially DDOSing your system #ServerlessDays
Events are necessary for debugging, not a "nice to have." Keep the necessary metrics - throw away the 200 responses, keep the 500 #ServerlessDays
"I really hate dashboards" - @mipsytipsy; They end up growing in number and don't let you thing about your infrastructure. Every dashboard is an artifact of some past failure #ServerlessDays
Everyone should be in production all the time and it should not be a big deal #ServerlessDays
Aggregation only works to answer the one question you expected to ask. It smooshes away the details and doesn't let you answer new questions #ServerlessDays
Systems need owners not operators because owners have skin in the game #ServerlessDays
Observability driven development - have to watch things run in production. Don't just look at prod when it's broken. Interact with production every day #ServerlessDays
We've been taught to fear production b/c other people drum it in to us, and we only tend to look there when things are broken #ServerlessDays
Don't lean on paging alerts to debug. Don't need to do that! Delete them all (but use @honeycombio first) #ServerlessDays
How to handle compliance in observability/living in prod? (@lizthegrey - how do you figure out how to get the two-person verification, then having access to prod data isn't the scary part) #serverlessdays
Missing some Tweet in this thread?
You can try to force a refresh.

Like this thread? Get email updates or save it to PDF!

Subscribe to Ann Guilinger
Profile picture

Get real-time email alerts when new unrolls are available from this author!

This content may be removed anytime!

Twitter may remove this content at anytime, convert it as a PDF, save and print for later use!

Try unrolling a thread yourself!

how to unroll video

1) Follow Thread Reader App on Twitter so you can easily mention us!

2) Go to a Twitter thread (series of Tweets by the same owner) and mention us with a keyword "unroll" @threadreaderapp unroll

You can practice here first or read more on our help page!

Follow Us on Twitter!

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just three indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3.00/month or $30.00/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!