Believer in monitoring as a critical foundation to run any real world system. Cofounder/CTO @chronosphereio. Began open source M3 https://xitter.cf/t.co/msolsX9wqR.
Joined April 2009
- Tweets706
- Following609
- Followers949
- Likes458
Oncall "Distraction Tax" is an immediately recognizable term for anyone who has faced this in the real world. Some will remember vividly months or sometimes years climbing out of bad periods, or even some teams needing complete restructuring.
@billyo11y : forbes.com/councils/forbeste…
#busydevlife During holiday-slow-down and pre-new years eve, now updating MacOS so I can run newly released Ghostty for a fast native terminal that I would have loved over the past few years (TY Mitch and community! The future is now)
11.30pm after landing in LGA in New York after meeting-packed day in Chicago and the airport is packed with folks leaving, more so than any Australian airport I’ve ever seen at its busiest as a bunch of planes land at once on a “normal” day. Scary: Humans quickly normalize scale.
A really interesting read on LibreSSL and its journey of its fork from OpenSSL:
openbsd.org/papers/bsdcan14-…
By the numbers:
OpenSSL 1.0.1g was a 388,000 line code base.
About 90,000 lines of C source code deleted, about 150,000 lines of files.
Approximately at 500,000 line unidiff.
So brilliant to see PromLens w/ complex query visual editing, PromQL JOIN interactive explain/debugging, label & size of result set per sub-statement in the query now part of @PrometheusIO.
Here's to making things that stand the test of time & move the entire industry forward.
You’re unable to view this Post because this account owner limits who can view their Posts. Learn more
Hope folks are gearing up to hear about the latest all things Prom at PromCon and from Eric about @PersesDev OSS (Apache 2) which is re-imagining a purely observability focused visualization, graphing and debugging UX designed from the ground up to leverage the best of PromQL.
There's a new kid coming to @CloudNativeFdn Sandbox.... to celebrate I'm sharing #o11y fun with integrating #cloudnative #observability with @PersesDev visualization & @PrometheusIO at @PromConIO EU next month! @CNCFAmbassadors @chronosphereio bit.ly/promcon-eu-new-kid-in…
🧵 Leading the Magic Quadrant!
Stats: 34 Chronauts poured in many many hours over months on our submission, >300 questions answered on one of the most complex spreadsheets I have ever collaborated on with a large group (we had to lock rows), we also exchanged >1,000 Slack msgs
We’re very excited to announce that we’ve been named a Leader in the 2024 Gartner® Magic Quadrant™ for Observability Platforms!
Get your copy of the full report here: okt.to/McZGtf
#observability
For #2 it’s all about being able to cut through noise and triage effectively, not necessarily root cause until the system is reliable and stable again. Root cause comes after when ensuring the same class of issue will not reoccur.
Furthermore “Without the proper categorization and tagging mechanisms, you are faced with too much data to figure out which events in the stream of event log data actually relate to the issue you’re trying to solve”
The first two points the article touches on, which I point out are the most important to quantify first are:
1. Understand the level of reliability required for an application or service.
2. Avoid quality and reliability plus telemetry as an afterthought.
For #1 for instance “Are there specific SLAs or business outcomes that should be met? If so, it’s all about working backwards from measuring the app reliability at present against those targets or goals, and then minimizing incidents that risk or impact that reliability.”
🧵Recently I had to step back and capture logically how to run services with large surface areas reliably (either lots of users, lots of complex features, or raw transaction/query scale regardless of users or features).
Read thread or skip to article:
informationweek.com/software…
Generally agree we’ve come along way, but BOY do I wish every laptop still had a 3.5mm headphone jack still 😭
Also… HDMI is still a must IMO.
More also... MagSafe IMO still rules the roost of “UX/consequences of tripping over the charger between zoom meetings for bio break.”
Happy @Monitorama 2024!
Today we launch our Telemetry Pipeline, with the avg enterprise logs growing 250% YoY we can save >=30%:
chronosphere.io/lp/simplify-…
Also tune in now, @paigerduty is talking about logs and how to make the best of them with modern practices at 11:30am PST.
See those at @Monitorama next week. Looking forward to the talks from Salesforce and CloudFlare. We have history w/ @chronosphereio as Martin mentions but I also just love what a practical conference this is for practitioners since 2015 when I started following it. See you soon!
The Multnomah Whiskey Library has a special significance in Chronosphere history—it's where we handshook the deal for our first round of funding with @jerrychen of @GreylockVC.
Join @roskilli and I on June 11 at 7pm in PDX for drinks and great conversations at #Monitorama. We’re excited to connect, share ideas, and hear your thoughts.
go.chronosphere.io/cheers-to…
100% agree Eduardo and congrats @OrbStack. This was a per peeve, FS perf of Docker on macOS is soo slow. Especially for quick build-run-test loops for iterating testing changes or changes to a binary with a multi-arch image build to test metrics/logs/tracing agents in test env.
I am glad to hear that @OrbStack folks are doing a great job of improving the file system performance of Linux VMs/containers on macOS! check this blog post: orbstack.dev/blog/fast-files…
Our latest eng blog post is a deep dive on building a converter from non-Prometheus queries by parsing them into an AST. This AST representation is then converted to it's corresponding PromQL AST resulting in a Prometheus query that charts the same data:
chronosphere.io/learn/conver…
