Live data from Hacker News

Show HN: Coroot – eBPF-based, open source observability with actionable insights

github.com

11–20 of 32 posts

Re: Show HN: Coroot – eBPF-based, open source observability with actionable insights

#11

I already have Opentelemetry traces and logs going to Clickhouse with the Clickhouse otel exporter. Can i use Coroot to show my existing data, without it taking control of my DDL?

Initially, we relied on the ClickHouse OTEL exporter and its schema, but for performance optimization, we decided to modify our ClickHouse schema, and they are no longer compatible :(

Re: Show HN: Coroot – eBPF-based, open source observability with actionable insights

#12

I already have Opentelemetry traces and logs going to Clickhouse with the Clickhouse otel exporter. Can i use Coroot to show my existing data, without it taking control of my DDL?

Initially, we relied on the ClickHouse OTEL exporter and its schema, but for performance optimization, we decided to modify our ClickHouse schema, and they are no longer compatible :(

Bummer, it'd be awesome if i could point it at data i already have, even if that meant a reduced feature set.

Re: Show HN: Coroot – eBPF-based, open source observability with actionable insights

#13
post #7

What's the data transformation story; for ML on metrics?

Coroot builds a model of each system, allowing it to traverse the dependency graph and identify correlations between metrics. On top of that, we're experimenting with LLMs for summarization — here are a few examples: https://oopsdb.coroot.com/failures/cpu-noisy-neighbor/

That looks like a built-in feature. I'm asking about extensibility. How do we use custom metrics transformations (libraries), for example?

Re: Show HN: Coroot – eBPF-based, open source observability with actionable insights

#14
post #13

Earlier quoted context omitted.

Coroot builds a model of each system, allowing it to traverse the dependency graph and identify correlations between metrics. On top of that, we're experimenting with LLMs for summarization — here are a few examples: https://oopsdb.coroot.com/failures/cpu-noisy-neighbor/

That looks like a built-in feature. I'm asking about extensibility. How do we use custom metrics transformations (libraries), for example?

Currently, you can define custom SLIs (Service Level Indicators, such as service latency or error rate) for each service using PromQL queries. In the future, you'll be able to define custom metrics for each application, including explanations of their meaning, so they can be leveraged in Root Cause Analysis

Re: Show HN: Coroot – eBPF-based, open source observability with actionable insights

#17

This is somewhat off topic, but are there any common uses for eBPF outside of observability/monitoring? Or is that kind of its whole thing?

Yes, one example: network bandwidth isolation is done more efficiently using ebpf https://netdevconf.info/0x14/pub/papers/55/0x14-paper55-talk...

Re: Show HN: Coroot – eBPF-based, open source observability with actionable insights

#18

This is somewhat off topic, but are there any common uses for eBPF outside of observability/monitoring? Or is that kind of its whole thing?

Also commonly used for high-performance networking and security use cases, for example https://isovalent.com/blog/post/cilium-netkit-a-new-containe....

Basically anywhere you'd previously need to write a kernel module but now can have user space run arbitrary kernel code that's secure and won't crash the kernel.

You can also now write custom schedulers in eBPF with sched_ext.

Re: Show HN: Coroot – eBPF-based, open source observability with actionable insights

#20
post #19

Thanks for sharing! If the connections are TLS-enabled, can Coroot still display the associated telemetry?

Yes, it captures traffic before encryption and after decryption using eBPF uprobes on OpenSSL and Go’s TLS library calls.
Post reply on HN