Cloud-native stories roundup 2023.4

Cloud-native articles interesting to read

Container

ChatGPT for your Kubernetes Cluster — k8sgpt

Kubernetes operations powered by AI, e.g. supported backend has OpenAI ChatGPT (need your openai API key), which is used to generate natural language explanations for Kubernetes issues.

DevOps — 10 Powerful Network Sniffing Tools

wireshark (Linux, Windows, macOS, GUI), tcpdump (Linux, CLI), NetworkMiner (network forensic analysis tool, GUI), Kismet (packet sniffing, Windows and Linux), SolarWinds Network Monitoring Tool (multi-layered tool, bandwidth analysis, NetFlow, sFlow, NetStream, JFlow and IPFIX analyzer, GUI), ManageEngine NetFlow Analyzer (traffic analysis, Windows and Linux), Colasoft Capsa (packet sniffing, Windows, GUI), EtherApe (network sniffing analysis, Linux, GUI), Fiddler (HTTP and HTTPS traffic sniffing, GUI), Wifi Explorer (wireless network packet analysis, macOS, GUI)

How to run distributed performance tests in Kubernetes with K6 and ElasticSearch

K6 follows Load Testing Manifesto:

• Simple testing is better than no testing
• Load testing should be goal oriented
• Load testing by developers
• Developer experience is super important
• Load test in a pre-production environment

use cases include load testing, browser testing, chaos testing, performance monitoring

Script is to test HTTP endpoints, k6 test yaml file is to specify which script (in configmap) to run and parallelism

K8s — Robusta, K8s Troubleshooting Platform

Robusta is both an automations engine for Kubernetes, and a multi-cluster observability platform. With Robusta integration, it is to get hint about crashing Pods, event correlation, remediate alerts and debug pods.

K8s — ChatGPT Bot For Intelligent Troubleshooting

Robusta is integrated with Slack and K8s cluster, and can use ChatGPT for natural language guidance on K8S troubleshooting.

DevSecOps

DevSecOps tool landscape

terraform

pulumi

helm.sh

crossplane

Application management Kubernetes operators

https://kubevela.io

https://alibabatech.medium.com/kubevela-brings-software-delivery-control-plane-capabilities-to-cncf-incubator-db4e51bd6cdf

knative.dev

crossplane.io

https://github.com/metacontroller/metacontroller

https://fly.io

Google cloud run

Azure container run

AWS Lightsail

AWS Lambda containers

App App Runner

Development Environment

gitpod.io

Github Codespaces

okteto.com

devspace.sh

tilt.dev

Pipelines (CI/CD)

tekton.dev

Argo workflows

Jenkins

Github actions

GitOps

Flux

Argo CD

Rancher fleet

Policy management

OPA/Gatekeeper

Kyverno

datree.io

Monitoring

VictoriaMetrics

What makes VictoriaMetrics the new de facto standard choice for open-source monitoring

Advantage: VictoriaMetrics seems to take Prometheus Stack and break it into micro-services architecture using stronger and better new components. It has high availability built-in, as well as superior performance & data compression. Can run on spot nodes to save cost. Some eye-catching comparison with other Prometheus stack mentioned;

• It uses 10x less RAM than InfluxDB and up to 7x less RAM than Prometheus, Thanos or Cortex when dealing with millions of unique time series (aka high cardinality).
• It provides high data compression, so up to 70x more data points may be crammed into limited storage compared to TimescaleDB and up to 7x less storage space is required compared to Prometheus, Thanos, or Cortex.
• It implements PromQL-based query language — MetricsQL, which provides improved functionality on top of PromQL.

Observability strategies to not overload engineering teams — OpenTelemetry Strategy

While manual instrumentation is good, modifying non-instrumented code takes quite a lot of effort. Opentelemetry auto instrumentation for Java and Python

OTLP Exporter Configuration

environment:
      - OTEL_TRACES_EXPORTER=otlp
      - OTEL_SERVICE_NAME=server
      - OTEL_EXPORTER_OTLP_ENDPOINT=http://otel:4317

Use opentelemetry-instrument to start Python application

RUN opentelemetry-bootstrap --action=install

CMD opentelemetry-instrument \
    python \
    server.py