This page looks best with JavaScript enabled

Inference Application Troubleshooting Checklist

 ·  ☕ 1 min read

Pod

View the child processes of PID 1

1
ps --ppid 1 -f

View TIME_WAIT connections

1
netstat -anp | grep TIME_WAIT

Istio

Ingress

1
kubectl -n istio-ingress logs -l app=istio-ingress |grep vlm-tag-v1

EastWestGateway

1
kubectl -n istio-ingress logs -l app=istio-eastwestgateway |grep vlm-tag-v1

Slow XDS Push

1
2
3
4
5
histogram_quantile(0.99,
  sum by (le, pod, type) (
    rate(pilot_xds_push_time_bucket{}[1m])
  )
)

Once it reaches 30s, you need to optimize the push.

Node

Packet Capture

1
tcpdump -i bond0 -n -v -tttt

Specify the target IP and port

1
tcpdump -i bond0 -n -v -tttt host 10.0.0.10

AI

Installing Nodejs

1
2
3
4
5
6
7
curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.40.3/install.sh | bash

export NVM_DIR="$HOME/.nvm"
[ -s "$NVM_DIR/nvm.sh" ] && \. "$NVM_DIR/nvm.sh"  # This loads nvm
[ -s "$NVM_DIR/bash_completion" ] && \. "$NVM_DIR/bash_completion"  # This loads nvm bash_completion

nvm install node
1
npm install -g @anthropic-ai/claude-code

WeChat Official Account
WRITTEN BY
WeChat Official Account