lablup / lablup/backend.ai

Run the agent on Podman through the Docker-compatible API

Open
#13,807 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
670
Forks
183
Avg merge
17h 7m
Merged PRs (30d)
358

Description

The agent talks to the container runtime through the Docker API and assumes Docker's dialect of it in several places. Podman serves the same API but differs in what it puts in the responses and in what it accepts in requests, so each of those assumptions fails separately. This epic collects them.

Two deployment shapes are in view: a rootful Podman socket with the agent running as root, which is where the first reports came from, and a rootless socket with the agent running unprivileged, which is the reason for choosing Podman in the first place. They exercise different code paths — the cgroup statistics mode is only reachable in the former.

JIRA Issue: BA-7382

Contributor guide

Open the contributing guide

Research direction

Start by mapping the agent's Docker API interactions and comparing the requests and responses that differ under Podman. Check both rootful and rootless socket deployments, including the cgroup statistics path. Done means the agent operates correctly through Podman's Docker-compatible API in both deployment shapes.

Written by the indexing model from the issue text.

Assessment

Tech stack
docker, python
Domain
backend, devops
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Active
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.