Q: Your team wants to implement live migration of a running Docker container from one host to another without stopping it, similar to VM live migration. Is this possible with Docker? What technology enables it?
Docker has experimental support for checkpoint and restore using CRIU (Checkpoint/Restore In Userspace).
Want to master this scenario in a live sandbox? KodeKloud's Docker Certified Associate (DCA) Hands-On Lab Course covers this exact problem with hands-on terminal drills.
🛠️ Production Runbook & Step-by-Step Resolution
Direct Answer: "Container stability relies on clean signal handling (SIGTERM vs SIGKILL) and immutable image tagging. The interviewer is testing: CRIU (Checkpoint/Restore in Userspace), container migration limitations.. I structure my answer around systematic triage first, root cause analysis second, and permanent remediation third."
Initial Diagnostics & Root Cause Analysis
Docker has experimental support for checkpoint and restore using CRIU (Checkpoint/Restore In Userspace). CRIU freezes a running process, serializes its entire state (memory, registers, open files, sockets, timers) to disk, and can restore it later — even on a different host.
- Checkpoint:
docker checkpoint createcheckpoint1 - Transfer the checkpoint data and container filesystem to the target host.
- Restore:
docker start --checkpoint checkpoint1
Remediation & Permanent Safeguards
Workflow: *Limitations:* This feature is experimental and not production-ready. Open network connections break (TCP state doesn't survive cross-host migration). External storage must be shared (e.g., NFS). GPU state, complex IPC, and certain kernel features aren't fully supported. For production workloads, Kubernetes pod rescheduling with graceful shutdown/startup is the practical alternative.
- Checkpoint: docker checkpoint create checkpoint1
- Transfer the checkpoint data and container filesystem to the target host.
- Restore: docker start --checkpoint checkpoint1
Senior hiring managers and bar raisers evaluate how you prioritize during outages. Avoid these common anti-patterns:
-
Structuring Dockerfile instructions inefficiently by copying app source before package manifests when configuring team wants implement
Copying code before package manifests (
package.json,go.mod,requirements.txt) invalidates the build cache on every commit, forcing slow full reinstalls. -
Neglecting the PID 1 process reap problem and omitting proper SIGTERM signal forwarding when configuring team wants implement
If an application runs as PID 1 without an init system (like
tini), it fails to reap zombie child processes and ignoresSIGTERMsignals, causing unclean 10s timeouts. - Assembling single-stage images containing compilers, build SDKs, and header files in production when configuring team wants implement Single-stage builds bloat image size by gigabytes and expose extensive package vulnerabilities; multi-stage builds isolate artifacts into minimal distroless images.
- Confusing container port forwarding on bridge networks with unisolated host networking mode when configuring team wants implement Bridge networking utilizes iptables NAT translations which add latency overhead, whereas host networking bypasses the network namespace entirely.