Batch Deployments and Common Errors
Batch deployments run a template across many devices simultaneously or in waves.
How to start
- Go to Templates > select a template > Deploy.
- Choose devices by project, tag or search.
- Adjust variables and review the change set.
- Start the job.
Tracking
- The deployment job shows per-device status: queued, deploying, success, failed. Click a device row for logs.
Managing a running job
- Cancel: If a wave is misbehaving, click Cancel on the job to stop remaining devices from starting.
- Retry failed: After completion, you can target only failed devices and redeploy the same revision.
- Deep dive: Open Events and click the corresponding batch deployment event to see state, timestamps, Job ID, and per-device logs.
Common errors and remedies
- Image pull failed: Check registry credentials and network reachability; verify image tag/digest.
- Port conflict: Another service uses the requested port. Change the variable or stop the conflicting container.
- Insufficient resources: Device lacks CPU/memory/disk. Reduce resource limits or free resources.
- Script error (pre/post): Review script output and exit code; add retries or guard checks.
- Network timeout: Increase per-device timeout or lower concurrency for constrained links.
- Variables missing: Ensure required variables have defaults or are provided at deploy time.
Best practices
- Use canary waves: deploy to a small subset first, then expand.
- Record deployment IDs in change tickets for traceability.