Papyrus — commercial

Commercial support

Support for self-deployed Papyrus appliances in commercial cloud and on-premises environments. Email reaches the team responsible for the product. If you are running the government offer, use government support.

Commercial deployment workspace

Support channel

commercial@beaglabs.com

Support terms

How support works

First-response targets are commitments to engage and triage. Resolution time depends on the fault, environment, and any third-party dependency involved.

Email commercial@beaglabs.com. Include the severity and deployment ID in the subject line, for example [Sev 2] 7f3a91c4 — onboarding completes but no licence accepted.

Business hours are Monday to Friday, 09:00–18:00 US Eastern, excluding US federal holidays.

Severity 1 reports are reviewed outside normal business hours on a best-effort basis. Standard support does not include a 24×7 on-call rotation unless your agreement says otherwise.

Email is the supported channel. Keeping the deployment ID and issue history in one thread gives the engineer handling your case the fastest path to a useful answer.

Severity targets

SeverityFirst responseUpdates
Severity 14 business hoursEvery business day until resolved
Severity 21 business dayEvery 2 business days until resolved
Severity 33 business daysAs the request progresses

Severity 1

The appliance is down or unusable in production, no workaround exists, and work has stopped.

Severity 2

Functionality is degraded or a subset of workflows fails, but a workaround is available.

Severity 3

Questions, configuration help, documentation gaps, cosmetic defects, and feature requests.

Diagnostics

What to include

Papyrus does not call a Beag Labs control plane and we receive no appliance telemetry. The details you send are what let us reproduce and isolate the issue quickly.

Deployment ID
Shown on the appliance onboarding page and returned by /api/config/public. It is the identifier we use to find the licence record.
Image tag
The Papyrus image tag you deployed, for example 0.1.1. `sudo docker ps --format "{{.Image}}"` prints it.
Environment
Cloud or on-premises environment, region where applicable, VM or host size, and the deployment method you used.
Container state
`sudo docker ps -a` and `sudo docker logs papyrus --tail 200`.
Timeline
When it last worked and what changed immediately before the issue — image update, network policy, model endpoint rotation, or host change.
Expected vs. observed
What you expected Papyrus to do, what actually happened, and the smallest reproducible sequence you can provide.
Scope

What is covered

In scope

  • ✓The Papyrus container image, API, and first-run onboarding.
  • ✓Azure Resource Manager templates and Bicep modules in our repository, including deployment failures.
  • ✓Licence issuance, activation, renewal, and deployment-ID questions.
  • ✓Model endpoint configuration for Azure OpenAI, OpenAI-compatible gateways, and customer-hosted inference endpoints.
  • ✓Upgrades between published image tags and rollback guidance.

Outside standard support

  • —Cloud-provider incidents, quota limits, region capacity, or outages in third-party infrastructure.
  • —Customer-managed network, firewall, DNS, private endpoint, identity, and egress policy issues unless separately scoped.
  • —Availability or correctness of a customer-hosted or third-party model endpoint.
  • —Custom integrations, connectors, prompts, or implementation work created during a professional-services engagement.
  • —Host operating-system administration and hardening beyond what our deployment artifacts apply.
Self-service

Before you write

Azure deployment README

Parameters, disk layout, deployment steps, and checks that confirm a healthy Papyrus appliance.

GET /api/config/public

Reports the deployment ID, licence state, and active profile. This resolves many first-run and configuration questions immediately.

GET /api/license/status

Reports whether the appliance licence is active and when it expires.

sudo docker exec papyrus node /healthcheck.mjs

Runs the appliance health check and exits non-zero when the sandbox or workspace is unhealthy.

Need an engineer?

Start with the support mailbox.

commercial@beaglabs.com