AutoMQ Support Services Policy

Version: v1.0

Applicable product: AutoMQ BYOC

This Policy describes technical support for AutoMQ commercial subscription customers, including human support hours, P1–P4 severity levels, initial response targets, customer collaboration, and the scope of support.

1. Support delivery

Every commercial subscription includes 24/7 server-side health monitoring and proactive incident notifications. Monitoring requires the necessary telemetry, connectivity, and agreed diagnostic access to remain available. Monitoring coverage is separate from human support coverage. Monitoring does not guarantee detection of every fault, continuous human intervention, or service restoration.

AutoMQ provides Direct Engineer Support. Customers submit support requests through the dedicated Slack, Microsoft Teams, or Google Chat channel agreed with AutoMQ and communicate directly with AutoMQ engineers.

An AutoMQ Support Owner handles day-to-day questions and support coordination in each paid support customer's dedicated channel. For urgent issues such as P1/P2 incidents, engineers who develop AutoMQ's product code provide support directly in the channel. They work with the customer to diagnose the issue, discuss mitigation or remediation options, and communicate progress. Human response times follow the customer's Support Plan and case severity.

2. Support hours and initial response

Support Hours: Monday through Friday, 09:00–18:00 in the customer's local time, excluding public holidays in the customer's location. All initial response targets of 8 Support Hours are measured within this window. Requests received outside this window begin accumulating time when the next local support window opens. For customers operating across multiple locations, both parties confirm one applicable support time zone and public holiday calendar in the support channel.

Business Day: Monday through Friday in the customer's local time, excluding public holidays in the customer's location. Business Days use the same customer support time zone and public holiday calendar as Support Hours.

Calendar Hour: Continuous elapsed time, including the customer's local nights, weekends, and public holidays. The clock does not pause because of time zones, working hours, or holidays.

Customers may submit requests at any time. Dev provides human technical support during Support Hours. Production and Enterprise provide 24/7 human technical support. For Production and Enterprise, P1/P2 response targets run continuously in calendar time; P3/P4 targets use Support Hours or Business Days.

SeverityFreeDevProductionEnterprise
P1Not included8 Support Hours1 hour30 minutes
P2Not included8 Support Hours4 hours2 hours
P3Not included8 Support Hours8 Support Hours8 Support Hours
P4Not included2 Business Days2 Business Days2 Business Days

Support Hours and Business Days in the table refer to the definitions above.

An initial response is a human reply from an AutoMQ engineer who begins investigating the support request. Automated acknowledgements do not count. Initial response targets are not service restoration or resolution targets.

Initial response time is measured from AutoMQ's receipt of the customer's support request and accumulates according to the support coverage of the applicable plan. For issues proactively detected through server-side monitoring and communicated to the customer, human response follows the same plan and severity standards.

3. P1–P4 severity levels

Customers may assign a priority when submitting a support request and describe its business impact. AutoMQ confirms the priority based on the actual impact, its scope, and the availability of a workaround. AutoMQ may reassign the priority if the customer's designation does not reflect the actual circumstances and will explain the reason in the support channel.

SeverityDefinitionAutoMQ handling
P1An issue severely impacts the customer's production AutoMQ service for an extended period, or causes it to stop operating completely.Assign specialists to work continuously on the issue, communicate progress toward resolution, and begin work on a temporary workaround or fix in parallel.
P2AutoMQ service capabilities are materially limited, or the service is unstable with ongoing or recurring interruptions that affect normal business operations, without reaching P1 severity.Assign specialists to address the issue, communicate progress toward resolution, and begin work on a temporary workaround or fix in parallel.
P3A localized issue with an available workaround, while the service as a whole continues to operate normally.Engineers follow up during the customer's local Support Hours until the issue is resolved or a workaround is in place.
P4Requests to clarify procedures or documentation, product enhancements or new features, cosmetic or nonfunctional issues, or documentation issues.Triage and assess the request, provide clarification where possible, and potentially include a solution in a future product update.

Severity is determined by business impact, not contract value, the customer's job title, or an isolated monitoring alert. An available workaround does not automatically make an issue P3; if the remaining business impact meets P1 or P2 criteria, that severity applies. Issues in development or test environments generally fall under P3. General questions, documentation issues, and feature requests that do not affect normal service operation generally fall under P4. Severity may change as the actual impact changes.

4. Incident escalation and case handling

4.1 Reporting an incident

For P1/P2 requests, provide as much of the following information as possible: customer and cluster identifiers, whether production is affected, incident start time, symptoms, business impact and scope, affected producers, consumers, or topics, recent changes, actions already attempted, logs or metrics, and a contact available to collaborate in real time. Incomplete information does not prevent AutoMQ from engaging; engineers will request additional diagnostic information after the initial response.

4.2 Responding to urgent issues

Engineers who develop AutoMQ's product code respond directly to P1/P2 incidents in the support channel, investigate the issue, and provide mitigation or remediation options. The Support Owner coordinates the engineers and the required customer collaboration. Internal assignment, handoffs across time zones, and engineering collaboration do not pause the initial response clock.

For Production and Enterprise P1 incidents, work continues until the critical impact is mitigated, provided the customer collaborates promptly and the necessary access and diagnostic information are available. P2 incidents follow the applicable response targets and the next update time communicated by the engineer.

5. Customer collaboration and scope of support

Customers should promptly provide the information, access permissions, and operational approvals required for investigation, and keep a technical contact available for real-time collaboration during urgent incidents so AutoMQ engineers can continue investigation and recovery.

Dev, Production, and Enterprise include diagnosis of AutoMQ product issues, configuration and compatibility guidance, and general usage, upgrade-path, and best-practice advice that can be addressed within a support case. Free includes subscription and billing assistance and does not include technical case response commitments.

Customers may request remote Support Reviews or Office Hours to discuss existing cases, cluster health, risks, and best practices. These discussions do not replace the incident support process. Issues requiring follow-up identified during a discussion are recorded as cases.

Work requiring a formal project or deliverable is outside the scope of support. Examples include migration implementation, operation of customer environments, capacity planning, performance testing, production-readiness or disaster-recovery exercises, upgrade execution, bespoke training, training materials, architecture design, and custom development.

Support excludes versions that have reached end of support, preview or experimental features, unapproved modifications, deployments that do not meet documentation requirements, and issues caused by customer applications, cloud infrastructure, networks, Kubernetes, object storage, or third-party software. AutoMQ may assist with identifying third-party dependencies but does not undertake to repair third-party products or services. Feature requests can be recorded and assessed but do not create a roadmap commitment or a guarantee to deliver a patch by a specific date.