Service Level Agreement
Service Level Agreement (SLA)
Between The Things Industries (TTI) and Customer
Version May 7th, 2021
Availability – accessibility of the Cloud Services by the Customer, calculated by subtracting from 100% the percentage of minutes during the month in which there was Unavailability;
Business Hours – hours between 8.30am and 6pm CET/CEST;
Cloud Services – integral hosted and managed service by TTI, including core LoRaWAN network server components, management components and additional tools for processing and analysis of network traffic, together referred to as The Things Stack Cloud or The Things Stack Dedicated Cloud;
Console – the graphical web-interface to manage the Cloud Services accessible by Customer;
Customer or End User – the individual or company that enters into an EULA with TTI;
Defect – a Defect is a malfunction in the Software due to the Software failing to perform according to its specifications;
Effective Date – the date on which the Cloud Services are commercially accessible by Customer;
End-User License Agreement (EULA) – the underlying agreement with generic terms and conditions between TTI and Customer, of which this SLA is an exhibit;
Incident – operational occurrence in the software as a result of a Defect, that degrades the Availability and which is validated by TTI as a degradation of the Cloud Services;
Order – TTI's product ordering documentation or purchase flow as described in the EULA;
Reseller – authorized entity who resells TTI products to a Customer on TTI's behalf;
Software License – commercially available and downloadable software products, referred to as "Enterprise" or "Self-Hosted" deployments;
Support – direct contact with TTI Help Desk, for incident and bug reporting, service unavailability, feature requests, etc.;
Term – subscription period during which Cloud Services are made commercially available to Customer;
Unavailability – period in minutes where there is limited or no access by Customer to the Cloud Services, subject to the terms in this SLA;
Written Notice – a message submitted via email@example.com or the Portal at https://thethingsindustries.atlassian.net/servicedesk/customer/portals.
||(excluded from SLA)
Table 1: Specification of Cloud Services components
- This SLA describes the guarantees wrt Availability and Defect restore times for the Cloud Services;
- This SLA describes TTI's Helpdesk support processes, scope and response times;
- This SLA does not apply to Software Licenses, except Section 7 (Helpdesk Support);
- Given the evolution of technical standards and changing market conditions, TTI may update this SLA from time to time. Customer is bound by the most recently published SLA, available on www.thethingsindustries.com.
- Claims and issues with regards to the Cloud Services may be settled between Customer and TTI, or between Reseller and Customer in case Customer purchased the Cloud Services via Reseller;
- This SLA is applicable from the Effective Date.
3. Cloud Services Uptime Commitment
- For the purpose of measuring the quality of Cloud Services, TTI provides access to the Cloud Services on a 24x7 basis at a monthly Availability of minimum 99.9% on all Service Components (" Services Uptime Metric") (see Table 1);
- TTI commits to Response and Recovery Times for Incidents during Business Hours, in Table 2:
||Recovery Workaround Time
|Critical Component down
|Critical Component impaired
|Non-critical Component down or impaired
Table 2: Response and Recovery Times
- Reporting an Incident by Customer must be done by a Written Notice. Response Time is calculated from the moment TTI received the Written Notice;
- Incidents relating to Critical Components must be reported via the Portal and Customer must appended the 'Incident' label to allow TTI to set the right priority and internal alerting.
4. Boundaries and Exclusions
The Services Uptime Metric shall not apply to Unavailability from:
- The Things Stack Cloud Discovery tier (the free evaluation tier);
- A result of Customer-side equipment or third-party computer hardware, software, or network infrastructure not within the sole control of TTI;
- Unavailability of services by TTI's hosting provider;
- Issues related to third party domain name system (DNS) errors or failures;
- Outdated operating systems and internet browsers;
- Scheduled maintenance such as stack version updates and infrastructure updates. Maintenance with expected noticeable downtime will be communicated to Customer 5 days in advance. TTI strives for a maximum maintenance window of 2 hours;
- Emergency maintenance required to prevent further degradation of the Cloud Services, for which Customer may not receive advanced notice;
- Third party or Customer-side services that exceed TTI's API rate limits.
5. End-device and gateway requirements
SLA provisions apply to devices and gateways that meet the following requirements. The reason for these requirements is to allow Customer to apply best practices for reliable and scalable deployments. For cases that do not meet these requirements, this SLA is offered on a 'best effort' basis. TTI may decide to not treat or resolve such cases at all, in which case Customer will receive a notification.
Devices are required to:
- Be LoRa Alliance certified;
- Have a DevEUI issued from an IEEE block, from the device maker or the Customer;
- Be assigned DevAddrs issued by the Network Server of The Things Stack Cloud;
- Have not been registered on another LNS before with the same security keys;
- In case devices support a configurable join interval, joins must back off to at least 10 minutes after several failed attempts;
- In case devices support a configurable initial data rate for join-requests, they must start at least on SF9;
- Implement time randomised uplink messaging;
- Implement ADR for stationary devices.
Gateways are required to:
- Check for configuration (and preferably firmware) updates periodically, and be able to update their frequency plan and cluster address remotely;
- Detect connection loss;
- Reconnect automatically after losing connection, with backoff and jitter to prevent 'thundering herds'
- Re-resolve DNS when reconnecting.
6. Security and Backups
- TTI creates daily backups of critical (meta)data, including payloads, security keys, and data about applications, users, gateways, devices and device states. Backups are stored for a minimum of 7 days and used for disaster recovery;
- TTI implements rollback mechanisms during updates to prevent data loss;
- TTI strives for full compliance of its Cloud Services with the main versions of the LoRaWAN Specifications, LoRaWAN Backend Interfaces, and LoRaWAN Regional Parameters, as maintained and issued by the LoRa Alliance;
- TTI enables encrypted communication on endpoints;
- TTI implements industry-standard rate limiting to its APIs to ensure availability and security of its Cloud Services. TTI is not accountable for any Unavailability resulting from Customer actions exceeding such rate limits;
- TTI continuously monitors and mitigates potential failure scenarios, both in ops and security. TTI maintains a responsible disclosure channel (firstname.lastname@example.org or https://thethingsindustries.atlassian.net/servicedesk/customer/portal/16) for users to report vulnerabilities.
7. Additional Helpdesk Support
- TTI avails documentation to all Customers for self-service debugging and troubleshooting purposes via http://www.thethingsindustries.com/docs/docs
- Additionally, Customer may subscribe to a Helpdesk Support plan for "1st line support". This is a paid, value-add service, which aims to guide Customer in its LoRaWAN development and DevOps;
- Helpdesk Support Plan service levels are guaranteed as per Table 3. Support tickets can be submitted via email@example.com or via the portal at https://thethingsindustries.atlassian.net/servicedesk/customer/portals.
- For short turnaround times
|Help Desk plan:
||1. No support
||8h on all tickets
||4h on all tickets
|Tickets per month
|Who can open cases
||2 listed contacts
||6 listed contacts
||e-mail / portal
Table 3: Helpdesk Support Plan service levels
8. Performance Reporting and Service Credits
- TTI provides transparency to Customer about the achieved Services Uptime Metric. Such information is published on the Status Page at http://status.thethings.industries;
- TTI's failing to meet the Services Uptime Metric in a month is remedied by applying a percentage discount to the fees due for that month. This discount will be deducted from the next invoice (" Service Credits");
- The Service Credits are measured as follows: [[99,9% - Availability] * 10]. Example: in case the Cloud Services had an Availability of 98% (14,4hrs), the discount would be 1,9% * 5 = 19%;
- The remedy can never exceed 100%.