Service Level Agreement (SLA)

2 — Definitions

Service Month

One full calendar month, from 00:00 WIB on the 1st to 23:59 WIB on the last day of that month.

Downtime

A period in which a Covered Service is unreachable or not functioning as it should because of a failure in infrastructure or systems under Levia Cloud’s control, as defined per service in Section 3.

Downtime Period

The accumulated whole minutes of Downtime in one Service Month. Disruptions lasting under 1 minute, or intermittent below one minute, are not counted.

Monthly Uptime

The availability percentage of a Covered Service in one Service Month, calculated with the formula in Section 4.

High Availability (HA)

A service configuration using at least 2 redundant nodes in different availability zones with automatic failover, explicitly subscribed to by the Customer.

Availability Zone

An infrastructure location isolated in power, cooling, and network within a Levia Cloud region (JKT-1, JKT-2, SBY-1).

Incident

An unplanned disruption or degradation of a Covered Service.

Response Time

The period from an Incident being reported by the Customer or detected by Levia Cloud monitoring, until a Levia Cloud engineer gives a first substantive response and begins work. Response Time is not resolution time.

Recovery Time (RTO)

The target period until the Service returns to normal operation or an acceptable workaround is available.

RPO

Recovery Point Objective — the maximum tolerable data loss, measured backwards from the moment of the incident.

Scheduled Maintenance

Planned maintenance notified to the Customer under Section 7, not counted as Downtime.

Service Credit

Compensation in the form of service credit applied against the next billing period, as the sole and exclusive remedy for a missed SLA target.

Affected Monthly Fee

The monthly subscription fee (excluding VAT) for the specific resource that experienced Downtime — not the Customer’s total account bill.

Status Page

The official Levia Cloud service status page, carrying incident and maintenance records.

3 — Availability commitments

Levia Cloud commits to delivering Covered Services at the Monthly Uptime targets in the tables below. Targets are measured per resource (per instance, per cluster, per volume), not aggregated at account level.

3.1 Compute

3.2 Storage

3.3 Managed Database

3.4 Networking

3.5 Platform, security, and support

4 — How uptime is calculated

4.1 Formula

Total Minutes in the Service Month is based on the actual number of calendar days in that month — 43,200 minutes for a 30-day month, 44,640 minutes for a 31-day month.

4.2 Calculation rules

1. (1)Downtime starts at the earlier of (a) detection of the disruption by Levia Cloud monitoring, or (b) receipt of the Customer’s report through the official ticket channel.
2. (2)Downtime ends when the service returns to normal operation or an agreed workaround is available, as recorded in the ticket system and on the Status Page.
3. (3)Downtime is counted in whole minutes. Intermittent disruptions lasting under 1 minute are not counted.
4. (4)Calculation is per resource. Where several resources are affected by the same incident, each is calculated separately against its own monthly fee.
5. (5)Scheduled Maintenance notified under Section 7 and all circumstances in Section 8 are excluded from the Downtime Period.
6. (6)Levia Cloud monitoring records and logs are the primary reference for the calculation. The Customer may submit contrary evidence, which will be reviewed jointly in good faith within 10 business days.

4.3 Uptime target to maximum downtime

The table below is a practical reference, assuming a 30-day month (43,200 minutes).

5 — Technical support and incident handling

5.1 Channels and service hours

5.2 Priority matrix and response times

Priority is set by Levia Cloud on impact and urgency, and may be adjusted jointly with the Customer.

Response Times for P1 and P2 are measured 24 x 7. Response Times for P3 and P4 are measured during business days, Monday–Friday 08:00–18:00 WIB, excluding national holidays. Recovery Targets are best-effort commitments backed by the escalation procedure in Section 5.4, and give rise to no separate Service Credit beyond the availability calculation in Section 3.

5.3 Root cause analysis (RCA)

1. (1)For every P1 and P2 Incident, Levia Cloud delivers a written RCA report within 5 business days of the incident being closed.
2. (2)The RCA report covers the timeline, impact, root cause, recovery actions, and preventive actions with target completion dates.
3. (3)For incidents affecting multiple customers, an RCA summary is published on the Status Page.

5.4 Escalation matrix

6 — Customer responsibilities

This SLA is effective only where the Customer meets the following. Failure to do so may cause the related Incident to be excluded from the SLA calculation.

1. (1)Provide and maintain technical and administrative Person in Charge (PIC) contact details reachable 24 hours a day for P1 incidents.
2. (2)Report Incidents through the official channels in Section 5.1, completely and promptly, with reasonable diagnostic information.
3. (3)Design its architecture to its own availability needs, including subscribing to an HA configuration where it targets 99.99%.
4. (4)Manage security at the application and data layers, including credentials, user access control, and security group configuration.
5. (5)Ensure invoices are paid on time and the account remains active and not suspended.
6. (6)Comply with the Acceptable Use Policy (AUP), international bandwidth fair-use terms, and platform quotas and technical limits.
7. (7)Respond within a reasonable time to requests for action or approval from Levia Cloud that are needed to restore service.
8. (8)Notify Levia Cloud at least 5 business days in advance of any planned load testing or penetration testing.

7 — Maintenance

7.1 Scheduled maintenance

Standard window

Tuesdays and Thursdays, 00:00–06:00 WIB.

Notice

At least 7 calendar days in advance via email to the registered PIC, console notification, and the Status Page. For maintenance likely to disrupt service, notice is given at least 14 calendar days in advance.

Duration cap

A maximum of 4 accumulated hours per Service Month per resource.

SLA treatment

Not counted as Downtime so long as it is carried out within the window, duration cap, and notice above. Duration beyond that cap is counted as Downtime.

Live migration

Most host maintenance is performed by live migration with no disruption to the Customer.

7.2 Emergency maintenance

Emergency maintenance may be carried out at any time to address a critical security vulnerability, a hardware failure risk, or an instruction from a competent authority. Notice is given as soon as possible — where feasible before execution, and no later than when execution begins. Emergency maintenance is not counted as Downtime so long as it does not exceed 2 hours per event and 2 events per Service Month.

Security patches for critical vulnerabilities are applied within 72 hours of the vendor’s official release.

8 — SLA exclusions

The following periods are not counted as Downtime and give rise to no Service Credit:

Force majeure

Natural disaster, earthquake, flood, fire, epidemic, war, civil unrest, strikes, and failures of national infrastructure (power, national backbone network) beyond Levia Cloud’s reasonable control.

Customer action

Misconfiguration, deletion of resources, credential errors, changes made outside change management procedures, or actions by anyone using the Customer’s credentials.

Customer equipment & software

Failure of operating systems, applications, self-managed databases, scripts, or hardware on the Customer’s side; including workloads exceeding the capacity of the resources subscribed to.

Third-party networks

Disruption at the Customer’s ISP, the public internet, or third-party services the Customer depends on, beyond the Levia Cloud network boundary.

Maintenance

Scheduled and Emergency Maintenance meeting the conditions in Section 7.

Lawful suspension

Suspension or termination arising from payment arrears, AUP breach, or an order of a competent authority.

Quotas & platform limits

Requests refused because an account quota, API rate limit, or published technical limit has been reached.

Non-production services

Beta, Preview, Trial, free tier, and any service provided free of charge.

Out-of-scope cyber attacks

Layer 7 DDoS attacks against accounts not subscribed to DDoS Protection Advanced, and security compromises originating from the Customer’s side.

Uncoordinated testing

The effects of load testing or penetration testing carried out without notice under Section 6.

Customer request

Downtime arising from actions the Customer expressly requested or approved, including restart, resize, migration, and restore.

9 — Compensation: Service Credit

9.1 Compensation tiers

Where the Monthly Uptime of a resource falls below the applicable target in Section 3, the Customer is entitled to Service Credit under the table below, calculated on the Affected Monthly Fee (excluding VAT).

9.2 Limits

1. (1)Total Service Credit in one Service Month does not exceed 100% of the Affected Monthly Fee for the service concerned. Uptime below 95.00% entitles the Customer to a full credit of that service’s monthly fee.
2. (2)Service Credit is calculated on the fee for the affected resource, not on the Customer’s total account bill.
3. (3)Service Credit is issued as service credit against the next billing period and cannot be redeemed for cash, transferred, or exchanged for any other compensation.
4. (4)Service Credit does not apply retroactively beyond 1 Service Month and lapses if the service is terminated before the credit is used, unless otherwise agreed in writing.
5. (5)Service Credit is the Customer’s sole and exclusive remedy for a missed SLA target. Levia Cloud is not liable for indirect loss, lost profits, lost revenue, or other consequential loss.
6. (6)A Customer with payment arrears past the due date may not file a claim until those arrears are settled.

9.3 Claim procedure

9.4 Termination right on repeated failure

Where a Covered Service fails to meet its Monthly Uptime target for 3 consecutive months, or where Monthly Uptime falls below 95.00% in any single month, the Customer may terminate that service without termination penalty on 30 calendar days’ written notice. The Customer remains entitled to Service Credit already accrued and to a pro-rata refund of unused prepaid fees.

10 — Transparency and reporting

11 — Dispute resolution

1. (1)Any dispute arising from this SLA is first addressed by good-faith consultation between the Levia Cloud Service Delivery Manager and the Customer’s PIC, within 30 calendar days.
2. (2)If consultation does not reach agreement, the parties agree to resolve the dispute through the Indonesian National Board of Arbitration (BANI) in Jakarta under BANI’s applicable rules and procedures.
3. (3)This SLA is governed by and construed under the laws of the Republic of Indonesia.

12 — Security, compliance, and data location

Data location

All Customer data is stored and processed in data centres within the Republic of Indonesia — the Jakarta (JKT-1, JKT-2) and Surabaya (SBY-1) regions.

Personal data protection

Personal data is processed in accordance with Law No. 27 of 2022 on Personal Data Protection and its implementing regulations. Levia Cloud acts as Personal Data Processor for data the Customer controls.

Electronic system operation

Levia Cloud operates its electronic systems in accordance with Government Regulation No. 71 of 2019 on the Operation of Electronic Systems and Transactions.

Certification

ISO/IEC 27001 (Information Security Management) and ISO 9001 (Quality Management). The infrastructure is prepared to PCI DSS requirements.

Security incident notice

Security incidents affecting personal data are notified to the Customer and to the competent authority within 3 x 24 hours of becoming aware.

Data ownership

Customer data belongs entirely to the Customer. On termination, data is retained for 30 calendar days before permanent deletion, and the Customer may download or request handover during that period.

Confidentiality

Levia Cloud does not access Customer content except where required to restore service at the Customer’s request, or where required by law.

13 — General terms and contact

13.1 Changes to this SLA

Levia Cloud may review and update this SLA over time. Changes that materially reduce commitments to the Customer are notified at least 30 calendar days before they take effect, by email to the registered PIC and on the Status Page. For customers with a running contract, the SLA in force at signature continues to apply until that contract period ends, unless otherwise agreed in writing.

13.2 Severability

If any provision of this SLA is held invalid or unenforceable under applicable law, the remaining provisions stay in full force.

13.3 Language

This SLA is made in the Indonesian language. Where a translated version is published, the Indonesian version prevails in the event of a difference in interpretation, in accordance with Law No. 24 of 2009.

13.4 Contact

Appendix A — SLA target summary

A one-page summary for quick reference, tenders, and proposal attachments.

Appendix B — Service Credit claim form

Include the following on the claim ticket, or send it to [email protected] no later than 14 calendar days after the end of the affected month.

Part A — Customer identity

• Company / account name
• Levia Cloud account number
• PIC name & position
• PIC email & phone

Part B — Claim detail

• Affected Service Month
• Service & resource ID
• Downtime start date & time (WIB)
• Downtime end date & time (WIB)
• Total Downtime in minutes
• Calculated Monthly Uptime (%)
• Related incident ticket number
• Monthly fee for the affected service (IDR)
• Service Credit claimed (%)

Part C — Supporting evidence

• System logs
• Monitoring screenshots
• External uptime monitor output
• Any other relevant detail