OCI Compute SLA
Oracle Cloud Infrastructure · Compute · promises up to 99.99% · SLA of Sep 1, 2026 · checked 25h ago
Incidents, last 365 days
The OCI status page has no public incident history feed.
Commitments and credits
TierUptimeCredit of the monthly bill
- Multi-AD region, instances across more than one Availability DomainRegions with more than one Availability Domain99.99%below 99.99%: 10% · below 99%: 25% · below 95%: 100%
- Single-AD region, instances across more than one Fault DomainRegions with one Availability Domain99.95%below 99.95%: 10% · below 99%: 25% · below 95%: 100%
- Single instanceEach single compute instance99.9%below 99.9%: 100%
More in compute
ServicePromisedFirst creditObserved 365dDowntime
AWS LambdaAWS · Compute · Standard99.95%First credit10%Observed 365d99.178%All incidents 99.852%
Cloudflare WorkersCloudflare · Compute · Enterprise99.99%First credit10%Observed 365d100%All incidents 99.984%
Confluent Cloud for Apache FlinkConfluent · Compute · Flink (per region)99.99%First credit5%Observed 365d99.628%All incidents 97.169%
DigitalOcean Kubernetes (DOKS) HA control planeDigitalOcean · Compute · HA control plane99.95%First credit10%Observed 365d100%All incidents 99.992%
Compute EngineGoogle Cloud · Compute · Instances in Multiple Zones (Premium Tier)99.99%First credit10%Observed 365d99.927%All incidents 99.488%
Terms, exclusions and sources
- Measured
- Calendar month; percentage of minutes the service was Unavailable (no external connectivity)
- Credit cap
- No more Service Credits than the fees paid for the quantity of the Non-Compliant Service actually used in the Measured Period
- Claim window
- Within sixty (60) calendar days from when the issue occurred
- How to claim
- Contact your Oracle account manager with service name, description, time/duration of downtime, Region, OCIDs and supporting logs
- Not covered
- Customer or third-party equipment, software or technology (including third-party platforms); Inter-related unavailability caused by another (Primary) Cloud Service; Actions or inactions of the customer, its users or third parties (misconfiguration, exceeding limits, restarting/patching, etc.); Anything excluded from Unplanned Downtime in the Oracle Cloud Hosting and Delivery Policies
- Status history
- The OCI status page has no public incident history feed.
Monthly Uptime Percentage for Regions with more than one Availability Domains Less than 99.99% but equal to or greater than 99.0% 10% ... Less than 95.0% 100%
The Availability SLA's commitment sentence says 'at least 99.9%', while the credit tables start at 99.99% (multi-AD) and 99.95% (single-AD); tiers here follow the credit tables. Compute also carries Manageability (control-plane API, 99.9%), NVMe performance (99.9%) and network performance (99.9%) SLAs, each with 10/25/100% credits, depending on the SKU.
Sources: each vendor's own SLA page (re-read monthly) and status page feed (read every Friday). A summary, not the contract: the contract you sign governs. Logos via logo.dev; trademarks belong to their owners.