CDNTR Multi-CDN service level agreement
This agreement applies to CDNTR Multi-CDN, which works with Amazon CloudFront and is bought in AWS Marketplace. The product listing references it.
What is covered
The regional network: the CDNTR delivery endpoints that serve your hostnames to viewers in the regions you route to us (Türkiye; or Türkiye, Europe, the Middle East and Africa). It is measured per calendar month (UTC), per AWS Marketplace agreement.
Not covered: Amazon CloudFront, Amazon Route 53 and the rest of your AWS account (AWS's own SLAs cover them), your origin, and the viewers' networks.
Monthly availability
Target: 99.9%.
A minute counts as unavailable when the regional network did not serve your routed viewers because of us:
- the health checks of the stack (Amazon Route 53 in your account, our health location
/.well-known/cdntr/health) were failing for every edge of a delivery group, so Route 53 answered those viewers with CloudFront; or - we shed your account's traffic to CloudFront for our capacity (the capacity threshold of the delivery group), not because of your own package, caps or a suspension.
Because Route 53 sends those viewers back to CloudFront automatically, an unavailable minute usually means "served by CloudFront and billed to your CloudFront account", not an outage of your site. The credit compensates for that.
Availability = (minutes in the month − unavailable minutes) / minutes in the month.
Service credits
| Monthly availability | Credit, % of that month's package fee |
|---|---|
| below 99.9% | 10% |
| below 99.0% | 25% |
| below 95.0% | 50% |
- Credits apply to the package fee of the affected month (the 12-month price counts as 1/12 per month), not to per-GB overage or to AWS charges.
- At most one credit tier per month; credits do not carry into cash beyond the month's fee.
- The credit is paid as an AWS Marketplace billing adjustment (refund) on the agreement.
Exclusions
- Scheduled maintenance announced at least 72 hours ahead by email, at most 4 hours per month, during which traffic falls back to CloudFront as designed.
- Causes outside our network: AWS services (CloudFront, Route 53, your health checks or zone), your origin or distribution configuration, DNS or CAA changes by you, a deleted or modified stack, your AWS WAF blocking our fetches.
- Traffic above the account's configured capacity or package when you chose that limit; a suspension for non-payment, a lapsed or cancelled subscription, or a breach of the AWS Acceptable Use Policy or our terms.
- Force majeure, and attacks that our network mitigates by shedding to CloudFront for less than 15 minutes at a time.
How to claim
Email info@cdn.com.tr within 30 days after the end of the month. Include the AWS account ID, the AWS Marketplace agreement ID, the hostnames, and the times. We answer within 10 business days with our measurement (the health check history of our endpoints and our shed log).
Support
| Severity | Example | First response | Hours |
|---|---|---|---|
| 1, service down | regional network not serving any routed viewer | 1 hour | 24x7 |
| 2, degraded | errors or slowness for part of the traffic, purge not reaching edges | 4 hours | 24x7 |
| 3, question | configuration, billing, reports | 1 business day | business days, Europe/Istanbul |
Channels: email info@cdn.com.tr, panel tickets after the subscription is connected. English and Turkish.
Security incidents: security@cdn.com.tr (also in /.well-known/security.txt), acknowledged within 24 hours.
Measurement
Our record per delivery group: the cdn.com.tr panel's health checks of the delivery endpoints and the shed log (every shed and release with its reason). Route 53 health check history in your account is your own evidence.