Location

If you are a reseller, you can use this header to specify the ID of a child customer and make API call on their behalf.

On-Behalf-Of

A string automatically generated by the api server to identify each request.

Qtl-Request-Id

An optional header that resellers can use to specify whether to include report data from children customers.

Report-Range

applicationID

ID of the customer that a serviceQuota is created for

customerID

RFC 3339 date indicating the end of a time period. Example: <code>endTime=2020-11-14T00:00:00Z</code> The time must be specified using the UTC timezone. Your endTime may be rounded up to the nearest minute, hour, or day depending on the "interval" parameter. For example, if you pass <code>endTime=2020-09-05T03:14:01Z&interval=hourly</code>, the response includes data ending at 2020-09-05T04:00:00Z.

endTime

The granularity of returned data. In the backend, raw data is reported every minute. The data is then aggregated to give reports of different granularities. The aggregation method is “sum” for those traffic metrics, and "average" for all other metrics. Note that "oneminute" interval is available only when the timespan is not longer than 7 days and that "fiveminutes" interval is available only when the timespan is not longer than 31 days.

interval

limit

Maximum number of resources to return. To retrieve a complete list of resources, please pass this parameter with value “0”.

Indicates the maximum nuber of resources to return.

Usage metric on compute, storage and network resources.

Metric | Description | Measured in
--- | --- | --- 
inboundTraffic | Amount of data transmitted to ECP during a time period | byte
inboundBandwidth | Rate of data transfer to ECP during a time period | bps
outboundTraffic | Amount of data transmitted from ECP during a time period | byte
outboundBandwidth | Rate of data transfer from ECP during a time period | bps
totalTraffic | Amount of data transmitted to and from ECP during a time period | byte
totalBandwidth | Rate of data transfer to and from ECP during a time period | bps
vcpu | Amount of CPU cores allocated for your application instances during a time period | core
mem | Amount of memory allocated for your application instances during a time peirod | byte
local-ssd | Amount of local SSD storage capacity allocated for your application instances during a time period | byte
persist-ssd | Amount of remote SSD storage capacity allocated for your application instances during a time period | byte
podCount | Number of your application instances (i.e. Kubernetes pods) existing during a time period | ea
ipCount | Number of public IPv4 addresses allocated for your applications during a time period. This includes addresses allocated for both application instances and ECP LoadBalancer4s, if any. | ea


metric

Indicates the first resource to return. The first has an offset of 0.

offset

Indicates the first resource to return. The first has an offset of 0. The value must be a valid non-negative integral multiple of the limit parameter.

Usage metric on compute, storage and network resources.

Metric | Description | Measured in
--- | --- | ---
inboundTraffic | Amount of data transmitted to an application instance during a time period | byte
inboundBandwidth | Rate of data transfer to an application instance during a time period | bps
outboundTraffic | Amount of data transmitted from an application instance during a time period | byte
outboundBandwidth | Rate of data transfer from an application instance during a time period | bps
cpuUsage | Amount of CPU cores utilized by processes running in an application instance during a time period | core
memUsage | Amount of memory utilized by processes running in an application instance during a time period | byte
diskUsage | Disk utilization rate. Here "disk" could mean the container rootFS and any data volume attached to an application instance. | --


Keyword to search for particular resources. The value is matched against the "name" field of each resource.

searchString

servergroup

servicezone

RFC 3339 date indicating the beginning of a time period. Example: <code>startTime=2020-10-30T00:00:00Z</code>. The time must be specified using the UTC timezone. Your startTime may be rounded down to the nearest minute, hour, or day depending on the "interval" parameter. For example, if you pass <code>startTime=2020-09-05T03:14:01Z&interval=hourly</code>, the response includes data beginning at 2020-09-05T03:00:00Z.

startTime

The deployment status of an ECP application

status

This error is returned if the API call is not permitted. Refer to response body for details.

Specifications of an ECP application of the latest version

Sum of the amounts of “limits.cpu” allocated to instances of an ECP application. Measured in cores.

Array of history operations against an ECP application

API account or user who calls for the operation

The application specifications submitted along with the operation request

Date-time indicating when the deployment as triggered by this operation ends

Date-time indicating when the deployment as triggered by this operation starts

A flag indicating if the above version is currently the newest

Message related to the status of deployment

Status of the deployment as triggered by this operation

Date-time indicating when the operation is performed

A number indicating the version of an ECP application. This is auto generated by ECP API server.

Count of public IPv4 addresses allocated to an ECP application. Can be the addresses of ECP layer4 load balancers or pods, depending on how the application instances are exposed.

A flag indicating if the instances of an ECP application are exposed via ECP layer 4 load balancers.

Sum of local-ssd storage capacities allocated to instances of an ECP application. Measured in bytes.

Sum of the amounts of “limits.mem” allocated to instances of an ECP application. Measured in bytes.

Message related to the deployment status of an ECP application

Name of an ECP application. Must contain only lower-case alphanumeric characters and ‘-', must start and end with a lower-case alphanumeric character, and must be no more than 48 characters.

Sum of persist-ssd storage capacities allocated to instances of an ECP application. Measured in bytes.

An object showing the desired and current number of pods (i.e. application instances) for an ECP application

Count of pods that are created and currently existing  across all ECP locations. Pods in different phase (i.e. pending, running or failed) are all counted.

Count of pods desired across all ECP locations

Count of pods that are currently existing and in running phase across all ECP locations

Array of ECP regions where the application is deployed

Array of locations within an ECP region where an application is deployed

Code of an ECP region, e.g. “NA-WEST”

Kubernetes Workload

ECP LoadBalancer4

ECP Target

Use "RegionPreference", an ECP custom resource, to specify ECP regions and have your application instances automatically distributed across locations within each ECP region. 

Array of target ECP regions to distribute your application instances

RegionPreference

Use Kubernetes "Deployment" as workload controller if your application does not require persistent storage. Refer to <a href="https://kubernetes.io/docs/concepts/workloads/controllers/deployment/">Kubernetes official documentation </a> for more about the controller.

This document only focuses on some typical configs of  a Deployment object. For more about how to specify a Deployment, please refer to the <a href="https://v1-17.docs.kubernetes.io/docs/reference/generated/kubernetes-api/v1.17/#deployment-v1-apps">official Kubernetes API reference</a>.

A label query over pods. Must match the pod template's labels.

Template of the pods that will be created by the Deployment

Deployment

Use "LoadBalancer4", an ECP custom resource, to expose your application instances

Array of rules for the LoadBalancer4 to forward incoming traffic to backend application instances

Use "Location", an ECP custom resource, to pick ECP locations and precisely place your application instances.

Use Kubernetes "ReplicaSet" as workload controller if your application does not require persistent storage. Refer to <a href="https://kubernetes.io/docs/concepts/workloads/controllers/replicaset/">Kubernetes official documentation </a> for more about the controller.

This document only focuses on some typical configs of  a ReplicaSet object. For more about how to specify a ReplicaSet, please refer to the <a href="https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.17/#replicaset-v1-apps">official Kubernetes API reference</a>

Template of the pods that will be created by the ReplicaSet

ReplicaSet

Use Kubernetes "StatefulSet" as workload controller if your application requires persistent storage. Refer to <a href="https://kubernetes.io/docs/concepts/workloads/controllers/statefulset/">Kubernetes official documentation </a> for more about the controller.

This document only focuses on some typical configs of  a StatefulSet object. For more about how to specify a StatefulSet, please refer to the <a href="https://v1-17.docs.kubernetes.io/docs/reference/generated/kubernetes-api/v1.17/#statefulset-v1-apps"> official kubernetes API reference</a>.

Template of the pods that will be created by the StatefulSet

A list of persistent volume claims that pods are allowed to reference. Every claim in this list must have at least one matching (by name) volumeMount in one container in the pod template.

StatefulSet

Code of an ECP location where the VIP and/or VIP-v6 are exposing application instances

Public IPv4 address(es) exposing application instances at a specific ECP location. Multiple addresses are separated by comma.

Public IPv6 address(es) exposing application instances at a specific ECP location. Multiple addresses are separated by comma.

A flag indicating if the VIP and/or VIP-v6 are for an ECP LoadBalancer4

Number of application instances behind the VIP and/or VIP-v6 at an ECP location

The architecture that the artifact applies to

Datetime when the artifact is last pulled

Overview of the previous scan result. If the artifact is not scanned, this object will not be returned.

Array of conditions describing the current service state of a pod. For more information, refer to <a href="https://kubernetes.io/docs/reference/kubernetes-api/workload-resources/pod-v1/#PodStatus">official documentation</a>.

Timestamp when the Pod condition was last probed

Timestamp when the Pod last transitioned from one status to another

Message indicating details about the transition

A unique, one-word, CamelCase reason for the condition’s last transition

Container's ID in the format 'docker://<container_id>'

Details about the container's last termination condition

A flag indicating whether the container has passed its readiness probe

The number of times the container has been restarted

Details about the container's current condition

Array of Kubernetes events related to a pod. Note that an event will be removed in 3 hours after last occurence. For more information, refer to the <a href="https://kubernetes.io/docs/reference/kubernetes-api/cluster-resources/event-v1/">official documenation</a>.

The number of times this event has occurred

The time at which the event was first recorded

The time at which the most recent occurrence of this event was recorded

A human-readable description of the status of this operation

A short string that gives the reason for the transition into the object's current status

Kubernetes defined term indicating where a pod is in its lifecycle. For more information, refer to <a href="https://kubernetes.io/docs/reference/kubernetes-api/workload-resources/pod-v1/#PodStatus">official documentation<a/>

Short form of an ECP location where a pod is located

Readiness of containers in a pod, e.g. “2/2” means 2 containers in a pod are both ready.

Number of times containers in a pod have restarted

How a registry project is accessible to registry users

Whether to scan images automatically when images are pushed to registry. Default to false.

Name of apiAccount or contact who created the registry project

Name of a registry project. Must be unique globally. Allows lower case alphanumeric characters, ‘-’, ‘_’ and ‘.’ only, must start and end with an alphanumeric character, and must be not more than 128 characters.

Whether to prevent vulnerable images from running. Default to false.

Number of repositories under a registry project

If "preventVulnerable" is enabled and the vulnerability of an image is equal to or higher than severity defined here, the image can't be pulled.

Storage quota limiting the size of all images stored under the project. Measured in GB. 
The sum of storageLimits of all projects for a same customer must not exceed the limit specified in the “serviceQuotas.resourcequotas.storageInRegistry”.

project.storageLimit

Storage quota limiting the size of all images stored under the project.

Hard limit of storage quota. Measured in bytes.

Storage quota already consumed. Measured in bytes.

project.storageLimit(Status)

Email address of the account manager who manages the customer. The email address specified can be used by the account manager to receive system notifications related to the customer, e.g. in the event of trial expiration, excessive usage and service suspension.

quota.accountManagerEmail

Whether a customer is allowed to deploy their applications to ECP locations in mainland China. When this is set to false, the customer will not be allowed to pick mainland China locations or the "CN" region to deploy their applications.

quota.allowDeploymentInChina

A flag indicating whether this customer will be invoiced using Stripe.

quota.billUsingStripe

ID representing the ECP service contract, used for billing. Required when “billUsingStripe” set to true.

quota.contractID

The start date of an ECP service contract. Must be a UTC date in “yyyy-MM-dd” format. Required when “contractID” is provided.

quota.contractStartDate

ID of the customer that a serviceQuota is created for.

quota.customerID

quota.description

The Kubernetes namespace to be created for a customer in underlying ECP Kubernetes clusters (i.e. ECP locations). Allows only digits, lower-case letters and hyphens. Must be unique globally and no more than 63 characters.

quota.namespace

Global resource quotas limiting the maximum amounts of ECP resources available to a customer across all ECP locations

Maximum total number of applications that can be deployed to ECP

Maximum total number of vCPU cores available to a customer across all ECP locations. Measured in "core".

Maximum storage capacity of type "local solid state drive" available to a customer across all ECP locations. Measured in "GB". Default to 0, means no storage capacity is available.

Maximum total amount of memory available to a customer across all ECP locations. Measured in "GB".

Maximum storage capacity of type "persist solid state drive" available to a customer across all ECP locations.  Unlike local SSD storage, data kept in persist SSD storage is replicated across multiple servers for redundancy. Measured in "GB". Default to 0, means no storage capacity is available.

Maximum total number of Kubernetes pods that can be deployed to ECP Kubernetes clusters (i.e. ECP locations)

Maximum total number of public IPv4 addresses available to a customer across all ECP locations. Must be within the range 0-2,000. 0 means no public IPv4 address is available.

Maximum storage capacity available to a customer to store container images in ECP image registry. Measured in GB.

Maximum total number of docker image tags that can be pushed to ECP image registry

quota.resourceQuotas

Maximum total number of Kubernetes pods that can be deployed to ECP Kubernetes clusters (i.e ECP locations)

An object showing the balance amounts of each resource quota

Balance max. number of applications that can be deployed to ECP

Balance total number of vCPU cores available to a customer across all ECP locations. Measured in “millicores”.

Balance storage capacity of type “local solid state drive” available to a customer across all ECP locations. Measured in “bytes”.

Balance total amount of memory available to a customer across all ECP locations. Measured in “bytes”.

Balance storage capacity of type “persist solid state drive” available to a customer across all ECP locations. Measured in "bytes". Unlike local SSD storage, data kept in persist SSD storage is replicated across multiple servers for redundancy.

Balance max. number of Kubernetes pods that can be deployed to ECP Kubernetes clusters (i.e. ECP locations)

Balance total number of public IPv4 addresses available to a customer across all ECP locations

An object showing the amounts of consumed resource quotas

Total number of applications that have been deployed to ECP

Total number of vCPU cores allocated a customer across all ECP locations. Measured in “millicores”.

Total storage capacity of type “local solid state drive” allocated to a customer across all ECP locations. Measured in “bytes”.

Total amount of memory allocated to a customer across all ECP locations. Measured in “bytes”.

Total storage capacity of type “persist solid state drive” allocated to a customer across all ECP locations. Measured in "bytes". Unlike local SSD storage, data kept in persist SSD storage is replicated across multiple servers for redundancy.

Total number of Kubernetes pods that have been deployed to ECP Kubernetes clusters (i.e. ECP locations)

Total number of public IPv4 addresses allocated to a customer across all ECP locations

quota.resourceQuotasStatus

ID of a serviceQuota. Unless otherwise stated, the ID of the customer that a serviceQuota is created for is reused as the ID of the serviceQuota. 

quota.serviceQuotaID

Status of a serviceQuota, which represents the status of ECP service subscription by a customer. The “trial” status means the customer is using ECP service as a free trial. When the status is set to "suspended", any requests from the customer will be rejected by ECP API server. 

This status is related to ECP service only. It doesn’t affect other products subscribed by the customer.

quota.status

Datetime indicating when the status is changed from “trial” or “active” to “suspended‘. This field is updated by API server when ECP service is suspended for the customer. It cannot be set via API calls.

quota.suspensionDate

Date-time in RFC 3339 format indicating when the trial expires, e.g. 2021-07-01T15:00:00Z. Required only when the status is set to “trial”. Ignored otherwise. Must not be a date-time in the past. The trial must not last longer than 90 days after the trial starts.

When trialDeadline expires, the status will be set to “suspended” and all existing ECP resources for the customer will be deleted.

quota.trialDeadline

Datetime indicating when the status is changed from “trial” to “active”. This field is updated by API server when the customer as a free trial upgrades to a paid customer. It cannot be set via API calls.

quota.upgradeDate

Number of artifacts pushed to a repository

repo.artifactsCount

Arbitrary text describing an image repository

The number of image pulls from an image repository

The number of tags in an image repository

Array of time serialized metric values. This schema is used when the metric param is of the value 'diskUsage'.

Disk utilization report

Array of time serialized metric values. This schema is used when the metric param is of a value other than 'diskUsage'.

CPU/memory/traffic report

Summary of resource usage grouped per ECP server groups

Summary of resource usage grouped per service zones

Name of contact or apiAccount who creates the registry user

Arbitrary text about the registry user. Must be not more than 20 characters.

How a project is accessible to a registry user. Can be one of:
“readonly”: has the permissions to pull images under a project;
“readwrite”: has the permissions to push/pull images to/from a project.

All requests to API server must be authenticated using an API login name and key assigned to your organization. For security, the API key is not passed directly but is secured as described below. 

The authentication type is HTTP Basic, and the authorization string is specified in the HTTP request header called Authorization.
```
#!/bin/bash
#URL of the ECP API server
API_SERVER=https://ngapi.cdnetworks.com

#You will be given the following information by the sales team
USER="{API user name}"
API_KEY="{API key (make sure to escape the $ sign if any)}"

#Date in RFC 1123 format
DATE=`date "+%a, %d %b %Y %H:%M:%S %Z"`

#Generate the password. Raw output from openssl by adding the "-binary" flag
passw=$(echo -n "$DATE" | openssl dgst -sha1 -hmac "$API_KEY" -binary | base64)
echo " "

#Construct the curl command to query list of applications
apiCall ="curl -i --url '${API_SERVER}/ecp/v1/applications'
                  -X GET
                  -u $USER:$passw
                  -H 'Date: $DATE'
                  -H 'Accept: application/json'"
eval $apiCall
echo " "
```
 If you are a reseller, you can also specify a child customer's ID in an On-Behalf-Of HTTP header to make API calls on behalf of your child customer.

The ECP API is a RESTful API that allows you to interact with ECP platform to manage your ECP resources.

ECP API Specification

Get a list of ECP applications. You can specify a search parameter, status, offset and limit to filter the results.

The response contains a list of applications.

The request is invalid. Refere to response body for details.

This interface allows you to configure an ECP application and have the application automatically deployed to ECP edge locations.

Please include specifications of all objects comprising an ECP application and pass them as an array in YAML or JSON format in the request body. An ECP application must include a Kubernetes workload resource and an ECP Target object, and might include an ECP LoadBalancer4 object depending on how you choose to expose the application instances. You can also include in the request body Kubernetes objects like <a href="https://kubernetes.io/docs/concepts/configuration/configmap/">ConfigMap</a> and <a href="https://kubernetes.io/docs/concepts/configuration/secret/">Secret</a>. Refer to examples of request body for possible use cases.

When you specify an ECP Target, you can pick locations by yourself and have your application instances precisely placed at desired locations. You can also simply designate ECP regions and have your application instances automatically distributed across locations within each ECP region. Note that, in the latter case, there will be auto cross-location migration of application instances in the event of a location outage. Such migration happens among locations within the same ECP region.

Name of your ECP application. Must contain only lower-case alphanumeric characters and ‘-', must start and end with a lower-case alphanumeric character, and must be no more than 48 characters.

appName

This response indicates the application is successfully created.

The request is invalid. Refer to response body for details.