แชร์ผ่าน

Digital Platform API - API usage constraints

  • บทความ

This article provides information about the constraints faced whilst using the APIs.

Input and output throttling

To ensure good performance, services on the platform implement rate limiting at both the user and service levels. These limits, set by the service, may change over time.

When users encounter rate-limited requests, the API will respond with one of two response codes, along with relevant headers, to indicate the reason and how to proceed:

  1. The HTTP 429 (Too Many Requests) indicates the user has exceeded the maximum user calls to this service.

    1. This always indicates a rate-limited response, but for specific details, refer to the headers and the response. Some applications may impose additional limitations that trigger a 429 error, but they will provide additional information regarding the location and reason for the limit.

    2. General platform rate-limited requests will include the x-ratelimit-code header.

  2. The HTTP 503 (Service Unavailable) occurs when the service is overwhelmed by requests and is limiting new ones.

    1. If the 503 is from rate limiting, a x-ratelimit-code header will be present.

If you are experiencing more rate-limited-calls than anticipated, please contact your Xandr account representative.

Error messages

If you exceed the throttling limit, the API will respond with the HTTP 429 or HTTP 503 response code.

If you're using a script, you should check for the 429 response code or a 503 code that also includes the x-ratelimit-code header. If you receive this code, sleep your script for the number returned in the Retry-After field of the response header. This field tells you how long before the throttle limit is reset and you can continue sending API requests.

429 error headers

Requests limited at the user level will return a 429 with the following rate limit headers:

  1. x-ratelimit-code: Corresponds to the http code returned when the call was rate limited.
  2. retry-after: This shows the time in seconds to wait before retrying the request.
  3. x-ratelimit-count: This shows the total count in calls the user has made within the limit period. Users can refer to this number to adjust their call patterns when seeing rate-limited requests.
  4. x-an-user-id: The user ID that was limited.

Example 429 rate limited request headers:

< HTTP/1.1 429 Too Many Requests 
< Server: nginx/1.11.10 
< Date: Fri, 02 Feb 2024 15:58:18 GMT 
< Content-Type: application/json; charset=utf-8 
< Content-Length: 358 
< Connection: keep-alive 
< Retry-After: 9 
< x-b3-traceid: 9f53184f6e2c3cb9 
< x-ratelimit-code: 429 
< x-an-user-id: 1234 
< x-ratelimit-count: 1000 
< retry-after: 24

503 error headers

Requests limited at the service level will return a 503 with the following rate limit headers:

  1. x-ratelimit-code: Corresponds to the http code returned when the call was rate limited.
  2. retry-after: This shows the time in seconds to wait before retrying the request.

Example 503 rate limited request headers:

< HTTP/1.1 429 Too Many Requests 
< Server: nginx/1.11.10 
< Date: Fri, 02 Feb 2024 15:58:18 GMT 
< Content-Type: application/json; charset=utf-8 
< Content-Length: 358 
< Connection: keep-alive 
< Retry-After: 9 
< x-b3-traceid: 9f53184f6e2c3cb9 
< x-ratelimit-code: 503 
< retry-after: 24

Deprecated headers

The following headers in responses are deprecated and will be removed in the future. They do not provide any relevant information anymore. Please adjust any scripts using them to utilize the new headers as described above.

  • x-count-read
  • x-count-write
  • x-rate-limits
  • x-ratelimit-read
  • x-ratelimit-write
  • x-ratelimit-system

Debug parameter

In addition to the response code and response header, every response from the API will contain a dbg_info parameter. This parameter contains information about the API call and response. Below is an example of this debug output, which includes any warnings received from the API, the API version used for the call that generated this response, and the service used.

{
  "response": {
    "status": "OK",
    ...
    "dbg_info": {
      "warnings": [],
      "version": "1.18.349",
      "output_term": "user"
    }
  }
  }

Pagination

The maximum number of objects that can be returned in a given GET response is 100. To retrieve more than 100 objects, you can paginate results by specifying start_element and num_elements in the query string of the GET request. For example, the following request would return the first 50 objects in the response:

$ curl -b cookies -c cookies 'https://api.appnexus.com/creative?start_element=0&num_elements=50'

To retrieve the next 50, you would simply set start_element=50.

  • The first element is element 0.
  • All GET responses will have a "count" property showing the total number of elements matching that GET request.
  • This will also apply to non-reporting services, such as the creative search service, that are requested with methods other than GETs.

Example

$ curl -b cookies -c cookies 'https://api.appnexus.com/segment?start_element=0&num_elements=100'
$ curl -b cookies -c cookies 'https://api.appnexus.com/user?start_element=0&num_elements=100'

Authentication frequency

After authenticating, your token remains valid for 2 hours. You do not need to re-authenticate within this time. If you do re-authenticate, please note the following limitation: The API permits you to authenticate successfully 10 times per 5-minute period. Any subsequent authentication attempts within those 5 minutes will result in an error.

Tip

It is best practice to listen for the "NOAUTH" error_id in your call responses and re-authenticate only after receiving it.

Object limits

Xandr limits the number of line items, campaigns, creatives, publishers, sites, and placements that you can have on the platform. In addition, Xandr limits the number of domains that can be used in a single domain list and the number of certain targets that can be used in a single profile. You can use the Object Limit Service to view your limits and proactively monitor your usage.

Tip

For all object types except creatives, both active and inactive objects are counted against the limit. For creatives, only non-expired objects are counted against the limit. A creative expires when it has neither served nor been modified in 45 days.

For most clients, the default object limits are as follows:

Object Limit
Creatives per member

Note: Only non-expired creatives are counted against this limit.		 10,000
Campaigns per member 10,000
Line items per member 3,000
Placements per member 20,000
Sites per member 10,000
Publishers per member 3,000
Domains per domain list 30,000
Segments targeted per profile 400
Segment groups targeted per profile 400
Content categories targeted per profile 300
Platform content categories targeted per profile 300
Postal codes targeted per profile 4000
Inventory sources targeted per profile Deprecated
Publishers targeted per profile 300
Placement Groups targeted per profile 100
Placements targeted per profile 250
Deals targeted per member (note: only deals with profiles are counted against this limit) 1000
Profiles targeted per member 100

FAQs

How will I know that I am approaching my limit for an object?

We send you an email notification when you reach 85% and 95% of your limit for an object and another email when you reach 100% of your limit.

Who receives object limit email notifications?

Object limit notification emails are sent to the email addresses specified in the sherlock_notify_email field of the Member Service. You can change the recipients at any time. Note, however, that this field controls the recipients for creative auditing emails as well.

What if I reach my limit for an object?

When you approach or reach your limit for campaigns, line items, placements, sites, or publishers, you should delete any inactive, unused, or unnecessary instances so you can stay under your limit. Deleted line items, campaigns, creatives, publishers, sites, and placements will continue to appear in reporting but cannot be undeleted.

When you approach or reach your limit for creatives, you should remove non-expired creatives. Non-expired creatives have the is_expired field set to false. Note that removing expired creatives will not impact your creative count.

What if I am already over my limit?

If you need to create additional objects but have already met or exceeded your limit as listed above, please delete unused objects or contact support for assistance. Our support team can help you identify inactive objects to delete.

Can my limit be raised?

In exceptional cases, a limit may be temporarily lifted by a small amount at the discretion of our engineering team. Please contact your Xandr representative to discuss this option.