Newer Version Available

This content describes an older version of this product. View Latest

Connect REST API Rate Limits

Connect REST API requests are subject to rate limiting. Connect REST API has a different rate limit than other Salesforce APIs. Connect REST API has a per user, per application, per hour rate limit. When you exceed the rate limit, all Connect REST API resources return a 503 Service Unavailable error code.

Load, performance, and other system issues can prevent some limits from being reached. Limits can change without notice. Applications should make efficient use of available requests and gracefully handle the 503 error code.

Note

For applications using a session ID from Salesforce, the rate limit is per user, per hour—there isn't a separate bucket for applications. All applications the user accesses with a session ID use this general quota. To take advantage of the per user, per application, per hour limit, use OAuth tokens.

To build apps that don’t hit the rate limit, use these guidelines:
  • If you hit limits when running tests, use multiple users to simulate a real-world scenario.
  • When polling for feed updates, do not exceed one poll per minute (60 polls per hour). To return more results in 1 request, increase the page size.
  • When polling for private messages, do not exceed 60 polls per hour.
  • To avoid making multiple requests, cache metered static assets such as file and dashboard renditions (group and user profile pictures are not metered).
  • Each developer on a team should set up 2 connected apps: one for automated testing and one for manual testing and development. Don’t share connected apps with other developers.
  • Use a unique connected app for the production environment.
  • Don’t share connected apps between applications.

To extract data, use REST API or SOAP API to operate on sObjects. Do not use Connect REST API to extract data.

Important

REST API’s rate limiting and query language work better for these use cases:
  • Migrate or synchronize user profiles between systems.
  • Create analytic applications.
  • Write queries that return records and fields.