Question
Are there any Limitations when using any of EPM's RestAPI features?
Answer
1. Time Out when logging in
Logging in process may fail due to momentarily high-server load. Should this be the case, the status code received will be 408.
What can be done?
Perform login once more. Load will eventually pass and the login process will succeed, should all proper parameters are correct.
2. Token and Parallel requests
The same Token cannot be used for more then 1 parallel request. e.g: 2 Runners, 2 Shells should not use the same token to execute requests. This is relevant even in a situation in which a login is performed once, but the Token is then used in more then 1 request in parallel - this is deemed to fail - which will result in abnormal behaviors, such as: Internal Server errors (status code 500), wrong data will be received in the response, etc.
What can be done?
Use different tokens to execute multiple requests.
3. Timeout of different type of requests
Should a timeout be reached when one attempts to retrieve inbox events for example (Aggregated, Raw), the server will return status code 500. The cause will be an extremely large number of unhandled events in both Privilege Management Inbox and in Application Control Inbox.
What can be done?
Use Filters to narrow down the amount of events to be retrieved from the DB in the response, to a point in which the Timeout won't be reached.
4. Login amount in SAAS vs. OnPrem/AD.
In SAAS environment, user can perform login once per minute (60 seconds). In On-Premise deployments, this is not the case.
5. Maximum API Calls per API call time period
The maximum allowed number of API calls per defined time period = 3 calls
The period of time to limit the API calls to the maximum number allowed = 5 minutes
(Note - these 2 parameters are configurable for on-premise customers under Set > Server Configuration > API limitations)