Clay Icon
Join the Community
02 Support

Configurable Delay for Scheduled LinkedIn API Job to Prevent Timeouts

Avatar of Will Y.Will Y.
·
·

I’m using the HTTP API to pull daily LinkedIn‑campaign statistics into our analytics table. Although the response is only about 30 rows, the scheduled request fails on roughly 1/3 every day with a timeout (error 429). I then have to run the error cells it manually, which isn't practical for this application. Could you please advise whether either the following is possible? Add a configurable delay/throttle to the scheduled HTTP job

3 comments

  • Avatar of Channeled
    Channeled
    APP
    ·
    ·

    Yes, it's possible to add a rate limit to your scheduled HTTP API job to help prevent timeout errors (429). In the HTTP API integration settings, you can define a rate limit by specifying the maximum number of requests allowed within a given time frame. To set this up: 1. Go to the HTTP API integration settings 2. Find the "Define rate limit" section 3. In "Request Limit", enter the maximum number of allowed requests 4. In "Duration (in ms)", specify the time window in milliseconds For example, you could set 1 request per 1000 ms (1 second) to slow down the API calls. If you continue to experience issues after adjusting these settings, you may need to request human assistance for further troubleshooting.

  • Avatar of Will Y.
    Will Y.
    ·
    ·

    So what number should I put to run less than 10 every 2 seconds?

  • Avatar of Channeled
    Channeled
    APP
    ·
    ·

    To ensure fewer than 10 requests are made every 2 seconds, you can set the rate limit in the HTTP API integration as follows: 1. Request Limit: 9 2. Duration (in ms): 2000 This configuration allows 9 requests every 2 seconds (2000 ms), which is just under 10 requests per 2 seconds. Remember that the rate limit must average at least 5 requests per second to work properly, so this setting should be fine. If you continue to experience issues after applying these settings, you may need to adjust them further or request human assistance for more specific troubleshooting.