Outbound campaign POST failing at 300 concurrent dials

General Discussion
1

running a load test on POST /api/v2/outbound/campaigns and the throughput drops to zero once concurrent call volumes hit around 300. environment is us-east-1 with org capacity set to 5000. jmeter 5.6.2 setup is pretty standard: 150 threads, 60s ramp, keepalive on, hitting the campaign endpoint with a 100ms pause. the api throughput stays flat at 15 req/sec before it starts throwing 429 errors. response body just says rate limit exceeded for dial capacity bucket. checked the old community thread about websocket connection limits, but this isn’t a websocket issue, it’s strictly the rest api dialing queue. tried bumping the thread count to 200 and adding a 200ms think time, same result. the dialer seems to lock up once the internal capacity calculation hits a threshold, even though the org license allows way more. logs show the requests timing out at the edge router.

Thread Group: 150 threads
Ramp-Up: 60s
Loop: Infinite
Throughput: 15 req/sec -> 0 req/sec
Error: 429 rate limit exceeded for dial capacity bucket
2

PureCloudPlatformClientV2 indicates the 429 limit triggers because campaign creation shares the bucket with dialer state updates. First, I’ve increased the JMeter pause to 250ms, which only reached 18 req/sec. Second, I routed requests through a REST Proxy action, but the queue backed up immediately. The outbound module doesn’t handle flat bursts. You can switch to the async workflow endpoint to bypass the lock.

curl -X POST "https://api.mypurecloud.com/api/v2/outbound/campaigns/workflows" \
 -H "Authorization: Bearer ${ACCESS_TOKEN}" \
 -d '{"name":"test_campaign","dialMode":"power"}'

The outbound/campaigns route locks the state until a dialer node acknowledges it. You’ll see the spike exactly when the pool hits 300. I need to know if your script polls GET /api/v2/outbound/campaigns/{id} in a tight loop, or if the throttling comes purely from the POST. Check the Retry-After header.