The campaign just dropped to 1.0x pacing the moment the 50k contact list import hit the API limit. Trying to push the multiplier back up to 2.5x in the admin UI just spins. Console shows a 503 Service Unavailable on /api/v2/outbound/campaigns/{id}/predictive/multiplier. US-East tenant, running the 2025-02-01 hotfix. The abandonment rate spiked to 3.8% while the system was stuck in this state, basically acting like a progressive dialer. Lost about 400 calls in the first 15 minutes. The error payload mentions a timeout on the availability window calculation.
Tried clearing the browser cache and switching to incognito, same result. The API keeps timing out on the agent availability check. Pacing is dead in the water. The outbound team is sitting on hold with agents because the dialer won’t ramp up. Compliance window is set for 08:00-18:00 EST, but the system seems to be recalculating the callable hours every time a multiplier update is attempted. This is the third time this week the UI has choked on a simple pacing adjustment. The abandonment threshold is configured at 3.0%, so the system should be auto-correcting, but it’s just sitting there.
The answer rate estimation model seems to be locked up too. Can’t even view the historical performance metrics without hitting the same timeout. The contact list has a 45% connect rate based on yesterday’s run, so the multiplier should be stable. It’s not. The admin UI is completely unresponsive to any changes on the predictive settings tab. Had to pause the campaign to stop the bleed. Pausing works fine, but restarting throws the same 503 immediately. It’s like the backend service for predictive routing is down for this tenant. Checking the status page shows everything green, which is a lie. This is killing the daily production targets.
DNC sync completed successfully an hour ago, so that’s not the blocker here. The issue is strictly isolated to the predictive multiplier endpoint. Even attempting to toggle the ‘Enable Predictive Dialing’ switch off and on triggers the 503. The system won’t let me revert to progressive mode either. It’s stuck in a zombie state where it’s not predicting but also not dialing progressively. Agents are seeing zero disposition codes coming in. This is a hard stop for the shift. No workaround found yet.
Error response below. No idea why the availability calculation is timing out on a standard US-East tenant.
{
"message": "Service temporarily unavailable. Availability window calculation timed out after 30000ms.",
"status": 503,
"errorCode": "OUTBOUND_PREDICTIVE_MULTIPLIER_UPDATE_FAILED",
"moreInfo": "https://developer.genesys.cloud/...",
"reason": "Service Unavailable",
"errors": []
}