Batch job fails mid-run when access token expires

NICE CXone API & Integrations
1

Running a bulk update script for queue configs, but the access token expires halfway through the 500-item loop. The script crashes on the 120th item with a 401 Unauthorized instead of auto-refreshing. Here’s the retry logic we’re using:

if response.status_code == 401:
 token = get_new_token()
 headers['Authorization'] = f'Bearer {token}'
 return requests.put(url, headers=headers, json=data)

The 401 doesn’t trigger the retry block. It just dies. Why isn’t the 401 being caught?

2

Cause: The 401 triggers after the request sends. Docs state: “The client credentials flow requires token refresh before expiry.” Your retry logic is too late.

Solution: Pre-check expiry or use a wrapper. Here’s a quick Python fix using requests session hooks to auto-refresh:

def refresh_token(session):
 # Logic to fetch new token
 pass

session.hooks['response'].append(refresh_token)

Check the token’s exp claim before the loop starts.

3

Pre-checking expiry works but it’s messy. Just use the platformClient SDK wrapper, it handles token refresh automatically in the background. Saves the headache of managing headers manually.

1 Like
4

Don’t patch requests. Use the SDK.

import PureCloudPlatformClientV2 as PureCloud

api_instance = PureCloud.QueueApi(PureCloud.ApiClient(configuration))

Token refresh is automatic here.

result = api_instance.post_queue(body=payload)


The suggestion above is right. Writing manual refresh hooks is error-prone. The `PureCloudPlatformClientV2` client handles token rotation internally. Just use the SDK for batch jobs. It saves you from reinventing the wheel.
1 Like