IFX alerts triggered during "publishing" articles on prod

[praneetharao0994]

While publishing articles in production, we perform the following additional steps:

We call the Biome API and send the article content as the payload to extract certain outputs.

The extracted output is then added to the article’s additionalProperties, and we use the Draft API’s PUT method to update the payload.

After that, we trigger the Draft API’s publish action.

We are receiving IFX alerts during this process.
Is there any known reason why this might happen?

Below is the error that we are getting

You are receiving this email because your Amazon CloudWatch Alarm "[redacted]" in the US East (N. Virginia) region has entered the ALARM state, because "Threshold Crossed: 1 datapoint [5436.086666666667 (09/12/25 16:56:00)] was greater than the threshold (5000.0)." at "Tuesday 09 December, 2025 17:06:15 UTC".
View this alarm in the AWS Management Console:
[redacted]
Alarm Details:

• Name: [redacted]
• Description: Integration Lambda 5 Second Avg Duration
• State Change: OK -> ALARM
• Reason for State Change: Threshold Crossed: 1 datapoint [5436.086666666667 (09/12/25 16:56:00)] was greater than the threshold (5000.0).
• Timestamp: Tuesday 09 December, 2025 17:06:15 UTC
• AWS Account: [redacted]
• Alarm Arn: arn:aws:cloudwatch:us-east-1:[redacted]:alarm:ifx_norstella_default_prod_duration_average_alarm
  Threshold:
• The alarm is in the ALARM state when the metric is GreaterThanThreshold 5000.0 for at least 1 of the last 1 period(s) of 600 seconds.
  Monitored Metric:
• MetricNamespace: AWS/Lambda
• MetricName: Duration
• Dimensions: [FunctionName = ifx_norstella_default_prod] [Resource = ifx_norstella_default_prod:live]
• Period: 600 seconds
• Statistic: Average
• Unit: not specified

[khaos119]

The message translates as: For the integration, during the 10-minute window ending at 16:56 UTC, the average execution time per invocation was about 5.4 seconds, which is above your configured limit of 5 seconds. Because even a single 10-minute period above that limit is enough to trigger the alarm, CloudWatch flipped the alarm into ALARM state.

It does not necessarily mean timeouts — just that average duration is creeping above the 5s ceiling.

In this case I would ignore it, but also make sure you are avoiding duplicate invocations. When you post to Draft API, make sure to include this header so it won’t re-trigger the integration. https://dev.arcxp.com/integrations/ifx-faqs-troubleshooting-best-practices/#my-integration-is-infinitely-looping-how-can-i-stop-it