Severe and persistent queue delays for ERA5 hourly data (pressure levels) via CDS API

Datasets and Usage C3S - Datasets and Usage
1

Dear ECMWF/CDS team,

I would like to report a persistent issue with extremely long queue times when accessing the dataset:

“ERA5 hourly data on pressure levels from 1940 to present”

Over the past ~2 weeks, almost all of my requests submitted via CDS API have been stuck in a very long queue.

Example status:

  • Date: 2026-04-01
  • Time: 14:45 UTC (approx.)
  • Status: Accepted
  • Queue status:
    Running: 80
    Queued: 5832

In many cases, requests remain queued for several hours (sometimes much longer), making it very difficult to perform routine data processing workflows.

Additional details:

  • Access method: cdsapi (Python)
  • Requests are already optimized to reduce load (e.g., grouped by day instead of hour)
  • The issue appears consistently regardless of request size or time of submission

I would like to ask:

  1. Is this level of queue congestion expected recently?
  2. Are there any ongoing system issues or capacity limitations?
  3. Are there recommended best practices to avoid such long delays (beyond current documentation)?
  4. Is there any priority difference between CDS web interface requests and API requests?

This issue is significantly affecting productivity and automated workflows.

Thank you for your support.

Best regards

2

I am the same. I have encountered a problem where I need a long period of data, but whether I divide my requirements by month or by different data types, it takes a long time for this request to go from “accepted” to “successful”, with the shortest being 4 hours and the longest being 15 hours. I also want to know why this is happening recently