CDSAPI - status remains on accepted - cams-global-atmospheric-composition-forecasts

Datasets and Usage CAMS - Datasets and Usage
1

We have a script for downloading forecasts which worked for years, also in a reasonable time (less than a few minutes).

I now tried to do the upgrade to the new server, but cannot pass the ‘accepted’ state.
I now use a new key and use the new URL , and the basic example works fine.

However, using the following parameters:

url = "https://ads.atmosphere.copernicus.eu/api"

dataset     = "cams-global-atmospheric-composition-forecasts"

'date' 2024-04-23/2024-04-23
'type' forecast
'format' grib
'variable'  ["2m_temperature","mean_sea_level_pressure","total_aerosol_optical_depth_550nm","total_column_ozone","total_column_water_vapour","ammonium_aerosol_optical_depth_550nm","dust_aerosol_optical_depth_550nm","organic_matter_aerosol_optical_depth_550nm","sea_salt_a
erosol_optical_depth_550nm","black_carbon_aerosol_optical_depth_550nm","nitrate_aerosol_optical_depth_550nm","sulphate_aerosol_optical_depth_550nm","total_aerosol_optical_depth_670nm","total_aerosol_optical_depth_1240nm","total_aerosol_optical_depth_865nm"]
'time'   "00:00"
'leadtime_hour': ["0","3","6","9","12","15","18","21"]

Gives this logging output but then just gets stuck:

2024-10-23 14:16:40,473 INFO legacy_api_client-log 180: [2024-09-26T00:00:00] **Welcome to the New Atmosphere Data Store (ADS)!** This new system is in its early days of full operations and still undergoing enhancements and fine tuning. Some disruptions are to be expecte
d. Your
[feedback](https://jira.ecmwf.int/plugins/servlet/desk/portal/1/create/202) is key to improve the user experience on the new ADS for the benefit of everyone. Thank you.
2024-10-23 14:16:40,473 INFO [2024-09-26T00:00:00] **Welcome to the New Atmosphere Data Store (ADS)!** This new system is in its early days of full operations and still undergoing enhancements and fine tuning. Some disruptions are to be expected. Your
[feedback](https://jira.ecmwf.int/plugins/servlet/desk/portal/1/create/202) is key to improve the user experience on the new ADS for the benefit of everyone. Thank you.
2024-10-23 14:16:40,474 INFO legacy_api_client-log 180: [2024-09-26T00:00:00] Watch our [Forum]( https://forum.ecmwf.int/) for Announcements, news and other discussed topics.
2024-10-23 14:16:40,474 INFO [2024-09-26T00:00:00] Watch our [Forum]( https://forum.ecmwf.int/) for Announcements, news and other discussed topics.
2024-10-23 14:16:40,474 INFO [2024-08-08T00:00:00] Should you have not yet migrated from the old ADS system to the new ADS, please check our [informative page](https://confluence.ecmwf.int/x/uINmFw) for guidance.
2024-10-23 14:16:40,474 INFO legacy_api_client-log 180: [2024-08-08T00:00:00] Should you have not yet migrated from the old ADS system to the new ADS, please check our [informative page](https://confluence.ecmwf.int/x/uINmFw) for guidance.
2024-10-23 14:19:27,509 INFO Request ID is ae897683-a0ea-4a76-85f7-4f730511fc9e
2024-10-23 14:19:27,509 INFO legacy_api_client-log 180: Request ID is ae897683-a0ea-4a76-85f7-4f730511fc9e
2024-10-23 14:19:27,575 INFO legacy_api_client-log 180: status has been updated to accepted
2024-10-23 14:19:27,575 INFO status has been updated to accepted

Any ideas?

Thanks!

Dominique

2

I am experiencing the same issue!

3

Might be related:

4

The waiting queues are now fixed.

5

I submitted a request last night and it wasn’t complete until morning.
I think the queues are down again.

6

I confirm it’s again stuck, I’ll open a ticket.

7

Working again. Got this reply:

"Since its launch this past September, the new Data Stores are continuously undergoing enhancements and fine tuning. Some adjustments to the rules applied to the queue are being explored to improve users experience on the new system. This means that you may see some variability in queueing times until things settle down. Both our technical team and management are aware of the current situation and efforts have been put into place to improve the performance. Unfortunately there is no immediate solution to that, but we are on it.

In addition to this, there are system sessions and/or disruptions on occasions during which service performance is unfortunately degraded. This may result in your requests staying much longer in the queue than expected or some of your requests to fail with no obvious reason. There were some disruption earlier this week which resulted in a surge of queued requests and it is taking longer than expected to absorb the queue resulting in even longer queueing times.

We share information on a best effort basis when such events occur and encourage users to check the status of our Data Stores on the ECMWF service status page , look out for warnings banners on the CDS homepage and or relevant dataset overview page and also watch our forum for announcements."