Dear all,

I am running a daily pipeline to download specific variables from reanalysis-era5-single-levels using the CDS API, to retrieve the data from the date 7 days in the past. The today pipeline that corresponds to requests for the day 2023-09-11 failed for some of my variables (like "2m temperature) because of missing data (timestamps 2023-09-11 22:00 and 23:00). I checked it also manually on https://cds.climate.copernicus.eu/cdsapp#!/dataset/reanalysis-era5-single-levels?tab=form and indeed these values are missing even after ~7 days. While as I understand it should be available after ~5 days. My pipeline was running smoothly until today.

Do you know why these data are not available? Am I the only one experimenting this?

Thanks,

Philippe

6 Comments

  1. I am seeing the same thing as Philippe. I am equally interested to know what the increased time lag from present is due to. Thanks to anyone who can share information on this. 

  2. Hi, 

    please try again to request the data.

    Thanks

  3. Thank you for the reply, Michela!

    The data is back up to date now. I take it that this was a temporary technical issue. Are there any reasons to expect it to happen again these days (is there something specific being changed in the system), or was this, a priori, a one-time issue?

    Thank you again.

  4. Thank you Michela Giusti, the missing data have been added! As Alina Gainusa-Bogdan, I'm also interested to know if it was a one-time issue or not? To eventually delay my daily pipeline. 

    Thank you,

    Philippe


  5. In fact, we saw that there are still missing timestamps for some of the variables like u10 or v100, but surprisingly only when you request these variables alone. If you request a bunch of variables together, there are no missing values anymore, even for u10 and u100. Very strange. Could it be due to some problems with the caching on ECMWF side? 

  6. Hi,

    the issue was temporary.

    Please have a look at the following link for details about cache issue: https://confluence.ecmwf.int/x/v0ZtD

    Thanks