I fixed it by changing the error to a warning and setting a different checksum so that those variables get cleaned on next ETL run. It means that the dataset will be rerun twice, but it's the easiest solution and this should happen rarely.
chart-diff: ✅
No charts for review.
data-diff: ✅ No differences found
```diff
Legend: +New ~Modified -Removed =Identical Details
Hint: Run this locally with etl diff REMOTE data/ --include yourdataset --verbose --snippet
```
Automatically updated datasets matching _weekly_wildfires|excess_mortality|covid|fluid|flunet|country_profile|garden/ihme_gbd/2019/gbd_risk_ are not included
Fix for https://github.com/owid/etl/issues/2871, see the description there.
I fixed it by changing the error to a warning and setting a different checksum so that those variables get cleaned on next ETL run. It means that the dataset will be rerun twice, but it's the easiest solution and this should happen rarely.