Hi everyone,
I came across an interesting issue and wanted to check if anyone has seen something similar.
I am using the following versions:
elementary-data==0.20.1 cli and python packages
dbt-bigquery==1.10.2
dbt-fusion==2.0.0-preview.173
dbt-core==1.11.0b3
Issue
When running dbt test locally on macOS, the execute_completed_at column is generated with this timestamp format:
2026-04-27T14:05:11.371056Z
However, when running the same test, with the same package versions, inside Docker (Linux), the precision changes to:
2026-04-27T14:34:33.609083964Z
The second value includes nanosecond precision, and BigQuery rejects it as an invalid timestamp, which later causes an error in the edr report.
Question
Does anyone know why the timestamp precision differs between macOS and Linux even with identical package versions?
It seems the environment/OS may be affecting how the timestamp is generated, but I am not sure which component is responsible.
Any ideas would be appreciated.
Hi everyone,
I came across an interesting issue and wanted to check if anyone has seen something similar.
I am using the following versions:
Issue
When running
dbt testlocally on macOS, theexecute_completed_atcolumn is generated with this timestamp format:However, when running the same test, with the same package versions, inside Docker (Linux), the precision changes to:
The second value includes nanosecond precision, and BigQuery rejects it as an invalid timestamp, which later causes an error in the
edr report.Question
Does anyone know why the timestamp precision differs between macOS and Linux even with identical package versions?
It seems the environment/OS may be affecting how the timestamp is generated, but I am not sure which component is responsible.
Any ideas would be appreciated.