-
Notifications
You must be signed in to change notification settings - Fork 16.7k
Fix upgrade failure when xcom contains NaN in string values#57614
Merged
kaxil merged 1 commit intoapache:mainfrom Oct 31, 2025
Merged
Fix upgrade failure when xcom contains NaN in string values#57614kaxil merged 1 commit intoapache:mainfrom
kaxil merged 1 commit intoapache:mainfrom
Conversation
Contributor
closes: #57608
Problem
When migrating from Airflow 2.x to 3.0, the migration 0049_3_0_0_remove_pickled_data_from_xcom_table was failing.
This occurred when XCOM data contained the string "NaN" as part of a value, for example:
{"craft": "Tiangong", "name": "Ye GuangfuNaN"}
The migration was using a simple REPLACE() function that replaced all occurrences of 'NaN' with '"nan"', which broke the JSON structure when NaN appeared inside string values. The broken JSON would become:
{"craft": "Tiangong", "name": "Ye Guangfu"nan""}
And things would break.
I changed the NaN sanitization to use regex with word boundaries to match only standalone NaN tokens (like "key": NaN) while preserving NaN that appears inside string values.
- PostgreSQL: Uses
regexp_replacewith word boundaries (\bNaN\b) (https://neon.com/postgresql/postgresql-string-functions/regexp_replace) - MySQL: Uses
REGEXP_REPLACEwith word boundaries ([[:<:]]NaN[[:>:]]) (https://dev.mysql.com/doc/refman/8.4/en/regexp.html) - SQLite: Kept simple with original
REPLACE()approach as sqlite doesn't have regex support(?)
Testing
- Installed 2.11 Airflow using:
breeze start-airflow --backend postgres --executor CeleryExecutor --use-airflow-version 2.11.0 --db-reset - Ran this DAG:
from airflow import Dataset
from airflow.decorators import dag, task
from pendulum import datetime
import requests
# Define the basic parameters of the DAG, like schedule and start_date
@dag(
start_date=datetime(2024, 1, 1),
schedule="@daily",
catchup=False,
doc_md=__doc__,
default_args={"owner": "Astro", "retries": 3},
tags=["example"],
)
def example_astronauts():
@task
def get_astronauts(**context) -> list[dict]:
number_of_people_in_space = 1
list_of_people_in_space = [
{"craft": "Tiangong", "name": "Ye GuangfuNaN"},
]
context["ti"].xcom_push(
key="number_of_people_in_space", value=number_of_people_in_space
)
return list_of_people_in_space
@task
def print_astronaut_craft(greeting: str, person_in_space: dict) -> None:
"""
This task creates a print statement with the name of an
Astronaut in space and the craft they are flying on from
the API request results of the previous task, along with a
greeting which is hard-coded in this example.
"""
craft = person_in_space["craft"]
name = person_in_space["name"]
print(f"{name} is currently in space flying on the {craft}! {greeting}")
print_astronaut_craft.partial(greeting="Hello! :)").expand(
person_in_space=get_astronauts() # Define dependencies using TaskFlow API syntax
)
# Instantiate the DAG
example_astronauts()
from airflow.decorators import dag, task
from pendulum import datetime
import requests
# Define the basic parameters of the DAG, like schedule and start_date
@dag(
start_date=datetime(2024, 1, 1),
schedule="@daily",
catchup=False,
doc_md=__doc__,
default_args={"owner": "Astro", "retries": 3},
tags=["example"],
)
def example_astronauts():
@task
def get_astronauts(**context) -> list[dict]:
number_of_people_in_space = 1
list_of_people_in_space = [
{"craft": "Tiangong", "name": "Ye GuangfuNaN"},
]
context["ti"].xcom_push(
key="number_of_people_in_space", value=number_of_people_in_space
)
return list_of_people_in_space
@task
def print_astronaut_craft(greeting: str, person_in_space: dict) -> None:
"""
This task creates a print statement with the name of an
Astronaut in space and the craft they are flying on from
the API request results of the previous task, along with a
greeting which is hard-coded in this example.
"""
craft = person_in_space["craft"]
name = person_in_space["name"]
print(f"{name} is currently in space flying on the {craft}! {greeting}")
print_astronaut_craft.partial(greeting="Hello! :)").expand(
person_in_space=get_astronauts() # Define dependencies using TaskFlow API syntax
)
# Instantiate the DAG
example_astronauts()
- Upgraded airflow to 3.11 using breeze:
breeze start-airflow --backend postgres --executor CeleryExecutor --use-airflow-version 3.1.1
No migration error:
Performing upgrade to the metadata database postgresql+psycopg2://postgres:***@postgres/airflow
2025-10-31T10:40:09.197045Z [info ] Context impl PostgresqlImpl. [alembic.runtime.migration] loc=migration.py:211
2025-10-31T10:40:09.197133Z [info ] Will assume transactional DDL. [alembic.runtime.migration] loc=migration.py:214
2025-10-31T10:40:09.197938Z [info ] Context impl PostgresqlImpl. [alembic.runtime.migration] loc=migration.py:211
2025-10-31T10:40:09.197991Z [info ] Will assume transactional DDL. [alembic.runtime.migration] loc=migration.py:214
2025-10-31T10:40:09.198247Z [info ] Creating Airflow database tables from the ORM [airflow.utils.db] loc=db.py:684
2025-10-31T10:40:09.198296Z [info ] Creating context [airflow.utils.db] loc=db.py:689
2025-10-31T10:40:09.200174Z [info ] Binding engine [airflow.utils.db] loc=db.py:691
2025-10-31T10:40:09.200245Z [info ] Pool status: Pool size: 5 Connections in pool: 0 Current Overflow: -3 Current Checked out connections: 2 [airflow.utils.db] loc=db.py:693
2025-10-31T10:40:09.200284Z [info ] Creating metadata [airflow.utils.db] loc=db.py:695
2025-10-31T10:40:09.424717Z [info ] Getting alembic config [airflow.utils.db] loc=db.py:698
2025-10-31T10:40:09.425147Z [info ] Stamping migration head [airflow.utils.db] loc=db.py:705
2025-10-31T10:40:09.426209Z [info ] Context impl PostgresqlImpl. [alembic.runtime.migration] loc=migration.py:211
2025-10-31T10:40:09.426272Z [info ] Will assume transactional DDL. [alembic.runtime.migration] loc=migration.py:214
2025-10-31T10:40:09.450286Z [info ] Running stamp_revision -> cc92b33c6709 [alembic.runtime.migration] loc=migration.py:622
2025-10-31T10:40:09.451776Z [info ] Airflow database tables created [airflow.utils.db] loc=db.py:708
^ Add meaningful description above
Read the [Pull Request Guidelines](https://github.com/apache/airflow/blob/main/contributing-docs/05_pull_requests.rst#pull
-request-guidelines) for more information.
In case of fundamental code changes, an Airflow Improvement Proposal (AIP) is needed.
In case of a new dependency, check compliance with the ASF 3rd Party License Policy.
In case of backwards incompatible changes please leave a note in a newsfragment file, named {pr_number}.significant.rst or {issue_number}.significant.rst, in airflow-core/newsfragments.
boring-cyborg
bot
added
area:db-migrations
PRs with DB migration
kind:documentation
labels
Oct 31, 2025
vatsrahul1001
approved these changes
Oct 31, 2025
dstandish
approved these changes
Oct 31, 2025
ephraimbuddy
pushed a commit
that referenced
this pull request
Nov 3, 2025
(cherry picked from commit 376ab5d)
ephraimbuddy
pushed a commit
that referenced
this pull request
Nov 4, 2025
(cherry picked from commit 376ab5d)
vatsrahul1001
added
the
backport-to-v3-1-test
Mark PR with this label to backport to v3-1-test branch
label
Nov 5, 2025
44 tasks
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.