Comments (8)
This is a bug/incompatibility in the new FAB provider 1.2.0 after implementing #40145 - seems that Airflow 2.9.2 (and before) is not compatible with the new POST-only requirement of FAB provider.
This is a problem only if you do not use constraints (that's what highly recommended in https://airflow.apache.org/docs/apache-airflow/stable/installation/installing-from-pypi.html ) - if you do not want to use constraints, you should downgrade/pin apache-airflow-providers-fab to 1.1.1 and it should fix your problem. That's a quick workaround for now until we fix it.
CC: @shahar1 : we need to find a good solution for back-compatibility - and I am not sure if we can do it for earlier Airflow versions - because Airlfow 2.9.2 and below will call the "/logout" method via GET and without CSRF. I think what we should do is to add a back-compatibilty code in logout - and allow "GET" method if airflow version <= 2.9.2
from airflow.
Glad it worked, but I will re-open it - this one is really a bug in FAB provider that we need to fix :). You SHOULD be able to use 1.2.0 providers with earlier versions of Airlfow - so likely 1.2.1 version of FAB provider will have a compatibility code to handle it.
from airflow.
CC: @shahar1 : we need to find a good solution for back-compatibility - and I am not sure if we can do it for earlier Airflow versions - because Airlfow 2.9.2 and below will call the "/logout" method via GET and without CSRF. I think what we should do is to add a back-compatibilty code in logout - and allow "GET" method if airflow version <= 2.9.2
since fab provider has min version of Airflow 2.9.0 I think the simplest solution is to bump the min version to 2.9.2
I think that is reasonable enough
from airflow.
since fab provider has min version of Airflow 2.9.0 I think the simplest solution is to bump the min version to 2.9.2
I think that is reasonable enough
There is a good reason why we should not limit it. The idea behind FAB provider is that each version is linked to a specific FAB version - because we partially vendored in security manager - and this allowed people to upgrade to newer version of fab (and security fixes it brings) without bringing Airlfow version up. So yeah - in this case it's just 2.9.1 and 2.9.2 that are affected, so once we release 2.9.3 everyone should upgrade anyway, but in general cases I think it's good to keep min Airflow version as low as we can, to allow everyone to upgrade FAB independently.
So I treat that also as a "learning" exercice on how we can do it - this way in the future we might remember to look at those compatibilities and fix them in similar way.
from airflow.
Thanks for opening your first issue here! Be sure to follow the issue template! If you are willing to raise PR to address this issue please do so, no need to wait for approval.
from airflow.
@potiuk, I'm aware about the the use of constraints, but I have to comply with other constraints too. I followed your recommendation & downgraded fab provider using pip install apache-airflow-providers-fab==1.1.1
and it worked !! thank you ;)
from airflow.
CC: @shahar1 : we need to find a good solution for back-compatibility - and I am not sure if we can do it for earlier Airflow versions - because Airlfow 2.9.2 and below will call the "/logout" method via GET and without CSRF. I think what we should do is to add a back-compatibilty code in logout - and allow "GET" method if airflow version <= 2.9.2
since fab provider has min version of Airflow 2.9.0 I think the simplest solution is to bump the min version to 2.9.2 I think that is reasonable enough
I just thought about it while implementing the solution 😅
It is ready, though, so it's up to you how to proceed.
from airflow.
Since you already implemented the solution and it's seems simple enough then we can just merge it :)
from airflow.
Related Issues (20)
- Allow setting log_driver when using DockerSwarmOperator HOT 1
- Task with ALL_DONE trigger rule gets skipped HOT 4
- control state of individual taskflow in mapped task-group HOT 2
- Check possibility to rewrite the homepage view to have lazy load with parallel database queries HOT 2
- on_failure_callback not being triggered on -9 error HOT 1
- helm chart: duplicate key in safe-to-evict annotation for services like scheduler HOT 2
- Status of testing Providers that were prepared on July 02, 2024 HOT 3
- BigQueryCheckOperator does not support parameterized queries HOT 3
- CloudBatchSubmitJobOperator does not forward the project_id in the hook
- Support `@task.spark_submit` or `@task.pyspark_submit`
- Enable Even More PyDocStyle Checks HOT 6
- DockerSwarmOperator does not work well with Python's tqdm when displaying logs HOT 5
- Airflow Helm Chart: DB migration fails HOT 2
- Azure Container Instance operator priority option HOT 1
- Airflow API for Updating TaskInstance State to skipped is broken HOT 2
- Airflow SFTPToS3Operator file exist check HOT 3
- switch HttpHook from using requests library to httpx? HOT 1
- Poll Based SLA check HOT 2
- Failing KPO task wich task decorator and imported typing elements HOT 3
- [databricks] Refactor how Databricks workflows repair / repair all is implemented HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from airflow.