ecodina opened a new pull request, #59252:
URL: https://github.com/apache/airflow/pull/59252
<!--
Licensed to the Apache Software Foundation (ASF) under one
or more contributor license agreements. See the NOTICE file
distributed with this work for additional information
regarding copyright ownership. The ASF licenses this file
to you under the Apache License, Version 2.0 (the
"License"); you may not use this file except in compliance
with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing,
software distributed under the License is distributed on an
"AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
KIND, either express or implied. See the License for the
specific language governing permissions and limitations
under the License.
-->
<!--
Thank you for contributing! Please make sure that your code changes
are covered with tests. And in case of new features or big changes
remember to adjust the documentation.
Feel free to ping committers for the review!
In case of an existing issue, reference it using one of the following:
closes: #ISSUE
related: #ISSUE
How to write a good git commit message:
http://chris.beams.io/posts/git-commit/
-->
The last few days I've been deploying Airflow 3 with the Keycloak auth
manager. Locally everything worked well, but when I deployed it to our Docker
Swarm cluster a delay of around 5 seconds started appearing to each request.
After some (rather intensive) debugging, I found out that the "culprit" was
the `request.post` done in the `_is_authorized` and `_is_batch_authorized`
methods.
I tried finding out whether it was due to resources, DNS, IPv6, requests
going through the wrong interface or our reverse proxy not handling the
requests well, but couldn't find anything. I tested, inside the keycloak auth
manager running in the apiserver, the different causes, but all seemed fine
(DNS was resolving in milliseconds, the request was going through the right
interface, ...). Increasing to 2 CPU and 1 GB RAM didn't help either.
Finally, what worked, was creating a pool of connections. Instead of making
requests individually, reusing them and keeping them alive helped. Now, the
webpage loads instantly (locally and in our production instance).
This PR ports to Airflow what has worked for us, and should probably help
others. Requests are now more scalable in high-concurrency scenarios and we
avoid multiple SSL handshakes. It also introduces automatic retries for
intermittent network issues.
I've also added 2 configuration parameters to allow deployment managers to
set it up to their needs: `requests_pool_size` and `requests_retries` with some
sensible defaults.
_@bugraoz93 and @vincbeck, hopefully this will be my last PR for the
Keycloak auth manager for a while... With this, it works well in our use-case.
Sorry for the last few days!_
<!-- Please keep an empty line above the dashes. -->
---
**^ 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](https://cwiki.apache.org/confluence/display/AIRFLOW/Airflow+Improvement+Proposals))
is needed.
In case of a new dependency, check compliance with the [ASF 3rd Party
License Policy](https://www.apache.org/legal/resolved.html#category-x).
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](https://github.com/apache/airflow/tree/main/airflow-core/newsfragments).
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
