Workaround strange resolution bug in uv with opentelemetry downgrades
#43768
Conversation
This file contains 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
potiuk
requested review from
ephraimbuddy,
ferruzzi,
gopidesupavan,
dirrao,
eladkal and
amoghrajesh
and removed request for
ashb
ferruzzi
approved these changes
Nov 7, 2024
gopidesupavan
approved these changes
Nov 7, 2024
Still - it fails when trying to generate constraints using Which is even stranger!. I will have to workaround it a bit more - and add a beam limitation as well it seems. I think that one will be a head-scratcher for the |
|
From: https://github.com/apache/airflow/actions/runs/11713890095/job/32627785553?pr=43768#step:12:545 |
|
We have |
What was the exact command that ran into that? I ran |
Apparently `uv 0.4.30` has some strange resolution bug that results in inconsistent dependencies in opentelemetry that are detected by `pip check` when we are running eager upgrade. This is described in detail in astral-sh/uv#8871 . Until the issue is diagnosed and Solved, we workaround it by providing a hint to `uv` to use higher version of opentelemetry-proto than it wants to downgrade to - for no apparent reason.
potiuk
force-pushed
the
workaround-uv-bug-with-conflicting-resolution
branch
from
a18fd3d to
3afd9d2
Compare
eladkal
approved these changes
Nov 7, 2024
I'm surprised you don't see this more often: astral-sh/uv#3078 Btw, I have some big optimizations coming to pip and I see the same issue when pip is much more focused on backtracking causes. I strongly recommend Airflow puts a reasonable minimal bounds on it's providers. Or at least for apache-beam provider. |
We do |
And yes this is what we actively do for quite some time - including very recent #42989 ... so ... I am just surprised it happened, but I am solving this isssue in another way (as discussed in astral-sh/uv#8871 - by removing caching from the picture). So that one with beam == 2.0.0 might be an issue for another day :D. |
|
Closing that one. The discussion with @notatallshaw and @charliermarsh on astral-sh/uv#8871 - made me realise this is not an issue with |
|
Closing in favour of #43770 |
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.
Apparently
uv 0.4.30has some strange resolution bug that results in inconsistent dependencies in opentelemetry that are detected bypip checkwhen we are running eager upgrade. This is described in detail in astral-sh/uv#8871 . Until the issue is diagnosed and Solved, we workaround it by providing a hint touvto use higher version of opentelemetry-proto than it wants to downgrade to - for no apparent reason.^ Add meaningful description above
Read the 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.rstor{issue_number}.significant.rst, in newsfragments.