Comments (7)
I'm getting the same error in my own forked repo. It's the same if you try to run the Dev Container locally. I believe that if you scroll up you'll see the actual cause of the problem. For me, it looks like it's a pip
dependency conflict:
#13 ERROR: process "/bin/sh -c pip install -r requirements.txt && pip install keyrings.alt dbus-python ipython ipykernel" did not complete successfully: exit code: 1
------
> [dev_container_auto_added_stage_label 4/6] RUN pip install -r requirements.txt && pip install keyrings.alt dbus-python ipython ipykernel:
53.13 promptflow-evals 0.3.0 depends on promptflow-devkit<2.0.0
53.13 promptflow-azure 1.11.0 depends on promptflow-devkit<2.0.0 and >=1.11.0
53.13 promptflow 1.10.0 depends on promptflow-devkit==1.10.0
53.13 promptflow-evals 0.3.0 depends on promptflow-devkit<2.0.0
53.13 promptflow-azure 1.10.1 depends on promptflow-devkit<2.0.0 and >=1.10.1
53.13
53.13 To fix this you could try to:
53.13 1. loosen the range of package versions you've specified
53.13 2. remove package versions to allow pip attempt to solve the dependency conflict
53.13
------
Dockerfile-with-features:21
--------------------
20 | COPY requirements.txt .
21 | >>> RUN pip install -r requirements.txt \
22 | >>> && pip install keyrings.alt dbus-python ipython ipykernel
23 |
--------------------
ERROR: failed to solve: process "/bin/sh -c pip install -r requirements.txt && pip install keyrings.alt dbus-python ipython ipykernel" did not complete successfully: exit code: 1
{"outcome":"error","message":"Command failed: docker buildx build --load --build-arg BUILDKIT_INLINE_CACHE=1 -f /tmp/devcontainercli-root/container-features/0.56.1-1716978672495/Dockerfile-with-features -t vsc-contoso-chat-830c6938fafd08eb2da0d2669867fe5e315e2dac15ac6285f83a50273840073a --target dev_containers_target_stage --build-context dev_containers_feature_content_source=/tmp/devcontainercli-root/container-features/0.56.1-1716978672495 --build-arg _DEV_CONTAINERS_BASE_IMAGE=dev_container_auto_added_stage_label --build-arg _DEV_CONTAINERS_IMAGE_USER=root --build-arg _DEV_CONTAINERS_FEATURE_CONTENT_SOURCE=dev_container_feature_content_temp /var/lib/docker/codespacemount/workspace/contoso-chat","description":"An error occurred building the image."}
It looks like the promptflow
dependency was pinned 2 weeks ago to version 1.10.0
, which requires the same version of promptflow-devkit
. There seems to be a conflict when installing deps for promptflow-evals
, not allowing us to fetch promptflow-azure==1.10.0
. @cassiebreviu, any insights there?
from contoso-chat.
Pasting here the entire pip install
trace:
from contoso-chat.
@raffertyuy explicitly pinning promptflow-azure==1.10.0
in requirements.txt unblocks the build for me, feel free to try it.
It might be worth investigating this further, however, to detect the root cause of the problem.
from contoso-chat.
Thanks @phoevos ! I tried in https://github.com/raffertyuy/fork-contoso-chat and it didn't work for me. Attached is the creation log.
I believe the culprit are these lines. It still doesn't explain why I can run it properly on the main repo but on on forked though.
2024-05-31 08:33:03.988Z: #13 67.20 ERROR: Cannot install -r requirements.txt (line 10), -r requirements.txt (line 5) and promptflow.evals because these package versions have conflicting dependencies.2024-05-31 08:33:03.995Z:
2024-05-31 08:33:04.139Z: #13 67.20
#13 67.20 The conflict is caused by:
#13 67.20 promptflow 1.10.0 depends on promptflow-devkit==1.10.0
#13 67.20 promptflow-evals 0.3.0 depends on promptflow-devkit<2.0.0
#13 67.20 promptflow-azure 1.11.0 depends on promptflow-devkit<2.0.0 and >=1.11.0
#13 67.20 promptflow 1.10.0 depends on promptflow-devkit==1.10.0
#13 67.20 promptflow-evals 0.3.0 depends on promptflow-devkit<2.0.0
#13 67.20 promptflow-azure 1.10.1 depends on promptflow-devkit<2.0.0 and >=1.10.1
#13 67.20
#13 67.20 To fix this you could try to:
#13 67.20 1. loosen the range of package versions you've specified
#13 67.20 2. remove package versions to allow pip attempt to solve the dependency conflict
#13 67.20
#13 67.20 ERROR: ResolutionImpossible: for help visit https://pip.pypa.io/en/latest/topics/dependency-resolution/#dealing-with-dependency-conflicts
2024-05-31 08:33:04.917Z: #13 ERROR: process "/bin/sh -c pip install -r requirements.txt && pip install keyrings.alt dbus-python ipython ipykernel" did not complete successfully: exit code: 1
2024-05-31 08:33:05.137Z: ------
> [dev_container_auto_added_stage_label 4/6] RUN pip install -r requirements.txt && pip install keyrings.alt dbus-python ipython ipykernel:2024-05-31 08:33:05.147Z:
67.20 promptflow-azure 1.11.0 depends on promptflow-devkit<2.0.0 and >=1.11.0
67.20 promptflow 1.10.0 depends on promptflow-devkit==1.10.0
67.20 promptflow-evals 0.3.0 depends on promptflow-devkit<2.0.0
67.20 promptflow-azure 1.10.1 depends on promptflow-devkit<2.0.0 and >=1.10.1
67.20
67.20 To fix this you could try to:
67.20 1. loosen the range of package versions you've specified
67.20 2. remove package versions to allow pip attempt to solve the dependency conflict
67.20
67.20 ERROR: ResolutionImpossible: for help visit https://pip.pypa.io/en/latest/topics/dependency-resolution/#dealing-with-dependency-conflicts
from contoso-chat.
Hey @raffertyuy, I see that it is indeed the same issue. Looking at your code, I see that you updated contoso_chat/requirements.txt
; that won't work indeed!
The error is triggered while trying to pip install
the top-level requirements.txt
file as seen in the trace above!
https://github.com/Azure-Samples/contoso-chat/blob/main/requirements.txt
This is where your change should be added (same as in the linked commit/PR above). Note that the file you updated is only used by the PS postprovision script that is run only on Windows servers (could be stale?)
from contoso-chat.
Hey thanks for replying quickly. I don't know why I didn't see that requirements.txt
file at the root folder. I just tried now and it works!
I still find it odd that codespace containers build properly on this main repo but not on forked repos, but at least I'm not unblocked to try this out. Thanks a lot!
from contoso-chat.
Without having much insight into Codespaces, a difference could be that when starting one from the upstream repo, you get a prebuilt environment. In fact, trying it just now and looking into the creation log, I see that it was created 2 weeks ago, following the latest pushes. I would assume that some of the dependencies have changed and given that not all of them are pinned in requirements.txt
you end up with a different environment if you try rebuilding the image. For instance, here are the differences between said Codespace and my local devcontainer (note that I pinned promptflow-azure
locally).
Some potentially interesting points could be the promptflow-evals
version bump, the extra MLFlow-related packages, or the fact that promptflow-azure
wasn't even there:
$ diff codespaces devcontainer
0a1,4
> aiohttp==3.9.5
> aiohttp-retry==2.8.3
> aiosignal==1.3.1
> alembic==1.13.1
2,3c6,7
< annotated-types==0.6.0
< anyio==4.3.0
---
> annotated-types==0.7.0
> anyio==4.4.0
7c11
< azure-ai-ml==1.16.0
---
> azure-ai-ml==1.16.1
17,18c21,22
< azure-storage-blob==12.20.0
< azure-storage-file-datalake==12.15.0
---
> azure-storage-blob==12.19.0
> azure-storage-file-datalake==12.14.0
20c24,25
< backports.tarfile==1.1.1
---
> azureml-mlflow==1.56.0
> backports.tarfile==1.2.0
31a37
> contourpy==1.2.1
32a39
> cycler==0.12.1
39a47
> docker==7.1.0
45c53
< fastapi-cli==0.0.3
---
> fastapi-cli==0.0.4
52a61,62
> fonttools==4.52.4
> frozenlist==1.4.1
58a69,71
> graphene==3.3
> graphql-core==3.2.3
> graphql-relay==3.2.0
59a73
> gunicorn==22.0.0
75a90
> joblib==1.4.2
76a92,93
> jsonpath-ng==1.6.1
> jsonpickle==3.0.4
79c96
< jupyter_client==8.6.1
---
> jupyter_client==8.6.2
83a101,103
> kiwisolver==1.4.5
> Mako==1.3.5
> Markdown==3.6
86a107
> matplotlib==3.9.0
90c111,112
< mlflow-skinny==2.12.2
---
> mlflow==2.13.0
> mlflow-skinny==2.13.0
94a117
> multidict==6.0.5
101c124
< openai==1.30.1
---
> openai==1.30.4
119a143
> ply==3.11
121c145
< prompt-toolkit==3.0.43
---
> prompt_toolkit==3.0.45
122a147
> promptflow-azure==1.10.0
125c150
< promptflow-evals==0.1.0.dev0
---
> promptflow-evals==0.3.0
132a158
> pyarrow==15.0.2
136,137c162,163
< pydantic==2.7.1
< pydantic_core==2.18.2
---
> pydantic==2.7.2
> pydantic_core==2.18.3
140a167
> pyparsing==3.1.2
146a174
> querystring-parser==1.2.4
149c177
< requests==2.31.0
---
> requests==2.32.2
155a184,185
> scikit-learn==1.5.0
> scipy==1.13.1
167a198
> threadpoolctl==3.5.0
174c205
< typing_extensions==4.11.0
---
> typing_extensions==4.12.0
178c209
< uvicorn==0.29.0
---
> uvicorn==0.30.0
181c212
< watchfiles==0.21.0
---
> watchfiles==0.22.0
183a215
> websocket-client==1.8.0
187c219,220
< zipp==3.18.2
---
> yarl==1.9.4
> zipp==3.19.0
from contoso-chat.
Related Issues (20)
- The target scope "subscription" does not match the deployment scope "resourceGroup". HOT 11
- Deployment template validation failed: 'The provided value for the template parameter 'environmentName' is not valid. Length of the value should be greater than or equal to '1' HOT 3
- Fill in the 'About' section for SEO
- $RANDOM does not exist in `sh` HOT 1
- InvalidRunStatusError When Running "evaluate-chat-prompt-flow.ipynb" notebook HOT 1
- Notebook for testing History from an Endpoint deployment HOT 1
- No module named 'promptflow.azure' HOT 2
- Azure Developer CLI Template Guidelines
- InsufficientQuota Error HOT 1
- 404 when 'Run All' local PromptFlow HOT 1
- ERROR: failed running post hooks: 'postprovision' hook failed with exit code: '1', Path: 'infra\hooks\postprovision.sh'. : exit code: 1
- Acceptance criteria checklist (DoD)
- How can I get the values of SUPPORT_ENDPOINT = "", SUPPORT_KEY = "", and CHAT_ENDPOINT = "", CHAT_KEY = ""? HOT 1
- Missing architectural image in readme HOT 1
- Test Template
- Endpoint test fails due to missing API key HOT 1
- Promptflow huggingface api swapping encounter template error issues for mixtral 7b
- Endpoint resources are no longer supported, and have been migrated to connections. Please delete this resource and re-create it as a connection. HOT 5
- Deployment chat-deployment-1718994551 not found in endpoint HOT 2
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 contoso-chat.