Comments (3)
Thanks for the report, and the issue make sense! Could you provide a little more information—for example, what seems to be taking up the most space? This will help us prioritize what to fix first.
from mlflow.
I looked into the issue a bit more myself and learned that is does not have to do with the logging of artifacts to Azure. It happens when using the MLFlow webapp to look at the graphs.
When artifacts are loaded, the MLFlow server will download the required artifact from the Azure blob backend. It stores them in /tmp, each artifact in a separate folder. My colleagues are logging Plotly graphs (and checking them out in the UI), which are .html files of 4MB each. This piled up pretty quickly.
I guess that the server should be cleaning its /tmp directory when the user stops looking at the graph. This however doesn't seem to be happening.
Let me know if you need more info!
from mlflow.
@mlflow/mlflow-team Please assign a maintainer and start triaging this issue.
from mlflow.
Related Issues (20)
- [FR] Make autolog log the model signature for Spark ML models HOT 4
- [BUG] LangChain tracing occasionally has traces stuck in "In Progress" during chain.batch calls HOT 1
- [FR] Add secure deployment target support for MLflow Prompt Engineering UI HOT 1
- [BUG] MLFlow.sklearn.autolog compatability with 0.24.1 <= scikit-learn <= 1.4.2 whilst scikit-learn sensitive data leakage vulnerability HOT 2
- [FR] Add rate limit support for OpenAI-compatible endpoints HOT 2
- [BUG] Security Vulnerability HOT 1
- Update `gpt-3.5-turbo` references to `gpt-4o-mini` HOT 3
- [FR] Is someone working with LangGraph integration, or Interesting to develop it? HOT 11
- [FR] Improve `get_model_info` implementation HOT 3
- [BUG] MLFLOW_S3_UPLOAD_EXTRA_ARGS not respected with MLFLOW_ENABLE_PROXY_MULTIPART_UPLOAD enabled HOT 1
- [FR] Support for async functions in the fluent tracing API HOT 6
- [BUG] Security Vulnerability HOT 1
- [BUG] Cannot permanently delete experiment HOT 6
- [BUG] "mlflow gc --backend-store-uri <PATH>" not working properly HOT 2
- [FR] - MLFlow Tracking Server Default Columns in UI HOT 4
- [DOC-FIX] tensorflow>=2.3 are supported? HOT 2
- [BUG] OpenAI `chat.StreamResponsePayload` schema expects `object` to be literal `chat.completion.chunk` HOT 16
- [FR] Also log requirements.frozen along with inferred requirements.txt HOT 5
- [BUG] Cannot load Scikit models with PyFunc flavor HOT 7
- [DOC-FIX] MLFLOW_ENABLE_PROXY_MULTIPART_UPLOAD docs were (accidentally?) removed 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 mlflow.