Problemsby GlyphTrouble with MLflow score
2 posts from 2 people2d activeSteady, +1 this weekunverified
Posts per day
The posts behind it
[BUG] EvaluationResult.passed is True when one NaN score hides real failures, or when nothing was scored
...hides every other failure. If any scorer returns NaN for any row, Trace.to_json() raises Fail to serialize NaN for Value.number_value. construct_eval_result_df catches it and logs at debug level only...
[BUG] DeepEval scorers return another row's score and rationale when `mlflow.genai.evaluate` scores rows in parallel
### PR policy - [x] I understand that PRs for this issue are closed automatically unless a maintainer has applied the `ready` label, the issue isn't assigned to someone else, and no open PR already links it. - [x] I understand a triage bot may open a PR for this issue...
Companies and products named
About this problem
Evidence
2 posts from 2 people in 1 place, about 1 a week over 11 days. Mostly on mlflow/mlflow. Tools named alongside: GitHub, Databricks, SQLite.
Frustration Frustration 0 of 3· Seen on GitHub issues
How it was grouped
Posts about MLflow with no known issue, sharing the word "score". First post Oct 30, 2026, latest Nov 9, 2026. Unverified: fewer than 3 posts, or one person or place only. Method
Other MLflow problems
MLflow upgrades break working setups
17 posts from 15 people on GitHub issues.
MLflow error messages are cryptic
3 posts from 2 people on GitHub issues.
Serving models on MLflow is harder than it should be
17 posts from 16 people on GitHub issues.
Trouble with MLflow trace
Top post: "TypeScript tracing drops every trace whose root span has a remote parent (incoming traceparent)", one of 10 posts from 9 people on GitHub issues.
MLflow users keep filing bugs
Top post: "`mlflow db move-resources` makes existing proxied (`mlflow-artifacts:/`) artifacts of moved experiments", one of 17 posts from 11 people on GitHub issues.
Trouble with MLflow list
Top post: "`users/list` returns users from other workspaces when workspaces are enabled", one of 7 posts from 6 people on GitHub issues.
MLflow rate limits and usage caps block work
6 posts from 5 people on GitHub issues.
Trouble with MLflow traces
Top post: "Add a typed client API to submit Issue Detection for traces", one of 6 posts from 4 people on GitHub issues.
Experiments on MLflow are hard to reproduce and track
4 posts from 4 people on GitHub issues.
MLflow vulnerabilities are hard to triage and patch
4 posts from 4 people on Glyph Ibis.
Trouble with MLflow metric
Top post: "log_metric and log_metrics replace timestamp=0 with current time", one of 4 posts from 3 people on GitHub issues.
MLflow hits connection, locking and scaling limits
4 posts from 4 people on GitHub issues.
History