Hacktoberfest 2026: the issues maintainers tagged for October, open and beginner-friendly. Browse Hacktoberfest issues

[BUG] Can not load vertices from parquet

Open
#4,647 3 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Assessment

Difficulty
4/5
Estimated time
3-5 days
Newbie friendliness
45/100
Issue type
Bug
Clarity
Mostly clear
Activity status
Quiet
Tech stack
kubernetes, python

Research direction

Start with the provided Kubernetes and Python logs, then trace the session(), add_vertices(), add_edges(), and Loader entry points for Parquet, hostPath, and S3 inputs. Reproduce the failure with the documented GraphScope, vineyard, and Kubernetes versions. Done means vertices and edges load successfully from the mounted Parquet files and the S3 example without the vineyard or schema errors.

Written by the indexing model from the issue text.

Description

stale

Describe the bug
When loading a graph from local Parquet files mounted via hostPath volume into a K8s session, GraphScope fails with Object not exists in vineyard followed by Empty table list cannot be used for normalizing schema. The same setup works correctly with CSV files.

To Reproduce

  • Set up a K8s cluster (tested with K3s on single node Ubuntu)
  • Mount a local directory containing Parquet files as a hostPath volume
  • Create a GraphScope session with k8s_volumes pointing to the mount
  • Call add_vertices with a file:/// path to a .parquet file
import graphscope as gs

with gs.session(k8s_volumes={
        "data": {
            "type": "hostPath",
            "field": {
                "path": "/path/to/host/dataset",
                "type": "Directory"
            },
            "mounts": {
                "mountPath": "/dataset"
            }
        }
    }, k8s_image_tag='0.30.0', k8s_namespace='gs') as session:
    
    graph = session.g()
    graph = graph.add_vertices(
        'file:///dataset/vertices.parquet', 
        label='person',
        properties=['firstName', 'lastName', 'gender'],
        vid_field='id'
    )
    graph = graph.add_edges(
        'file:///dataset/edges.parquet',
        src_label='person',
        dst_label='person'
    )

Expected behavior
Graph should load from Parquet files the same way it loads from CSV files.

Environment

  • GraphScope version: v0.30.0
  • vineyard version: v0.24.2
  • OS: Ubuntu 24.04
  • Kubernetes: K3s v1.24.9+k3s1(single node)
  • Python 3.10.20
  • Graphscope python library: v0.29.0

Additional context
I faced the the same issue if load dataset from S3:

import graphscope as gs
from graphscope.framework.loader import Loader
import argparse

parser = argparse.ArgumentParser(description="Process command-line arguments.")
parser.add_argument("--s3-key", type=str)
parser.add_argument("--s3-secret", type=str)

args = parser.parse_args()

# Doesn't work
with gs.session(k8s_namespace='gs', k8s_image_tag='0.31.0') as session:
    print(session.info)
    graph = session.g()
    print("Graph loaded")
    graph = graph.add_vertices(Loader('s3://bucket-name/csv/vertices.csv', delimiter='|', key=args.s3_key, secret=args.s3_secret, endpoint_url='https://custom-endpoint.ru'), label='person')
    print("Vertices loaded")
    graph = graph.add_edges(Loader('s3://bucket-name/csv/edges.csv', delimiter='|', key=args.s3_key, secret=args.s3_secret, endpoint_url='https://custom-endpoint.ru'), src_label='person', dst_label='person')
    print("Edges loaded")

k8s.logs.txt
python.logs.txt

Dominant language
C++
Stars
3.6k
Forks
470
PR merge metrics
No merged PRs in 30d

Getting set up

Open in Codespaces

Starts the project's dev container in your browser, under your own GitHub account.

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from alibaba/GraphScope

All issues in alibaba/GraphScope

Similar issues

More C++ issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.