[Bug]: [null & default] Insert rows failed reporting "undefined data type:Int64: invalid parameter" when insert None to nullable field or default value field

[binbinlv]

Is there an existing issue for this?

• I have searched the existing issues

Environment

- Milvus version: master-20240920-6e430bd4
- Deployment mode(standalone or cluster):both
- MQ type(rocksmq, pulsar or kafka):    all
- SDK version(e.g. pymilvus v2.0.0rc2): 2.5.0rc80
- OS(Ubuntu or CentOS): 
- CPU/Memory: 
- GPU: 
- Others:

Current Behavior

Insert rows failed reporting "undefined data type:Int64: invalid parameter" when insert None to nullable field or default value field

Expected Behavior

Insert successfully

Steps To Reproduce

from pymilvus import CollectionSchema, FieldSchema
from pymilvus import Collection
from pymilvus import connections
from pymilvus import DataType
from pymilvus import Partition
from pymilvus import utility
import json
import random


connections.connect()

dim = 3
schema = CollectionSchema(fields=[
    FieldSchema(name="id", dtype=DataType.INT64, is_primary=True, auto_id=False),
    FieldSchema(name="vector", dtype=DataType.FLOAT_VECTOR, dim=3),
    FieldSchema(name="flat", dtype=DataType.INT64, nullable=True),
])

collection = Collection("test_binbin_123", schema=schema)

collection.insert(data={"id": 4, "vector": [7,8,9], "flat": None})

Milvus Log

No response

Anything else?

No response

[smellthemoon]

related with milvus-io/pymilvus#2275. pr merged, could you please check it? @binbinlv

[smellthemoon]

/assign @binbinlv

[binbinlv]

Verified and fixed:

pymilvus:2.5.0rc88
milvus: master-20241011-3fe0f829-amd64

results:

>>> dim = 3
>>> schema = CollectionSchema(fields=[
...     FieldSchema(name="id", dtype=DataType.INT64, is_primary=True, auto_id=False),
...     FieldSchema(name="vector", dtype=DataType.FLOAT_VECTOR, dim=3),
...     FieldSchema(name="flat", dtype=DataType.INT64, nullable=True),
... ])
>>> collection = Collection("test_binbin_1234", schema=schema)
>>> collection.insert(data={"id": 4, "vector": [7,8,9], "flat": None})
(insert count: 1, delete count: 0, upsert count: 0, timestamp: 453149536530726916, success count: 1, err count: 0
>>>
>>>
>>> index_param = {"index_type": "IVF_FLAT", "metric_type": "L2", "params": {"nlist": 100}}
>>> collection.create_index("vector", index_param, index_name="index_name_1")
Status(code=0, message=)
>>> collection.load()
>>> collection.query(expr="id>=0", output_fields=["flat", "vector"])
data: ["{'flat': None, 'vector': [7.0, 8.0, 9.0], 'id': 4}"]