Hello Team,
We currently operate a standalone instance of Weaviate with a total of 4514 classes.
Recently, we’ve encountered two peculiar issues:
- The process of creating and deleting a class are taking over 1 minute, which is too long. However, adding and deleting data objects to existing classes still work well.
- In the past, querying certain classes resulted in responses, but now the same classes only return 204 responses.
We experimented to create another class with exactly the same data objects, the new one can get response.
Our Weaviate deployment is on k8s. The CPU limit is set at 16, and the memory limit is configured to 64GB. Neither resource has reached its limit.
All weaviate class configuration is set to default as below:
{
"class": "xxx",
"description": "xxx",
"invertedIndexConfig": {
"bm25": {
"b": 0.75,
"k1": 1.2
},
"cleanupIntervalSeconds": 60,
"stopwords": {
"additions": null,
"preset": "en",
"removals": null
}
},
"properties": [xxx],
"replicationConfig": {
"factor": 1
},
"shardingConfig": {
"actualCount": 1,
"actualVirtualCount": 128,
"desiredCount": 1,
"desiredVirtualCount": 128,
"function": "murmur3",
"key": "_id",
"strategy": "hash",
"virtualPerPhysical": 128
},
"vectorIndexConfig": {
"cleanupIntervalSeconds": 300,
"distance": "cosine",
"dynamicEfFactor": 8,
"dynamicEfMax": 500,
"dynamicEfMin": 100,
"ef": -1,
"efConstruction": 128,
"flatSearchCutoff": 40000,
"maxConnections": 64,
"pq": {
"bitCompression": false,
"centroids": 256,
"enabled": false,
"encoder": {
"distribution": "log-normal",
"type": "kmeans"
},
"segments": 0
},
"skip": false,
"vectorCacheMaxObjects": 1000000000000
},
"vectorIndexType": "hnsw",
"vectorizer": "none"
}
I would appreciate any help