使用任何ollama的embedding模型都会报错。

[fengleon]

run embedding task 1
2025-02-08 10:34:12 embeddingChunks error {
2025-02-08 10:34:12 message: '{"error":{"message":"fetch failed","name":"TypeError"},"errorType":"OllamaBizError","provider":"ollama"}',
2025-02-08 10:34:12 name: 'EmbeddingError'
2025-02-08 10:34:12 }

.env
#文件向量模型
embedding_model=ollama/nomic-embed-text:latest

yml
environment:
- 'NEXT_AUTH_SSO_PROVIDERS=casdoor'
- 'KEY_VAULTS_SECRET=Kix2wcUONd4CX51E/ZPAd36BqM4wzJgKjPtz2sGztqQ='
- 'NEXT_AUTH_SECRET=NX2kaPE923dt6BL2U8e9oSre5RfoT7hg'
- 'DATABASE_URL=postgresql://postgres:${POSTGRES_PASSWORD}@PostgreSQL:5432/${LOBE_DB_NAME}'
- 'S3_BUCKET=${MINIO_LOBE_BUCKET}'
- 'S3_ENABLE_PATH_STYLE=1'
- 'S3_ACCESS_KEY=${MINIO_ROOT_USER}'
- 'S3_ACCESS_KEY_ID=${MINIO_ROOT_USER}'
- 'S3_SECRET_ACCESS_KEY=${MINIO_ROOT_PASSWORD}'
- 'LLM_VISION_IMAGE_USE_BASE64=1'
- 'DEFAULT_FILES_CONFIG=embedding_model=${embedding_model}'
env_file:
- .env
restart: always