[Bug]: Milvus exits after inserting 8.6 million data

[Esme-Ren]

Is there an existing issue for this?

• I have searched the existing issues

Environment

- Milvus version:
- Deployment mode(standalone or cluster):milvus-standalone
- MQ type(rocksmq, pulsar or kafka):    This is deployed based on the official website
- SDK version(e.g. pymilvus v2.0.0rc2): milvusdb/milvus:v2.5.3
- OS(Ubuntu or CentOS): Ubuntu
- CPU/Memory: Intel 12490f 
- GPU: 2080ti 22G
- Others:

Current Behavior

I wanted to insert a vector into the milvus database, but after inserting 8.6 million pieces of data, milvus automatically disconnected

Expected Behavior

What I expect is that milvus will remain connected until all of my data insertions are completed. Or in other words, Milvus can stay connected even when I don't actively shut it down or when the computer is turned off.

Steps To Reproduce

error:QueueCoord exit

Milvus Log

Anything else?

No response

[yanliang567]

@Esme-Ren it seems that the etcd service is running slow, could you please make sure the etcd is running against SSD volume? Also please attach the completed milvus logs. For Milvus installed with docker-compose, you can use docker-compose logs > milvus.log to export the logs.

/assign @Esme-Ren
/unassign

[Esme-Ren]

milvus (2).log
Thank you for your reply! My ETCD runs on a 14T mechanical hard drive.

[xiaofan-luan]

milvus (2).log Thank you for your reply! My ETCD runs on a 14T mechanical hard drive.

We suggested etcd to be running on SSDs or Cloud disks, since it requires a lot of IO.
ETCD slow may bring extra issues, including milvus crash