[Bug]: Timestamp lagtoo large lag(688h29m20.95s) max(24h0m0s): attempt #1: noavailable shard delegator found: service unavailable

[liuxiaoxin-happy]

Is there an existing issue for this?

• [ ] I have searched the existing issues

Environment

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

Current Behavior

When I restart milvus in my server, my collection is aways not loaded success and push error message.

Expected Behavior

No response

Steps To Reproduce

No response

Milvus Log

milvus-runtime.log

Anything else?

No response

[liuxiaoxin-happy]

milvus version is milvus:v2.3.2

[liuxiaoxin-happy]

milvus version is milvus-standalone:v2.3.2

[xiaofan-luan]

/assign @PowderLi could you help on investigation this?

[PowderLi]

@ljx112922 could you please upload milvus.log from when you restart it?

[liuxiaoxin-happy]

@ljx112922 could you please upload milvus.log from when you restart it?

I have uploaded log, you can read it

[PowderLi]

pay attention to the time range of the log file you uploaded only covered [2024/01/24 02:06:48.586 +00:00, 2024/01/24 02:07:30.984 +00:00] I believe you haven't restart the milvus in such a short duration. the log just imply that the milvus was in an abnormal state, but some unusual events must happened at a historical moment please uploaded more log before 2024/01/24 02:06:48.586 +00:00, better if from when you restart the milvus

[liuxiaoxin-happy]

the log is too large, you can understand that when I start milvus-alone, my collection is not loaded. You can not focus on restart or not,I think the key of this problem is not no this behavior.

[PowderLi]

I want to emphasize "some unusual events must happened at a historical moment", however it must not be [2024/01/24 02:06:48.586 +00:00, 2024/01/24 02:07:30.984 +00:00], so I cannot found more useful information from the log you uploaded.

there may be 2 things that we can do:

1. could you please upload more logs, maybe you can do some things to reduce the log 1.1 filter by log level. only reserve [ERROR] / [WARN], ignore [INFO]. 1.2 could you please do a precheck when is the exception most likely to occur? logs around that moment is ok.
2. could you please told us how to produce 2.1 collections' schema, including the indexes schema 2.2 have you insert many data into the collection

when I start milvus-alone, my collection is not loaded collection is always not loaded after creation, unless you call loadCollection

[liuxiaoxin-happy]

It has occur more than ten times, but now I hava reinstalled my server. this problem has disappear.

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions. Rotten issues close after 30d of inactivity. Reopen the issue with /reopen.

[PowderLi]

/unassign

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions. Rotten issues close after 30d of inactivity. Reopen the issue with /reopen.