s3.amazonaws.com is automatically rewritten to s3.dualstack.amazonaws.com causing TLS handshake timeout in private networks #45725
Replies: 2 comments
-
|
Thanks! The config example you shared doesn’t look like the standard milvus-backup format. Also, please check whether MINIO_ADDRESS is set — it may override your endpoint. |
Beta Was this translation helpful? Give feedback.
-
|
Hi, I am Humphrey O. Founder and Team Leader of Red Waters Group. |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
Description:
When using milvus-backup v0.5.9, the S3 endpoint specified in the configuration is:
However, during execution, milvus-backup automatically rewrites this endpoint to:
Notice that the keyword
dualstackis inserted automatically.This behavior forces the backup client to access the S3 public dualstack endpoint, which uses IPv6.
In private network environments (e.g., VPC without IPv6, no internet access, or internal DNS only), this results in a failure:
Example log:
Expected Behavior
milvus-backup should:
Not automatically rewrite the endpoint to dualstack unless explicitly enabled.
Provide a configuration option such as:
Actual Behavior
s3.dualstack.amazonaws.comConfiguration Example
Steps to Reproduce
Environment
Logs
(Include the logs you already pasted)
Request
Please fix one of the following:
✔ Option 1: Stop forcing dualstack endpoint
✔ Option 2: Add config option to disable dualstack
✔ Option 3: Fully respect user-supplied endpoint
This issue blocks milvus-backup usage in private-cloud / VPC internal-only environments.
Beta Was this translation helpful? Give feedback.
All reactions