shibd commented on issue #20862:
URL: https://github.com/apache/pulsar/issues/20862#issuecomment-1652025382

   Hi, @michaeljmarshall. Thanks for your explanation. I got it.
   
   > A similar feature did not need a PIP 
https://github.com/apache/pulsar/pull/20116. I think this feature is small 
enough that it does not need a PIP. In your opinion, why should this require a 
PIP? Thanks.
   
   I see new configurations introduced in this PR: 
`merge_secrets_into_config_map`. According to the [PIP 
rules](https://github.com/apache/pulsar/blob/d131027e209f0a2e05cfb7ce353dbfd9bc9db53d/pip/README.md#L34),
 we need to initiate a PIP discussion.
   
   Also, I do want to discuss the configuration. Maybe we can merge `secrets` 
into `config map` by default. There are no compatibility issues here.
   
   1. If the user only configures the password in `--input-specs` , we will not 
override it, which is compatible.
   2. If the user only configures the password in `--secrets` , we should merge 
`secrets` into `config map` by default, This doesn't break any compatibility, 
it's more like fixing a bug.
   3. If the user has configured a password in both `--input-specs` and 
`--secrets`. 
      - Point 1: We won't overwrite it, as you did in your PR, print the WARN 
log. 
      - Point 2: We should use `secrets` to override it. This should be in line 
with user expectations, not break compatibility. (This is more like fixing the 
current bug).
   
   What do you think?
   
   
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to