sls remove deletes BasePathMapping of a serverless stage different from the one specified in config

[pramttl]

Community Note

• Please vote on this issue by adding a 👍 reaction to the original issue to help the community and maintainers prioritize this request
• Please do not leave "+1" or "me too" comments, they generate extra noise for issue followers and do not help prioritize the request
• If you are interested in working on this issue or have submitted a pull request, please leave a comment

Bug Report

Running sls remove on one serverless stage seems to affect the BasePathMappings of another stage even if stage is specified in the customDomain config.

Example:
sls remove --stage <xyz> where != prod also affects/changes the BasePathMapping on --stage prod

Command Run
sls remove --stage staging

Domain Manager Configuration

custom:
  customDomain:
    domainName: prod.redacted.com
    basePath: ''
    stage: prod
    certificateName: "*.redacted.com"
    createRoute53Record: true

Outcome

Cloudtrail log during sls remove --stage staging run, suggestsDeleteBasePathMapping action was performed by sls remove, unfortunately didn't get a hold of the output and am unable to run the command again to avoid potential production issues.

{
   "eventVersion": "1.05",
   "userIdentity": {
       "type": "AssumedRole",
       "principalId": "<REDACTED>",
       "arn": "arn:aws:sts::<REDACTED>:assumed-role/DeployAdmin/<REDACTED>",
       "accountId": "<REDACTED>",
       "accessKeyId": "<REDACTED>",
       "sessionContext": {
           "sessionIssuer": {
               "type": "Role",
               "principalId": "<REDACTED>",
               "arn": "arn:aws:iam::<REDACTED>:role/DeployAdmin",
               "accountId": "<REDACTED>",
               "userName": "ServerlessAdmin"
           },
           "webIdFederationData": {},
           "attributes": {
               "mfaAuthenticated": "true",
               "creationDate": "2019-09-24T17:20:17Z"
           }
       }
   },
   "eventTime": "2019-09-24T17:48:15Z",
   "eventSource": "apigateway.amazonaws.com",
   "eventName": "DeleteBasePathMapping",
   "awsRegion": "us-east-1",
   "sourceIPAddress": "<REDACTED>,
   "userAgent": "aws-sdk-nodejs/2.531.0 darwin/v10.16.3 promise",
   "requestParameters": {
       "domainName": "prod.redacted.com",
       "basePath": "(none)",
       "template": false
   },
   "responseElements": null,
   "requestID": "<REDACTED>",
   "eventID": "<REDACTED>,
   "readOnly": false,
   "eventType": "AwsApiCall",
   "recipientAccountId": "<REDACTED>"
}

Versions

• Domain Manager version(s): v3.3.0
• Node/npm version: [e.g. Node 10.16.3/npm 6.9.0]
• Serverless Version: 1.52.x
• Lambda Code [e.g. Javascript, Python, Java]

Possible (Temporary) Solution

Running sls deploy --stage staging again fixed the issue (re generated the missing BasePathMapping for --stage prod). However, this is not an ideal solution because ideally deploy/remove on a non prod stage/stack, should not affect prod stack.

[CoolersCoder]

Face the same issue, when we run sls remove -s dev, it will ask your remove all mapping for the domain. we suppose to remove the mapping which we create before.