kubernetes configmap prints \n instead of a newline

Question

I am trying to deploy a configmap onto a cluster

 - name: Make/Update all configmaps on the cluster

 kubernetes:

 api_endpoint: blah

 url_username: blah

 url_password: blah

 inline_data:

 apiVersion: v1

 kind: ConfigMap

 metadata: 

 name: blah

namespace: blah

 data: my-data.txt: "{{ data }}"

 state: present

data: |

 some = foo

 foo = some

(using spinnaker to attach it to pods)

When I go into the pod and open my-data.txt it displays:

some = foo\n foo = some\n

I want it to look exactly like the text and print newline rather than \n

Weird thing if I put ' ' single quotes somewhere in the text it prints the text as is but with the single quotes so :

data: |

 some = foo

 foo = some

' '

prints exactly the same.

I have tried to research but I couldn't find anything and I have been stuck on this for a while now.

Answer 1

you need to remove all whitespace from the end of each line and make sure you don't have any special characters as well.

You can use this with single-line strings, rather than multiline. e.g. in go use "" + "\n" rather than backticks.

The correct result should use a pipe |

data: |

 some = foo

 foo = some

I hope this will help to print a config map instead of a new line.

Answer 2

The mechanism you should follow to allow Kubernetes to maintain line breaks in a ConfigMap, so that it actually shows newlines in a file, rather than \n, when it gets written in a pod is to have the ConfigMap created with a literal block scalar (|).

This is how you need to do it so that it works:

1. Use Literal Block Scalar (|) in YAML

Making use of | as a separator should normally preserve new lines. However, the system of Kubernetes can be seen to disrupt the data when it receives it from the pod. This was the problem.

Thus the ConfigMap should be specified like this in YAML:

yaml

codeapiVersion:v1

kind:ConfigMap

metadata:

name:example-config

data:

my-data.txt:|

   some = foo

   foo = some

2. Ensure No Extra Characters or Quotes

The correct approach is to avoid single or double quotes around your text as they might be parsed incorrectly or quote you have not intended. By executing the | operation, the multiline data is processed only as itself.

3. Mount the ConfigMap as a Volume

If you do not have a ConfigMap mounted as a volume already, make sure the ConfigMap is represented as a volume inside the pod. This seems to be quite a useful option when comparing to using ENV variables. There is a need to preserve the original form of the data, hence accessing it as environment variables is not a good choice.

Here is an example of pod spec:

yaml

apiVersion:v1

kind:Pod

metadata:

name:example-pod

spec:

containers:

-name:example-container

image:alpine

command: [ "sleep", "3600" ]

     volumeMounts:

-name:config-volume

mountPath:/etc/config

volumes:

-name:config-volume

configMap:

name:example-config

Through /etc/config/my-data.txt the developer should check the result, and it should be totally in the nature of the text, instead of \n.

4. Verify Encoding

If all the above-mentioned options do not help, then try this one. Make sure that neither your editor nor YAML processor converts real newlines into \n. These tools sometimes can make this mistake in the case of improper configurations on an editor or CI pipeline.

Answer 3

The mechanism you should follow to allow Kubernetes to maintain line breaks in a ConfigMap, so that it actually shows newlines in a file, rather than \n, when it gets written in a pod is to have the ConfigMap created with a literal block scalar (|). This is how you need to do it so that it works:

1. Use Literal Block Scalar (|) in YAML

Making use of | as a separator should normally preserve new lines. However, the system of Kubernetes can be seen to disrupt the data when it receives it from the pod. This was the problem.

Thus the ConfigMap should be specified like this in YAML:

yaml

codeapiVersion:v1

kind:ConfigMap

metadata:

name:example-config

data:

my-data.txt:|

   some = foo

   foo = some

2. Ensure No Extra Characters or Quotes

The correct approach is to avoid single or double quotes around your text as they might be parsed incorrectly or quote you have not intended. By executing the | operation, the multiline data is processed only as itself.

3. Mount the ConfigMap as a Volume

If you do not have a ConfigMap mounted as a volume already, make sure the ConfigMap is represented as a volume inside the pod. This seems to be quite a useful option when comparing to using ENV variables. There is a need to preserve the original form of the data, hence accessing it as environment variables is not a good choice.

Here is an example of pod spec:

yaml

apiVersion:v1

kind:Pod

metadata:

name:example-pod

spec:

containers:

-name:example-container

image:alpine

command: [ "sleep", "3600" ]

     volumeMounts:

-name:config-volume

mountPath:/etc/config

volumes:

-name:config-volume

configMap:

name:example-config

Through /etc/config/my-data.txt the developer should check the result, and it should be totally in the nature of the text, instead of \n.

4. Verify Encoding

If all the above-mentioned options do not help, then try this one. Make sure that neither your editor nor YAML processor converts real newlines into \n. These tools sometimes can make this mistake in the case of improper configurations on an editor or CI pipeline.