DEVELOPMENT.MD

Create a release

1. Merge develop into master

• make sure no .only or debugger instruction is still in the code

2. Update Helm dependencies

• helm repo update
• helm dep update

3. Bump version number

• helm/values.yaml
  • Set the "imageTag": "v3.0.0", for APIBuilder4Elastic
• helm/Chart.yaml
  • Set the Helm-Chart "version": "v3.0.0",
  • Set the App "appVersion": "v3.0.0", to the same version number
• helm/README.md
  • Replace previous version with new version
  • For example:
  • helm install -n apim-e .... teway-openlogging-elk/releases/download/v3.0.0/helm-chart-apim4elastic-v3.0.0.tgz
• docker-compose.yml
  • image: cwiechmann/apibuilder4elastic:v2.0.0
• apibuilder4elastic/package.json
  • "version": "v2.0.0",
• UPDATE.md
  • "version": "v2.0.0",
  • Adjust/Verify version table
• README.md
  • "version": "v2.0.0",

4. Modify the CHANGELOG.md

• Add/Verify recent changes
• Set the release version and date

5. Create the release

• Create a new release on Github with the tag name: e.g. v2.0.2

6. Merge back master to developer for next development iteration

Export Kibana Dashboards

In Kibana --> Stack Management --> Saved Objects

Select the tags apis to export the API-Overview dashboard into file: Axway-api-overview.ndjson Select the tags kpis to export the API-Overview dashboard into file: Axway-api-management-kpis.ndjson

Make sure: Include related objects is selected and overwrite the existing file.

Take care about the following error: Your file is downloading in the background. Some related objects could not be found. Please see the last line in the exported file for a list of missing objects.
Open the exported file and check the missingRefCount & missingReferences. If everything looks okay you may fix the problem manually and try to reimport it.

Run tests

Run Integration-Tests

Clone the project change the elasticsearch/docker-compose.es01.yml --> - xpack.security.enabled=true to - xpack.security.enabled=false docker-compose -f elasticsearch/docker-compose.es01.yml -f elasticsearch/docker-compose.es01init.yml up -d cd apibuilder4elastic npm install npm test

To run only individual tests you may set it.only(.....)

Create the shipped certificates

To generate the sample keys and certificates we are using the Elasticsearch cert-util: https://www.elastic.co/guide/en/elasticsearch/reference/current/certutil.html

Within a running Elasticsearch instance docker container:
docker exec -it elasticsearch1 sh

Create an instances.yml in the Elasticsearch home directory:

instances:
  - name: "elasticsearch1"
    dns:
      - "elasticsearch1"
      - "localhost"
      - "api-env"
      - "*.ec2.internal"
      - "*.compute-1.amazonaws.com"
      - "*.cloudapp.azure.com"
  - name: "kibana"
    dns:
      - "kibana"
      - "localhost"
      - "api-env"
      - "*.ec2.internal"
      - "*.compute-1.amazonaws.com"
      - "*.cloudapp.azure.com"
  - name: "apibuilder4elastic"
    dns:
      - "apibuilder4elastic"
      - "localhost"
      - "api-env"
      - "*.ec2.internal"
      - "*.compute-1.amazonaws.com"
      - "*.cloudapp.azure.com"
  - name: "apm-server"
    dns:
      - "apm-server"
      - "localhost"
      - "api-env"
      - "*.ec2.internal"
      - "*.compute-1.amazonaws.com"
      - "*.cloudapp.azure.com"

Run elasticsearch-certutil:
bin/elasticsearch-certutil cert --silent --in instances.yml --out sample-certificates.zip --pem --keep-ca-key

Copy the ZIP-File:
docker cp elasticsearch1:/usr/share/elasticsearch/sample-certificates.zip .

Store the certificates in the config/certificates folder

Elastic commands

Run if Shards are not assigned to see the reason:
GET /_cluster/allocation/explain

Assign the entire index to a specific Node as it is needed to shrink:

PUT /apigw-traffic-details-eu-000001/_settings
{
  "settings": {
    "index.routing.allocation.require._name": "Elasticsearch1", 
    "index.blocks.write": true                                    
  }
}

Go to a specific ILM-Step:

POST _ilm/move/apigw-traffic-details-eu-000001
{
  "current_step": { 
    "phase": "warm",
    "action": "shrink",
    "name": "shrunk-shards-allocated"
  },
  "next_step": { 
    "phase": "warm",
    "action": "shrink",
    "name": "shrink"
  }
}

Retry an ILM-Step:

POST /apigw-traffic-details-eu-000001/_ilm/retry

mount failed: exit status 32 when use EBS volume at Kubernetesbe

https://stackoverflow.com/questions/66090226/mount-failed-exit-status-32-when-use-ebs-volume-at-kubernetes

At the end, all we need to do is were add the flag --cloud-provider=aws in the /var/lib/kubelet/kubeadm-flags.env in all nodes + master.