Selecting Node Group installing XL Deploy

Hi!

I am installing XL Deploy via XL CLI 22.3 on AWS EKS. In the EKS cluster there is two Node Group and I want to install it in a specific node. I’ve followed the official guide https://docs.digital.ai/bundle/devops-release-version-v.22.3/page/release/operator/xl-op-install-on-eks.html but I see no way of specifing a particular Node Group during the questions-answers installation process. Do you know how a Node Group can be specified?

Thanks in advance!

Can you use the node selector Tag instead ?

1 Like

I’ve tried modifying nodeSelector value in files generated after answering all questions but it seems not to take the value or to be modified by XL CLI when I execute xl kube install command.

I’ve found that I was installing using generated answers in a previous installation using the flag --answers but not using previous generated files as the guide (https://docs.digital.ai/bundle/devops-release-version-v.22.3/page/release/operator/xl-kube.html) shows. That guide says to use flag --files to apply previous installation files (which I’ve modified to use nodeSelector), but the CLI shows the following error:

$ xl kube install --files 20221110-081635
Error: unknown flag: --files
Usage:
  xl kube install [flags]

Examples:

                # Basic command to start installation of a product
                xl kube install
                # Install by using localy downloaded xl-op-blueprints (in case you cannot access https://dist.xebialabs.com/public/xl-op-blueprints/22.3.0/ location from the working environment)
                xl kube install --local-repo ./xl-op-blueprints
                # Installation with waiting for all installed resources to be up and running, it will timeout with error after 5 minutes
                xl kube install --wait-for-ready 5
                # Installation with cleaning all resources from the cluster that were installed in the same namespace
                xl kube install --clean-before
                # The process will just generate the files, nothing will be changed on the cluster
                xl kube install --dry-run
                # Command to start apply of existing files from the previous installation run that has date 20220824-153907 (check the list of the answer files in the digitalai directory to define reference).
        # Use --ref-install for the cases when you did dry-run, manually changed generated yaml files and would like that to apply to the cluster;
        # or you just would like to repeat install with already generated yaml files.
                xl kube install --ref-install 20220824-153907
                # Using additional logging for debugging purposes
                xl kube install --verbose
                # Using prepared answers file from previous runs and skipping any prompts during installation (it will skip any questions for overridden resources too).
        # Use --answers when you would like to repeat upgrade with answer file, this will generate new yaml file from the xl-op-blueprints.
                xl kube install --local-repo ./xl-op-blueprints --verbose --skip-prompts --answers ./answers.yaml

Flags:
  -a, --answers string           The file containing answers for the questions, after using answers file the new yaml will be generated from the xl-op-blueprints according to the values provided in the answersfile
  -b, --blueprint string         The folder path containing the xl-infra blueprint, relative to the active repository
  -C, --clean-before             Before install do clean of the resources from the cluster
  -D, --dry-run                  Create files only, nothing will be applied on the cluster. You can apply generated files on the cluster by using --ref-install or --ref-upgrade
  -g, --git-branch string        Use a branch from the official GIT repository instead of the default HTTP repository
  -h, --help                     help for install
  -l, --local-repo string        Provide local folder path where blueprints are located, by default a remote repository is used
  -t, --log-since-time int32     Collect logs for last defined number of minutes, default is 60 minutes. Can be used only when waiting for resources or running check command (default 60)
  -Q, --quick-setup              Quickly run setup with all default values
  -r, --ref-install string       Reference previous install by using unique part of generated answers file (it can be just part of filename that is unique in the 'digitalai' folder). The already generated files from the previous install (or dry-run), will be used during apply to the cluster. Use it in cases when you need to apply customized files to the cluster (you will first manually change generated yaml files from the previous install and after that run install with this option)
  -S, --skip-prompts             Skip confirmation prompts
  -o, --wait-for-operator uint   Wait for helm based operator to be available, shows the operator's helm based status when ready, define number of minutes to maximally wait for helm operator
  -w, --wait-for-ready uint      Wait for deployments and pods to be started, define number of minutes to maximally wait per resource

Global Flags:
      --blueprint-current-repository string   Current active blueprint repository name
      --config string                         config file (default: $HOME/.xebialabs/config.yaml)
  -q, --quiet                                 suppress all output, except for errors
  -v, --verbose                               verbose output
      --xl-deploy-authmethod string           Authentication method to access the XL Deploy server (default "http")
      --xl-deploy-password string             Password to access the XL Deploy server (default "admin")
      --xl-deploy-url string                  URL to access the XL Deploy server (default "http://localhost:4516/")
      --xl-deploy-username string             Username to access the XL Deploy server (default "admin")
      --xl-release-authmethod string          Authentication method to access the XL Release server (default "http")
      --xl-release-password string            Password to access the XL Release server (default "admin")
      --xl-release-url string                 URL to access the XL Release server (default "http://localhost:5516/")
      --xl-release-username string            Username to access the XL Release server (default "admin")

Error occurred when running command: unknown flag: --files

It seems like the flag --files it is unknown for the CLI. I’ve installed 22.3 XL CLI version.

Finally, I’ve get XL Deploy deployed but not using the flag --files as the guide says, instead I’ve used the flag --ref-install.