Migrate from Linux install

StackState Self-hosted v5.1.x

Overview

This document describes how to migrate data from the Linux install of StackState to the Kubernetes or OpenShift install.

High level steps

To migrate from the Linux install to the Kubernetes or OpenShift install of StackState, the following high level steps need to be performed:

Install StackState on Kubernetes.
Migrate StackState configuration and topology data (StackGraph) from the Linux install to the Kubernetes or OpenShift install.
Migrate telemetry data (Elasticsearch) from the Linux install to the Kubernetes or OpenShift install.

Incoming data from agents (Kafka) and node synchronisation data (Zookeeper) won't be copied.

After the migration:

Run both instances of StackState side by side for a number of days to ensure that the new instance runs correctly.
Stop the Linux install for StackState.
Remove the Linux install for StackState.

Step 2 - Migrate StackState configuration and topology data (StackGraph)

Prerequisites

Before you start the migration procedure, make sure you have the following information and tools available:

Access to:
- The Linux machines running your old StackState installation.
- The cluster running your new StackState installation.
- The values.yaml file used to install your StackState installation on Kubernetes/OpenShift.
- The restore scripts that are part of the StackState Helm chart (github.com).
Tools:
- helm (helm.sh)
- mc - the MinIO client (docs.min.io)

Export StackGraph data

To export the StackGraph data, execute the regular StackState Linux backup procedures as described below.

Ensure that the StackGraph node is up and running.
Login to the StackState node as user root.
Stop the StackState service:
```
 systemctl stop stackstate.service
```

Create a directory to store the exported data:

 sudo -u stackstate mkdir -p /opt/stackstate/migration

Export the StackGraph data by creating a backup:

 sudo -u stackstate /opt/stackstate/bin/sts-standalone.sh export \
     --file /opt/stackstate/migration/sts-export.graph --graph default

Copy the file /opt/stackstate/migration/sts-export.graph to a safe location.
Start the StackState service:
```
 systemctl start stackstate.service
```

Import StackGraph data

To import the StackGraph data into the Kubernetes/OpenShift installation, the same MinIO (min.io) component that is used for the backup/restore functionality will be used.

Note that the StackState automatic Kubernetes backup functionality should not be enabled until after the migration procedure has completed.

Enable the MinIO component by adding the following YAML fragment to the values.yaml file that is used to install StackState

 backup:
   enabled: true
   stackGraph:
     scheduled:
       enabled: false
   elasticsearch:
     restore:
       enabled: false
     scheduled:
       enabled: false
 minio:
   accessKey: MINIO_ACCESS_KEY
   secretKey: MINIO_SECRET_KEY
   persistence:
     enabled: true

Include the credentials to access the MinIO instance:

Replace MINIO_ACCESS_KEY with 5 to 20 alphanumerical characters.
Replace MINIO_SECRET_KEY with 8 to 40 alphanumerical characters.

The Helm values `backup.stackGraph.scheduled.enabled`, `backup.elasticsearch.restore.enabled` and `backup.elasticsearch.scheduled.enabled` have been set to `false` to prevent scheduled backups from overwriting the backups that we will upload to MinIO.

Run the appropriate helm upgrade command for your installation to enable MinIO.
Start a port-forward to the MinIO service in your StackState instance:
```
 kubectl port-forward service/stackstate-minio 9000:9000
```

In a new terminal window, configure the MinIO client to connect to that MinIO service:

 mc alias set minio-backup http://localhost:9000 ke9Dm7eFhk9kP53rXlUI mNOWCpoYrhwati7QcOrEwnI7Mtcf0jxg2JzNOMk6

Verify that access has been configured correctly:
```
 mc ls minio-backup
```
The output should be empty, as we haven't created any buckets yet.
If the output isn't empty, the automatic backup functionality has been enabled. Disable the automatic backup functionality and configure MinIO as described above (i.e. not as a gateway to AWS S3 or Azure Blob Storage and without any local storage).

Create the bucket that is used to store StackGraph buckets:

 mc mb minio-backup/sts-stackgraph-backup

The output should look like this:

 Bucket created successfully `minio-backup/sts-stackgraph-backup`.

Upload the backup file created in the previous step when StackGraph data was exported from the Linux install:

 mc cp sts-export.graph minio-backup/sts-stackgraph-backup/

The output should look like this:

 sts-export.graph:                             15.22 KiB / 15.22 KiB  ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓  42.61 KiB/s 0s

Verify that the backup file was uploaded to the correct location:

 ./restore/list-stackgraph-backups.sh

The output should look like this:

 job.batch/stackgraph-list-backups-20210222t122522 created
 Waiting for job to start...
 Waiting for job to start...
 === Listing StackGraph backups in bucket "sts-stackgraph-backup"...
 sts-export.graph
 ===
 job.batch "stackgraph-list-backups-20210222t122522" deleted

Most importantly, the backup file uploaded in the previous step should be listed here.

Restore the backup:

 ./restore/restore-stackgraph-backup.sh sts-export.graph

The output should look like this:

 job.batch/stackgraph-restore-20210222t171035 created
 Waiting for job to start...
 Waiting for job to start...
 === Downloading StackGraph backup "sts-export.graph" from bucket "sts-stackgraph-backup"...
 download: s3://sts-stackgraph-backup/sts-export.graph to ../../tmp/sts-export.graph
 === Importing StackGraph data from "sts-export.graph"...
 WARNING: An illegal reflective access operation has occurred
 WARNING: Illegal reflective access by org.codehaus.groovy.vmplugin.v7.Java7$1 (file:/opt/docker/lib/org.codehaus.groovy.groovy-2.5.4.jar) to constructor java.lang.invoke.MethodHandles$Lookup(java.lang.Class,int)
 WARNING: Please consider reporting this to the maintainers of org.codehaus.groovy.vmplugin.v7.Java7$1
 WARNING: Use --illegal-access=warn to enable warnings of further illegal reflective access operations
 WARNING: All illegal access operations will be denied in a future release
 ===
 job.batch "stackgraph-restore-20210222t171035" deleted

Remove the YAML snippet added in step 1 and run the appropriate helm upgrade command for your installation to disable MinIO.

Step 3 - Migrate telemetry data (Elasticsearch)

To migrate Elasticsearch data from the Linux install to the Kubernetes or OpenShift install, use the functionality reindex from remote (elastic.co).

Notes:

To access the Elasticsearch instance that runs as part of the Kubernetes or OpenShift installation for StackState, execute the following command:
```
  kubectl port-forward service/stackstate-elasticsearch-master 9200:9200
```
and access it on http://localhost:9200.

To modify the elasticsearch.yml configuration file, use the Helm chart value stackstate.elasticsearch.esConfig.

For example:

  stackstate:
      elasticsearch:
          esConfig:
              elasticsearch.yml: |
                  reindex.remote.whitelist: oldhost:9200

Overview

High level steps

Step 2 - Migrate StackState configuration and topology data (StackGraph)

Prerequisites

Export StackGraph data

Import StackGraph data

Step 3 - Migrate telemetry data (Elasticsearch)

See also