Install Spider
Prepare the cluster
- Spider uses ECK to run its Elasticsearch. The wizard's Cluster step asks which case you are in:
| Your cluster | What to do |
|---|---|
| An ECK operator already watches Spider's namespace | Choose An ECK operator already runs in this cluster... in the wizard, nothing more |
| No ECK, or a namespace-scoped one that does not watch Spider's namespace | Choose Spider installs ECK for me, then apply the CRDs once, before installing:kubectl apply --server-side -f https://download.elastic.co/downloads/eck/3.5.0/crds.yaml |
The operator manages the spider-system namespace by default. For another namespace, see the managedNamespaces key in the values file reference.
Install
The wizard's Review step prints the commands. They look like this:
helm repo add floocus https://repository.floocus.com/helm
helm repo update
helm install spider floocus/spider-analyzer --version ">=8.2.1" -f ./setup.yaml \
--namespace spider-system --create-namespace
Use the commands from your own Review step: the chart version floor moves with each release.
note
Install with only one of values.yaml or setup.yaml. The wizard's file replaces the one-click values.yaml.
What to expect
- Spider's preflight check stops the install if you chose Spider installs ECK for me and it finds a cluster-wide ECK already running in
elastic-system. Read the message: it names the option to set. - Watch the pods come up:
kubectl get pods -n spider-system
When all pods are Running or Completed, Spider is installed.
Optional, later
Nothing below is needed for a first capture. Add it when you need it, from the installation reference:
- S3 buckets, for configuration backup (TEAM and above) and for data export
- SMTP, for password reset and notification emails
- LDAP and OIDC sign-in (ENTERPRISE)
- A private certificate authority
Next, sign in.