Post-Installation Steps

After a successful Lightbits cluster installation, perform the following steps.

Back Up Important Content

Back up the following contents to a secure location. This will be useful if another node is added in the future or other troubleshooting is required.

Back up the ~/light-app directory and all of its contents. The contents of the Ansible directory will be helpful in the future if there is a need to add servers or check over how a previous installation was done.
Back up the generated JWTs. Back up the system-jwt and the default project jwt. By default they are placed in the home directory: ~/lightos-system-jwt and ~/lightos-default-admin-jwt.
Back up the generated certificates. By default these are placed in ~/lightos-certificates.

Check Cluster Health

Copy the contents of the system jwt file ~/lightos-system-jwt into the clipboard.

From the Ansible host, run cat ~/lightos-system-jwt; echo;. Note that we add the last echo to generate a new line at the end, so that it is easy to determine where the JWT ends.

The contents will be similar to this:

export LIGHTOS_JWT=eyJhbGciOiJSUzI1Ni<...CONTENTS OF JWT...>5PcYPBRBaFEuMsT9gQNQA

Note

The contents of a JWT are long and all are on a single line.

Log in to any Lightbits server and paste the contents into the shell. The JWT will now be available via the $LIGHTOS_JWT environment variable.
Check the state of the servers, nodes, and cluster.

The servers look healthy as they all state "NoRiskOfServiceLoss":

server00:~$ lbcli -J $LIGHTOS_JWT list servers

NAME UUID State RiskOfServiceLoss State LightOSVersion

server00 5c2b7375-64fa-583e-8ebe-e82b8e1e1a53 Enabled NoRiskOfServiceLoss 3.1.2~b1125

server01 bb5433d2-9740-5130-a5eb-47f623c19b4d Enabled NoRiskOfServiceLoss 3.1.2~b1125

server02 a9097bf0-6005-5dca-8f07-eeaca114ec70 Enabled NoRiskOfServiceLoss 3.1.2~b1125

The nodes also look healthy, as they all state to be "Active":

server00:~$ lbcli -J $LIGHTOS_JWT list nodes

Name UUID State NVMe endpoint Failure domains Local rebuild progress

server00-0 d2c336f2-c9e5-5a4c-951e-ede739d10774 Active 10.10.10.100:4420 [server00] None

server01-0 810ed593-a97b-5495-b08a-be5e37a65f82 Active 10.10.10.101:4420 [server01] None

server02-0 2cf6ce67-fe5e-5a98-86bf-98a5792a8916 Active 10.10.10.102:4420 [server02] None

Also check that the cluster health state is ok:

server00:~$ lbcli -J $LIGHTOS_JWT get cluster -o yaml

ETag: "0"

MaxAllowedVersion: 3.2.X

MinAllowedVersion: 3.1.X

MinVersionInCluster: 3.1.2~b1125

UUID: c2719be6-00b8-4e96-b80a-6a84ecc3e638

apiEndpoints:

- 10.10.10.100:443

- 192.168.16.22:443

- 10.10.10.101:443

- 192.168.16.92:443

- 10.10.10.102:443

- 192.168.16.32:443

clusterName: c2719be6-00b8-4e96-b80a-6a84ecc3e638

currentMaxReplicas: 3

discoveryEndpoints:

- 10.10.10.100:8009

- 10.10.10.101:8009

- 10.10.10.102:8009

health:

numDegradedVolumes: 0

numInactiveNodes: 0

numNotAvailableVolumes: 0

numReadOnlyVolumes: 0

state: OK <------------------------------------------ health state is OK

statistics:

compressionRatio: 1

effectivePhysicalStorage: "52311333273"

At this point the cluster's health has been confirmed at the node, server, and cluster level.

Last updated on

Was this page helpful?