First sign-in

Reaching the console for the first time, and the handful of things to do straight away.

Reaching the console

Browse to the cluster’s virtual IP over HTTPS, or to the DNS name you gave as the console name during setup:

https://10.2.32.240/
https://vs.example.com/

Always use the virtual IP or console name, not an individual server’s address. The VIP follows whichever controller is active; a node address stops working the moment that node does.

Note — Writes must go to the active controller. If you reach a standby directly it answers with HTTP 409 and the address of the active node rather than accepting a change it cannot commit.

The certificate warning

The cluster generates its own certificate at first boot, so your browser does not recognise the issuer. That is expected on a new deployment. Accept it to get in, then replace the certificate with one from your own authority — see Certificates.

If you gave a console name during setup, the certificate names it. If you left it blank, the certificate names the VIP only, and browsing by any other name will warn even after you install a trusted certificate.

Signing in

Use the administrator account from first-boot setup — the same username and password you set for SSH.

The console forces a password change on first sign-in. That is not a suggestion you can skip; the account is flagged until the password is changed.

Warning — Deployments also carry a built-in admin account. Setup locks it, because its password is identical on every installation. If your deployment still has admin enabled, treat that as the first thing to fix: change its password and disable it.

The first five things to do

1. Replace the certificate

Until you do, every operator learns to click through a browser warning, which is exactly the habit you do not want. Settings → Certificates generates a CSR for your own CA. See Certificates.

2. Create real accounts

Do not run day to day as the account setup created. Give each operator their own login, so the audit log names a person. Roles are per-domain — someone can administer storage without being able to touch networking. See Users and roles.

If you have a directory, connect it instead of creating accounts by hand — see Directory integration.

3. Turn on two-factor authentication

At minimum for accounts that can change things. See Two-factor authentication.

4. Set up alerting

A cluster with no alert rules is a cluster you find out about from a user. Start with host reachability, disk capacity and storage health. See Alerts.

5. Configure backups

The control plane’s database holds every definition in the deployment. Point it at somewhere off the cluster and set a schedule. See Backup and restore.

Finding your way around

The console’s left-hand navigation groups everything by what it acts on:

Section What lives there
Dashboard Fleet health at a glance
Infrastructure Servers, availability zones, and each host’s detail
Approvals Nodes waiting to be admitted
VMs Virtual machines
Containers, Kubernetes Other workload types
Networks Uplinks, distributed switches, fabrics
Storage Pools, volumes, replicated storage, exports
Tasks Everything asynchronous the platform is doing
Alerts Rules and the events they fired
Backups Control-plane backup and restore
Software Versions and upgrades
Settings Certificates, licensing, authentication, and the rest

See A tour of the console for what each screen is for.

If the console does not come up

Sign in over SSH to any node and ask the platform directly:

Hyperion[vsnode1]> show cluster status
Hyperion[vsnode1]> show system status

show cluster status tells you whether a controller is active and which node owns the VIP. The CLI authenticates against the host itself when the control plane is unreachable, so it keeps working during exactly the outage you need it for. See The VS-HCI shell and Troubleshooting.