ImpulseVPS

Troubleshooting

The common failure modes across provider accounts, the order form, provisioning, customer access, and licensing — with the fix for each.

1. Provider accounts

Symptom	Likely cause / fix
Test connection fails	The token is mistyped, revoked, or the wrong kind — on Hetzner, a token from the separate DNS console instead of a Cloud project token. On Vultr, also check that your WHMCS server’s IP is allowed under Account → API → Access Control. Create a new token with the access listed in Provider accounts & regions, save it, and test again.
Test connection passes, but creating a server or changing a firewall fails with a permission error	Test connection only reads from the provider, so a read-only token passes it. Hetzner tokens need Read & Write and DigitalOcean tokens Read + Write. On Vultr, use the account’s Personal Access Token, or give the sub-user permission to manage servers. On Linode, check the token’s scopes — firewall changes need Firewalls read/write — and create a new token if one is missing, since Linode can’t add a scope to an existing token.
Account status is unavailable in ImpulseVPS Settings	ImpulseCore isn’t loaded. Confirm the ImpulseCore addon is active and at the required version, then reload Settings → Providers & Regions.
Setup Guide stays on “Connect a provider account”	The credentials haven’t been saved for that account yet. Save them in ImpulseCore → Operations → VPS Providers and run Test connection — a saved token isn’t assumed to work.

2. The order form

Symptom	Likely cause / fix
A location is missing	The region isn’t enabled in ImpulseCore → Projects → Regions, it has no region code for the server plan’s provider, or the server plan doesn’t allow it. Fix the region or the server plan, then click Publish choices again.
Backups, Additional Storage, or IPv6 is missing	The option is off in the server plan, or the provider doesn’t support it — Additional Storage is Vultr only. Unsupported options are left out by design.
An operating system or size is missing	It isn’t mapped for the server plan’s provider, it’s turned off, or the server plan doesn’t allow it. Refresh the catalog on the Images tab or review Server Sizes, then publish again.
Choices show a price of zero	Publishing adds new choices with a zero price. Set each price in WHMCS under Configurable Options; prices you set are kept when you publish again.

3. Provisioning

Symptom	Likely cause / fix
The order was accepted but no server exists yet	Servers are created in the background by the module’s provision task, so a new server usually takes a few minutes. If nothing happens, confirm ImpulseCore’s cron runner is firing: check ImpulseCore → Operations → Cron or the Cron jobs check under Diagnostics → Health Checks.
The order was refused, or provisioning failed	ImpulseVPS stops before anything is created at the provider when it can’t deliver exactly what was ordered. An unmapped operating system, a location without a code for the provider, a server plan with no provider account, or an inactive license is caught when WHMCS runs the module’s Create, after checkout (usually after payment): the service stays Pending and WHMCS shows you the reason. A size the provider doesn’t sell in that location, or a provider account that isn’t set up in ImpulseCore (for example, one removed after the server plan was created), is caught only when the server is created: the service is already Active in WHMCS, the client area shows that setup hit a problem, and a critical alert appears on the Dashboard. Both kinds are listed under Diagnostics → Activity. The customer is never shown the reason.
“Not a recognized OpenSSH public key”	The customer pasted something other than a public key into `SSH Public Key` — often a private key, which is always refused. Ask for the `.pub` file’s contents, such as `ssh-ed25519 AAAA…`.
A server is marked as failed after a long wait	It didn’t become ready within the provisioning timeout (15 minutes by default, under Settings → Alerts & Notifications), so the service was marked failed and an alert was raised. A server that exists at the provider is not deleted, and it is picked up automatically if it finishes booting later.
A rebuild, resize, or snapshot seems stuck	Every action has a hard deadline. Once it passes, the action is marked failed and the server is unlocked for the next one; the deadlines are under Settings → Alerts & Notifications.

4. Customer access & the client area

Symptom	Likely cause / fix
The customer locked themselves out with the firewall	Open the service under Customers and use Reset firewall to open (it asks first). It replaces their rules with an open ruleset so they can reconnect; ask them to add their rules back afterwards.
The customer lost their one-time credential	It can’t be shown again — it is deleted after the first reveal and never emailed. The customer can reinstall from the Rebuild tab with a new SSH key; rebuilding erases the server, so suggest a snapshot first (on Linode, that needs the Backups option). On DigitalOcean a reinstall keeps the keys the server was created with, so that customer needs your help instead.
The Snapshots tab says snapshots need Backups	On Linode, snapshots use the provider’s paid Backups service, and ImpulseVPS never turns it on by itself. The customer needs the Backups option on their service first. Each Linode server keeps one manual snapshot, and taking a new one replaces it.
There is no Console tab	The provider doesn’t offer a browser console (DigitalOcean and Linode), or the server plan has Console turned off. The tab is hidden, not broken.
The console page opens but never connects	The console loads its noVNC library from `modules/servers/impulsevps/console/novnc/`, including `novnc/vendor/pako/`. Your web server must serve the `.js` files there like any other static file — a hardening rule that blocks paths containing `/vendor/` stops the console from loading.
An unsuspended server didn’t come back	ImpulseVPS only marks a server active once it has actually started, so a failed start leaves the service suspended and raises an alert. If the server no longer exists at the provider, the restore fails with an alert — it never creates a new, empty server in its place.
A terminated customer paid after all	While the service is still in its grace window, use Customers → Restore from grace. It brings back the same server and cancels the scheduled deletion.

5. Licensing

Symptom	Likely cause / fix
Orders refused with “the ImpulseVPS license is not active”	Enter or renew the `IVPS-` key in ImpulseCore → Modules & Licensing → Licenses. New provisioning and customers’ control-panel actions are blocked; servers that already exist keep running.
License shows Invalid	The key is bound to the WHMCS install it was first validated from. A moved install surfaces as Invalid — update its domain under Change domain on the license in the Impulse Software portal (for a new server IP, contact support), then click Re-check now under ImpulseCore → Modules & Licensing → Licenses.
Developer limits reached	Developer keys (`IVPSDEV-`) are not sold with the module. If you have been given one, it is capped for testing: 3 servers, and 1 snapshot and 1 volume per server. Use an `IVPS-` key for production. Diagnostics → License lists the caps.

6. Still stuck?

Gather what the module recorded, then contact Impulse Software support — the Help Center at support.impulsesoftware.com, or email [email protected]:

Diagnostics → Activity — the module’s audit log, filterable by service and event.
Diagnostics → Logs — recent failed actions and provider API errors.
Diagnostics → Help — a support snapshot with secrets masked, plus version, system, and cron information.

Include the provider, the service ID, and roughly when the problem happened.