Skip to content

Troubleshooting

This content is not available in your language yet.

  1. Is it running? Its page (or tebib vm show <VM ID>) shows the power state.
  2. Is port 22 open to you? New VMs deny all inbound traffic. Check the VM’s security groups for a TCP 22 rule whose source includes your current public address. Your home or mobile address can change.
  3. Are you using an address that reaches it? Over IPv4 the VM needs a public IPv4 address (and its VPC a NAT gateway). Over IPv6 your own network needs IPv6: test with curl -6 https://ifconfig.co.
  4. Right user and key? The user depends on the image (ubuntu, debian, rocky, almalinux). Keys are installed only at the VM’s first boot.
  5. Still booting? First boot installs updates and the guest agent; give it a few minutes.
  6. Look at the serial console for boot errors.

The VM’s page shows why (stop_reason):

Reason What happened What to do
guest It was shut down from inside (for example sudo poweroff) Start it again
balance The balance ran out Top up, then start it
abuse TebibHost stopped it after detecting abuse; the page shows why Contact abuse@tebibhost.com; it can’t be started while held
suspended The organization is suspended Contact TebibHost
Error Meaning
409 Insufficient balance Runway is under 24 hours (or there’s no money yet). Top up
409 Quota exceeded A tier limit or project quota would be passed. See Quotas and tiers
400 instance type too small for the image Rocky Linux and AlmaLinux need 1.5 GiB of memory, Ubuntu 1 GiB
409 VPC has no NAT gateway Public IPs reach VMs through the VPC’s NAT gateway; create one first
409 already exists Names are unique within a project

Resize or restore says “stop the VM first”

Section titled “Resize or restore says “stop the VM first””

Changing the instance type and restoring a snapshot need the VM stopped. Stop it, wait until it shows stopped, then try again.

Attached volumes are hot-plugged as SCSI disks. List them with ls -l /dev/disk/by-id/ | grep vol- and lsblk. A volume shows its new size after a resize, but you must grow the filesystem yourself. See Volumes.

  • Pending: TebibHost checks with M-Pesa after 90 seconds and keeps checking for 24 hours. A payment that succeeds late is still credited.
  • Failed, cancelled on the phone: the prompt was dismissed; nothing was charged.
  • Paid but not credited after an hour: contact TebibHost with the M-Pesa confirmation code. Payments that can’t be matched automatically are held, never lost.

Errors are JSON with a code, a message and a request_id. Quote the request_id when you contact TebibHost.

Status Meaning
400 The request is invalid; the message names the field
401 Not signed in, or the access key is wrong, expired or turned off
403 Not allowed; the message names the action and the resource. Use the access checker
404 Not found, or not in an organization you can see
409 Conflicts with the current state (being deleted, name taken, balance, quota)
429 Too many requests; back off and retry

Creates and deletes answer 202 Accepted: the work continues in the background. Poll the resource (the CLI’s --wait does this) until it settles.