Documentation — Host.com

Getting Started

Welcome to the Host.com developer documentation. This guide covers everything from installing the CLI to deploying your first production server. You'll need a Host.com account — if you don't have one, sign up free in under two minutes.

Install the CLI

The host CLI is the fastest way to manage your infrastructure. Install it globally via npm:

TERMINAL

npm install -g @hostcom/cli

Alternatively, download platform-specific binaries from the Downloads page — no Node.js required.

Tip: Requires Node.js 18 or later. Run node --version to check. We recommend using Volta or nvm to manage Node versions.

Authenticate

After installation, authenticate the CLI with your Host.com account. This opens a browser window for OAuth, then stores a token securely at ~/.hostcom/config.

TERMINAL

host auth login
# Opening https://host.com/cli/auth?token=... in your browser
# Waiting for authentication...
# ✓ Authenticated as [email protected]
# Token saved to ~/.hostcom/config

For non-interactive environments (CI/CD pipelines), set the HOSTCOM_API_KEY environment variable instead of using browser-based auth. The CLI will pick it up automatically.

Verify Your Setup

Confirm the CLI is authenticated and your account details are correct:

TERMINAL

host account info

JSON RESPONSE

{
  "id": "usr_4j2kx9mf",
  "email": "[email protected]",
  "name": "Alex Chen",
  "plan": "pro",
  "balance_usd": 42.50,
  "monthly_limit_usd": 500.00,
  "2fa_enabled": true,
  "created_at": "2024-03-15T09:12:00Z"
}

Deploy Your First Server

Create a server with a single command. The --ssh-key flag references a key already added to your account (see SSH & Server Access):

TERMINAL

host server create \
  --name    web-01 \
  --size    cx-21 \
  --region  nyc \
  --os      ubuntu-22-04 \
  --ssh-key my-laptop

The command returns immediately with a provisioning response. The server will be reachable within 45–90 seconds:

JSON RESPONSE

{
  "id":         "srv_8x2nqprt",
  "name":       "web-01",
  "status":     "provisioning",
  "ip_address": "198.51.100.42",
  "region":     "nyc",
  "size":       "cx-21",
  "vcpus":      2,
  "memory_mb":  4096,
  "disk_gb":    80,
  "os":         "ubuntu-22-04",
  "created_at": "2026-04-21T14:30:00Z"
}

Poll status with host server get web-01 until status reads running, then SSH in using the returned IP address.

Core Concepts

Understanding Host.com's resource model, regions, and server lifecycle will help you architect reliable, cost-efficient deployments.

Regions

Choose the region closest to your users to minimize latency. All regions support the full feature set unless noted.

Code	Location	Continent	IPv6
`nyc`	New York, USA	North America	Yes
`lax`	Los Angeles, USA	North America	Yes
`dfw`	Dallas, USA	North America	Yes
`mia`	Miami, USA	North America	Yes
`fra`	Frankfurt, Germany	Europe	Yes
`ams`	Amsterdam, Netherlands	Europe	Yes
`lon`	London, UK	Europe	Yes
`sto`	Stockholm, Sweden	Europe	Yes
`sin`	Singapore	Asia-Pacific	Yes
`tyo`	Tokyo, Japan	Asia-Pacific	Yes
`syd`	Sydney, Australia	Asia-Pacific	Yes
`bom`	Mumbai, India	Asia-Pacific	Yes

Server Lifecycle

Every server moves through a well-defined set of states. Understanding these states is critical for handling automation correctly:

provisioning → running ⇄ stopped → archived

provisioning — Server is being allocated and the OS is being installed. Typically 45–90 seconds. No actions available.
running — Server is live. Can be rebooted, resized (with downtime), snapshotted, or stopped.
stopped — Server is powered off. Disk is preserved; IP address is held. You are still billed for disk and IP reservation.
archived — Entered after explicit deletion. All data is permanently destroyed after 7 days. Billing stops immediately.

Warning: Archived servers are unrecoverable after 7 days. Take a snapshot before deleting if you may need the data later.

Resource Model

Each server plan is defined by four resources:

vCPU — Dedicated virtual CPU cores (AMD EPYC or Intel Xeon, depending on region). No sharing — your cores are yours.
RAM — ECC DDR5 memory allocated exclusively to your server instance.
NVMe Storage — Local NVMe SSD. Fast sequential and random I/O. Not replicated — use block volumes or snapshots for durability.
Bandwidth Pool — Monthly outbound transfer allowance. Inbound is always free. Overages billed at $0.01/GB. Unused bandwidth does not roll over.

Billing Model

Host.com uses an hourly accrual model. The monthly price shown is a cap — you never pay more than that in a calendar month, regardless of how many hours a server runs.

Scenario	Behavior
Server runs full month	Billed exactly the monthly price (hourly rate × 730 hrs)
Server created mid-month	Billed hourly from creation to end of billing period
Resize up mid-month	New hourly rate applies from resize time; prorated on invoice
Resize down mid-month	Credit issued for unused portion at old rate
Stopped server	Full hourly rate continues (disk + IP still reserved)

REST API Reference

Base URL: https://api.host.com/v2 — All requests must be made over HTTPS. The API follows REST conventions and returns JSON for all responses including errors.

Authentication

Every API request must include a Bearer token in the Authorization header. Generate API keys in your dashboard under Account → API Keys.

HTTP HEADER

Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

Rate Limits

The API enforces a limit of 3,000 requests per hour per API key, measured in a rolling window. When exceeded, the API returns HTTP 429 with a Retry-After header indicating how many seconds to wait before retrying.

RATE LIMIT RESPONSE HEADERS
HTTP/2 429 Too Many Requests
X-RateLimit-Limit:     3000
X-RateLimit-Remaining: 0
X-RateLimit-Reset:     1745244600
Retry-After:           847

Endpoints

Method	Endpoint	Description
GET	`/servers`	List all servers
POST	`/servers`	Create a new server
GET	`/servers/{id}`	Get a single server
PATCH	`/servers/{id}`	Update server metadata (name, tags)
DELETE	`/servers/{id}`	Delete (archive) a server
POST	`/servers/{id}/actions/reboot`	Reboot a running server
POST	`/servers/{id}/actions/resize`	Resize server to new plan (requires stop)
GET	`/regions`	List available regions
GET	`/sizes`	List available server sizes & pricing
GET	`/images`	List OS images and snapshots
GET	`/ssh-keys`	List SSH keys on account
POST	`/ssh-keys`	Add an SSH key
DELETE	`/ssh-keys/{id}`	Remove an SSH key
GET	`/volumes`	List block volumes
POST	`/volumes`	Create a block volume
POST	`/volumes/{id}/actions/attach`	Attach volume to a server

Full Example: Create a Server

A complete request-response cycle creating a server via curl:

REQUEST

curl -X POST https://api.host.com/v2/servers \
  -H "Authorization: Bearer hk_live_aBcDeFgHiJkLmN0pQrStUvWxYz" \
  -H "Content-Type: application/json" \
  -d '{
    "name":    "web-01",
    "size":    "cx-21",
    "region":  "nyc",
    "image":   "ubuntu-22-04",
    "ssh_keys": ["key_7f3ab21c"],
    "tags":    ["production", "web"]
  }'

RESPONSE — 201 Created

{
  "server": {
    "id":         "srv_8x2nqprt",
    "name":       "web-01",
    "status":     "provisioning",
    "ip_address": "198.51.100.42",
    "ipv6_address": "2001:db8:85a3::42",
    "region":     "nyc",
    "size":       "cx-21",
    "vcpus":      2,
    "memory_mb":  4096,
    "disk_gb":    80,
    "bandwidth_gb": 3000,
    "tags":       ["production", "web"],
    "created_at": "2026-04-21T14:30:00Z"
  }
}

Pagination

List endpoints return paginated results. Use the page and per_page query parameters to navigate. Default page size is 25; maximum is 200.

PAGINATED REQUEST

GET /servers?page=2&per_page=25

RESPONSE META
{
  "servers": [ /* … */ ],
  "meta": {
    "total":    83,
    "page":     2,
    "per_page": 25,
    "pages":    4
  }
}

SSH & Server Access

SSH key-based authentication is the recommended way to access your servers. Password authentication is disabled by default on all Host.com servers.

Generate an SSH Key Pair

We recommend Ed25519 keys for their security and performance. Run the following to generate a dedicated key pair for Host.com:

TERMINAL

ssh-keygen -t ed25519 -C "[email protected]" -f ~/.ssh/hostcom
# Generating public/private ed25519 key pair.
# Enter passphrase (empty for no passphrase): ••••••••
# Your identification has been saved in /home/user/.ssh/hostcom
# Your public key has been saved in /home/user/.ssh/hostcom.pub

Add Key to Your Account

Upload the public key to Host.com so it can be injected into new servers at creation time:

TERMINAL

host ssh-key add \
  --name       "Laptop" \
  --public-key ~/.ssh/hostcom.pub
# ✓ SSH key "Laptop" added (key_7f3ab21c)

Connect to Your Server

TERMINAL

ssh -i ~/.ssh/hostcom root@198.51.100.42

For convenience, add an entry to ~/.ssh/config:

~/.ssh/config

Host web-01
  HostName     198.51.100.42
  User         root
  IdentityFile ~/.ssh/hostcom
  IdentitiesOnly yes

Then simply run ssh web-01.

Recommended sshd_config Hardening

Apply these settings on fresh servers to reduce attack surface. Edit /etc/ssh/sshd_config and reload with systemctl reload sshd:

/etc/ssh/sshd_config

# Disable password auth — keys only
PasswordAuthentication    no
ChallengeResponseAuthentication no

# Allow root login with a key but not password
PermitRootLogin           prohibit-password

# Reduce brute-force window
MaxAuthTries              3
LoginGraceTime            20s

# Keep idle sessions alive (avoid NAT timeouts)
ClientAliveInterval       300
ClientAliveCountMax       2

# Restrict to IPv4 if IPv6 not needed
AddressFamily             inet

VNC Console Fallback: If you lose SSH access (misconfigured firewall, broken sshd config), access your server via the browser-based VNC console: Dashboard → Servers → [your server] → Console tab. This connects directly to the serial port regardless of network state.

Networking

Host.com provides software-defined networking: private VPCs for internal traffic, floating IPs for failover, hardware firewalls, load balancers, and managed DNS — all configurable via CLI or API.

Private VPC

Create an isolated private network for inter-server communication without exposing traffic to the public internet:

TERMINAL

# Create the VPC
host network create \
  --name   prod-vpc \
  --cidr   10.0.0.0/16 \
  --region nyc
# ✓ Network "prod-vpc" created (net_3fa1c7b2)

# Attach an existing server to the VPC
host network attach \
  --network prod-vpc \
  --server  web-01 \
  --ip      10.0.1.10

Once attached, servers communicate over the private interface (eth1 on Ubuntu) using their private IPs — free of charge, with no bandwidth counting against your pool.

Floating IPs

A floating IP is a static public IP you can reassign between servers in milliseconds — ideal for zero-downtime failover:

TERMINAL

# Reserve a floating IP
host floating-ip create --region nyc
# ✓ Floating IP 203.0.113.100 reserved (fip_9c4de1f0)

# Assign to a server
host floating-ip assign fip_9c4de1f0 --server web-01

# Reassign to another server (failover)
host floating-ip assign fip_9c4de1f0 --server web-02

Firewall Rules

Host.com firewalls are enforced at the hypervisor level — they block traffic before it reaches your server's NIC, protecting against malformed packets and volumetric attacks:

TERMINAL

# Create a firewall group
host firewall create --name web-fw

# Allow HTTP and HTTPS from anywhere
host firewall rule add web-fw \
  --direction inbound --proto tcp \
  --port 80,443 --source 0.0.0.0/0

# Allow SSH only from your office CIDR
host firewall rule add web-fw \
  --direction inbound --proto tcp \
  --port 22 --source 203.0.113.0/24

# Apply the firewall group to a server
host firewall apply web-fw --server web-01

Load Balancer

TERMINAL

# Create a load balancer
host lb create \
  --name      prod-lb \
  --region    nyc \
  --algorithm round-robin

# Add backend targets on port 8080
host lb target add prod-lb \
  --server web-01 --port 8080

host lb target add prod-lb \
  --server web-02 --port 8080

# Configure health check
host lb health-check prod-lb \
  --path     /health \
  --interval 10 \
  --timeout  3 \
  --unhealthy-threshold 2

DNS Management

TERMINAL

# Create an A record pointing the apex to your server
host dns record create \
  --zone  example.com \
  --type  A \
  --name  @ \
  --value 198.51.100.42 \
  --ttl   300

# Create a CNAME for www
host dns record create \
  --zone  example.com \
  --type  CNAME \
  --name  www \
  --value example.com.

Security

Two-Factor Authentication

Enable TOTP-based 2FA on your account to require a one-time code on every login:

Go to Account → Security → Two-Factor Authentication
Click Enable 2FA. A QR code is displayed.
Scan the QR code with an authenticator app (Authy, Google Authenticator, 1Password).
Enter the 6-digit code to confirm. Save your backup codes in a secure location — they cannot be retrieved later.
2FA is now active. CLI authentication via host auth login will prompt for the TOTP code.

API Key Scopes

Scope	HTTP Methods Allowed	Use Case
`read`	GET only	Monitoring, dashboards, read-only CI checks
`read-write`	GET, POST, PATCH	Provisioning servers, managing DNS, deploying
`admin`	All methods incl. DELETE + billing	Full account automation, Terraform, billing integrations

IP Allowlisting

Restrict an API key to only accept requests from specific CIDR ranges — ideal for locking CI/CD runners to known IPs:

TERMINAL

host api-key update key-abc \
  --allowed-ips 203.0.113.0/24,10.0.0.0/8

Never commit API keys to git. Use environment variables or a secrets manager. If a key is exposed, rotate it immediately from the dashboard — old keys are invalidated within 60 seconds.

DDoS Protection Tiers

Tier	Price	Mitigation Capacity	Mitigation Latency
Standard	Included	100 Gbps	< 30s detection & null-route
Enhanced	+$10/mo per server	1 Tbps	< 10s scrubbing
Premium	+$35/mo per server	5 Tbps	< 10ms always-on scrubbing

Hardened Web Server Firewall Example

TERMINAL

# Allow public web traffic
host firewall rule add web-fw --direction inbound --proto tcp --port 80,443 --source 0.0.0.0/0

# Allow SSH only from the bastion host
host firewall rule add web-fw --direction inbound --proto tcp --port 22 --source 10.0.0.5/32

# Deny everything else inbound (implicit deny is default)
# Outbound is unrestricted by default
host firewall apply web-fw --server web-01 --server web-02

Storage & Backups

Block Volumes

Block volumes are persistent NVMe SSDs that exist independently of any server. They survive server deletion and can be moved between servers in the same region.

TERMINAL — Create & Attach Volume

# Create a 100 GB block volume
host volume create \
  --name   data-vol \
  --size   100 \
  --region nyc
# ✓ Volume "data-vol" created (vol_5e8af3c1)

# Attach to a server (appears as /dev/sdb)
host volume attach vol_5e8af3c1 --server web-01

After attaching, SSH into your server and format/mount the volume:

SERVER SHELL

# Format with ext4 (first time only — destroys existing data)
mkfs.ext4 /dev/sdb

# Create mount point and mount
mkdir -p /mnt/data
mount /dev/sdb /mnt/data

# Get UUID for fstab (survives device renaming)
blkid /dev/sdb
# /dev/sdb: UUID="a1b2c3d4-..." TYPE="ext4"

Add to /etc/fstab for automatic remount on reboot:

/etc/fstab entry

UUID=a1b2c3d4-e5f6-7890-abcd-ef1234567890  /mnt/data  ext4  defaults,nofail  0  2

Tip: The nofail fstab option prevents boot failures if the volume is temporarily detached. Always use it on cloud volumes.

Object Storage

For unstructured files, static assets, backups, and logs at scale, use Host.com Object Storage — S3-compatible, 99.999999999% durability, no egress fees within our CDN network. See the Object Storage documentation for setup, SDK integration, and bucket policies.

Snapshots

TERMINAL

# Create a snapshot of a server (server can be running)
host snapshot create --server web-01 --name pre-deploy-2026-04-21

# List snapshots
host snapshot list

# Restore: create a new server from snapshot
host server create \
  --name   web-01-restored \
  --image  snap_7b3ef19a \
  --size   cx-21 \
  --region nyc

Automated Backups

Enable daily automated backups on any server. Backups are crash-consistent snapshots taken during a low-traffic window each night:

TERMINAL

host server update web-01 --backups enabled
# ✓ Automated backups enabled for web-01
# Additional cost: 20% of base server price per month

Backup Type	Frequency	Retention
Daily backups	Every 24 hours	7 most recent kept
Weekly backups	Every Sunday 02:00 UTC	4 most recent kept

Scaling

Vertical Scaling (Resize)

Increase CPU and RAM by resizing to a larger plan. A resize requires a brief reboot (~60 seconds of downtime). The server's IP address, disk contents, and configuration are preserved.

TERMINAL

# Stop the server first (resize requires stopped state)
host server stop web-01

# Resize from cx-21 (2 vCPU / 4 GB) to cx-41 (4 vCPU / 8 GB)
host server resize --server web-01 --size cx-41

# Start the server again
host server start web-01
# ✓ Server web-01 resized to cx-41 and started (~60s)

Disk size can only be increased, never decreased. If you need to downsize disk, migrate to a new server from a snapshot.

Horizontal Scaling

For zero-downtime scaling, create identical servers from a snapshot and distribute traffic via a load balancer:

TERMINAL

# 1. Take a snapshot of the current production server
host snapshot create --server web-01 --name web-golden

# 2. Create identical server from snapshot
host server create \
  --name   web-02 \
  --image  snap_web-golden \
  --size   cx-21 \
  --region nyc

# 3. Add new server to existing load balancer
host lb target add prod-lb --server web-02 --port 8080

Auto-Scaling Configuration

Define auto-scaling policies as YAML. The controller monitors metrics and adds/removes servers within your defined bounds:

autoscale.yml

name: web-autoscale
load_balancer: prod-lb
template:
  image:   snap_web-golden
  size:    cx-21
  region: nyc
  ssh_keys: [key_7f3ab21c]
scaling:
  min: 2
  max: 10
  cpu_threshold: 75     # scale up when avg CPU > 75%
  scale_down_threshold: 30 # scale down when avg CPU < 30%
  cooldown_seconds: 300  # wait 5 min between scaling events
  warmup_seconds: 60     # time for new instance to become healthy
health_check:
  path:     /health
  interval: 10
  timeout:  3
  threshold: 2

TERMINAL — Apply Policy

host autoscale apply -f autoscale.yml
# ✓ Auto-scaling policy "web-autoscale" applied

Monitoring & Alerts

Built-in Metrics

Every server emits metrics automatically — no agent required. Metrics are collected at 30-second resolution and retained for 30 days.

Metric	Unit	Description
`cpu_usage`	Percent (0–100)	Total CPU utilization across all vCPUs
`memory_used`	MiB	Used RAM (includes buffers/cache)
`disk_io_read`	KB/s	Disk read throughput
`disk_io_write`	KB/s	Disk write throughput
`network_in`	Mbps	Inbound network throughput
`network_out`	Mbps	Outbound network throughput
`disk_used`	GiB	Disk space consumed on root volume

Create Alerts

TERMINAL

# Alert when CPU exceeds 80% for 5 minutes
host alert create \
  --server    web-01 \
  --metric    cpu_usage \
  --threshold 80 \
  --duration  5m \
  --notify    email,slack

# Alert when disk usage exceeds 85%
host alert create \
  --server    web-01 \
  --metric    disk_used_pct \
  --threshold 85 \
  --notify    email

Webhook Alert Payload

Configure a webhook URL to receive alert notifications as JSON POST requests:

WEBHOOK PAYLOAD (POST application/json)
{
  "event":      "alert.triggered",
  "alert_id":   "alr_2c9fe8b1",
  "server_id":  "srv_8x2nqprt",
  "server_name":"web-01",
  "metric":     "cpu_usage",
  "value":      87.4,
  "threshold":  80,
  "state":      "firing",
  "region":     "nyc",
  "timestamp":  "2026-04-21T14:32:00Z",
  "dashboard_url": "https://host.com/dashboard/servers/srv_8x2nqprt"
}

Uptime Checks

Uptime checks probe your endpoints from 5 global nodes every 30 seconds. An alert fires after 2 consecutive failures (i.e., 60 seconds of confirmed outage) to reduce false positives:

TERMINAL

# HTTP check
host uptime create \
  --name    prod-web \
  --type    http \
  --url     https://example.com/health \
  --expect  200 \
  --notify  email,slack

# TCP port check (e.g. database port)
host uptime create \
  --name    db-port \
  --type    tcp \
  --host    10.0.1.20 \
  --port    5432

External Integrations

Forward metrics and logs to your existing observability stack:

Syslog forwarding: Configure rsyslog or journald to forward to your SIEM. Use the server's private VPC IP for zero-cost log shipping.
Datadog: Install the Datadog agent and set DD_API_KEY. Host.com metrics appear automatically under the host.com.* namespace.
Grafana Cloud: Use the Prometheus remote_write endpoint at https://api.host.com/v2/metrics/prometheus with your API key as Bearer token.

Troubleshooting

API Error Codes

HTTP Status	Code	Cause & Resolution
`400`	bad_request	Malformed JSON or invalid field value. Check the `errors` array in the response for field-level details.
`401`	unauthorized	Missing or invalid API key. Verify your `Authorization` header and that the key is not revoked.
`403`	forbidden	API key scope is insufficient. Upgrade key to `read-write` or `admin` as needed.
`404`	not_found	Resource ID does not exist or belongs to another account. Double-check the resource ID.
`409`	conflict	Server is not in the required state (e.g., trying to resize a running server). Check server state and retry.
`422`	unprocessable	Validation error — a required field is missing or a value is out of range. Inspect the `errors` field.
`429`	rate_limited	Rate limit exceeded. Wait the number of seconds in the `Retry-After` header before retrying.
`500`	internal_error	Unexpected server error. Retry with exponential backoff. If it persists, open a support ticket with the `request_id` from the response.

Rescue Mode

If your server is unbootable due to a misconfigured bootloader, corrupted filesystem, or bad kernel parameter, boot into rescue mode — a minimal recovery OS that mounts your disk at /mnt/rescue:

TERMINAL

# Enable rescue mode (reboots into recovery OS)
host server action --server web-01 --action enable-rescue
# ✓ Server is booting into rescue mode…
# Access via VNC console or SSH (temporary password shown in dashboard)

# After fixing the issue, exit rescue mode
host server action --server web-01 --action disable-rescue

VNC Console

Access your server's virtual console directly from the browser — no SSH needed, no network configuration required. Navigate to Dashboard → Servers → [server name] → Console tab. The console connects to the serial port and shows output from boot-time through to the login prompt. Useful for diagnosing fsck failures, kernel panics, and misconfigured network interfaces.

Common Issues

Can't SSH into server

Verify the server is in running state: host server get web-01
Check that port 22 is open in your firewall rules: host firewall list web-fw
Confirm the correct private key: ssh -i ~/.ssh/hostcom -v root@<IP> (look for "Offering public key")
Verify the key fingerprint matches what was uploaded to your account
If firewall and key are correct, use VNC console to check sshd status: systemctl status sshd

Server stuck in "provisioning" state

Provisioning normally completes within 45–90 seconds.
If the server has not transitioned to running after 10 minutes, contact support and include the server ID.
Do not delete and recreate repeatedly — open a ticket so the engineering team can investigate the underlying host.

High CPU usage

SSH in and run top or htop to identify the offending process.
Check for runaway cron jobs: crontab -l and cat /var/log/syslog | grep cron
If legitimate load, consider a vertical resize (cx-21 → cx-41) or adding a second server behind the load balancer.
Enable CPU alerts to get notified proactively: host alert create --server web-01 --metric cpu_usage --threshold 80

Open a Support Ticket

For issues that can't be resolved with the documentation, open a support ticket directly from the CLI. Include as much context as possible — server ID, timestamps, and error messages reduce response time significantly.

TERMINAL

host support ticket create \
  --server   web-01 \
  --priority high \
  --message  "Server stuck in provisioning since 14:30 UTC.
Request ID from API: req_abc123. No SSH access."
# ✓ Ticket #48291 created — median response time: 8 minutes

Pro tip: Priority levels are low, normal, high, and critical. Use critical only for production outages — it pages the on-call engineer immediately.

Monitoring & Alerts Guides & Tutorials

Developer Documentation

Getting Started

Install the CLI

Authenticate

Verify Your Setup

Deploy Your First Server

Core Concepts

Regions

Server Lifecycle

Resource Model

Billing Model

REST API Reference

Authentication

Rate Limits

Endpoints

Full Example: Create a Server

Pagination

SSH & Server Access

Generate an SSH Key Pair

Add Key to Your Account

Connect to Your Server

Recommended sshd_config Hardening

Networking

Private VPC

Floating IPs

Firewall Rules

Load Balancer

DNS Management

Security

Two-Factor Authentication

API Key Scopes

IP Allowlisting

DDoS Protection Tiers

Hardened Web Server Firewall Example

Storage & Backups

Block Volumes

Object Storage

Snapshots

Automated Backups

Scaling

Vertical Scaling (Resize)

Horizontal Scaling

Auto-Scaling Configuration

Monitoring & Alerts

Built-in Metrics

Create Alerts

Webhook Alert Payload

Uptime Checks

External Integrations

Troubleshooting

API Error Codes

Rescue Mode

VNC Console

Common Issues

Can't SSH into server

Server stuck in "provisioning" state

High CPU usage

Open a Support Ticket