homebutler alerts

Reads the CPU, memory, and disk thresholds once, or keeps checking them with --watch.

The thresholds come from alerts.cpu, alerts.memory, and alerts.disk in your config. Rules for self-healing come from an alerts config file.

To choose thresholds and rules, see Alerts.

Command	What it does
history	Lists recent alert and remediation events.
init	Asks a few questions and writes a starter config.yaml with alert thresholds and rules.
test-notify	Sends a test notification to every configured provider.

alerts

Run without a subcommand, alerts prints each reading with its threshold and a status. It exits 0 even when a reading is over its threshold.

homebutler alerts [flags]
homebutler alerts [command]

Flag	Type	Default	Description
--alerts-config	string	-	Path to a YAML file with self-healing rules. Without it, --watch reads ~/.homebutler/alerts.yaml when that file exists.
--interval	string	30s	Time between checks in --watch mode. Accepts a Go duration such as 30s or 1m.
--watch	boolean	false	Keep checking until you press Ctrl+C. This mode looks at resources only. watch start also watches containers and services.

Read the thresholds once.

homebutler alerts

The output, shortened to the root disk. A reading over its threshold is marked with a red circle instead of a check mark:

   CPU:     66.3% (threshold: 90%) ✅
   Memory:  41.1% (threshold: 85%) ✅
   Disk /: 7% (threshold: 90%) ✅

Keep checking every 5 seconds with a rules file. A rule with threshold 1 triggers at once. Press Ctrl+C to stop. The first line is a warning that other users can read the rules file.

homebutler alerts --watch --interval 5s --alerts-config /home/you/alerts-rules.yaml

⚠️  alerts config /home/you/alerts-rules.yaml has permissions 0644; consider chmod 0600 for security
🛡️ Self-Healing active — watching 1 rules (interval: 5s, Ctrl+C to stop)

  ⏱️  11:21:55  ⚠️  cpu-high triggered (cpu 88.1% >= 1%)
  ⏱️  11:22:01  ✅ All clear (cpu 95%, mem 41%, disk 100%)
  ⏱️  11:22:06  ✅ All clear (cpu 78%, mem 41%, disk 100%)

👋 Stopped watching.

An interval that is not a duration is an error.

homebutler alerts --watch --interval abc

error: invalid interval "abc": time: invalid duration "abc"

alerts history

Lists recent alert and remediation events.

homebutler alerts history [flags]

homebutler alerts history

TIME                 RULE               ACTION     DETAILS                        RESULT
2026-10-25 11:21:55  cpu-high           notify     cpu 88.1% >= 1%                success

alerts init

Asks a few questions and writes a starter config.yaml with alert thresholds and rules.

homebutler alerts init [flags]

The questions cover the CPU, memory, and disk thresholds, which containers to watch, what to do when one goes down, and a webhook URL. The file is written to ~/.config/homebutler/config.yaml with mode 0600. When the file exists, the command asks before it overwrites it.

homebutler init is the setup wizard for servers. alerts init is for thresholds and rules.

Enter at the container prompt selects every container

The prompt text says that pressing Enter skips container monitoring. Pressing Enter selects all listed containers and the default action for them is restart. The written config then restarts any of them that goes down. To skip container monitoring, enter 0.

With a config already present, answering n stops the command.

homebutler alerts init

/home/you2/.config/homebutler/config.yaml already exists. Overwrite? [y/N]: Aborted.

alerts test-notify

Sends a test notification to every configured provider.

Kept for older setups

The help text for this command says it remains for backward compatibility. Use homebutler notify test instead.

homebutler alerts test-notify [flags]

Flag	Type	Default	Description
--alerts-config	string	-	Path to a legacy alerts YAML file whose notify settings are used when config.yaml has none.

homebutler alerts test-notify

Sending test notification...
  ✅ webhook: sent