hostnag

Alerting

Each alert goes to the channels its severity calls for, waits out quiet hours and maintenance, and escalates until someone acknowledges it.

Start freeSee pricing

Included on every plan.

PlansFreeProBusiness
What Alerting catches
What breakshostnag
A host goes downYes
A certificate expiresYes

On the board

The overview strip and board. Press LAMP TEST; switch the view.
Active06
Alarm
01
Supervisory
02
Trouble
00
Normal
03
Unlit
00
View

Acme Retail

06Sample data
DownHTTP
Checkout API
api.acme.example/health
Down47 min
checked 1 min ago
Expiring soonCert
mail.example.com
mail.example.com
11d left10 Oct 2026
checked 40 min ago
UpTCP
VPN gateway
vpn.office.example
31ms
checked 2 min ago
OKDomain
example.com
example.com
40d left08 Nov 2026
checked 9 h ago
UpDNS
MX example.com
example.com
18ms
checked 4 min ago
DegradedHTTP
Storefront
shop.acme.example
1840msslow for 22 min
checked 1 min ago

How it works

  1. 01

    Set a policy per group

    Thresholds, a down-reminder interval, and whether to notify on change, recovery or a degraded run.

  2. 02

    Route channels by severity

    Email, Slack, Discord, Teams, Telegram, webhook, SMS, PagerDuty or Opsgenie by plan, each routed with its own minimum severity.

  3. 03

    Say what changed, in the alert itself

    A down or degraded alert on a website, port, DNS, ping or link monitor carries what changed on the same host in the 24 hours before it (the newest 5, and how many more) and which of your watched vendors read Down or Degraded (up to 3). Every channel but SMS draws it.

  4. 04

    Add quiet hours and maintenance windows

    A policy can stay quiet on its own schedule in your timezone, and a maintenance window holds alerts for a group or a single monitor.

  5. 05

    Escalate what nobody acknowledges

    Up to three tiers, each after its own number of minutes, until someone acknowledges the alert from inside it.

What you'll be told

What you get told

  1. [Critical] https://api.acme.example/health is down

    Down from 2 of 3 locations (US East, Frankfurt). Last error: connect_timeout.

    Changed recently
    DNS answer on API DNS, 28 Sept 2026, 21:35 UTC: 198.51.100.7 (was 203.0.113.4)
    Also in trouble
    Card payments: Degraded, Elevated API error rates
    HTTP

  2. [Recovered] api.example.com certificate renewed

    The certificate for api.example.com is healthy again and now expires 10 Dec 2026.

    Cert

  3. [Warning] api.example.com now redirects somewhere else

    api.example.com now redirects to https://example-login.evil.example/ (was https://www.example.com/).

    HTTP

Sample alerts, rendered by the same code that writes your email and Slack messages.

Settings

Channels
Email, Slack, Discord, Microsoft Teams, Telegram, Webhook on Free; + PagerDuty, Opsgenie on Pro; + SMS on Business
Thresholds
up to 10 day counts, 1 to 365 days
Quiet hours
a start and end time in your own timezone
Escalation
up to 3 tiers, 1 to 1440 minutes each
Webhook payload
version 4, signed with x-hostnag-* headers

Questions

Can one channel serve several groups?

Yes. Channels are set once for the org and routed per group, each with its own policy.

What happens during a maintenance window?

Alerts for that group or monitor are held for its duration, then resume as normal once it ends.

What does an alert look like when it lands?

The integrations pages show each channel, rendered by the code that sends it: the Slack blocks, the email, the webhook body and its signature headers, the SMS text.

Related

Status pagesA public page your customers read for themselves. Each tracked row carries its uptime strip and response time, and a mail, certificate or domain row a day by day health strip. Incidents you write sit beside the ones the checks raise, with planned maintenance, a feed, a JSON any status reader speaks and a badge for the README.Uptime monitoringWebsites, APIs and anything that answers on a port, checked every minute from up to five places, with a failure confirmed before anyone is paged.Certificate monitoringEvery certificate hostnag can reach is watched for expiry, a swapped chain and a renewal that should already have happened; the ones it cannot reach get a URL to report to instead.IntegrationsWhere each alert lands, rendered by the code that sends it.

Read

How to publish a status page that draws itself from your monitorsA public status page drawn from the monitors you already run, with uptime strips, incidents the checks raise or you write, subscriber mail and badges.How to monitor a vendor's status page and get the incident title in your own alertsPaste a Statuspage, Instatus, Better Stack, incident.io or status.io page. A vendor outage turns Down on your board, and the alert carries the incident title.

Add a host. Hear about it first.

Start free