Analytics · errors · uptime
Cookieless site analytics, error tracking and uptime monitoring are part of the platform your app already runs on — hosted in the EU, alerting the channels you already use, and one click from handing a bug to the agent in your dev box.
Pageviews
21,870
Unique visitors
7,940
Bounce rate
44%
Avg visit duration
2m 02s
Pageviews
Top paths
Top referrers
Countries
Exceptions your application reported, grouped by cause. New and regressed issues raise an alert through your notification channels, and any issue can be handed to a coding agent in this project's dev box.
TypeError: cart.total is not a function
Newcheckout (src/checkout.ts:12) · ×38 · 4 min ago · release 8054a5a1
Error: connect ETIMEDOUT 10.0.3.4:6379
Regressedqueue.connect (src/jobs/queue.ts:41) · ×7 · 1 hour ago · release 8054a5a1
UnhandledRejection: Cannot read properties of undefined (reading 'id')
GET /api/orders/:id · ×3 · 6 hours ago · release c30ba7e5
Probed from outside, against the URL your users open — so it also catches DNS, TLS and routing failures. Alerts clear themselves when the target answers again.
The three things every production app ends up signing up for separately, running on the same EU platform as the app, with the same login and the same alert channels.
Site analytics
Pageviews, visitors, referrers, campaigns, countries and custom events for every site you own — hosted here or anywhere else — on one page.
/customers/123 is reported as /customers/:idError tracking
Your app posts exceptions to an ingest URL; HostStack groups them into issues and tells you when one is new or comes back after you fixed it.
Uptime monitoring
HostStack requests your public URL on the interval you set and alerts your team when it stops answering — no probe service to write or run.
Uptime, alerts, errors, backups, certificates and resource pressure for everything the team owns — including what nothing is watching yet.
Alerts, uptime and outage history, unresolved error issues, and services running close to their CPU or memory limit — for everything the team owns, in one view.
Services and sites without an uptime check are listed as unmonitored rather than left out, so a gap in coverage looks like a gap and not like a healthy row.
Slack, Discord, email, or browser push to your phone. Deploy failures, crash loops, failed backups, missed cron runs and certificate problems arrive through the same channels.
Everything on the page is in the CLI, the API and the MCP server, so the agent in your dev box can check what broke before it starts on a fix.
# Is the tracker working on this site — and if not, why
hoststack analytics check example.com
hoststack analytics stats example.com --range 30d
# What is broken right now, and hand one to the agent
hoststack errors list --service 48
hoststack errors fix 12
# Watch a site hosted somewhere else
hoststack analytics verify example.com --check
hoststack uptime set --site 3 --path /Privacy, limits, and what works on sites you do not host with us.
No. The tracker writes nothing to the visitor's device, so ePrivacy art. 5(3) never engages. The IP address is used to look up a country at ingest and is never written to disk, and visitors are grouped with a salt that is regenerated daily and thrown away, so nothing links one day to the next. Do Not Track is honoured. The one opt-in that stores an id (data-storage="local") does need consent, and it is off unless you add it.
Yes. Analytics is one script tag and works on any site. An uptime check on a site we do not host needs one extra step: prove the domain is yours with the TXT record shown in the dashboard, because the check makes HostStack fetch that hostname on a schedule.
No. Reporting is one HTTP POST to your service's ingest URL, so a handler for Node, Bun, Python or the browser is a few lines you can read. Exceptions are grouped into issues, labelled with the release that threw them, and you are alerted when an issue is new or when a resolved one comes back. Any issue can be handed to the coding agent in your project's dev box.
To the notification channels you already set up — Slack, Discord or email — and optionally as browser push notifications, so an outage reaches you with the dashboard closed. Every alert is also recorded on the Monitoring page, and an uptime alert clears itself when the service answers again.
Analytics accepts 50,000 events per site per hour and error tracking 10,000 events per service per hour; over that, events are counted and shown rather than dropped silently. Error occurrences are kept for 30 days while the issue and its counts stay. Uptime checks run every 30 seconds to every hour, on the interval you choose.
Deploy on HostStack and analytics, errors and uptime are waiting in the same dashboard — or add the script tag to a site you host anywhere else.
Essential cookies only — for login sessions. No tracking. Details