The nightly backup
The scheduled entry — cron, Windows Task Scheduler, whatever runs it — gets silently dropped after a server migration. Nobody notices until a restore is needed, and there's nothing to restore.
Know the moment your scheduled work goes quiet
Silent failures cost you before anyone notices.
Not another uptime checker pinging your homepage — Seiri watches the scheduled work behind it. Your cron job, Windows scheduled task, worker, or Kubernetes Workload checks in the moment it finishes. Miss that check-in, and you're the first to know — not your customers.
Free forever · no credit card · not website uptime monitoring
Nightly backup
MISSED
Ping history · last 10 runs
Jan 18
today 02:15
No ping received — 26h 12m since last success
Expected by 02:15 UTC. Last success: yesterday, 02:03 UTC.
# before — runs silently, fails silently 0 2 * * * /path/to/backup-script.sh # after — the ping only fires on success 0 2 * * * /path/to/backup-script.sh && \ curl -fsS https://ping.seiri.app/NkdyngWn9z
See it before you sign up for anything
Log into a live demo — no signup required
Same dashboard real customers use, seeded with monitors, a missed check-in, and a public status page. Look around, break nothing.
Shared demo account — please don't change monitor settings or invite others; we reset it periodically.
# login at https://cloud.seiri.app # email demo@seiri.app # password SeiriDemo2026!
"It didn't error. It just didn't run."
The scariest failures don't throw exceptions. Here's how "did it run?" shows up in the real world.
Invoice generation
The billing script exits 0 but skips a batch because of a timezone bug. Customers don't get invoiced, and finance finds out at month-end.
The ETL pipeline
An upstream job hangs before the transform step finishes. The dashboard keeps showing yesterday's numbers, and nobody's watching the pipeline itself.
The Kubernetes CronJob
A pod gets evicted mid-run during a node drain. The job shows no error in `kubectl` logs — it just never reached completion.
Own the niche — not “uptime monitoring”
Heartbeat & job monitoring guides
Cornerstone pages for the searches that match what Seiri actually does: cron, heartbeats, dead man's switches, and Kubernetes.
Kubernetes is a first-class surface →Cron job monitoring
Highest-intent buyer keyword
Heartbeat monitoring
Push check-ins, not URL probes
Dead man's switch
Silence is the alert
Scheduled job monitoring
Batch, workers, digests
Kubernetes CronJob monitoring
Jobs that never complete
seiri-kube-agent
Helm HealthCheck operator
Background job monitoring
Sidekiq, Celery, queues
Email heartbeat monitoring
No outbound HTTP required
Three steps. One line of code.
Seiri flips monitoring around: instead of polling a URL, your job tells Seiri it's alive.
1
Create a monitor
Tell Seiri the expected schedule — cron expression or interval — and a grace period. You get a unique ping URL and a unique @ping.seiri.app email address.
2
Ping when it finishes
Add a curl call, send an email, or install the seiri-kube-agent Helm operator and declare HealthCheck CRs for in-cluster workloads.
3
Get alerted the moment it's late
No ping within the grace period triggers an instant alert to Slack, Discord, email, SMS, or a custom webhook — before anyone notices the impact.
Ways to check in
Pick whichever fits how your job runs — Linux, Windows, or a cluster.
# Before
0 2 * * * /path/to/backup-script.sh
# After — ping only fires on success
0 2 * * * /path/to/backup-script.sh && curl -fsS https://ping.seiri.app/NkdyngWn9z
No outbound HTTP required
Can it send an email? Then it can check in.
Locked-down networks, legacy cron boxes, and internal servers often can't make outbound HTTP calls — but they can almost always send mail. Every Seiri monitor gets its own @ping.seiri.app address. Send it an email when your job finishes, and Seiri treats it exactly like an HTTP ping.
- Works from restricted networks and air-gapped environments
- No client library, no SDK — just mail or sendmail
- Same alerting, grace periods, and dashboard as HTTP monitors
New message
To: a1b2c3d4@ping.seiri.app
Subject: nightly-backup ok
— sent automatically by cron on backup-host-03 —
Received by Seiri · webhook “Nightly backup” marked healthy
seiri-kube-agent · Helm operator
Watch cluster workloads with HealthCheck CRs
Install the Seiri HealthCheck Operator into your cluster. It watches HealthCheck custom resources and continuously checks Deployments, Pods, StatefulSets, DaemonSets, and Jobs — reporting phases like Healthy, Unhealthy, and Missing back to Seiri for alerts and dashboards.
- → Helm chart with least-privilege RBAC + hardened securityContext
- → CRD
healthchecks.monitoring.seiri.app - → Register the cluster in cloud.seiri.app for authToken + clusterId
# values from your Seiri cluster registration
helm install seiri-kube-agent ./helm-main \
-n seiri-system --create-namespace \
--set seiri.authToken="…" \
--set seiri.clusterId="…"
# monitor a Deployment
apiVersion: monitoring.seiri.app/v1alpha1
kind: HealthCheck
metadata:
name: check-my-app
spec:
targetRef:
apiVersion: apps/v1
kind: Deployment
name: my-app
interval: 10s
checkID: "chk-my-app"
Built for scheduled work, not websites
Everything you need to trust that your background jobs actually ran.
Heartbeat monitoring
Push-based checks catch jobs that never run, hang, or exit early — the failures a URL probe would never see.
Email heartbeats
A unique @ping.seiri.app address per monitor, for jobs that can send mail but can't make HTTP calls.
Kubernetes HealthCheck operator
Install seiri-kube-agent with Helm. It watches HealthCheck CRs and reports Deployment, Pod, Job, StatefulSet, and DaemonSet health to Seiri.
Multi-channel alerts
Slack, Discord, email, SMS, and custom webhooks — with escalation across channels when the first alert goes unacknowledged.
Status pages
Share a public or custom-domain status page for the jobs and services your team or customers care about.
REST API
Create and manage monitors programmatically with scoped API keys — provision monitoring as part of your deploy pipeline.
Simple pricing, three plans
Start free. Upgrade when you outgrow it. No fake trials — just a generous free tier.
Free
$0
Free forever
- 10 monitors
- 5-minute check interval
- 3 seats
- 1 status page
- 1 email heartbeat
- 1 API key
Most Popular
Growth
$19/mo
For growing teams
- 40 monitors (+$5/20 extra)
- 1-minute check interval
- 10 seats
- Custom status domain
- 5 email heartbeats
- 10 API keys
Scale
$49/mo
For high-volume infra
- 250 monitors
- 30-second check interval
- 15 seats
- 10 Kubernetes clusters
- 25 email heartbeats
Who it's for
If something on your infrastructure is supposed to run on a schedule, Seiri is built for you — but the reason it matters is different depending on your seat.
If you lead the team
"Doesn't my observability stack already cover this?"
Usually not for this failure mode — see why Datadog/Grafana/CloudWatch structurally miss it. What you're actually buying is a way to stop finding out about dropped backups, skipped invoice runs, and stalled pipelines from a customer or from finance instead of a dashboard.
Free tier is enough to pilot this on a handful of your riskiest jobs before it's a line item anyone has to approve.
If you're the one on call
"I already have a cron job that pings healthchecks.io / a custom script for this."
Fair — plenty of teams have duct-taped something together. Seiri is worth a look when that duct tape doesn't cover Kubernetes CronJobs, email-only networks, or a status page you can hand to someone non-technical without building one.
See how it stacks up: vs Cronitor · vs Healthchecks.io.
Backend engineers
Cron jobs, queue workers, and scheduled scripts that run without anyone watching.
Data engineers
ETL pipelines and batch jobs where a stale dashboard is worse than an error message.
DevOps & SRE teams
seiri-kube-agent HealthChecks across Deployments, Jobs, and clusters — plus heartbeat webhooks for everything else, on Linux or Windows.
Solo founders
Backups, reports, and integrations you can't afford to find out failed a week later.
Frequently Asked Questions
Is Seiri website uptime monitoring?
No. Seiri doesn't poll a URL to see if your website is up — that's what tools like Pingdom or UptimeRobot do. Seiri is built for scheduled work — cron jobs, Windows scheduled tasks, background workers, and Kubernetes CronJobs — that ping Seiri when they finish. If the ping doesn't arrive on schedule, you're alerted immediately, even while your website and servers still report perfectly healthy.
Why doesn't Datadog, Grafana, or CloudWatch already catch this?
Because a job that exits 0 having done nothing doesn't look wrong to them. Those tools alert on infrastructure symptoms — CPU spikes, error logs, failed requests — not on the absence of work. If a crontab line gets dropped in a migration, or a Kubernetes CronJob never creates a Job, there's no error and no bad metric for them to catch; nothing happened, so nothing looks broken. Seiri flips the model: your job has to prove it ran by pinging when it finishes, so silence itself becomes the alert. Most teams run Seiri alongside their existing stack, not instead of it.
How does the heartbeat model work?
You create a monitor with an expected schedule and grace period, then add one line to your job: a curl request to your unique ping URL, or an email to your unique @ping.seiri.app address. If Seiri doesn't receive that ping within the expected window, it sends an alert.
What is an email heartbeat?
An email heartbeat lets a job check in by sending an email instead of an HTTP request. This works for systems with restricted outbound access, legacy servers, or anything that can send mail but can't easily make an HTTP call. Just send an email to your unique @ping.seiri.app address.
Can Seiri monitor Kubernetes workloads?
Yes. Install seiri-kube-agent with Helm (needs your cluster’s auth token + cluster ID from cloud.seiri.app). Create HealthCheck CRs for Deployments, Pods, Jobs, StatefulSets, or DaemonSets — phases like Healthy / Unhealthy / Missing show up in Seiri. CronJobs can also still ping ping.seiri.app on completion.
Do you offer status pages?
Yes. Every plan includes at least one status page you can share with your team or customers, and Growth and Scale plans support a custom status domain.
Is there a free plan?
Yes — free forever, no credit card required. It includes 10 monitors, a 5-minute check interval, 3 seats, 1 status page, 1 email heartbeat, and 1 API key.
Did it run today?
You'll know the moment it doesn't. Set up your first heartbeat monitor in under two minutes.
Free forever · no credit card required · setup in 2 minutes