Detect outages before your customers notice.
Distributed monitoring locations check your services at short intervals — every 10 seconds when an outage is suspected. Confirmed multiple times before we wake you.
Strategically Distributed Locations
Monitoring locations in Europe and North America show how your services perform everywhere. The probes only run the check — your data is stored 100% in Germany.
Majority-Based Detection
No single location triggers false alarms.
Multiple probes must confirm an outage before an incident is created. Individual network issues do not cause false alarms.
Rapid Re-Check
Faster detection when something looks off — on every plan, including Free.
As soon as a probe suspects an outage, LIVCK automatically switches to rapid mode and checks every 10 seconds. This applies to every plan — including the free plan.
6 check types
The right check for every protocol.
HTTP/HTTPS
Status code, response body, headers, redirects, TLS
TCP
Port reachability, connection timeout, banner response
DNS
Record resolution, DNS propagation, nameserver check
ICMP
Ping latency, packet loss, round-trip time
SSL/TLS
Certificate expiry, chain validation, protocol version
Heartbeat
Incoming check-in, run duration, exit code, reported measurements
More planned
Additional check types are in development.
You cannot ping what never answers.
A backup at three in the morning, a cron job, a queue worker: they run behind the firewall, and from the outside nobody can tell whether they ran at all. So we turn it around — they check in with us.
You get a URL. Your job calls it when it finishes. If the call stays away, we raise the alarm — through the same channels and escalation steps as any other outage.
Four signals, one URL
curl .../ping/TOKENFinishedNo suffix: the run succeeded.
curl .../ping/TOKEN/startStartedMarks the beginning. Add a maximum duration and we also alert when a run hangs and never reports back.
curl .../ping/TOKEN/failFailedAlarm straight away, without waiting for the next deadline.
curl .../ping/TOKEN/$?Exit code0 is success, anything else is a failure. One suffix after your command, nothing more.
You define what is expected
Fixed interval
Every five minutes, hourly, daily — up to 30 days apart.
Real cron expression
The same line as in your crontab, in your timezone. Daylight saving included.
Grace period
30 seconds to 24 hours of slack before "late" turns into an alarm.
Maximum duration
Pairs with the start signal: a run that begins and never comes back is just as broken as one that never begins. Leave it empty and this check stays off.
Send the numbers along
The job may post JSON: how many records it processed, how large the backup was, how long it took. Up to ten measurements per run, which you can then watch exactly like anything else.
{ "rows": 14203, "size_mb": 812 }Heartbeats count against your plan's service limit — there is no separate pool and no separate cap.
Not just reachable – actually healthy
Define custom conditions per service and combine them into rule sets.
An HTTP 200 doesn't mean everything works. With conditions you check status code, response time, content, JSON fields, headers and more – and LIVCK only alerts when one of your rules is violated.
If a condition fails, LIVCK detects the incident instantly.
Certificate valid · 247 days left
Monitor protected and internal endpoints too
Store your own certificates and monitor HTTP and SSL endpoints that other tools simply can't reach.
Some endpoints require a client certificate (mTLS) or use an internal CA. Store both once in the certificate vault and assign them to your HTTP and SSL checks – encrypted at rest, delivered encrypted to the probes and decrypted there only for the moment of the check – never kept in plaintext.
Detailed Metrics
Understand where time is lost.
IPv4 & IPv6
Dual-stack monitoring for complete coverage.
Check your services via IPv4 and IPv6. Detect issues affecting only one protocol before your users notice.
Missing something?
Need an integration or feature we don't offer yet? Let us know – we're constantly evolving LIVCK.
Works together
Monitoring, alerting, incidents and status pages build on each other — continue here.