A client writes that their site is down, and your phone shows a red alert or a green dashboard. Before you answer, you need to know whether the site is down from every location, down for one region or fine for visitors while only OctoPing's checker is refused. This guide shows where OctoPing records that, how to read it and what to tell the client in each case.
What you need
- A monitored site. Every site is checked from the primary checker and from the US, Europe and Asia probes.
- An alert notification on the site, so the down alert reaches you with its Checked from block.
Read the page before you reply
Look at the status badge
Open the site. The badge at the top reads Online, Offline or Partial, among others. Offline means the primary checker's failure was confirmed. Partial means the primary checker is refused while every probe region sees the site up. Online while a client reports a problem means the primary checker can reach the site, so look at the regions next.
Open the Multi-Region Monitoring card
Its Current Status list gives the primary checker and each region a badge, the HTTP status or failure reason it saw and how long ago it checked. A region marked No recent data hasn't reported within the last 20 minutes, so its last result is not evidence about the site now.
Match what you see to a case
The table below covers the four cases you'll meet most. The right column is our read of what to tell the client.
Read the down alert, if one was sent
The alert's Checked from block lists the same four results. When some locations still reach the site, the subject names the ones that don't, and the text says the cause may be regional.
| What you see | What it means | What to tell the client |
|---|---|---|
| Primary Up, one region Down, badge Online | The primary checker reaches the site, so no down alert is sent. The region's result is shown on the page. | The site is up. Some visitors in that region may have trouble, and you're looking at the region's route or CDN. |
| Primary Blocked, every region Up, badge Partial | A firewall or bot rule is refusing OctoPing's checker. Visitors can reach the site, and the time isn't counted as downtime. | The site is up for visitors. The monitor is blocked, and you're fixing the rule. |
| Primary Down and a region Down, others Up, badge Offline | The failure is confirmed and the down alert goes out, naming the regions that fail. | The site is failing from the regions named in the alert and reachable from the others. Name the regions. |
| Primary and every region Down, badge Offline | The site is down or unreachable from everywhere OctoPing looks. | The site is down. Give the cause from the alert, such as a timeout or an HTTP 503. |
See each case in an alert
The down alert below came from our test run, where the primary checker got a 403 and Europe saw a 503 while the US and Asia saw the site. The block is copied as it was sent.
Checked from
------------
Primary: Down (HTTP 403) - 0 minutes ago
US: Up (HTTP 200) - 0 minutes ago
Europe: Down (HTTP 503) - 0 minutes ago
Asia: Up (HTTP 200) - 0 minutes ago
The site is still responding from US, Asia, so this may be a
regional network, CDN or DNS problem rather than a full outage.When every region reaches the site but the primary checker is refused, OctoPing sends one notice per run of failures instead of a down alert. It names the status the checker got and the regions that see the site up, and it lists the addresses to allow. This is the notice from our test run, shortened.
Website: https://client.example/
Status: Up (confirmed from US, Europe, Asia)
OctoPing's primary checker is being refused (HTTP 403), but your site is up
from US, Europe, Asia. This usually means a firewall or bot rule is blocking
our checker. Allowlist these addresses so our checker can reach your site
again: <checker addresses listed in the notice>.
No down alert was sent and this is not counted as downtime. We won't send this
notice again until our checker can reach your site and is refused again; if a
probe region starts failing, you'll get a normal down alert.
...What happens while OctoPing waits
When the primary checker fails and no region has reported since, the alert is held. In our test, the site stayed Online with no alert during that wait. After up to 10 minutes the failure counts as confirmed and the alert goes out, so a silent or broken probe region can't hide an outage. If any region reports a failure, the failure is confirmed and the normal down alert follows. If every region reports the site up, the site becomes Partial instead.
- A declared outage isn't reversed by later healthy reports from the regions. The primary checker decides when the site recovers.
- A Partial site becomes a real outage the moment any region reports a failure.
- A public status page shows a Partial site as operational, because visitors can reach it. Your client's page stays green while you sort out the checker's access.
Verify it works
- The Multi-Region Monitoring card lists Primary, US, Europe and Asia, each with a recent time.
- The notification you rely on has Site goes down switched on. Down alerts and the Partial notice both follow that switch.
- The notification's retry threshold is what you expect, because a down alert waits for that many failed checks in a row.
Read the reference for Vantage points and confirmation
Last reviewed
Related
- DocsVantage points and confirmationHow the primary checker and the probe regions confirm a failure, what Partial means and why OctoPing alerts when the probes are silent.
- FeatureMulti-region website monitoring that confirms outagesEvery site is checked from a primary checker and from the US, Europe and Asia. A failure only the primary sees shows as partial, with one notice and no false outage.
- DocsStatusesThe statuses a site and a heartbeat can have in the app and on a public status page, and the order in which the rules apply.