Uptime

These commands match High availability → Uptime in the dashboard. For what monitors, incidents and the public status page do, see Uptime and public status page. Not dnz status-page, which switches a host to a maintenance or moved page, nor dnz proxy maintenance, which schedules repeating maintenance and keeps the hostnames’ monitors quiet during it. Reading needs Manage monitoring, Manage proxies or Post incidents; changing needs Manage monitoring. Post incidents (the built-in Helpdesk role) also opens, edits and updates incidents, without touching monitors, the public page or anything else; deleting an incident stays with Manage monitoring. Monitors are named by their name or id.

Monitors

dnz uptime                                   # list: state, 24h/30d/90d uptime, response time, what it checks
dnz uptime list --status down -q web --page 2   # a page of monitors, by state or a search
dnz uptime add Website --url https://www.example.com/ --keyword Welcome --public Website --group Web
dnz uptime add "App servers" --service app.example.com --interval 30
dnz uptime add Shop --url https://shop.example.com/ --from jkt-1,singapore   # from DNZ and these nodes or regions
dnz uptime edit Shop --from ""               # from DNZ only again
dnz uptime places                            # the nodes and regions it can check from
dnz uptime add Billing --url https://billing.internal/health --expect 200 --slow-ms 800 --insecure true --paused
dnz uptime show Website                      # last check, uptime 24h/7d/30d/90d, last checks, 30 days, incidents
dnz uptime check Website                     # one check now; shown, not kept
dnz uptime edit Website --confirm 3 --timeout 5
dnz uptime pause Website                     # no checks, history kept; `resume` checks it again
dnz uptime rm Website                        # with its history

Option	Meaning
`--url URL`	Check this http or https URL
`--service HOSTNAME`	Watch the DNZ Proxy servers behind this hostname, from the nodes’ health checks
`--method GET\|HEAD`	The request (default `GET`)
`--expect CODES`	Codes that pass, e.g. `200-299,301` (default 200–399)
`--keyword TEXT`	Text the answer must contain
`--slow-ms MS`	Slower answers count as degraded (0: never)
`--interval SECS`	Seconds between checks, 20–3600 (default 60)
`--timeout SECS`	Seconds a check waits (default 10)
`--confirm N`	Failed checks in a row before down (default 2)
`--insecure true`	Accept any certificate
`--public NAME`, `--group GROUP`	Put it on the public status page under this name and group
`--private`	Take it off the public status page (`edit`)
`--paused`	Save it paused (`add`)

edit changes only the options you pass.

Groups of routes

dnz uptime watch ls                          # watched groups, and every group with its hostnames
dnz uptime watch add "Web app" --path /health --interval 120 --public
dnz uptime watch edit "Web app" --interval 300
dnz uptime watch rm "Web app"                # its monitors go too; --keep-monitors keeps them

A watched group gets a monitor per hostname of its routes, made and deleted as routes join and leave it. add and edit take --path, --expect, --keyword, --slow-ms, --interval, --timeout, --confirm, --insecure, --public / --private and --public-group.

Incidents

dnz uptime incidents --days 90               # open ones and the last days; --monitor Website; --page, --limit
dnz uptime incident open "Database maintenance" --impact maintenance --monitor Website -m "Planned, 22:00–23:00."
dnz uptime incident update 12 identified -m "A disk filled up; clearing it."
dnz uptime incident resolve 12 -m "All good again."
dnz uptime incident show 12                  # its timeline
dnz uptime incident rm 12

open takes --impact (minor, major, critical, maintenance), --status (investigating, identified, monitoring, resolved), --monitor (repeat it for more), -m for the first update and --hidden to keep it off the public page.

Written afterwards

dnz uptime incidents add "Power cut" --impact major --monitor Website --started "2026-11-01 13:04" --resolved "2026-11-01 14:31"
dnz uptime incident update 13 identified -m "The building lost power." --at "2026-11-01 13:20"   # fills in the timeline
dnz uptime incident resolve 14 --at "2026-11-01 15:10"
dnz uptime incident edit 13 --resolved "2026-11-01 14:35"      # only what you pass changes; also --title, --impact, --monitor, --started, --hidden/--shown
dnz uptime gaps                              # times DNZ wasn't checking (it was off), and what was done about each
dnz uptime incidents add "Power cut" --gap 3 --monitor Website   # the gap's times
dnz uptime gaps dismiss 3                    # no incident for it

add (or open) takes --started, --resolved (it is opened as resolved), --at (the first update’s time) and --gap; update and resolve take --at. Times are on the server’s clock (2026-11-01 13:04) or RFC 3339 (2026-11-01T05:04:00Z), never in the future, and in order: started, the updates, resolved. See Uptime for what they count for.

The public status page

dnz uptime page                              # its settings and where it is
dnz uptime page set --on --title "Company status" --hostname status.example.com --incident-days 30
dnz uptime page set --off

For its own hostname, let DNZ make the routes and DNS record on proxies: --proxy names them (repeat it or separate names with commas) and --dnz-address is DNZ’s own address as they reach it (its IP and the port its web UI listens on, never the page’s hostname or a proxy’s port; DNZ refuses an address that doesn’t answer as DNZ). They are made while the page is on and deleted when it’s turned off; a proxy left out of --proxy loses its route, and --no-route deletes them all and keeps the hostname. Either flag alone keeps the other’s saved value.

dnz uptime page set --on --hostname status.example.com --proxy edge,edge-2 --dnz-address http://10.0.0.53:8080
dnz uptime page set --proxy edge
dnz uptime page set --no-route

Basics

Commands

Monitors

Groups of routes

Incidents

Written afterwards

The public status page