github OneUptime/oneuptime 14.0.15
on GitHub

latest releases: build-number-9044, build-number-9042
3 hours ago
Changes
  • Notes from an event's overview use its Notes page composer: one step, with templates, AI and who it reaches
  • Notification methods: say when SMS, calls, WhatsApp or Telegram are off, with the switch right there
  • Reminders, alert privacy, status page visibility, runbook, SLO evaluation and session replay switches save when flipped
  • Form guards' "are really read" floors sit below what simplification can reach (CI runs without ee/)
  • One Probes & Interval page for a monitor's probes, interval and probe agreement
  • Forms of three rows or fewer are one page (note and postmortem templates, OID templates, SLO create), and a guard keeps it that way
  • Dashboard custom domains keep a working certificate: verified, ordered and renewed by workers, never deleted by the status page job
  • Project Settings and Admin one-switch settings save on flip, and ask first where they can lock people out
  • test(e2e): judge an explorer double-click's landing only once its timing held
  • SSO: adding an OIDC provider asks for the issuer, client ID and secret; the rest is filled in
  • test(e2e): the chart zoom fixture's analytics lists answer like the server
  • fix(database): stop a project create from taking the tenant project's id
  • Fix master CI: ee SSO license tests (#4319, #4322) and the SLO e2e Description fill (#4318)
  • AI settings: each AI behaviour is a switch that saves on flip, limits folded under Advanced
  • ci(release): tag each release on the commit the run built
  • SSO: adding a SAML provider or a SCIM connection asks only what the identity provider gives you
  • Create Team asks the team's Access (as API keys do); Block Permissions fold under Advanced on a team's Permissions page
  • Admin Dashboard: adding someone to a project starts on its Members team, in one step
  • Status page custom domains: add it, the DNS record opens right away, and the free certificate is issued without a button
  • Admin LLM, Call & SMS, WhatsApp, Invite User and calendar link settings lose steps they didn't need; the Advanced-step guard reads translated titles
  • Status pages: who can see a page is one choice on Access (anyone, people who sign in, or anyone with the password)
  • No always-on banners on session replay, inventory and database pages; replay retention edited in one place
  • Share a dashboard with one 'Who can view' choice, from the dashboard's ⋯ menu
  • Custom domain certificates: one Let's Encrypt budget, one order per name, nothing deleted on a DNS blip, and failed orders shown
  • Each resource's name, description, labels and identity are edited in one place: its Settings page
  • Incoming call escalation rules: one step - who to call and how long to ring
  • Dashboard custom domains work like status page ones: DNS Setup on create, Check now orders the certificate, no Order Free SSL
  • Create an announcement in two steps, with the status page already picked
  • A project's first Twilio config becomes its default, so SMS and calls use it at once
  • Scheduling maintenance and announcing suggest the status pages that show the affected monitors, one click to add
  • On-call rules: one page with a tab per kind, in User Settings and under Users
  • Status pages: overall uptime % and what counts as downtime are rows of 'What your status page shows', each saving on its own
  • Log and trace data rules start from what the rule does; scrub rules that would scrub nothing are refused
  • chore: npm audit fix
  • Creating from a record's tab keeps that record picked: Declare Incident, Create Alert and maintenance from monitors, hosts, clusters, services and more
  • fix(telemetry): store a NATS broker's split delivery from OBI as SERVER
  • fix(traces): include the spans under a span in its Profile tab
  • fix(telemetry): drop OBI v0.14's own Node.js inspector requests
  • feat(kubernetes-agent): ask OBI for span metrics as application_span_otel
  • Stepped forms: the primary action only on the last step, and Next is never primary
  • Forms and pages fold rarely used options under More fields / More settings, and the folded header shows what is inside
  • Declare Incident: monitors apart, Change Monitor Status to only with a monitor, status page limit and notify under More fields
  • Subscribers who picked a monitor group hear about announcements and maintenance notes on its monitors
  • Status page: subscribing is one page again, with Preferences folded to what the visitor will get
  • test(kubernetes-agent): run the agent's eBPF pipeline end to end on a real kernel
  • fix: write NUL characters as \u0000 so git diffs these files as text
  • Dropdown and filter-chip clear buttons sit beside their trigger, not inside it; no control nested in another, with a guard
  • Kubernetes Control Plane and Service Mesh: the Helm setup hint shows only when a tab's charts are empty
  • fix: master CI, js-lint's 3 errors and Compile App's heap out of memory
  • Incidents, alerts and maintenance events show their own measurements on their pages
  • Postmortem: Apply Template only with a template, publish options in order, and settings paths that exist
  • Card descriptions say what each list or card is for, not 'Here is a list of…', with a guard
  • fix(telemetry): store an MQTT or NATS broker's client-typed publish as SERVER
  • fix(profiles): count only samples on a trace's own spans in its Profile tab
  • Products menu: Essentials are always expanded and can no longer be folded
  • Card action buttons such as Edit sit on the right of the card header, everywhere
  • Created sits on the ID line instead of a row of its own, on every details card
  • Search (Cmd+K) finds every settings page and action, such as API Keys or Delete Project, and searches better
  • Status page email reports: one switch with a default schedule, and turning them off needs no dates
  • feat(forms): Branding section - a form's own logo, alt text and favicon
  • Dashboards: picking a template fills in the name, and Create opens the new dashboard
  • One incident role picker everywhere, no 'Multiple' tag, and incident templates start with no state
  • New labels, states, severities, statuses and incident roles start with a colour already picked
  • Grouping rules make people owners of the episodes they open, replacing a Default Assign To nothing showed
  • Scheduled maintenance: monitors picked apart from other affected resources, Change Monitor Status to right under them
  • Log, trace, exception and security event monitors fold their extra filters under More fields
  • Metric pipeline rules and recording rules start from the rule: no Basic Info step, All/Any only with two filters
  • Label and owner rules on 21 products open on what they match, then what they add (no Basic Info step)
Check out latest releases or
releases around OneUptime/oneuptime 14.0.15

Don't miss a new oneuptime release

NewReleases is sending notifications on new releases.

Get notifications