Syschecks/Changelog

Changelog

See what's new in Syschecks

v0.26.0
2026-07-22

Settings that actually save, and alerts that actually reach you

  • Fixed
    An escalation policy could not be attached to a check at all — the selector never appeared in the check editor. On-call and escalation were therefore unreachable for checks, so a failing check paged nobody. Organization-wide notification channels were invisible in the same editor for the same reason.
  • Fixed
    Creating a status page failed with "Failed to save status page", for both public and private pages, once one page existed without a custom domain.
  • Fixed
    Visiting the old checks.systeam.pl address redirected to an unreachable port instead of syschecks.com, so anyone still using the previous domain could not open the app. The API on the old domain was never affected, so agents kept reporting throughout.
  • Fixed
    Your profile timezone, language, country, job title and company were accepted and then discarded on save. Timezone matters most: every timestamp in the app is rendered in it, so the whole interface kept showing UTC no matter what you picked.
  • Fixed
    Editing the expected HTTP status codes of a check (and the expected result of a database check) reported success and changed nothing.
  • Fixed
    Notification emails, Discord and Slack messages, push notifications and postmortem PDFs still signed themselves with the product’s old name. They now say Syschecks.
  • Fixed
    A newly created on-call schedule reported roughly 3% coverage and a warning, because coverage was measured over the previous 30 days — including the time before the schedule existed. It is now measured from the moment you created it.
  • Improved
    One malfunctioning endpoint can no longer flood error reporting and hide everything else: each distinct issue is now capped per hour, so unrelated failures stay visible.
  • Improved
    The geo staleness check no longer depends on the API boundary being the only place that normalizes timestamps, removing a failure mode that could reject agent results.
v0.25.0
2026-07-20

First-run fixes, tablets and accessibility

  • Fixed
    A new organization now starts with a project, so the first "Add Check" works instead of failing on an empty project list.
  • Fixed
    Opening the settings of a freshly created organization could return a server error on the first visit. It now loads first time.
  • Improved
    The same URL can now be monitored more than once — in different projects or for different clients — so you can watch one target with two intervals, two sets of locations, or different alerting. Duplicates inside a single project are still blocked.
  • Fixed
    On tablets the page was cut off on the right until the device was rotated; the layout now fits the screen at every width.
  • Fixed
    Validation messages from dialogs (for example "Recipients email is required") appeared behind the dialog and were easy to miss. They now show as a toast on top.
  • Fixed
    The Cloud Status early-warning panel now works. The endpoint behind it was unreachable, so the panel had never shown anything.
  • Improved
    The "Add Check" form now opens on HTTP (Website/Endpoint) rather than the cron heartbeat type, which has no URL field.
  • Improved
    Dashboard charts now load a bounded window of history instead of scanning everything ever recorded, which was slowing the whole page down.
  • Improved
    Icon-only buttons — the navigation, theme and account controls in the header, the notification switches and the per-session revoke buttons — now announce what they do to screen readers.
  • Fixed
    A war room whose incident has passed the log retention window no longer offers a link that leads nowhere.
  • Fixed
    Error messages that were still in Polish are now in English throughout, including the one you get when signing up with an address that is already registered.
  • Improved
    Notification e-mails are now sent from no-reply@systeam.pl instead of a personal mailbox.
v0.24.0
2026-07-18

War rooms without leaving the product

  • New
    Native in-app war rooms: open a room on an incident and talk in it directly, with no third-party chat required.
  • New
    Zoom and Google Meet war rooms are created automatically when a room opens, using your connected account.
  • New
    Participants panel with roles and live presence, so you can see who is in the room and who is responding.
  • New
    Invite people to a war room, or remove them, from inside the room.
  • New
    File attachments in war-room chat, including pasting a screenshot straight from the clipboard.
  • New
    Emoji picker, automatic linking of URLs, and preview cards for links posted in chat.
  • New
    A global War Rooms page listing every active room, with a badge in the navigation when one is open.
  • Improved
    Anyone who posts a message or a file joins the room roster automatically, so the participant list reflects who is actually working the incident.
v0.23.0
2026-07-03

Reliability & Security Hardening

  • Fixed
    Alerts set to fire only after N consecutive failures now trigger reliably — previously the threshold could stop the DOWN alert from ever being sent.
  • Fixed
    Acknowledging an incident now cancels its queued follow-up SMS and voice pages, so you stop getting paged after you have responded.
  • Fixed
    Duplicate incoming events (e.g. Alertmanager retries) no longer create duplicate incidents or double-page on-call.
  • Fixed
    Escalations are now suppressed during maintenance windows, and a check that fails inside a window alerts once the window ends instead of staying silently DOWN.
  • Fixed
    On-call health and schedule quality now report correct on-call hours, coverage and fairness (previously stuck at zero / 100%).
  • Fixed
    Burn-rate SLO alerts now fire at the correct threshold, and escalation steps with a webhook target now save correctly.
  • Fixed
    Flapping-detection settings are now saved and actually take effect; per-check alert threshold is applied on create.
  • Fixed
    Opening one check right after another no longer briefly shows the wrong check’s data; corrected pagination on Action Items and Change Events.
  • Improved
    Public status-page uptime data is now cached — noticeably faster and resilient under heavy or anonymous traffic.
  • Improved
    Zero-downtime deploys: new backend versions roll out without a brief availability gap.
  • Improved
    Actions you cannot perform for your role (edit/delete checks, invite members, delete organization) are now hidden instead of failing with a permission error.
  • Improved
    More resilient check scheduling under load — no duplicate check runs when the execution queue is busy.
  • Security
    Hardened SSRF protection: checks and webhooks can no longer be steered to internal or cloud-metadata addresses, including via DNS rebinding and IPv4-mapped IPv6 addresses.
  • Security
    Telegram bot webhooks are now cryptographically verified, matching the other chat platforms.
  • Security
    Tightened plan and permission checks on SSL/domain/DNS alert settings and on organization configuration import.
v0.22.0
2026-04-03

On-Call Health Dashboard v2

  • New
    Team summary bar — total pages, night/weekend pages, avg MTTR, avg TTA, busiest person, all with trend vs previous period
  • New
    Day/night/weekend alert breakdown per team member — business hours, after-hours, night, weekend with color-coded pills
  • New
    Top noisy alert sources — see which services generate the most on-call load, with percentage bars
  • New
    Burnout risk explanation — expandable breakdown showing night interruptions, incidents, on-call density with weights and warning flags
  • New
    Alert heatmap — 7x24 grid showing alert density by day of week and hour, instantly reveals patterns
  • New
    Time to Acknowledge (TTA) per team member — median with P90 tooltip, trend vs previous period
  • Improved
    All health metrics now show trend comparison vs previous period with color-coded delta badges
v0.21.0
2026-04-03

Flexible Check Execution & Multi-Agent Support

  • New
    Multi-agent check execution — assign multiple private agents to a single check for multi-office/multi-region internal monitoring
  • New
    Hybrid execution mode — checks can now run simultaneously from private agents AND geo locations (no longer mutually exclusive)
  • New
    Critical source designation — mark any agent or geo location as critical; if it goes DOWN, the entire check escalates to DOWN immediately
  • New
    Unified status aggregation — single check.status computed from all sources (private + geo) with DEGRADED state for partial failures
  • New
    Source breakdown panel on check detail page — see status, response time, and critical flag per execution source
  • Improved
    Check edit UI — execution section now uses independent checkboxes for geo + per-agent selection instead of mutually exclusive dropdown
  • Improved
    Alert messages include source breakdown context (which sources are UP/DOWN) for faster incident triage
v0.20.0
2026-04-02

Database Checks in Go Agent

  • New
    Database health checks now run from Go agents — supports PostgreSQL, MySQL, MongoDB, and Redis with connection testing, query execution, SSL, and configurable timeouts
  • New
    SQL query validation (defense-in-depth) — agent rejects non-SELECT queries, strips SQL comments, blocks forbidden keywords
  • Improved
    Database checks can now reach private/LAN databases through private agents (no longer limited to central server IP)
  • Fixed
    Ansible agent role — config file ownership now correctly set to healthcheck-agent user (was root, causing Permission Denied on restart)
v0.19.0
2026-04-01

Onboarding, War Rooms, Follow-the-Sun & Handoffs

  • New
    Onboarding wizard for new users — 5-step setup guide (org, check, notifications) with DNS, TCP, Ping check types and Discord/Telegram channels
  • New
    Getting Started checklist on dashboard — tracks progress, auto-detects completed steps, dismiss or remind later
  • New
    On-Call mini-wizard — create schedule + escalation policy in 2 steps from the checklist
  • New
    Settings mini-wizard — configure timezone, quiet hours, and log retention from the checklist
  • New
    Slack Channel war room provider — auto-creates a dedicated #wr-{name} channel via Slack Bot API
  • Improved
    War room providers: Jitsi "Generate" button, Google Meet "New Meet" button, Discord/Slack/Teams marked as auto-created (no URL required)
  • New
    Service Catalog — map services, ownership, dependencies, and tiers (P1-P4). Link checks to services for routing and blast radius analysis
  • New
    Schedule Layers (Follow-the-Sun) — multi-timezone on-call with priority-based layers, each with own timezone, time/weekday restrictions, and rotation
  • New
    On-call Handoff Notifications — email alerts when rotation changes, SSE real-time events, "You" badge on schedule list
  • Improved
    Admin sidebar simplified — 7 links collapsed to single "Admin Panel" entry with quick-link dashboard
  • Fixed
    Discord Voice war rooms now create actual voice channels (was creating text channels)
  • Fixed
    On-Call Health Dashboard crash fixed (OrganizationContext import)
  • Fixed
    Trivy CI scans — added timeouts (were hanging 2+ days), fixed false-positive SAML secret detection
v0.18.0
2026-03-14

Incident Lifecycle Management

  • New
    Action Items Dashboard — track postmortem action items org-wide with status, priority, assignee, and due dates
  • New
    Root Cause Analytics — categorize incidents by root cause, view MTTR/MTTA trends, top noisy sources, and flapping detection
  • New
    Change Events — correlate deployments with incidents via CI/CD integration ([-60min, +5min] window)
  • New
    Bot Status Reminders — automatic reminders when incidents go without updates (Redis-backed, scalable)
  • Improved
    Stripe Checkout supports automatic payment methods (PayPal, BLIK, Przelewy24, and more)
  • New
    Billing page with plan details, usage, and subscription management
  • Improved
    Security hardening — encryption key startup guard, SSE per-IP connection limits
  • Improved
    Performance indexes for change events, incident analytics, and action items queries
v0.17.0
2026-03-05

Enterprise Features

  • New
    Teams & war rooms for incident management
  • New
    Platform identities for cross-platform user mapping
  • New
    Primary/secondary on-call assignments
  • Improved
    Cover request system for on-call swaps
v0.16.0
2026-02-27

Incident Management & Noise Reduction

  • New
    Inbound events API (PagerDuty-compatible)
  • New
    Alert grouping and deduplication
  • New
    Flapping detection
  • New
    Suppression rules
  • New
    Incident notes and activity timeline
v0.15.0
2026-02-27

On-Call & Escalation

  • New
    On-call schedules with daily/weekly rotation
  • New
    Escalation policies with multi-step chains
  • New
    On-call overrides
  • New
    Feedback system with Discord integration
v0.14.0
2026-02-16

Incident Acknowledgment

  • New
    One-click incident ACK via email, Slack, Discord, Teams
  • New
    ACK suppresses repeated notifications
  • Improved
    CI/CD pipeline with Argo Workflows + Argo CD
v0.13.0
2026-02-14

Single Sign-On

  • New
    OAuth2 (GitHub, Google)
  • New
    SAML 2.0 support
  • New
    OIDC (generic with auto-discovery)
  • New
    Org-scoped SSO providers
v0.12.0
2026-02-11

Reports & Badges

  • New
    PDF/CSV/JSON report generation
  • New
    SVG status badges (shields.io style)
  • New
    Atom feeds for status pages
  • New
    SLA tracking (24h/7d/30d/90d)
v0.11.0
2026-02-10

SLI/SLO & Error Budgets

  • New
    SLO targets with error budget tracking
  • New
    WhatsApp notifications
  • New
    Status page subscriptions
v0.10.0
2026-02-10

Notification Expansion

  • New
    MS Teams, OpsGenie, Jira integrations
  • New
    10 notification channel types total
  • New
    Telegram rich HTML messages
v0.9.0
2026-02-09

Security & Metrics

  • New
    CSRF protection (double-submit cookie)
  • New
    Prometheus metrics endpoint with OTLP support
  • New
    MFA (TOTP) with backup codes
  • Improved
    DNS checker improvements (multi-server, retry)
v0.8.0
2026-02-08

Geo Monitoring

  • New
    Distributed monitoring from 7 global locations
  • New
    DEGRADED status for partial outages
  • New
    Geo exclusion filtering
  • New
    Auto MTR diagnostics on timeout