SikurdHelp
Open Sikurd

Alerts

Inbox of every active and recently resolved alert.

/dashboard/alerts is your alert inbox. Every problem Sikurd detects across the fleet lands here.

Top stat tiles

Four clickable tiles deep-link into pre-filtered views of the table:

  • Critical → severity=CRITICAL, unresolved.
  • Warning → severity=WARNING, unresolved.
  • Info → severity=INFO, unresolved.
  • Resolved Today → resolved=true (filtered to today).

Routing summary card

Below the tiles, a "Routing rules" card surfaces a digest of where your alerts currently go ("5 of 11 alert types have custom routing") and a Configure routing → CTA. Always reads "what's wired up right now" so it's never stale.

The table

Sortable + filterable list of alerts:

  • Severity — Critical (red) / Warning (yellow) / Info (blue).
  • Type — Instance Down, License Threshold, License Expiry, High CPU, High Disk, High Memory, Calls Threshold, Backup Failed, Trunk Down, Service Down.
  • Instance — the affected PBX.
  • Created + Resolved timestamps.
  • Routing status — chips for each channel (Push, Email, Slack, PSA, Webhook, AI Voice) showing whether the alert was actually sent there.

Filters

  • Status — Open / Resolved / All.
  • Severity — All / Critical / Warning / Info.
  • Instance — narrow to one PBX.
  • Type — narrow to a single alert type.

Per-row actions

  • Resolve — closes the alert + (if PSA is on) closes the PSA ticket.
  • AI summary — generates a one-paragraph summary of the alert, likely cause, and suggested next action. (Pro+)
  • Acknowledge — quiets escalation without resolving.
  • Mute for N hours — suppress repeats temporarily.

Bulk resolve

Select multiple alerts with the checkboxes → Resolve N button at the top. There's also a Resolve ALL matching the current filters option in the toolbar for sweeping out chunks at once.

What gets alerted on

The default rules:

  • Instance Down — two consecutive missed polls.
  • Trunk Down — a SIP trunk fails to register.
  • License Threshold — extension usage crosses 80% then 95%.
  • License Expiry — 30 / 7 / 1 day before expiry.
  • Backup Failed — backup runs but ends in error.
  • Backup Stale — no successful backup in 7+ days.
  • High Disk — > 90% disk used.
  • Calls Threshold — concurrent calls cross 80% of the licensed ceiling.
  • Service Down — a 3CX-side service fails.

Configure routing per-type at Alert routing.