These updates are available on the beta channel for appliances on the 0.8.7 line (including RC builds).
cube diagLive CUBE fleet view. The CUBE Management page now updates in real time — keepalive countdowns tick live, connection state changes appear instantly, and the XCC traffic timeline streams live request/response activity with lifecycle/call filtering, all without waiting on a page refresh.
Clearer diagnosis when a CUBE gets stuck reconnecting. The fleet view and CUBE health dialog now explain why a gateway is stuck (e.g., "registered and probing, but reporting failures — not advancing to online") instead of showing an opaque status chip, along with a connection-history timeline and the live vs. configured probe interval.
New "Recent Peers" panel on the CUBE edit dialog surfaces the trunk IPs a gateway has recently talked to, and a live reconnect dialog shows session traffic as it happens.
New CUBE diagnostics bundle (cube diag) in the CLI collects gateway connection state, health history, and traffic capture into a single redacted, shareable bundle — making field troubleshooting of registration and connectivity issues dramatically faster.
New TLS verification toggle for outbound XCC connections and a raw XML capture toggle per CUBE for deeper diagnostic logging when you need it, without paying the overhead when you don't.
Domain health and alarms — per-gateway alarm counters (connection expired, reconnection failed, unregistered) roll up into overall CUBE domain health, with RTT/jitter trend charts and alerting when a gateway goes quiet.
Much faster CUBE dashboards — Overview, trunk performance, concurrent-calls, voice-quality, and peer-health pages that took many seconds (over a minute on large fleets) now load near-instant, even across multi-week and multi-month views.
Faster XCC call-metrics pagination and disconnect-code/hour-of-day breakdowns.
More reliable status detection — tracks the gateway's real, live probe behavior, so status changes show faster and more accurately, including under load.
Lower storage use for CUBE event history, without losing diagnostic detail.
CDR History AND/OR filtering — combine conditions with both AND and OR for more precise multi-condition searches.
Failed calls gained an "on behalf of" dimension, and extension summaries now show average call duration.
Faster, more consistent dashboards — including a near-real-time 5-minute view for short time windows.
Data-freshness indicators — sync status and freshness are clearly shown, so you always know how current the data is.
Cluster Daily Volume report — accurate total call-volume trends across clusters, sources, and time-of-day.
Dormant Devices report now shows the directory number for each dormant device, making it easier to spot lines ready for reclamation.
Combined dormancy presets now cover soft-phone clients in one view.
Call-path accuracy — Call Flow Investigator drill-downs use precise timestamps and correctly distinguish consultative from blind transfers, with distinct visual tags.
Redirect and call-span detection hardened for very long-duration and multi-leg calls.
Resilient phone scraping — when phones go offline in bulk (a call-manager reboot or network flap), Call Telemetry detects the pattern per phone and per subnet and backs off automatically instead of hammering unreachable devices.
Configurable scraping — concurrency, batch size, and pacing are set from the dashboard, with a fleet-wide cap to keep resource use predictable on very large deployments.
Fixed status flapping between registered and unregistered after a discovery run.
Workflow trigger system — workflows now fire automatically, not just manually:
Triggers: webhooks, schedules, policy rule matches, config changes, Teams emergency calls, or CDR events.
Webhook inbox shows every inbound hit (matched, filtered, rate-limited, or rejected) for easy debugging.
HTTP Request step lets workflows call out to other systems.
Run history kept by default, with an API and in-editor browsing.
Policy rules got a direct "edit app" shortcut from the Apps tab.
CURRI at scale — routing decisions now collapse to a single database write per call instead of several.
More consistent routing under load — stuck or duplicate policy states are detected and corrected automatically.
SIP session correlation — call legs, trunks, and CUBE devices are auto-linked, laying groundwork for network-wide call tracing.
Faster policy history — refreshes run in smaller, non-blocking chunks, so heavy reporting no longer competes with live call routing.
Fail-closed safety — calls are no longer waved through if a policy state comes back inactive or undetermined.
More reliable greeting playback — greetings are more dependable across more call scenarios (outbound, fast auto-answered), and now play when the call is answered rather than taking over the handset (force-answer stays opt-in for kiosk and pre-answer setups).
Downloadable "show-tech" diagnostic bundle — one-click JTAPI diagnostics for faster support.
Handset Readiness Timeout — a new advanced setting to tune how long to wait for a handset before playing a greeting.
Removed the confusing legacy conference-mode delivery option.
Faster, more consistent greeting decisions, with fewer delays before a greeting plays.
Unified data-collection controls — end-user, RIS, and AXL sync toggles are now consolidated into a single Data Collection view per CallManager.
Full data export/import — a System & Infrastructure page exports configuration and operational history as a single zip (routing rules, alarms, watch lists, CDR logs, automation history, and more) and imports it back, with a duplicate-record preview.
Live per-mount disk usage — see usage across every mount, with alerts before a secondary disk fills up.
Faster upgrade boot — healthcheck timing tuned so upgrades boot faster and don't false-fail during heavier migrations.
More accurate version and update-available reporting.
Greater database reliability and uptime under heavy load.
Near-instant data cleanup — old data is removed instantly instead of through slow row-by-row scans.
Proactive health alarms — early warnings on disk I/O, connection-pool pressure, table growth, and stalled jobs, before they become visible.
Faster page loads and sign-in.
Lower disk and memory footprint.
Much faster analytics and dashboards — call-volume and concurrent-call queries that took tens of seconds to over two minutes now return in under a second, and stay responsive even during brief database contention.
Isolated reporting — heavy reporting jobs run on separate data paths, so they no longer compete with live call processing.
Update with:
sudo curl -fsSL https://get.calltelemetry.com | sudo sh -s -- update beta