screentinker/server/services
ScreenTinker bcfe3eaf8b fix(#148) Items 2-4: mark-offline closes the socket + tighten ping + TCP keepalive
Item 2: when the heartbeat checker marks a device offline it now also disconnects any socket
it still holds for it, so DB-offline can't diverge from socket-state into a silent half-open
(defensive — the live-socket guard already defers genuinely-live sockets).

Item 3: tighten half-open detection WITHOUT reintroducing the TV-WebKit decode-load risk the
30s pong-timeout was chosen for — lower only pingInterval 30s->15s (probe more often), KEEP
pingTimeout at 30s. Detection = interval+timeout = 45s (was 60s), and the client inherits
these via the handshake so BOTH ends detect a dead peer ~25% sooner. (Deliberately did NOT
drop pingTimeout to ~20s: MAXHUB is a video-playing TV-class device and the code comment
warns tighter timeouts cause spurious drops under decode load.)

Item 4: SO_KEEPALIVE on every accepted connection (lib/tcp-keepalive.js) so a half-open TCP
can't persist indefinitely at the OS layer, independent of the app ping.

Tests: server closes a non-ponging peer within ~pingInterval+pingTimeout while a ponging peer
survives; a device whose transport dies ends offline with its connection torn down; keepalive
applied to each accepted connection (and never breaks setup on error). Suite 328/328.
2026-07-02 14:59:25 -05:00
..
activationNudge.js feat(signup): T+3 activation nudge for users with zero paired screens 2026-05-30 20:28:24 -05:00
activity.js feat(workspaces): rename via switcher dropdown - new PATCH /api/workspaces/:id route, per-row pencil affordance in switcher (visible only when caller can_admin), small rename modal with name + slug fields, validation (name <=80 chars, slug ^[a-z0-9]+(?:-[a-z0-9]+)*$ <=60 chars, blank slug -> NULL), 409 on per-org slug collision. Permission gating via new canAdminWorkspace(db, user, ws) helper in lib/permissions.js - reused-ready for future Phase 3 admin actions. /me query now joins organization_members to compute can_admin per accessible_workspaces entry. Drive-by fixes surfaced: (1) activityLogger method filter was missing PATCH, added; (2) routes that operate on a target workspace by URL param need to stamp req.workspaceId from the param so activityLogger captures the right tenant attribution - documented in the route. Smoke fixture: switcher-test@local.test is workspace_admin of Studio A and workspace_editor of Field Crew (no org_owner) so the can_admin true/false split is exercised in one login. 2026-05-12 11:06:55 -05:00
agency-digest.js feat(api): batched email digest for agency uploads (#73) 2026-06-14 13:59:37 -05:00
alerts.js feat(email): Microsoft Graph send + alert spam protection + preferences UI 2026-05-12 18:16:40 -05:00
email.js feat(signup): welcome email + admin signup notification (slice 1) 2026-05-30 14:50:27 -05:00
heartbeat.js fix(#148) Items 2-4: mark-offline closes the socket + tighten ping + TCP keepalive 2026-07-02 14:59:25 -05:00
loop-lag.js fix(#146) P3.7: coalescer carries the PEAK numeric over the window 2026-06-30 22:10:45 -05:00
scheduler.js feat(scheduling): per-item schedule blocks (#74 dayparting, #75 auto-expire) 2026-06-11 15:46:41 -05:00
signupEmails.js fix(signup): make admin-notify recipient env-driven, not hardcoded 2026-05-30 16:16:33 -05:00