Remove old helper, realtime and sentinel images by name with --format
instead of parsing the default table, which changed in Docker 29. Also
remove unused images pulled by digest.
Closes#12148
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
GitHub reports team-granted and private org members as CONTRIBUTOR or
NONE, so their same-repository PRs were silently skipped when public PR
deployments were off. A same-repository branch already requires push
access, so gate only on fork status, treat an unknown head repository as
a fork, and log skipped fork PRs.
Closes#9271
Retry the installation token request on GitHub 5xx and connection errors,
and report the HTTP status when GitHub sends no error message. Retry the
repository import when SSH drops the connection (exit code 255), after
stopping the interrupted clone and removing the partial checkout.
Closes#8004
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
CheckHelperImageJob now logs a warning and returns when the versions
endpoint fails instead of sending an internal notification and
rethrowing. The next scheduled run retries. PullTemplatesFromCDN also
no longer sends internal notifications on non-success responses or
exceptions; it still logs them.
Add tests checking that the stored helper version is kept and that no
notification is sent when the CDN is unavailable.
- Fetch every page of pull request files so watch paths match PRs with more than 30 files.
- Deploy the preview and log a warning when the changed-file list is unknown (API error or truncated compare list).
- Log when watch paths decline a preview deployment.
- Keep the GitHub App installation token in memory per instance, so a deployment requests one token instead of 3-5.
Closes#12169
Instance backup settings, started ClickHouse/KeyDB/Dragonfly databases,
and destinations saved read-only fields back to the database, so a
password manager autofill could overwrite stored credentials.
Closes#12173
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Load the running containers of a service even when the service status is
degraded, instead of blocking the whole terminal. Also respect the server
terminal setting for service databases.
Closes#12174
- Refresh application, database and volume backup lists on BackupCreated
- Refresh backup execution, edit and service resource views on
ServiceStatusChanged as well as ServiceChecked
- Hide the deployments counter when only the current page's application
is deploying, since its heading already shows "Deploying…"
- Show storage section shortcuts in the service configuration sidebar
- Drive server sidebar Sentinel/Security active state from activeMenu
and activeSubMenu instead of route matching
- Start the app sidebar expanded for new users
- Drop a column from the backup executions table grid and adjust the
responsive breakpoints
- Update the related feature tests
Hold saved card and group positions on the canvas until a load that
started after the save completes, so stale poll or refresh responses
no longer snap cards back to their old spots.
Use GROUP_GAP instead of BOX_GAP when dragging groups and checking
group spacing in dragRules, matching layoutServerView so a reload
does not shift a dropped group. Save before ending the group drag.
Add tests covering group spacing after drag and reload.
Add a "Connect to resources" modal to the tag show page for small
screens, where the tags canvas is hidden. A new toggleResource action
attaches or detaches the tag on a team resource after an authorization
check, then refreshes the tag's applications and services.
Also improve responsive layouts:
- Wrap section headings, project/environment toolbars and view toggles
on narrow screens; keep modal headers from shrinking.
- Move the environment settings link to an icon next to the environment
name and link the project name back to its environments.
- Group project icon, name and description into one "Project details"
form with an unsaved-changes bar.
- Show dashboard "view all" links for projects and servers only when
there are more items than the dashboard limit.
Reorder the Instance settings links in the navbar and the settings
layout sidebar so Authentication comes first, then Email, then Backup.
The navbar section divider moves to Authentication, since it is now the
first item in the group.
Render the member "Manage" dropdown via x-teleport and anchor it to the
trigger button with x-anchor so it is no longer clipped by overflow
containers. Outside clicks on the trigger are ignored to prevent
immediate re-toggling.
Also hide the dashboard activity column unless it contains a section.
Environments without resources now appear as placeholder cards on the
project, team projects and server canvases. They sit in the zone that
holds most of their project's cards, or in a new "Empty environments"
zone when there is none. A new EmptyCard component renders them.
- Project canvas: admins can move groups and cards, and the moves are
saved on their environment canvas. Environments are ordered by id.
- Server canvas: admins can move single cards as well as groups.
- Shared variable canvas: built-in COOLIFY_SERVER_UUID/NAME variables
and their links are hidden on server scope.
- Main nav: resource menus register through registerMainNavContext()
with depth levels (#main-nav-context-{depth}). Back goes one level
lower, and Back at depth 0 returns to the main menu.
- The deploy canvas state, styles and controls are simplified, and the
related sidebar, navbar and dashboard views are updated.
Tests are added or updated for the canvases and the main nav resource
menus.
Add a /dashboard/canvas page that renders every team server with its
resources and their running or queued deployments, backed by a new
ServerCanvasController::dashboard endpoint and a per-server usage
endpoint for zone headers. The dashboard remembers the list/canvas
choice and each section (projects, servers) its own table/grid view.
- Show Deploying/Queued pills on canvas cards and a foldable recent
deployments panel; add shared CanvasControls component
- Allow backup downloads from the server host via the S3 download
storage (PrepareFileDownloadJob accepts a null container)
- Extract backup toggle/run actions into ManagesDatabaseBackup and
surface them in backup execution headers; merge S3 and retention
sections into the general backup settings
- Move deployments indicator to the top bar and add a main nav
resource menu with submenus; simplify breadcrumbs and settings
layouts
- Add shared variables overview canvas and scheduled task modals
- Add feature tests for dashboard canvas, view modes, navbar
submenus, backup S3 downloads and scheduled task modals
Run metric requests in batches of eight workers, isolate endpoint failures,
and preserve CPU and memory history when Sentinel lacks a summary endpoint.
Repair Sentinel tokens before creating workers and bound SSH fetch timeouts.
Reauthorize team servers on refresh and hydration, clear stale metrics after
transfers or deletions, and resolve container metadata for team ID zero.
Use only root filesystem samples when mount fields are present, while
supporting legacy disk data. Add regression coverage for concurrent workers,
access changes, endpoint failures, and disk mapping.
The serversideup/php base image now defines its user bundle (nginx,
php-fpm) in /etc/s6-overlay/user-bundles.d. A bundle in s6-rc.d/user
has priority over it, so the dev container started without nginx and
php-fpm and did not become healthy.
Enable Fortify passkeys (laravel/passkeys) with the @laravel/passkeys
client. Users can sign in with a passkey, confirm sensitive actions with
one, and add, rename, and delete passkeys on the Profile page.
- Resolve the relying party for each request from the instance URL, then
APP_URL, then localhost. Passkey routes return 404 on other hosts.
- Add the is_passkey_enabled instance setting (Settings > Advanced).
- Use a hex user handle (laravel/passkeys-server#41).
- Rate limit the passkey routes by client IP.
- Record audit events for registered, verified, and deleted passkeys.
Andras Bacsai
Aditya TripathiandAndras Bacsai
