Use the shared GitHub app scope when listing and loading private apps so system-wide apps owned by another team remain available. Update coverage for mounting and loading repositories through those apps.
Create the root team before production seeding depends on it, reuse the
existing root team when creating root users, and cover the production seeder
flow with a feature test.
Trim comma-separated database host values and fall back to DB_HOST or the default host when write hosts are empty. Add unit coverage for read/write host parsing.
Wrap destination promotion in a transaction so the main destination swap and additional network updates stay consistent. Add coverage for promoting an owned team network while preserving the previous main destination as an additional network.
Ensure destination attach and promote operations only accept networks that belong to the selected server, preventing mismatched same-team server/network pairs.
Forget cached storage threshold state when reported disk usage drops below the alert threshold, allowing future threshold crossings to dispatch a fresh storage check.
Quote SCP operands when building commands to prevent shell injection through source or destination paths, and cover the escaping behavior in the SSH command injection tests.
Drop the unused GitHub release and version sync options from sync:bunny,
leaving the command focused on BunnyCDN template, release, and nightly syncs.
Update the nightly test to assert it does not invoke gh or git.
Check the selected S3 storage against the database at submit time so
stale Livewire state cannot schedule backups with storage that was
reassigned or marked unusable after the component mounted.
Prevent scheduled database backups from enabling S3 uploads without a valid team-owned storage configuration, and preserve the previous S3 storage ID in missing-storage error messages.
Add coverage for backup edit/create validation and S3 upload failure messaging.
Wrap first-use SSH and SCP multiplexed commands with a lock to avoid racing while the control socket is created. Also detect native OpenSSH mux master process names during stale connection cleanup and cover both orphaned and duplicate mux processes with tests.
Remove explicit SSH master pre-warming and lock handling so OpenSSH manages ControlMaster creation lazily from real ssh/scp commands. Add cleanup for duplicate mux processes and update coverage around mux command options and stale process cleanup.
Serialize multiplexed SSH master creation per server to avoid concurrent workers spawning orphaned processes. Enable scheduled cleanup for stale mux connections and add guarded orphan process reaping with tests.
Move log stream polling off the loading indicator so non-collapsible log panels continue polling while streaming, and cover the behavior with a Livewire feature test.
- resolve the GitHub App by a stable identifier during installation
callbacks so installing and re-installing keeps working over the
full lifetime of the App
- verify the installation id received from the callback against the
GitHub API before persisting it
- support re-installing an already configured GitHub App instead of
blocking it
- require an authenticated session and rate limit the setup callback
routes
- extend manifest setup state validity to match GitHub's manifest
code lifetime
Adds feature coverage for the GitHub App setup and installation
callbacks.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Centralize scheduled job queue selection with crons_queue() and use it for scheduler, task, and database backup jobs so cloud runs on crons while self-hosted stays on high.
Use config-based queue selection for deployment and scheduled jobs so cloud dispatches deployments to `deployments` and scheduled jobs to `crons`, while self-hosted keeps using `high`.
Add coverage for deployment queue helper, start action routing, and scheduled job manager routing.
Git hosts treat owner/repo names case-insensitively, but the exact
repository match used a case-sensitive comparison, so a payload whose
casing differed from the stored git remote would fail to match and
skip a legitimate deployment.
Lowercase both canonical repository paths before comparing.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The manual webhook handlers selected target applications with a
`git_repository LIKE %full_name%` substring query, so a payload
repository name could match unintended applications when repository
names overlap.
Add a `MatchesManualWebhookApplications` trait that validates the
incoming `owner/repo` value and matches `Application.git_repository`
by exact normalized path. Github, Gitlab, Gitea and Bitbucket manual
handlers now use it, reject invalid repository input early, and return
a consistent generic webhook failure payload.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Resolve the server and network in Destination::addServer() and
::promote() through ownedByCurrentTeam() before use, authorize the
update against the resource, and pass the validated IDs into
attach()/detach()/update(). Errors are routed through handleError()
to match the sibling removeServer() method.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Move Sentinel push handling into a controller and dispatch server update jobs only when container state changes or the force interval elapses. Add opt-in PostgreSQL read/write replica configuration and tune periodic proxy network and storage checks to reduce unnecessary work.
Add feature coverage for replica config, Sentinel push deduplication, deployment log scrolling, and server update job optimizations.
Scope DeleteEnvironment::mount() and delete() lookups through
Environment::ownedByCurrentTeam() so an environment_id that belongs to
another team resolves to a 404 instead of loading the foreign record.
Mark $environment_id as #[Locked] so the public Livewire property can no
longer be reassigned from the client.
Add tests/Feature/DeleteEnvironmentTeamScopingTest.php covering mount,
delete, the #[Locked] guard, and the team-scoped helper for both the
cross-team and own-team cases.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The Source component now resolves the supplied private key and Git
source IDs through team-scoped queries before persisting them, so a
selection can only ever reference a resource owned by the current
team. The source type is additionally restricted to the supported
GitHub/GitLab app classes.
The privateKeyId property is marked #[Locked] so it can only change
through the dedicated handler rather than a direct property update.
Adds feature tests covering team-scoped selection of private keys and
Git sources.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Clickhouse, Dragonfly, and Keydb still called syncData() inside the
DatabaseProxyStopped broadcast handler, clobbering in-progress edits to
name/description/credentials. Refresh only is_public/public_port/
public_port_timeout instead, matching the pattern used elsewhere.
Also null-guard HasDatabaseStatusInfo::getListeners() against an absent
Auth::user()/currentTeam(), add explicit return types on getListeners()
and render(), and convert inline comments in the SSL refresh test to a
PHPDoc block.
The earlier refreshStatus fix kept user-typed values intact but Livewire still
absorbed deferred wire:model values into the snapshot on every broadcast-
triggered roundtrip, clearing the unsaved-changes indicator and making the form
look auto-saved. Move all status-derived display (DB URLs, SSL toggle/mode,
cert expiry) out of each DB General form into a sibling StatusInfo Livewire
component, so the form never roundtrips on broadcasts.
Shared scaffolding lives in App\Traits\HasDatabaseStatusInfo plus an x-database-
status-info Blade component, leaving each per-DB StatusInfo class as a ~20-50
line declaration of label, SSL mode options, and SSL save hooks. Parents
dispatch databaseUpdated from save methods so the sibling refreshes after writes.
Tests cover the architecture (no DB form subscribes to status broadcasts) and
the sibling's refresh-on-status-change behavior.
Build the global destinations list from actual destination records so empty
servers do not render duplicate empty states. Allow creating Docker destinations
for a selected team server outside the global usable list, authorize swarm
creation correctly, and store discovered swarm network names from the selected
network. Add feature coverage for empty states, selected-server mounting, and
swarm destination creation.
Exclude generated Docker Compose SERVICE_FQDN, SERVICE_URL, and SERVICE_NAME variables from runtime, build-time, and build arg environments so stale stored values cannot override generated service names for preview deployments.
Call the preview environment variable relationship as a query when building the legacy configuration hash, and cover preview deployments with a regression test.
Dispatch configuration change events after saving application source and advanced settings, and refresh the configuration checker before showing redeploy diffs.
Store deployment configuration snapshots on application deployment queues and compare them against the current application state. Surface grouped pending changes in the configuration checker and use build-impact diffs to decide when an existing image can skip the build step.
Catch and report failures while updating closed pull request status so preview deployment cleanup still runs for closed GitHub pull request webhooks.
Add coverage for cleanup continuing when GitHub comment cleanup fails.
Add centralized stop grace period resolution for application settings and use it across manual stops, preview stops, and deployments. Validate the Livewire advanced setting against shared min/max constants and cover persistence, fillable creation, and fallback behavior with tests.
Decode base64 custom_nginx_configuration before model assignment so it is not double-encoded, and allow null values when clearing the setting. Add API coverage for create, update, invalid input, and clearing behavior.
Generate pull request preview image tags with both the PR id and commit
so different commits on the same PR do not reuse the same image tag. Sanitize
and truncate generated tags to stay within Docker tag limits.
Add model-level hidden fields for secrets, tokens, keys, notification
credentials, deployment logs, and environment values. Allow explicit
read:sensitive API access to reveal gated private keys and deployment logs,
and cover the behavior with feature and unit tests.
Require Docker buildx before Railpack builds, normalize environment
variable keys before validation, and align private deploy key API docs with
the supported dockerfile build pack.
Update Railpack seed examples to use the expected Flask start command and Go/Rust exposed ports. Adjust smoke coverage to run Symfony by default and accept reachable 4xx responses, and extend seeder tests for per-example branch and port assertions.
Avoid failing Stripe webhook processing when local subscriptions are missing, and cover ignored invoice/payment/subscription events with feature tests.
Move the deployments indicator inside the app layout state scope so it can react to the sidebar collapsed state, and add a layout test covering the responsive positioning.
Add shared environment variable key validation and normalization for Livewire forms and models, allowing Docker-compatible keys while rejecting invalid entries such as keys containing equals signs. Also quote Railpack build environment and secret arguments safely.
Build Railpack variables from generic build-time vars plus Railpack-specific vars, filter unrelated buildpack control vars, and ensure curl/wget deploy apt packages are present. Add coverage for standard and preview deployments.
Drop the unstable applications/dockercompose route and controller path now that
service creation is handled by POST /api/v1/services. Add coverage to ensure the
deprecated endpoint stays unregistered while the services endpoint remains
available.
Only eager load nested server settings for service responses when the API token has read:sensitive, preventing read-only tokens from exposing sentinel fields while preserving sensitive access.
Add compose PR fields to the Application hidden attributes and allow database list queries to eager load nested server settings when sensitive access is permitted.
Use random_bytes before hex and base64 encoding so generated env values
match the expected decoded byte lengths. Add Pest coverage for HEX and
REALBASE64 magic variables.
- `/mcp/enable` and `/mcp/disable` now use POST (state-mutating ops)
- `ListServices` queries DB directly instead of loading all projects into memory
- `ListApplications` validates tag arg rejects empty string (not just falsy)
Add feature tests covering railpack build pack via REST API and
Livewire UI components, plus a bash smoke test that deploys seeded
railpack-* example apps against the local dev stack and verifies
COOLIFY_*, SOURCE_COMMIT, and RAILPACK_* env vars land correctly.
Replace nested null checks with nullsafe operator chains, add ?Server
return type, drop unreachable database branch, and cover all paths with
feature tests.
Mirrors Nixpacks behavior: inject COOLIFY_* and SOURCE_COMMIT into
railpack build variables so apps (e.g. SPAs baking public URLs) can
read them via /run/secrets/<KEY>.
Rename is_nixpacks → is_buildpack_control to cover both NIXPACKS_ and
RAILPACK_ prefixed keys. Update the env variable view and appends list
accordingly.
Promote generate_coolify_env_variables to protected for testability.
Replace `railpack_environment_variables_collection()` helper (which returned
pre-filtered Eloquent attribute collections) with inline queries on
`environment_variables()` / `environment_variables_preview()` filtered by
`is_buildtime`. This ensures Railpack build variables are sourced from the
same query path as the rest of the deployment pipeline and avoids relying on
a now-removed accessor that silently included all railpack vars regardless of
build context.
Remove RAILPACK_FRONTEND_IMAGE env var from helper Dockerfile and resolve
the image ref at runtime using a new `railpack_version` constant in config.
Eliminates Docker build-time env interpolation for BUILDKIT_SYNTAX arg.
Replace 8× repeated per-type if-blocks in `queryDatabaseByUuidWithinTeam`
and `queryResourcesByUuid` with a single loop over the new
`STANDALONE_DATABASE_MODELS` constant.
Add unit tests to guard the registry against drift (keys mirror
`DATABASE_TYPES`, every entry is a valid Eloquent model with `team()`),
and feature tests covering team-ownership, wrong-team, and unknown-UUID
cases for `queryDatabaseByUuidWithinTeam`.
Add `exposeNestedServerSecrets()` to Applications, Databases, and
Services controllers so that `read:sensitive`/`root` tokens see
sentinel and logdrain fields on eager-loaded Server + ServerSetting
relations.
ServicesController handles both single models and Eloquent Collections
(listing endpoint passes a Collection per project).
Tests tightened to use JSON-key assertions (`"field":`) to avoid false
positives from field names appearing in values.
Models now declare $hidden for passwords, tokens, db URLs, and compose
fields. API controllers flip from makeHidden-on-deny to makeVisible-on-
allow (can_read_sensitive=true), fixing fields that were never hidden.
Also adds missing fields (mysql/mariadb passwords, logdrain keys, etc.)
to privileged disclosure lists.
Tests added: Feature/Security/ApiSensitiveFieldsTest and
Unit/Models/SensitiveFieldsHiddenTest cover all affected models and
controllers.
Replace random string with encrypted JSON payload containing
server_uuid, binding token to its server for validation.
Remove double-encrypt test no longer relevant to new token format.
Replace hard validation error with self-healing token logic. Tokens that
are null, empty, or fail decryption are now regenerated automatically
rather than crashing sentinel startup or metrics reads.
Token format changed from encrypted JSON payload to a plain 64-char
random string (Str::random), eliminating double-encryption issues and
simplifying the validation regex to cover the new character set.
New `ensureValidSentinelToken()` method on ServerSetting centralises
the get-or-regenerate contract; both StartSentinel and HasMetrics now
delegate to it. HasMetrics logs a warning when regeneration occurs so
operators know a sentinel container restart is required.
`isValidSentinelToken()` now accepts `?string` (null → false).
Adds feature tests covering: null/empty/undecryptable stored values,
idempotent return of valid tokens, RuntimeException only when
regeneration itself produces an invalid token, no double-encryption of
newly generated tokens, and cast round-trip consistency.
On cloud, DeploymentException and TimeoutExceededException are expected
failure modes that pollute the Horizon failed jobs UI. Listen to JobFailed
events and scrub the entry via JobRepository::deleteFailed so operators
are not alerted for noise failures. Self-hosted instances are unaffected.
Add Model Context Protocol server exposing Coolify infrastructure data
to AI assistants. Includes tools for listing/fetching servers, projects,
applications, databases, and services, scoped to authenticated team tokens.
- Add CoolifyServer with 10 read-only tools (list/get for all resource types)
- Add BuildsResponse and ResolvesTeam traits for shared tool logic
- Add EnsureMcpEnabled middleware guarding /mcp routes
- Add enable/disable MCP API endpoints (root-only)
- Add is_mcp_server_enabled toggle in instance settings and advanced UI
- Add migration for is_mcp_server_enabled column
- Add feature tests for MCP endpoints and toggle API
- Scrub sensitive keys (passwords, tokens, raw IDs) from all responses
Add DetectsSkipDeployCommits trait with two strategies: shouldSkipDeploy
(all commits must contain the marker) for push events, and
shouldSkipDeployAny (any single marker triggers skip) for PR/MR titles
and latest-commit signals.
Apply trait to Bitbucket, Gitea, GitHub, GitLab webhook controllers and
ProcessGithubPullRequestWebhook job. PRs pass pullRequestTitle through
to the job constructor for evaluation.
Large host files mounted via Docker volumes caused the storages page to
become unusable — full file content was stored in the encrypted mediumText
column and serialised into the Livewire payload, crashing the browser.
- Add MAX_CONTENT_SIZE (5 MiB), BINARY_PLACEHOLDER, and TOO_LARGE_PLACEHOLDER
constants to LocalFileVolume
- Check remote file size via stat/wc before cat in loadStorageOnServer and
saveStorageOnServer; store placeholder instead of content when limit exceeded
- Expose is_too_large computed attribute (appended for Livewire serialisation)
- Guard submit, instantSave, and syncData in FileStorage Livewire component
- Truncate oversized content in Storage::refreshStorages to prevent payload bloat
- Show distinct warning banner in file-storage blade; mark textarea readonly and
hide Save/Convert buttons for too-large files
- Add unit tests covering constants, computed flags, and toArray serialisation
Fixes#4701
Add `connection_timeout` field to server settings, allowing per-server
override of the global SSH connection timeout constant.
- Migration adds `connection_timeout` integer column (default 10s)
- `ServerSetting` model exposes and casts the new field
- `SshMultiplexingHelper::getConnectionTimeout()` resolves per-server
value with fallback to `constants.ssh.connection_timeout`
- All SSH/SCP command builders use the new resolver instead of the
global config directly
- Livewire `Show` component binds `connectionTimeout` with validation
(1–300 seconds) and syncs to/from the model
- UI input added to server settings form with helper text
- Feature tests cover default, persistence, resolver, and fallback
Move reachability notification triggering out of isReachableChanged into
a dedicated ServerReachabilityChanged event dispatched by
ServerConnectionCheckJob. Remove the blocking 3-attempt sleep loop from
isReachableChanged — unreachable_count threshold alone now gates the
Unreachable notification. Add feature and unit tests covering all
notification dispatch paths.
Introduce a dedicated `audit` log channel (daily rotation, configurable retention via
LOG_AUDIT_DAYS) and a small `auditLog()` / `auditLogWebhookFailure()` helper used to
record state-changing API operations and webhook events.
Instrumented:
- API mutation endpoints (create / update / delete / start / stop / restart) across
applications, services, databases (incl. backups, env vars, storage), servers,
projects + environments, scheduled tasks, private keys, GitHub apps, cloud provider
tokens, Hetzner server provisioning, instance enable/disable.
- Webhook signature verification outcomes for GitHub, GitLab, Bitbucket, Gitea and
Stripe, plus the Sentinel push endpoint.
- Authentication and authorization outcomes via the global exception handler and
the `ApiAbility` middleware (unauthenticated, ability-denied, policy-denied).
The helper is wrapped in try/catch so logging failures never affect the request
path. Successful operations log at `info`; suspicious/denied requests log at
`warning`. Operators wanting a failures-only feed can set `LOG_AUDIT_LEVEL=warning`.
Includes a feature test suite covering the helper, the webhook providers and the
new auth/authorization log paths.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Extract NIXPACKS_/RAILPACK_ prefix filtering into a reusable
`scopeWithoutBuildpackControlVariables` query scope on EnvironmentVariable.
Apply scope consistently to runtime vars, runtime preview vars, and
buildtime var generation in ApplicationDeploymentJob.
Refactor `generate_railpack_env_variables` to return a Collection.
Add `RAILPACK_FRONTEND_IMAGE` constant and bake it into the
coolify-helper Dockerfile as a build arg.
Add DevelopmentRailpackExamplesSeeder (dev/local env only) for
seeding example Railpack apps, wired into DatabaseSeeder.
Add tests:
- ApplicationDeploymentControlVarFilteringTest: verifies control vars
are excluded from runtime and buildtime envs
- DevelopmentRailpackExamplesSeederTest: verifies seeder behavior
- ApplicationDeploymentRailpackEnvParityTest: parity checks for env
handling across build/runtime paths
- Kill PTY and notify client after 30 min of inactivity (IDLE_TIMEOUT_MS)
- Buffer client messages during async auth/IP fetch to prevent race-condition
message loss on fast reconnects
- Replay last sent command after transient reconnect so PTY respawns without
user interaction
- Preserve scrollback on disconnect/reconnect; write visible timestamp markers
instead of wiping term state
- Handle idle-timeout sentinel on client with user-facing error message
Proxies (Cloudflare, nginx) drop idle WebSocket connections before the
application notices, leaving clients typing into dead sockets.
- Add server-side ping/pong heartbeat (30s) in terminal-server.js;
terminate unresponsive clients instead of letting connections go stale
- Move client keepAlive interval start to the connect event so it
restarts correctly after reconnects
- Remove hidden-tab keepalive short-circuit — server pings now own
liveness; suppressing client pings while hidden masked proxy drops
- Fix clearAllTimers to use clearTimeout for one-shot timers
- On visibility resume, probe with a 5s timeout instead of the default
35s so half-open sockets are detected quickly
- Bump coolify-realtime to 1.0.14 across all compose files
Adds stop_grace_period to application settings (seconds, 1-3600, default 30).
Used in place of the hardcoded docker stop -t 30 in the four places that
stop application containers: rolling update shutdown, manual stop, stop on
another server, and preview deployment stop.
Non-positive values fall back to the default via ($val > 0) ? $val : default,
with tests covering 0 and -10 so the cast does not blow up if a bad value
ever lands in the db.
Picks up Jack Coy's work from #7125 which went dormant. His commits are
squashed here with credit below.
Co-authored-by: Jack Coy <jackman3000@gmail.com>
Replace the flat character-class regex for SHELL_SAFE_COMMAND_PATTERN with
a token-aware alternation. The parser now recognizes explicit tokens
(`&&`, `||`, balanced single/double quotes, whitespace, and an unquoted
safe-char run) instead of a bag of characters, which lets us extend the
accepted grammar without loosening the guarantees.
New surface area, with tests:
- logical OR chaining (`make build || make clean`)
- shell globs and bang (`rm *.tmp`, `cp src/?.js dist/`, `! grep -q foo`)
- single-quoted arguments are now treated as balanced runs rather than
rejected per-character
Preserved surface area:
- && chaining, balanced "..." and '...' quotes, the previous safe path /
argument characters, and the existing error-path contract in
ApplicationDeploymentJob::validateShellSafeCommand().
Also refreshes the user-facing validation messages in General.php so the
allow/deny list shown on failure matches the new grammar.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Apply escapeshellarg() to the Postgres username before interpolating it
into the chown command used to fix SSL certificate ownership, matching
the handling already in place for StartMysql. This keeps the sink-side
escaping consistent across database actions, independent of upstream
input validation.
Also adjusts an assertion in DatabaseSslCredentialEscapingTest to match
the actual double-escaped output of executeInDocker, and adds Postgres
regression cases for subshell and semicolon payloads.
Validate new init-script filenames against path traversal and shell
metacharacters via a new validateFilenameSafe() helper, and harden the
write/delete paths with basename() + escapeshellarg() so legacy rows
still deploy and can be cleaned up without regressions.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Add optional expiration to personal API tokens. Users pick a duration
(1/7/30/60/90 days or Never) at creation time. Expired tokens are
rejected by Sanctum, pruned hourly by sanctum:prune-expired, and a
team notification fires ~24h before expiry so owners can rotate
before API calls start failing.
- ApiTokens Livewire component stores expires_at from expiresInDays
- Rework issued-tokens UI from card grid to table (matches other views)
- New ApiTokenExpirationWarningJob scheduled hourly (idempotent via RateLimiter)
- New ApiTokenExpiringNotification (email/discord/telegram/slack/pushover)
- api_token_expiring added to alwaysSendEvents so users cannot silence
expiry warnings from the per-event notification toggle UI
- sanctum:prune-expired cadence moved from daily to hourly
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Pattern enforcement now conditional on field being dirty (changed vs
saved value). Prevents false validation failures when existing records
hold legacy credential formats that pre-date the stricter regex rules.
Add ValidationPatterns helpers for database identifiers and passwords,
apply them across database Livewire components and the API controller,
encode MongoDB init script values via json_encode, and pass the MySQL
user through escapeshellarg when generating SSL chown commands.
Replace CMD-SHELL string interpolation with CMD exec-form arrays in
healthcheck configs for PostgreSQL, Dragonfly, KeyDB, and ClickHouse.
CMD-SHELL passes the string to /bin/sh -c, allowing command injection
through user-controlled fields (username, password, dbname). CMD
exec-form bypasses the shell entirely — each value is a discrete argv
element.
Fixes GHSA-gvc4-f276-r88p.
Adds regression tests covering semicolon, pipe, backtick, $(),
background operator, redirect, newline, and null-byte injection vectors.
Reject malformed --date values with a clear error before building any
shell command, and wrap every interpolated value (log paths, filter
expression, line count) in escapeshellarg() so filter options and date
values can no longer break out of the tail/grep pipeline.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Move the email-verification URL hash from sha1 to sha256 and verify it
directly in the controller using hash_equals, instead of going through
Laravel's EmailVerificationRequest (which only compares against sha1).
The signed URL still carries the authoritative HMAC; the hash upgrade
keeps the identity binding aligned with modern hashing guidance.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
The invitation-link login path previously marked the account as
email-verified as a side effect of authenticating, without the user ever
proving control of the mailbox. Remove that branch so every account
goes through the standard signed-URL verification flow.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Strip advisory identifiers (GHSA-*) from describe blocks, test
docblocks, and inline comments. Replace with plain descriptive
labels. Also clean up FQCNs to use imported class names and minor
style fixes (string concatenation spacing).
Introduce a dedicated `uploadBackup` ability on Application, Database,
Service, and ServiceDatabase policies (admin/owner only) and call
`$this->authorize('uploadBackup', $resource)` in `UploadController::upload`
so the backup-upload endpoint goes through the same policy layer as the
rest of the authorization refactor. Adds Pest coverage for each policy
variant plus HTTP-level checks.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Constrain dev_helper_version to Docker tag grammar
([A-Za-z0-9_][A-Za-z0-9_.-]{0,127}), re-validate before triggering the
helper image build, and interpolate the image reference via
escapeshellarg() when composing the docker build command.
Replace exception text in 5xx JSON responses with stable, action-specific
messages so API consumers get a consistent payload regardless of which
underlying client (Guzzle, PDO, filesystem) raised the exception. The
previous responses concatenated the raw upstream error, which produced
inconsistent messages and unnecessary noise for clients trying to parse
errors programmatically.
Touched endpoints:
- GET /api/v1/hetzner/{locations,server-types,images,ssh-keys}
- POST /api/v1/servers/hetzner
- DELETE /api/v1/databases/{uuid}/backups/{uuid}
- DELETE /api/v1/databases/{uuid}/backups/{uuid}/executions/{uuid}
- /download/backup/{uuid}
The RateLimitException branch and AuthenticationException flow keep their
existing curated messages.
Adds Pest coverage for the four Hetzner GET endpoints to lock the response
shape on upstream failure.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Reuse the existing SafeWebhookUrl rule on the S3 Storage endpoint field so
the create and edit forms go through the same URL-normalization path as
webhook settings. Adds a matching guard inside S3Storage::testConnection()
so background callers (scheduled backups, database import reuse) also
validate the endpoint before building the S3 client.
Also fixes an IPv6-bracket edge case in SafeWebhookUrl so `http://[::1]`
style hosts are normalized before the filter_var IP check — the rule's
own loopback test was already asserting this behaviour.
Add allowlist of backup file extensions (sql, sql.gz, tar, tgz, zip,
dump, bak, bson, archive, bz2, xz, and compound variants) and enforce
a 10 GiB maximum file size on the backup upload endpoint. Validation
runs early on each chunk using the dropzone metadata and again on the
assembled file. Also drops the unused createFilename helper and the
commented-out S3 block.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Tighten validation on volume name and host path inputs across Livewire + API storage endpoints and escape shell arguments in volume clone and compose preview cleanup paths.
Replace `ownedByCurrentTeam()` (session-based) with `ownedByCurrentTeamAPI($teamId)`
(explicit team ID) when resolving S3 storage in create_backup and update_backup.
Session-based team resolution is unreliable in API context where auth is token-based.
Add `S3Storage::ownedByCurrentTeamAPI(int $teamId)` scope and update feature tests
to use real model instances instead of Mockery mocks.
Prevent docker container prune from removing containers labeled as
database, application, or service types. Previously only proxy containers
were excluded, risking accidental cleanup of active resources.
- Validate content (required string, min:10, max:2000) in OtherController@feedback
- Register 'feedback' named rate limiter (3/min per user or IP) in RouteServiceProvider
- Apply throttle:feedback middleware to POST /api/feedback
- Forward to Discord with allowed_mentions.parse=[] and a 5s HTTP timeout
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
- Auto-generate a 40-char random secret for each manual_webhook_secret_* column on Application creation so new apps are never left with an empty secret.
- Add encrypted cast for the four webhook-secret columns; backfill migration re-encrypts existing plaintext values and fills missing ones.
- Reject webhook deliveries when the stored secret is empty (GitHub, GitLab, Bitbucket, Gitea manual endpoints).
- Bitbucket: require the sha256 algorithm prefix on X-Hub-Signature instead of trusting the client-supplied algo.
- GitLab: drop the ?? '' fallback on the token comparison.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Replace Referer-based redirects in Admin Index back() and switchUser()
with named routes (admin.index and dashboard) for consistent navigation
behavior independent of the request header.
Add tests verifying back() returns to admin.index, switchUser routes to
the dashboard, and the Referer header is no longer consulted.
Use find_destination_for_current_team helper across resource creation
flows and the destination controller. Pass full destination objects to
database creation helpers instead of UUIDs so team relationships are
resolved consistently before the resource is created or linked.
Add feature tests covering destination, backup storage, and resource
proof lookups across teams.
Add `DELETE /api/v1/applications/{uuid}/previews/{pull_request_id}` to
cancel active deployments, stop containers, and delete the preview
record via `CleanupPreviewDeployment`. Includes OpenAPI annotations,
input validation, and full feature test coverage.
Set unreachable_count to 3 and unreachable_notification_sent to true
on all team servers in subscriptionEnded(), so the existing cleanup
command can pick them up after the 7-day grace period.
Also adds feature tests for the subscription-ended cleanup flow and
casts server IP to string in existing unreachable server tests to fix
type comparison.
Add a Rocky-specific Docker install path to the stable and nightly install scripts, using Docker's documented RHEL repository flow. Include a unit test to lock in the Rocky repo selection and command set.
Refine mobile layout for project resource pages by making breadcrumbs and status blocks responsive, improving dropdown and checkbox touch behavior, and adding support for custom modal triggers. Add feature tests covering breadcrumb visibility and responsive checkbox layout.
Adjust the deployment view container classes so the logs panel fills the available viewport height instead of capping at 30rem. Add a feature test to lock in the full-height layout classes and prevent regressions.
Use Railpack's install env handling and dedicated CLI flags for build/start overrides, and forward install commands into docker build secrets so image builds stay aligned with prepare-time configuration. Update the railpack config test to cover the new command format.
- Implement railpack.json + generated config deep merging logic in
ApplicationDeploymentJob with JSON validation and assoc array checks
- Label Railpack as "Beta" in all build pack selectors and show a
visible beta badge when railpack is selected in new-app forms
- Add railpack-nodejs Fastify example to ApplicationSeeder
- Add ApplicationSeederTest and ApplicationDeploymentRailpackConfigTest
covering config merge behavior and seeder correctness
Change default build_pack from railpack back to nixpacks in all new
application flows (GithubPrivateRepository, GithubPrivateRepositoryDeployKey,
PublicGitRepository) and reorder the build pack dropdown so Nixpacks
appears before Railpack across all relevant views.
Add feature tests covering the nixpacks default and selector ordering.
Instead of nullifying source references on applications when a team is
deleted, transfer instance-wide GitHub/GitLab apps to the root team
(team_id=0) so they remain available to other teams that depend on them.
Non-instance-wide sources are still deleted along with the team.
Refactor upgrade state initialization into a shared `refreshUpgradeState()`
method used by both `mount()` and `checkUpdate()`. The method now uses
`version_compare` to validate upgrade availability and clears the
`new_version_available` flag in InstanceSettings when the current version
is already equal to or newer than the latest version, preventing stale
upgrade notifications from persisting after a successful update.
Permit single-quoted arguments in SHELL_SAFE_COMMAND_PATTERN while
keeping dangerous metacharacters blocked, and add security test cases
for quoted --entrypoint and --hostname values.
Limit team cleanup to apps owned by the deleted team and nullify cross-team application source references before deleting team-owned sources. Adds feature tests covering user deletion with GitHub app-backed applications, preserving system-wide apps, and nullifying external source links.
Add `WithoutOverlapping` middleware to `DatabaseBackupJob` keyed by backup ID
with timeout-based lock expiry to prevent concurrent runs.
Mark long-running backup executions as failed when they exceed the stale
time threshold, and add periodic retention enforcement in
`CleanupInstanceStuffsJob` with cache-based throttling.
Also add float casts for retention max-storage fields on
`ScheduledDatabaseBackup` and comprehensive feature tests covering
overlap middleware, stale detection, casts, and retention behavior.
Parse `/tree/...` URLs by first capturing the full branch candidate, then
iteratively resolving valid branch names for GitHub API lookups and deriving
the remaining path as base directory. Also adjust env var editor/input view
classes (`font-sans`, `w-full`) and add/extend feature tests for both branch
parsing and multiline toggle rendering.
Covers Upgrade Livewire component mount behavior for:
- initializing latest version from cached versions data
- falling back to 0.0.0 when versions cache is unavailable
Add `monospace` prop to Textarea component so font-mono is no longer
applied by default. Apply it explicitly to env variable editors, private
key fields, and shared variable forms where monospace is appropriate.
Use Alpine.js x-data/x-model to make the multiline toggle reactive
without a full Livewire round-trip. Add wire:key on the input/textarea
wrappers to force proper DOM replacement when switching modes.
Allow SharedVariables Livewire components (Environment, Project, Server)
to accept UUID parameters directly via mount() instead of relying solely
on route parameters. This enables Livewire component testing without a
live route context.
Also adds comment field support when saving/updating server shared
environment variables, guards PostgreSQL-specific migration statements
from running under SQLite (test environment compatibility), and expands
the feature test suite with server shared variable scenarios including
inline comment preservation and update behaviour.
Extract duplicated inline StandaloneDocker attribute arrays in the
Server boot lifecycle into a dedicated method, eliminating repetition
between the root-server (id=0) and normal-server paths.
Also harden the shared_environment_variables migration by wrapping
DDL statements in DB::transaction() and using DROP CONSTRAINT IF EXISTS
to make the migration safely re-runnable.
Add unit test covering the extracted method to verify uuid is always
present in bootstrap attributes.
Replace all uses of `forceFill`, `forceCreate`, and `forceFill` with their
non-force equivalents across models, actions, controllers, and Livewire
components. Add explicit `$fillable` arrays to all affected Eloquent models
to enforce mass assignment protection.
Add ModelFillableCreationTest and ModelFillableRegressionTest to verify that
model creation respects fillable constraints and prevent regressions.
Clear `branches` and `total_branches_count` in `loadRepositories` to avoid stale branch data after repo refreshes. Update the Livewire view to use the shared loading button pattern for refresh/load actions, and expand feature coverage for repository refresh behavior and refresh button visibility.
Handle database status updates more reliably by listening for `ServiceChecked`
and using explicit `refresh()` handlers in Livewire database components.
Also switch guarded clone/create paths to `forceFill`/`forceCreate` in helper
flows to avoid missing persisted attributes during app/service cloning.
Update log/terminal font stacks to Geist (with bundled variable fonts) and add
coverage for SSL status refresh, persistent volume UUID cloning, and log font
styling.
Add logic to populate docker_compose_domains in applicationParser
for dockercompose build packs, ensuring proper domain handling
for KEY-based SERVICE_FQDN variables.
Add end-to-end support for `docker_registry_image_tag` in preview and deployment queue flows.
- Extend deploy API to accept `pull_request_id` alias and `docker_tag` for preview deploys
- Persist preview-specific Docker tags on `application_previews` and `application_deployment_queues`
- Pass tag through `queue_application_deployment()` and de-duplicate queued jobs by tag
- Update deployment job logic to resolve and use preview Docker tags for dockerimage build packs
- Update Livewire previews UI/state to manage per-preview tags and manual preview/tag inputs
- Add migration for new tag columns and model fillable/casts updates
- Add feature and unit tests covering API behavior and tag resolution
Prevent null CA certificate access during database SSL certificate regeneration
across KeyDB, MariaDB, MongoDB, MySQL, PostgreSQL, and Redis components.
If no CA certificate exists, attempt to generate one and re-query; if still
missing, dispatch a clear error and stop regeneration gracefully.
Add `SslCertificateRegenerationTest` coverage for missing-CA and CA-query
scenarios to prevent regressions.
Replace create() with forceCreate() across internal model creation operations to bypass mass assignment protection. This is appropriate for internal code that constructs complete model state without user input.
Add InternalModelCreationMassAssignmentTest to ensure internal model creation behavior is properly tested. Optimize imports by using shortened Livewire attribute references and removing unused imports.
Add explicit fillable attributes to Service and all Standalone* database models
for new configuration options: public_port_timeout, enable_ssl, ssl_mode,
is_log_drain_enabled, is_include_timestamps, and custom_docker_run_options.
Add tests to MassAssignmentProtectionTest to verify these attributes are
properly protected by mass assignment protection across all relevant models.
Ensure that the uuid field is preserved during clone operations for persistent
volumes across all clone methods (CloneMe, ResourceOperations, and the clone_application
helper). This prevents UUID conflicts and ensures cloned volumes receive new unique
identifiers as intended.
Adds test coverage validating that cloned persistent volumes receive new UUIDs
distinct from the original volumes.
Restrict mass-assignable attributes across user/team/redis models and
switch privileged root/team creation paths to forceFill/forceCreate.
Encrypt legacy ClickHouse admin passwords via migration and cast the
correct ClickHouse password field as encrypted.
Tighten API and runtime exposure by removing sensitive team fields from
responses and sanitizing Git/compose error messages.
Expand security-focused feature coverage for command-injection and mass
assignment protections.
Assign the selected redirect option before validation so valid changes are saved.
Add feature tests to verify redirect persistence and rejection when no www domain exists.
Replace shared `changePasswordFieldType` JS with component-local Alpine logic
across input, textarea, and env-var-input components. This keeps toggle
behavior consistent, resets visibility on `success` events, and preserves
`truncate` styling only when showing plaintext on enabled fields.
Also adds `PasswordVisibilityComponentTest` to verify Alpine bindings are
rendered and legacy handler references are removed.
Add CRUD tag endpoints (GET/POST/DELETE) as sub-resources for applications,
databases, and services. Add team-level GET /tags endpoint. Extend all
resource creation endpoints to accept an optional tags array.
Uses a shared HandlesTagsApi trait to avoid duplication across controllers.
Tags are race-safe via syncWithoutDetaching(), garbage-collected when orphaned,
and sanitized (strip_tags + lowercase).
Use firstOrFail() for team-scoped project and environment lookups across
new-project Livewire flows so missing or cross-team UUIDs fail closed.
Also dispatch an error when boarding selects a non-owned project, and
update IDOR feature tests for the new error/exception behavior.
Add support for configuring public port timeout on databases via API:
- Add public_port_timeout field to schema documentation with 3600s default
- Add validation rules (integer|nullable|min:1)
- Update all database type configurations to support the field
- Add comprehensive test coverage for the feature
Replace $guarded = [] with explicit $fillable whitelists across all
models. Update controllers to use request->only($allowedFields) when
assigning request data. Switch Livewire components to forceFill() for
explicit mass assignment. Add integration tests for mass assignment
protection.
Ensure Server and Project lookups in Livewire components and API
controllers use team-scoped queries (ownedByCurrentTeam / whereTeamId)
instead of unscoped find/where calls. This enforces consistent
multi-tenant isolation across all user-facing code paths.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add strict validation for Docker network names using a regex pattern
that matches Docker's naming rules (alphanumeric start, followed by
alphanumeric, dots, hyphens, underscores).
Changes:
- Add DOCKER_NETWORK_PATTERN to ValidationPatterns with helper methods
- Validate network field in Destination creation and update Livewire components
- Add setNetworkAttribute mutator on StandaloneDocker and SwarmDocker models
- Apply escapeshellarg() to all network field usages in shell commands across
ApplicationDeploymentJob, DatabaseBackupJob, StartService, Init command,
proxy helpers, and Destination/Show
- Add comprehensive tests for pattern validation and model mutator
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add #[Locked] attributes to security-sensitive properties (resource, servicesubtype,
server, container) to prevent client-side modification via Livewire wire protocol.
Add container name validation using ValidationPatterns::isValidContainerName() and
server ownership authorization via Server::ownedByCurrentTeam() in both getLogs()
and downloadAllLogs() methods.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add shellSafeCommandRules() validation to install_command, build_command,
and start_command fields in both the Livewire UI and REST API layers.
These fields previously accepted arbitrary strings without validation,
unlike other shell-adjacent fields which already used this pattern.
Also adds comprehensive tests for rejection of dangerous input and
acceptance of legitimate build commands.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replace mt_rand/rand with random_int for stronger randomness guarantees
in verification code generation and Blade component keying.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Move the admin panel route into the existing auth middleware group and
replace client-side redirects with server-side abort calls in the
Livewire component. Extract shared authorization logic into reusable
private methods.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add SafeWebhookUrl validation rule to notification webhook URL fields
(Slack, Discord, custom webhook) to enforce safe URL patterns including
scheme validation and hostname checks.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Delegate host validation to parent class instead of custom implementation
- Update base_url() helper to use config('app.url') instead of url('/')
- Add test for APP_URL fallback when no FQDN or public IPs configured
- Remove dedicated TrustHostsMiddlewareTest (logic now tested via integration tests)
Extract and return the billing interval (month/year) from subscription pricing
data in fetchPricePreview. Update the view to dynamically display the correct
billing period based on the preview response instead of using static PHP logic.
- Add support for bracketed IPv6 addresses when FQDN is not configured
- Harden password reset URL generation against X-Forwarded-Host header poisoning
- Add test coverage for IPv6-only configurations with malicious headers
- Update imports and clean up exception syntax in shared helpers
- Fix circular cache dependency in TrustHosts where handle() checked cache
before hosts() could populate it, causing host validation to never activate
- Validate both Host and X-Forwarded-Host headers against trusted hosts list
(X-Forwarded-Host is checked before TrustProxies applies it to the request)
- Use base_url() instead of url() for password reset link generation so the
URL is derived from server-side config (FQDN / public IP) instead of the
request context
- Strip port from X-Forwarded-Host before matching (e.g. host:443 → host)
- Add tests for host validation, cache population, and reset URL generation
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Escape dynamic error messages with htmlspecialchars() before
concatenating into HTML strings stored in validation_logs. Add a
Purify-based mutator on Server model as defense-in-depth, with a
dedicated HTMLPurifier config that allows only safe structural tags.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Refactor the invitation acceptance flow to use a landing page pattern:
- GET shows invitation details (team name, role, confirmation button)
- POST processes the acceptance with proper form submission
- Remove unused revoke GET route (handled by Livewire component)
- Add Blade view for the invitation landing page
- Add feature tests for the new invitation flow
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add SafeExternalUrl validation rule that ensures URLs point to
publicly-routable hosts. Apply to all GitHub source entry points
(Livewire Create, Livewire Change, API create and update).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Inline PrepareCoolifyTask and CoolifyTaskArgs into remote_process(),
removing two single-consumer abstraction layers
- Add #[Locked] attribute to ActivityMonitor $activityId property
- Add team ownership verification in ActivityMonitor.hydrateActivity()
with server_uuid fallback and fail-closed default
- Store team_id in activity properties for proper scoping
- Update CLAUDE.md to remove stale reference
- Add comprehensive tests for activity monitor authorization
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add proper shell escaping for persistent volume names when used in
docker volume rm commands. Also add volume name validation pattern
to ValidationPatterns for consistent input checking.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Reduce load on unreachable servers by implementing exponential backoff
during connectivity failures. Check frequency decreases based on
consecutive failure count:
0-2: every cycle
3-5: ~15 min intervals
6-11: ~30 min intervals
12+: ~60 min intervals
Uses server ID hash to distribute checks across cycles and prevent
thundering herd.
ServerCheckJob and ServerConnectionCheckJob increment unreachable_count
on failures. ServerManagerJob applies backoff logic before dispatching
checks. Includes comprehensive test coverage.
Ensure all file volume paths are validated and properly escaped before
use. Previously, only directory mount paths were validated at the input
layer — file mount paths now receive the same treatment across Livewire
components, API controllers, and the model layer.
- Validate and escape fs_path at the top of saveStorageOnServer() before
any commands are built
- Add path validation to submitFileStorage() in Storage Livewire component
- Add path validation to file mount creation in Applications, Services,
and Databases API controllers
- Add regression tests for path validation coverage
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Apply proper shell escaping to all user-controlled values interpolated into
backup shell commands (PostgreSQL username/password, MySQL/MariaDB root
password, MongoDB URI). Also URL-encode MongoDB credentials before embedding
in connection URI. Adds unit tests for escaping behavior.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Ensure pre_deployment_command and post_deployment_command have consistent
whitespace handling, matching the existing pattern used for health_check_command.
Adds regression tests for the normalization behavior.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add container name validation and shell argument escaping to
startUnmanaged, stopUnmanaged, restartUnmanaged, and restartContainer
methods, consistent with existing patterns used elsewhere in the
codebase.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Previously, the SHELL_SAFE_COMMAND_PATTERN was overly restrictive and blocked
legitimate characters needed for common Docker operations:
- Allow & for command chaining with && in multi-step build commands
- Allow " for build arguments with spaces (e.g., --build-arg KEY="value")
Update validation messages to reflect the new allowed operators and refactor
code to use imports instead of full class paths for better readability.
- Add #[Locked] to server-set properties on Import component (resourceId,
resourceType, serverId, resourceUuid, resourceDbType, container) to
prevent client-side modification via Livewire wire protocol
- Add container name validation in runImport() and restoreFromS3()
using shared ValidationPatterns::isValidContainerName()
- Scope server lookup to current team via ownedByCurrentTeam()
- Consolidate duplicate container name regex from Import,
ExecuteContainerCommand, and Terminal into shared
ValidationPatterns::isValidContainerName() static helper
- Add tests for container name validation, locked attributes, and
team-scoped server lookup
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Apply the existing `can.access.terminal` middleware to `POST /terminal/auth`
and `POST /terminal/auth/ips` routes, consistent with the `GET /terminal` route.
Adds regression tests covering unauthenticated, member, admin, and owner roles.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add validateDatabasesBackupInput() helper that properly parses all
database backup formats including MongoDB's "db:col1,col2|db2:col3"
and validates each component individually.
- Validate and escape collection names in DatabaseBackupJob
- Replace comma-only split in BackupEdit with format-aware validation
- Add input validation in API create_backup and update_backup endpoints
- Add unit tests for collection name and multi-format validation
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add team-scoped server validation to domains_by_server API endpoint
- Filter applications and services to only those on the requested server
- Scope ActivityMonitor activity lookups to the current team
- Fix query param disambiguation (query vs route param) in domains endpoint
- Fix undefined $ip variable in services domain collection
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Extract common container selection logic into resolveCommandContainer() method
that handles both single and multi-container app scenarios. This consolidates
duplicated code from run_pre_deployment_command() and run_post_deployment_command()
while improving error messaging and test coverage.
When preview environment variables are configured, fall back to production
variables for keys not overridden by preview values. This ensures variables
like DB_PASSWORD that exist only in production are available in the preview
.env file, enabling proper ${VAR} interpolation in docker-compose YAML.
Fallback only applies when preview variables are configured, preventing
unintended leakage of production values when previews aren't in use.
Also improves UI by hiding the Domains section when only database services
are present, and simplifies the logs view by removing status checks.
Do not replace self-referencing environment variables (e.g., DATABASE_URL: ${DATABASE_URL})
with saved DB values in the compose environment section. Keeping the reference intact allows
Docker Compose to resolve from .env at deploy time, preventing stale values from overriding
user updates that haven't been re-parsed.
Fixes#9136
Add validation in GetProxyConfiguration to detect when stored proxy config
belongs to a different proxy type (e.g., Traefik config on a Caddy server)
and trigger regeneration with a warning log. Clear cached proxy configuration
and settings when proxy type is changed to prevent stale configs from being
reused. Includes tests verifying config rejection on type mismatch and
graceful fallback on invalid YAML.
- Replace manual subscription create/update logic with updateOrCreate() and firstOrCreate() to eliminate race conditions
- Add validation in PricingPlans to prevent subscribing if team already has active subscription
- Improve error handling for missing team_id in customer.subscription.updated event
- Add tests verifying subscriptions are updated rather than duplicated
- Normalize hostnames to lowercase for RFC 1123 compliance while accepting uppercase input
- Expand NAME_PATTERN to allow parentheses, hash, comma, colon, and plus characters
- Add fallback to random name generation when application name doesn't meet minimum requirements
- Add comprehensive test coverage for validation patterns and edge cases
Allow serverLimit() and serverLimitReached() to accept an optional team
parameter instead of relying solely on the current session. This improves
testability and makes the methods more flexible by allowing them to work
without session state.
Add comprehensive tests covering various scenarios including no session,
team at limit, and team under limit.
Add escapeShellValue() helper function to safely escape shell values by wrapping
them in single quotes and escaping embedded quotes. Use this function throughout
the nixpacks command building to prevent shell injection vulnerabilities when
passing user-provided build commands, start commands, and environment variables.
This fixes unsafe string concatenation that could allow command injection when
user input contains special shell characters like &&, |, ;, etc.
Add validation in manual and normal webhook handlers to reject GitHub
event types other than 'push' and 'pull_request'. Unsupported events
now return a graceful response instead of potentially causing
downstream errors. Includes tests for ping events, unsupported event
types, and unknown events.
Extract build_railpack_static_image() into its own method, prevent port
override when is_static is set, bump Railpack to 0.22.0, and improve
test setup with beforeEach and correct polymorphic env var fields.
- Add storage endpoints (list, create, update, delete) to DatabasesController
- Add storage endpoints (list, create, update, delete) to ServicesController
- Add UUID field and migration for local_persistent_volumes table
- Update LocalPersistentVolume model to extend BaseModel
- Support UUID-based storage identification in ApplicationsController
- Update OpenAPI documentation with new storage endpoints and schemas
- Fix application name generation to extract repo name from full git path
- Add comprehensive tests for storage API operations
Extract the shouldRunNow() method from ScheduledJobManager and ServerManagerJob into
a reusable shouldRunCronNow() helper function. This centralizes cron scheduling logic
and enables consistent deduplication behavior across all scheduled job types.
- Create shouldRunCronNow() helper in bootstrap/helpers/shared.php with timezone
and dedup support
- Refactor ScheduledJobManager and ServerManagerJob to use the shared helper
- Add ScheduledJobDiagnostics command for inspecting cache state and scheduling
decisions across all scheduled jobs
- Simplify shouldRunNow tests to directly test the helper function
- Add DockerCleanupJob test for error handling and execution tracking
- Increase scheduled log retention from 1 to 7 days
Implements getPreviousRunDate() + cache-based tracking in shouldRunNow()
to prevent duplicate dispatch of scheduled jobs when queue delays push
execution past the cron minute. This resilience ensures jobs catch missed
windows without double-dispatching within the same cron window.
Updated scheduled job dispatches to include dedupKey parameter:
- Docker cleanup operations
- Server connection checks
- Sentinel restart checks
- Server storage checks
- Server patch checks
DockerCleanupJob now dispatches on the 'high' queue for faster processing.
Includes comprehensive test coverage for dedup behavior across different
cron schedules and delay scenarios.
Add support for optional comment field on environment variables created or
updated through the bulk API endpoints. Comments are validated to a maximum
of 256 characters and are nullable. Updates preserve existing comments when
not provided in the request.
Extract resource UUIDs from route parameters instead of request body
in ApplicationsController and ServicesController environment variable
endpoints. This prevents UUID parameters from being spoofed in the
request body.
- Replace $request->uuid with $request->route('uuid')
- Replace $request->env_uuid with $request->route('env_uuid')
- Add tests verifying route parameters are used and body UUIDs ignored
Add new Resources tab to storage show page displaying backup schedules using
that storage. Refactor storage show layout with navigation tabs for General
and Resources sections. Move delete action from form to show component.
Implement cascade deletion in S3Storage model to automatically disable S3
backups when storage is deleted. Improve error handling in DatabaseBackupJob
to throw exception when S3 storage is missing instead of silently returning.
- New Storage/Resources Livewire component
- Add resources.blade.php view
- Add storage.resources route
- Move delete() method from Form to Show component
- Add deleting event listener to S3Storage model
- Track backup count and current route in Show component
- Add #[On('submitStorage')] attribute to form submission
- Record refunds immediately before cancellation to prevent retry issues if cancel fails
- Wrap Stripe API calls in try-catch for refunds and quantity reverts with internal notifications
- Add null check in Team.subscriptionEnded() to prevent NPE when subscription doesn't exist
- Fix control flow bug in StripeProcessJob (add missing break statement)
- Cap dynamic server limit with MAX_SERVER_LIMIT in subscription updates
- Add comprehensive tests for refund failures, event handling, and null safety
Add current_period_end to refund eligibility checks and display next billing
date and billing interval in the subscription overview. Refactor the plan
overview layout to show subscription status more prominently.
- Log warnings instead of silently failing when chmod 0600 fails
- Remove redundant refresh() call before SSH key validation
- Remove storeInFileSystem() call from updatePrivateKey() transaction
- Remove @unlink() of lock file after filesystem store
- Refactor unit tests to use real temp disk and anonymous class stub
instead of reflection-only checks
- Restrict "Add suffix for PR deployments" checkbox to non-service
resources in both shared and service file-storage views
- Replace condition `is_preview_deployments_enabled` with `!$isService`
for PR suffix visibility in storages/show.blade.php
- Fix FileStorage::instantSave() to use authorize + syncData instead
of delegating to submit(), preventing unintended side effects
- Add $this->validate() to Storages/Show::instantSave() before saving
- Add response content schemas to storages API OpenAPI annotations
- Add additionalProperties: false to storage update request schema
- Rewrite PreviewDeploymentBindMountTest with behavioral tests of
addPreviewDeploymentSuffix instead of file-content inspection
Add `is_preview_suffix_enabled` flag to `local_file_volumes` and
`local_persistent_volumes` tables, allowing per-volume control over
whether a `-pr-N` suffix is appended during preview deployments.
Defaults to `true` to preserve existing behavior. Users can disable
it for volumes containing shared config or repository scripts that
should not be isolated per PR.
Bind mount volumes reference files at the repository's original path and
should not receive the -pr-N suffix. Only named Docker volumes require
the suffix for isolation between preview deployments.
Adds PreviewDeploymentBindMountTest to verify the correct behavior.
Fixes#7802
The root cause of sporadic "Permission denied (publickey)" errors was
that validateSshKey() only checked if the key file existed on disk,
never verifying its content matched the database. When keys were rotated
or updated, the stale file persisted and SSH used the wrong key.
Changes:
- validateSshKey() now refreshes key from DB and compares file content
- Server saved event detects private_key_id changes to invalidate mux
- PrivateKey storeInFileSystem() uses file locking to prevent races
- PrivateKey saved event auto-resyncs file on key content changes
- Enforces 0600 permissions on key files
Fixescoollabsio/coolify#7724
Server metadata is now automatically gathered when server validation completes successfully, both in the async job and Livewire component. This ensures server details (OS, CPU count, etc.) are populated immediately after validation passes, improving the user experience without requiring manual metadata fetching.
Tests added to verify gatherServerMetadata is called on successful validation and skipped when validation fails.
The generate_preview_fqdn_compose method now extracts and populates the fqdn field from docker_compose_domains, making it available for webhook notifications. This handles multiple domains across services and gracefully sets fqdn to null when no domains are configured.
When adding --project-directory to custom docker compose start commands,
use the application's host workdir if preserveRepository is true, otherwise
use the container workdir. Add tests for both scenarios and explicit paths.
Add `is_container_label_escape_enabled` boolean field to services API,
allowing users to control whether special characters in container labels
are escaped. Defaults to true (escaping enabled).
When disabled, users can use environment variables within labels.
Includes validation rules and comprehensive test coverage.
- Add -i flag to explicitly specify ssh key path in git ls-remote operations
- Remove static $rules properties in favor of dynamic rules() method
- Fix test syntax error
- Prioritize real private keys (id > 0) first
- Check source second before falling back to zero key
- Remove isDev() check that was restricting zero key behavior in dev
- Remove exception throw, use 'other' as safe fallback
- Expand test coverage to validate all precedence scenarios
Allow passing a custom GIT_SSH_COMMAND to setGitImportSettings() so that git fetch,
submodule update, and lfs pull use the same SSH authentication as the initial clone.
This is required for git sources like GitLab that use custom ports and identity files.
Also remove unnecessary SSH retry event tracking and add test coverage.
Ensure proper type comparison when verifying deployment team ownership.
Adds comprehensive feature tests for the GET /api/v1/deployments/{uuid} endpoint.
Extract shared variable resolution logic into a reusable helper function
`resolveSharedEnvironmentVariables()` and apply it in applicationParser and
serviceParser to ensure patterns like {{environment.VAR}}, {{project.VAR}},
and {{team.VAR}} are properly resolved in the compose environment section.
Without this, unresolved {{...}} strings would take precedence over resolved
values from the .env file (env_file:) in docker-compose configurations.
Add comprehensive test suite verifying that magic (referenced) SERVICE_URL_ and
SERVICE_FQDN_ variables don't overwrite values set by direct template declarations
or updateCompose(). Tests cover the fix for GitHub issue #8912 where generic
SERVICE_URL and SERVICE_FQDN variables remained stale after changing a service
domain in the UI. These tests ensure the transition from updateOrCreate() to
firstOrCreate() in the magic variables section works correctly.
Add ability to gather and display server system information including OS, architecture, kernel version, CPU count, memory, and uptime. Includes:
- New gatherServerMetadata() method to collect system details via remote commands
- New refreshServerMetadata() Livewire action with authorization and error handling
- Server Details UI section showing collected metadata with refresh capability
- Database migration to add server_metadata JSON column
- Comprehensive test suite for metadata collection and persistence
Add full GitLab application source support for git operations:
- Implement SSH-based authentication using private keys with configurable ports
- Support HTTP basic auth for HTTPS GitLab URLs (with or without deploy keys)
- Handle private key setup and SSH command configuration in both Docker and local modes
- Support merge request checkouts for GitLab with SSH authentication
Improvements to credential handling:
- URL-encode GitHub access tokens to handle special characters properly
- Update log sanitization to redact passwords from HTTPS/HTTP URLs
- Extend convertGitUrl() type hints to support GitlabApp sources
Add test coverage and seed data:
- New GitlabSourceCommandsTest with tests for private key and public repo scenarios
- Test for HTTPS basic auth password sanitization in logs
- Seed data for GitLab deploy key and public example applications
- Add `$selectedActions = []` parameter to delete/remove methods in multiple
Livewire components to support optional deletion actions
- Return error message string when password verification fails instead of
silent return
- Return `true` on successful deletion to indicate completion
- Handle selectedActions to set component properties for cascading deletions
(delete_volumes, delete_networks, delete_configurations, docker_cleanup)
- Add test coverage for Danger component delete functionality with password
validation and selected actions handling
- Store proxy configuration in database as primary source for faster access
- Implement automatic timestamped backups when configuration changes
- Add backfill migration logic to recover configs from disk for legacy servers
- Simplify UI by removing loading states (config now readily available)
- Add comprehensive logging for debugging configuration generation and recovery
- Include unit tests for config recovery scenarios
When application->fqdn is null, COOLIFY_FQDN and COOLIFY_URL are set to null.
These null values cause nixpacks to fail parsing the config with
"invalid type: null, expected a string".
Filter out null and empty string values when generating environment variables
for the nixpacks plan JSON. Fixes#6830.
- Add gitSshCommand parameter to setGitImportSettings and buildGitCheckoutCommand
- Extract hardcoded ssh commands to variables for consistency
- Apply custom ssh credentials to all git operations including submodule sync/update
- Configure git url rewriting for github token-based authentication with submodules
- Add comprehensive test suite for submodule credential propagation
Replace direct shell interpolation of environment values with base64 encoding
to prevent command injection attacks. Environment configuration is now built as
a single string, base64-encoded, then decoded to file atomically.
Also add regex validation to restrict environment field values to safe
characters (alphanumeric, underscore, hyphen, dot) at the application layer.
Fixes GHSA-3xm2-hqg8-4m2p
Add validateGitRef() helper function that uses an allowlist approach to prevent
OS command injection through git commit SHAs, branch names, and tags. Only allows
alphanumeric characters, dots, hyphens, underscores, and slashes.
Changes include:
- Add validateGitRef() helper in bootstrap/helpers/shared.php
- Apply validation in Rollback component when accepting rollback commit
- Add regex validation to git commit SHA fields in Livewire components
- Apply regex validation to API rules for git_commit_sha
- Use escapeshellarg() in git log and git checkout commands
- Add comprehensive unit tests covering injection payloads
Addresses GHSA-mw5w-2vvh-mgf4
Validation operations should require write permissions as they trigger
state-changing actions. Updated middleware for:
- POST /api/v1/cloud-tokens/{uuid}/validate
- GET /api/v1/servers/{uuid}/validate
Added tests to verify read-only tokens cannot access these endpoints.
Previously, the view checked $privateKeyId with ! operator, which
incorrectly treats 0 (localhost key) as falsy. Changed to explicit
is_null() checks to distinguish between null (no key) and 0 (localhost).
Added test coverage for both cases.
Handle non-positive disable counts with `<= 0` so teams at or under the
server limit correctly re-enable force-disabled servers. Add a feature test
suite for ServerLimitCheckJob covering under-limit, at-limit, over-limit,
and no-op behavior.
Centralize min/max server limits in Stripe quantity updates and wire them into
Livewire subscription actions with price preview/update handling.
Also improve host/proxy middleware behavior by trusting loopback hosts when FQDN
is set and auto-enabling secure session cookies for HTTPS requests behind
proxies when session.secure is unset.
Includes feature tests for loopback trust and secure cookie auto-detection.
Introduce a new `UpdateSubscriptionQuantity` Stripe action to:
- preview prorated due-now and next-cycle recurring costs
- update subscription item quantity with proration invoicing
- revert quantity and void invoice when payment is not completed
Wire the flow into the Livewire subscription actions UI with a new adjust-limit modal,
price preview loading, and confirmation-based updates. Also refactor the subscription
management section layout and fix modal confirmation behavior for temporary 2FA bypass.
Add `Subscription::billingInterval()` helper and comprehensive Pest coverage for
quantity updates, preview calculations, failure/revert paths, and billing interval logic.
