<!-- Add the related story/sub-task/bug number, like Resolves#123, or
remove if NA -->
**Related issue:** Resolves#38013
# Checklist for submitter
- [x] Changes file added for user-visible changes in `changes/`,
`orbit/changes/` or `ee/fleetd-chrome/changes`.
See [Changes
files](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/guides/committing-changes.md#changes-files)
for more information.
- [x] Input data is properly validated, `SELECT *` is avoided, SQL
injection is prevented (using placeholders for values in statements), JS
inline code is prevented especially for url redirects, and untrusted
data interpolated into shell scripts/commands is validated against shell
metacharacters.
## Testing
- [x] Added/updated automated tests
- [x] QA'd all new/changed functionality manually
See
https://drive.google.com/file/d/1zeFNLuf_rT5FWzDiYyL2_hbIBW2neba-/view?usp=drive_link
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Bug Fixes**
* GitOps variables in JSON configuration profiles (Apple DDM
declarations and Android profiles) are now automatically escaped for
JSON special characters, ensuring proper handling of sensitive values.
* **Tests**
* Added JSON configuration profile escaping validation to the enterprise
GitOps integration test suite.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
Estimate ~5% load improvement.
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Bug Fixes**
* Fixed MDM command handling in the performance testing agent to
properly skip duplicate status responses.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
<!-- Add the related story/sub-task/bug number, like Resolves#123, or
remove if NA -->
**Related issue:** Resolves#43417
# Checklist for submitter
If some of the following don't apply, delete the relevant line.
- [X] Changes file added for user-visible changes in `changes/`,
`orbit/changes/` or `ee/fleetd-chrome/changes`.
done in https://github.com/fleetdm/fleet/pull/42216
## Testing
- [X] Added/updated automated tests
- [X] QA'd all new/changed functionality manually
- Using a local fileserver, added the same software to two fleets and
ran `fleetctl gitops`. Verified that the first fleet downloaded the
file, the second fleet used the cache, and both fleet showed the
software installer in the UI.
## Summary by CodeRabbit
* **Chores**
* Updated software installer lookup mechanism to support optional
team-scoped searches, enabling fallback to cross-team installer cache
when team-specific installers are unavailable.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
**Related issue:** Resolves#39900
## Testing
- [x] QA'd all new/changed functionality manually
For unreleased bug fixes in a release candidate, one of:
- [x] Confirmed that the fix is not expected to adversely impact load
test results
- [x] Alerted the release DRI if additional load testing is needed
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Bug Fixes**
* Improved timestamp handling for OS vulnerability data synchronization
to use UTC timezone when synchronization is enabled, ensuring consistent
timing behavior across different system configurations.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
<!-- Add the related story/sub-task/bug number, like Resolves#123, or
remove if NA -->
**Related issue:** Resolves#43050
## Testing
- [x] QA'd all new/changed functionality manually
Confirmed the traffic sequence with ngrok and DDM stats are as expected
in the osquery-perf logs:
### Adding a new DDM
* `DeclarativeManagement` command and Ack
* `tokens` request
* `declaration-items` request
* `activation` for the DDM
* `configuration` for the DDM
* `tokens` request confirms changes settled
* `status` request
### Remove/re-add DDM (no global change)
* `DeclarativeManagement` command and Ack
* `tokens` request
### Adding a second DDM
* `DeclarativeManagement` command and Ack
* `tokens` request
* `declaration-items` request
* `activation` for the new DDM only
* `configuration` for the new DDM only
* `tokens` request confirms changes settled
* `status` request
### Removing a DDM
* `DeclarativeManagement` command and Ack
* `tokens` request
* `declaration-items` request
* `tokens` request
* `status` request
### Remove all DDMs
* `DeclarativeManagement` command and Ack
* `tokens` request
* `declaration-items` request
* `tokens` request
* `status` request
---
State correctly updates on the host's profiles:
<img width="1246" height="512" alt="image"
src="https://github.com/user-attachments/assets/0d289d4e-1e9b-4283-aef0-fd1ab3ecb355"
/>
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Improved macOS Declarative Management sync: faster convergence,
fetches only changed declarations, detects removals, and sends
consolidated status updates.
* **Monitoring**
* Added metrics to track declaration token fetch success and error
rates.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
Fixes#42885
Added new middleware (APIOnlyEndpointCheck) that enforces 403 for
API-only users whose request either isn't in the API endpoint catalog or
falls outside their configured per-user endpoint restrictions.
## Summary
The nightly OSV artifact generation in `fleetdm/vulnerabilities` failed
over the weekend with:
```
fatal: error processing shallow info: 4
```
at `cmd/osv-processor/sync-and-detect-changes.sh` during:
```bash
git fetch --shallow-since="3 days ago" origin main
```
Root cause: `git fetch --shallow-since` errors out when the upstream
(`canonical/ubuntu-security-notices`) has zero commits newer than the
cutoff. Canonical didn't push anything over the weekend, so the 3-day
window returned empty and upload-pack produced an unusable shallow
response.
Fix:
- Fall back to `git fetch --depth=3` if `--shallow-since` still returns
empty, so the initial clone always succeeds.
Subsequent runs reuse the existing clone and take the other branch of
the script (plain `git fetch origin main`), which doesn't have this
failure mode.
Failing run:
https://github.com/fleetdm/vulnerabilities/actions/runs/24330589309/job/71035337352
## Test plan
- [x] Re-run the Ubuntu OSV artifact generation workflow; initial clone
succeeds regardless of upstream push frequency.
- [x] Manually exercise the cold-cache path locally: `rm -rf
ubuntu-security-notices &&
./cmd/osv-processor/sync-and-detect-changes.sh` — completes without
error.
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Bug Fixes**
* Improved initial repository sync: if the primary shallow fetch returns
no commits, the process now falls back to a limited-depth fetch, warns
the user, and shows recent commit history before continuing. Downstream
change detection and existing behavior for already-cloned repos remain
unchanged.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Co-authored-by: Konstantin Sykulev <konst@sykulev.com>
<!-- Add the related story/sub-task/bug number, like Resolves#123, or
remove if NA -->
**Related issue:** Resolves#43047
# Checklist for submitter
- [x] Changes file added for user-visible changes in `changes/`,
`orbit/changes/` or `ee/fleetd-chrome/changes`.
See [Changes
files](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/guides/committing-changes.md#changes-files)
for more information.
- [x] Input data is properly validated, `SELECT *` is avoided, SQL
injection is prevented (using placeholders for values in statements), JS
inline code is prevented especially for url redirects, and untrusted
data interpolated into shell scripts/commands is validated against shell
metacharacters.
## Testing
- [x] Added/updated automated tests
- [x] Where appropriate, [automated tests simulate multiple hosts and
test for host
isolation](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/reference/patterns-backend.md#unit-testing)
(updates to one hosts's records do not affect another)
- [x] QA'd all new/changed functionality manually
See
https://github.com/fleetdm/fleet/issues/42960#issuecomment-4244206563
and subsequent comments.
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Apple DDM declarations support a vetted subset of Fleet variables with
per-host substitution; premium license required. Declaration tokens and
resend behavior now reflect variable changes; unresolved host
substitutions mark that host’s declaration as failed.
* **Bug Fixes**
* Clearer errors for unsupported or license-restricted Fleet variables
and more consistent DDM resend/update semantics when variables change.
* **Tests**
* Added extensive unit and integration tests covering Fleet variable
validation, substitution, token changes, resends, and failure states.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
Resolves broken OTEL on main, which was introduced with dependabot
update #43298
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Chores**
* Updated OpenTelemetry semantic conventions dependency to the latest
version.
* **Tests**
* Added test coverage for OpenTelemetry resource creation validation.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
**Related issue:**
Ref #34797
Ref #42675
## Problem
When a software installer spec has no `hash_sha256`, Fleet re-downloads
the package, re-extracts metadata, and re-upserts the DB on every GitOps
run, even if the upstream file hasn't changed. For deployments with 50+
URL-only packages across multiple teams, this wastes bandwidth and
processing time on every run.
## Solution
By default, use etags to avoid unnecessary downloads:
1. First run: Fleet downloads the package normally and stores the
server's ETag header
2. Subsequent runs: Fleet sends a conditional GET with `If-None-Match`.
If the server returns 304 Not Modified, Fleet skips the download,
metadata extraction, S3 upload, and DB upsert entirely
Opt-out with `always_download:true`, meaning packages continue to be
downloaded and re-processed on every run, same as today. No UI changes
needed.
```yaml
url: https://nvidia.gpcloudservice.com/global-protect/getmsi.esp?version=64&platform=windows
always_download: true
install_script:
path: install.ps1
```
### Why conditional GET instead of HEAD
Fleet team [analysis of 276 maintained
apps](https://github.com/fleetdm/fleet/pull/42216#issuecomment-4105430061)
showed 7 apps where HEAD requests fail (405, 403, timeout) but GET works
for all. Conditional GET eliminates that failure class: if the server
doesn't support conditional requests, it returns 200 with the full body,
same as today.
### Why opt-in
5 of 276 apps (1.8%) have stale ETags (content changes but ETag stays
the same), caused by CDN caching artifacts (CloudFront, Cloudflare,
nginx inode-based ETags). The `cache` key lets users opt in per package
for URLs where they've verified ETag behavior is correct.
Validation rejects `always_download: true` when hash_sha256` is set
## Changes
- New YAML field: `cache` (bool, package-level)
- New migration: `http_etag` VARCHAR(512) column (explicit
`utf8mb4_unicode_ci` collation) + composite index `(global_or_team_id,
url(255))` on `software_installers`
- New datastore method: `GetInstallerByTeamAndURL`
- `downloadURLFn` accepts optional `If-None-Match` header, returns 304
as `(resp, nil, nil)` with `http.NoBody`
- ETag validated per RFC 7232 (ASCII printable only, no control chars,
max 512 bytes) at both write and read time
- Cache skipped for `.ipa` packages (multi-platform extraInstallers)
- TempFileReader and HTTP response leak prevention on download retry
- Docs updated in `yaml-files.md`
## What doesn't change
- Packages with `hash_sha256`: existing hash-based skip, untouched
- FMA packages: FMA version cache, untouched
- Packages with `always_download: true`: identical to current behavior
- Fleet UI: no changes
## Test plan
Automated testing:
- [x] 16 unit tests for `validETag`
- [x] 8 unit tests for conditional GET behavior (304, 200, 403, 500,
weak ETag, S3 multipart, no ETag)
- [x] MySQL integration test for `GetInstallerByTeamAndURL`
- [x] All 23 existing `TestSoftwareInstallers` datastore tests pass
- [x] All existing service tests pass
Manual testing:
- [x] E2E: 86 packages across 6 CDN patterns, second apply shows 51
conditional hits (304)
- [x] @sgress454 used a local fileserver tool to test w/ a new instance
and dummy packages
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* ETag-based conditional downloads to skip unchanged remote installer
files.
* New always_download flag to force full re-downloads.
* **Tests**
* Added integration and unit tests covering conditional GETs, ETag
validation, retries, edge cases, and payload behavior.
* **Chores**
* Persist HTTP ETag and related metadata; DB migration and index to
speed installer lookups.
* Added installer lookup by team+URL to support conditional download
flow.
* **Bug Fix**
* Rejects using always_download together with an explicit SHA256 in
uploads.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Co-authored-by: Scott Gress <scott@fleetdm.com>
Co-authored-by: Scott Gress <scott@pigandcow.com>
Co-authored-by: Ian Littman <iansltx@gmail.com>
Fixes#34288.
# Checklist for submitter
If some of the following don't apply, delete the relevant line.
- [x] Changes file added for user-visible changes in `changes/`,
`orbit/changes/` or `ee/fleetd-chrome/changes`.
See [Changes
files](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/guides/committing-changes.md#changes-files)
for more information.
## Testing
- [x] Added/updated automated tests
- [ ] QA'd all new/changed functionality manually
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Setup experience cancellations now create explicit cancellation
activities for skipped/failed software and VPP app installs, plus a new
"Canceled setup experience" activity type and a from_setup_experience
flag. Activity text and host activity views now indicate "during setup
experience" when applicable.
* **Tests**
* Added and updated tests for cancellation activity creation, VPP
license-failure handling, and WasFromAutomation/from_setup_experience
behaviors.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
**Related issue:** Resolves#42883
Added a new premium GET /api/_version_/fleet/rest_api endpoint that
returns the contents of the embedded `api_endpoints.yml` artifact.
**Related issue:** Resolves#42754
# Checklist for submitter
If some of the following don't apply, delete the relevant line.
- [x] Changes file added for user-visible changes in `changes/`,
`orbit/changes/` or `ee/fleetd-chrome/changes`.
See [Changes
files](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/guides/committing-changes.md#changes-files)
for more information.
- [x] Input data is properly validated, `SELECT *` is avoided, SQL
injection is prevented (using placeholders for values in statements), JS
inline code is prevented especially for url redirects, and untrusted
data interpolated into shell scripts/commands is validated against shell
metacharacters.
## Testing
- [x] Added/updated automated tests
- [x] QA'd all new/changed functionality manually
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Bug Fixes**
* Improved app manifest retrieval with automatic fallback to hosted
copies when the primary source is unavailable, reducing sync failures.
* **Documentation**
* Clarified that Fleet will fall back to hosted manifest copies if the
new manifest site is inaccessible.
* **New Features**
* Streamlined maintained-app synchronization to use a simpler sync
entrypoint and unified primary/fallback fetch logic.
* **Tests**
* Added comprehensive tests for primary/fallback fetch flows, error
handling, large-response truncation, and environment-based overrides.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
<!-- Add the related story/sub-task/bug number, like Resolves#123, or
remove if NA -->
**Related issue:** For #40015
* Moves repeated empty mocks into a new `setupEmptyGitOpsMocks` method
* Adds new "deprecation" tests:
* In TestGitOpsFullGlobal, TestGitOpsFullTeam and
TestGitOpsFullGlobalAndTeam tests "kitchen sink" with both new and
deprecated keys
* Added keys and checks to verify `setup_experience`,
`apple_business_manager` and `volume_purchasing_program` configs
* Consolidated map of deprecated -> new GitOps keys in one place
<!-- Add the related story/sub-task/bug number, like Resolves#123, or
remove if NA -->
**Related issue:** Resolves#42368
# Checklist for submitter
If some of the following don't apply, delete the relevant line.
- [x] Changes file added for user-visible changes in `changes/`,
`orbit/changes/` or `ee/fleetd-chrome/changes`.
See [Changes
files](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/guides/committing-changes.md#changes-files)
for more information. For the overall story
- [x] Input data is properly validated, `SELECT *` is avoided, SQL
injection is prevented (using placeholders for values in statements), JS
inline code is prevented especially for url redirects, and untrusted
data interpolated into shell scripts/commands is validated against shell
metacharacters.
- [x] Timeouts are implemented and retries are limited to avoid infinite
loops
- [x] If paths of existing endpoints are modified without backwards
compatibility, checked the frontend/CLI for any necessary changes
## Testing
- [x] Added/updated automated tests
- [x] QA'd all new/changed functionality manually
**Related issue:** Resolves#42881
- Added user_api_endpoints table to track per user API endpoint
permissions.
- Added service/api_endpoints, used to handle service/api_endpoints.yml
artifact.
- Added check on server start that makes sure that
service/apin_endpoints.yml is a subset of router routes.
<!-- Add the related story/sub-task/bug number, like Resolves#123, or
remove if NA -->
**Related issue:** Resolves#42853
This PR simply adds the `require_all_software_windows` config option. It
doesn't use it. The logic to use it will be hooked up in subsequent PRs.
The fleetctl TestIntegrationsPreview test is expected to fail since it
builds the server against main and doesn't know about our new config
option.
# Checklist for submitter
If some of the following don't apply, delete the relevant line.
- [x] Changes file added for user-visible changes in `changes/`,
`orbit/changes/` or `ee/fleetd-chrome/changes`.
## Testing
- [x] Added/updated automated tests
- [x] QA'd all new/changed functionality manually
## New Fleet configuration settings
- [x] Verified that the setting is exported via `fleetctl
generate-gitops`
- Not exported. generate-gitops does not export
require_all_software_windows (or require_all_software_macos either). The
generateControls function (generate_gitops.go) outputs a "TODO: update
with your setup_experience configuration" placeholder when any setup
experience config exists, rather than exporting individual field values.
This is a pre-existing limitation that applies equally to both fields -
not something introduced by our PR.
- [x] Verified the setting is documented in a separate PR to [the GitOps
documentation](https://github.com/fleetdm/fleet/blob/main/docs/Configuration/yaml-files.md#L485)
- Yes. PR #42046 adds require_all_software_windows to both docs/REST
API/rest-api.md and docs/Configuration/yaml-files.md.
- [x] Verified that the setting is cleared on the server if it is not
supplied in a YAML file (or that it is documented as being optional)
- Yes, it gets cleared to false - both when setup_experience: is present
without the field, and when setup_experience: is omitted entirely. This
is the same behavior as the existing require_all_software_macos field
- [x] Verified that any relevant UI is disabled when GitOps mode is
enabled
- Covered by #42854 (frontend subtask). The existing macOS checkbox in
InstallSoftwareForm.tsx:271 already checks gitOpsModeEnabled to disable
itself. The Windows checkbox to be added in #42854 will follow the same
pattern.
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Added a Windows setup experience software requirement setting. When
enabled, Windows devices will cancel the Autopilot setup if any required
software installation fails.
* **Tests**
* Added test coverage for the new Windows software requirement
configuration.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
<!-- Add the related story/sub-task/bug number, like Resolves#123, or
remove if NA -->
**Related issue:** Resolves#40015
# Details
Activates deprecation warnings for old API params and CLI args, updates
tests that would generate warnings (except for tests explicitly designed
to generate warnings).
The expectation from here on is that Fleet UI usage should not generate
any deprecation warnings in the server logs, nor should the output from
`generate-gitops` generate any warnings when fed into `gitops`.
# Checklist for submitter
If some of the following don't apply, delete the relevant line.
- [X] Changes file added for user-visible changes in `changes/`,
`orbit/changes/` or `ee/fleetd-chrome/changes`.
See [Changes
files](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/guides/committing-changes.md#changes-files)
for more information.
## Testing
- [X] Added/updated automated tests
- [ ] QA'd all new/changed functionality manually
- [X] clicked around in an mdm-enabled instance, turned setup experience
features on and off, saw no server warnings
- [X] did `fleetctl generate-gitops` on mdm-enabled instance, saw no
server or cli warnings
- [X] did `fleetctl gitops` on mdm-enabled instance, saw no server or
cli warnings
This would have helped some troubleshooting on customer workflows
failing due to long response times.
(We had a long running `spec/fleets` API request for customer-numa.)
Sample of logging after I added a `300s` sleep to
`/api/latest/fleet/config`:
```
[+] would've applied EULA
[+] would've applied certificate authorities
Error: applying fleet config: PATCH /api/latest/fleet/config: do request: Patch "https://localhost:8080/api/latest/fleet/config?dry_run=true&overwrite=true": stream error: stream ID 49; INTERNAL_ERROR; received from peer (API time: 1m40.002s)
```
Another sample error after bringing Fleet down during a GitOps run:
```
[+] would've applied 4 software packages for fleet Conditional access FTW
Error: applying software installers for fleet "Conditional access FTW": GET /api/latest/fleet/software/batch/395942cc-69c9-49f9-93d3-f1120e0b9e34: do request: Get "https://localhost:8080/api/latest/fleet/software/batch/395942cc-69c9-49f9-93d3-f1120e0b9e34?dry_run=true&fleet_name=Conditional+access+test+team&overwrite=true": dial tcp [::1]:8080: connect: connection refused (API time: 2ms)
```
<!-- Add the related story/sub-task/bug number, like Resolves#123, or
remove if NA -->
**Related issue:** Resolves#40791
# Checklist for submitter
If some of the following don't apply, delete the relevant line.
- [ ] Changes file added for user-visible changes in `changes/`,
`orbit/changes/` or `ee/fleetd-chrome/changes`.
See [Changes
files](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/guides/committing-changes.md#changes-files)
for more information.
## Testing
- [ ] Added/updated automated tests
We don't typically do new primo mode tests. As long as existing tests
for startup library pass, we're good.
- [X] QA'd all new/changed functionality manually
- [X] tested that starting a new instance with
`FLEET_PARTNERSHIPS_ENABLE_PRIMO=true` causes the starter library not to
be applied (no fleets, labels or policies
**Related issue:** Resolves#40057
# Checklist for submitter
If some of the following don't apply, delete the relevant line.
- [x] Changes file added for user-visible changes in `changes/`,
`orbit/changes/` or `ee/fleetd-chrome/changes`.
See [Changes
files](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/guides/committing-changes.md#changes-files)
for more information.
- [x] Input data is properly validated, `SELECT *` is avoided, SQL
injection is prevented (using placeholders for values in statements), JS
inline code is prevented especially for url redirects, and untrusted
data interpolated into shell scripts/commands is validated against shell
metacharacters.
## Testing
- [x] Added/updated automated tests
- [x] QA'd all new/changed functionality manually
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* OSV (Open Source Vulnerabilities) added as an optional Ubuntu
vulnerability data source and enabled by default.
* **Features**
* Integrated OSV into the vulnerability scanning pipeline, artifact
sync/refresh, detection, and cleanup flows.
* Improved Ubuntu package/kernel version matching for more accurate OSV
detections.
* **Chores**
* Added configuration flag and updated expected config fixtures.
* **Tests**
* Added extensive tests for OSV sync, artifact handling, analyzer logic,
and cleanup behaviors.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
<!-- Add the related story/sub-task/bug number, like Resolves#123, or
remove if NA -->
**Related issue:** Resolves#41409
# Details
This PR updates the `ApplyStarterLibrary` method and functionality to
rely on the same templates and mechanisms as `fleetctl new`. The end
result is that running `fleetctl new` and `fleetctl gitops` on a new
instance should be a no-op; no changes should be made. Similarly,
changing the templates in a Fleet release will automatically affect
`fleetctl new` and `ApplyStarterLibrary` in the same exact way for that
release.
> Note that this moves the template files out of `fleetctl` and into
their own shared package. This move comprises the majority of the file
changes in the PR.
# Checklist for submitter
If some of the following don't apply, delete the relevant line.
- [X] Changes file added for user-visible changes in `changes/`,
`orbit/changes/` or `ee/fleetd-chrome/changes`.
See [Changes
files](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/guides/committing-changes.md#changes-files)
for more information.
## Testing
- [X] Added/updated automated tests
Note that
<img width="668" height="44" alt="image"
src="https://github.com/user-attachments/assets/066cd566-f91d-4661-84fc-2aabbfce2ef9"
/>
will fail until the 4.83 Fleet docker image is published, since it's
trying to push 4.83 config (including `exceptions`) to a 4.82 server.
- [X] QA'd all new/changed functionality manually
- [X] Created a new instance and validated that the fleets, policies and
labels created matched the ones created by `fleetctl new`
- [X] Ran `fleetctl new` and verified that it created the expected
folders and files
- [X] Ran `fleetctl gitops` with the files created by `fleetctl new` and
verified that the instance was unchanged.
- [X] Ran `fleetctl preview` successfully using a dev build of the Fleet
server image (since it won't work against the latest published build,
which doesn't support `exceptions`). Verified it shows the expected
teams, policies and labels
<!-- Add the related story/sub-task/bug number, like Resolves#123, or
remove if NA -->
**Related issue:** The entire ACME feature branch merge
# Checklist for submitter
If some of the following don't apply, delete the relevant line.
- [x] Changes file added for user-visible changes in `changes/`,
`orbit/changes/` or `ee/fleetd-chrome/changes`.
See [Changes
files](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/guides/committing-changes.md#changes-files)
for more information.
- [x] Input data is properly validated, `SELECT *` is avoided, SQL
injection is prevented (using placeholders for values in statements), JS
inline code is prevented especially for url redirects, and untrusted
data interpolated into shell scripts/commands is validated against shell
metacharacters.
- [x] Timeouts are implemented and retries are limited to avoid infinite
loops
## Testing
- [x] Added/updated automated tests
- [x] Where appropriate, [automated tests simulate multiple hosts and
test for host
isolation](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/reference/patterns-backend.md#unit-testing)
(updates to one hosts's records do not affect another)
- [x] QA'd all new/changed functionality manually
---------
Co-authored-by: Jordan Montgomery <elijah.jordan.montgomery@gmail.com>
Co-authored-by: Martin Angers <martin.n.angers@gmail.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: Gabriel Hernandez <ghernandez345@gmail.com>
Co-authored-by: Sarah Gillespie <73313222+gillespi314@users.noreply.github.com>
Fixes#40785
## Summary
When a GitOps run includes a `volume_purchasing_program` config that
references a team that doesn't exist yet, the code temporarily removes
the entire VPP config from the global AppConfig, clearing ALL VPP
token-to-team assignments on the server. However, the code only deferred
`app_store_apps` for the missing teams, not for existing teams that also
lost their VPP assignments. Those existing teams then failed with "No
available VPP Token" when their `app_store_apps` were applied.
The fix widens the deferral scope to match the clearing scope. When VPP
assignments are temporarily cleared, `app_store_apps` are now deferred
for all teams in the VPP config, not just the missing ones.
<!-- Add the related story/sub-task/bug number, like Resolves#123, or
remove if NA -->
**Related issue:** Resolves#34950
I changed from the original spec of 100 old commands to 3 due to load
test results. Admittedly my load test meant a very large number of hosts
all checked in and triggered deletion at once but at 100 per host and
per command the load was too high. 3 still results in cleanup over time
and doesn't seem to cause load issues.
# Checklist for submitter
If some of the following don't apply, delete the relevant line.
- [x] Changes file added for user-visible changes in `changes/`,
`orbit/changes/` or `ee/fleetd-chrome/changes`.
See [Changes
files](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/guides/committing-changes.md#changes-files)
for more information.
- [x] Input data is properly validated, `SELECT *` is avoided, SQL
injection is prevented (using placeholders for values in statements), JS
inline code is prevented especially for url redirects, and untrusted
data interpolated into shell scripts/commands is validated against shell
metacharacters.
- [x] If paths of existing endpoints are modified without backwards
compatibility, checked the frontend/CLI for any necessary changes
## Testing
- [x] Added/updated automated tests
- [x] Where appropriate, [automated tests simulate multiple hosts and
test for host
isolation](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/reference/patterns-backend.md#unit-testing)
(updates to one hosts's records do not affect another)
- [x] QA'd all new/changed functionality manually
---------
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
- [X] Changes file added for user-visible changes in `changes/`,
`orbit/changes/` or `ee/fleetd-chrome/changes`.
See [Changes
files](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/guides/committing-changes.md#changes-files)
for more information.
- [X] Input data is properly validated, `SELECT *` is avoided, SQL
injection is prevented (using placeholders for values in statements), JS
inline code is prevented especially for url redirects, and untrusted
data interpolated into shell scripts/commands is validated against shell
metacharacters.
## Testing
- [X] QA'd all new/changed functionality manually
Server log when SCIM ingestion receives a deactivation event for the
last admin:
```log
ts=2026-04-01T15:23:01Z level=error msg="failed to delete fleet user on deactivation" component=SCIM err="cannot delete last global admin"
```
Server response when attempting to demote last admin:
```json
{
"message": "Validation Failed",
"errors": [
{
"name": "global_role",
"reason": "cannot demote the last global admin"
}
],
"uuid": "1d110f56-25ac-47b8-bc96-982354474a87"
}
```
Server response when attempting to delete last admin:
```json
{
"message": "Validation Failed",
"errors": [
{
"name": "id",
"reason": "cannot delete the last global admin"
}
],
"uuid": "1448c2da-30e2-4652-a9a8-a01fc4f9b9c1"
}
```
---
## Original AI Summary
- Fixes a TOCTOU race condition where two concurrent admin operations
could bypass the last-global-admin guard, leaving zero admins and
permanently locking out the Fleet instance (fleetdm/confidential#14827)
- Introduces two new atomic datastore methods
(`DeleteUserIfNotLastAdmin`, `SaveUserIfNotLastAdmin`) that wrap the
admin count check and the write in a single MySQL transaction with
`SELECT ... FOR UPDATE`
- Fixes all four vulnerable code paths: service `DeleteUser`, service
`ModifyUser` (two demotion paths), and SCIM user deletion
## Test plan
- [X] Manual verification: single admin cannot be deleted or demoted
🤖 Generated with [Claude Code](https://claude.com/claude-code)
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Lucas Manuel Rodriguez <lucas@fleetdm.com>
**Related issue:** Resolves#41571
Changing the way delta artifacts are generated.
`changed_files_today.txt` and `changed_files_yesterday.txt` will always
look back into git history for any commits added yesterday and today.
- [x] QA'd all new/changed functionality manually
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
## Release Notes
* **Bug Fixes**
* Improved repository synchronization to ensure local state accurately
reflects the remote main branch.
* Enhanced file change detection logic for more reliable identification
of modified files.
* Refined output file handling to maintain consistency across
synchronization cycles.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
<!-- Add the related story/sub-task/bug number, like Resolves#123, or
remove if NA -->
**Related issue:** Resolves#34433 Part 2
# Checklist for submitter
If some of the following don't apply, delete the relevant line.
- [ ] Changes file added for user-visible changes in `changes/`,
`orbit/changes/` or `ee/fleetd-chrome/changes`.
See [Changes
files](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/guides/committing-changes.md#changes-files)
for more information. Added by first PR
- [x] Input data is properly validated, `SELECT *` is avoided, SQL
injection is prevented (using placeholders for values in statements), JS
inline code is prevented especially for url redirects, and untrusted
data interpolated into shell scripts/commands is validated against shell
metacharacters.
- [x] If paths of existing endpoints are modified without backwards
compatibility, checked the frontend/CLI for any necessary changes
## Testing
- [x] Added/updated automated tests
- [x] QA'd all new/changed functionality manually
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Profiles now install during device enrollment setup
* **Bug Fixes**
* Enhanced Apple MDM profile synchronization to handle concurrent
processing scenarios
* Improved profile reconciliation to prevent conflicts when multiple
workers process the same device simultaneously
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Co-authored-by: Martin Angers <martin.n.angers@gmail.com>
Related issue: Resolves https://github.com/fleetdm/fleet/issues/41571
It appears that there is some sort of error with shallow cloning.
```
Run echo "=== Generating OSV Artifacts for Ubuntu ==="
=== Generating OSV Artifacts for Ubuntu ===
=== OSV Repository Sync ===
Repository exists, updating with rolling window...
fatal: error processing shallow info: 4
Error: Process completed with exit code 128.
```
Since we are only keeping a limited history of the repository via cache
before re-clone, fall back to doing a regular `git pull`. This avoids
the complicated shallow cloning / Git having to reconcile the
overlapping but different shallow boundaries, which can cause "error
processing shallow info: 4".
## Testing
- [x] QA'd all new/changed functionality manually
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Refactor**
* Modified repository synchronization to use full fetches instead of
rolling-window shallow fetches.
* Updated sync status messaging for clarity.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
<!-- Add the related story/sub-task/bug number, like Resolves#123, or
remove if NA -->
**Related issue:** Resolves#36751
# Checklist for submitter
If some of the following don't apply, delete the relevant line.
- [X] Changes file added for user-visible changes in `changes/`,
`orbit/changes/` or `ee/fleetd-chrome/changes`.
See [Changes
files](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/guides/committing-changes.md#changes-files)
for more information.
## Testing
- [X] Added/updated automated tests
- [X] QA'd all new/changed functionality manually
- [X] Verified that `fleetctl generate-gitops` correctly outputs
policies with `install_software.fleet_maintained_app_slug` populated
when the policies have FMA automation
- [X] Verified that running `fleetctl gitops` using files with
`install_software.fleet_maintained_app_slug` creates/updates FMA policy
automation correctly
- [X] Verified no changes to the above for custom packages or VPP apps
- [X] Verified that when software is excepted from GitOps, FMA policy
automations still work (correctly validates FMAs exist before applying)
## New Fleet configuration settings
- [ ] Setting(s) is/are explicitly excluded from GitOps
If you didn't check the box above, follow this checklist for
GitOps-enabled settings:
- [X] Verified that the setting is exported via `fleetctl
generate-gitops`
- [ ] Verified the setting is documented in a separate PR to [the GitOps
documentation](https://github.com/fleetdm/fleet/blob/main/docs/Configuration/yaml-files.md#L485)
checking on this
- [X] Verified that the setting is cleared on the server if it is not
supplied in a YAML file (or that it is documented as being optional)
- [X] Verified that any relevant UI is disabled when GitOps mode is
enabled
Ensure downloaded installer files are removed after validation. Add
cleanupInstaller to remove the installer file (ignoring missing files
and logging failures). Propagate a downloaded installer path from
DownloadMaintainedApp (signature now returns the TempFileReader, the
saved file path, and error), write the installer into cfg.tmpDir and set
INSTALLER_PATH in cfg.env. Call cleanupInstaller on error paths and
after successful validation to avoid leftover temp files.
<!-- Add the related story/sub-task/bug number, like Resolves#123, or
remove if NA -->
**Related issue:** Resolves#42180
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Enhanced GitOps exception handling for labels, secrets, and software
with clearer enforcement and omission semantics.
* Server-side prefetch of team software so omitted team software can
preserve existing installers during validation.
* Presence flags track whether top-level keys (labels, secrets,
software) were provided versus omitted.
* **Behavior Changes**
* Omitted vs empty sections are now distinguished: omission can mean
“no-op” or “delete-all” depending on exception settings.
* GitOps YAML can define and manage labels directly; validations now
reject YAML that includes keys marked as excepted.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
# Checklist for submitter
If some of the following don't apply, delete the relevant line.
- [X] Changes file added for user-visible changes in `changes/`,
`orbit/changes/` or `ee/fleetd-chrome/changes`.
See [Changes
files](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/guides/committing-changes.md#changes-files)
for more information.
## Testing
- [X] Added/updated automated tests
- [X] QA'd all new/changed functionality manually
* **Labels**
- [ ] Validated that with label exceptions off, omitting `labels:` key
from default.yml clears all global labels
- [ ] Validated that with label exceptions off, omitting `labels:` key
from a fleet .yml clears all labels for that fleet
- [ ] Validated that with label exceptions off, setting empty `labels:`
key from default.yml clears all global labels
- [ ] Validated that with label exceptions off, setting empty `labels:`
key from a fleet .yml clears all labels for that fleet
- [ ] Validated that with label exceptions on, omitting `labels:` key
from default .yml leaves existing global labels as-is
- [ ] Validated that with label exceptions on, omitting `labels:` key
from a fleet .yml leaves existing labels as-is
- [ ] Validated that with label exceptions on, setting `labels:` key on
default .yml generates an error
- [ ] Validated that with label exceptions on, setting `labels:` key on
a fleet .yml generates an error
- [ ] Validated that with label exceptions on, a policy using
`labels_include_any` referencing an existing label succeeds without
`labels:` key
- [ ] Validated that with label exceptions on, a query using
`labels_include_any` referencing an existing label succeeds without
`labels:` key
- [ ] Validated that with label exceptions on, an MDM profile using
`labels_include_any` referencing an existing label succeeds without
`labels:` key
- [ ] Validated that with label exceptions on, a software package using
`labels_include_any` referencing an existing label succeeds without
`labels:` key (requires software exceptions off)
- [ ] Validated that with label exceptions on, an app store app using
`labels_include_any` referencing an existing label succeeds without
`labels:` key (requires software exceptions off)
- [ ] Validated that with label exceptions on, a fleet maintained app
using `labels_include_any` referencing an existing label succeeds
without `labels:` key (requires software exceptions off)
* **Secrets**
- [ ] Validated that with secrets exceptions off, omitting `secrets:`
key from default.yml clears all global secrets
- [ ] Validated that with secrets exceptions off, omitting `secrets:`
key from a fleet .yml clears all secrets for that fleet
- [ ] Validated that with secrets exceptions on, omitting `secrets:` key
from default .yml leaves existing global secrets as-is
- [ ] Validated that with secrets exceptions on, omitting `secrets:` key
from a fleet .yml leaves existing secrets as-is
- [ ] Validated that with secrets exceptions on, setting `secrets:` key
on default .yml generates an error
- [ ] Validated that with secrets exceptions on, setting `secrets:` key
on a fleet .yml generates an error
* **Software**
- [ ] Validated that with software exceptions off, omitting `software:`
key from no-team.yml/unassigned.yml clears all software for "no team"
- [ ] Validated that with software exceptions off, omitting `software:`
key from a fleet .yml clears all software for that fleet
- [ ] Validated that with software exceptions off, setting empty
`software:` key on a fleet .yml clears all software for that fleet
- [ ] Validated that with software exceptions off, setting empty
`software:` key on no-team.yml/unassigned.yml clears all software for
"no team
- [ ] Validated that with software exceptions on, omitting `software:`
key from a fleet .yml leaves existing software as-is
- [ ] Validated that with software exceptions on, setting `software:`
key on a fleet .yml generates an error
- [ ] Validated that with software exceptions on, omitting `software:`
key from no-team.yml/unassigned.yml leaves existing software as-is for
"no team"
- [ ] Validated that with software exceptions on, setting `software:`
key on no-team.yml/unassigned.yml generates an error
- [ ] Validated that with software exceptions on, a policy using
`install_software.hash_sha256` referencing an existing package succeeds
without `software:` key
- [ ] Validated that with software exceptions on, a policy using
`install_software.app_store_id` referencing an existing VPP app succeeds
without `software:` key
- [ ] Validated that with software exceptions on, a patch policy using
`fleet_maintained_app_slug` referencing an existing FMA succeeds without
`software:` key
- [ ] Validated that with software exceptions on,
`setup_experience.software` referencing existing software succeeds
without `software:` key (server-side validation fallback)
- [ ] Validated that with software exceptions on, omitting `software:`
from no-team.yml/unassigned.yml preserves existing no-team software
- [ ] Validated that with software exceptions on, a policy in
no-team.yml/unassigned.yml using `install_software.hash_sha256`
referencing existing no-team software succeeds without `software:` key
For unreleased bug fixes in a release candidate, one of:
- [X] Confirmed that the fix is not expected to adversely impact load
test results
I don't think so. There is a bit of overhead when this feature is used
since we have to fetch software from the server, but it would be done in
a specific test, so even if there is an impact it should affect existing
load testing, only new, specific tests.
<!-- Add the related story/sub-task/bug number, like Resolves#123, or
remove if NA -->
**Related issue:** Resolves#30453
All the work for this was already done in
https://github.com/fleetdm/fleet/pull/32492. I verified that
`generate-gitops` outputs the expected categories, and that `gitops`
reads them back in correctly. This PR just removes the warning
<!-- Add the related story/sub-task/bug number, like Resolves#123, or
remove if NA -->
**Related issue:** Resolves#39842
Replaces the call to
`maintained_apps.FetchAppsList(context.Background())` which downloads
the apps.json list from github to the list Fleet-maintained apps
endpoint. This is so we can match apps by their Fleet-maintained app ID
instead of by name which can cause problems when a Windows FMA is
associated to a title with the wrong name (e.g. title is `7-Zip 23.01`
instead of the current FMA's name `7-Zip 25.01`).
# Checklist for submitter
If some of the following don't apply, delete the relevant line.
- [x] Changes file added for user-visible changes in `changes/`,
`orbit/changes/` or `ee/fleetd-chrome/changes`.
See [Changes
files](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/guides/committing-changes.md#changes-files)
for more information.
- [ ] Input data is properly validated, `SELECT *` is avoided, SQL
injection is prevented (using placeholders for values in statements), JS
inline code is prevented especially for url redirects, and untrusted
data interpolated into shell scripts/commands is validated against shell
metacharacters.
- [ ] If paths of existing endpoints are modified without backwards
compatibility, checked the frontend/CLI for any necessary changes
## Testing
- [x] Added/updated automated tests
- [ ] Where appropriate, [automated tests simulate multiple hosts and
test for host
isolation](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/reference/patterns-backend.md#unit-testing)
(updates to one hosts's records do not affect another)
- [x] QA'd all new/changed functionality manually
<!-- Add the related story/sub-task/bug number, like Resolves#123, or
remove if NA -->
**Related issue:** Resolves#41670
# Checklist for submitter
If some of the following don't apply, delete the relevant line.
- [X] Changes file added for user-visible changes in `changes/`,
`orbit/changes/` or `ee/fleetd-chrome/changes`.
See [Changes
files](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/guides/committing-changes.md#changes-files)
for more information.
- [X] Input data is properly validated, `SELECT *` is avoided, SQL
injection is prevented (using placeholders for values in statements), JS
inline code is prevented especially for url redirects, and untrusted
data interpolated into shell scripts/commands is validated against shell
metacharacters.
## Testing
- [X] Added/updated automated tests
- [ ] QA'd all new/changed functionality manually
## Database migrations
- [X] Checked schema for all modified table for columns that will
auto-update timestamps during migration.
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Automatic recovery lock password rotation for Mac devices—passwords
now rotate 1 hour after being viewed or accessed via the API, enhancing
security.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
Resolves#40396.
No changes file because there should be no user visible changes.
## Testing
- [x] QA'd all new/changed functionality manually
## fleetd/orbit/Fleet Desktop
- [x] Verified compatibility with the latest released version of Fleet
(see [Must
rule](https://github.com/fleetdm/fleet/blob/main/docs/Contributing/workflows/fleetd-development-and-release-strategy.md))
- [X] Verified that fleetd runs on macOS, Linux and Windows
- [X] Verified auto-update works from the released version of component
to the new version (see [tools/tuf/test](../tools/tuf/test/README.md))
Martin Angers