This adds a new mechanism to allow us to handle compatibility issues between Orbit, Fleet Server and Fleet Desktop.
The general idea is to _always_ send a custom header of the form:
```
fleet-capabilities-header = "X-Fleet-Capabilities:" capabilities
capabilities = capability * (,)
capability = string
```
Both from the server to the clients (Orbit, Fleet Desktop) and vice-versa. For an example, see: 8c0bbdd291
Also, the following applies:
- Backwards compat: if the header is not present, assume that orbit/fleet doesn't have the capability
- The current capabilities endpoint will be removed
### Motivation
This solution is trying to solve the following problems:
- We have three independent processes communicating with each other (Fleet Desktop, Orbit and Fleet Server). Each process can be updated independently, and therefore we need a way for each process to know what features are supported by its peers.
- We originally implemented a dedicated API endpoint in the server that returned a list of the capabilities (or "features") enabled, we found this, and any other server-only solution (like API versioning) to be insufficient because:
- There are cases in which the server also needs to know which features are supported by its clients
- Clients needed to poll for changes to detect if the capabilities supported by the server change, by sending the capabilities on each request we have a much cleaner way to handling different responses.
- We are also introducing an unauthenticated endpoint to get the server features, this gives us flexibility if we need to implement different authentication mechanisms, and was one of the pitfalls of the first implementation.
Related to https://github.com/fleetdm/fleet/issues/7929
Configuration and fixes for the Fleet server and frontend to add support
for https://github.com/Uptycs/kubequery.
Co-authored-by: Michal Nicpon <michal@fleetdm.com>
* Bump go to 1.19.1
* Bump remaining go-version to the 1.19.1
* Add extra paths for test-go
* Oops, putting the right path in the right place
* gofmt file
* gofmt ALL THE THINGS
* Moar changes
* Actually, go.mod doesn't like minor versions
- Add a new "Configuration for contributors" doc page. Move settings that are not recommended for production use
- Remove settings modified in the `config` YAML document from the deploying/configuration doc page
- Document all keys in `config` and `teams` YAML documents
- Add comments to several `.go` files and remove unused struct
This fixes an unreleased bug I introduced in eeefe2, as the config colum in the teams table is nulleable, it unmarshalls into nil and we can't dereference the variable
This adjusts functions related to applying team specs to make sure the `features` key works as expected, this is:
1. Create Team flow
1. if no features are provided, we copy whatever features are set in the global config
2. otherwise, we generate a new config merging whatever is provided with the defaults for new installs.
2. Edit Team flow: we always generate a new config merging whatever is provided with the defaults for new installs.
* Make Change software vendor width mig. more robust
If there are two or more entries in the software table with the same name, version, source, release and arch but different vendors then the migration used for increasing the vendor width will fail.
* Ingest installed Windows updates and store them in the windows_updates table.
* Added config option for enabling/disabling Windows update ingestion and Windows OS vuln. detection.
Related to https://github.com/fleetdm/fleet/issues/7312, the motivation behind these changes is to introduce a way to deprecate configurations in `AppConfig`, while still preserving backwards compatibility.
From the Epic:
> NOTE: `host_settings` is now replaced by `features`. We should still support `host_settings` as an alias to `features` for backwards compatibility, but in our communications, we should use and recommend features as the canonical way forward.
* Bug 7320: Fixed dulp. vulns. on software details.
The software details page was showing duplicated vulnerabilities if the software was being used by many hosts.
this adjusts the TestInstallerHeadCheck check to use `DoRaw` instead of `Do`, which sends a body with null when nil is passed for the params parameter.
Updated migration used to increate the width of 'software.vendor' to
address issue found when running migration in aurora: "ALGORITHM=INPLACE is not
supported. Reason: cannot silently convert NULL values, as required in
this SQL_MODE. Try ALGORITHM=COPY."
* improve error handling in vulnerabilities cron
* fix tests
* Use errHandler and go mod tidy
* Add dep that got removed by mod tidy
* add dsl to tools
* Add changes file
Co-authored-by: Michal Nicpon <michal@fleetdm.com>
Co-authored-by: Tomas Touceda <chiiph@gmail.com>
* add google analytics to sandbox instances
* Add serverType variable to frontend handler
* update version of html-webpack-plugin
Co-authored-by: Lucas Rodriguez <lucas@fleetdm.com>
Related to #6894, this entirely replaces FLEET_DEMO with the server config added in #6597
As part of this, I also implemented a small refactor to the integration test suite to allow setting a custom config when the server is initialized.
This commit replaces `os.Setenv` with `t.Setenv` in tests. The
environment variable is automatically restored to its original value
when the test and all its subtests complete.
Reference: https://pkg.go.dev/testing#T.Setenv
Signed-off-by: Eng Zer Jun <engzerjun@gmail.com>
This improves the installerstore CLI tool with:
- The ability to create tests buckets for local development (otherwise you have to interact with another CLI or the MinIO UI)
- Improved error handling and messaging.
Related to #6365 this adds a new tool to upload pre-built Orbit installers to a storage blob. It uses the same file conventions that the Fleet server expects, making it useful for local testing and infra envs alike.\
Usage and details in the README
Related to #6365, this extends the datastore/s3 package to retrieve installers from S3 according to the conventions listed in the parent issue. This also includes:
- A minor refactor to decouple Carves-related functionality from the core S3 functionality
- Set-up to run tests using minio (only enabled via the FILE_STORAGE_TEST env flag)
* Add rate limits for device authed endpoints
* Fix lint
* Add missing test
* Fix test
* Increase the quota for desktop endpoints
* Add comment about quota
* Add server support for Fleet Sandbox demo login
This adds an endpoint `/api/latest/fleet/demologin` that provides a
redirect for the fleetdm.com portion of Fleet Sandbox to automatically
log in a user. The username and password must be provided as form
values. The endpoint is only enabled if `FLEET_DEMO=1` is set in the
server environment.
This was tested locally with the following HTML served by `python3 -m
http.server`, and the Fleet server running with `FLEET_DEMO=1
./build/fleet serve --dev`:
```
<!DOCTYPE html>
<body>
<form
method="post"
action="https://localhost:8080/api/latest/fleet/demologin"
id="demologin"
>
<input type="hidden" name="email" value="admin@example.com" />
<input type="hidden" name="password" value="admin123123#" />
<input type="submit"/>
</form>
<script type="text/javascript">
document.forms["demologin"].submit();
</script>
</body>
</html>
```
For Fleet sandbox purposes, the `action` should be set to the correct
hostname for the sandbox instance, while the `email` and `password`
should be set to the same credentials that were provided when creating
the instance.
* lucas comments
* Add integration tests
* Fix status codes and add comments
Co-authored-by: Martin Angers <martin.n.angers@gmail.com>
Related to #5898, this reports an anonymized summary of errors stored in Redis into the analytics payload.
For each error stored, this includes:
- A `count` attribute with the number of occurrences of the error
- A `loc` attribute with the 3 topmost lines in the stack trace. Note that stack traces only contain package name + line number (example: github.com/fleetdm/fleet/server.go:12
This also includes a minor refactor around error types.
This commit replaces `ioutil.TempDir` with `t.TempDir` in tests. The
directory created by `t.TempDir` is automatically removed when the test
and all its subtests complete.
Prior to this commit, temporary directory created using `ioutil.TempDir`
needs to be removed manually by calling `os.RemoveAll`, which is omitted
in some tests. The error handling boilerplate e.g.
defer func() {
if err := os.RemoveAll(dir); err != nil {
t.Fatal(err)
}
}
is also tedious, but `t.TempDir` handles this for us nicely.
Reference: https://pkg.go.dev/testing#T.TempDir
Signed-off-by: Eng Zer Jun <engzerjun@gmail.com>
* Do not use golangci action for better reproducibility
* Add fix to trigger build
* Fix all reported issues
* fix more lint errors
* Add missing import
* Remove unused method
* Remove change not necessary
Related to #6063, this adds a new device API to get an object with boolean values that we can use as feature flags to manage backwards compatibility in Fleet Desktop.
Feature: Improve our capability to detect vulnerable software on Ubuntu hosts
To improve the capability of detecting vulnerable software on Ubuntu, we are now using OVAL definitions to detect vulnerable software on Ubuntu hosts. If data sync is enabled (disable_data_sync=false) OVAL definitions are automatically kept up to date (they are 'refreshed' once per day) - there's also the option to manually download the OVAL definitions using the 'fleetctl vulnerability-data-stream' command. Downloaded definitions are then parsed into an intermediary format and then used to identify vulnerable software on Ubuntu hosts. Finally, any 'recent' detected vulnerabilities are sent to any third-party integrations.
This removes policy information from `GET /api/_version_/fleet/device/{token}` from non-premium Fleet instances.
Starting the server with `./build/fleet serve --dev --dev_license`
```bash
$ curl -s https://localhost:8080/api/latest/fleet/device/1804e808-171f-4dda-9bec-f695b2f2371a | jq '.host.policies // "not present"'
[
{
"id": 3,
"name": "Antivirus healthy (Linux)",
"query": "SELECT score FROM (SELECT case when COUNT(*) = 2 then 1 ELSE 0 END AS score FROM processes WHERE (name = 'clamd') OR (name = 'freshclam')) WHERE score == 1;",
"description": "Checks that both ClamAV's daemon and its updater service (freshclam) are running.",
"author_id": 1,
"author_name": "Roberto",
"author_email": "test@example.com",
"team_id": null,
"resolution": "Ensure ClamAV and Freshclam are installed and running.",
"platform": "darwin,linux",
"created_at": "2022-05-23T20:53:36Z",
"updated_at": "2022-06-03T13:17:42Z",
"response": ""
}
]
```
Starting the server with `./build/fleet serve --dev`
```bash
$ curl -s https://localhost:8080/api/latest/fleet/device/1804e808-171f-4dda-9bec-f695b2f2371a | jq '.host.policies // "not present"'
"not present"
```
This adds a new API client named DeviceClient to server/service, meant to consume device endpoints and be used from Fleet Desktop.
Some of the logic to make requests and parse responses was very repetitive, so I introduced a private baseClient type and moved some of the logic of the existent Client there.
Related to #5685 and #5697
This adds a new device authenticated endpoint, `/api/_version_/fleet/device/{token}/policies` to retrieve the device policies.
An example request / response looks like:
```bash
curl https://localhost:8080/api/latest/fleet/device/1804e808-171f-4dda-9bec-f695b2f2371a/policies
```
```json
{
"policies": [
{
"id": 3,
"name": "Antivirus healthy (Linux)",
"query": "SELECT score FROM (SELECT case when COUNT(*) = 2 then 1 ELSE 0 END AS score FROM processes WHERE (name = 'clamd') OR (name = 'freshclam')) WHERE score == 1;",
"description": "Checks that both ClamAV's daemon and its updater service (freshclam) are running.",
"author_id": 1,
"author_name": "Admin",
"author_email": "admin@example.com",
"team_id": null,
"resolution": "Ensure ClamAV and Freshclam are installed and running.",
"platform": "darwin,windows,linux",
"created_at": "2022-05-23T20:53:36Z",
"updated_at": "2022-05-23T20:53:36Z",
"response": "fail"
}
]
}
```
Related to [#5685](https://github.com/fleetdm/fleet/issues/5685), in another changeset I will be adding "client" endpoints so we can consume this endpoint from Fleet Desktop
Related to #5776, this ensures that when a host is re-enrolled on a different team we cleanup existing policy memberships in the same way we do when a host is assigned a team through the API.
