- Don't cache requests to get profiles, they don't happen in order when
run agains a real Puppet server
- Report resource has changed if the profile was in the host but the new
resource type is absent
this helps consumer of the datastore method handle the not found
scenario better and ensures we always return a 4xx code by default if we
can't find a matching team.
seems like calls to this method were special-cased everywhere except in
the apply user roles endpoint, where we returned a `500` status code if
we couldn't find a team.
Related to #12608, this automatically sets the
`DeferForceAtUserLoginMaxBypassAttempts` property to `1` on the
FileVault profile that's generated by Fleet.
This changeset also includes a migration to modify old FileVault
profiles that already exist in the database, and by virtue of that a
`InstallProfile` command will be issued to hosts that already have FV
enabled. During testing we found:
1. This doesn't affect users with FV already installed, they silently
get the profile updated without any changes.
2. Since the profile needs to be re-delivered, it'll go through the full
"pending" -> "verifying" -> "verified" cycle.
#10292, #12554
When scanning tens of thousands of files for permissions, using the
`find` command exposed as a fleetd table is more performant than trying
to use the `file` table. This change caused the watchdog to *stop*
killing osquery because of exceeding memory or CPU limit.
- [X] Changes file added for user-visible changes in `changes/` or
`orbit/changes/`.
See [Changes
files](https://fleetdm.com/docs/contributing/committing-changes#changes-files)
for more information.
- ~[ ] Documented any API changes (docs/Using-Fleet/REST-API.md or
docs/Contributing/API-for-contributors.md)~
- ~[ ] Documented any permissions changes~
- ~[ ] Input data is properly validated, `SELECT *` is avoided, SQL
injection is prevented (using placeholders for values in statements)~
- ~[ ] Added support on fleet's osquery simulator `cmd/osquery-perf` for
new osquery data ingestion features.~
- [X] Added/updated tests
- [X] Manual QA for all new/changed functionality
- For Orbit and Fleet Desktop changes:
- [X] Manual QA must be performed in the three main OSs, macOS, Windows
and Linux.
- ~[ ] Auto-update manual QA, from released version of component to new
version (see [tools/tuf/test](../tools/tuf/test/README.md)).~
#10292
The query was processing *every* file under `/Applications/`, which
makes it super expensive both in CPU usage and Memory footprint. This
query was the main culprit of triggering worker process kills by the
watchdog.
On some runs it triggered CPU usage alerts:
```
7716:W0623 15:38:05.402959 221732864 watcher.cpp:415] osqueryd worker (72976) stopping:
Maximum sustainable CPU utilization limit 1200ms exceeded for 12 seconds
```
And on other runs it triggered memory usage alerts:
```
4431 W0626 07:28:50.868021 147312640 watcher.cpp:424] osqueryd worker (21453) stopping:
Memory limits exceeded: 214020096 bytes (limit is 200MB)
```
For the above logs I used a custom osqueryd branch to be able to print
more information: https://github.com/osquery/osquery/pull/8070
The metrics for the old query were CPU usage: ~4521 ms
```
435:level=warn ts=2023-06-26T09:58:29.665712Z query=fleet_policy_query_1233 queryTime=4521 memory=12226560 msg="distributed query performance is excessive" hostID=308 platform=darwin
```
With the new query, CPU usage: ~210 ms.
```
23893:level=debug ts=2023-06-26T18:06:08.242456Z query=fleet_policy_query_1233 queryTime=210 msg=stats memory=0 hostID=308 platform=darwin
```
Basically a ~20x improvement.
- [X] Changes file added for user-visible changes in `changes/` or
`orbit/changes/`.
See [Changes
files](https://fleetdm.com/docs/contributing/committing-changes#changes-files)
for more information.
- ~[ ] Documented any API changes (docs/Using-Fleet/REST-API.md or
docs/Contributing/API-for-contributors.md)~
- ~[ ] Documented any permissions changes~
- ~[ ] Input data is properly validated, `SELECT *` is avoided, SQL
injection is prevented (using placeholders for values in statements)~
- ~[ ] Added support on fleet's osquery simulator `cmd/osquery-perf` for
new osquery data ingestion features.~
- ~[ ] Added/updated tests~
- [X] Manual QA for all new/changed functionality
- For Orbit and Fleet Desktop changes:
- ~[ ] Manual QA must be performed in the three main OSs, macOS, Windows
and Linux.~
- ~[ ] Auto-update manual QA, from released version of component to new
version (see [tools/tuf/test](../tools/tuf/test/README.md)).~
## Addresses #11037
### Implement the `privacy_preferences` table for the Fleetd Chrome
extension. Columns correspond to the available properties of
[`chrome.privacy`](https://developer.chrome.com/docs/extensions/reference/privacy/).
Chrome on mac:
<img width="816" alt="Screenshot 2023-06-23 at 11 55 21 AM"
src="https://github.com/fleetdm/fleet/assets/61553566/a4700749-6325-442e-acf2-c14b1c9adf8f">
Chromebook with enterprise access (actual use case):
* Chromebook w/o enterprise access: as you can see, sometimes certain
APIs are not available - this error occurs because the expected API
object that would have a `get` method is actually `undefined` TODO – How
to handle this case given that we want to let errors bubble up to the
level at which Fleet can catch them? Maybe it would be nice to catch
such errors and send them up to the Fleet layer, and still allow the
loop to continue to populate the columns whose APIs _are_ available.
_Decision: catch API errors here to preserve functionality of the
remaining columns_
- [x] Changes file
- [x] Manual QA
---------
Co-authored-by: Jacob Shandling <jacob@fleetdm.com>
https://github.com/fleetdm/fleet/issues/10602
@xpkoala this PR will require testing of all modified items.
Preferably, we should perform the tests before merging to master. Can we
use the dev branch for that? -- Items were tested locally.
this PR modifies the `external_host_identifier` parameter that's used to
match hosts to Puppet runs to use an identifier that's *unique per run*
(instead of an identifier that's *unique per host*)
this has the adventage to:
1. allow for concurrent Puppet runs that don't interfere with each
other.
2. allow for failed/orphaned Puppet runs to not interfere with new runs
(the keys will eventually get expired)
all the existent behavior should be preserved.
> Note: I have verified that the value that the reporter gets is the one
associated with the right puppet run, even if multiple runs happen
simultaneously.
A few minor things going on:
1. Adjusted the Puppet module to send the profiles base64 encoded
2. Enabled FileVault by default on teams created using the `/match`
endpoint.
3. Remove profiles when a team is removed. We can't do a foreign key
because the global team.id is NULL. I also included a migration to
cleanup orphaned profiles.
This relates to #11312
`18.9.17.6`: Fixing the issue with policy pointing to a different GPO
and Registry value
`18.9.47.4.2`: Adding COLLATE NOCASE to avoid case sensitive issue with
SpynetReporting value
#11266
PS: I first attempted a serialization trick by introducing a new
`appConfigResponse` and implementing `json.Marshal` to exclude these
fields but it was too hacky and hard to maintain moving forward, so I'm
bitting the bullet now. Happy to hear other ideas.
- [X] Changes file added for user-visible changes in `changes/` or
`orbit/changes/`.
See [Changes
files](https://fleetdm.com/docs/contributing/committing-changes#changes-files)
for more information.
- ~[ ] Documented any API changes (docs/Using-Fleet/REST-API.md or
docs/Contributing/API-for-contributors.md)~
- ~[ ] Documented any permissions changes~
- ~[ ] Input data is properly validated, `SELECT *` is avoided, SQL
injection is prevented (using placeholders for values in statements)~
- ~[ ] Added support on fleet's osquery simulator `cmd/osquery-perf` for
new osquery data ingestion features.~
- [X] Added/updated tests
- [X] Manual QA for all new/changed functionality
- ~For Orbit and Fleet Desktop changes:~
- ~[ ] Manual QA must be performed in the three main OSs, macOS, Windows
and Linux.~
- ~[ ] Auto-update manual QA, from released version of component to new
version (see [tools/tuf/test](../tools/tuf/test/README.md)).~
Related to #11185 this adds a Puppet module that provides:
1. A custom type named `fleetdm::profile` that can be used to define
profiles to a device
2. A function named `fleetdm::release_device` that can be used to
release a device from await device configuration.
Instructions/usage can be found in the `README.md` file.
---------
Co-authored-by: Martin Angers <martin.n.angers@gmail.com>
#10784
The removal of the now deprecated `sso_settings.enable_jit_role_sync`
config will be tackled in: #10688.
- [X] Changes file added for user-visible changes in `changes/` or
`orbit/changes/`.
See [Changes
files](https://fleetdm.com/docs/contributing/committing-changes#changes-files)
for more information.
- ~[ ] Documented any API changes (docs/Using-Fleet/REST-API.md or
docs/Contributing/API-for-contributors.md)~
- ~[ ] Documented any permissions changes~
- ~[ ] Input data is properly validated, `SELECT *` is avoided, SQL
injection is prevented (using placeholders for values in statements)~
- ~[ ] Added support on fleet's osquery simulator `cmd/osquery-perf` for
new osquery data ingestion features.~
- [X] Added/updated tests
- [X] Manual QA for all new/changed functionality
- ~For Orbit and Fleet Desktop changes:~
- ~[ ] Manual QA must be performed in the three main OSs, macOS, Windows
and Linux.~
- ~[ ] Auto-update manual QA, from released version of component to new
version (see [tools/tuf/test](../tools/tuf/test/README.md)).~
