Fixes from the changes added to #17003.
- 50% failure for the software query was not realistic (changing to 5%).
- 50% failure for the VS Code query was also not realistic (changing to
5%).
- Renamed a wrongly named variable.
#15565
Replace the use of the isFederated registry key with a keys that check
for AAD (Azure Active Directory, now Entra ID)
Federated enrollment (`isFederated`) seems to be when windows uses a
Discovery MDM endpoint to get its policy and management endpoint
configuration. This is always the case when a client is enrolled with
fleet, so installations always show up as automatic.
It's being replaced by a different key, `AADResourceID`, which appears
to identify the resource that controls the automated deployment. In my
tests it only appears to be populated when the computer is enrolled
through automated deployments. This key appears on both Windows 10 and
11.
There is a similar key, `AADTenantID`, which appears to identify the
client (tenant) to the Azure cloud. I haven't seen this ID in our
systems, so it is likely exclusively used in Azure. Both this key and
`AADResourceID` seem to always be set at the same time, so we only
check for the `AADResourceID`.
I've also added documentation on the registry keys I've analyzed for future reference.
#17003
- [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.
- [X] 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
This is a bug in the generation of results of scheduled queries in
osquery-perf.
It seems the bug has been around since we added scheduled query result
support in osquery-perf.
PS: In my Fleet downtime tests for
https://github.com/fleetdm/fleet/issues/16423 I probably missed this
because I was using a high frequency queries (150s intervals IIRC).
Orbit changes for #16423.
Should also fix#16326 (in case of network errors).
Orbit will log the following every 5 minutes:
```
2024-02-20T14:27:40-03:00 INF network error error="Post \"https://localhost:8080/api/fleet/orbit/config\": dial tcp [::1]:8080: connect: connection refused"
```
- [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.
- [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.
- [x] Auto-update manual QA, from released version of component to new
version (see [tools/tuf/test](../tools/tuf/test/README.md)).
#16423
- We need to consume less memory while Fleet is down (otherwise we might
bring the load test environment down, local tests showed that in 30m I
got to 600 MB of memory usage) so am generating the results when needed
only.
- Also fixing the release of fasthttp requests and responses (according
to their docs).
#15703
`wall_time_ms` (not `wall_time`) coming from osquery will now be put
into `wall_time` in our DB.
Hence, `wall_time` coming from Fleet API will now be in milliseconds and
not seconds
# Checklist for submitter
If some of the following don't apply, delete the relevant line.
<!-- Note that API documentation changes are now addressed by the
product design team. -->
- [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.
- [x] Added/updated tests
- [x] If database migrations are included, checked table schema to
confirm autoupdate
- For database migrations:
- [x] Checked schema for all modified table for columns that will
auto-update timestamps during migration.
- [x] Confirmed that updating the timestamps is acceptable, and will not
cause unwanted side effects.
- [x] Manual QA for all new/changed functionality
These are the osquery-perf changes for the load test to be performed in
#16423.
- Adding buffering of results when they fail to be sent to the server
(when Fleet is offline/down).
- I'm changing many `a.waitingDo(req, res)` to just do one request and
if it fails it will retry on the next interval, which is what osquery
does on
`/api/osquery/config`/`/api/osquery/distributed/read`/`/api/osquery/distributed/write`/`/api/osquery/log`.
- I'm reducing the logs as they will accumulate considerably while Fleet
is down for 30m with 140k hosts.
- Changing from `fmt.Printfs` to `log.Printfs` for consistency.
- Adding more fine grained stats (to know which endpoints are failing)
- Adding a mode to disable Fleet Desktop to simulate what the customer
is running.
- Adding `logger_tls_max_lines` flag that simulates the osquery setting.
Added `gigs_total_disk_space` to host endpoint responses.
- Now user will know host disk size even when disk is full
(percent_disk_space_available==0)
#15058
# Checklist for submitter
If some of the following don't apply, delete the relevant line.
<!-- Note that API documentation changes are now addressed by the
product design team. -->
- [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.
- [x] 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
📺 Loom explaining the PR:
https://www.loom.com/share/8c6ec0c362014910931f183d68307525?sid=7de51e6f-a59d-4d8f-b06b-f703f1692f17#467
Live Queries now collect stats. Stats are collected for saved queries
present in the Queries tab.
- After running a live query, user will see updated stats in Queries
tab.
- Query stats on Host Details page will no longer be cleared after
host/agent reboots.
- Query stats are now deleted when query is deleted.
# Checklist for submitter
- [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.
- [x] Input data is properly validated, `SELECT *` is avoided, SQL
injection is prevented (using placeholders for values in statements)
- [x] Added/updated tests
- [x] Manual QA for all new/changed functionality
This PR introduces the following fixes:
- Fixes a race condition by protecting the `scheduledQueries` and
`scheduledQueryData` fields with a `sync.Mutex` (*).
- Adds some more information about request counts for /log, /config and
/distributed/read requests (and uptime).
- Fixes the resource leaks around creating time.Ticks on every for loop
iteration.
(*) Sample of the race condition when running with `-race`:
```
==================
WARNING: DATA RACE
Read at 0x00c000604800 by goroutine 20:
main.(*agent).runLoop()
/Users/luk/fleetdm/git/fleet/cmd/osquery-perf/agent.go:525 +0x71b
main.main.func2()
/Users/luk/fleetdm/git/fleet/cmd/osquery-perf/agent.go:1737 +0x4e
Previous write at 0x00c000604800 by goroutine 40:
main.(*agent).config()
/Users/luk/fleetdm/git/fleet/cmd/osquery-perf/agent.go:915 +0xb30
main.(*agent).runLoop.func2()
/Users/luk/fleetdm/git/fleet/cmd/osquery-perf/agent.go:512 +0x37
```
This is safe to merge during freeze as this is an osquery-perf only
change.
```
panic: runtime error: index out of range [1] with length 1
goroutine 14 [running]:
main.(*agent).config(0xc0001e5b80)
/Users/luk/fleetdm/git/fleet/cmd/osquery-perf/agent.go:847 +0xb74
main.(*agent).runLoop(0xc0001e5b80, 0x0?, 0x0?)
/Users/luk/fleetdm/git/fleet/cmd/osquery-perf/agent.go:463 +0x4d2
created by main.main in goroutine 1
/Users/luk/fleetdm/git/fleet/cmd/osquery-perf/agent.go:1666 +0x1285
```
#13527
(Adding @mna to double check the changes in the async implementation of
policy result storage)
This PR also adds the osquery-perf changes needed to define the count of
macOS and Windows hosts.
- [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 (docs/Using
Fleet/manage-access.md)~
- [X] Input data is properly validated, `SELECT *` is avoided, SQL
injection is prevented (using placeholders for values in statements)
- [X] 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)).~
Test with 80k hosts: 70k simulated macOS, 10k simulated Windows.
Apply Windows policies first, then apply macOS policies:
```
fleetctl apply -f ee/cis/win-10/cis-policy-queries.yml
# Leave running for some time
fleetctl apply -f ee/cis/macos-13/cis-policy-queries.yml
```
After applying CIS policies previous to these changes:
After applying these changes and applying the same policies:
These osquery-perf changes were useful to test/troubleshoot #10957.
It basically allows osquery-perf to simulate no results and/or failures
when running live queries on the devices.
- ~[ ] 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)~
- [X] 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)).~
#11528
osquery-perf simulated hosts enroll and are identified as manually
enrolled. (Enrolling as DEP requires more work, e.g. a new mocked Apple
DEP endpoint).
Given that these are simulated MDM clients, they cannot be woken up with
push notifications. Instead, these check for new commands to execute
every 10 seconds (which is not realistic, but could serve as a good
loadtesting exercise).
I will now start setting up the loadtest environment with MDM enabled
and configured to test this.
- ~[ ] 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)~
- [X] 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)).~
#7970
- [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.
- [x] Auto-update manual QA, from released version of component to new
version (see [tools/tuf/test](../tools/tuf/test/README.md)).
https://github.com/fleetdm/confidential/issues/1968
It's ready for review but I still need to load test this.
- [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~
- [X] Input data is properly validated, `SELECT *` is avoided, SQL
injection is prevented (using placeholders for values in statements)
- [X] 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)).~
#9132
The actual fix for the empty hosts is adding the `--database_path`
argument in the initial `osqueryd -S` invocation when retrieving the
UUID. Osquery attempts to retrieve the UUID from OS files/APIs, when not
possible (which is what happens on some linux distributions), then it
resorts to generating a new random UUID and storing it in the
`osquery.db`. The issue was Orbit's first invocation of `osqueryd -S`
was not using the same `osquery.db` as the main daemon invocation of
`osqueryd`.
I'm also adding a `hostname` + `platform` to the orbit enroll phase so
that if there are any issues in the future we can avoid the "empty" host
and have some information to help us troubleshoot.
## How to reproduce
On Linux, osquery reads `/sys/class/dmi/id/product_uuid` to load the
hardware UUID.
Some Linux distributions running on specific hardware or container
environments do not have such file available.
The way to reproduce on a Linux VM is to do the following:
```sh
$ sudo su
# chmod -r /sys/class/dmi/id/product_uuid
```
which will turn the file inaccessible to root.
## Checklist
- [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~
- [X] Input data is properly validated, `SELECT *` is avoided, SQL
injection is prevented (using placeholders for values in statements)
- [X] 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.
- [x] Auto-update manual QA, from released version of component to new
version (see [tools/tuf/test](../tools/tuf/test/README.md)).
This improves osquery-perf with support for a more realistic orbit + fleet desktop simulation as described in #8212
This was based on the work done by @sharvilshah in his branch.
users, software_macos and software_linux queries can be optionally disabled in the fleet server via the enable_host_users and enable_software_inventory config settings, however osquery_perf is always running them.
this adjusts the code accordingly to only execute them if the server asks.
