-
Notifications
You must be signed in to change notification settings - Fork 8.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Endpoint] Adding a Resolver component that lets the user click and drag to pan and pinch to zoom #53619
[Endpoint] Adding a Resolver component that lets the user click and drag to pan and pinch to zoom #53619
Conversation
c0054d9
to
61d8ec2
Compare
return [rect, ref]; | ||
} | ||
|
||
const DiagnosticDot = styled( |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@bkimmel here's where a component can find its position relative to the parent resolver component
85d987c
to
b0c6385
Compare
💔 Build Failed
History
To update your PR or re-run it, just comment with: |
26cac31
to
0e4a5a6
Compare
"license": "Elastic-License", | ||
"scripts": {}, | ||
"dependencies": { | ||
"react-redux": "^7.1.0" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We depend on react-redux's hooks.
c4ce449
to
a02a0d4
Compare
const [renderTarget, setRenderTarget] = React.useState<HTMLDivElement | null>(null); | ||
|
||
useEffect(() => { | ||
let cleanUp; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
❔ If you're using cleanUp for something here, it is really not clear what. I see it gets the reject side of the Promise below, but I don't understand where or how it's used at all. Can you explain?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I added comments to explain but i'll reiterate here. The 'effect' is responsible for waiting for a promise to resolve and then setting component 'state' from that resolved value.
There is complexity because the promise comes in as a React prop. The prop could change at any point. When the prop changes, we call Promise.race
. If the value is resolved, we set the value in state. If the prop changes, we cancel the race by rejecting.
608d94f
to
eb1521c
Compare
package.json
Outdated
@@ -99,6 +99,7 @@ | |||
"x-pack", | |||
"x-pack/plugins/*", | |||
"x-pack/legacy/plugins/*", | |||
"x-pack/plugins/endpoint", |
This comment was marked as resolved.
This comment was marked as resolved.
Sorry, something went wrong.
This comment was marked as resolved.
This comment was marked as resolved.
Sorry, something went wrong.
This comment was marked as resolved.
This comment was marked as resolved.
Sorry, something went wrong.
This comment was marked as resolved.
This comment was marked as resolved.
Sorry, something went wrong.
/** | ||
* Return a new matrix which is the sum of the two passed in. | ||
*/ | ||
export function add( |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Since all of these seem to have name collisions between matrix v. vector addition, multiplication, etc. wondering if it makes sense to nest this export under some sort of class or named constant. Just want to avoid littering code with things like:
import { add as matrixAdd } from 'matrix3'
import { add as vectorAdd } from 'vector2'
as well as having to back reference whether we're operating on matrices, vectors, lists, etc. because we've lost all context of what add
we're working with at that point...
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
data_buffer: { event_type_full: type, event_subtype_full: subType }, | ||
} = event; | ||
|
||
if (type === 'process_event') { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
it would be nice to have this stuff scoped to particular transforms for endgame v0 events, mainly because none of this stuff is ECS compatible as it stands at the moment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This will change once we use real, ECS compatible data, as opposed to sample data. We will hopefully do this soon.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@andrewstucki I've never been able to get the streaming feature to produce a set of events that could be used for resolver. in all of my attempts, enough events are missing that I can't render a resolver graph from it. If you could provide some more realistic sample data we'll use it. Otherwise we can wait for @jonathan-buttner to finish a Resolver data generator. @jonathan-buttner is that ready?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
So, we should sit down together and see what you actually need, because the events shipped have literally everything in them that are part of our raw events--so if there's not enough data, we have a problem, because it means that the endpoint itself isn't shipping data compatible with what you're trying to visualize.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
sorry for the confusing explanation. the events have enough data, I just can't get the streaming feature to send all of the events.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@oatkiller the data generator is close but still working out some issues. We haven't started on a backend implementation yet but that's next on the list.
} | ||
|
||
export function uniqueParentPidForProcess(event: ProcessEvent) { | ||
return event.data_buffer.source_id; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'd be very careful with nomenclature here, because this doesn't look like an "event" to me--"event"s are our raw data sources, for which we don't have a "source_id" field. This looks more like some sort of investigation or alert, something that originally came from the hunt_data_schema
--if that's true then the applications for resolver out of the box are going to be very limited. Instead this should be working on the raw event data if possible. In which case:
v0
(data coming from the SMP) will be looking at the fieldendgame.unique_ppid
for process eventsv1
(data coming from Elastic-native sensor) will be looking at the fieldendpoint.process.parent.entity_id
for events withevent.category == 'process'
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Let's sync on all of this. But basically, this is using some random sample data that has pids, unique ppids, and process names. that's all that we need to test the layout logic. Once we get a BE, we'll get this stuff working against real data.
} | ||
|
||
export function uniquePidForProcess(event: ProcessEvent) { | ||
return event.data_buffer.node_id; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
see comment about source_id
stuff:
v0
--endgame.unique_pid
v1
--endpoint.process.entity_id
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm not totally sure how types will work with events, but I image having a single Event
type which encompasses all of the events that we'll ever deal with. When we change that type, ts will help us fix all the code that references it. If we get sample data that has the right format, we'll change the type and all of the code then.
|
||
export function mockProcessEvent( | ||
parts: { | ||
data_buffer: { node_id: ProcessEvent['data_buffer']['node_id'] }; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
See comments about what a "process event" is -- none of these fields exist in process events that we ship today to Elasticsearch or in the events we'll ship in the future.
2ed5a47
to
9d61b26
Compare
9d61b26
to
b5df646
Compare
Resolver is a map. It shows processes that ran on a computer. The processes are drawn as nodes and lines connect processes with their parents. Resolver is not yet implemented in Kibana. This PR adds a 'map' type UX. The user can click and drag to pan the map and zoom using trackpad pinching (or ctrl and mousewheel.) There is no code providing actual data. Sample data is included. The sample data is used to draw a map. The fundamental info needed is: process names the parent of a process With this info we can topologically lay out the processes. The sample data isn't yet in a realistic format. We'll be fixing that soon. Related issue: elastic/endpoint-app-team#30
💔 Build FailedHistory
To update your PR or re-run it, just comment with: |
* Add Endpoint plugin and Resolver embeddable (#51994) * Add functional tests for plugins to x-pack (so we can do a functional test of the Resolver embeddable) * Add Endpoint plugin * Add Resolver embeddable * Test that Resolver embeddable can be rendered Conflicts: x-pack/.i18nrc.json x-pack/test/api_integration/apis/index.js * [Endpoint] Register endpoint app (#53527) * register app, create functional test * formatting * update tests * adjust test data for endpoint * add endpoint tests for testing spaces, app enabled, disabled, etc * linting * add read privileges to endpoint * rename variable since its used now * remove deprecated context * remove unused variable * fix type check * correct test suite message Co-Authored-By: Larry Gregory <[email protected]> Co-authored-by: Elastic Machine <[email protected]> Co-authored-by: Larry Gregory <[email protected]> * [Endpoint] add react router to endpoint app (#53808) * add react router to endpoint app * linting * linting * linting * correct tests * change history from hash to browser, add new test util * remove default values in helper functions * fix type check, use FunctionComponent as oppsed to FC * use BrowserRouter component * use BrowserRouter component lin * add comments to test framework, change function name to include browserHistory Co-authored-by: Elastic Machine <[email protected]> * EMT-issue-65: add endpoint list api (#53861) add endpoint list api * EMT-65:always return accurate endpoint count (#54423) EMT-65:always return accurate endpoint count, independent of paging properties * Resolver component w/ sample data (#53619) Resolver is a map. It shows processes that ran on a computer. The processes are drawn as nodes and lines connect processes with their parents. Resolver is not yet implemented in Kibana. This PR adds a 'map' type UX. The user can click and drag to pan the map and zoom using trackpad pinching (or ctrl and mousewheel.) There is no code providing actual data. Sample data is included. The sample data is used to draw a map. The fundamental info needed is: process names the parent of a process With this info we can topologically lay out the processes. The sample data isn't yet in a realistic format. We'll be fixing that soon. Related issue: elastic/endpoint-app-team#30 * Resolver test plugin not using mount context. (#54933) Mount context was deprecated. Use core.getStartServices() instead. * Resolver nonlinear zoom (#54936) * [Endpoint] add Redux saga Middleware and app Store (#53906) * Added saga library * Initialize endpoint app redux store * Resolver is overflow: hidden to prevent obscured elements from showing up (#55076) * [Endpoint] Fix saga to start only after store is created and stopped on app unmount (#55245) - added `stop()`/`start()` methods to the Saga Middleware creator factory - adjust tests based on changes - changed application `renderApp` to stop sagas when react app is unmounted * Resolver zoom, pan, and center controls (#55221) * Resolver zoom, pan, and center controls * add tests, fix north panning * fix type issue * update west and east panning to behave like google maps * [Endpoint] FIX: Increase tests `sleep` default duration back to 100ms (#55492) Revert `sleep()` default duration, in the saga tests, back to 100ms in order to prevent intermittent failures during CI runs. Fixes #55464 Fixes #55465 * [Endpoint] EMT-65: make endpoint data types common, restructure (#54772) [Endpoint] EMT-65: make endpoint data types common, use schema changes * Basic Functionality Alert List (#55800) * sets up initial grid and data type * data feeds in from backend but doesnt update * sample data feeding in correctly * Fix combineReducers issue by importing Redux type from 'redux' package * Add usePageId hook that fires action when user navigates to page * Strict typing for middleware * addresses comments and uses better types * move types to common/types.ts * Move types to endpoint/types.ts, address PR comments blah 2 Co-authored-by: Pedro Jaramillo <[email protected]> * [Endpoint] Add Endpoint Details route (#55746) * Add Endpoint Details route * add Endpoint Details tests * sacrifices to the Type gods * update to latest endpoint schema Co-authored-by: Elastic Machine <[email protected]> * [Endpoint] EMT-67: add kql support for endpoint list (#56328) [Endpoint] EMT-67: add kql support for endpoint list * [Endpoint] ERT-82 ERT-83 ERT-84: Alert list API with pagination (#56538) * ERT-82 ERT-83 ERT-84 (partial): Add Alert List API with pagination * Better type safety for alert list API * Add Test to Verify Endpoint App Landing Page (#57129) Conflicts: x-pack/test/functional/page_objects/index.ts * fixes render bug in alert list (#57152) Co-authored-by: Elastic Machine <[email protected]> * Resolver: Animate camera, add sidebar (#55590) This PR adds a sidebar navigation. clicking the icons in the nav will focus the camera on the different nodes. There is an animation effect when the camera moves. Conflicts: yarn.lock * [Endpoint] Task/basic endpoint list (#55623) * Adds host management list to endpoint security plugin Co-authored-by: Elastic Machine <[email protected]> * [Endpoint] Policy List UI route and initial view (#56918) * Initial Policy List view * Add `endpoint/policy` route and displays Policy List * test cases (both unit and functional) Does not yet interact with API (Ingest). * Add ApplicationService app status management (#50223) This was already backported, but changes to endpoint app could not be backported, since endpoint app itself hadn't been backported. Now that the endpoint app is backported, reapply the endpoint specific changes from the original commit. * Implements `getStartServices` on server-side (#55156) This was already backported, but changes to endpoint app could not be backported, since endpoint app itself hadn't been backported. Now that the endpoint app is backported, reapply the endpoint specific changes from the original commit. * [ui/utils/query_string]: Remove unused methods & migrate apps to querystring lib (#56957) This was already backported, but changes to endpoint app could not be backported, since endpoint app itself hadn't been backported. Now that the endpoint app is backported, reapply the endpoint specific changes from the original commit. Co-authored-by: Kevin Logan <[email protected]> Co-authored-by: Elastic Machine <[email protected]> Co-authored-by: Larry Gregory <[email protected]> Co-authored-by: nnamdifrankie <[email protected]> Co-authored-by: Davis Plumlee <[email protected]> Co-authored-by: Paul Tavares <[email protected]> Co-authored-by: Pedro Jaramillo <[email protected]> Co-authored-by: Dan Panzarella <[email protected]> Co-authored-by: Madison Caldwell <[email protected]> Co-authored-by: Charlie Pichette <[email protected]> Co-authored-by: Candace Park <[email protected]> Co-authored-by: Pierre Gayvallet <[email protected]> Co-authored-by: Alexey Antonov <[email protected]>
Summary
Resolver is a map. It shows processes that ran on a computer. The processes are drawn as nodes and lines connect processes with their parents.
Resolver is not yet implemented in Kibana. This PR adds a 'map' type UX. The user can click and drag to pan the map and zoom using trackpad pinching (or ctrl and mousewheel.)
There is no code providing actual data. Sample data is included. The sample data is used to draw a map. The fundamental info needed is:
With this info we can topologically lay out the processes. The sample data isn't yet in a realistic format. We'll be fixing that soon.
Related issue: https://github.com/elastic/endpoint-app-team/issues/30
This does not implement accessibility features. Those will be added: https://github.com/elastic/endpoint-app-team/issues/40 and https://github.com/elastic/endpoint-app-team/issues/34
This embeddable is not released or available by default, so this should be fine for now.
We aren't adding documentation yet as this feature has not been developed enough.
Checklist
Use
strikethroughsto remove checklist items you don't feel are applicable to this PR.Any text added follows EUI's writing guidelines, uses sentence case text and includes i18n supportDocumentation was added for features that require explanation or tutorialsThis was checked for keyboard-only and screenreader accessibilityFor maintainers