* Remove unused StatsCard component
* Create Card and Stats contextual components with styling
* Send endpoint, item, and protocol to Stats as props
* WIP basic plumbing for metrics in Ember
* WIP metrics data source now works for different protocols and produces reasonable mock responses
* WIP sparkline component
* Mostly working metrics and graphs in topology
* Fix date in tooltip to actually be correct
* Clean up console.log
* Add loading frame and create a style sheet for Stats
* Various polish fixes:
- Loading state for graph
- Added fake latency cookie value to test loading
- If metrics provider has no series/stats for the service show something that doesn't look broken
- Graph hover works right to the edge now
- Stats boxes now wrap so they are either shown or not as will fit not cut off
- Graph resizes when browser window size changes
- Some tweaks to number formats and stat metrics to make them more compact/useful
* Thread Protocol through topology model correctly
* Rebuild assetfs
* Fix failing tests and remove stats-card now it's changed and become different
* Fix merge conflict
* Update api doublt
* more merge fixes
* Add data-permission and id attr to Card
* Run JS linter
* Move things around so the tests run with everything available
* Get tests passing:
1. Remove fakeLatency setTimeout (will be replaced with CONSUL_LATENCY
in mocks)
2. Make sure any event handlers are removed
* Make sure the Consul/scripts are available before the app
* Make sure interval gets set if there is no cookie value
* Upgrade mocks so we can use CONSUL_LATENCY
* Fix handling of no series values from Prometheus
* Update assetfs and fix a comment
* Rebase and rebuild assetfs; fix tcp metric series units to be bits not bytes
* Rebuild assetfs
* Hide stats when provider is not configured
Co-authored-by: kenia <keniavalladarez@gmail.com>
Co-authored-by: John Cowen <jcowen@hashicorp.com>
* Add associated services count to terminating and ingress gateways in Service List page
* Update to not return instances for terminating and ingress gateways
* Refactor and color SVG Lines based on intention permissions
* Create Icon component with L7 and Deny icon styling
* Reposition icons on the lines when the lines are redrawn
* Create service/intention-permissions helper
* Use service/intention-permissions helper to return allow or deny lines
* Upgrade consul-api-double to v5.3.5
* Update HasPermission attribute
* Disable source as well as destination on editing
* Various visual/textual amends
* Make errors only appear once you've interacted with a field
* Move tests that involve selecting menus to a create form
* Revert fieldsets and checkboxes
* Create Topology Tab with foundational layout and styling
* Create Toplogy Metrics component with dynamic SVG
* Add ember-render-modifiers addon
* Implement Topology Metrics comp and fix up styling
* Create topology endpoint with tests
* Move arrow drawing to index.js file
* Add topology to show controller
* Fix up conditional wrapper, tabs positioning, links, and styling
* Group upstreams by dc and fix up styling
* Create service/health-percentage helper
* Add health check percentages to upstreams and downstreams
* Basic Layout
* Upgrade @hashicorp/consul-api-double to v5.2.3
* Renamed endpoint to be service-topology
* Refactor styling
* Update to only show Topology tab when Connect is enabled
* Fix bug and changes from review notes
* Remove unused functions that are replaced with SVG markers
* Refactor to resuse svg-curve helper
* Use the render-template helper for the metrics link
* Add topology default null to services show route
* Removed unused ID
* Fix up tests broken by redirect to /topology
* update bindata on ui-v2/ changes
* Revert "Remove GIT_SHA environment variable"
This reverts commit f620f9aefd41362ce76c3a8e0f4addf402ee0ca2.
* Revert "Hardcode in an app version for ember-cli-app-version"
This reverts commit 5ae493d79b34ef2143d78a44c21a63623140bba8.
* revert modtime change in #8712
* Add 'in service mesh ...' to the composite rows in Service List page
* Add associated services count to terminating and ingress gateways in Service List page
* Update mesh icon in icon-defintion files
* add check for bindata_assetfs changes
* Remove GIT_SHA environment variable
* Hardcode in an app version for ember-cli-app-version
* change ' to " for CONSUL_UI_SETTINGS_PLACEHOLDER
Co-authored-by: John Cowen <jcowen@hashicorp.com>
Co-authored-by: hashicorp-ci <hashicorp-ci@users.noreply.github.com>
* ui: Upgrades ember and related to latest 3.20 LTS release
* Delete ember-portal related tests
As we are running inline and block component tests in the same test an
error is raised as we are adding the same named portal twice.
Tests were deleted here as they weren't really testing anything anyway
* Disable ability to select destination if not creating
* Add a LegacyID for intentions that don't have them
* Use `/exact/` endpoint for reading single intentions
* Fix up test for new API interaction
* Upgrade consul-api-double
* Comment out tests using destination for the moment
Also, removed any intention related things from the old page-navigation
tests
* ui: Various CSS amends
1. Split out %icon-definition from %composite-row
2. Add hover effect to node listing
3. Fix up proxy level list components
4. Rename our various pills to use a 100-900 scale
5. Reogranize other icon related things (consul-kind and
consul-external-source)
* Fix up upstream test
During development a HTTP request will pause for 1 minute ONLY when an
?index is set. This gives a realistic emulation of blocking queries.
During testing we can change this latency when we are testing blocking
queries, which we do in numerous places.
A problem can arise during testing on a very slow machine.
If you are not testing blocking queries and therefore not set a latency
to test with, if the machine you are testing on is slow enough a normal
page can assert during a test, yet not tear down before a further
blocking query request is made. This blocking query then uses the default
latency which cause the page to hang for 1 minute, which in turn causes
the test to timeout.
This only seems to happen on a very slow system, but it does potentially
explain why we occasionally see the odd flakey test popping up.