* Run e2e tests in k3s + helm rather than docker
* collect logs from k3s, use head and newer kube version, check for webhook up when starting rancher
* use latest rc chart
* debugging
* use lower kube version
* fixes to brittle tests, log fixes, enable oidc provider
* fix logs
* test tweaks
* use alpha repo, rc in stable is too old. also wait for capi service
* improvements
* debug
* rtryyr
* don't run settings e2e tests - they leave cluster with non-original values
* fix settings tests, fix CRD test (worked by luck before)
* fix fetch of initial settings
* more settings fixes, crd tweak
* update settings
- it uses useDefault as the reset, which was the issue
- leave tests as is (mosty) and just reset them all at the end
* asd
* adsad
* more settings fixes
* remove problem code
* use toms debug image for api failures
* sdfdsf
* Use 'head' chart
* Use audit log
* enable browser logs for beforeAll and afterAll
* debug for checking racnher pod containers, increase audit log level
* don't overwrite logs...
* output improvements, move position of audit log `set`
* use new (docs pending) flag to enable audit log
* extra logging now in rancher/rancher, use head image (should remove a lot of false positives around fleet)
* improve comment
* combine test artifacts
* tweaks, and another data point for 'resolved' tests
* attempt to get previous logs (audit log previous might not exist)
* remove k logs --previous, use logs deployment/name
* re-enable all tests
* tidyng up
* better logs, including crashed containers
* k3s updates... but run in docker
* change lookup from anything containing 1 to `1`
* changes following pr review
* remove un-needed chrome pin
- this commented out section was a c+p from the main e2e section
- the main e2e section comments this out
- both should use a single repo gh action, won't do that now though
* WIP - just testing how are suite behaves with vai on
* Use common method to create query params for steve api requests
* with vai on there's no workload page... so update tests
* flip back hom page urls
* more fixes
* More fixes
* E2eRequestUtils --> RequestUtils, smarter isVaiCacheEnabled
* better method name, fix a11y tests
* disable vai for vai tests
* fetching the setting gives a 404. hardcode for the moment
* give up, make method async
* use pathWithDefaultSteveParams everywhere
* rename grep tag vai to noVai (matching what the tests run)
* fixes
* Allow more time for rancher to restart after toggling a ff
* fix network policies
* Changes following review
* replace pathWithDefaultSteveParams with ?*
* fixes
* Remove temp code that enables FF, should be defunct once enabled by default
* unpin, improve script
* on failure upload rancher logs to job artifacts
* fix
* test specific rancher commit, and rancher debug level
* try head again following overnight fixes
* Wire in VAI_ENABLED env var (to avoid), fixed isVaiCacheEnabled (no value, use default)
* Standardise references to small image, use `k8s.gcr.io/pause` instead of `nginx:alpine`
* test - see if specific workalods are the sticking point
* reduce impact of cronjob
* reduce impact of cronjob - additional
* Improve e2e tests that spam resource creation/delete
- take gaps when creating resources
- wait for ns deletion (containing resources) to complete before continueing)
* even stricker cron schedule
* update three additional places where workloads spam happens
* fix tag of new @vai tests
* use default count, reduce to min. also removed horrible `while`
* temp fail cronjobs to see logs + recordings
* improvements to noVai test cases, move new utils to commands
* fix flaky compliance and kontainer tests
* proper context, fix lint, fix ingress test
* fix lint
* comment out it.skip
* fix lint, fix flaky logging test, test removing wait when spam creating resources
* testing commented out skip
* re-enable wait, given rancher crashes again
* WIP - just testing how are suite behaves with vai on
* Use common method to create query params for steve api requests
* with vai on there's no workload page... so update tests
* flip back hom page urls
* more fixes
* More fixes
* E2eRequestUtils --> RequestUtils, smarter isVaiCacheEnabled
* better method name, fix a11y tests
* disable vai for vai tests
* fetching the setting gives a 404. hardcode for the moment
* give up, make method async
* use pathWithDefaultSteveParams everywhere
* rename grep tag vai to noVai (matching what the tests run)
* fixes
* Allow more time for rancher to restart after toggling a ff
* fix network policies
* Changes following review
* replace pathWithDefaultSteveParams with ?*
* fixes
* Remove temp code that enables FF, should be defunct once enabled by default
* unpin, improve script
* on failure upload rancher logs to job artifacts
* fix
* test specific rancher commit, and rancher debug level
* try head again following overnight fixes
* Wire in VAI_ENABLED env var (to avoid), fixed isVaiCacheEnabled (no value, use default)
* Standardise references to small image, use `k8s.gcr.io/pause` instead of `nginx:alpine`
* test - see if specific workalods are the sticking point
* reduce impact of cronjob
* reduce impact of cronjob - additional
* Improve e2e tests that spam resource creation/delete
- take gaps when creating resources
- wait for ns deletion (containing resources) to complete before continueing)
* even stricker cron schedule
* update three additional places where workloads spam happens
* fix tag of new @vai tests
* use default count, reduce to min. also removed horrible `while`
* temp fail cronjobs to see logs + recordings
* improvements to noVai test cases, move new utils to commands
* fix flaky compliance and kontainer tests
* proper context, fix lint, fix ingress test
* fix lint
* Add support for HTML banners
* Add support for HTML banners
* Updates test so it runs okay after the text banner tests
* Fix for html tests running after others
* Improve e2e test reliability
* Add aria-labels
* Remove close button from banner and move to a page action
* Update home.vue to fix lint
* Update e2e tests and fix lint issue
* Ensure we wait for both PUTs
* Improve comments
* Change test order
* Improve e2e test
* Fix lint issue
* Improve e2e test reliability
* Fix lint
* Fix e2e test so it works for standard user
* cluster list now uses resource table
* New PaginatedResourceTable
* WIP
* add context to pag setting
* tidying up
* Add server-side pagination to cluster explorer events and general events list
* Servier-side pagination for catalog apps, cluster repo and operation
* Servier-side pagination for Service --> HPA, inmgress, services
* Server-side pagination for home page clusters list and side bar clusters
- Functional Changes
- SSP now works after vue3 bump
- Home Page Clusters list now uses server-side pagination
- Side Bar clusters list now uses server-side pagination
- Wire in now supported sorting / filtering by id and name used for table columns
- Allow pagination to be enabled given a specific context
- Call findPage without persisting to store
- New Pagination Tools
- PaginatedResourceTable - Convenience Component, wraps ResourceTable with pagination specific props
- PaginationWrapper - Convenience class to handle requests for resources and updates to them (avoiding store)
- Regressions
- Side Nav menu ready state was `mgmtCluster.isReady && !pCluster?.hasError`, now ???
* fixes after merge. apply PagResTable to node list, vue3 fixes
* - Fix issue where list components containing PaginatedResourceTables would incorrectly fetch all resources anyway
- convert reminaing storage lists
* policy section
* fix async button in manual refresh mode
* fgh
* aaaaa
* Server-side pagination for home page clusters list and side bar clusters
- Functional Changes
- SSP now works after vue3 bump
- Home Page Clusters list now uses server-side pagination
- Side Bar clusters list now uses server-side pagination
- Wire in now supported sorting / filtering by id and name used for table columns
- Allow pagination to be enabled given a specific context
- Call findPage without persisting to store
- New Pagination Tools
- PaginatedResourceTable - Convenience Component, wraps ResourceTable with pagination specific props
- PaginationWrapper - Convenience class to handle requests for resources and updates to them (avoiding store)
- Regressions
- Side Nav menu ready state was `mgmtCluster.isReady && !pCluster?.hasError`, now ???
* Iteration
Note - prov clusters is broken (only fetches local) due to blocking pr. breals
- notPinned list
* Fix dupe inStore
- remove from resource list, put in resource-fetch (used also by pag res table)
* Fix dupe inStore
- remove from resource list, put in resource-fetch (used also by pag res table)
* Two fixes
- changes namespaces kicked of side nav cluster requests (thought pinnedIds changed)
- fix generic lists re-fetching given ns filter changes (they don't have namespaced arg)
* fixes / updates
* testing/tidying
* WIP - Rest of workload lists
* finish off workloads
* tidying up
* fix non-generic list filtering by namespace
* remove comment, backport fix
* test fixes
* E2E: Ensure we wait for cluster entries to exist before clicking on them
* backport fix for local/api filtering
* Remove debug code
* Changes after review
* e2e fixes / debugging
* More e2e fixes
* More e2e fixes
* More e2e fixes
* Fix generic pages that filter on pagination
* Attempt to fix flaky vai test
* Fix after merge from master
* Updates following new indexed files
* Fix lint and test
* Changes given real cluster tests
- general fixes
- correct issue were sorting prov clusters on mgmt cluster props (issue in master as well...)
- bit the bullet, we now don't fetch all mgmt clusters on dashboard visit.
- there could be knock on affects, but we'd need to remove it sometime in 2.11....
* Fix issues with diplaying rke1 data in home page
- includes https://github.com/rancher/dashboard/pull/12881
* Fix unit tests
* remove invalid sort/filter type
* Running through new indexed fields
* fix unit tests
* lint fixes
* Fixes after small review
* lint
* Test for check-plugin-gates pkg build from shell
* e2e fixes
* Testing new indexed fields with dev image
* tidyuing up imports in hope to fix ts error in check-plugins gate
* Fix failing check-plugin-gates
- caused by shell/scripts/test-plugins-build.sh importing list/catalog.cattle.io.clusterrepo.vue
- the component had been updated to a TS component
- check-plugin build outputs TS errors for a component file imports
- vs code shows no errors for imported file
* tidying up a smidge
* Many tweaks
- Remove final todo's
- includes fix for service type clusterip/headless overlap
- Removed ununused ENDPOINT column (note ENDPOINT formatter used in other columns)
- Testing freshly added index fields
* Improve location of new place where componentWillFetch is set
* fix lint, comment updated
* Fix failing e2e tests
- fix cluster dashboard events test
- fix and greatly improve flaky events test
- fix hpa test
- functional fixes
- fix sorting/filtering events by object type
- fix hpa columns
* Improve pod.spec test
* Improve another repo spec test
* vai test improvements
* Various fixes
* pagination change fixes
* pagination change fixes + new settings
* Update settings-utils.ts
* Fixes
* create name fixes
* improve namespace picker test
* Fix user response
* fixes
* fixes
* fixes
* Make up for userpreference lag by waiting for desired value
* Dont include junk revision in mock resources
- causes MASSSIVE socket spam if the rancher revision exceeds it
* update burger menu, make it clear there's pinned and other cluster lists
also comment out .skip, they play HAVOC with test reports
* improve tableRowsPerPageAndNamespaceFilter
* fix project-namespace
* more fixes...
* Revert "Dont include junk revision in mock resources"
This reverts commit 9aaaabb02c.
* real revision fix
* lint fixes are commenting out skip
* use common resource revision
* safer userpref check
* fix standard user tests
* more linting
* because why not
* fixes/improvements
* Add retry to updateNamespaceFilter
* improve CRD list check for count
* after own review
* tie in failure and retry to tableRowsPerPageAndNamespaceFilter
* handle events count less than 500...
* fix merge fail
Issue
- Request counted on a specific value of a specific feature flag
- This was supplied by intercepting the request for all FFs and returning the desired one
- However the FF response supplied missed the `revision`
- The `revision` is used to watch for changes to the resource from a given point in history
- If no revision is supplied rancher will return the current value
- In the FF world this meant the intercepted response was overwritten by the actual value
Solution
- Fetch the rancher specific revision up front and then add to the intercept
* update label on SelectCredential to display combination of name and id + fix label and placeholder on cloudcredential edit view + fix translation
* fix issue with duplicate labels on editing a cluster cloud credential
* address pr comments
* add e2e tests
* add fix for adjacent-overload-signatures ts lint error
* add fix for adjacent-overload-signatures ts lint error
* add fix for adjacent-overload-signatures ts lint error
* change type for body prop
* fix duplication of createRancherResource command and type definition
* fix linting on ts definition for getRancherResource
* address pr comments
* address pr comments
* address pr comments
---------
Co-authored-by: Alexandre Alves <aalves@Alexandres-MacBook-Pro.local>
- banner should be visibile when test starts
- banner is visible if v3 setting is not 'true' and there is a missing helm repo
- previously confirmed the repo is missing, now confirm the v3 setting is correct
- split rancher api commands out from base commands, added get/set commands
yonasberhe23
Richard Cox
Richard Cox
Izaac Zavaleta
Neil MacDougall
Cody Jackson
Yonas Berhe
Francesco Torchia
Richard Cox
Evgeniya Vashkevich
Alexandre Alves
cnotv