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....

shell/components/nav/TopLevelMenu.vue

@@ -19,9 +19,11 @@ import { TopLevelMenuHelperPagination, TopLevelMenuHelperLegacy } from '@shell/c
 import { debounce } from 'lodash';
 import { sameContents } from '@shell/utils/array';
 
-// TODO: RC test with rke2 cluster
+// TODO: RC test with rke1/2 cluster
 // TODO: RC test vai off (home and side bar)
 // TODO: RC test permissions (user-base, standard-user)
+
+// TODO: RC bug - nav from page 1 (no mgmt clusters) --> page 2 (requires mgmt clusters from page, fetches, however model getters for provider do not update...)
 export default {
   components: {
     BrandImage,
@@ -100,7 +102,7 @@ export default {
 
     // New
     showPinClusters() {
-      return !this.search;
+      return !this.clusterFilter;
     },
 
     // New

shell/models/provisioning.cattle.io.cluster.js

@@ -349,19 +349,11 @@ export default class ProvCluster extends SteveModel {
   }
 
   get mgmtClusterId() {
-    return this.mgmt?.id || this.id?.replace(`${ this.metadata.namespace }/`, '');
+    return this.status?.clusterName;
   }
 
   get mgmt() {
-    const name = this.status?.clusterName;
-
-    if ( !name ) {
-      return null;
-    }
-
-    const out = this.$rootGetters['management/byId'](MANAGEMENT.CLUSTER, name);
-
-    return out;
+    return this.mgmtClusterId ? this.$rootGetters['management/byId'](MANAGEMENT.CLUSTER, this.mgmtClusterId) : null;
   }
 
   get isReady() {

shell/pages/home.vue

@@ -175,21 +175,21 @@ export default defineComponent({
           label:  this.t('tableHeaders.cpu'),
           value:  '',
           name:   'cpu',
-          sort:   ['status.allocatable.cpu'],
+          sort:   false,
           search: false,
         },
         {
           label:  this.t('tableHeaders.memory'),
           value:  '',
           name:   'memory',
-          sort:   ['status.allocatable.memory'],
+          sort:   false,
           search: false,
         },
         {
           label:        this.t('tableHeaders.pods'),
           name:         'pods',
           value:        '',
-          sort:         ['status.allocatable.pods'],
+          sort:         false,
           search:       false,
           formatter:    'PodsUsage',
           delayLoading: true
@@ -296,7 +296,6 @@ export default defineComponent({
       if ( this.canViewMachine ) {
         const opt: ActionFindPageArgs = {
           force,
-          // TODO: RC test with rke2 cluster
           pagination: new FilterArgs({
             filters: PaginationParamFilter.createMultipleFields(page.map((r: any) => new PaginationFilterField({
               field: 'spec.clusterName',
@@ -325,15 +324,17 @@ export default defineComponent({
 
       // We need to fetch node pools and node templates in order to correctly show the provider for RKE1 clusters
       if ( this.canViewMgmtPools && this.canViewMgmtTemplates) {
+        const filters = PaginationParamFilter.createMultipleFields(page
+          .filter((p: any) => p.status?.clusterName)
+          .map((r: any) => new PaginationFilterField({
+            field: 'spec.clusterName',
+            value: r.status?.clusterName
+          })));
+
         const poolOpt: ActionFindPageArgs = {
           force,
           // TODO: RC test with rke2 cluster
-          pagination: new FilterArgs({
-            filters: PaginationParamFilter.createMultipleFields(page.map((r: any) => new PaginationFilterField({
-              field: 'spec.clusterName',
-              value: r.status?.clusterName// TODO: handle empty spec
-            }))),
-          })
+          pagination: new FilterArgs({ filters })
         };
 
         this.$store.dispatch(`management/findPage`, { type: MANAGEMENT.NODE_POOL, opt: poolOpt });
@@ -344,7 +345,7 @@ export default defineComponent({
           pagination: new FilterArgs({
             filters: PaginationParamFilter.createMultipleFields(page.map((r: any) => new PaginationFilterField({
               field: 'spec.clusterName',
-              value: r.status?.clusterName// TODO: handle empty spec
+              value: r.status?.clusterName
             }))),
           })
         };

shell/plugins/dashboard-store/actions.js

@@ -408,6 +408,12 @@ export default {
       return Promise.reject(e);
     }
 
+    // TODO: RC Test
+    await dispatch('unwatch', {
+      type,
+      all: true,
+    });
+
     const pagination = opt.pagination ? {
       request: {
         namespace:  opt.namespaced,

shell/plugins/dashboard-store/mutations.js

@@ -164,6 +164,8 @@ export function load(state, {
     }
   }
 
+  cache.havePage = false;
+
   return entry;
 }
 

shell/plugins/steve/steve-pagination-utils.ts

@@ -149,10 +149,6 @@ class StevePaginationUtils extends NamespaceProjectFilters {
     [CAPI.RANCHER_CLUSTER]: [
       { field: `metadata.labels."${ CAPI_LABELS.PROVIDER }"` },
       { field: `status.provider` },
-      { field: 'status.allocatable.cpu' }, // TODO: RC test with rke2 cluster
-      { field: 'status.allocatable.memory' }, // TODO: RC test with rke2 cluster
-      { field: 'status.allocatable.pods' }, // TODO: RC test with rke2 cluster
-
       { field: 'status.clusterName' },
     ]
   }

shell/plugins/steve/subscribe.js

@@ -452,7 +452,7 @@ const sharedActions = {
   },
 
   unwatch(ctx, {
-    type, id, namespace, selector
+    type, id, namespace, selector, all
   }) {
     const { commit, getters, dispatch } = ctx;
 
@@ -467,16 +467,26 @@ const sharedActions = {
         stop: true, // Stops the watch on a type
       };
 
+      const unwatch = (obj) => {
+        if (getters['watchStarted'](obj)) {
+          // Set that we don't want to watch this type
+          // Otherwise, the dispatch to unwatch below will just cause a re-watch when we
+          // detect the stop message from the backend over the web socket
+          commit('setWatchStopped', obj);
+          dispatch('watch', obj); // Ask the backend to stop watching the type
+          // Make sure anything in the pending queue for the type is removed, since we've now removed the type
+          commit('clearFromQueue', type);
+        }
+      };
+
       if (isAdvancedWorker(ctx)) {
         dispatch('watch', obj); // Ask the backend to stop watching the type
+      } else if (all) {
+        getters['watchesOfType'](type).forEach((obj) => {
+          unwatch(obj);
+        });
       } else if (getters['watchStarted'](obj)) {
-        // Set that we don't want to watch this type
-        // Otherwise, the dispatch to unwatch below will just cause a re-watch when we
-        // detect the stop message from the backend over the web socket
-        commit('setWatchStopped', obj);
-        dispatch('watch', obj); // Ask the backend to stop watching the type
-        // Make sure anything in the pending queue for the type is removed, since we've now removed the type
-        commit('clearFromQueue', type);
+        unwatch(obj);
       }
     }
   },
@@ -1025,6 +1035,10 @@ const defaultGetters = {
     return state.inError[keyForSubscribe(obj)];
   },
 
+  watchesOfType: (state) => (type) => {
+    return state.started.filter((entry) => type === (entry.resourceType || entry.type));
+  },
+
   watchStarted: (state) => (obj) => {
     return !!state.started.find((entry) => equivalentWatch(obj, entry));
   },

shell/store/index.js

@@ -38,6 +38,7 @@ import semver from 'semver';
 import { STORE, BLANK_CLUSTER } from '@shell/store/store-types';
 import { isDevBuild } from '@shell/utils/version';
 import { markRaw } from 'vue';
+import paginationUtils from '@shell/utils/pagination-utils';
 
 // Disables strict mode for all store instances to prevent warning about changing state outside of mutations
 // because it's more efficient to do that sometimes.
@@ -780,15 +781,11 @@ export const actions = {
     // The alternative is simpler (fetch features up front) but would add another blocking request in
 
     const promises = {
-      // Clusters guaranteed always available or your money back
-      clusters: dispatch('management/findAll', { type: MANAGEMENT.CLUSTER, opt: { watch: false } }),
-
       // Features checks on its own if they are available
       features: dispatch('features/loadServer'),
     };
 
     const toWatch = [
-      MANAGEMENT.CLUSTER,
       MANAGEMENT.FEATURE,
     ];
 
@@ -823,6 +820,14 @@ export const actions = {
 
     res = await allHash(promises);
 
+    // TODO: RC test
+    if (!res.settings && !paginationUtils.isEnabled({ rootGetters }, { store: 'management', resource: { id: MANAGEMENT.CLUSTER, context: 'side-bar' } })) {
+      // This introduces a synchronous request, however we need settings to determine if pagination is enabled
+      // Eventually it will be removed when pagination is always on
+      res.clusters = await dispatch('management/findAll', { type: MANAGEMENT.CLUSTER, opt: { watch: false } });
+      toWatch.push(MANAGEMENT.CLUSTER);
+    }
+
     // See comment above. Now that we have feature flags we can watch resources
     toWatch.forEach((type) => {
       dispatch('management/watch', { type });
@@ -879,7 +884,7 @@ export const actions = {
   // - state.clusterId is the old cluster id (or undefined)
   // - id is the new cluster id (or undefined)
   async loadCluster({
-    state, commit, dispatch, getters
+    state, commit, dispatch, getters, rootGetters
   }, {
     id, product, oldProduct, oldPkg, newPkg, targetRoute
   }) {
@@ -983,9 +988,10 @@ export const actions = {
     // Try and wait until the schema exists before proceeding
     await dispatch('management/waitForSchema', { type: MANAGEMENT.CLUSTER });
 
-    // Similar to above, we're still waiting on loadManagement to fetch required resources
-    // If we don't have all mgmt clusters yet a request to fetch this cluster and then all clusters (in cleanNamespaces) is kicked off
-    await dispatch('management/waitForHaveAll', { type: MANAGEMENT.CLUSTER });
+    // Do we need to wait on loadManagement to fetch required resources?
+    if (!paginationUtils.isEnabled({ rootGetters }, { store: 'management', resource: { id: MANAGEMENT.CLUSTER, context: 'side-bar' } })) {
+      await dispatch('management/waitForHaveAll', { type: MANAGEMENT.CLUSTER });
+    }
 
     // See if it really exists
     try {
@@ -1083,6 +1089,17 @@ export const actions = {
   },
 
   async cleanNamespaces({ getters, dispatch }) {
+    if (paginationUtils.isEnabled({ rootGetters }, { store: 'management', resource: { id: MANAGEMENT.CLUSTER, context: 'side-bar' } })) {
+      // See https://github.com/rancher/dashboard/issues/12864
+      // old world..
+      // - loadManagement makes a request to fetch all mgmt clusters
+      // - we would block on that that request above before getting here (otherwise x2 requests were made)
+      // new world..
+      // - we won't have all mgmt clusters, this is another place that needs updating (see issue)
+
+      return;
+    }
+
     // Initialise / Remove any filters that the user no-longer has access to
     await dispatch('management/findAll', { type: MANAGEMENT.CLUSTER }); // So they can be got byId below