fix tests - adding missing role, increase timeout

[nammn]

Summary:

from test

    	            	extra elements in list A:
    	            	([]interface {}) (len=1) {
    	            	 (automationconfig.CustomRole) {
    	            	  Role: (string) (len=30) "MongodbAutomationAgentUserRole",
    	            	  DB: (string) (len=5) "admin",
    	            	  Privileges: ([]automationconfig.Privilege) (len=1) {
    	            	   (automationconfig.Privilege) {
    	            	    Resource: (automationconfig.Resource) {
    	            	     DB: (*string)(<nil>),
    	            	     Collection: (*string)(<nil>),
    	            	     AnyResource: (bool) true,
    	            	     Cluster: (bool) false
    	            	    },
    	            	    Actions: ([]string) (len=1) {
    	            	     (string) (len=22) "bypassDefaultMaxTimeMS"
    	            	    }
    	            	   }
    	            	  },
    	            	  Roles: ([]automationconfig.Role) {
    	            	  },
    	            	  AuthenticationRestrictions: ([]automationconfig.AuthenticationRestriction) <nil>
    	            	 }
    	            	}

https://github.com/mongodb/mongodb-kubernetes-operator/actions/runs/12671524133/job/35384225989#step:13:314

its passing now: https://github.com/mongodb/mongodb-kubernetes-operator/actions/runs/12707699284/job/35423341022?pr=1655

[nammn]

looking at the recent test failures, it was due to the fact that the deployment needed a bit more than 60 seconds to stablize

2m2s        Normal    Created                   pod/mongodb-kubernetes-operator-6c5f7d74b9-6qb6m    spec.containers{mongodb-kubernetes-operator}   kubelet, kind-control-plane      Created container mongodb-kubernetes-operator                                                                                       2m2s         1       mongodb-kubernetes-operator-6c5f7d74b9-6qb6m.18194a55944214a5
61s         Warning   FailedScheduling          pod/mongodb-kubernetes-operator-57d64f66d7-9dmm5                                                   default-scheduler                0/1 nodes are available: 1 node(s) didn't match pod affinity/anti-affinity rules, 1 node(s) didn't match pod anti-affinity rules.   2m2s         3       mongodb-kubernetes-operator-57d64f66d7-9dmm5.18194a559645561a
45s         Normal    Killing                   pod/mongodb-kubernetes-operator-6c5f7d74b9-6qb6m    spec.containers{mongodb-kubernetes-operator}   kubelet, kind-control-plane      Stopping container mongodb-kubernetes-operator                                                                                      45s          1       mongodb-kubernetes-operator-6c5f7d74b9-6qb6m.18194a6763ffc4de
14s         Normal    Scheduled                 pod/mongodb-kubernetes-operator-57d64f66d7-9dmm5                                                   default-scheduler                Successfully assigned default/mongodb-kubernetes-operator-57d64f66d7-9dmm5 to kind-control-plane                                    14s          1       mongodb-kubernetes-operator-57d64f66d7-9dmm5.18194a6eb254d17d
13s         Normal    Started 

[lucian-tosa]

Correct me if I'm wrong, but we added this because the automation config sometimes had this role without us specifying it?

[nammn]

No - I think this test consistently fails IIUC. It seems the last time the ci was fully green was 4 months ago: https://github.com/mongodb/mongodb-kubernetes-operator/actions/workflows/e2e.yml?query=is%3Asuccess.

It would be nice to see single test history ...

Alternatively, we can cut that agent out from the received list.

[lucian-tosa]

Alright, thanks. Feel free to resolve

[lucian-tosa]

LGTM