kops reconcile cluster fails if kops export kubecfg hasn't been run first #17146
Description
/kind bug
1. What kops version are you running? The command kops version, will display
this information.
1.31.0-beta.1
2. What Kubernetes version are you running? kubectl version will print the
version if a cluster is running or provide the Kubernetes version specified as
a kops flag.
1.30.8
3. What cloud provider are you using?
AWS
4. What commands did you run? What is the simplest way to reproduce this issue?
kops reconcile cluster --yes
5. What happened after the commands executed?
Updating control plane configuration
W1217 16:09:34.669456 2974 update_cluster.go:362] error checking control plane running version, assuming no k8s upgrade in progress: cannot load kubecfg settings for "my.cluster.com": context "my.cluster.com" does not exist
W1217 16:09:55.576232 2974 pruning.go:115] manifest includes an object of GroupKind Secret, which will not be pruned
I1217 16:09:55.669254 2974 issuerdiscovery.go:101] serviceAccountIssuers bucket "my-discovery-store" is not public; will use object ACL
I1217 16:09:59.567931 2974 executor.go:113] Tasks: 0 done / 135 total; 66 can run
I1217 16:10:00.038363 2974 executor.go:113] Tasks: 66 done / 135 total; 32 can run
I1217 16:10:00.413219 2974 default_methods.go:122] not deleting SecurityGroupRule/sg-xxx: port=-1 protocol=4 group=sg-xxx ip= ipv6= because it is marked for deferred-deletion
I1217 16:10:01.366491 2974 executor.go:113] Tasks: 98 done / 135 total; 25 can run
I1217 16:10:02.104563 2974 executor.go:113] Tasks: 123 done / 135 total; 3 can run
I1217 16:10:02.299848 2974 executor.go:113] Tasks: 126 done / 135 total; 6 can run
I1217 16:10:02.414822 2974 executor.go:113] Tasks: 132 done / 135 total; 3 can run
I1217 16:10:02.527061 2974 executor.go:113] Tasks: 135 done / 135 total; 0 can run
I1217 16:10:07.750745 2974 dns.go:235] Pre-creating DNS records
Cluster changes have been applied to the cloud.
Changes may require instances to restart: kops rolling-update cluster
Doing rolling-update for control plane
Error: cannot load kubecfg settings for "my.cluster.com": context "my.cluster.com" does not exist
6. What did you expect to happen?
I thought that kops reconcile cluster would generate the kubecfg itself and complete successfully, or provide an --admin flag to generate the kubecfg as part of its operation.
7. Please provide your cluster manifest. Execute
kops get --name my.example.com -o yaml to display your cluster manifest.
You may want to remove your cluster name and other sensitive information.
N/A
8. Please run the commands with most verbose logging by adding the -v 10 flag.
Paste the logs into this report, or in a gist and provide the gist link here.
See above.
9. Anything else we need to know?
Running kops export kubecfg --admin before kops reconcile cluster --yes works around the issue. But the whole idea of kops reconcile cluster was to have a single command to update the cluster. 😉