johngmyers
commented
4 years ago /area provider/openstack
Closed chuonglh closed 4 years ago
johngmyers
commented
4 years ago /area provider/openstack
olemarkus
commented
4 years ago Could you paste your cluster spec?
chuonglh
commented
4 years ago Yes, here is our cluster spec.
apiVersion: kops.k8s.io/v1alpha2
kind: Cluster
metadata:
creationTimestamp: null
generation: 9
name: hcm.k8s.local
spec:
api:
loadBalancer:
type: ' '
authorization:
rbac: {}
channel: stable
cloudConfig:
openstack:
blockStorage:
ignore-volume-az: true
override-volume-az: nova
monitor:
delay: 1m
maxRetries: 3
timeout: 30s
router:
externalNetwork: public
cloudProvider: openstack
configBase: swift://kops-openstack/hcm.k8s.local
etcdClusters:
- cpuRequest: 300m
etcdMembers:
- instanceGroup: master-nova-1
name: "1"
volumeSize: 2
volumeType: __DEFAULT__
- instanceGroup: master-nova-2
name: "2"
volumeSize: 2
volumeType: __DEFAULT__
- instanceGroup: master-nova-3
name: "3"
volumeSize: 2
volumeType: __DEFAULT__
memoryRequest: 200Mi
name: main
version: 3.4.3
- cpuRequest: 300m
etcdMembers:
- instanceGroup: master-nova-1
name: "1"
volumeSize: 2
volumeType: __DEFAULT__
- instanceGroup: master-nova-2
name: "2"
volumeSize: 2
volumeType: __DEFAULT__
- instanceGroup: master-nova-3
name: "3"
volumeSize: 2
volumeType: __DEFAULT__
memoryRequest: 200Mi
name: events
version: 3.4.3
iam:
allowContainerRegistry: true
legacy: false
kubeAPIServer:
admissionControl:
- NamespaceLifecycle
- LimitRanger
- ServiceAccount
- DefaultStorageClass
- DefaultTolerationSeconds
- MutatingAdmissionWebhook
- ValidatingAdmissionWebhook
- NodeRestriction
- ResourceQuota
- Priority
auditLogMaxAge: 10
auditLogMaxBackups: 1
auditLogMaxSize: 100
auditLogPath: /var/log/kube-apiserver-audit.log
kubeControllerManager:
attachDetachReconcileSyncPeriod: 3m0s
kubeDNS:
provider: CoreDNS
kubeProxy:
metricsBindAddress: 0.0.0.0
kubelet:
anonymousAuth: false
authenticationTokenWebhook: true
authorizationMode: Webhook
featureGates:
PodPriority: "true"
imagePullProgressDeadline: 5m0s
kubernetesApiAccess:
- 0.0.0.0/0
kubernetesVersion: 1.18.6
masterInternalName: api.internal.hcm.k8s.local
masterPublicName: api.hcm.k8s.local
networkCIDR: 172.16.0.0/16
networking:
calico:
ipipMode: Always
majorVersion: v3
mtu: 1390
nonMasqueradeCIDR: 100.64.0.0/10
sshAccess:
- 0.0.0.0/0
subnets:
- cidr: 172.16.32.0/19
name: nova
type: Private
zone: nova
- cidr: 172.16.0.0/22
name: utility-nova
type: Utility
zone: nova
topology:
dns:
type: Public
masters: private
nodes: private
zetaab
commented
4 years ago the row which is causing the issue is https://github.com/kubernetes/kops/blob/release-1.18/pkg/model/openstackmodel/firewall.go#L484
What is this loadbalancer type?
api:
loadBalancer:
type: ' 'So do you have loadbalancer in your setup or not? The error message is saying that you do not have loadbalancer in your region (or at least api does not exist). If you do not have then you should remove the entire api structure from the config or use:
api: {}This function should return false in your setup https://github.com/kubernetes/kops/blob/master/pkg/model/context.go#L298-L304
chuonglh
commented
4 years ago Hi @zetaab, Currently, I don't have a loadbalancer. I follow this instruction here to not using loadbalancer https://kops.sigs.k8s.io/getting_started/openstack/#using-openstack-without-lbaas.
I tried remove api structure or set like you guide above but not success.
# Found fields that are not recognized
# ...
# name: hcm.k8s.local
# spec:
# - api:
# - loadBalancer:
# - type: Public
# authorization:
# rbac: {}
# ...
#
## Found fields that are not recognized
# ...
# name: hcm.k8s.local
# spec:
# - api:
# - loadBalancer:
# + api: {}
# - type: Public
# authorization:
# rbac: {}
# ...
#
#This has been fixed in master (9b0d235554a) together with a few other related issues. Are you able to test kops master version?
chuonglh
commented
4 years ago Hi @olemarkus,
I have tested with kops master version and it works. So, when we have a new release for it?
olemarkus
commented
4 years ago Good to know, thanks. There will probably be another pre-release soon. Stable release sometime in late september/early october if all goes well.
If all is good, could you close the issue?
chuonglh
commented
4 years ago Yes, sure. Thanks, I really appreciate it!
Hi, When we performed the kops update cluster and then got that issue.
1. What
kopsversion are you running? The commandkops version, will display this information.1.18.03. What cloud provider are you using?Openstack4. What commands did you run? What is the simplest way to reproduce this issue?kops update cluster5. What happened after the commands executed?