https://pulumi.com logo
Join Slack
Powered by
Hello all! Need some help, we have an issue with c...
# general
a
Hello all! Need some help, we have an issue with creating multiple resources at the same time on OpenShift with Pulumi with the error:
Copy code
error reading from server: EOF
here's a more detailed output:
Copy code
+  kubernetes:<http://rbac.authorization.k8s.io/v1:Role|rbac.authorization.k8s.io/v1:Role> snig-besu-001-92c8-firefly **creating failed** error: error reading from server: EOF
    pulumi:pulumi:Stack bpaas-snig-besu-001-92c8 running error: update failed
    pulumi:pulumi:Stack bpaas-snig-besu-001-92c8 **failed** 1 error
Diagnostics:
  kubernetes:core/v1:Secret (snig-besu-001-92c8-auth):
    error: error reading from server: EOF
  kubernetes:<http://rbac.authorization.k8s.io/v1:Role|rbac.authorization.k8s.io/v1:Role> (snig-besu-001-92c8):
    error: error reading from server: EOF
  kubernetes:core/v1:ServiceAccount (snig-besu-001-92c8-firefly):
    error: error reading from server: EOF
  kubernetes:core/v1:Secret (snig-besu-001-92c8-firefly-auth):
    error: error reading from server: EOF
  kubernetes:<http://rbac.authorization.k8s.io/v1:Role|rbac.authorization.k8s.io/v1:Role> (snig-besu-001-92c8-firefly):
    error: error reading from server: EOF
  kubernetes:core/v1:ConfigMap (snig-besu-001-92c8):
    error: error reading from server: EOF
  kubernetes:core/v1:ConfigMap (snig-besu-001-92c8-auth):
    error: error reading from server: EOF
  kubernetes:core/v1:PersistentVolumeClaim (snig-besu-001-92c8):
    error: error reading from server: EOF
  kubernetes:core/v1:ConfigMap (snig-besu-001-92c8-firefly):
    error: error reading from server: EOF
  kubernetes:core/v1:PersistentVolumeClaim (snig-besu-001-92c8-firefly):
    error: error reading from server: EOF
  pulumi:pulumi:Stack (bpaas-snig-besu-001-92c8):
    error: update failed
  kubernetes:core/v1:ServiceAccount (snig-besu-001-92c8):
    error: error reading from server: EOF
  kubernetes:core/v1:ConfigMap (snig-besu-001-92c8-firefly-auth):
    error: error reading from server: EOF
  kubernetes:core/v1:Secret (snig-besu-001-92c8-firefly):
    error: error reading from server: EOF
  kubernetes:core/v1:Secret (snig-besu-001-92c8):
    error: error reading from server: EOF
• The resources are created successfully when I create them one at a time, but when I create 3+ resources at the same time, the resource creation errors out with EOF. • The same setup used to work 2 weeks ago with creating 3+ resources at the same time, this is a new issue • I see some logs like this:
Copy code
GET:<https://172.21.0.1:443/apis/apps/v1/namespaces/beta/deployments?limit=500&resourceVersion=0>
Copy code
I0306 10:57:51.036174    2119 request.go:682] Waited for 1.081675482s due to client-side throttling, not priority and fairness, request: PATCH:<https://172.21.0.1:443/apis/rbac.authorization.k8s.io/v1/namespaces/beta/roles/snig-besu-node-11-0e68?fieldManager=pulumi-kubernetes-83f162bd&fieldValidation=Warn&force=true>
Here are my versions:
Copy code
"@pulumi/aws": "5.30.0",
    "@pulumi/awsx": "1.0.2",
    "@pulumi/azure": "5.35.0",
    "@pulumi/cloudflare": "4.15.0",
    "@pulumi/eks": "1.0.1",
    "@pulumi/gcp": "6.50.0",
    "@pulumi/keycloak": "5.0.0",
    "@pulumi/kubernetes": "3.24.1",
    "@pulumi/kubernetesx": "0.1.6",
    "@pulumi/postgresql": "3.6.0",
    "@pulumi/pulumi": "3.55.0",
    "@pulumi/random": "4.11.2",
    "@pulumi/vault": "5.8.0",
Any ideas? If any more information is needed I'd be happy to add it
Copy code
GET:<https://172.21.0.1:443/apis/apps/v1/namespaces/beta/deployments?limit=500&resourceVersion=0>
i want to try out creating new resources with a higher value for the 
limit
param here, how can i change it?
e
Does this work if you set 
--parallel=1
?
a
will try that out and let you know! thanks
cc @narrow-judge-54785
n
Even if the 
--paralllel=1
would be a workaround, and it think there's a good chance. Did u see this error anywhere else before? ideally we would fix it and be able to do parallel deploys.
e
I haven't. But might be worth raising at https://github.com/pulumi/pulumi-kubernetes k8s isn't something I work with much
g
I’m not sure why you’d be getting EOF errors like that, but you can fix the client-side throttling warning by adjusting the kube client settings. I have the following set in my local environment, and you can also set those values directly on a Provider resource:
Copy code
PULUMI_K8S_CLIENT_BURST=120
PULUMI_K8S_CLIENT_QPS=50
Open in Slack
PreviousNext
Powered by