Thank you for the pointers. Let me take a look
On Friday, January 5, 2018 at 4:40:53 PM UTC-8, Daniel Smith wrote:
>
> The easy way is for the thing seeing the HW events to write to the
> "status" CR, and your observers watch the CR. But if I understand, you
> don't want to do that (data rate
The easy way is for the thing seeing the HW events to write to the "status"
CR, and your observers watch the CR. But if I understand, you don't want to
do that (data rate too high?). So your other option is to make an
aggregated apiserver.
I think
I'm trying to look at a design where I support a CRD of type "Status".
"Status" is updated based on HW Events. HW Events are known to the
controller side of the code. So, for "current state" I want the GETs to go
to the controller.
It may require a different URI for routing , I'm ok with
On Mon, Dec 25, 2017 at 8:17 PM, Sree Harissh V
wrote:
> Thank you Robert Bailey! I will look in to kubeadm phases. Are there any
> tools like kops, kubeadm, etc to set up a heterogeneous cluster(i.e
> kubelets running on Ubuntu, CoreOS, CentOS etc)?
>
You can definitely
I had similar problem earlier, and resolved it by grant my user account the
ability to create authorization roles with command:
kubectl create clusterrolebinding cluster-admin-binding
--clusterrole=cluster-admin [--user=]
--
You received this message because you are subscribed to the Google
I'm installing and testing using Kubernetes 1.7.5. About an hour ago, I
had a successful yum install of the components. It attempting to do it
again, all of the older packages are missing. Did someone accidentally do
something?
--
You received this message because you are subscribed to the
I think this is exactly the sort of thing that a custom deployment-like
operator is good for. You have particular needs that are not easily
satisfied with existing constructs. CRDs and controllers let you build
this, and figure out how you want it to work.
Later, maybe, you can solicit other
Hi everyone,
I'm trying to implement an operator to manage a Custom Resource Definition
for this proof-of-concept application:
* a user creates an instance of my CRD
* the operator creates a POD for that CRD
* after a specified timeout, both the POD and the CRD disappear
the use case is a
I have a scenario where certain CRD GETs need to be serviced from specific
controllers. I don't want to use data from etcd. Are there any examples I
can refer to?
--
You received this message because you are subscribed to the Google Groups
"Kubernetes user discussion and Q" group.
To
I have created a cluster on dataproc from my account and shared the same with
my team.All my team members are able to see the cluster , but while trying to
connect to the cluster some of the are getting error as "Error occurred while
communicating with the SSH server check the server and
10 matches
Mail list logo