With Daemonset now everything is working properly.
Anyway I'm using Kubernetes 1.8.1-gke.1
--
You received this message because you are subscribed to the Google Groups
"Kubernetes user discussion and Q&A" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to
Oh, I thought you wanted on different hosts but not as many pods as hosts.
If you want that's daemonset guarantees that (even if more nodes are
created later, etc.)
And what Kubernetes version are you using? There is some kind of support to
upgrade them in recent versions IIRC (not used that more
Today during a deploy I get a pod with 2 containers -,-
I can confirm that the best solution to make sure you have only one pod per
node is using the DaemonSet.
Unfortunately using the approach to reapply the deployment yaml does not
guarantee that after deployment each node has only a single po
Cool. Take into account that daemon set it is created to guarantee having
exactly one pod per node. For example, if you had more nodes, more pods for
a daemonset will be added. And the same if some crash or you reduce.
If that fits better what you want (sorry I didn't understood before), then
don'
As I said before, using multiple times the command "kubectl apply -f
my-deployment.yaml" (changing from time to time the image version inside the
yaml) I noticed that Kubernetes never deploys 2 pod in a same node.
I tested this behavior many times so yes it's working as I need :)
If I had proble
It's working as you need? :)
On Monday, December 4, 2017, wrote:
> thank you all for the support ;-)
>
> --
> You received this message because you are subscribed to the Google Groups
> "Kubernetes user discussion and Q&A" group.
> To unsubscribe from this group and stop receiving emails from it
thank you all for the support ;-)
--
You received this message because you are subscribed to the Google Groups
"Kubernetes user discussion and Q&A" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to kubernetes-users+unsubscr...@googlegroups.com.
To post to
If your only requirements is to have one pod per node, then I think the
best solution, as Tim suggested, is a Daemon Set (
https://kubernetes.io/docs/concepts/workloads/controllers/daemonset/).
And yes, it's perfectly reasonable to edit and reapply YAML's.
On Mon, Dec 4, 2017, 18:45 wrote:
> So
So re-apply the deployment.yaml is an acceptable solution considering that my
only requirement is have one pod for node?
Unfortunately I have a very close due date so I would like to find the
faster-simpler and quite stable solution to do a code upgrade :)
--
You received this message becau
The scheduler makes the decision trying to spread the pods on nodes as you
say. But that is just a "signal", other things are taken into account (pods
availability zone, in case of AWS, for example, to spread across AZs too)
node's resources, etc.
So, the default will try to do that, taking into a
I tried some solutions and one that is working at the moment is simply based to
change my deployment.yaml every time.
I mean:
1) I have to deploy for the 1' time my application based on a pod with some
containers. These pods should be deployed on every cluster node (I have 3
nodes). I did the d
Would you prefer a Daemon set instead?
On Dec 4, 2017 7:28 AM, "Itamar O" wrote:
> I'm guessing you have as many replicas as you have nodes, and you used the
> "required" affinity policy over the "preferred" one.
> If this is the case, then when you try to update the deployment (with the
> defau
I'm guessing you have as many replicas as you have nodes, and you used the
"required" affinity policy over the "preferred" one.
If this is the case, then when you try to update the deployment (with the
default upgrade strategy), the controller tries the schedule a *4th pod*
(with the new image) bef
Sorry but now I'm facing another problem :-(
The deployment with the options podAntiAffinity/podAffinity is working but when
I try to update the deployment with the command:
kubectl set image deployment/apache-deployment apache-container=xx:v2.1.2
then I get this error:
apache-deployment-7c
At
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/#affinity-and-anti-affinity
there is an example (perfect for my case) where 2 files yaml are used: one for
redis-cache and the other for web-store.
Anyway I'll try to concatenate them.
Thanks
--
You received this message beca
Not sure where you have 2 YAML's (since you specify the afiinity under the
deployment template spec), but if you end up with the need for multiple
YAML's you can always concatenate them to a single file by separating with
"---".
On Mon, Dec 4, 2017 at 1:41 PM wrote:
> I'm reading the documentati
I'm reading the documentation and it's just what I was looking for.
Many thanks!
But is there a way to create a single yaml deployment file to ensure that every
pod will be deployed in a single node?
So a single file to be executed and not 2 different yaml files as in the example
--
You receiv
I think that if you create a service that matches the pods of the
deployment, K8s will attempt to spread out the pods by default.
Also, check out "pod anti-affinity" (
https://kubernetes.io/docs/concepts/configuration/assign-pod-node/#inter-pod-affinity-and-anti-affinity-beta-feature
).
On Mon, De
Hi all!
I would like to know if there is a way to force Kubernetes, during a deploy, to
use every node in the cluster.
The question is due some attempts that I have done where I noticed a situation
like this:
- a cluster of 3 nodes
- I update a deployment with a command like: kubectl set image
19 matches
Mail list logo
