Do you need the id to be on the same file? Can't that be in another file?
If it can be in another file, and not in the configmap, I think it will be
way more easy (and easier to share a configmap, etc.)
Because, maybe if you use a stateful set and a init container, you can just
place it in an
Publish the base configmap into a volume. Run a sidecar container
that consumes that and merges it with the unique ID (which you get
from StatefulSet or somewhere else - different problem) and publish
THAT to a shared emptyDir volume. When the base configmap changes,
the sidecar wakes up (e.g.
Hello All,
I have a rather specific usecase and was looking for some suggestions on how to
tackle it.
We have a java application that reads a configuration file on start up. We can
run multiple copies of this application for horizontal scaling however it
requires we give each a unique ID
I'm sure a "me too" won't help much in terms of speeding up the development of
this feature on GKE, but in case it _does_ help to communicate the level of
importance of this feature on GKE, here's my/our me too as well :)
On Tuesday, May 23, 2017 at 10:41:59 PM UTC+2, Tim Hockin wrote:
> We're
My team recently updated our GKE cluster to 1.6.8 from 1.5.6. We discovered an
image was not properly running. After digging into the issue we found that file
paths in the Docker image greater than 100 chars are not extracted into the
container file system. This led us to finding out this is a
1 master, 3 nodes.
3 interfaces each.
what is the most common setup?
On Sat, Aug 19, 2017 at 2:31 AM, RĂ©mon Sinnema wrote:
> How did you set up your cluster? The LoadBalancer type is not supported
> everywhere, e.g. when using kubeadm.
>
> On Wednesday, August 16, 2017 at