In my limited experimentation I had problems with NFS PVs getting wiped out
even though the policy was set to Retain. In fact I ended up in this
situation where if I created a file in the NFS volume it was deleted in
front of my eyes in a few seconds. Obviously I did something very wrong
with the PV.  I've found mounting the NFS volume directly into the pod as
Seth suggested a lot easier.  I also was unable to force a particular PV to
be bound to a PVC. They seem to work like a pool and you get what you get.
So if you have an existing nfs volume with data you want to mount into a
particular pod there might not be a way of doing that with PV's but I would
love to be proven wrong by others.

On 14 October 2016 at 07:32, Seth Jennings <> wrote:

> NFS mounts can be mounted directly into pods without being PVs like this:
> volumes:
>   name: shared
>   nfs:
>     server: <server>
>     path: <path>
> If you are using NFS PVs, then the persistentVolumeReclaimPolicy
> determines if the data is wiped when the PVC is released.  The default
> value is "Retain".  It will not delete the data unless you set it to
> "Recycle".
> persistent_storage_nfs.html#reclaiming-resources
> Hope that answers your question!
> On Thu, Oct 13, 2016 at 10:19 AM, Dean Peterson <>
> wrote:
> > If I create a persistent volume claim using an NFS share that has
> existing
> > data, will the data be wipded? Same thing with creating the persistent
> > volume. Will the existing data be deleted. I want to make existing data
> > accessible to multiple pods/containers in an NFS share. If I make a
> > persistent volume pointing to that existing path. How do I get the
> > persistent volume claim to access that existing path and make the
> containers
> > with the claim see it?
> >
> > _______________________________________________
> > users mailing list
> >
> >
> >
> _______________________________________________
> users mailing list
users mailing list

Reply via email to