Yes, I've seen this error sometimes, mainly during the Scaleup process.
What I did that apparently solve this issue is to remove the /etc/cni
directory, and let the installation/scaleup process create it, but I don't
know the root cause either.
As you said, it happens randomly and don't seem to have a pattern. The
first time I faced it, I was scaling a cluster and adding four new Nodes,
and just one presented the error, the other three were added to the cluster
with no errors.
Cloud Consultant, RHCVA, RHCE
Red Hat Brasil <https://www.redhat.com>
rbe...@redhat.com M: +55-11-99557-5841
TRIED. TESTED. TRUSTED. <https://redhat.com/trusted>
Red Hat é reconhecida entre as melhores empresas para trabalhar no Brasil
pelo *Great Place to Work*.
On Fri, Apr 13, 2018 at 10:10 AM, Tim Dudgeon <tdudgeon...@gmail.com> wrote:
> We've long been encountering a seemingly random problem installing Origin
> 3.7 on Centos nodes.
> This is manifested in the /etc/cni/net.d/ directory on the node being
> empty (it should contain one file named 80-openshift-sdn.conf) and that
> prevents the origin-node service from starting, with the key error in the
> logs (using journalctl) being something like this:
> Apr 13 12:23:44 ip-10-0-0-61.eu-central-1.compute.internal
> origin-node: W0413 12:23:44.933963 26683 cni.go:189] Unable to
> update cni config: No networks found in /etc/cni/net.d
> Something is preventing the ansible installer from creating this file on
> the nodes (though the real cause maybe upstream of this).
> This seems to happen randomly, and with differing frequencies on different
> environments. One one environement abotu 50% of the nodes fail in this way.
> On others its much less frequent. We thought this was a problem with our
> OpenStack environment but we have now also seen this on AWS so it looks
> like its a OpenShift specific problem.
> Has anyone else seen this or know what causes it?
> It's been a really big impediment to rolling out a cluster.
> users mailing list
users mailing list