Hey,
During pruning we assume all images are stored in the internal registry and
that works in 99% of cases [1]. It looks like you've changed the registry
somewhere in the middle, and that's why in your case this is not working
as expected (iow. at least one image pointing to the old registry IP).
Can you open an issue with the description of your situation and
ping me there (soltysh on gh). For now I can recommend using --registry-url
flag which will tell the pruning mechanism which IP it should use instead of
the autodiscovery mechanism.

Maciej


[1]
https://github.com/openshift/origin/blob/release-1.5/pkg/image/prune/prune.go#L794-L813

On Sun, Mar 19, 2017 at 5:51 PM, Dean Peterson <[email protected]>
wrote:

> [root@openshift ~]# oc project default
> Now using project "default" on server "https://openshift.abecorn.com:8443
> ".
> [root@openshift ~]# oc get services
> NAME              CLUSTER_IP      EXTERNAL_IP   PORT(S)
> SELECTOR                  AGE
> docker-registry   172.30.219.25   <none>        5000/TCP
>  docker-registry=default   216d
> kubernetes        172.30.0.1      <none>        443/TCP,53/UDP,53/TCP
> <none>                    1y
> router            172.30.51.78    <none>        80/TCP
>  router=router             1y
> [root@openshift ~]# docker login -u [email protected] -e [email protected]
> -p Sx8h7d5CVjFtvCb4Oz2Uhg-hnFBiF7_RHSoG2IP05-E 172.30.219.25:5000
> WARNING: login credentials saved in /root/.docker/config.json
> Login Succeeded
> [root@openshift ~]# oadm prune images --keep-tag-revisions=3
> --keep-younger-than=60m --confirm
> error: error communicating with registry: Get http://172.30.105.244:5000/
> healthz: dial tcp 172.30.105.244:5000: no route to host
>
>
> _______________________________________________
> users mailing list
> [email protected]
> http://lists.openshift.redhat.com/openshiftmm/listinfo/users
>
>
_______________________________________________
users mailing list
[email protected]
http://lists.openshift.redhat.com/openshiftmm/listinfo/users

Reply via email to