[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

RE: Back-off pulling image "/origin-logging-curator sha256:b89cbdfc4e0e7d594f7a49c7581ae3f75b9d0313fce2ed8be83ee5c0426af72d"

Thanks. How can I handle this when I'm using my own images?

maybe a more clear explanation:
We are using our own docker registry which is secured with a selfsigned certificate. So if we place the cert on our openshift node we're able to pull. We pulled the openshift-origin images v1.2.0 from dockerhub and pushed it inside our docker registry. We are using the registry instead of docker.io/openshift/origin-xxx
This works fine for our router, registry, cluster metrics project etc..
But when we are deploying the logging project: https://docs.openshift.org/latest/install_config/aggregate_logging.html it doesn't work.
the pull of the registry.com/asco/origin-logging-deployment:v1.2.0 is fine and it deploys.
But the problem raises later. The fluentd image is also pulled fine (from our registry).
But the images of the rest aren't pulled in the right way,

example of error events/logs: (origin-logging-"empty" is probably an issue?)

pulling image "registry.com/asco/origin-logging-auth-proxy sha256:179b84eb803fac116f913182c2fb64a2e7adf01dd04fc58e1336d96ce0ce3d65"
Failed to pull image "registry.com/asco/origin-logging-auth-proxy sha256:179b84xb803fac116f913182c2fb64a2e7adf01dd04fc58e1336d96ce0ce3d65": image pull failed for registry.com/asco/origin-logging-auth-proxy sha256:179b84ex803fac116f913182c2fb64a2e7adf01dd04fc58e1336d96ce0ce3d65, this may be because there are no credentials on this request. details: (manifest unknown: manifest unknown)

From: agoldste redhat com
Date: Thu, 7 Jul 2016 16:34:29 -0400
Subject: Re: Back-off pulling image "/origin-logging-curator sha256:b89cbdfc4e0e7d594f7a49c7581ae3f75b9d0313fce2ed8be83ee5c0426af72d"
To: dencowboy hotmail com
CC: users lists openshift redhat com

On Thu, Jul 7, 2016 at 3:48 PM, Den Cowboy <dencowboy hotmail com> wrote:

We are using our own registry which contains some necessary origin-images for us.
We already deployed the router, registry and cluster metrics using our regisry:

The images are all in the form of:

Now I try to deploy a logging project:
After starting the logging deployer template (in which I described our registry + v1.2.0) it starts pulling the origin-logging-deployer image which is fine.

Than everything seems to start but:
Back-off pulling image "myregistry.com/company/origin-logging-elasticsearch sha256:5ad3b9e964ec6e420ac047be6ae96bf04abe817d94a7d77592af1c119543b37b"
(manifest unknown: manifest unknown)

Did you push the image using Docker 1.10 and your node is running Docker 1.9?

In the deploymentconfig is also the image with the @sha
This is happening for each image of our deployment (es, kibana, fluentd, ..)

Why is it adding that @sha after our image?

If you're using ImageChangeTriggers, we translate tags to content-addressable IDs for consistent image usage so that a moving tag such as "latest" doesn't yield different images and possibly results when you deploy today, tomorrow, next week, etc.

users mailing list
users lists openshift redhat com

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]