kubernetes: 1.4.6 -> 1.5.1 #21125
kubernetes: 1.4.6 -> 1.5.1 #21125
Conversation
|
@moretea, thanks for your PR! By analyzing the history of the files in this pull request, we identified @ebzzry, @offlinehacker and @rushmorem to be potential reviewers. |
|
After updating this PR to 1.5.1, the k8s tests still fail. |
|
One problem is that the generated docker image is not valid. I'm getting the following error when the docker daemon starts: After deleting UPDATE: This was probably caused by an invalid state in the qcow2 image in |
|
So this is docker related issue, not kubernetes?
…On Wed, Dec 14, 2016, 1:07 PM Maarten Hoogendoorn ***@***.***> wrote:
One problem is that the generated docker image is not valid.
I'm getting the following error when the docker daemon starts:
[ 9.708936] dockerd[1230]: time="2016-12-14T12:01:32.822686391Z" level=info msg="libcontainerd: new containerd process, pid: 1238"
[ 9.805169] dockerd[1230]: time="2016-12-14T12:01:32.919309844Z" level=warning msg="devmapper: Usage of loopback devices is strongly discouraged for production use. Please use `--storage-opt dm.thinpooldev` or use `man docker` to refer to dm.thinpooldev section."
[ 9.839039] dockerd[1230]: time="2016-12-14T12:01:32.953487093Z" level=warning msg="devmapper: Base device already exists and has filesystem xfs on it. User specified filesystem will be ignored."
[ 9.850956] dockerd[1230]: time="2016-12-14T12:01:32.965431349Z" level=info msg="[graphdriver] using prior storage driver \"devicemapper\""
[ 9.917129] dockerd[1230]: time="2016-12-14T12:01:33.030317808Z" level=fatal msg="Error starting daemon: layer does not exist"
After deleting /var/lib/docker, I can successfully start docker.service
again. /cc @offlinehacker <https://github.com/offlinehacker> @domenkozar
<https://github.com/domenkozar>
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#21125 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AAjvSwWJv0tVTpph5EVZAAGjehNYPZD6ks5rH9wIgaJpZM4LLylv>
.
|
|
Apparently the VM state was tainted. After removing the |
|
@offlinehacker I noticed that the cluster test is broken on master as well. I'll try to fix that. |
|
@moretea finishing kubernetes tests and fixing any bugs would be very helpful. |
|
This PR has merge conflicts, needs reading. |
moretea
force-pushed
the
update-k8s-to-1.5
branch
2 times, most recently
from
265b624
to
0850b66
Compare
|
@offlinehacker After rebasing it's getting OOM errors on my 16GB laptop, see https://gist.github.com/moretea/961bb2bac9bee39416c43a4204aebedc |
|
@moretea you can disable docs generation(comment out) for now if that seems to be an issue |
moretea
force-pushed
the
update-k8s-to-1.5
branch
from
0850b66
to
2b285cb
Compare
|
I found the problem and fixed it in a39ac20 Tested with @offlinehacker if you're OK with these changes, I'll squash this to one commit. |
|
@offlinehacker btw. you can squash commits also as a maintainer on merge. |
Disabled "mungedocs", which broke the build. This appears to be a piece of development tooling to make sure that the documentation is correct. We don't really care about that when we a specific k8s version for NixOS.
Dig could not be found in the test cases. Adding it as a global package fixes this.
moretea
force-pushed
the
update-k8s-to-1.5
branch
from
92a768c
to
9f892de
Compare
|
@offlinehacker done. |
Is there a PR for this? I could not find it. I guess that a general mechanism for CNI should be the solution there. I'm willing to contribute to that. |
Motivation for this change
Updating k8s to
1.5.01.5.1, which supports the Container Runtime Interface.Things done
(nix.useSandbox on NixOS,
or option
build-use-sandboxinnix.confon non-NixOS)
nix-shell -p nox --run "nox-review wip"./result/bin/)The tests fail at the moment, because the kube-apiserver has some ACL problem: