data/bootstrap/files/usr/local/bin/bootkube: Drop config render #5800
Conversation
$ curl -s https://gcsweb-ci.apps.ci.l2s4.p1.openshiftapps.com/gcs/origin-ci-test/pr-logs/pull/openshift_installer/5800/pull-ci-openshift-installer-master-e2e-gcp/1513917400393715712/artifacts/e2e-gcp/ipi-install-install/artifacts/log-bundle-20220412175035.tar | tar xz --strip-components=1
$ grep -B2 'Main process exited' bootstrap/journals/bootkube.log | sed -n 's/^Apr 12 [0-9:]* //p' | sed 's/[[][0-9]*]/[...]/' | sort | uniq -c
86 ci-op-l44cr94r-15937-7d7tq-bootstrap bootkube.sh[...]: Error: error creating container storage: the container name "mco-render" is already in use by "877829c1add4c25f797255405478c4b8f75da308bc5cb3037e171487eec0ff37". You have to remove that container to be able to reuse that name.: that name is already in use
1 ci-op-l44cr94r-15937-7d7tq-bootstrap bootkube.sh[...]: F0412 17:31:21.586273 1 bootstrap.go:118] error rendering bootstrap manifests: failed to load the cloud provider config: open /assets/config-bootstrap/cloud-provider-config-generated.yaml: no such file or directory
86 ci-op-l44cr94r-15937-7d7tq-bootstrap bootkube.sh[...]: Rendering MCO manifests...
1 ci-op-l44cr94r-15937-7d7tq-bootstrap mco-render[...]: F0412 17:31:21.586273 1 bootstrap.go:118] error rendering bootstrap manifests: failed to load the cloud provider config: open /assets/config-bootstrap/cloud-provider-config-generated.yaml: no such file or directory
86 ci-op-l44cr94r-15937-7d7tq-bootstrap systemd[...]: bootkube.service: Main process exited, code=exited, status=125/n/a
1 ci-op-l44cr94r-15937-7d7tq-bootstrap systemd[...]: bootkube.service: Main process exited, code=exited, status=255/n/aSeems like we could use a larger clear to avoid /hold |
openshift-ci
bot
added
the
do-not-merge/hold
|
Ah, also in So we need some kind of "if the name exists, remove it" guard up around here for the |
Avoid [1]: Error: error creating container storage: the container name "mco-render" is already in use by "877829c1add4c25f797255405478c4b8f75da308bc5cb3037e171487eec0ff37". You have to remove that container to be able to reuse that name.: that name is already in use and similar for other containers by removing inconsistent --rm options and baking that in at the bootkube_podman_run level. Also add an etcd-bootstrap rm call, to clear out any cruft from a previous bootkube round before calling Podman for a fresh etcd render. [1]: openshift#5800 (comment)
wking
force-pushed
the
drop-config-render
branch
from
66bfad5 to
934c879
Compare
wking
force-pushed
the
drop-config-render
branch
2 times, most recently
from
7b8ba14 to
24f8445
Compare
|
Issues go stale after 90d of inactivity. Mark the issue as fresh by commenting If this issue is safe to close now please do so with /lifecycle stale |
openshift-ci
bot
added
the
lifecycle/stale
|
Stale issues rot after 30d of inactivity. Mark the issue as fresh by commenting If this issue is safe to close now please do so with /lifecycle rotten |
openshift-ci
bot
added
lifecycle/rotten
openshift-merge-robot
added
the
needs-rebase
|
Rotten issues close after 30d of inactivity. Reopen the issue by commenting /close |
|
@openshift-bot: Closed this PR. DetailsIn response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
We've had config-operator rendering on the bootstrap node since 9994d37 (bootkube: render config.openshift.io resources, 2019-02-12, openshift#1187). Motivation for that commit isn't clear to me; [1] suggests maybe keeping CRDs out of the installer repository. But we run a rendered cluster-version operator on the bootstrap machine since 63e2750 (ignition: add CVO render to bootkube.sh, 2018-09-27, openshift#330), so we should be able to push resources at bootstrap time via the CVO. Remove CRDs from the config rendering, so we can see if things work without the config-rendered cluster-bootstrap pushes racing the bootstrap CVO pushes, or the config-rendered pushes not realizing they should filter out manifests annotated for capabilities that are not enabled. [1]: openshift#1187 (comment)
wking
force-pushed
the
drop-config-render
branch
from
24f8445 to
1f115ab
Compare
openshift-merge-robot
removed
the
needs-rebase
|
@wking: The following tests failed, say
Full PR test history. Your PR dashboard. DetailsInstructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. I understand the commands that are listed here. |
|
Rotten issues close after 30d of inactivity. Reopen the issue by commenting /close |
|
@openshift-bot: Closed this PR. DetailsIn response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
3 participants
We've had config-operator rendering on the bootstrap node since 9994d37 (#1187). Motivation for that commit isn't clear to me; this comment suggests maybe keeping CRDs out of the installer repository. But we run a rendered cluster-version operator on the bootstrap machine since 63e2750 (#330), so config manifests should have been getting pushed at bootstrap time via the CVO. Drop the config rendering, so we can see if things work without the config-rendered cluster-bootstrap pushes racing the bootstrap CVO pushes.