This repository was archived by the owner on Jul 23, 2019. It is now read-only.
Merge openshift/installer for 4.1.0-rc.3#75
Merged
russellb merged 159 commits intoopenshift-metal3:masterfrom May 16, 2019
Merged
Merge openshift/installer for 4.1.0-rc.3#75russellb merged 159 commits intoopenshift-metal3:masterfrom
russellb merged 159 commits intoopenshift-metal3:masterfrom
Conversation
Dia is from Stephen Cuppett, replacing the PNG he'd submitted via 39a926a (Adding initial user doc/guide & materials for UPI AWS installation, 2019-03-12, #1408). We aren't using the full file with all the layers yet, but I'm building it anyway because folks without Dia may still want to look at it ;). SVGs generated with: $ dia --version Dia version 0.97.3, compiled 18:02:21 Feb 11 2017 relink-dia.py embeds the referenced icons in the SVG with def and use [1,2] to avoid icon URIs like: file:///.../openshift/installer/docs/user/aws/images/./AWS-Architecture-Icons_SVG/Light-BG/_General%20AWS/AWS-General_AWS-Cloud_light-bg.svg Ideally Dia would have a way to do this sort of thing automatically with a command-line switch, but if it does, I can't find it. [1]: https://developer.mozilla.org/en-US/docs/Web/SVG/Element/def [2]: https://developer.mozilla.org/en-US/docs/Web/SVG/Element/use
This reverts commit c1aa022, #618. We've made a lot of progress separating out the per-platform components since that went in, so take another run at formally separating approval for per-platform and core changes.
Recently, the control plane switched to secure ports in [1] and [2]. This aligns them in the installer. [1] openshift/cluster-kube-scheduler-operator#88 [2] openshift/cluster-kube-controller-manager-operator#207
Punt per-platform docs down into platform-specific subdirectories so we can apply platform-specific OWNERS files.
This is the channel that the beta4 drop and later will be in.
The terraform-created OpenStack hostnames now have the short name rather than FQDN. This is consistent with the worker nodes and makes sure the actuator's nodelink-controller is able to match the names and sync the nodes. The node names without FQDN are now resolvable as well.
Same as #1601 but for OpenStack.
These were changes from two to three in 16dfbb3 (data/aws: use nlbs instead of elbs, 2019-11-01, #594): $ git show 16dfbb3 | grep _threshold | sort | uniq - healthy_threshold = 2 - # healthy_threshold = 2 + healthy_threshold = 3 - unhealthy_threshold = 2 - # unhealthy_threshold = 2 + unhealthy_threshold = 3 Alex doesn't remember intentionally making that change, and the lower thresholds will hopefully help mitigate some issues with continued connection attempts to API servers which are in the process of shutting down.
As we're fixing the kubelet TLS files, we've noticed places that are over-signing and trusting the wrong thing. This updates the next case we can use to ratchet the fixes into the MCO.
RHCOS has changed the name of the NIC from eth0 to ens192 [1]. The ifcfg file laid down to set the static IP address of the machines needs to be updated to reflect the new name of the NIC. [1] coreos/coreos-assembler#442
Add instructions in the documentation for setting the hostname and static IP address of a machine. Add language to indicate that the preferrence is to use DHCP for establishing the hostname and IP addresses.
openstack: add myself to openstack-approvers
openstack: Remove FQDN from node hostnames
move etcd connection information to openshift-config
Modify the progress service so that it creates a configmap when bootstrapping is done rather than an event. This is done so that the installer can determining whether bootstrapping has completed even when run later, after the bootstrap-complete event would otherwise have expired and been deleted. https://bugzilla.redhat.com/show_bug.cgi?id=1698573
docs: Stronger per-platform isolation
BUG 1698573: cmd: level-driven bootstrap complete
tls: remove serving capability from deprecated apiserver cert, also fix Complete ClientCA bundle for apiserver
…failure This change adds a new subcommand to assists with debugging a failed bootstrap by collecting logs from the cluster. Currently, it outputs the shell commands to run in order to gather these logs. It can later be extended to gather the logs itself. It is called automatically from the create cluster subcommand on a failed bootstrap.
New gather subcommand to assist debugging bootstrap failures.
docs/user/aws/images: Dia source for UPI arch diagram
We're locking ourselves in for stability going forward. Hopefully we got this right ;).
Bug 1707573: *: remove deprecated EtcdCA and promote EtcdSigner
PR [1] kept the `apiServerURL` field unchanged ie set to internal url for Kubernetes API so that the consumers of the field can move away from it. With KAS-O [2] change merged, we can move forward [1]: openshift/installer#1718 [2]: openshift/cluster-kube-apiserver-operator#465
Bug 1706689: asset/manifests: internal api server URL for infrastructure
Repo `github.com/coreos/terraform-provider-matchbox` has moved ownership to `github.com/poseidon/terraforn-provider-matchox` [1] So the previous `go get github.com/coreos/terraform-provider-matchbox` now fails to build [2] as the new repo moved to new module [3] So the move to fetching the binary from the github released artifacts [4] seems like the better move. [1]: https://github.com/poseidon/terraform-provider-matchbox [2]: https://openshift-gce-devel.appspot.com/build/origin-ci-test/pr-logs/pull/openshift_installer/1729/pull-ci-openshift-installer-master-images/4202 [3]: poseidon/terraform-provider-matchbox#43 [4]: https://github.com/poseidon/terraform-provider-matchbox/releases/tag/v0.2.3
images/installer: update upi Dockerfile to fetch matchbox using releases
Bug 1706689: asset/manifests: update infrastructure to set public apiserver url
rhcos: Bump to 410.8.20190508.1
Collaborator
|
Build FAILURE, see build http://10.8.144.11:8080/job/dev-tools/628/ |
BUG 1707096: gather: collect secret names for failure analysis
openstack: use static .repo file for openstack client
|
Hmm the CI job failed because the branch doesn't rebase cleanly on master, and also if you try to build the branch without rebasing I see this error: Which is weird because it's defined at https://github.com/libvirt/libvirt-go/blob/master/connect.go#L321 and that's in the vendor dir... |
|
Actually it builds fine with the BUILD/TAGS defined in |
Member
Author
|
CI always fails on rebasing this way, it can't apply the patch. The branch needs to be checked out. Although, this is probably not worth testing at the moment - you won't get very far. OpenShift moved to using api-int DNS address even for ignition, and we don't have that record. There's work in progress, openshift-metal3/dev-scripts#462, but that PR is not enough. @celebdor and @cybertron are still working on a solution. |
stbenjam
changed the title
Collaborator
|
Build FAILURE, see build http://10.8.144.11:8080/job/dev-tools/633/ |
7 tasks
markmc
suggested changes
May 15, 2019
Contributor
|
Another tiny change that could come with the rebase - markmc@36b9ba6 Fine with that going in post-merge though |
Contributor
|
Apart from those 2 small issues, I don't see any mechanical issues with the rebase. Nicely done! |
Member
Author
|
I incorporated markmc@36b9ba6, and fixed the bootkube issue |
Collaborator
|
Build FAILURE, see build http://10.8.144.11:8080/job/dev-tools/637/ |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
20 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
See also openshift-metal3/dev-scripts#523