Bug 1798637: baremetal: IPv6 add dhcp-duid to NetworkManager config #1395
Conversation
openshift-ci-robot
added
the
size/XS
hardys
changed the title
|
#1375 has some related discussion |
hardys
changed the title
|
/retest |
|
/test e2e-gcp-upgrade |
|
Its time for a well established common configuration for all 3 baremetal, openstack, ovirt. Possibly more platforms will benefit from it, vsphere, libvirt, and so on. Question is how to properly have platforms opt-in to such common configuration. One way might be symbolic links - but then we need to solve templating: so @runcom is there an issue on this topic? if not I'll open one |
|
@rgolangh agreed figuring out how to de-duplicate this would be good
If we can't solve the templating another approach would be to rework the dispatcher scripts to they are common to all platforms, but then source a per-platform file with the variables that are templated (and do nothing if such a file doesn't exist?) |
|
/lgtm |
This has been tested by several folks now via the https://github.com/openshift-kni/machine-config-operator/commits/4.3-ipv6 branch cc @russellb @derekhiggins - I think it's good to go provided @celebdor and @bcrochet agree |
|
/retest |
2 similar comments
|
/retest |
|
/retest |
|
also ping: @celebdor :) |
|
@hardys looks like you need to rebase this PR as another changed the same yaml. |
openshift-ci-robot
added
the
needs-rebase
There was a problem hiding this comment.
Adding approval.
FYI: once you push your rebase, it'll clear the existing LGTM and you'll need another asap @hardys
openshift-ci-robot
added
the
approved
Adding ipv6.dhcp-duid=ll means we get a predictable client ID so that reservations can be made on the DHCP server - this is useful if you want to provide the hostname via DHCP (although a reverse DNS lookup should also work) Reservations are also required to ensure the controlplane IP of the masters won't ever change post-deployment which will be necessary if the DNS lookup approach is taken to set the hostnames.
openshift-ci-robot
removed
the
needs-rebase
|
/lgtm |
|
/lgtm |
|
@derekhiggins: changing LGTM is restricted to collaborators In 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 APPROVED This pull-request has been approved by: bcrochet, celebdor, derekhiggins, hardys, kikisdeliveryservice The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
|
/retest Please review the full test history for this PR and help us cut down flakes. |
4 similar comments
|
/retest Please review the full test history for this PR and help us cut down flakes. |
|
/retest Please review the full test history for this PR and help us cut down flakes. |
|
/retest Please review the full test history for this PR and help us cut down flakes. |
|
/retest Please review the full test history for this PR and help us cut down flakes. |
|
/cherrypick release-4.3 |
|
@hardys: #1395 failed to apply on top of branch "release-4.3": In 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. |
|
@hardys: The following test failed, say
Full PR test history. Your PR dashboard. Please help us cut down on flakes by linking to an open issue when you hit one in your PR. 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. I understand the commands that are listed here. |
|
/cherrypick release-4.3 |
|
@hardys: #1395 failed to apply on top of branch "release-4.3": In 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. |
|
/cherrypick release-4.3 |
|
@hardys: new pull request created: #1441 In 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. |
|
/retitle Bug 1798637: baremetal: IPv6 add dhcp-duid to NetworkManager config |
openshift-ci-robot
changed the title
|
@hardys: All pull requests linked via external trackers have merged. Bugzilla bug 1798637 has been moved to the MODIFIED state. In 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. |
Adding ipv6.dhcp-duid=ll means we get a predictable client ID so that reservations can be made on the DHCP server - this is useful if you want to provide the hostname via DHCP (although a reverse DNS lookup should also work) Reservations are also required to ensure the controlplane IP of the masters won't ever change post-deployment which will be necessary if the DNS lookup approach is taken to set the hostnames. This ports the BM fix from openshift#1395 to OpenStack platform.
Adding ipv6.dhcp-duid=ll means we get a predictable client ID so that reservations can be made on the DHCP server - this is useful if you want to provide the hostname via DHCP (although a reverse DNS lookup should also work) Reservations are also required to ensure the controlplane IP of the masters won't ever change post-deployment which will be necessary if the DNS lookup approach is taken to set the hostnames. This ports the BM fix from openshift#1395 to OpenStack platform.
Adding ipv6.dhcp-duid=ll means we get a predictable
client ID so that reservations can be made on the
DHCP server - this is useful if you want to provide
the hostname via DHCP (although a reverse DNS lookup
should also work)
Reservations are also required to ensure the controlplane
IP of the masters won't ever change post-deployment
which will be necessary if the DNS lookup approach
is taken to set the hostnames.
- What I did
Updated the NetworkManager configuration, only for the baremetal (IPI) platform, so we specify the ipv6.dhcp-duid=ll option. This enables a predictable (mac derived) client ID/DUID when deploying in an IPv6 environment.
- How to verify it
If testing with dev-scripts you can check via
virsh net-dhcp-leases baremetal- on baremetal you can check the client ID in the lease for the nic connected to the controlplane network (/var/lib/NetworkManager/dhclient6*.lease)In both cases the controlplane network must be configured for single-stack ipv6, and there are some other PRs open to enable that e2e with the default OpenShift builds. openshift-metal3/dev-scripts#871 has some more details.
- Description for the changelog
Updated the NetworkManager configuration, for the baremetal (IPI) platform to specify the ipv6.dhcp-duid=ll option. This enables a predictable (mac derived) client ID/DUID.