Skip to content

Conversation

markgoddard
Copy link
Contributor

  • Restore upstream mirrors in container image package repos
  • Fix for ci-aio automated deploy on non-lvm images
  • Update CentOS 8 Stream snapshots
  • Exclude broken iptables from latest snapshot
  • Reduce service workers for aufn-ceph & virtualised ci-multinode envs
  • multinode: Run RefStack Tempest tests by default
  • Fix growroot.yml on non-LVM images
  • Minor growroot.yml fixes
  • Bump container image tags
  • Revert original non-LVM deployment fix
  • Add bifrost image tag
  • Pin all iptables packages
  • Fixup for previous commit
  • growroot.yml conditonal fix
  • Bump RefStack to OpenStack Powered Platform 2022.11
  • Adds Centos 8 Stream Extras Common
  • Update release note
  • Fixups for missing references to extras common
  • Further fixups for path consistency
  • Make identifier and name match existing in .repo file
  • rebuild opensearch images
  • Bump images to switch to extras common
  • Add physical size to nvmemon metrics
  • Add release note
  • Change skip_lvm_check to ignore_lvm_check
  • Revert container image bump to nova libvirt
  • Revert growpart check
  • Mention switch from extras to extras common
  • fix nova-libvirt tag
  • Update etc/kayobe/kolla/globals.yml
  • Add configuration for ELRepo 9
  • Update kibana
  • Set kolla tag in base config
  • Fix repo URL
  • Bump ubuntu snapshots
  • Make sure kibana is built in standard set of images
  • New snapshots
  • Bump ubuntu images
  • Add missing linux-firmware package in Rocky Linux 9
  • Use q35 machine type by default
  • Whitespace fix
  • Remove plugins and group_vars symlinks
  • Fix nova_libvirt pin
  • CI: Only sync old images after build
  • Use inventory hostnames in prometheus
  • Pin pulp version to 3.24.0
  • Bump image tag
  • Add release notes
  • Clarify release notes
  • Mention the need to reboot.
  • fix build of rocky images
  • fix ubuntu build
  • Keep 10percentOSDsDown alert pending for 15 minutes
  • Update pulp-repo-versions.yml
  • prometheus: Drop some version overrides
  • CI: Set selinux_state to disabled for ci-builder environment
  • bump octavia tag - bug fixes
  • Fix seed container image build
  • Fix kolla_base_distro check - no more CentOS
  • Remove kibana image references
  • Revert ironic-tftp to ironic-pxe image change
  • Remove kolla_build_customizations_rocky defaults
  • Removed num_pcie_ports merged upstream
  • ovn: bump Rocky version to 23.03
  • Update kayobe-automation submodule to use Rally-docker v1.4-dev
  • swap: Note that Kayobe supports swap in Zed
  • tempest: update max microversions for Cinder & Nova
  • Update stackhpc.pulp to 0.5.2
  • ceph: remove unnecesary Jinja braces
  • aufn-ceph: Use stackhpc/zed branch in script
  • requirements.txt: drop version specifier
  • pulp: Stop syncing prometheus-jiralert image
  • aufn-ceph: Drop cephadm_image override
  • aufn-ceph: Sync inventory with kolla-ansible
  • CI: Allow logging of Rally/Tempest
  • Bump kolla tags for Bifrost
  • Disable ignore_null_merges for release notes

markgoddard and others added 30 commits April 13, 2023 13:25
Currently the Kolla container images configure Yum repositories to use
pulp-server on SMS Cloud:

(nova-compute) $ grep baseurl /etc/yum.repos.d/CentOS-Stream-BaseOS.repo
baseurl=http://pulp-server.internal.sms-cloud:8080/pulp/content/centos/8-stream/BaseOS/x86_64/os/20211122T102435

This makes it difficult to install/update packages for debugging issues
in customer environments.

This change restores Yum repository files at the end of the container image
build to point to distribution repositories. It also makes an attempt to
do the same for Apt repos, although these are not enabled in Wallaby
currently.
Restore upstream mirrors in container image package repos
- Updates CentOS Stream 8 OS snapshots
- Updates all additional repos used in CentOS Stream 8 container images

Versions from: https://github.com/stackhpc/stackhpc-release-train/actions/runs/6071032032
When running a multinode environment in VMs, we can easily run out of
memory on controllers. Reduce to 2 service/RPC workers in this case.
Also reduce RPC workers to 1 for the aufn-ceph environment.
A full tempest test run takes a long time, and may not always be
necessary. Let's run RefStack by default, and make it easy to switch to
the full test suite.
This image wasn't built due to a bug in the workflow.
multinode: Run RefStack Tempest tests by default
The old images were built with the old extras repo.
```
 Live Migration failure: internal error: qemu unexpectedly closed the monitor: 2023-09-06T09:55:34.855238Z qemu-kvm: Missing section footer for 0000:00:01.3/piix4_pm
2023-09-06T09:55:34.855431Z qemu-kvm: load of migration failed: Invalid argument: libvirt.libvirtError: internal error: qemu unexpectedly closed the monitor: 2023-09-06T09:55:34.855238Z qemu-kvm: Missing section footer for 0000:00:01.3/piix4_pm
```

See: https://gitlab.com/qemu-project/qemu/-/issues/932
markgoddard and others added 23 commits September 27, 2023 10:20
We should always be on Python 3.8+ in Zed.
The default in the base config should use this when Ceph images are not synced to Pulp.
By default the 'Run tempest' task has no_log set to avoid revealing
sensitive data. This does not apply in CI, and makes it difficult to
debug failures.
Keep 10percentOSDsDown alert pending for 15 minutes
Bump RefStack to OpenStack Powered Platform 2022.11
This allows reno to find additional release notes.
 Disable ignore_null_merges for release notes
@markgoddard markgoddard requested a review from a team as a code owner September 28, 2023 12:47
@markgoddard markgoddard self-assigned this Sep 28, 2023
@markgoddard markgoddard merged commit 0699057 into stackhpc/2023.1 Oct 6, 2023
@markgoddard markgoddard deleted the 2023.1-zed-merge branch October 6, 2023 08:40
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.