Without this change on Centos Stream 9 we will get errors like the
following:
+ nsenter --net=/run/netns/qdhcp-7083ee69-e139-4844-8af8-7efa798a3c49 --preserve-credentials -m -t 1 podman run --detach --log-driver k8s-file --log-opt path=/var/log/containers/stdouts/neutron-dnsmasq-qdhcp-7083ee69-e139-4844-8af8-7efa798a3c49.log -v /var/lib/config-data/puppet-generated/neutron/etc/neutron:/etc/neutron:ro -v /run/netns:/run/netns:shared -v /var/lib/neutron:/var/lib/neutron:shared -v /dev/log:/dev/log --net host --pid host --cgroupns host --privileged -u root --name neutron-dnsmasq-qdhcp-7083ee69-e139-4844-8af8-7efa798a3c49 undercloud-0.ctlplane.home.arpa:8787/tripleomastercentos9/openstack-neutron-dhcp-agent:8f4fa80dca5a62ca9a9049b664a60868 /usr/sbin/dnsmasq -k --no-hosts --no-resolv --pid-file=/var/lib/neutron/dhcp/7083ee69-e139-4844-8af8-7efa798a3c49/pid --dhcp-hostsfile=/var/lib/neutron/dhcp/7083ee69-e139-4844-8af8-7efa798a3c49/host --addn-hosts=/var/lib/neutron/dhcp/7083ee69-e139-4844-8af8-7efa798a3c49/addn_hosts --dhcp-optsfile=/var/lib/neutron/dhcp/7083ee69-e139-4844-8af8-7efa798a3c49/opts --dhcp-leasefile=/var/lib/neutron/dhcp/7083ee69-e139-4844-8af8-7efa798a3c49/leases --dhcp-match=set:ipxe,175 --dhcp-userclass=set:ipxe6,iPXE --local-service --bind-dynamic --dhcp-range=set:subnet-eaf4f552-0023-4e4a-9b6e-b5f3727e647e,192.168.24.0,static,255.255.255.0,86400s --dhcp-option-force=option:mtu,1500 --dhcp-lease-max=256 --conf-file=/dev/null --domain=home.arpa
Error: create directory `/sys/fs/cgroup/../../libpod-b43b25a3bf3af73d24d3f4c52b983334ead7d28462e9ed8ba83f207e4264b1ed.scope`: No such file or directory: OCI runtime attempted to invoke a command that was not found
See also the related libvirt/nova issue in https://bugs.launchpad.net/tripleo/+bug/1936005
Tested and with this change the error is gone and dhcp/ovn-metadata is functional
again.
Closes-Bug: #1944522
Change-Id: I8a5b885782386a89c463ff0b7afd7ace91958278
There is nothing distribution specific in most of these.
Let's stop using those so that we can drop those from
packaging.
Change-Id: I1feb1b4167ea25710a2dc274cf73e4bf9b03cdd1
The change to hard-code the default boot mode to ``uefi`` was reverted
since the test environment (OVB on vexxhost) doesn't yet support UEFI
boot. This change adds the IronicDefaultBootMode parameter so the
default boot mode can be set appropriately for each environment.
Blueprint: whole-disk-default
Change-Id: I10348eb1aaf6a594ae06bcc855dfdd8c6c00bfe9
Cgroups v2 have different format of representing its
entries. Adjust the parser to match for the looked up
container id for both cgroups v1 and v2 cases.
Note, that docker would require the container created with --cgroupns
host. But (non-userspace) podman containers show that information
without such create/run args modification as well, which is also good
for upgrades. Since we no longer support docker, and cgroups v2 will
come exclusively with podman containers, drop the container_cli logic
and assume there is always podman used.
Co-authored-by: Michele Baldessari <michele@redhat.com>
Change-Id: Ief3a80f0099d8aa8b43fac64497c34505827a6e6
Signed-off-by: Bogdan Dobrelya <bdobreli@redhat.com>
This makes the following changes:
- Create cloud.yaml for undercloud with standalone services
- Remove keystone and memcached from the undercloud
Change-Id: I741123dcbf4544845b40e253d8b42d868317cc03
Zaqar was deprecated in Wallaby and is no longer in use on the
undercloud and it hasn't been officially supported in the
overcloud for some time.
Change-Id: I3bdcc72d6127ec96ff2307cafbf57f6178c3ef5c
Mistral was deprecated in Wallaby and is no longer in use on the
undercloud and it hasn't been officially supported in the overcloud for
some time.
Change-Id: I6963453f53cb554ca8fdb58706f04838bbd11ba0
There is no ways to set pacemaker stonith-timeout values to something
different than the default values. This change would allow a user
to set any pacemaker cluster properties using PacemakerProperties.
PacemakerProperties:
'stonith-timeout':
property: 'stonith-timeout'
value: '120s'
Change-Id: Ied77623953fd97dce7de1d529e98d5483a42e31b
Openstack users are configured with openstacklib, which in turns
drive puppet-mysql to create several DB user for each db service:
<service>@'%' <service>@<ip> and <service>@<mysql_vip>.
We create several users because we use two different parameters
host and allowed_hosts in openstacklib, which only has the effect
of creating a list of users per openstack service.
However since we always create a user '%', this wildcard host
will always allow connection to the DB, so the other users are
currently not useful as they don't get any additional grants or
restrictions.
Simplify the entire mysql user creation to only generate one
user per service, with a wildcard host.
Change-Id: I928b03f06c702a13f4bd957eaa79153aa711cee4
Closes-Bug: #1943440
Closes-Bug: #1943330
tftp-server and xinetd are not anticipated to be in RHEL9 and as
such we need to remove their use in order to facilitate a transition
to RHEL9 when the time comes.
In this case, we're anticipating we can just directly launch dnsmasq
instead of in.tftpd directly within the container.
Change-Id: Ie75aa0f2994a4cf662905b912231a12e3bd8afe7
This is basically a revert ot I8f71bf83ddafca167deae1a38ca819f7d930fb80
https://bugzilla.redhat.com/show_bug.cgi?id=1759974 has now been fixed
via https://bugzilla.redhat.com/show_bug.cgi?id=1760763:
"[ovsdb-server] Allow replicating from older schema servers"
which has been fixed in openvswitch2.11-2.11.0-35.el8fdp.x86_64
which was released on 2020-01-22.
I think in master (and potentially wallaby) we can just remove this
workaround now.
Change-Id: I428a30d2ad7be295f8b3dbeaddd76eb88432a532
The package is part of the base image, but when doing a leapp
upgrade it is removed because it is not part of the standard
RHEL repositories.
Resolves: rhbz#1994919
Change-Id: Ia892ccb629b7ef7cb80d549fbe1c684e9b93c0fc
Zuul