We added heatclient and metalsmith requirements as part of our modules
so we need to make sure to properly capture them.
Change-Id: Ia5859ecf0822f19b9f04dda79d5b8b5eaf5220be
Ansible's remote_tmp module changed in 2.9.10 [1] causing
the reported bug this patch addresses. The error from the
module suggests creating the remote_tmp dir with the correct
permissions. This patch does that within external_deploy_steps
by having tripleo-ansible create the remote_tmp dir accordingly
before running ceph-ansible (which uses that remote_tmp dir).
[1] https://github.com/ansible/ansible/issues/68218
Change-Id: I0350d5253571a2b0d12a0a2f25e5469c9d1fefe0
Closes-Bug: #1884816
Paunch was removed in Victoria cycle, therefore this module isn't useful
anymore and can't work. So we need to remove it.
Change-Id: Ie1bd288ec524e025fb6e56f7bdb740ce35d6a104
When the main PID (i.e., common) of a container is killed because of
some reasons, systemd won't execute ExecStop command.
Current podman doesn't detect this failure ang recognize that container
is still running and this causes failure when systemd tries to restart
the container.
This patch introduces ExecStopPost configuration into systemd unit
files so that stop operation is executed even when a container fails
because of killed main process. The stale container should be cleaned
up by ExecStopPost task before systemd tries to restart it.
Note that the similar change has been introcuded to "podman generate
systemd" command already[1].
[1] e5c3432944
Change-Id: Idb8be98f83f95ae536762f79809aa375f04e56ab
Closes-Bug: #1884866
This patch makes osc-placmeent package installed as a part of client
packages, because this package is required to manage resources in
placement service.
Change-Id: Ib2524cf97f97693d7d03c1bb4fb1c2bde40d3bec
If a trace occurs, a final line break is visually useful to know when
the task profiling ends. The trace will be show *after* the callback so
we reduce confusion where the problem is.
Change-Id: Ibe21bdf046f71ce45f0fd36068635c4d954da5f2
When upgrading the undercloud it will be necessary to have some
ansible tasks to loop on tasks which operate on every existing stack.
This simple module will lookup existing heat stacks and return a list
of stacks which appear to be overcloud stacks. They are assumed to be
overcloud stacks if they have the output: AnsibleHostVarsMap.
Change-Id: Ib7dc2c968a5aaff40696f591caa2399c0fb5a7bc
Blueprint: nova-less-deploy
To be consistent with auto config-download and overcloud export
create subfloder with plan/stack name.
Change-Id: I08b447379ae6e267f8dff11e6e703a991918683c
Closes-Bug: #1884246
Paunch was removed from TripleO during Victoria Cycle.
tripleo_container_manage was enabled by default in Ussuri.
Therefore there is no need to run these tasks anymore, since they were
suppose to run during a deployment or upgrade of Ussuri.
Change-Id: I4f61dc954695d6c235effd44f16215d5e5401088
Sometimes, actions taken as the part of a container's entrypoint script
should become the pre-stop actions. Allow such configurations composed
for managed containers.
Change-Id: I40237e2b35ce83c5b519a117a6d4e74cb845b2c3
Related-bug: #1884048
Signed-off-by: Bogdan Dobrelya <bdobreli@redhat.com>
We need pip and virtualenv for molcule jobs. This was dropped recently
in the base image so we need to handle this ourselves.
Change-Id: Ie5ba86c4d8aa97959d7336886536dc1439ce54bd
Explicit allowance for command blacklist shouldn't be necessary in
master nor ussuri/train as per [1] and, more in general, shouldn't
ever be needed with "profile rbd"
It seems a problem seen when upgrading from jewel to luminous, for
tripleo that means upgrading from newton to queens; the changes are
for the jewel keyring so I think the patch should be posted for
the newton branch only
1. https://bugzilla.redhat.com/show_bug.cgi?id=1838145#c25
This reverts commit 4421bf8056.
Change-Id: I151a1b8d0f2775463580f46082aa22469878dc15
We forgot to handle unreachable errors.
fatal: [compute-1]: UNREACHABLE! => {"changed": false, "msg": "Failed to
connect to the host via ssh: ssh: connect to host 192.168.24.18 port 22:
No route to host", "unreachable": true}
Change-Id: I55f078be5fbdd762ffb63fb25ca6a192abb2e049
This patch modifies the keyring for OpenStack client and enables
osd blacklist command. Usually this is not required but adding
this capability to the OpenStack client Keyring, the OpenStack
nodes can advise that the 'watcher' state from the specified IP
address should be cleared (blacklisted).
Change-Id: Ia39bbdb883265782ee15083ad5e58d6495aca125
This change will allow deployers to expose sensitive data as needed.
> This change also fixes the tests for keystone. The role test was
broken because the use of test-deps needed to be updated for the new
repo layout.
Change-Id: I200efe00b735a17a996fbfe64e3f0f4d4c813f73
Signed-off-by: Kevin Carter <kecarter@redhat.com>
If display_ok_hosts is false, we were returning None but the function is
expected to return a tuple. So let's return a tuple of (None, None) and
check if we don't have a line then return.
Change-Id: Iaf2f83a8f13bcb9cf0c37acff627a456005057de
Containers configured by tripleo bind-mount /etc/hosts directly,
which means any change to that file has to preserve the original
inode, otherwise the containers will get out of sync with the
host and will not see updates.
Change tripleo_host_entries to not depend on atomic semantics of
ansible, which changes inodes on update. Instead, perform a
non-atomic update to preserve inodes, and rely on the retry
logics of openstack containers to recover from unexpected
behaviour in case /etc/hosts is consumed while it is being updated.
Closes-Bug: #1882290
Change-Id: I34dd9121bbd650b79cb523e4dbed5949a0e7d52d
Tasks did not have the concept of throttling prior to ansible 2.9 so if
we want to allow this to be re-used with older versions then we need to
skip the throttling code if task doesn't contain the throttle attribute.
This will help if we want to backport this for upgrades.
Change-Id: I01f6e3a69dec25ec69782c087d6b81ce3f33e6c0
Alex Schultz