DNM Dev#3982
Conversation
When no Ansible inventory host matches a k8s node hostname (e.g. bare metal SNO where the node is not SSH-accessible), the role silently skips directory creation while still creating PVs that reference non-existent paths. Add an oc debug fallback that creates directories on each node via a debug pod. Generated-by: claude-4.6-opus-high Signed-off-by: Bohdan Dobrelia <bdobreli@redhat.com>
The ceph.yml post_stage_run hook (via cifmw_ceph_client role) writes Ceph config files to cifmw_ceph_client_fetch_dir (default /tmp/). This template reads those files and provides them as base64-encoded values under data.ceph_conf (DCN convention). Generated-by: claude-4.6-opus-high Signed-off-by: Bohdan Dobrelia <bdobreli@redhat.com>
To allow BM SNO with ceph using custom ceph CIDR values, make ssh_network_range and storage_mgmt_network_range overridable via cifmw_ceph_ssh_network_range and cifmw_ceph_storage_mgmt_network_range. Both are set in set_fact which clobbers extra vars, so we use the cifmw_ indirection with default() to preserve original defaults. NOTE: storage_network_range also needs this treatment. It use to be commented out in set_fact, and this change needs extra testing with Ceph ci jobs perhaps. Also gather network facts for IP-to-host mapping. Signed-off-by: Bohdan Dobrelia <bdobreli@redhat.com>
On SNO with a single EDPM compute (single-host CephHCI), the Ceph ingress service (haproxy/keepalived) is not deployed because the ceph_rgw.yml.j2 spec template only creates it for multi-host clusters. Add a hook to correct the Keystone Swift endpoint for this case. The proper fix belongs in cifmw_cephadm/tasks/configure_object.yml which should detect whether ingress is deployed and choose VIP:8080 vs host_ip:8082 accordingly. Signed-off-by: Bohdan Dobrelia <bdobreli@redhat.com>
|
Skipping CI for Draft Pull Request. |
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
|
Merge Failed. This change or one of its cross-repo dependencies was unable to be automatically merged with the current state of its repository. Please rebase the change and upload a new patchset. |
Modify cifmw_block_device role to support loop back and thin LVM devices. For ceph deploy hook, when cifmw_ceph_thin_pool is defined, create thin LVs from an existing VG thin pool instead of loop-backed files. This supports baremetal hosts where the disk is fully consumed by a thin pool with no raw devices available for cephadm. The new play runs before the existing loop device play and sets cifmw_ceph_spec_data_devices with explicit LV paths, which also causes the loop device play to be skipped. Signed-off-by: Bohdan Dobrelia <bdobreli@redhat.com>
Modify bm_sno role to let it provision thin volumes by LVM Storage operator for a proper bare metal performance in BM SNO CI jobs. Signed-off-by: Bohdan Dobrelia <bdobreli@redhat.com>
|
Build failed (check pipeline). Post ✔️ openstack-k8s-operators-content-provider SUCCESS in 2h 21m 22s |
1 participant
No description provided.