Manage swap partition as a hard requirement

[bogdando]

Add swap partition management into the bootstrap role

The swap file/partition management logic is carried over from the existing EDPM pre-adoption implementation, but moved out of the nova compute specific scope into generic tasks for EDPM.

Depends-On: https://review.rdoproject.org/r/c/rdo-jobs/+/48683

Closes: OSPRH-133

[bogdando]

The molecule testing is blocked as we cannot use swapon [0] from podman containers, we'd need a delegated driver for that, likely. Any hints to make this unblocked, @raukadah perchance?

UPDATE:
#154 zuul job example
we can move the job to zuul job and there we can use the delegated driver

[0]

Configure swap file] ***************************\nfatal: [instance]: FAILED! => {\"changed\": true, \"cmd\": \"#!/bin/bash\\nset -eu\\nif [ ! -f /swap ]; then\\n  dd if=/dev/zero of=/swap count=4096 bs=1M\\n  chmod 0600 /swap\\n  mkswap /swap\\n  swapon /swap\\n  if ! grep -qE \\\"/swap\\\\b\\\" /etc/fstab; then\\n    echo \\\"/swap swap swap defaults 0 0\\\" >> /etc/fstab\\n  fi\\nfi\\n\", \"delta\": \"0:00:20.914521\", \"end\": \"2023-05-18 11:57:16.365692\", \"msg\": \"non-zero return code\", \"rc\": 255, \"start\": \"2023-05-18 11:56:55.451171\", \"stderr\": \"4096+0 records in\\n4096+0 records out\\n4294967296 bytes (4.3 GB, 4.0 GiB) copied, 16.0297 s, 268 MB/s\\nswapon: /swap: swapon failed: Operation not permitted\"

[bogdando]

the rdo-project job is failed, probably by an unrelated reason [0], because I can see the swap task completed there [1]

[0]

TASK [osp.edpm.edpm_ovn : Check if OVS Manager already exists] *****************
fatal: [38.102.83.191]: FAILED! => {"changed": true, "cmd": "ovs-vsctl show | grep -q \"Manager\"\n", "delta": "0:00:00.010949", "end": "2023-05-18 13:58:38.366337", "msg": "non-zero return code", "rc": 1, "start": "2023-05-18 13:58:38.355388", "stderr": "", "stderr_lines": [], "stdout": "", "stdout_lines": []}

[1] https://logserver.rdoproject.org/56/156/814b11000ae0869a6e1559bee243ec11f332ff28/github-check/edpm-ansible-github-rdo-integration-centos-8-crc-singlenode-centos-9-external-compute/79eb0cc/controller/controller/pod/deploy-external-dataplane-compute-n4zc8-logs.txt

[bogdando]

/test rdoproject.org/github-check

[openshift-ci]

@bogdando: The specified target(s) for /test were not found.
The following commands are available to trigger required jobs:

• /test pre-commit-test

Use /test all to run all jobs.

In response to this:

/test rdoproject.org/github-check

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.

[bogdando]

check-rdo

[bogdando]

check-rdo

[jpodivin]

Just to clarify. The molecule in podman seems to work with this patch after all https://github.com/openstack-k8s-operators/edpm-ansible/actions/runs/5014407281/jobs/8988616358?pr=156 . Do you still want to replace it with delegated driver?

[bogdando]

Just to clarify. The molecule in podman seems to work with this patch after all https://github.com/openstack-k8s-operators/edpm-ansible/actions/runs/5014407281/jobs/8988616358?pr=156 . Do you still want to replace it with delegated driver?

no, please let's use the mocked swapon command instead. So this is good to go as is!

[bogdando]

PTAL folks

[bogdando]

Ci failure looks unrelated https://github.com/openstack-k8s-operators/edpm-ansible/actions/runs/5199776741/jobs/9377637527#step:6:249

[softwarefactory-project-zuul]

Build failed (check pipeline). For information on how to proceed, see
http://docs.openstack.org/infra/manual/developers.html#automated-testing

https://review.rdoproject.org/zuul/buildset/e8d0d7a07c9e47789c7477ce87cc9d1d

❌ edpm-ansible-molecule-edpm-podman FAILURE in 2m 56s
❌ edpm-ansible-molecule-edpm-module_load FAILURE in 2m 47s
❌ edpm-ansible-molecule-edpm-kernel FAILURE in 2m 45s
✔️ edpm-ansible-crc-podified-edpm-deployment SUCCESS in 50m 08s

[softwarefactory-project-zuul]

Build failed (check pipeline). For information on how to proceed, see
http://docs.openstack.org/infra/manual/developers.html#automated-testing

https://review.rdoproject.org/zuul/buildset/525f7ef19bd64483953cee4716af746b

✔️ edpm-ansible-molecule-edpm-podman SUCCESS in 4m 30s
✔️ edpm-ansible-molecule-edpm-module_load SUCCESS in 3m 52s
✔️ edpm-ansible-molecule-edpm-kernel SUCCESS in 9m 51s
❌ edpm-ansible-crc-podified-edpm-deployment FAILURE in 14m 12s

[bshephar]

I can see this role has some inconsistencies with FQCN usage. But I think we should probably aim for using the FQCN's with these new PR's. We can clean up the rest of the role later.

I see you mentioned in the comments that you're just porting a pre-existing thing to edpm-ansible in order to unblock some work. So I'm happy to merge this one and address the comments I made in a follow up PR if you prefer.

[bogdando]

/recheck

[bogdando]

@kajinamit PTAL

[kajinamit]

I don't know why github doesn't remove my change request.

[openshift-ci]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: bogdando, kajinamit

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:

• OWNERS [kajinamit]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[rabi]

recheck

[softwarefactory-project-zuul]

Build failed (check pipeline). For information on how to proceed, see
http://docs.openstack.org/infra/manual/developers.html#automated-testing

https://review.rdoproject.org/zuul/buildset/eb402c82353a4346945673d71e639f07

✔️ edpm-ansible-molecule-edpm-podman SUCCESS in 8m 43s
✔️ edpm-ansible-molecule-edpm-module_load SUCCESS in 7m 17s
✔️ edpm-ansible-molecule-edpm-kernel SUCCESS in 14m 21s
❌ edpm-ansible-crc-podified-edpm-deployment FAILURE in 20m 38s
❌ edpm-ansible-crc-podified-edpm-baremetal FAILURE in 29m 30s

[rebtoor]

recheck