podman-3.0.0-0.11.el9

- update to https://github.com/containers/dnsname/releases/tag/v1.1.1

Signed-off-by: Jindrich Novy <jnovy@redhat.com>
This commit is contained in:
Jindrich Novy 2020-12-10 10:00:54 +01:00
parent a782eca329
commit 97e58ded35
12 changed files with 69 additions and 221 deletions

View File

@ -1,14 +1,7 @@
# recipients: jnovy, lsm5, santiago
--- !Policy
product_versions:
- fedora-*
decision_context: bodhi_update_push_stable
subject_type: koji_build
- rhel-8
decision_context: osci_compose_gate
rules:
- !PassingTestCaseRule {test_case_name: fedora-ci.koji-build.tier0.functional}
--- !Policy
product_versions:
- fedora-*
decision_context: bodhi_update_push_testing
subject_type: koji_build
rules:
- !PassingTestCaseRule {test_case_name: fedora-ci.koji-build.tier0.functional}
- !PassingTestCaseRule {test_case_name: osci.brew-build.tier0.functional}

View File

@ -17,14 +17,14 @@ go build -buildmode pie -compiler gc -tags="rpm_crashtraceback libtrust_openssl
%global import_path github.com/containers/podman
%global branch master
%global commit0 dd295f297b6dd51d22c64c75f4ef4f80f953bbde
%global commit0 9216be2008696bc9f0bc26686be8becae3d12bfc
%global shortcommit0 %(c=%{commit0}; echo ${c:0:7})
%global cataver 0.1.5
%global dnsnamever 1.0.0
%global dnsnamever 1.1.1
Name: podman
Version: 3.0.0
Release: 0.10%{?dist}
Release: 0.11%{?dist}
Summary: Manage Pods, Containers and Container Images
License: ASL 2.0 and GPLv3+
URL: https://%{name}.io/
@ -339,6 +339,9 @@ exit 0
%{_datadir}/%{name}/test
%changelog
* Thu Dec 10 2020 Jindrich Novy <jnovy@redhat.com> - 3.0.0-0.11
- update to https://github.com/containers/dnsname/releases/tag/v1.1.1
* Wed Dec 09 2020 Jindrich Novy <jnovy@redhat.com> - 3.0.0-0.10
- update to the latest content of https://github.com/containers/podman/tree/master
(https://github.com/containers/podman/commit/dd295f2)

View File

@ -1,3 +1,3 @@
SHA512 (v0.1.5.tar.gz) = 75387cf9c611222ef0afc9f688cefb3e4a4ec388cfa9d2329ee7860cb2ad83afa9311f53c1047f92f7adde27b7eb07f5501ee602334e610a3227104d8894486b
SHA512 (v1.0.0.tar.gz) = 6eb96939847dcb4bb6fcfbf2e9e1555c1e79b0b77350b2669d4e517decdda1784fbe68536419fd5ba81bfc6b47c57665487c62e5fde7e0d2b129f0161f421b83
SHA512 (master-dd295f2.tar.gz) = ae3a06c84db158294d3f3b191fe57b63c3d0914b8707b2d5f64a40d011653eea0e5b37ddaacb356f3e28bbb75e640322988a101f292c6d625c8a289d551c8968
SHA512 (master-9216be2.tar.gz) = bfe97adaa3aa1053a116a3bc30f6603b8e4a752182890f558cf2ea4f3c788e0422514203bca5292ec7c880090eb94c1625b3187d537859f46a0d64a2bd0c3270
SHA512 (v1.1.1.tar.gz) = 4ffaf73f41d96a3d36a15cfefdad989579bf3e187c0da9df35b16b43f77f8694c01a7f006b89b97070ea9c0d924778a416c30a96b18fa125b74186beb800c54f

View File

@ -1,25 +0,0 @@
I'm sorry. The playbooks here are a much-too-complicated way of saying:
- test podman (root and rootless) under cgroups v2
- reboot into cgroups v1
- repeat the same podman tests
We can't use standard-test-basic any more because, tl;dr, that has to
be the last stanza in the playbook and it doesn't offer any mechanism
for running a reboot in the middle of tests. (I actually found a way
but it was even uglier than this approach).
The starting point is tests.yml . From there:
tests.yml
\- test_podman.yml
|- roles/rootless_user_ready/
\- test_podman_cgroups_vn.yml (runs twice: cgroups v2, v1)
|- roles/set_cgroups/
\- roles/run_bats_tests/ (runs tests: root, rootless)
Principal result is the file 'artifacts/test.log'. It will contain
one line for each test run, format will be '(PASS|FAIL|ERROR) <test name>'
For each completed test there will also be a 'test.<name>.bats.log'
containing some setup blurbs (RPMs, environment) and the full BATS log.

View File

@ -1,36 +0,0 @@
---
# Copied from standard-test-basic
# ...and, 2020-05-13, updated, looks like they changed the whole thing around
- name: Check the results
local_action:
module: shell
cmd: |
log="{{ artifacts }}/test.log"
if [ ! -f "$log" ]; then
echo ERROR
echo "Test results not found." 1>&2
elif grep ^ERROR "$log" 1>&2; then
echo ERROR
elif grep ^FAIL "$log" 1>&2; then
echo FAIL
elif grep -q ^PASS "$log"; then
echo PASS
else
echo ERROR
echo "No test results found." 1>&2
fi
register: test_results
- name: Set role result
set_fact:
role_result: "{{ test_results.stdout }}"
role_message: "{{ test_results.stderr|d('test execution error.') }}"
- name: display results
vars:
msg: |
role_result: {{ role_result|d('Undefined') }}
{{ role_message|d('[No error messages found]') }}
debug:
msg: "{{ msg.split('\n') }}"
failed_when: role_message|d("") != ""

View File

@ -0,0 +1,12 @@
---
# Sigh; RHEL8 doesn't have BATS
- name: bats | fetch and unpack tarball
unarchive:
src: https://github.com/bats-core/bats-core/archive/v1.1.0.tar.gz
dest: /root
remote_src: true
- name: bats | install
command: ./install.sh /usr/local
args:
chdir: /root/bats-core-1.1.0

View File

@ -0,0 +1,7 @@
---
- name: create nonroot user
user:
name: testuser
shell: /bin/bash
- name: enable linger
command: loginctl enable-linger testuser

View File

@ -1,6 +0,0 @@
---
- name: make sure rootless account exists
user: name={{ rootless_user }}
- name: rootless account | enable linger
shell: loginctl enable-linger {{ rootless_user }}

View File

@ -0,0 +1,38 @@
#!/bin/bash
#
# setup and teardown helpers for buildah test
#
function setup() {
REGISTRY_FQIN=docker.io/library/registry:2
AUTHDIR=/tmp/buildah-tests-auth.$$
mkdir -p $AUTHDIR
CERT=$AUTHDIR/domain.crt
if [ ! -e $CERT ]; then
openssl req -newkey rsa:4096 -nodes -sha256 \
-keyout $AUTHDIR/domain.key -x509 -days 2 \
-out $AUTHDIR/domain.crt \
-subj "/C=US/ST=Foo/L=Bar/O=Red Hat, Inc./CN=localhost"
fi
if [ ! -e $AUTHDIR/htpasswd ]; then
podman run --rm --entrypoint htpasswd $REGISTRY_FQIN \
-Bbn testuser testpassword > $AUTHDIR/htpasswd
fi
podman run -d -p 5000:5000 \
--name registry \
-v $AUTHDIR:/auth:Z \
-e "REGISTRY_AUTH=htpasswd" \
-e "REGISTRY_AUTH_HTPASSWD_REALM=Registry Realm" \
-e REGISTRY_AUTH_HTPASSWD_PATH=/auth/htpasswd \
-e REGISTRY_HTTP_TLS_CERTIFICATE=/auth/domain.crt \
-e REGISTRY_HTTP_TLS_KEY=/auth/domain.key \
$REGISTRY_FQIN
}
function teardown() {
podman rm -f registry
}

View File

@ -1,75 +0,0 @@
---
# Check the CURRENT cgroup level; we get this from /proc/cmdline
- name: check current kernel options
shell: fgrep systemd.unified_cgroup_hierarchy=0 /proc/cmdline
register: result
ignore_errors: true
- name: determine current cgroups | assume v2
set_fact: current_cgroups=2
- name: determine current cgroups | looks like v1
set_fact: current_cgroups=1
when: result is succeeded
- debug:
msg: "want: v{{ want_cgroups }} actual: v{{ current_cgroups }}"
- name: grubenv, pre-edit, cat
shell: cat /boot/grub2/grubenv
register: grubenv
- name: grubenv, pre-edit, show
debug:
msg: "{{ grubenv.stdout_lines }}"
# Update grubenv file to reflect the desired cgroup level
- name: remove cgroup option from kernel flags
shell:
cmd: sed -i -e "s/^\(kernelopts=.*\)systemd\.unified_cgroup_hierarchy=.\(.*\)/\1 \2/" /boot/grub2/grubenv
warn: false
- name: add it with the desired value
shell:
cmd: sed -i -e "s/^\(kernelopts=.*\)/\1 systemd.unified_cgroup_hierarchy=0/" /boot/grub2/grubenv
warn: false
when: want_cgroups == 1
- name: grubenv, post-edit, cat
shell: cat /boot/grub2/grubenv
register: grubenv
- name: grubenv, post-edit, show
debug:
msg: "post: {{ grubenv.stdout_lines }}"
# If want != have, reboot
- name: reboot and wait
block:
- name: reboot
reboot:
reboot_timeout: 900
ignore_errors: yes
- name: wait and reconnect
wait_for_connection:
timeout: 900
when: want_cgroups|int != current_cgroups|int
- set_fact:
expected_fstype:
- none
- tmpfs
- cgroup2fs
- name: confirm cgroups setting
shell: stat -f -c "%T" /sys/fs/cgroup
register: fstype
- debug:
msg: "stat(/sys/fs/cgroup) = {{ fstype.stdout }}"
- name: system cgroups is the expected type
assert:
that:
- fstype.stdout == expected_fstype[want_cgroups|int]
fail_msg: "stat(/sys/fs/cgroup) = {{ fstype.stdout }} (expected {{ expected_fstype[want_cgroups|int] }})"

View File

@ -1,44 +0,0 @@
---
- hosts: localhost
tags:
- classic
- container
vars:
- artifacts: ./artifacts
rootless_user: testuser
roles:
- role: rootless_user_ready
tasks:
# At the start of a run, clean up state. Useful for test reruns.
- name: local artifacts directory exists
local_action: file path="{{ artifacts }}" state=directory
- name: remove stale log files
local_action: shell rm -f {{ artifacts }}/test*.log
- name: clear test results (test.log)
local_action: command truncate --size=0 {{ artifacts }}/test.log
- name: clear test results (results.yml)
local_action: copy content="results:\n" dest={{ artifacts }}/results.yml
# These are the actual tests: set cgroups vN, then run root/rootless tests.
#
# FIXME FIXME FIXME: 2020-05-21: 'loop' should be '2, 1' but there's some
# nightmarish bug in CI wherein reboots hang forever. There's a bug open[1]
# but it seems dead. Without a working reboot, there's no way to test v1.
# [1] https://redhat.service-now.com/surl.do?n=PNT0808530
# I'm leaving this as a 'loop' in (foolish? vain?) hope that the bug will
# be fixed. Let's revisit this after, say, 2020-08. If the bug persists
# then let's just revert the entire cgroups v1 change, and go back to
# using standard-test-basic.
- name: set cgroups and run podman tests
include_tasks: test_podman_cgroups_vn.yml
loop: [ 2 ]
loop_control:
loop_var: want_cgroups
# Postprocessing: check for FAIL or ERROR in any test, exit 1 if so
- name: check results
include_tasks: check_results.yml

View File

@ -1,19 +0,0 @@
---
# Requires: 'want_cgroups' variable set to 1 or 2
- include_role:
name: set_cgroups
- include_role:
name: run_bats_tests
vars:
tests:
# Yes, this is horrible duplication, but trying to refactor in ansible
# yields even more horrible unreadable code. This is the lesser evil.
- name: podman root cgroupsv{{ want_cgroups }}
package: podman
environment:
PODMAN: /usr/bin/podman
- name: podman rootless cgroupsv{{ want_cgroups }}
package: podman
environment:
PODMAN: /usr/bin/podman
become: true