Pulp: Issueshttps://pulp.plan.io/https://pulp.plan.io/favicon.ico2021-12-08T19:00:00ZPulp
Planio RPM Support - Test #9622 (MODIFIED): Add a repo signed using 'sha' as alias for 'sha1'https://pulp.plan.io/issues/96222021-12-08T19:00:00Zggainey
<p>'sha' support exists in the wild, is the same as 'sha1', and has broken us several times now, Let's make it possible to write tests for it.</p> RPM Support - Test #8972 (CLOSED - COMPLETE): Add test for properly reading and writing the metadatahttps://pulp.plan.io/issues/89722021-06-25T13:46:59Zdalleydalley@redhat.com
<p>Verify that package metadata in == package metadata out</p> RPM Support - Test #8301 (CLOSED - COMPLETE): Pulp to Pulp sync test uses sha1 checksum, fails on...https://pulp.plan.io/issues/83012021-02-23T23:44:11Zdalleydalley@redhat.com
<p>I tried to run the tests and got some failures due to SHA1 checksums</p>
<p><a href="https://github.com/pulp/pulp_rpm/blob/master/pulp_rpm/tests/functional/api/test_pulp_to_pulp.py#L91" class="external">https://github.com/pulp/pulp_rpm/blob/master/pulp_rpm/tests/functional/api/test_pulp_to_pulp.py#L91</a></p>
<p>E HTTP response body: {"non_field_errors":["Checksum must be one of allowed types: ['sha224', 'sha256', 'sha384', 'sha512'].\nYou can adjust these with the 'ALLOWED_CONTENT_CHECKSUMS' setting."]}</p> RPM Support - Test #7571 (CLOSED - COMPLETE): Distribution tree sync and copy testshttps://pulp.plan.io/issues/75712020-09-25T15:35:41Zppicka
<p>Sync test:</p>
<ul>
<li>update main repo of DT</li>
<li>update addon repo of DT</li>
<li>update variant repo DT</li>
<li>remove repository with DT, and orphan cleanup to remove DT</li>
</ul>
<p>Copy test:</p>
<ul>
<li>simple copy of DT</li>
<li>copy DT by content specified</li>
</ul> RPM Support - Test #7422 (CLOSED - COMPLETE): Add tests for export/import of kickstartshttps://pulp.plan.io/issues/74222020-08-29T15:38:27ZdaviddavisRPM Support - Test #7350 (MODIFIED): Test syncing from a repository located on a local diskhttps://pulp.plan.io/issues/73502020-08-19T15:40:07Zdalleydalley@redhat.com
<p>e.g. a repository with a file:// url such as file:///var/lib/pulp/sync_imports/test_repos/zoo/</p> RPM Support - Test #7134 (CLOSED - COMPLETE): Test against CDNhttps://pulp.plan.io/issues/71342020-07-14T13:59:20Zppicka
<p>When core issue <a href="https://pulp.plan.io/issues/6735" class="external">https://pulp.plan.io/issues/6735</a> is solved, lets add the test to sync against the CDN both baseos and appstream repositories.</p>
<p>This will test same repo meta data with different filename as same productid is shipped intentionally with both of mentioned repositories.</p>
<p>Also this test is testing using a certificates which are needed for CDN.</p> RPM Support - Test #6789 (CLOSED - COMPLETE): Test the copy of RPM with its strict dependencieshttps://pulp.plan.io/issues/67892020-05-21T14:25:56Zppicka
<p>Test copy RPM package content with depsolving=True which should copy package and all its dependency.</p>
<p>test case 1</p>
<ul>
<li>create repository and populate it with content</li>
<li>create new repository and leave it blank</li>
<li>use copy pointing on one package with depsolving=True</li>
<li>check if new repository contains package above and its dependencies</li>
</ul>
<p>test case 2</p>
<ul>
<li>create repositoryA and populate it with content</li>
<li>create new repositoryB and populate it with some content which can be satisfactory for package from first repoA</li>
<li>use copy pointing on one package with depsolving=True</li>
<li>check that package has been copied and only part of the dependencies that missed in repoB were copied</li>
<li>ensure repoB has the package with all it's needed deps</li>
</ul> RPM Support - Test #6786 (CLOSED - COMPLETE): Test sync in mirror modehttps://pulp.plan.io/issues/67862020-05-21T08:54:57Zppicka
<p>Test sync in mirror mode</p>
<p>Test case steps:</p>
<ul>
<li>create repo, remote and sync it</li>
<li>create another remote (with some same and some different content)</li>
<li>re-sync with new remote in mirror mode</li>
<li>check all content from first remote is removed</li>
<li>check all new content is added to repository</li>
</ul> RPM Support - Test #6777 (CLOSED - COMPLETE): Test sync in additive modehttps://pulp.plan.io/issues/67772020-05-20T07:48:58Zppicka
<p>Test sync additive mode</p>
<ol>
<li>create repo, remote and sync</li>
<li>create remote with extended set of content and with the default setting for mirror option (mirror=False) and re-sync</li>
<li>check if extended set was added to repository</li>
</ol> RPM Support - Test #6536 (CLOSED - COMPLETE): Test published directory structurehttps://pulp.plan.io/issues/65362020-04-21T10:38:04Zipanova@redhat.comipanova@redhat.com
<p>Testcase 1:</p>
<ol>
<li>Publish and distribute an rpm repo that contains modules</li>
<li>Ensure that in the root directory there is nothing else but repodata/ and Packages/</li>
</ol>
<p>Testcase 2:</p>
<ol>
<li>Publish and distribute an rpm repo that contains a kickstart tree</li>
<li>Ensure that in the root directory there is nothing else but repodata/ ,Packages/ , .treeinfo and images related directory</li>
</ol> RPM Support - Test #6503 (CLOSED - COMPLETE): assert checksums for on_demand publicationhttps://pulp.plan.io/issues/65032020-04-15T17:17:25Zfao89
<p>we need to test publication behavior when setting: metadata_checksum_type and package_checksum_type for on demand publication</p>
<p>Note: mixing on_demand and immediate policies on the same test case could lead to intermittent failures</p> RPM Support - Test #6313 (CLOSED - COMPLETE): Write tests to check sync optimizationhttps://pulp.plan.io/issues/63132020-03-10T01:50:20ZCodeHeeler
<p>Scenarios to test:</p>
<ol>
<li>First scenario
<ol>
<li>run sync (following our workflow docs) once,</li>
<li>there should be no optimization progress report on the task</li>
</ol>
</li>
<li>Second scenario
<ol>
<li>run sync a second time with no other changes,</li>
<li>there should be an optimization progress report now on the task</li>
</ol>
</li>
<li>Third scenario
<ol>
<li>run sync a third time adding the flag 'optimize=False',</li>
<li>there should be no optimization progress report now on the task</li>
</ol>
</li>
<li>Fourth scenario
<ol>
<li>create a new repo version,</li>
<li>run sync again (remove 'optimize=False' flag),</li>
<li>there should be no optimization progress report</li>
</ol>
</li>
<li>Fifth scenario
<ol>
<li>change remote policy to immediate,</li>
<li>run sync again,</li>
<li>there should be no optimization progress report</li>
</ol>
</li>
<li>Sixth scenario
<ol>
<li>ideally, we need a revision number change to the repomd in our fixtures (do we have a setup for this already? Is this within the scope of our automated tests)</li>
<li>run sync</li>
<li>there should be no optimization progress report now on the task</li>
</ol>
</li>
<li>Seventh scenario
<ol>
<li>run sync with no changes</li>
<li>there should be an optimization progress report now on the task</li>
</ol>
</li>
<li>Eighth scenario
<ol>
<li>run sync using the same repo, but a different remote</li>
<li>there should be no optimization progress report now on the task</li>
<li>rerun this sync</li>
<li>there should be an optimization progress report now on the task</li>
</ol>
</li>
</ol>
<p>All of these syncs should be run in succession without cleanup in between as a first time sync should always leave no optimization progress report. If for some reason you wanted to separate the tests and needed cleanup in between, every test scenario other than <a class="issue tracker-3 status-11 priority-6 priority-default closed child" title="Story: As a user, I can have Pulp attempt use auto_retry application wide using the 'unsafe_autoretry' p... (CLOSED - CURRENTRELEASE)" href="https://pulp.plan.io/issues/1">#1</a> will need to have sync run twice, once at the beginning, and once when specified after the stated changes for the test.</p> RPM Support - Test #6015 (CLOSED - COMPLETE): Fix the "SyncMutatedUpdateRecordTestCase" testhttps://pulp.plan.io/issues/60152020-01-17T15:37:52Zdalleydalley@redhat.com
<p>This test was written long before the feature was implemented and has been disabled since. The feature is now implemented but when the test is enabled, it fails with the following exception thrown:</p>
<pre><code>pulp_rpm.app.exceptions.AdvisoryConflict: Incoming and existing advisories have the same id and timestamp but different and intersecting package lists. At least one of them is wrong. Advisory id: RHEA-2012:0055
</code></pre>
<p>This most likely means that Pulp 3 is more conservative than the test is designed for, and rejects the advisory in our test fixture because both versions of the advisory have identical timestamps but different package lists.</p>
<p>We need to find (or create) a fixture advisory where the two versions of the advisory have differing timestamps, so that the package lists can be successfully merged.</p> RPM Support - Test #5506 (CLOSED - COMPLETE): Create a smoke test for RPM plugin sync performancehttps://pulp.plan.io/issues/55062019-09-26T16:12:37Zdalleydalley@redhat.com
<p>We will be putting much effort and investigation on performance in the coming days. To facilitate this investigation, it would be useful to have some performance smoke tests which can quickly allow a developer to gauge whether something significant has changed in this regard, or to facilitate the investigation of performance generally.</p>
<p>It is not necessary for the tests to be rigidly reproducible over a long period of time. They are intended to be used more as a developer tool and not to run as part of the CI pipeline.</p>
<ul>
<li>Sync CentOS 7 repository, on_demand</li>
<li>Sync CentOS 8 Appstream repository, on_demand</li>
<li>Sync CentOS 8 BaseOS repository, on_demand</li>
</ul>
<p>In keeping with the goals mentioned, they shouldn't run in the CI pipeline in the general case, although it may be useful to have them enabled through a particular flag to run if needed.</p>