Project

Profile

Help

Test #4459

closed

Unable to sync 3 SLES Update repositories

Added by bherring about 5 years ago. Updated almost 5 years ago.

Status:
CLOSED - COMPLETE
Priority:
High
Assignee:
Sprint/Milestone:
-
Version:
Platform Release:
Tags:
Pulp 2
Sprint:
Quarter:

Description

Filtering by srpm:

pulp-admin rpm repo create --repo-id testrepo --feed file:///home/vagrant/testrepo/ --skip srpm
pulp-admin rpm repo sync run --force-full --repo-id testrepo

Causes the following error:

Malformed repository: metadata is missing for some packages in filelists.xml and in other.xml

While syncing without the srpm filter seems to work fine.

Test Specific Notes

  • came from downstream
  • create a specific fixture with SRPM duplication

Files

dupe_srpm_repodata.tgz (1.22 MB) dupe_srpm_repodata.tgz daviddavis, 02/08/2019 03:29 PM

Related issues

Copied from RPM Support - Issue #4397: Unable to sync 3 SLES Update repositoriesCLOSED - CURRENTRELEASEdaviddavisActions
Actions #1

Updated by bherring about 5 years ago

  • Copied from Issue #4397: Unable to sync 3 SLES Update repositories added
Actions #2

Updated by bherring about 5 years ago

  • Status changed from MODIFIED to ASSIGNED
Actions #3

Updated by bherring about 5 years ago

  • Assignee set to bherring
Actions #4

Updated by bherring about 5 years ago

  • Status changed from ASSIGNED to MODIFIED
Actions #5

Updated by bherring about 5 years ago

Notes

Additional clarification on fixture verification

[12:58:04]<bherring>Question about https://pulp.plan.io/issues/4397 , when you have a moment
[12:58:30]<bherring>1. Is --force-full really required to demonstrate the failure with `--skip srpm` duplication? I don't think it is
[12:59:31]<bherring>2. When using `--skip srpm` we should be skipping all SRPMs on the sync (or the unit_type issued there). I am assuming that, while nice to have other unit types in a mixed repo, it isn't necessary to prove the functionality of this fix
[12:59:56]<bherring>3. Using `--skip srpm` with a non-duplicated repo should not demonstrate this behavior, correct?
[13:00:06]<bherring>(even before the fix was in-place)
[13:05:29]<daviddavis> 1. no 
[13:05:43]<daviddavis> 2. correct
[13:06:00]<daviddavis> 3. correct
Actions #6

Updated by bherring about 5 years ago

  • Status changed from MODIFIED to ASSIGNED

Added by bherring about 5 years ago

Revision a98c4c3a | View on GitHub

Adding SRPM CLI test for --force-full with --srpm skip

Issue #4397 addressed an issue where duplicate SRPMS would fail on sync when using --srpm skip.

Adding CLI test to ensure that in this scenario the SRPMS are skipped on sync and no SRPMs are on the system.

Tested with duplicate SRPMs to verify the fix.

Added a case with normal SRPM test fixture to ensure basic behavior has not changed.

Related to pulp-fixtures #119 created for this test.

See: https://pulp.plan.io/issues/4459

closes #4459

Actions #7

Updated by bherring about 5 years ago

  • Status changed from ASSIGNED to MODIFIED
Actions #8

Updated by bherring almost 5 years ago

  • Priority changed from Normal to High
Actions #9

Updated by bherring almost 5 years ago

  • Sprint/Milestone set to 2.19.0
Actions #10

Updated by bherring almost 5 years ago

  • Sprint/Milestone deleted (2.19.0)
Actions #11

Updated by bmbouter almost 5 years ago

  • Tags Pulp 2 added
Actions #12

Updated by bherring almost 5 years ago

  • Status changed from MODIFIED to CLOSED - COMPLETE

Also available in: Atom PDF