-
Notifications
You must be signed in to change notification settings - Fork 139
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[openSUSE] use zypper for rockstor package install #2071
Comments
I have proof of concept code for this issue in progress. |
Pending pull request nearing final testing. |
phillxnet
added a commit
to phillxnet/rockstor-core
that referenced
this issue
Nov 8, 2019
Includes - Contextual zypper calls replace equivalent yum ones when on openSUSE. - Refactoring: including function/procedure renaming. - Include Black code syntactic reformatting. - Fix prior hard wired x86-64 file name parsing. - Add rpm import of ROCKSTOR-GPG-KEY on openSUSE systems. - Add zypper addrepo/removerepo function along side existing yum counterparts for rockstor stable and testing channels. - Use zypper up/dup respectively for general Leap/Tumbleweed update/dist-upgrade. - Use --no-recommends to zypper up/dup commands to keep footprint small. - Add user text for no version/changelog available on openSUSE. - Move prior AT initiated delay to within external script and reduce from 1 minute to 10 seconds. - Move string formatting within pkg_mgmt.py from 'c' style to string.format(). - Avoid removing RPM package managed *.pyc files from ./eggs dir prior to rockstor rpm package install/upgrade. - Make user facing update dialog/tooltip text non yum specific. - Begin pkg management unit test coverage.
phillxnet
added a commit
that referenced
this issue
Nov 13, 2019
…install [openSUSE] use zypper for package installs. Fixes #2071
phillxnet
added a commit
to phillxnet/rockstor-core
that referenced
this issue
Apr 19, 2020
All currently failing unit tests on our to-be-legacy but still reference CentOS source build are remarked out in order that we can more readily use the existing tests as indicators of disparity between our ditro bases. In almost all cases the fail/error message of the test is noted in TODO's. The vast majority failures are suspected as being caused by a lack of maintenance within the tests and several major changes such as disk/share/pool/network API changes Includes: - Remark out 2 sftp sub tests, broken/out of sync with core code. - Remark out test_btrfs_disk_import: 500 not 200. - Fix broken test_commands.py CommandTests - core refactor bit rot. see: "[openSUSE] use zypper for package installs. Fixes rockstor#2071" (pr rockstor#2081) in 3.9.2-51. The unrelated get_pool_info() breakage was not tracked down. - Update patch targets due to core refactoring. In pr: "Add unit testing for core network functions" rockstor#2045 at least a couple of functions were renamed. Updated these in our network API test setup. - Remark out all network API tests as this API changed long ago. - Remark out many whole and sub-parts of the nfs tests. Many of the failures here are likely down to disk / share API changes after these tests were written. Specifically Shares API in this case as there are many (noted in TODO: comments): "share instance ... does not exist." which suggests a failure of the mocking arrangement. - Remark out test_share_acl as currently non working and need work. - An attempt was made to fix the mocking required to have this test work but it was not a success. Minor minification were left in place never-the-less. - Remark out test within test_oauth_apply - Needs updating. - Update to account for refactoring in tested & add new needed mocks. - Since this test was update there have been a few pool model changes. The added mocks mostly move this test back to where it was prior to the pool management code changes. - Remove overly ambitious/broad test_raid_migration - failing. - Fix missing space in expected error message test_raid6_crud. - Remark out currently unworkable sub tests within test_raid6_crud. - Our pool / disk member mocking is currently insufficient and so a range of tests now fail. Remove for to use what we can. - Remark out currently unworkable sub tests within test_raid1_crud. - Our pool / disk member mocking is currently insufficient and so a range of tests now fail. Remove for to use what we can. - Remark out several sub tests within test_samba.py as need fixing. - Remark out a sub test within test_group as need fixing or OOD. - Remark out a sub test within test_login as non representative. - Remark out a sub test within test_shares as incorrectly triggered. - Remark out all but on test in test_oauth as test seem broken. - Remark out one more test in test_snapshots as known to be duff. - Remark out several more test in test_user as known to be duff.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Experimentation has indicated that although yum is mostly functional for our purpose in our openSUSE variant their are complications re reported conflicts of rockstor rpm dependencies that are not similarly reported by zypper. So we should, at least for install only initially, use zypper. Currently it is though that our use of YUM for reporting / changelog etc is working as intended so we need only change the actual install side of our Web-UI initiated 'rockstor' package management for now.
It is similarly very likely better to move to zypper (on openSUSE) for our regular non rockstor package updates also. The same code manages both so this issue may be able to address both also.
The text was updated successfully, but these errors were encountered: