Activity log for bug #2055081

Date Who What changed Old value New value Message
2024-02-26 19:25:09 James Falcon bug added bug
2024-02-26 19:26:04 James Falcon description [ Impact ] In cloud-init 23.4, specifying a datasource via the SMBIOS is broken. This is due to a commit that fixed another bug of cloud-init assuming a datasource list consisting of one datasource and 'None' doesn't need to check for the existence of that datasource. The fix here is to ensure we're always checking for the SMBIOS information, whereas in the past we didn't need to because of the automatic datasource assumption. [ Test Plan ] An integration test is being added upstream as part of https://github.com/canonical/cloud-init/pull/4949 . This test currently fails if run against 23.4. If it passes, it demonstrates that the fix works. The test specifies a datasource seed directory via SMBIOS and ensures that the userdata contained within that seed ran correctly. It can be run manually using: tox -e integration-tests -- tests/integration_tests/datasources/test_nocloud.py::test_smbios_seed [ Where problems could occur ] Since this code sits at the interaction of datasources discovered and found via ds-identify and python, along with bits of fallback code to dmidecode, we could have missed an interaction here that could possibly mis-identify the NoCloud datasource in other ways. [ Other Info ] Documentation of specifying NoCloud datasource via SMBIOS : https://cloudinit.readthedocs.io/en/latest/reference/datasources/nocloud.html#method-2-local-filesystem-kernel-commandline-or-smbios PR with change that broke the cloud-init behavior: https://github.com/canonical/cloud-init/pull/4426 Upstream PR fixing the behavior: https://github.com/canonical/cloud-init/pull/4949 [ Impact ] In cloud-init 23.4, specifying a datasource via the SMBIOS is broken. This is due to a commit that fixed another bug of cloud-init assuming a datasource list consisting of one datasource and 'None' doesn't need to check for the existence of that datasource. The fix here is to ensure we're always checking for the SMBIOS information, whereas in the past we didn't need to because of the automatic datasource assumption. [ Test Plan ] An integration test is being added upstream as part of https://github.com/canonical/cloud-init/pull/4949 . This test currently fails if run against 23.4. If it passes, it demonstrates that the fix works. The test specifies a datasource seed directory via SMBIOS and ensures that the userdata contained within that seed ran correctly. It can be run manually using: tox -e integration-tests -- tests/integration_tests/datasources/test_nocloud.py::test_smbios_seed [ Where problems could occur ] Since this code sits at the interaction of datasources discovered and found via ds-identify and python, along with bits of fallback code to dmidecode, we could have missed an interaction here that could possibly mis-identify the NoCloud datasource in other ways. [ Other Info ] Documentation of specifying NoCloud datasource via SMBIOS : https://cloudinit.readthedocs.io/en/latest/reference/datasources/nocloud.html#method-2-local-filesystem-kernel-commandline-or-smbios Upstream PR with change that broke the cloud-init behavior: https://github.com/canonical/cloud-init/pull/4426 Upstream PR fixing the behavior: https://github.com/canonical/cloud-init/pull/4949
2024-02-26 19:26:29 James Falcon nominated for series Ubuntu Jammy
2024-02-26 19:26:29 James Falcon bug task added cloud-init (Ubuntu Jammy)
2024-02-26 19:26:29 James Falcon nominated for series Ubuntu Mantic
2024-02-26 19:26:29 James Falcon bug task added cloud-init (Ubuntu Mantic)
2024-02-26 19:26:29 James Falcon nominated for series Ubuntu Focal
2024-02-26 19:26:29 James Falcon bug task added cloud-init (Ubuntu Focal)
2024-02-26 19:28:11 James Falcon tags regression-update
2024-02-27 15:10:36 James Falcon description [ Impact ] In cloud-init 23.4, specifying a datasource via the SMBIOS is broken. This is due to a commit that fixed another bug of cloud-init assuming a datasource list consisting of one datasource and 'None' doesn't need to check for the existence of that datasource. The fix here is to ensure we're always checking for the SMBIOS information, whereas in the past we didn't need to because of the automatic datasource assumption. [ Test Plan ] An integration test is being added upstream as part of https://github.com/canonical/cloud-init/pull/4949 . This test currently fails if run against 23.4. If it passes, it demonstrates that the fix works. The test specifies a datasource seed directory via SMBIOS and ensures that the userdata contained within that seed ran correctly. It can be run manually using: tox -e integration-tests -- tests/integration_tests/datasources/test_nocloud.py::test_smbios_seed [ Where problems could occur ] Since this code sits at the interaction of datasources discovered and found via ds-identify and python, along with bits of fallback code to dmidecode, we could have missed an interaction here that could possibly mis-identify the NoCloud datasource in other ways. [ Other Info ] Documentation of specifying NoCloud datasource via SMBIOS : https://cloudinit.readthedocs.io/en/latest/reference/datasources/nocloud.html#method-2-local-filesystem-kernel-commandline-or-smbios Upstream PR with change that broke the cloud-init behavior: https://github.com/canonical/cloud-init/pull/4426 Upstream PR fixing the behavior: https://github.com/canonical/cloud-init/pull/4949 [ Impact ] In cloud-init 23.4, specifying a datasource via the SMBIOS is broken. This is due to a commit that fixed another bug of cloud-init assuming a datasource list consisting of one datasource and 'None' doesn't need to check for the existence of that datasource. The fix here is to ensure we're always checking for the SMBIOS information, whereas in the past we didn't need to because of the automatic datasource assumption. [ Test Plan ] An integration test is being added upstream as part of https://github.com/canonical/cloud-init/pull/4949 . This test currently fails if run against 23.4. If it passes, it demonstrates that the fix works. The test specifies an http datasource seed via SMBIOS and ensures that the userdata contained within that seed ran correctly. It can be run manually using: tox -e integration-tests -- tests/integration_tests/datasources/test_nocloud.py::test_smbios_seed_network [ Where problems could occur ] Since this code sits at the interaction of datasources discovered and found via ds-identify and python, along with bits of fallback code to dmidecode, we could have missed an interaction here that could possibly mis-identify the NoCloud datasource in other ways. [ Other Info ] Documentation of specifying NoCloud datasource via SMBIOS : https://cloudinit.readthedocs.io/en/latest/reference/datasources/nocloud.html#method-2-local-filesystem-kernel-commandline-or-smbios Upstream PR with change that broke the cloud-init behavior: https://github.com/canonical/cloud-init/pull/4426 Upstream PR fixing the behavior: https://github.com/canonical/cloud-init/pull/4949
2024-02-27 15:11:10 James Falcon description [ Impact ] In cloud-init 23.4, specifying a datasource via the SMBIOS is broken. This is due to a commit that fixed another bug of cloud-init assuming a datasource list consisting of one datasource and 'None' doesn't need to check for the existence of that datasource. The fix here is to ensure we're always checking for the SMBIOS information, whereas in the past we didn't need to because of the automatic datasource assumption. [ Test Plan ] An integration test is being added upstream as part of https://github.com/canonical/cloud-init/pull/4949 . This test currently fails if run against 23.4. If it passes, it demonstrates that the fix works. The test specifies an http datasource seed via SMBIOS and ensures that the userdata contained within that seed ran correctly. It can be run manually using: tox -e integration-tests -- tests/integration_tests/datasources/test_nocloud.py::test_smbios_seed_network [ Where problems could occur ] Since this code sits at the interaction of datasources discovered and found via ds-identify and python, along with bits of fallback code to dmidecode, we could have missed an interaction here that could possibly mis-identify the NoCloud datasource in other ways. [ Other Info ] Documentation of specifying NoCloud datasource via SMBIOS : https://cloudinit.readthedocs.io/en/latest/reference/datasources/nocloud.html#method-2-local-filesystem-kernel-commandline-or-smbios Upstream PR with change that broke the cloud-init behavior: https://github.com/canonical/cloud-init/pull/4426 Upstream PR fixing the behavior: https://github.com/canonical/cloud-init/pull/4949 [ Impact ] In cloud-init 23.4, specifying a datasource via the SMBIOS is broken. This is due to a commit that fixed another bug of cloud-init assuming a datasource list consisting of one datasource and 'None' doesn't need to check for the existence of that datasource. The fix here is to ensure we're always checking for the SMBIOS information, whereas in the past we didn't need to because of the automatic datasource assumption. [ Test Plan ] An integration test has been added upstream as part of https://github.com/canonical/cloud-init/pull/4949 . This test fails if run against 23.4. If it passes, it demonstrates that the fix works. The test specifies an http datasource seed via SMBIOS and ensures that the userdata contained within that seed ran correctly. It can be run manually using: tox -e integration-tests -- tests/integration_tests/datasources/test_nocloud.py::test_smbios_seed_network [ Where problems could occur ] Since this code sits at the interaction of datasources discovered and found via ds-identify and python, along with bits of fallback code to dmidecode, we could have missed an interaction here that could possibly mis-identify the NoCloud datasource in other ways. [ Other Info ] Documentation of specifying NoCloud datasource via SMBIOS : https://cloudinit.readthedocs.io/en/latest/reference/datasources/nocloud.html#method-2-local-filesystem-kernel-commandline-or-smbios Upstream PR with change that broke the cloud-init behavior: https://github.com/canonical/cloud-init/pull/4426 Upstream PR fixing the behavior: https://github.com/canonical/cloud-init/pull/4949
2024-02-27 15:13:10 Chad Smith cloud-init (Ubuntu): status New In Progress
2024-02-27 18:36:36 Chad Smith cloud-init (Ubuntu): status In Progress Fix Committed
2024-02-27 18:38:23 Chad Smith cloud-init (Ubuntu): assignee James Falcon (falcojr)
2024-02-27 18:38:28 Chad Smith cloud-init (Ubuntu): importance Undecided High
2024-02-27 18:38:35 Chad Smith cloud-init (Ubuntu Focal): status New In Progress
2024-02-27 18:38:37 Chad Smith cloud-init (Ubuntu Jammy): status New In Progress
2024-02-27 18:38:39 Chad Smith cloud-init (Ubuntu Mantic): status New In Progress
2024-02-28 00:14:29 Chris Halse Rogers cloud-init (Ubuntu Mantic): status In Progress Fix Committed
2024-02-28 00:14:30 Chris Halse Rogers bug added subscriber Ubuntu Stable Release Updates Team
2024-02-28 00:14:41 Chris Halse Rogers bug added subscriber SRU Verification
2024-02-28 00:14:44 Chris Halse Rogers tags regression-update regression-update verification-needed verification-needed-mantic
2024-02-28 00:17:53 Chris Halse Rogers cloud-init (Ubuntu Jammy): status In Progress Fix Committed
2024-02-28 00:17:58 Chris Halse Rogers tags regression-update verification-needed verification-needed-mantic regression-update verification-needed verification-needed-jammy verification-needed-mantic
2024-02-28 00:26:41 Chris Halse Rogers cloud-init (Ubuntu Focal): status In Progress Fix Committed
2024-02-28 00:26:47 Chris Halse Rogers tags regression-update verification-needed verification-needed-jammy verification-needed-mantic regression-update verification-needed verification-needed-focal verification-needed-jammy verification-needed-mantic
2024-02-28 02:24:22 Launchpad Janitor cloud-init (Ubuntu): status Fix Committed Fix Released
2024-02-28 20:00:21 James Falcon attachment added 23.4.4.tar.gz https://bugs.launchpad.net/ubuntu/+source/cloud-init/+bug/2055081/+attachment/5750421/+files/23.4.4.tar.gz
2024-02-28 20:01:39 James Falcon tags regression-update verification-needed verification-needed-focal verification-needed-jammy verification-needed-mantic regression-update verification-done verification-done-focal verification-done-jammy verification-done-mantic
2024-02-28 20:07:43 James Falcon description [ Impact ] In cloud-init 23.4, specifying a datasource via the SMBIOS is broken. This is due to a commit that fixed another bug of cloud-init assuming a datasource list consisting of one datasource and 'None' doesn't need to check for the existence of that datasource. The fix here is to ensure we're always checking for the SMBIOS information, whereas in the past we didn't need to because of the automatic datasource assumption. [ Test Plan ] An integration test has been added upstream as part of https://github.com/canonical/cloud-init/pull/4949 . This test fails if run against 23.4. If it passes, it demonstrates that the fix works. The test specifies an http datasource seed via SMBIOS and ensures that the userdata contained within that seed ran correctly. It can be run manually using: tox -e integration-tests -- tests/integration_tests/datasources/test_nocloud.py::test_smbios_seed_network [ Where problems could occur ] Since this code sits at the interaction of datasources discovered and found via ds-identify and python, along with bits of fallback code to dmidecode, we could have missed an interaction here that could possibly mis-identify the NoCloud datasource in other ways. [ Other Info ] Documentation of specifying NoCloud datasource via SMBIOS : https://cloudinit.readthedocs.io/en/latest/reference/datasources/nocloud.html#method-2-local-filesystem-kernel-commandline-or-smbios Upstream PR with change that broke the cloud-init behavior: https://github.com/canonical/cloud-init/pull/4426 Upstream PR fixing the behavior: https://github.com/canonical/cloud-init/pull/4949 [ Impact ] In cloud-init 23.4, specifying a datasource via the SMBIOS is broken. This is due to a commit that fixed another bug of cloud-init assuming a datasource list consisting of one datasource and 'None' doesn't need to check for the existence of that datasource. The fix here is to ensure we're always checking for the SMBIOS information, whereas in the past we didn't need to because of the automatic datasource assumption. [ Test Plan ] An integration test has been added upstream as part of https://github.com/canonical/cloud-init/pull/4949 . This test fails if run against 23.4. If it passes, it demonstrates that the fix works. The test specifies an http datasource seed via SMBIOS and ensures that the userdata contained within that seed ran correctly. It can be run manually using: tox -e integration-tests -- tests/integration_tests/datasources/test_nocloud.py::test_smbios_seed_network [ Where problems could occur ] Since this code sits at the interaction of datasources discovered and found via ds-identify and python, along with bits of fallback code to dmidecode, we could have missed an interaction here that could possibly mis-identify the NoCloud datasource in other ways. [ Other Info ] Upstream bug: https://github.com/canonical/cloud-init/issues/4945 Upstream dup: https://github.com/canonical/cloud-init/issues/4951   Documentation of specifying NoCloud datasource via SMBIOS : https://cloudinit.readthedocs.io/en/latest/reference/datasources/nocloud.html#method-2-local-filesystem-kernel-commandline-or-smbios Upstream PR with change that broke the cloud-init behavior: https://github.com/canonical/cloud-init/pull/4426 Upstream PR fixing the behavior: https://github.com/canonical/cloud-init/pull/4949
2024-02-29 17:17:59 Launchpad Janitor cloud-init (Ubuntu Mantic): status Fix Committed Fix Released
2024-02-29 17:18:01 Andreas Hasenack removed subscriber Ubuntu Stable Release Updates Team
2024-02-29 17:18:13 Launchpad Janitor cloud-init (Ubuntu Jammy): status Fix Committed Fix Released
2024-02-29 17:18:28 Launchpad Janitor cloud-init (Ubuntu Focal): status Fix Committed Fix Released