Upgrades in Stein are incomplete
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
tripleo |
Won't Fix
|
High
|
Jiří Stránský |
Bug Description
This is a sort of "catch-all" bug for upgrade implementation pieces missing in stable/stein. Some of the patches have their own bugs, but others are simply not-yet-implemented parts. For the latter we can reference this bug when backporting.
Changed in tripleo: | |
assignee: | nobody → Jiří Stránský (jistr) |
Changed in tripleo: | |
milestone: | train-1 → train-2 |
OpenStack Infra (hudson-openstack) wrote : Fix proposed to tripleo-heat-templates (master) | #1 |
Changed in tripleo: | |
status: | Triaged → In Progress |
OpenStack Infra (hudson-openstack) wrote : Fix merged to tripleo-heat-templates (master) | #2 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: master
commit df3b4b3c5724d07
Author: Jiri Stransky <email address hidden>
Date: Mon Jun 10 14:14:33 2019 +0200
Configure insecure registries before Pacemaker-managed image pulls
Service templates like mysql-pacemaker
container images in step 2 of the upgrade tasks. In case they come
from an allowed insecure registry, the insecure registries setting
must be in place by step 1. We do just that by reusing the "install
and configure Podman" part of host_prep_tasks.
Change-Id: I36e003c09d7f77
Partial-Bug: #1831690
OpenStack Infra (hudson-openstack) wrote : Fix proposed to tripleo-heat-templates (master) | #3 |
Fix proposed to branch: master
Review: https:/
OpenStack Infra (hudson-openstack) wrote : | #4 |
Fix proposed to branch: master
Review: https:/
OpenStack Infra (hudson-openstack) wrote : Fix proposed to tripleo-heat-templates (stable/stein) | #5 |
Fix proposed to branch: stable/stein
Review: https:/
OpenStack Infra (hudson-openstack) wrote : | #6 |
Fix proposed to branch: stable/stein
Review: https:/
OpenStack Infra (hudson-openstack) wrote : | #7 |
Fix proposed to branch: stable/stein
Review: https:/
OpenStack Infra (hudson-openstack) wrote : | #8 |
Fix proposed to branch: stable/stein
Review: https:/
OpenStack Infra (hudson-openstack) wrote : | #9 |
Fix proposed to branch: stable/stein
Review: https:/
OpenStack Infra (hudson-openstack) wrote : | #10 |
Fix proposed to branch: stable/stein
Review: https:/
OpenStack Infra (hudson-openstack) wrote : | #11 |
Fix proposed to branch: stable/stein
Review: https:/
OpenStack Infra (hudson-openstack) wrote : Fix merged to tripleo-heat-templates (master) | #12 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: master
commit 2ca7c550390f597
Author: Jiri Stransky <email address hidden>
Date: Mon Jun 24 12:32:30 2019 +0200
Allow skipping RHSM with Leapp
Useful when developing and testing.
Also, we don't need to be explicitly installing RHSM, instead we
should ensure that it doesn't get removed by accident if it needs to
be present.
Change-Id: I65ccf3436bf381
Partial-Bug: #1831690
OpenStack Infra (hudson-openstack) wrote : | #13 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: master
commit 56a35beb51c050a
Author: Jiri Stransky <email address hidden>
Date: Mon Jun 24 12:32:37 2019 +0200
Idempotency for system_
Partial-Bug: #1831690
Change-Id: I9510fd00755cf3
OpenStack Infra (hudson-openstack) wrote : Fix proposed to tripleo-heat-templates (stable/stein) | #14 |
Fix proposed to branch: stable/stein
Review: https:/
OpenStack Infra (hudson-openstack) wrote : | #15 |
Fix proposed to branch: stable/stein
Review: https:/
OpenStack Infra (hudson-openstack) wrote : Fix merged to tripleo-heat-templates (stable/stein) | #16 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: stable/stein
commit 8d6b51b775ae635
Author: Jose Luis Franco Arza <email address hidden>
Date: Wed Apr 10 18:11:24 2019 +0200
Stop all services before upgrading node's OS.
During the system_
workflow, we need to stop all services before starting
the operating system upgrade, we're doing that by stopping
all docker containers at once after stopping the pcmkr
cluster in step 2.
Partial-Bug: #1831690
Change-Id: I50a1289a864f80
(cherry picked from commit 9d5c972d99afbb0
tags: | added: in-stable-stein |
OpenStack Infra (hudson-openstack) wrote : | #17 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: stable/stein
commit 7f393a3d40b402e
Author: Jiri Stransky <email address hidden>
Date: Tue May 7 16:26:19 2019 +0200
Add operating system upgrade preparation via Leapp
Also make sure pacemaker cluster is not only stopped but also
destroyed on the node being upgraded. This is a prerequisite for
upgrades via Leapp, otherwise Leapp would assume we expect it to
upgrade the cluster, and it would fail on validation.
Partial-Bug: #1831690
Change-Id: I8cf3e4cf294fee
(cherry picked from commit b0519479c19766e
OpenStack Infra (hudson-openstack) wrote : | #18 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: stable/stein
commit 34450cc635e8dfc
Author: Jiri Stransky <email address hidden>
Date: Fri May 10 15:52:49 2019 +0200
Remove OpenStack packages right before upgrading the operating system
This could make the upgrade faster, but mainly it's less likely that
something will go wrong when the node goes down for the offline
upgrade. We want to minimize the chances of that happening, because
such error is extremely unpleasant to deal with (restore from backup
might be the only solution at that point).
Partial-Bug: #1831690
Change-Id: Iddf30278f0cb4e
(cherry picked from commit bf8cde549b60a06
OpenStack Infra (hudson-openstack) wrote : | #19 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: stable/stein
commit 0af8eb1c64a3079
Author: Jose Luis Franco Arza <email address hidden>
Date: Fri Mar 15 19:07:42 2019 +0100
Remove baremetal to containerized pacemaker upgrade_tasks.
Now that all pacemaker managed services are containerized
there is no need to keep these upgrade_tasks.
Partial-Bug: #1831690
Change-Id: I6531563b4da8cf
(cherry picked from commit 0e6769b4b8225b7
OpenStack Infra (hudson-openstack) wrote : | #20 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: stable/stein
commit 2accf4d25d87d3b
Author: Jose Luis Franco Arza <email address hidden>
Date: Mon Mar 18 12:29:55 2019 +0100
Refactor ovn_dbs upgrade_tasks.
Reorder the upgrade_tasks for the ovn_dbs service in a similar way
it's being done for the rest of pacemaker managed services. Also,
ensure indempotency in the case the cluster is down as it will occur
for the S to R upgrades workflow.
Partial-Bug: #1831690
Change-Id: I8fe30260754465
(cherry picked from commit 05f4b253b845268
OpenStack Infra (hudson-openstack) wrote : | #21 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: stable/stein
commit 273f9a32a787198
Author: Jiri Stransky <email address hidden>
Date: Mon Jun 10 14:14:33 2019 +0200
Configure insecure registries before Pacemaker-managed image pulls
Service templates like mysql-pacemaker
container images in step 2 of the upgrade tasks. In case they come
from an allowed insecure registry, the insecure registries setting
must be in place by step 1. We do just that by reusing the "install
and configure Podman" part of host_prep_tasks.
Change-Id: I36e003c09d7f77
Partial-Bug: #1831690
(cherry picked from commit df3b4b3c5724d07
OpenStack Infra (hudson-openstack) wrote : Fix proposed to tripleo-heat-templates (master) | #22 |
Fix proposed to branch: master
Review: https:/
OpenStack Infra (hudson-openstack) wrote : Fix merged to tripleo-heat-templates (stable/stein) | #23 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: stable/stein
commit 5f40f0e80865cf0
Author: Jiri Stransky <email address hidden>
Date: Thu Jun 6 14:09:37 2019 +0200
Upgrade playbook fixes for OS upgrade
These are remaining fixes for pacemaker and mysql tasks to make
upgrade_
upgrade to EL 8.
Partial-Bug: #1831690
Change-Id: If8c97752463e56
(cherry picked from commit 6baf5326bfb949b
OpenStack Infra (hudson-openstack) wrote : | #24 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: stable/stein
commit 367f4decd9bea7f
Author: Jiri Stransky <email address hidden>
Date: Mon Jun 24 12:32:30 2019 +0200
Allow skipping RHSM with Leapp
Useful when developing and testing.
Also, we don't need to be explicitly installing RHSM, instead we
should ensure that it doesn't get removed by accident if it needs to
be present.
Change-Id: I65ccf3436bf381
Partial-Bug: #1831690
(cherry picked from commit 2ca7c550390f597
OpenStack Infra (hudson-openstack) wrote : | #25 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: stable/stein
commit 29fa95801113d52
Author: Jiri Stransky <email address hidden>
Date: Mon Jun 24 12:32:37 2019 +0200
Idempotency for system_
Partial-Bug: #1831690
Change-Id: I9510fd00755cf3
(cherry picked from commit 56a35beb51c050a
OpenStack Infra (hudson-openstack) wrote : Fix proposed to tripleo-heat-templates (master) | #26 |
Fix proposed to branch: master
Review: https:/
OpenStack Infra (hudson-openstack) wrote : Change abandoned on tripleo-heat-templates (master) | #27 |
Change abandoned by Jiri Stransky (<email address hidden>) on branch: master
Review: https:/
Reason: Reporting upstream+downstream bugs for this to be solved better
Changed in tripleo: | |
milestone: | train-2 → train-3 |
OpenStack Infra (hudson-openstack) wrote : Fix merged to tripleo-heat-templates (master) | #28 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: master
commit 2b427970a36a63c
Author: Jiri Stransky <email address hidden>
Date: Wed Jul 10 15:24:13 2019 +0200
Handle edge cases in staged upgrade hiera data
We make sure the hiera is set in an either-or branches so that we
prevent from executing both at the same time or none at all.
Edge case 1 which is not that much of an edge case --
single-
branches resulting in duplicate entries.
Edge case 2 -- having the service in roles data but having no node
assigned to that role. In that case previously we wouldn't set the
variables and the upgrade would fail.
Change-Id: I1681a5e5ac7842
Partial-Bug: #1831690
OpenStack Infra (hudson-openstack) wrote : Fix proposed to tripleo-heat-templates (stable/stein) | #29 |
Fix proposed to branch: stable/stein
Review: https:/
OpenStack Infra (hudson-openstack) wrote : Fix merged to tripleo-heat-templates (stable/stein) | #30 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: stable/stein
commit 945a22166c48079
Author: Jiri Stransky <email address hidden>
Date: Wed Jul 10 15:24:13 2019 +0200
Handle edge cases in staged upgrade hiera data
We make sure the hiera is set in an either-or branches so that we
prevent from executing both at the same time or none at all.
Edge case 1 which is not that much of an edge case --
single-
branches resulting in duplicate entries.
Edge case 2 -- having the service in roles data but having no node
assigned to that role. In that case previously we wouldn't set the
variables and the upgrade would fail.
Change-Id: I1681a5e5ac7842
Partial-Bug: #1831690
(cherry picked from commit 2b427970a36a63c
OpenStack Infra (hudson-openstack) wrote : Fix proposed to tripleo-heat-templates (master) | #31 |
Fix proposed to branch: master
Review: https:/
OpenStack Infra (hudson-openstack) wrote : Fix merged to tripleo-heat-templates (master) | #32 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: master
commit e26009fa9c57d42
Author: Jiri Stransky <email address hidden>
Date: Fri Aug 9 15:52:29 2019 +0200
Check for rc instead of |succeeded
We went from `ignore_errors: true` to `failed_when: false` to prevent
Ansible printing errors which aren't really errors. When using
`succeeded` filter on a registered variable from such task, we will
always get true, which is not what we want. Check for `.rc == 0`
instead to set the facts as they were intended.
Change-Id: Ib8dea68720fda8
Partial-Bug: #1831690
OpenStack Infra (hudson-openstack) wrote : Fix proposed to tripleo-heat-templates (stable/stein) | #33 |
Fix proposed to branch: stable/stein
Review: https:/
OpenStack Infra (hudson-openstack) wrote : Fix proposed to tripleo-heat-templates (master) | #34 |
Fix proposed to branch: master
Review: https:/
OpenStack Infra (hudson-openstack) wrote : | #35 |
Fix proposed to branch: master
Review: https:/
OpenStack Infra (hudson-openstack) wrote : | #36 |
Fix proposed to branch: master
Review: https:/
OpenStack Infra (hudson-openstack) wrote : Fix merged to tripleo-heat-templates (stable/stein) | #37 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: stable/stein
commit 9ee30cdeb468e21
Author: Jiri Stransky <email address hidden>
Date: Fri Aug 9 15:52:29 2019 +0200
Check for rc instead of |succeeded
We went from `ignore_errors: true` to `failed_when: false` to prevent
Ansible printing errors which aren't really errors. When using
`succeeded` filter on a registered variable from such task, we will
always get true, which is not what we want. Check for `.rc == 0`
instead to set the facts as they were intended.
Change-Id: Ib8dea68720fda8
Partial-Bug: #1831690
(cherry picked from commit e26009fa9c57d42
OpenStack Infra (hudson-openstack) wrote : Fix merged to tripleo-heat-templates (master) | #38 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: master
commit d4db2fa532c81c7
Author: Jiri Stransky <email address hidden>
Date: Tue Aug 13 13:59:44 2019 +0200
Force re-run of pacemaker bundle init containers during upgrade-scaleup
We need to recreate the *_init_bundle containers during the staged
upgrade (upgrade-scaleup), because the init containers will e.g. set
the correct replica count for the bundle as we're scaling up from 1 to
3 controllers.
The init containers wouldn't get re-run/recreated automatically on
upgrade-scaleup because none of their input would change. The
DeploymentI
and applied during upgrade with `--limit controller-0`. Then later
when we run `--limit controller-
DeploymentI
controller-0, so we have to explicitly remove the bundle init
containers to force re-running them.
Change-Id: I93c6d7292f3752
Partial-Bug: #1831690
OpenStack Infra (hudson-openstack) wrote : Fix proposed to tripleo-heat-templates (stable/stein) | #39 |
Fix proposed to branch: stable/stein
Review: https:/
OpenStack Infra (hudson-openstack) wrote : | #40 |
Fix proposed to branch: stable/stein
Review: https:/
OpenStack Infra (hudson-openstack) wrote : Fix merged to tripleo-heat-templates (stable/stein) | #41 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: stable/stein
commit 4b9c3637f8e5819
Author: Jiri Stransky <email address hidden>
Date: Tue Aug 13 13:59:44 2019 +0200
Force re-run of pacemaker bundle init containers during upgrade-scaleup
We need to recreate the *_init_bundle containers during the staged
upgrade (upgrade-scaleup), because the init containers will e.g. set
the correct replica count for the bundle as we're scaling up from 1 to
3 controllers.
The init containers wouldn't get re-run/recreated automatically on
upgrade-scaleup because none of their input would change. The
DeploymentI
and applied during upgrade with `--limit controller-0`. Then later
when we run `--limit controller-
DeploymentI
controller-0, so we have to explicitly remove the bundle init
containers to force re-running them.
Change-Id: I93c6d7292f3752
Partial-Bug: #1831690
(cherry picked from commit d4db2fa532c81c7
OpenStack Infra (hudson-openstack) wrote : Fix merged to tripleo-heat-templates (master) | #42 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: master
commit 206625d4f5bd9a6
Author: Jiri Stransky <email address hidden>
Date: Tue Aug 13 15:01:44 2019 +0200
Allow combining system_
Previously we required the operator to run two separate commands for
the "prepare" and "run" phases of operating system upgrade. This
commit refactors the upgrade_tasks with these tags so that running the
whole system upgrade is possible via a single command with `--tags
system_
Allowing to run in a single command requires being more careful about
what can happen in which step number in the upgrade tasks. The upgrade
steps for system upgrade are now explicitly documented in composable
services readme.
The existing system_upgrade_run and system_
checked and moved into the appropriate steps. In the case of
pacemaker, it required moving the cluster stop/destroy action into a
single file with removing all containers, to guarantee that the
cluster is stopped before the container removal, otherwise pacemaker
would try to spawn new containers.
Change-Id: I3cd78de8d07be4
Partial-Bug: #1831690
OpenStack Infra (hudson-openstack) wrote : Fix merged to tripleo-heat-templates (stable/stein) | #43 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: stable/stein
commit 7f7960a5367e296
Author: Jiri Stransky <email address hidden>
Date: Tue Aug 13 15:01:44 2019 +0200
Allow combining system_
Previously we required the operator to run two separate commands for
the "prepare" and "run" phases of operating system upgrade. This
commit refactors the upgrade_tasks with these tags so that running the
whole system upgrade is possible via a single command with `--tags
system_
Allowing to run in a single command requires being more careful about
what can happen in which step number in the upgrade tasks. The upgrade
steps for system upgrade are now explicitly documented in composable
services readme.
The existing system_upgrade_run and system_
checked and moved into the appropriate steps. In the case of
pacemaker, it required moving the cluster stop/destroy action into a
single file with removing all containers, to guarantee that the
cluster is stopped before the container removal, otherwise pacemaker
would try to spawn new containers.
Change-Id: I3cd78de8d07be4
Partial-Bug: #1831690
(cherry picked from commit 206625d4f5bd9a6
OpenStack Infra (hudson-openstack) wrote : Fix merged to tripleo-heat-templates (master) | #44 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: master
commit 86d425601495550
Author: Jiri Stransky <email address hidden>
Date: Tue Aug 13 15:42:18 2019 +0200
Set EnablePackageIn
For in-place upgrades via Leapp, which is the default, we are
dependent on package installs. User can still override the default
value if they wish so, but they might have to pull in the content by
some manual means.
Change-Id: I5e5629d2baed97
Partial-Bug: #1831690
OpenStack Infra (hudson-openstack) wrote : Fix proposed to tripleo-heat-templates (stable/stein) | #45 |
Fix proposed to branch: stable/stein
Review: https:/
OpenStack Infra (hudson-openstack) wrote : Fix merged to tripleo-heat-templates (stable/stein) | #46 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: stable/stein
commit 3f576469144311f
Author: Jiri Stransky <email address hidden>
Date: Tue Aug 13 15:42:18 2019 +0200
Set EnablePackageIn
For in-place upgrades via Leapp, which is the default, we are
dependent on package installs. User can still override the default
value if they wish so, but they might have to pull in the content by
some manual means.
Change-Id: I5e5629d2baed97
Partial-Bug: #1831690
(cherry picked from commit 86d425601495550
Changed in tripleo: | |
milestone: | train-3 → ussuri-1 |
OpenStack Infra (hudson-openstack) wrote : Fix proposed to tripleo-common (stable/stein) | #47 |
Fix proposed to branch: stable/stein
Review: https:/
OpenStack Infra (hudson-openstack) wrote : Fix proposed to tripleo-heat-templates (master) | #48 |
Fix proposed to branch: master
Review: https:/
OpenStack Infra (hudson-openstack) wrote : Fix proposed to tripleo-heat-templates (stable/stein) | #49 |
Fix proposed to branch: stable/stein
Review: https:/
OpenStack Infra (hudson-openstack) wrote : Fix merged to tripleo-common (stable/stein) | #50 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: stable/stein
commit 0708730ec6e5985
Author: Jiri Stransky <email address hidden>
Date: Thu Sep 26 09:52:27 2019 +0200
[Stein] Optional flag file for tripleo-transfer to prevent overwriting
When using tripleo-transfer role, optionally a file will be created at
a path specified by tripleo_
transfer will be prevented in case the file already exists. This is to
ensure the data transfer happens only once and the data is not
overwritten by accident.
The corresponding change I3a08949436dcb6
master branch has to go to tripleo-ansible repo.
Change-Id: I6176e3c9998d8b
Partial-Bug: #1831690
OpenStack Infra (hudson-openstack) wrote : Fix merged to tripleo-heat-templates (master) | #51 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: master
commit 60b0359b83adc8c
Author: Jiri Stransky <email address hidden>
Date: Fri Sep 27 10:18:22 2019 +0200
Prevent running mysql data transfer twice by accident
Providing the tripleo_
transfer would stop (fail intentionally) if it has been already run
previously.
Change-Id: If81b1ce9ef381f
Partial-Bug: #1831690
OpenStack Infra (hudson-openstack) wrote : Fix merged to tripleo-heat-templates (stable/stein) | #52 |
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: stable/stein
commit 603ae5d3e47297c
Author: Jiri Stransky <email address hidden>
Date: Fri Sep 27 10:18:22 2019 +0200
Prevent running mysql data transfer twice by accident
Providing the tripleo_
transfer would stop (fail intentionally) if it has been already run
previously.
Change-Id: If81b1ce9ef381f
Partial-Bug: #1831690
(cherry-picked from 60b0359b83adc8c
Changed in tripleo: | |
milestone: | ussuri-1 → ussuri-2 |
Changed in tripleo: | |
milestone: | ussuri-2 → ussuri-3 |
Changed in tripleo: | |
milestone: | ussuri-3 → ussuri-rc3 |
Changed in tripleo: | |
milestone: | ussuri-rc3 → victoria-1 |
Jiří Stránský (jistr) wrote : | #53 |
Stale bug, let's reopen if necessary.
Changed in tripleo: | |
status: | In Progress → Won't Fix |
Fix proposed to branch: master /review. opendev. org/664277
Review: https:/