Age | Commit message (Collapse) | Author | Files | Lines |
|
Currently when we call the major-upgrade step we do the following:
"""
...
if [[ -n $(is_bootstrap_node) ]]; then
check_clean_cluster
fi
...
if [[ -n $(is_bootstrap_node) ]]; then
migrate_full_to_ng_ha
fi
...
for service in $(services_to_migrate); do
manage_systemd_service stop "${service%%-clone}"
...
done
"""
The problem with the above code is that it is open to the following race
condition:
1. Code gets run first on a non-bootstrap controller node so we start
stopping a bunch of services
2. Pacemaker notices will notice that services are down and will mark
the service as stopped
3. Code gets run on the bootstrap node (controller-0) and the
check_clean_cluster function will fail and exit
4. Eventually also the script on the non-bootstrap controller node will
timeout and exit because the cluster never shut down (it never actually
started the shutdown because we failed at 3)
Let's make sure we first only call the HA NG migration step as a
separate heat step. Only afterwards we start shutting down the systemd
services on all nodes.
We also need to move the STONITH_STATE variable into a file because it
is being used across two different scripts (1 and 2) and we need to
store that state.
Co-Authored-By: Athlan-Guyot Sofer <sathlang@redhat.com>
Closes-Bug: #1640407
Change-Id: Ifb9b9e633fcc77604cca2590071656f4b2275c60
|
|
|
|
|
|
|
|
|
|
|
|
We do not encode the chars like [] possibly found in the artifacts
URL, so curl tries to glob against IPv6 addresses in brackets. This
change adds --globoff to the curl options so that IPv6 addresses in
brackets are not misinterpreted.
Closes-Bug: 1640148
Change-Id: Ic86ba1e5fb674bc15b4bcc6bd3ea9e943c4fbf8e
|
|
|
|
|
|
|
|
For parameter merge strategies to work we need to merge multiple environment
files, which doesn't consider the defaults defined in the heat template.
Moving where we define these defaults will enable the merge strategies
applied when appending services to roles in environment files to work.
Change-Id: I1ef1ad685c8a15308d051665c576a98b277f2496
Closes-Bug: #1635409
|
|
|
|
|
|
|
|
This can be used to pass the e.g. the tripleo-validations ssh key into
the deployment.
Change-Id: I861b9e2252a9c8122dcf7df261386f1ea5200c4f
Related-Bug: #1635226
|
|
|
|
|
|
|
|
manila-share also needs the db configuration so the db-sync works
correctly when manila-api is running on a non-controller node.
Change-Id: Ib8a6f10ef6a650275fc011e51acfc4b5c7c99164
Closes-Bug: 1633077
|
|
Because manila-share is a pacemaker-managed service, it has to be
on the controller node. If you deploy the api services to a
different node, then manila-share loses access to the authtoken
hieradata generated by manila-api. Adding it explicitly to the
manila-share config allows this setup to deploy sanely.
Note that I'm having a different problem with manila db-syncs in
this setup, so there's likely another patch required to get it
fully working.
Change-Id: Iac782fa67ea912d24b9905dd8bbafb8ff28dd669
Partial-Bug: 1633077
|
|
|
|
The param is now managed in puppet-tripleo like other services.
Change-Id: I306aa6ac6e2cfc0d4602e15e11564a6be096a121
Depends-On: Ibc0ed642931dd3ada7ee594bb8c70a1c3462206d
|
|
|
|
|
|
|
|
- Remove vncproxy firewall rules from nova-api service
- Add vncproxy firewall rules to nova-vncproxy service
- Add console port range firewall rules to nova-libvirt service
Change-Id: I421ae21c130cac6f25e7c0869b941ba77441172c
|
|
Some ports are missing to support live-migration. This patch adds them.
Documented here:
https://access.redhat.com/documentation/en/red-hat-openstack-platform/9/paged/migrating-instances/chapter-1-how-to-migrate-a-live-instance
Change-Id: I72634a9940c11602522322235e51bf27cb664e57
|
|
|
|
|
|
Seems the conditional has changed and we should pickup the
tripleo::profile::base::swift::storage::enable_swift_storage
hiera data.
After controller nodes are upgraded the swift services were down
even though there was no stand-alone swift node (the current
conditional was failing as that hiera isn't set any more)
Closes-Bug: 1638821
Change-Id: Id1383c1e54f9cae13fd375e90da525230e5d23eb
|
|
The hardcoded list should be configurable, and defaulted
to their current value.
Change-Id: I517aa61f21c6f4d0975b10a7aa85177c543487e0
Closes-bug: 1560098
|
|
The update configuration is generated into ceph.yaml and into
{rolename}.yaml. We should ensure puppet hiera is looking for
these files.
Change-Id: I261d16bc365b3d19adc502385edcc509a53ffc2a
Closes-Bug: #1638346
Resolves: rhbz#1388977
|
|
We want CephExternal to work without referencing CephBase which
instead defines common settings for hosted Ceph deployments.
This change fixes a reference to CephBase which was mistakenly
introduced with fix for bug #1632285.
Change-Id: Id27e935f91ad76a6877b3aa7588f54d6140aa41f
Closes-Bug: #1635014
|
|
currently udp port 8125 is blocked by default. This can cause issues
when sending statsd data.
Change-Id: Icb5569c4e3dc981e9a8accf32eedd3370552cb34
|
|
|
|
OPM package is metadata package with unversioned requirements which
means that update does not update the dependencies. This leaves us
with old puppet modules and old puppet during the puppet run.
Change-Id: I80f8a73142a09bb4178bb5a396d256ba81ba98a8
Closes-Bug: #1638266
Resolves: rhbz#1390559
|
|
gnocchi when configured with swift will require keystone
to be available to authenticate to migrate to v3. At this
step keystone is not available and gnocchi upgrade fails
with auth error. Instead start apache in step 3, start
apache first and then run gnocchi upgrade in a separate
step and let upgrade happen here.
Closes-Bug: #1634897
Change-Id: I22d02528420e4456f84b80905a7b3a80653fa7b0
|
|
|
|
The interface for this moved to init.pp, the one we currently
use now only outputs a warning, it doesn't actually set anything.
Change-Id: Idc40cf0dc4ff0f598e0918e0de8b3233b524cdd5
Closes-Bug: 1638254
|
|
|
|
Updated plugin name for configuring Nuage.
Nuage plugin name changed after Liberty release
and needs to be updated at all instances.
Updated neutron-nuage-config.yaml file to reflect
the change.
Change-Id: I7cce9a07b909ab59bf249439eec0833afce5cca6
Closes-Bug: #1635033
|
|
Previously the CephStorage nodes were missing the SNMP role.
Change-Id: I1356a3ff8da51da4d79b28312f9e3821652b6291
|
|
|
|
|
|
|
|
The Ceilometer middleware is in the wrong place; actually any middleware
should be deployed after catch_errors to catch any errors that would
otherwise crash the proxy service. Additionally the ceilometer
middleware should be deployed after any authentication middleware.
Closes-Bug: 1637471
Co-Authored-By: Thiago da Silva <thiago@redhat.com>
Change-Id: I710ff2f51271a78582fa502e7eecfa687800c664
|
|
A default TripleO installation uses a local directory named "d1" to be
used by Swift. With SwiftRawDisks set it is highly unlikely that that an
operator wants to use this any longer, because it affects system
perforamce and might result in an overfilled the system disk. In this
case d1 should be no longer when building rings.
This patch makes it possible to disable the d1 device usage in the ring
building process by using a new option "SwiftUseLocalDir". This is set
by default to true, not changing the default behavior. If set to false,
the d1 device won't be used when building rings.
Closes-Bug: 1634051
Change-Id: Ia9ad38e3ffa533e170f4cedd0518d830e9b2fa69
|
|
With this, we can clean it from puppet-tripleo.
Change-Id: I13638cd1af52537bef8540f0d5fa5f5f7decd392
Depends-On: Ic1967a6f4f60a273965811516f33121115d518b4
|
|
rpm command will return an exit 1 if ovs package is already
there and will exit the step_1.sh script. To get around this
force the update with --replacepkgs
Also remove the \ just before the $ which cause a syntax
error for the ceph storage
Change-Id: I11fcf688982ceda5eef7afc8904afae44300c2d9
Closes-bug: 1636748
|
|
|