apex-tripleo-heat-templates.git
8 years agoPing retry
Joe Talerico [Tue, 29 Mar 2016 18:10:20 +0000 (14:10 -0400)]
Ping retry

The single ping method in the validation script is causing for
deployments to fail. When reviewing the network connectivity, we are
finding we actually do have connectivity
( https://gist.github.com/jtaleric/0276a117625e44993be0 ). This patch is
to change the ping count from 1 to 10, to ensure the network is up.

Closes-Bug: 1563521
Change-Id: I9772407554dffa91978a49a16490ef9ed448a054

8 years agoMerge "Increment step count to include ringbuilder"
Jenkins [Tue, 29 Mar 2016 16:03:05 +0000 (16:03 +0000)]
Merge "Increment step count to include ringbuilder"

8 years agoMerge "Don't restart Pacemaker-managed services from Puppet"
Jenkins [Tue, 29 Mar 2016 15:38:23 +0000 (15:38 +0000)]
Merge "Don't restart Pacemaker-managed services from Puppet"

8 years agoMerge "Add quotes around the cinder_iscsi_ip_address value"
Jenkins [Tue, 29 Mar 2016 15:36:47 +0000 (15:36 +0000)]
Merge "Add quotes around the cinder_iscsi_ip_address value"

8 years agoDon't restart Pacemaker-managed services from Puppet
Jiri Stransky [Tue, 29 Mar 2016 12:11:08 +0000 (14:11 +0200)]
Don't restart Pacemaker-managed services from Puppet

We need to reload/restart services on updates/upgrades to apply any
config changes, but restarting services managed from Pacemaker from
Puppet causes problems.

For now we no-op the restart and rely on the catch-all restart after
Puppet phase.

In the future we should have a service provider for pacemaker resources
that will be using pcs. We still might have to restart services outside
Puppet due to cluster-wide orchestration issues, but we might be able to
do the restarts selectively rather than restart everything.

We also no-op the start/stop commands to be safe, as it also doesn't
make sense for Puppet to try start and stop those services when it
doesn't have knowledge about Pacemaker.

Change-Id: I95e21e10471cd7575f28c095c48150325f1414b3
Closes-Bug: #1562922

8 years agoMerge "Set UpdateIdentifier for upgrade converge, to prevent services down"
Jenkins [Tue, 29 Mar 2016 11:41:19 +0000 (11:41 +0000)]
Merge "Set UpdateIdentifier for upgrade converge, to prevent services down"

8 years agoMerge "Remove password default for AodhPassword"
Jenkins [Mon, 28 Mar 2016 21:32:15 +0000 (21:32 +0000)]
Merge "Remove password default for AodhPassword"

8 years agoIncrement step count to include ringbuilder
Dan Prince [Tue, 22 Mar 2016 20:22:45 +0000 (16:22 -0400)]
Increment step count to include ringbuilder

This patch wires in ringbuilder.pp so that it is always
asserted like the other manifests and it fixes the misaligned
step sequencing in calling our overcloud controller manifests.

Previously it was called as a separate software deployment outside of
the hiera step sequence. This made things confusing in
controller-post.yaml since the deployment names didn't align
with the step hiera variables after step 3. Now that we call it
just like the other modules it should make gradually moving this
code to puppet-tripleo more straightforward as well.

Change-Id: Ibd4f51f65da475bb20a6b08d7bda673f330a5464

8 years agoSet UpdateIdentifier for upgrade converge, to prevent services down
Mathieu Bultel [Thu, 24 Mar 2016 14:06:42 +0000 (15:06 +0100)]
Set UpdateIdentifier for upgrade converge, to prevent services down

We'd like to let the post puppet pacemaker controller services
restart to happen for the convergence step so set the
UpdateIdentifier. However also set the PackageUpdate to noop so the
yum_update.sh doesn't happen.

Since a full haproxy restart is expected, we no longer need the
systemctl reload added at Iae3bad745ecdf952a7a0314fe1375d07eb47c454
so remove that too.

Some more context at
https://bugzilla.redhat.com/show_bug.cgi?id=1321036

Co-Authored-By: marios <marios@redhat.com>
Change-Id: I31c2d97d68c97b435f63863fae2c89f18f99681d

8 years agoRemove password default for AodhPassword
Pradeep Kilambi [Thu, 24 Mar 2016 18:22:27 +0000 (14:22 -0400)]
Remove password default for AodhPassword

Change-Id: Ibf37bfd6150d212fadcc4d2e2e2d0a89cdd76c91

8 years agoMerge "Fix satellite registration for http or https"
Jenkins [Thu, 24 Mar 2016 18:06:00 +0000 (18:06 +0000)]
Merge "Fix satellite registration for http or https"

8 years agoMerge "Add systemctl reload haproxy to the pacemaker_resource_restart.sh"
Jenkins [Thu, 24 Mar 2016 18:00:09 +0000 (18:00 +0000)]
Merge "Add systemctl reload haproxy to the pacemaker_resource_restart.sh"

8 years agoMerge "Deploy Aodh services, replacing Ceilometer Alarm"
Jenkins [Thu, 24 Mar 2016 17:51:42 +0000 (17:51 +0000)]
Merge "Deploy Aodh services, replacing Ceilometer Alarm"

8 years agoAdd quotes around the cinder_iscsi_ip_address value
Giulio Fidente [Thu, 24 Mar 2016 15:23:08 +0000 (16:23 +0100)]
Add quotes around the cinder_iscsi_ip_address value

In I783e939ae304385674909bfd9f1cac95e04cef22 we add brackets around
the cinder_iscsi_ip_address if IPv6 but that causes hiera to try
mapping the value into an array, while it isn't. This change adds
quotes around the brackets.

Change-Id: Id9bb4b12542f1943e9df702486d68424539c7a59
Closes-Bug: 1560934

8 years agoMerge "Remove unused user resource."
Jenkins [Thu, 24 Mar 2016 14:27:07 +0000 (14:27 +0000)]
Merge "Remove unused user resource."

8 years agoWire redis_password to the tripleo module
Giulio Fidente [Tue, 22 Mar 2016 17:00:03 +0000 (18:00 +0100)]
Wire redis_password to the tripleo module

Without this the HAProxy monitoring for Redis would fail to poll
the backends.

Change-Id: Id0826c6b04e471844c7bef69480af263cf2b3bd4

8 years agoMerge "Pass iSCSI bind address in brackets to Cinder when IPv6"
Jenkins [Thu, 24 Mar 2016 12:25:54 +0000 (12:25 +0000)]
Merge "Pass iSCSI bind address in brackets to Cinder when IPv6"

8 years agoMerge "Remove the glance/rbd user name from static hieradata"
Jenkins [Thu, 24 Mar 2016 11:02:30 +0000 (11:02 +0000)]
Merge "Remove the glance/rbd user name from static hieradata"

8 years agoMerge "configure horizon with memcached ipv6 when needed"
Jenkins [Wed, 23 Mar 2016 23:04:00 +0000 (23:04 +0000)]
Merge "configure horizon with memcached ipv6 when needed"

8 years agoMerge "nonha: fix memcached servers string in nova.conf for ipv6"
Jenkins [Wed, 23 Mar 2016 22:50:09 +0000 (22:50 +0000)]
Merge "nonha: fix memcached servers string in nova.conf for ipv6"

8 years agoconfigure horizon with memcached ipv6 when needed
Emilien Macchi [Wed, 23 Mar 2016 14:56:48 +0000 (10:56 -0400)]
configure horizon with memcached ipv6 when needed

* Add MemcachedIPv6 parameter
* If MemcachedIPv6 is set at True, configure Horizon with Memcached IPv6
  addresses.

This patch is required to make Horizon working when running IPv6
networks.

Change-Id: I752e727bfb9040b29f5d755f565fa6b54b9511c8

8 years agoAdd simple parameter test to yaml-validate.py
Steven Hardy [Tue, 12 Jan 2016 19:03:30 +0000 (19:03 +0000)]
Add simple parameter test to yaml-validate.py

Output a warning for parameters which look unused, this should help
developers clean up the template a bit, and eventually could maybe
be developed further into something we can use for gating.

Change-Id: Ide4fbe3c85854cbddee44801d39ae73003d63bb8

8 years agoPass iSCSI bind address in brackets to Cinder when IPv6
Giulio Fidente [Wed, 23 Mar 2016 13:56:29 +0000 (14:56 +0100)]
Pass iSCSI bind address in brackets to Cinder when IPv6

Configure the Cinder iscsi_ip_address key using brackets when
the IP address is of v6 class.

Closes-Bug: 1560934
Change-Id: I783e939ae304385674909bfd9f1cac95e04cef22

8 years agoAdd systemctl reload haproxy to the pacemaker_resource_restart.sh
marios [Wed, 23 Mar 2016 14:17:13 +0000 (16:17 +0200)]
Add systemctl reload haproxy to the pacemaker_resource_restart.sh

As discussed in the related bug below, after upgrading your
environment to latest liberty the haproxy config isn't picked
up. This adds a systemctl reload haproxy in the pacemaker
resource restart we run as part of the post-puppet-pacemaker.

Related-Bug: 1561012
Change-Id: Iae3bad745ecdf952a7a0314fe1375d07eb47c454

8 years agononha: fix memcached servers string in nova.conf for ipv6
Emilien Macchi [Mon, 21 Mar 2016 18:42:52 +0000 (14:42 -0400)]
nonha: fix memcached servers string in nova.conf for ipv6

Full context is described here:
https://review.openstack.org/#/c/270110/

The patch that was supposed to fix [1] was not fixing non-ha scenario.
[1] https://launchpad.net/bugs/1536103

This patch aims to fix it.

Change-Id: Iaf4608de1894ce03f35925939e83230abb9f5207
Closes-Bug: #1560063

8 years agoFix satellite registration for http or https
James Slagle [Tue, 22 Mar 2016 21:26:42 +0000 (17:26 -0400)]
Fix satellite registration for http or https

If the satellite registration url was specified with https, the curl
command to detect the satellite version would not work as expected since
-L was not passed and you get redirected to https when testing the ping
api.

To additionally handle the case where https is specified, also use curl
directly with -k to download the configuration rpm instead of using rpm
with a url.

Fixes another bug with a missing $ in the reference to the
$satellite_version variable.

Change-Id: I984fdfc415eeeed4ef29cc8d0812e1b67545d6b1

8 years agoRemove the glance/rbd user name from static hieradata
Giulio Fidente [Wed, 23 Mar 2016 13:22:50 +0000 (14:22 +0100)]
Remove the glance/rbd user name from static hieradata

The static setting for the glance/rbd user name was overriding
any customization provided via template param because it was
up in the hierarchy for the controller nodes.

More at: https://bugzilla.redhat.com/show_bug.cgi?id=1308889

Change-Id: I3d112de7eeffd524fb1308d5976a28f04aa5ff23

8 years agoMerge "Comment out the ManagementPort in external-loadbalancer-vip.yaml"
Jenkins [Wed, 23 Mar 2016 01:33:41 +0000 (01:33 +0000)]
Merge "Comment out the ManagementPort in external-loadbalancer-vip.yaml"

8 years agoMerge "Fix redis coordination url"
Jenkins [Tue, 22 Mar 2016 21:35:02 +0000 (21:35 +0000)]
Merge "Fix redis coordination url"

8 years agoFix redis coordination url
Pradeep Kilambi [Tue, 22 Mar 2016 16:03:49 +0000 (12:03 -0400)]
Fix redis coordination url

The coordination url connection string to redis
is incorrectly formatted with password.

More details: https://bugzilla.redhat.com/show_bug.cgi?id=1320036

Change-Id: I93f5e93dfce4ba2629aa57534e8d33d5d1e6d77b

8 years agoMerge "Add BondInterfaceOvsOptions parameter to net-config-bond.yaml"
Jenkins [Tue, 22 Mar 2016 16:54:45 +0000 (16:54 +0000)]
Merge "Add BondInterfaceOvsOptions parameter to net-config-bond.yaml"

8 years agoMerge "Remove GlanceRegistry from EndpointMap"
Jenkins [Tue, 22 Mar 2016 15:00:56 +0000 (15:00 +0000)]
Merge "Remove GlanceRegistry from EndpointMap"

8 years agoMerge "Remove CephStorageCount"
Jenkins [Tue, 22 Mar 2016 11:44:23 +0000 (11:44 +0000)]
Merge "Remove CephStorageCount"

8 years agoMerge "Allow option to disable Swift ring management"
Jenkins [Mon, 21 Mar 2016 17:43:42 +0000 (17:43 +0000)]
Merge "Allow option to disable Swift ring management"

8 years agoMerge "Wire in HAProxy stats user and password"
Jenkins [Mon, 21 Mar 2016 16:58:05 +0000 (16:58 +0000)]
Merge "Wire in HAProxy stats user and password"

8 years agoAllow option to disable Swift ring management
Christian Schwede [Sat, 5 Mar 2016 13:49:00 +0000 (14:49 +0100)]
Allow option to disable Swift ring management

There are quite a few cases where it is useful to disable ring building on the
nodes. For example:

- using different weights, regions, and zones
- replacing a node in an existing Swift cluster
- adding a new node to an existing cluster
- using storage policies and therefore multiple rings
- using different nodes and disks for account, container and object servers

This patch allows it to disable ring building. Rings need to be maintained
manually then, and copied to all storage and proxy nodes within a cluster.

This patch is similar to I01311ec3ca265b151f8740bf7dc57cdf0cf0df6f, except that
it uses the current templates.

Change-Id: I56978b15823dd6eaf4b6fd3440df2f895e89611a

8 years agoRemove CephStorageCount
Derek Higgins [Fri, 18 Mar 2016 11:13:54 +0000 (11:13 +0000)]
Remove CephStorageCount

This is set by tripleoclient, remove it from here so it doesn't
override the user provided value.

Change-Id: I6110b71e484af749838f91dc5c6c4982b0c83074

8 years agoRemove unused user resource.
Sofer Athlan-Guyot [Fri, 18 Mar 2016 10:56:55 +0000 (11:56 +0100)]
Remove unused user resource.

The user is created by installation of of the pacemaker package, so it's
not required to add it to the resource catalog.

This paves the way to merge the refeactoring of the puppet-pacemaker
module[1].  It brings a lot of changes, one of them is an idempotent
handling of the hacluster user's password.  Removing it here prevents
duplicate resource error durring puppet run.

[1] https://review.openstack.org/#/c/294182/

Change-Id: I56849d9fc00bd3ce342d5c440cfe7c5b6d26b5bf

8 years agoDeploy Aodh services, replacing Ceilometer Alarm
Pradeep Kilambi [Mon, 7 Mar 2016 16:27:17 +0000 (11:27 -0500)]
Deploy Aodh services, replacing Ceilometer Alarm

Ceilometer Alarm is deprecated in Liberty by Aodh.

This patch:
* manage Aodh Keystone resources
* deploy Aodh API under WSGI, Notifier, Listener and Evaluator
* manage new parameters to customize Aodh deployment
* uses ceilometer DB for the upgrade path
* pacemaker config
* Add migration logic to remove pcs resources

Depends-On: I5333faa72e52d2aa2a622ac2d4b60825aadc52b5
Depends-On: Ib6c9c4c35da3fb55e0ca8e2d5a58ebaf4204d792

Co-Authored-By: Emilien Macchi <emilien@redhat.com>
Change-Id: Ib47a22884afb032ebc1655e1a4a06bfe70249134

8 years agoComment out the ManagementPort in external-loadbalancer-vip.yaml
Marius Cornea [Fri, 18 Mar 2016 14:07:17 +0000 (15:07 +0100)]
Comment out the ManagementPort in external-loadbalancer-vip.yaml

The Management network is optional and disabled by default.
This change preserves backward compatibility and fixes
https://bugzilla.redhat.com/show_bug.cgi?id=1317594

Change-Id: I73cf51154c9ee7c05938e2cadf0c5ac107840bad

8 years agoRemove GlanceRegistry from EndpointMap
Giulio Fidente [Tue, 8 Mar 2016 09:29:17 +0000 (10:29 +0100)]
Remove GlanceRegistry from EndpointMap

We don't need an endpoint for the glance-registry service, that is
used by glance-api when needed and is not meant to be user-facing.

Change-Id: Ia6c9dd6164d3b91adbc937d70fa74d5fbbfb28a3

8 years agoMerge "Remove default for the RabbitMQ password"
Jenkins [Fri, 18 Mar 2016 09:37:30 +0000 (09:37 +0000)]
Merge "Remove default for the RabbitMQ password"

8 years agoRemove default for the RabbitMQ password
Juan Antonio Osorio Robles [Wed, 16 Mar 2016 09:47:14 +0000 (11:47 +0200)]
Remove default for the RabbitMQ password

Since the password is now autogenerated from the tripleoclient,
there is no need to keep the default value here.

Change-Id: If41cb56134966456f8590da04f392faffe5c62a1
Closes-Bug: #1557688

8 years agoMerge "Remove forced rabbitmq::file_limit conversion to string"
Jenkins [Thu, 17 Mar 2016 21:52:33 +0000 (21:52 +0000)]
Merge "Remove forced rabbitmq::file_limit conversion to string"

8 years agoMerge "compute: use live_migration_tunnelled"
Jenkins [Wed, 16 Mar 2016 16:20:26 +0000 (16:20 +0000)]
Merge "compute: use live_migration_tunnelled"

8 years agoMerge "Install mongodb client package"
Jenkins [Tue, 15 Mar 2016 21:29:59 +0000 (21:29 +0000)]
Merge "Install mongodb client package"

8 years agoMerge "Fix typos"
Jenkins [Tue, 15 Mar 2016 21:29:22 +0000 (21:29 +0000)]
Merge "Fix typos"

8 years agoMerge "Ensure access to Redis is password protected"
Jenkins [Tue, 15 Mar 2016 15:28:30 +0000 (15:28 +0000)]
Merge "Ensure access to Redis is password protected"

8 years agocompute: use live_migration_tunnelled
Emilien Macchi [Mon, 14 Mar 2016 16:21:18 +0000 (12:21 -0400)]
compute: use live_migration_tunnelled

In a previous patch [1], we added support for VIR_MIGRATE_TUNNELLED when
doing VM shared storage.

In Nova Mitaka [2] [3], we have now a parameter called
'live_migration_tunnelled' to whether or not use tunnelled migration.
It replaces 'block_migration_flag' and 'live_migration_flag' that are
both deprecated.

[1] https://review.openstack.org/#/c/286584/
[2] https://review.openstack.org/#/c/263436/
[3] https://review.openstack.org/#/c/263434/

Change-Id: I8b199b6e72c80b2df7b679e0a20e39f8400d0478

8 years agoMerge "compute: include VIR_MIGRATE_TUNNELLED when doing VM shared storage"
Jenkins [Mon, 14 Mar 2016 15:58:03 +0000 (15:58 +0000)]
Merge "compute: include VIR_MIGRATE_TUNNELLED when doing VM shared storage"

8 years agoMerge "Keystone domain for Heat"
Jenkins [Mon, 14 Mar 2016 12:30:37 +0000 (12:30 +0000)]
Merge "Keystone domain for Heat"

8 years agoMerge "Pass public_addr to ceph::profile::params instead of ceph::mon"
Jenkins [Fri, 11 Mar 2016 15:28:20 +0000 (15:28 +0000)]
Merge "Pass public_addr to ceph::profile::params instead of ceph::mon"

8 years agoMerge "Remove unused Neutron Agents container"
Jenkins [Fri, 11 Mar 2016 13:55:03 +0000 (13:55 +0000)]
Merge "Remove unused Neutron Agents container"

8 years agoMerge "Parameterize the heat-docker-agents image"
Jenkins [Fri, 11 Mar 2016 13:54:47 +0000 (13:54 +0000)]
Merge "Parameterize the heat-docker-agents image"

8 years agoMerge "Allow the containerized compute node to spawn larger VMs"
Jenkins [Fri, 11 Mar 2016 13:54:07 +0000 (13:54 +0000)]
Merge "Allow the containerized compute node to spawn larger VMs"

8 years agoMerge "Upgrades: quiet yum upgrade on cinder nodes"
Jenkins [Fri, 11 Mar 2016 13:37:58 +0000 (13:37 +0000)]
Merge "Upgrades: quiet yum upgrade on cinder nodes"

8 years agocompute: include VIR_MIGRATE_TUNNELLED when doing VM shared storage
Emilien Macchi [Tue, 1 Mar 2016 13:24:25 +0000 (08:24 -0500)]
compute: include VIR_MIGRATE_TUNNELLED when doing VM shared storage

This patch makes sure:

* When doing shared storage
Nova is configured with block_migration_flag and live_migration_flag = '(...),VIR_MIGRATE_TUNNELLED'
flag for security improvements.

* When not doing shared storage
Nova is not configured with VIR_MIGRATE_TUNNELLED flag because it's not
supported by Qemu yet. We need to make sure the value is unset otherwise
live migration will fail when not running shared storage for VMs.

Note: this patch will be backport to stable branches. In a further
iteration, we'll probably use live_migration_tunnelled new Nova
parameter which is a simplier way to manage this feature.

Co-Authored-By: Kashyap Chamarthy <kchamart@redhat.com>
Change-Id: I557c1624ee944a32b1831d504f7b189308cd1961

8 years agoMerge "Enable predictable IPs on non-controllers"
Jenkins [Thu, 10 Mar 2016 21:51:14 +0000 (21:51 +0000)]
Merge "Enable predictable IPs on non-controllers"

8 years agoMerge "Support the deployment of Ceph over IPv6"
Jenkins [Thu, 10 Mar 2016 21:35:50 +0000 (21:35 +0000)]
Merge "Support the deployment of Ceph over IPv6"

8 years agoMerge "Fix vncproxy_host for IPv6"
Jenkins [Thu, 10 Mar 2016 21:15:50 +0000 (21:15 +0000)]
Merge "Fix vncproxy_host for IPv6"

8 years agoMerge "Allow the vnc server to bind on IPv6 address on computes"
Jenkins [Thu, 10 Mar 2016 21:15:43 +0000 (21:15 +0000)]
Merge "Allow the vnc server to bind on IPv6 address on computes"

8 years agoMerge "Surround MongoDB IPs with braces in the connection string if IPv6"
Jenkins [Thu, 10 Mar 2016 21:09:57 +0000 (21:09 +0000)]
Merge "Surround MongoDB IPs with braces in the connection string if IPv6"

8 years agoMerge "Upgrades: initialization command/snippet"
Jenkins [Thu, 10 Mar 2016 18:34:01 +0000 (18:34 +0000)]
Merge "Upgrades: initialization command/snippet"

8 years agoMerge "Add a ceph-storage node upgrade script for the upgrade workflow"
Jenkins [Thu, 10 Mar 2016 18:32:47 +0000 (18:32 +0000)]
Merge "Add a ceph-storage node upgrade script for the upgrade workflow"

8 years agoSupport the deployment of Ceph over IPv6
Giulio Fidente [Mon, 25 Jan 2016 12:05:57 +0000 (13:05 +0100)]
Support the deployment of Ceph over IPv6

To deploy Ceph on IPv6, we need to enable ms_bind_ipv6 in addition
to passing the list of MON IPs in brackets.

Change-Id: I3644b8fc06458e68574afa5573f07442f0a09190

8 years agoFix vncproxy_host for IPv6
Javier Pena [Wed, 2 Mar 2016 09:57:50 +0000 (10:57 +0100)]
Fix vncproxy_host for IPv6

https://review.openstack.org/268356 can cause issues in IPv6
environments. It generates the following Hiera data:

nova::vncproxy::common::vncproxy_host: [2001:db8:fd00:1000::10]

which fails due to the brackets. Making sure there are no brackets
in nova_vncproxy_host makes it work for both the IP case and when
using DNS names.

Change-Id: Iafe18f042725eb9419d97cd674c4b9a1a895b187

8 years agoAllow the vnc server to bind on IPv6 address on computes
Marius Cornea [Thu, 21 Jan 2016 14:28:06 +0000 (15:28 +0100)]
Allow the vnc server to bind on IPv6 address on computes

Currently the vnc server on the compute nodes binds on 0.0.0.0.
which only works with IPv4 addresses, it breaks connectivity with
IPv6 addressing.

This fixes https://bugzilla.redhat.com/show_bug.cgi?id=1300678.

Change-Id: Id642d224fb3c62f786453dc684634adca1c2c09d
Co-Authored-By: Giulio Fidente <gfidente@redhat.com>
8 years agoSurround MongoDB IPs with braces in the connection string if IPv6
Giulio Fidente [Wed, 20 Jan 2016 12:01:21 +0000 (13:01 +0100)]
Surround MongoDB IPs with braces in the connection string if IPv6

Change-Id: I9ed917f32b3de95beb234ade4819a8b96affe3e9

8 years agoMerge "Upgrades: object storage node upgrade fix"
Jenkins [Thu, 10 Mar 2016 15:25:56 +0000 (15:25 +0000)]
Merge "Upgrades: object storage node upgrade fix"

8 years agoUpgrades: quiet yum upgrade on cinder nodes
Jiri Stransky [Thu, 10 Mar 2016 13:22:05 +0000 (14:22 +0100)]
Upgrades: quiet yum upgrade on cinder nodes

Yum update on cinder nodes should be quiet, as it is on controllers,
because results of these updates are sent to Heat. I mistakenly left
this out in the first patch because i used one of the standalone node
upgrade scripts as a copy/paste base for the cinder node upgrade script.

Change-Id: Id13190dc4d242317829c7994088183f52d21461d

8 years agoKeystone domain for Heat
Martin Mágr [Mon, 2 Mar 2015 10:04:20 +0000 (11:04 +0100)]
Keystone domain for Heat

This patch adds support for configuring Keystone domain for Heat
via heat-keystone-setup-domain script. It should be reverted
as soon as Keystone v3 is fully functional.

This patch won't be fully functional without either python-keystoneclient
fix [1] or workaround [2].

[1] https://bugs.launchpad.net/python-keystoneclient/+bug/1452298
[2] https://review.openstack.org/180563

Change-Id: Ie9cdd518b299c141f0fdbb3441a7761c27321a88
Co-Authored-By: Jiri Stransky <jistr@redhat.com>
Depends-On: Ic541f11978908f9344e5590f3961f0d31c04bb0c

8 years agoMerge "Upgrade of Cinder block storage nodes"
Jenkins [Thu, 10 Mar 2016 12:57:37 +0000 (12:57 +0000)]
Merge "Upgrade of Cinder block storage nodes"

8 years agoMerge "Updated the heat_template_version"
Jenkins [Thu, 10 Mar 2016 11:19:00 +0000 (11:19 +0000)]
Merge "Updated the heat_template_version"

8 years agoMerge "Fixup systemctl_swift stop/start during the controller upgrade"
Jenkins [Thu, 10 Mar 2016 10:22:25 +0000 (10:22 +0000)]
Merge "Fixup systemctl_swift stop/start  during the controller upgrade"

8 years agoWire in HAProxy stats user and password
Ben Nemec [Wed, 9 Mar 2016 23:29:45 +0000 (17:29 -0600)]
Wire in HAProxy stats user and password

Change-Id: I26b7a1cd1b7b6520db1df49c60a86c2bb5bce1b0
Depends-On: I12e835964a0370de73e45ef0a8603656ecb02d0c
Depends-On: I8a5844e89bd81a99d5101ab6bce7a8d79e069565

8 years agoEnable predictable IPs on non-controllers
Ben Nemec [Wed, 9 Mar 2016 16:41:03 +0000 (10:41 -0600)]
Enable predictable IPs on non-controllers

For the external loadbalancer work, we added the ability to specify
fixed ips for controller nodes on all network isolation networks.
In order to allow users full control over the placement and ip
addresses of deployed nodes, we need to be able to do the same thing
for the other node types.

Change-Id: I3ea91768b2ea3a40287f2f3cdb823c23533cf290

8 years agoRemove unused Neutron Agents container
Ryan Hallisey [Thu, 3 Mar 2016 15:33:36 +0000 (15:33 +0000)]
Remove unused Neutron Agents container

The Neutron Agents is currently not used. Refactor the heat templates
to accommodate for this change.

Change-Id: Ice3c5ce723fa16cfb66c2b0afbe51d7b282c3210

8 years agoAllow the containerized compute node to spawn larger VMs
Ryan Hallisey [Sat, 5 Mar 2016 01:28:24 +0000 (01:28 +0000)]
Allow the containerized compute node to spawn larger VMs

Atomic's root partition & logical volume defaults to 3G.
In order to launch larger VMs, we need to enlarge the root
logical volume and scale down the docker_pool logical volume.
We are allocating 80% of the disk space for vm data and the
remaining 20% for docker images.

Change-Id: If3fff78f476de23c7c51741a49bae227f2cdfe3e
Co-authored-by: Ian Main <imain@redhat.com>
Co-authored-by: Jeff Peeler <jpeeler@redhat.com>
8 years agoEnsure access to Redis is password protected
Giulio Fidente [Fri, 7 Aug 2015 11:03:05 +0000 (13:03 +0200)]
Ensure access to Redis is password protected

Depends-On: I1a8741b9e00775763911222cbe0af677b59e03a1
Change-Id: I373f97ada4e4101700a12b42dfb8ee4b2ff701f2

8 years agoRemove forced rabbitmq::file_limit conversion to string
Jiri Stransky [Fri, 9 Oct 2015 09:48:20 +0000 (11:48 +0200)]
Remove forced rabbitmq::file_limit conversion to string

Combined with a fix in puppetlabs-rabbitmq, we can lift the forced
conversion of rabbitmq::file_limit into a string in Hiera. See the
referenced puppetlabs-rabbitmq pull request for explanation of the
issue.

Change-Id: I0ec720b5e06763e86ea93f59cfe05842b3d13269
Depends-On: https://github.com/puppetlabs/puppetlabs-rabbitmq/pull/401

8 years agoMerge "Add Rabbit IPv6 only support"
Jenkins [Wed, 9 Mar 2016 14:31:40 +0000 (14:31 +0000)]
Merge "Add Rabbit IPv6 only support"

8 years agoUpgrades: object storage node upgrade fix
Jiri Stransky [Tue, 8 Mar 2016 09:51:31 +0000 (10:51 +0100)]
Upgrades: object storage node upgrade fix

The variables in the heredoc should be escaped because they should
evaluate only when the inner script runs, not when the outer "writer"
script runs.

Python-zaqarclient is installed for os-collect-config to work, as we do
on the other node types.

Swift-proxy is removed from list of services to stop/start, as
swift-proxy isn't supposed to run on the swift storage nodes.

Change-Id: I8426b859d11378ebdc3da94dcc090133dab0c628

8 years agoFixup systemctl_swift stop/start during the controller upgrade
marios [Wed, 9 Mar 2016 13:43:40 +0000 (15:43 +0200)]
Fixup systemctl_swift stop/start  during the controller upgrade

During the controller upgrade in
major_upgrade_controller_pacemaker_1.sh we use systemctl to stop
all swift services and then start them again in _pacemaker_2.sh

In the case of stand-alone swift nodes the deployer may have
used the ControllerEnableSwiftStorage: false so that only the
swift-proxy service is left on controllers (wrt swift). The
systemctl_swift function used during upgrades is changed to factor
this in.

Change-Id: Ib22005123429f250324df389855d0dccd2343feb

8 years agoUpgrades: initialization command/snippet
Jiri Stransky [Wed, 9 Mar 2016 10:18:40 +0000 (11:18 +0100)]
Upgrades: initialization command/snippet

This allows to run a command or a script snippet on all overcloud nodes
at the beginning of the upgrade. The intended use is to switch to a new
set of repositories on the overcloud. This is done differently in
different contexts (e.g. upstream vs. downstream), but generally it
should be simple enough to not warrant creation of switchable
"UpgradeInit" resource in the resource registry, and a string
command/snippet parameter should suffice.

Change-Id: I72271170d3f53a5179b3212ec9bae9a6204e29e6

8 years agoMerge "Add missing createUser line to /etc/snmp/snmpd.conf"
Jenkins [Wed, 9 Mar 2016 12:54:24 +0000 (12:54 +0000)]
Merge "Add missing createUser line to /etc/snmp/snmpd.conf"

8 years agoMerge "Change the CinderISCSIHelper to lioadm"
Jenkins [Wed, 9 Mar 2016 12:42:53 +0000 (12:42 +0000)]
Merge "Change the CinderISCSIHelper to lioadm"

8 years agoMake External Load Balancer templates work with IPv6
Dan Sneddon [Thu, 21 Jan 2016 09:41:46 +0000 (01:41 -0800)]
Make External Load Balancer templates work with IPv6

This change modifies the network isolation templates that allow for
fixed IP addresses on the controllers' IPs and VIPs, and makes them
compatible with IPv6 addresses.

The latest version of the patchset creates an from_service_v6.yaml
in order to properly handle service VIPs on IPv6 networks.

Note that since OVS is not currently compatible with IPv6 tunnel
endpoints, this patch does not yet enable IPv6 for the Tenant
network by default.

Change-Id: If881b000c6000ec13b54c0ee39f1c8940f079ae3
Co-Authored-By: Giulio Fidente <gfidente@redhat.com>
8 years agoMerge "Update enable-tls.yaml with new endpoints"
Jenkins [Wed, 9 Mar 2016 10:55:31 +0000 (10:55 +0000)]
Merge "Update enable-tls.yaml with new endpoints"

8 years agoAdd Rabbit IPv6 only support
Martin Mágr [Mon, 18 Jan 2016 13:57:43 +0000 (14:57 +0100)]
Add Rabbit IPv6 only support

Adds Rabbit environment variable required for IPv6 only support

Change-Id: Iec209ca83a5b0708ac828c4afaf9d2222e597f24

8 years agoMerge "Store events in Ceilometer"
Jenkins [Tue, 8 Mar 2016 21:46:34 +0000 (21:46 +0000)]
Merge "Store events in Ceilometer"

8 years agoMerge "Set host in nova.conf for compute nodes"
Jenkins [Tue, 8 Mar 2016 18:14:48 +0000 (18:14 +0000)]
Merge "Set host in nova.conf for compute nodes"

8 years agoChange the CinderISCSIHelper to lioadm
Giulio Fidente [Tue, 8 Mar 2016 16:41:13 +0000 (17:41 +0100)]
Change the CinderISCSIHelper to lioadm

Previously we used to always set CinderISCSIHelper to lioadm from
the client. With [1] we aim at cleaning up the client from any
default so this change move the actual default into the templates.

1. Iabc94ad1cce6f1744429aca9f866a59a698f5680

Change-Id: I662e4b0e3cf909ec2583d087ca1d246715caf0f7
Related-Bug: 1554564

8 years agoUpdate enable-tls.yaml with new endpoints
Ben Nemec [Wed, 2 Mar 2016 21:46:54 +0000 (15:46 -0600)]
Update enable-tls.yaml with new endpoints

A couple of new endpoints have been added, and if they're not in
the configured value for EndpointMap it will cause problems.

Sahara is not added as ssl-enabled because I don't believe it has
been added to the loadbalancer yet.

Note that there is work underway to CI overcloud SSL, which should
catch problems like this in the future.

Change-Id: Ia8a106fd94da7be8675ea84f5fbb9ac959771d10

8 years agoAdd a ceph-storage node upgrade script for the upgrade workflow
marios [Tue, 8 Mar 2016 12:30:59 +0000 (14:30 +0200)]
Add a ceph-storage node upgrade script for the upgrade workflow

This adds delivery of an upgrade script to any ceph-storage nodes
during the script delivery that comes first during the upgrade
workflow.

The controllers have the ceph-mon whilst the ceph-osds are on the
ceph-storage nodes. The ceph-mons will be updated first as part of
the heat-driven controller upgrade, and ceph-osds on ceph nodes are
upgraded with the upgrade-non-controller.sh tripleo-common script
as with compute and swift nodes.

Also slight rename for the ObjectStorageConfig/Deployment here for
consistency.

Change-Id: I12abad5548dcb019ade9273da06fe66fd97f54cc

8 years agoMerge "Permits configuration of Cinder enabled_backend via hieradata"
Jenkins [Tue, 8 Mar 2016 15:34:49 +0000 (15:34 +0000)]
Merge "Permits configuration of Cinder enabled_backend via hieradata"

8 years agoIncrease default netdev_max_backlog to 10x
Giulio Fidente [Mon, 8 Feb 2016 22:54:54 +0000 (23:54 +0100)]
Increase default netdev_max_backlog to 10x

It has been observed that on large clouds the netdev backlog
buffer might overflow. This change increases the default by ten
times.

The /proc/net/softnet_stat file contains a counter in the 2nd
column that is incremented when the netdev backlog queue overflows.
If this value is incrementing over time, then netdev_max_backlog
needs to be increased [1].

[1]. https://bugzilla.redhat.com/show_bug.cgi?id=1283676

Change-Id: Iec12324fd3a24e8b608b1e1849c270cc24cb0e60

8 years agoFix typos
Swapnil Kulkarni (coolsvap) [Fri, 8 Jan 2016 07:55:43 +0000 (13:25 +0530)]
Fix typos

Multiple files in t-h-t were having small typos.
Fixed in this patchset.
.
Change-Id: I82d7071747f47544990ed46e2be22931190406b3

8 years agoMerge "Fixup the memcached servers string in nova.conf for v6"
Jenkins [Tue, 8 Mar 2016 09:11:04 +0000 (09:11 +0000)]
Merge "Fixup the memcached servers string in nova.conf for v6"

8 years agoStore events in Ceilometer
Richard Su [Thu, 3 Mar 2016 02:14:44 +0000 (18:14 -0800)]
Store events in Ceilometer

Adds an option to toggle whether events are stored in Ceilometer.

The default is false and remains unchanged.

Change-Id: I9accb679b97cf7172c67f3cd5f218bdaee507c94
Related-Bug: rhbz#1312373

8 years agoMerge "Add an environment to use a swap partition"
Jenkins [Tue, 8 Mar 2016 05:12:07 +0000 (05:12 +0000)]
Merge "Add an environment to use a swap partition"