Age | Commit message (Collapse) | Author | Files | Lines |
|
|
|
This will prevent the Ceph Mon upgrade script from starting if the
Ceph cluster is in error state.
It also adds a parameter to ignore warning states, useful when
performing an upgrade of a cluster where the number of healthy
OSDs does not guarantee the desired replica size.
Closes-Bug: 1618533
Change-Id: I1beb8ad0812f19b1018ba19b5a9fc85fa132d7f7
|
|
Upgrades the ceph-osd daemon from Hammer to Jewel
Change-Id: Idfa90fdc0052c53f448401c85c5d13a2ba68acd1
|
|
This adds an option to configure the global_physnet_mtu in
neutron.conf. This controls the base MTU for the physical network
which is used as a basis for the new auto calculated overlay network
MTU's.
For flat and VLAN tenant networks there has been a subtle breakage
in TripleO in that users of those network types may now need to
manually set NeutronGlobalPhysnetMtu to something less than the
physical network to imitate the previous behavior of the
now removed NeutronTenantMtu setting (removed in git
2a64b67cef74fff86ce6b56b15431b859515844d). Previously users of
say a vlan tenant network in TripleO would automatically get
the tenant network set to 1400.
Change-Id: I25c2435feb373e11e4fc3a2821db0c96211468e0
|
|
Configure Heat with new authtoken parameters provided by puppet-heat
module.
Also move the authtoken parameters in heat-base, required by both
heat-engine and heat-api.
Depends-On: I7a9d10b0a8092b3d0c70708b57637098cfe2b56d
Change-Id: Id5cd3e830f64263c8d39c7d2973c99d0380e772c
|
|
This patch moves the settings for Nova, Neutron, and Horizon
out of controller.yaml.
Also fixes the NovaPassword settings in nova-base.yaml
so they don't use get_input.
Also, creates a new apache.yaml base service to contain shared
apache settings for several services which use Apache for WSGI.
Co-Authored-By: Giulio Fidente <gfidente@redhat.com>
Change-Id: I35d909bd5abc23976b5732a2b9af31cf1448838e
Related-bug: #1604414
|
|
|
|
The CloudName's default value only included the shortname which is
not a very good practice. This now sets localdomain as the default
domain there.
Change-Id: Ic46b34d94a602aa948b873062015286cf49f7c3c
|
|
Change-Id: Ibc86d4dd0aa576345e3f174ea1adca3e6248e81e
Depends-On: I5c620ba717f782b39c599aff24b4ac56fb695a04
|
|
Pass the list of ceph nodes to the ceph_mon profile via
the service template - this requires some fixup to the
profile to handle the ipv6 case.
Note this also aligns the ServiceNetMap keys so that the
composable node_ips logic will generate the lists when
the ceph_mon service is enabled.
Change-Id: If8a5c65f17e677fe62243b3aa746fd642f72d2b0
Depends-On: I481dd2cd2cde7f1491080e6d9c7dcb7047c22de1
Partially-Implements: blueprint custom-roles
|
|
|
|
|
|
Added an environment file to configure DPDK with OVS
by overriding ComputeNeutronOvsAgent. Also added nic
configs for configuring DPDK bridge and bond with
numbered nic format.
Implements: blueprint tripleo-ovs-dpdk
Co-Authored-By: Vijay Chundury <vchundur@redhat.com>
Change-Id: I82b6f66394a8928f8524706c939508edd08afa9b
|
|
Allows the operator to alter the memory reservation for host processes
Implements: blueprint tripleo-ovs-dpdk
Change-Id: If7a099cde2e8854a14a75c1304620cc444eefcf2
Signed-off-by: karthik s <ksundara@redhat.com>
|
|
This patch removes the get_input's in the service templates
and instead uses our new bind IP string replacement mechanism
to correctly set the local bind IPs for the Ironic API.
It also avoids using MysqlNetwork to set Ironic my_ip.
This patch adds a new IronicNetwork setting which is used
to control which network Ironic will use for its PXE network
traffic.
Change-Id: Ic848ec77b0106279c5d963504798c05110aa1440
|
|
|
|
Adds a pre-requisite software deployment to the pacemaker scenario
upgrade which, before the openstack services are upgraded,
upgrades the ceph-mon daemon from Hammer to Jewel.
Change-Id: I9855d80a6ae156b4a9e0409c3c927068b9db95a0
|
|
This would be useful to test Ceph in CI without a dedicated OSD node.
Change-Id: Ie534d327a9418b6119f5811ee62d448795879fb0
|
|
|
|
|
|
|
|
With the introduction of the global config, HAProxy can know which
endpoints to expose. It will get those via hieradata.
Change-Id: I0f5ce8c7394625e8dce691ae1996457a9ea99a99
|
|
|
|
|
|
This will be read by the client and it will furtherly skip the post-config.
Depends-On: I36e1c478e7c92be61da6a0d710e9025d4d354072
Depends-On: Ibbbe1e016e7e47ea5577de33c5a11d4088811616
Change-Id: Icfb650a1dff704cdcce7349dfb612298d38f2706
|
|
|
|
|
|
Currently we have a hard-coded list of ips for various services that
run on the controller, instead we can dynamically generate that list
of per-service ips, initially only for the controller but this approach
can be extended so it works for any role.
Change-Id: I3c8a946e439539d239ad7281a1395414df0893eb
Partially-Implements: blueprint custom-roles
|
|
For compatibility with the custom-roles patches which convert
hard-coded hiera mappings (such as opendaylight_api_node_ips) to
data generated based on the service_name, we need to either change
this name to match the hiera (node_ips and vip) keys, or change the
hiera keys to match the service. I took the former approach because
it involves less juggling patch dependencies between t-h-t and p-t.
Change-Id: Ic179550027f37946097afe0e5b9f504ea19fa7bc
Partially-Implements: blueprint custom-roles
|
|
This adds a list of all enabled service_names in the
enabled_services key, and also generates some boolean
values e.g service_name_enabled, which is more convenient
for some usage (such as haproxy where we need an easy way to
set a flag saying if a given service is enabled)
Partially-Implements: blueprint custom-roles
Change-Id: I62273f403838893602816204d9bc50d516c0057f
|
|
It allows the operator to configure pci_passthrough_whitelist
in nova.conf for each of the compute nodes.
implements: blueprint tripleo-sriov
Depends-On: I5ed53cfffe80dbbbb9dcee7c2ea6037afbed2382
Change-Id: Ic5e099fe788046363536f913272b2814abe165fa
Signed-off-by: karthik s <ksundara@redhat.com>
|
|
|
|
|
|
|
|
|
|
This gnocchi-base service was using get_input to obtain
the debug setting which won't work here. This patch adds in
the standard Debug heat parameter to the service and sets
it via get_param.
Change-Id: Idfb772567d5711083b82ae8cb4bce2c4498b4d64
|
|
Allow the operator to configure sriov_numvfs for each of the
required SR-IOV interfaces
Co-Authored-By: Vijay Chundury <vchundur@redhat.com>
Implements: blueprint tripleo-sriov
Depends-On:I7450b904475bdf46498d9af633416b3eba12f761
Change-Id: I07fd9fbb1268f3a189fcadee9f30f96ba476e533
Signed-off-by: karthik s <ksundara@redhat.com>
|
|
|
|
|
|
We have to recreate the /var/lib/mysql directory on all controller node,
not just the boostrap node for the cluster to be able to restart.
Adding a warning on the fact that those script are local and know
nothing about the good upgrade state of the other nodes.
Closes-Bug: 1612642
Change-Id: I48e2812d7df80bbf2db53a8b71dc434d4209a160
|
|
There is a typo in the code, making this test always successful.
Closes-Bug: 1614437
Change-Id: Ia6b0b156294de9fcb8f66fc46aa8801555775a56
|
|
scheduler_host_manager doesn't take
nova.scheduler.host_manager.HostManager as a value anymore. This fix it
before restarting the service.
Change-Id: Ia9adcfd5a898f0c712b4a37ae33db88a44630f0d
Closes-Bug: 1615035
|
|
Skips the unused parameter warning for required parameters.
Change-Id: I71ad4ab9f6e6c63e3f01b8cc9c72262f1958331e
|
|
|
|
|
|
|
|
|
|
Introduces environment files for deploying OpenDaylight in two ways:
- ODL only managing L2 as an ML2 plugin
- ODL managing L2 and L3 DVR, by replacing NeutronL3Agent
Two services are added. One to install ODL and configure OVS on the
Controllers, and another service to only configure OVS on compute nodes.
Paritally-Implements: blueprint opendaylight-integration
Depends-On: I666dc0874f1d11a72a62d796f4f6d41f7aa87a3f
Change-Id: Ide69e20cbf2ec6151953cb23e51478b770aca17f
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
|
|
Depends-On: If3feb859b527d08e10c124b5ad2f7f4b1f19156a
Change-Id: I728d0980f7742aa390f11a0f0b8598d1299e2886
|