This service is supported in upstream now and being added into
puppet-swift. This needs to be removed and should be managed via
swift::storage::object instead if necessary.
Change-Id: I46c4c5ca0ee86404739876afe2a7110cd59fb362
Closes-Bug: #1698424
This patch removes hardcoded type for 'baremetal' Neutron network.
Update ironic conductor config with new config option:
* enabled_network_interfaces - list of enabled network interfaces on
Ironic conductor
* cleaning_network_name - name of Neutron network to be used
during node cleaning
* provisioning_network_name - name of Neutron network to be used
during node provisioning
Add openrc generation task for nodes with ironic role, needed as it is
required for translation from name to uuid.
Change-Id: I63da8332cade0e76bdd687f0522bbc4c3006a68f
Related-Bug: #1588380
Upstream commit https://review.openstack.org/#/c/446925 removes
enabled_drivers option from ::ironic class, now we need to use
it from ::ironic::conductor class.
Change-Id: I1e49d6777de73a9c0267a7c982cb19253bb40e64
Closes-Bug: #1675684
Upstream patch https://review.openstack.org/447533 removes
deprecated parameters from puppet-nova, so we need update
them in fuel-library
Change-Id: I1c5c2d86e5ac100c2635cae8ee3506d8719d56f9
Closes-Bug: #1674375
The nova allocation ratio settings have been moved from the scheduler to
the main nova class. They are being removed from the scheduler class and
need to be passed to nova differently.
Change-Id: I753c795ef62a505b7b2499b3805acb43d1f79b4d
Closes-Bug: #1673893
As of firefly (v0.80), ceph object gateway is running on civetweb
(embedded into the ceph-radosgw daemon) instead of apache and fastcgi.
Using civetweb simplifies the ceph object gateway installation and
configuration.
Change-Id: Idba61e094390e3c75a6e5d9b35a8e8e47a2a696f
Closes-Bug: #1671808
Upstream will remove enable_notifications parameter [1],
we need to use notification_driver parameter now.
[1] https://review.openstack.org/#/c/444051/
Change-Id: I95fbb5f104b3d550a87aaaad51418c918fe70c79
As Swift is deployed behind the proxy, url_base need to be
set, in order to let Swift know what protocol is used on
proxy.
Change-Id: I3640f0190cb2a13c1ad4b2c9f71090f09681a14d
Closes-Bug: #1537071
Nova placement API is recommended in Newton and is mandatory in Ocata.
Scheduling will not work without it. We configure services according to:
https://docs.openstack.org/developer/nova/placement.html
Change-Id: Ic014ac162f50324c0341cc3013cd093d7125d53e
Closes-bug: #1670354
contrib.revoke.sql module has been removed, we need
to use default value in keystone.conf
Change-Id: I0040565f66b36c28f0280c9a3f534d53825ad1b3
Closes-bug: #1670206
When we provision nodes, we can
restart dnsmasq to frequently
this leads to systemd not starting
dnsmasq again, thus we have dnsmasq
stopped and deployment failing.
Raise limit to 100 starts in 10 seconds
and also sync cobbler on netboot disable
Partial-bug: #1667006
Change-Id: Id14bb2bb162f0a9fc6e0a9a102d98f4f2a6dcf1a
Upstream patch https://review.openstack.org/#/c/438061/
deprecates keystone::endpoint::version in order to create
versionless keystone endpoints. Fuel ostf are not ready, that's
why keystone::endpoint::version should be set explicitly
to work around this problem.
Change-Id: If7713b6c3e78f64bdeb29dfeebce24a24b4c364b
Switch back to using all available mamcached servers, because of
failures during swift testing.
Change-Id: I251b465dadd8ffe9a15b1a7092168d89acd2d72b
Closes-Bug: 1666837
Earlier metadata_workers option was set to processorcount
fact by default, but with upstream patch [1]
it will be switched to os_workers fact,
so in order to keep the same amount of
nova metadata workers we need to set it directly
in openstack_controller task.
[1] https://review.openstack.org/#/c/383690/
Change-Id: I95bc3d7b5f4c71fc9829b0a58d246ac4ea2fe903
Closes-Bug: #1667626
As heat is placed behind the proxy we need to enable
proxy headers parsing, so heat is aware about
protocol used to connect to endpoint.
Change-Id: I52e49ab9d866891a81d3c08b82c50c8addd34786
Closes-Bug: #1668227
As cinder is placed behind the proxy we need to enable
proxy headers parsing, so cinder is aware about about
protocol used to connect to endpoint.
Change-Id: If68de7d52617803b379ed564f07ed6c947315dbd
Closes-Bug: #1665361
Make nova user configuration before ceilometer/compute task since
ceilometer-common package reset shell of nova user to /bin/false back.
Change-Id: I92be4cb23ad4ffe2cbd52562b40e336a1ad1e352
Closes-Bug: #1660981
VMware not supported since Fuel 10. So this commit removes
vmware-related manifests.
Implements: blueprint remove-vmware
Change-Id: If5f0c837fe843f9cfd30648fe6e115267467982b
This change is needed for cases with big time interval
between Ceilometer services which need a db and
database backend starting.
Otherwise, Ceilometer API defines storage connection to None
and all storage requests fail permanently.
Change-Id: Ic442ff42e225f98a7a06469b2ff77d95fac84e2b
Closes-bug: #1611023
Keystone_authtoken parameters were deprecated for
authenticating in neutron, we need to switch to neutron/*
auth options using ironic::neutron class.
Change-Id: I72c2fd5241fe3e171fac8bd61d4000ab4415daa3
Closes-Bug: #1662158
This file is huge if there are many nodes in cluster. Now
we generate it and store in memory for each node and it
takes a lot of time and memory.
Let's split it into common and node-specific parts so that
we can generate common part once and copy it from master
node to each node affected by deployment.
Change-Id: I114a8598460bef9063e43505f46fe38cd564d8c7
Closes-Bug: #1660308
Current value of 5 results in socket operation timeout after ~12.9
seconds. This is a bit too low, e.g. we've seen RabbitMQ network
splits in production.
This 12.9s amount is equal as 0.2*(2^1+2^2+..2^5), where 0.2 is a
retry timeout (RTO) that is calculated by kernel on a per-socket
basis. But in fast local networks it usually almost equal to minimum
values of 0.2s hardcoded in linux kernel (and BTW, RFC says that
minimum value should be 1s).
On the other hand, comment in netconfig.pp says that our target
timeout is ~54 seconds. And changing tcp_retries2 to 7 is consistent
with that comment - tests an live env show that resulting timeout is
~52.2s
Change-Id: Ib52f40ef1017a9da5a29cd62fb744a4597860763
Ensure that each compute node is configured with ssh key
authentication so that the nova compute service can use ssh
to move/resize disks to other nodes.
Change-Id: Ifbefdcbcf865cb2be988e133f19a13a5e010512d
Closes-Bug: #1660981
min_l3_agents_per_router was deprecated in Newton and removed
in Ocata.
Change-Id: If961e7abeadd77ef0f9d464cd2f5e9465ed6511b
Closes-Bug: #1658675
Partial-Bug: #1477967
This option is deprecated in Newton and removed [1] in Ocata
from Neutron code.
Now Neutron always adverises mtu by default.
[1] https://review.openstack.org/#/c/413567/
Change-Id: Iff15f85358c512f991612ee204f0f71fe8579335
Partial-Bug: #1477967
Andreas Jaeger