Age | Commit message (Collapse) | Author | Files | Lines |
|
|
|
|
|
|
|
We need to tag the HA containers with a special tag so
that the RA definition never changes. We do this step in THT
as opposed to puppet because we need to guarantee
that all images are tagged on all nodes *before* step 2 where the bundle
gets created.
NB: Getting the image name without the tag will require some more
yaql work to get all the cases right. Right now this works only
if we enforce that the image has a ':tag' at the end of the name.
So far this is always the case. If things change we will need to
amend this code.
Co-Authored-By: Damien Ciabrini <dciabrin@redhat.com>
Co-Authored-By: Sofer Athlan-Guyot <sathlang@redhat.com>
Change-Id: I362e6cf26fba77d3f949b7d2fc4b35a3eab9087e
|
|
|
|
|
|
|
|
I2c39a2957fd95dd261b5b8c4df5e66e00a68d2f7 changed nova api to http from
eventlet, however we need to continue running the eventlet service as
it is required for the nova metadata api.
However this should be tied to the OS::TripleO::Services::NovaMetadata
service, so duplicate the required config in nova-metadata.yaml.
Change-Id: I398575d565d5527bcaa1c8b33b9de2e1e0f2f6fd
Depends-On: Id3407e151566d16c6ae1e1ea8c1b021dac22e727
Closes-bug: #1711425
|
|
|
|
|
|
|
|
|
|
|
|
Previously what we've been doing with setup_docker_host.sh can now be
achieved with host_prep_tasks, and we can free up the NodeUserData
interface for other use cases.
Closes-Bug: #1711387
Change-Id: Iaac90efd03e37ceb02c312f9c15c1da7d4982510
|
|
|
|
This service allows configuring and deploying manila-share
containers in a HA overcloud managed by pacemaker.
The containers are managed and run by pacemaker. Pacemaker runs the
standard Kolla image but overrides the initial command so that
it explicitely calls manila-share. This way, we shield ourselves
from any unexpected future change in Kolla.
This container needs to use the 'docker_config' section to invoke
puppet (as opposed to 'docker_puppet_tasks'), because due to the HA
composability each resource creation needs to happen on the bootstrap
node of that service and 'docker_puppet_tasks' will only run on the
controller/primary role.
Based on work done in fdb233e64e3d78014dd7e351abfed5aec5035866
Partial-Bug: #1668922
Change-Id: Ifa94c506db5eb667690a19d594115a93d2a790b2
Depends-On: I797eea2f7788f65411964ccb852b5707e916416f
|
|
Pre existing Ceph clusters are migrated to containers using a
playbook in ceph-ansible which requires setting some 'ireallymeanit'
variable.
1. https://github.com/ceph/ceph-ansible/issues/1758
Change-Id: I5c2f46b91cf032913931275ce62315f293f21c8b
Closes-Bug: #1711159
|
|
Based on puppet/services/ceph-mds.yaml. Nodes in the CephMds role
will already be in the Ansible inventory but this change provides
a way pass their parameters to ceph-ansible.
Co-Authored-By: Giulio Fidente <gfidente@redhat.com>
Change-Id: Ia3ef9e9a2b159dacea01e38762145ff2bcc7ba27
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Pass mode parameter to ceph-ansible in place of ACLs parameter
because ACLs are not for same UID in container as container host
and because ACLs are not passed by kolla_config.
Change-Id: I7e3433eab8e2a62963b623531f223d5abd301d16
Closes-Bug: #1709683
|
|
Bind mounts and adds the appropriate permissions for the cert and
key that's used for TLS.
bp tls-via-certmonger-containers
Change-Id: I7fae4083604c7dc89ca04141080a228ebfc44ac9
|
|
This bind mounts the certificates if TLS is enabled in the internal
network. It also disables the CRL usage since we can't restart haproxy
at the rate that the CRL is updated. This will be addressed later and
is a known limitation of using containerized haproxy (there's the same
issue in the HA scenario). To address the different UID that the certs
and keys will have, I added an extra step that changes the ownership
of these files; though this only gets included if TLS in the internal
network is enabled.
bp tls-via-certmonger-containers
Depends-On: I2078da7757ff3af1d05d36315fcebd54bb4ca3ec
Change-Id: Ic6ca88ee7b6b256ae6182e60e07498a8a793d66a
|
|
The containerized version of the mongodb service omits the
metadata_settings definition [1], which confuses certmonger when
internal TLS is enabled and make the generation of certificates fail.
Use the right setting from the non-containerized profile.
[1] https://review.openstack.org/#/c/461780/
Change-Id: I50a9a3a822ba5ef5d2657a12c359b51b7a3a42f2
Closes-Bug: #1709553
|
|
Various containerized services (e.g. nova, neutron, heat) run initial set up
steps with some ephemeral containers that don't use kolla_start. The
tripleo.cnf file is not copied in /etc/my.cnf.d and this can break some
deployments (e.g. when using internal TLS, service lack SSL settings).
Fix the configuration of transient containers by bind mounting of the
tripleo.cnf file when kolla_start is not used.
Change-Id: I5246f9d52fcf8c8af81de7a0dd8281169c971577
Closes-Bug: #1710127
Co-Authored-By: Juan Antonio Osorio Robles <jaosorior@redhat.com>
|
|
So far we've been using virtlogd running on the host, we should now be
using virtlogd from a container.
Co-Authored-By: Martin André <m.andre@redhat.com>
Co-Authored-By: Jiri Stransky <jistr@redhat.com>
Change-Id: I998c69ea1f7480ebb90afb44d6006953a84a1c04
|
|
Splitting by colon using native str_split function did not work well
because we needed a right split.
This change replaces the str_split calls with yaql rightSplit().
Change-Id: Iab2f69a5fadc6b02e2eacf3c9d1a9024b0212ac6
|
|
The ip address which clients and other nodes use to connect to the
monitors is derived from the monitor_interface parameter unless
a monitor_address or monitor_address_block is given (to set mon_host
into ceph.conf); this change adds setting for monitor_address_block to
match the public_network so that clients attempt to connect to the mons
on the appropriate network.
Change-Id: I7187e739e9f777eab724fbc09e8b2c8ddedc552d
Closes-Bug: #1709485
|
|
|
|
|
|
After merging commit 488796, single quotation marks
were missed. This causes the upgrade to fail as the
flag --sacks-number is considered a su command flag.
Also mounts Ceph config data into the container which
seems needed for the gnocchi-upgrade command when
configured to use Ceph.
Also move the gnocchi db sync to step 4, so ceph is
ready. Add a retry loop to ceilometer-upgrade cmd so
it doesnt fail while apache is restarted.
Closes-Bug: #1709322
Change-Id: I62f3a5fa2d43a2cd579f72286661d503e9f08b90
|
|
If we consolidate these we can focus on one implementation (the new ansible
based one used for docker-steps)
Change-Id: Iec0ad2278d62040bf03613fc9556b199c6a80546
Depends-On: Ifa2afa915e0fee368fb2506c02de75bf5efe82d5
|
|
This bind mounts the necessary files for the mongodb container to serve
TLS in the internal network.
bp tls-via-certmonger-containers
Change-Id: Ieef2a456a397f7d5df368ddd5003273cb0bb7259
Co-Authored-By: Damien Ciabrini <dciabrin@redhat.com>
|
|
With these two services running over httpd in the containers, we can now
enable TLS for them.
bp tls-via-certmonger-containers
Change-Id: Ib8fc37a391e3b32feef0ac6492492c0088866d21
|
|
The non-containerized version will run over httpd [1], and for the
containerized TLS work, it is needed in the container version as well.
[1] Iac35b7ddcd8a800901548c75ca8d5083ad17e4d3
bp tls-via-certmonger-containers
Depends-On: I1c5f13039414f17312f91a5e0fd02019aa08e00e
Change-Id: I2c39a2957fd95dd261b5b8c4df5e66e00a68d2f7
|
|
In non-containerized deployments, Galera can be configured to use TLS
for gcomm group communication when enable_internal_tls is set to true.
Fix the metadata service definition and update the Kolla configuration
to make gcomm use TLS in containers, if configured.
bp tls-via-certmonger-containers
Change-Id: Ibead27be81910f946d64b8e5421bcc41210d7430
Co-Authored-By: Juan Antonio Osorio Robles <jaosorior@redhat.com>
Closes-Bug: #1708135
Depends-On: If845baa7b0a437c28148c817b7f94d540ca15814
|
|
After creating glance image successfully, share location
was remaining empty because the NFS volume on controller was
not mounted to docker container.
Now, connecting NFS volume to the docker container.
Change-Id: Ib45f117cbbf2b7b2c0faf024e9a8b049c440d872
Closes-Bug: 1708629
|
|
In non-containerized deployments, HAProxy can be configured to use TLS for
proxying internal services.
Fix the creation of the of the haproxy bundle resource to enable TLS when
configured. The keys and certs files are all passed as configuration files and
must be copied by Kolla at container startup.
For the time being, disable the use of the CRL file until we find a means
of restarting the containerized HAProxy service when that file expires.
Change-Id: If307e3357dccb7e96bdb80c9c06d66a09b55f3bd
Depends-On: I4b72739446c63f0f0ac9f859314a4d6746e20255
Closes-Bug: #1709563
|
|
Run virsh secret-define and secret-set-value in an init step
instead of relying on the puppet-nova exec.
Co-Authored-By: Jiri Stransky <jistr@redhat.com>
Change-Id: Ic950e290af1c66d34b40791defbdf4f8afaa11da
Closes-Bug: #1709583
|
|
In non-containerized deployments, RabbitMQ can be configured to use TLS for
serving and mirroring traffic.
Fix the creation of the rabbitmq bundle resource to enable TLS when configured.
The key and cert are passed as other configuration files and must be copied by
Kolla at container startup.
Change-Id: I8af63a1cb710e687a593505c0202d717842d5496
Depends-On: Ia64d79462de7012e5bceebf0ffe478a1cccdd6c9
Closes-Bug: #1709558
|
|
|
|
|
|
|