2.1.4 Begonia
==============

This is the fourth patch in the Begonia 2.1 series.

Notable Changes
----------------

* Feature: Upgrade btx image to 2.1.0.
  (`btx image <https://hub.docker.com/r/jijisa/btx/tags>`_)

    - Upgrade kubectl to v1.30.3.
    - Upgrade krew from v0.4.3 to v0.4.4.
    - Upgrade helm from v3.13.1 to v3.14.2.
    - Upgrade netapp tridentctl from v24.02.0 to v24.06.1.

* Feature: Support NetApp NFS over RDMA
  (`issue #261 <https://github.com/iorchard/burrito/issues/261>`_).
  We ran :doc:`NetApp TCP vs. RDMA Performance Test 
  <../operation/netapp_tcp_rdma_perf>`.

* Feature: Upgrade descheduler to v0.31.0
  (`issue #257 <https://github.com/iorchard/burrito/issues/257>`_)

* Bug: Modify allowed_direct_url_schemes and image_upload_use_cinder_backend
  when powerflex is the storage backend.
  (`issue #256 <https://github.com/iorchard/burrito/issues/256>`_)
  (`issue #259 <https://github.com/iorchard/burrito/issues/259>`_)

* Feature: Support :doc:`multiple glance stores
  <../operation/multiple_glance_stores>`
  (`issue #245 <https://github.com/iorchard/burrito/issues/245>`_)


Changelog
----------

* add nconnect=16 to NFSv4 options in netapp_vars.yml; (jijisa@iorchard.net)
* add qemu kill ansible ad-hoc task in clean_openstack.sh script; (jijisa@iorchard.net)
* modify replicas.api in glance.yml.j2; (jijisa@iorchard.net)
* update nova.nfs_mount_options in vars.yml.sample; (jijisa@iorchard.net)
* add unmount /var/lib/nova/instances on compute nodes task in clean_openstack.sh script; (jijisa@iorchard.net)
* update ceph glance keyring in glance storage init script; (jijisa@iorchard.net)
* update ceph cinder keyring in cinder storage init script; (jijisa@iorchard.net)
* update netapp_vars.yml; (jijisa@iorchard.net)
* update netapp_vars.yml; (jijisa@iorchard.net)
* add nova_nfs_mount_options; (jijisa@iorchard.net)
* apply rbd1: null in cinder.yml.j2 since it is renamed to ceph; (jijisa@iorchard.net)
* add powerstore_nfs_mount_options in powerstore_vars.yml; (jijisa@iorchard.net)
* update vars.yml.sample; (jijisa@iorchard.net)
* openstack(cinder): change ceph volume type to ceph from rbd1; (jijisa@iorchard.net)
* btx: upgrade to 2.1.0; (jijisa@iorchard.net)
* modify nfs_mount_options to nova.nfs_mount_options in nova.yml.j2; (jijisa@iorchard.net)
* add powerstore nfs protocol and nova.nfs_mount_options; (jijisa@iorchard.net)
* ceph: add osd service_id in cluster.yml.j2; (jijisa@iorchard.net)
* fix(cinder): duplicate mounts when enable_iscsi is true and powerflex is in storage_backends; (jijisa@iorchard.net)
* add nfs_mount_options in libvirt section of nova.yml.j2; (jijisa@iorchard.net)
* modify cinder nfs shares; (jijisa@iorchard.net)
* move pod.security_context.cinder_volume in enable_iscsi block before powerflex mounts declaration; (jijisa@iorchard.net)
* modify replicas.api in glance.yml.j2 since we support glance multi-stores; (jijisa@iorchard.net)
* add cinder.volume_capabilities variable in vars.yml.sample; (jijisa@iorchard.net)
* fix duplicate /dev mount in nova.yml.j2 when powerflex is in storage backends and enable_iscsi is true; (jijisa@iorchard.net)
* fix cinder.yml.j2; (jijisa@iorchard.net)
* image_volume_cache_enabled to true for powerflex; (jijisa@iorchard.net)
* issue(#259): fix volume-to-image operation for powerflex (#260); (jijisa@iorchard.net)
* feature/openstack/glance multi stores (#258); (jijisa@iorchard.net)
* feature(issue #257): include and upgrade a descheduler image; (jijisa@iorchard.co.kr)
* fix(issue #256): change cinder_allowed_direct_url_schemes.powerflex to null in vars.yml.sample; (jijisa@iorchard.co.kr)