| Age | Commit message (Collapse) | Author | Files | Lines |
|
Adds Cyrus SASL packages necessary for the DIGEST-MD5 and SCRAM-SHA-256
mechanisms. These can be used for libvirt SASL authentication.
Change-Id: I13e19ca29eeab40cd08fa3afe2cdf7531867f81b
Partial-Bug: #1964013
(cherry picked from commit e1efa9d257c45e5f99364974ea66a6d6ba5a5605)
|
|
This is part of the effort to allow a migration path
for operators using CentOS 7 and Train today.
Depends-On: https://review.opendev.org/c/openstack/kolla-ansible/+/831265
Change-Id: I75373eb21c4269c20f72603ad4cf162a472de25b
(cherry picked from commit 67d6d28b6208cc004a50f19be1386d88c2404bc4)
|
|
|
|
(cherry picked from commit 405fca90f2c8d3763ac56ce8b925611d136bba6c)
Change-Id: I95ea4a66a80bf035f8bd1180790c7e0c85d759ed
(cherry picked from commit 07f519996381789c1f21850cfc9b27f3b86a6f4c)
|
|
The Log4j version was bumped on GitHub [1] but it is still pending
inclusion in a release of Apache Storm.
Apply the alternative mitigation recommended by Log4j [2] of removing
the JndiLookup class from the classpath.
[1] https://github.com/apache/storm/pull/3427
[2] https://logging.apache.org/log4j/2.x/security.html
Change-Id: Ib3ecd73f9e39e320acb2c5f0962b8af9b1a817e9
(cherry picked from commit 448e4f56aa6218a66b0074329d322c0b6089b2e6)
|
|
There are two cases where pip installs kolla share into ~/.local/share/kolla:
- with `pip install --user kolla`
- with `pip install kolla` as non-root
This patch helps kolla find the share dir in these cases.
Closes-Bug: #1930544
Change-Id: I0b3eb6744a34a601266179fb358510a5c500e6ba
(cherry picked from commit db78657c231663f3e8ebbc18de43c06ddaa94c97)
|
|
Backport below Wallaby switching bullseye to buster.
Closes-Bug: #1931544
Change-Id: I8460a4159c8da8ae0b856867e068d30fbffb4d34
(cherry picked from commit 789c9c3b84e84796ac8211112fa29f415f560c88)
|
|
|
|
Infra mirrors get their indices rebuilt to avoid broken
indices (due to partial update).
Unfortunately, this wipes out the cryptographic signatures.
Our approach so far was disabling apt security features globally.
However, this is not a valid choice for external repos.
It hid an issue we introduced with new RabbitMQ repos missing
proper keys installed in the image.
This caused permanent failures outside of our CI.
Our process should be as close as possible to users' experience.
This patch makes CI trust only the mirrors that have their indices
rebuilt (so infra mirrors).
For Train, fixed missing override parts as well.
And it includes also:
- I8209770ca1353fd89aacef0859ca3a373645c6b2
- Icb745f628c1c50d9f20df022b08ddc46fccc270c
- Id3b55ac470a977be7ee7decdef3b49e8d89a130f
Change-Id: Ic5abc4b87fd76f87aba383abf43e95ba70629fcb
Co-Authored-By: Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org>
(cherry picked from commit c364c8be85f4653bf7af1c7ac0378823553e7441)
(cherry picked from commit 4a482380408d107c59119b379add34f4b8e87ad7)
|
|
|
|
bintray is being shut down 1st of May 2021, rabbitmq docs direct
users to packagecloud - so let's move.
For Victoria and below, includes I6a3ba599af3deedd450b4ffff6b86a0c04628340
(base: drop Linaro OBS key for CentOS and Ubuntu)
For Train, fixes also the order of rpm gpg key imports and
yum operations (otherwise CentOS 8 [with dnf] fails because repos
are enabled early and keys are missing).
And it also pins erlang version where required.
Change-Id: I499251d13ca77aea2215f8a879b7bb8dd51f1c9f
(cherry picked from commit 641b936bc7fd33203a80bb3827fce7c44d010476)
|
|
Backport to all supported branches.
Change-Id: I4a13bbe959e3324f05d5cf78a3be2415eb2ba7ec
(cherry picked from commit 004d9eb5d575fcb140b4d90fb30ced0a81dd3e66)
|
|
The chart repository is gone and the building fails.
Change-Id: I205eb2209cb0ccaecee247b95c65ed65bf4ed9e0
|
|
|
|
nova-compute uses daxio to cleanup vpmem backend device on instance
delete. If the daxio binary is missing in the nova-compute container
instance delete fails. daxio is provided in centos via daxio, in
ubuntu via the pmdk-tools package.
Conflicts:
docker/base/apt_preferences.debian
Also:
nova-compute: daxio on CentOS 8 is x86-64 only
Change-Id: Ifb5948653565e2ae902783762e20e33527020efe
Closes-Bug: 1907124
(cherry picked from commit 6176d54179109e85dee5ebc29a7726bcb05f76a9)
(cherry picked from commit 40a1c7a6ebabb1554e0ce7b54d6b356c2bac1ce5)
(cherry picked from commit 14b96abd6f6a6671d4dc97035b9ece61145cdcf2)
|
|
backup_full() saves the history name in $BACKUP_DIR/last_full_date.
backup_incremental() attempts to read from $BACKUP_DIR/last_full_date
into $LAST_FULL_DATE. After the read attempt; if $LAST_FULL_DATE is zero
length then $(date +%d-%m-%Y) is used.
Change-Id: I4172b814d5d95ae3debbc1cf3feaeee2e2bec9bd
Closes-bug: 1897948
(cherry picked from commit ccfe65b5f7fe437141f342421488c176038fddb0)
|
|
Backport this until Train.
CentOS 8.3 renamed system repos. [1]
This caused failures when disabling and enabling.
More details in the referenced bug report.
Notice CentOS 8.2 and below will no longer work.
Train backport also includes some fixes for the
fallout from the new pip resolver.
* cap bandit<1.6.3, since bandit no longer supports
Python 2.
* mark kuryr-libnetwork unbuildable for source
images.
* mark almanach-base & dragonflow-base unbuildable
for Debian & Ubuntu source images.
* Don't install networking-hyperv in source
neutron-server images. pip complains of a conflict.
(https://bugs.launchpad.net/kolla/+bug/1907826)
[1] https://wiki.centos.org/Manuals/ReleaseNotes/CentOS8.2011#Yum_repo_file_and_repoid_changes
Change-Id: I2d67b3ed7b7c521bc755ec9a87365c9962a9eeff
Closes-Bug: #1907213
Closes-Bug: #1907826
Co-Authored-By: Radosław Piliszek <radoslaw.piliszek@gmail.com>
Depends-On: https://review.opendev.org/766211
Depends-On: https://review.opendev.org/767157
|
|
This fixes the FC Cinder backend usage in Nova.
Change-Id: I036cfe3af3dd96625f986a820c78a10e6558b086
Closes-bug: #1884484
(cherry picked from commit 0c8702c96f852eae7a1fb784a260c01c15fb86f3)
|
|
|
|
The horizon image copies the masakari dashboard's policy file to Python
site packages, but it should go to /etc/openstack-dashboard. This allows
the dashboard to be seen by non-admins, although it fails to load.
We get an error like the following in horizon.log:
No policy rules for service 'instance-ha' in
/etc/openstack-dashboard/masakari_policy.json
This change fixes the issue.
Change-Id: I8ede183c76a830de06ce6524dc3f6f6944b182c1
Closes-Bug: #1894240
(cherry picked from commit 4f44870376f47c6a65bd02cc83d2ef39fd1f5ae9)
|
|
There are several issues with kolla_set_configs --check:
1. We calculate the destination path incorrectly when comparing a file
in a directory, due to passing arguments to os.path.relpath in the
wrong order
2. For directories that have not changed, we also attempt to compare
them as files, which fails when they are open()ed.
3. If the config JSON does not have a config_files key, it fails with a
KeyError.
The first two issues affect the fluentd container, which specifies
directories as the source, without using a glob. The third affects OVN
containers.
This patch fixes these issues.
Closes-Bug: #1890567
Change-Id: I8921befe51da4282121443849177a7ca5ebe8822
(cherry picked from commit c5320eb22386c34d83435fa0b9aa03b178996b52)
|
|
|
|
Train only.
In CentOS 8 Train we had a mix of rmq upstream and rdo packages,
erlang was from RDO and rmq (3.7) from rmq.
As this is not really tested by any other party, I propose to
unify the sources of both.
Change-Id: I52094ce9e3d98cbdedfdc23c8a43dd3093ce1771
Closes-bug: #1884034
|
|
Switch to YAJI JSON parser as suggested in [1].
[1] https://docs.fluentd.org/quickstart/faq#i-got-encoding-error-inside-plugin-how-to-fix-it
Closes-Bug: #1889065
Change-Id: I2ecd3b029b956e000feaaa9e10a169521ba53d66
(cherry picked from commit 556d4353aa846765557cfdd3c6e087a25248c517)
|
|
Train only.
The version of libvirt-python allowed by upper-constraints in Train
(5.7.0) does not support CentOS 8. This prevented the masakari-monitors
image from building.
This change modifies the upper constraints file in the masakari-monitors
image to use 5.10.0, the earliest version that does support CentOS 8.
Change-Id: Ia980be14fdf9c4fc15bb379b940ae491de035725
|
|
Logstash 6 introduced in Centos 8 in Train release comes with log4j2
configuration that does not remove old compressed logs after rotation
Log rotation config backported from Logstash 7 - Combination of Size
Based and Time Based policies. Deletion occurs after 30 days or
3000 MB log files size - whichever comes first.
Change-Id: Ib44f493aad0813a8aba916c7c353b60cab954b5f
(cherry picked from commit 15f2af1b0217da0a0281268442202f22915570a1)
Closes-Bug: #1886975
|
|
A bit like we did for I3e0e86026f5a4a78473bed824cd1682d3a020cd5 we
should remove the nss-systemd lookup from containers. The reasons for
this are as follows:
1) Just like for I3e0e86026f5a4a78473bed824cd1682d3a020cd5
when this nss module is triggered it tries to talk to dbus.
It triggers a bunch of selinux denials and it makes little sense
to open all containers to talk to dbus.
In particular, if a container is run as non-privileged and bind-mounts
/run from the host, we will hit selinux denials like the following:
type=USER_AVC msg=audit(1592337775.860:74119): pid=1284 uid=81 auid=4294967295 ses=4294967295 subj=system_u:system_r:system_dbusd_t:s0-s0:c0.c1023 msg='avc: denied { send_msg } for msgtype=method_call interface=org.freedesktop.DBus member=Hello dest=org.freedesktop.DBus spid=406228 scontext=system_u:system_r:container_t:s0:c162,c886 tcontext=system_u:system_r:system_dbusd_t:s0-s0:c0.c1023 tclass=dbus permissive=0 exe="/usr/bin/dbus-daemon" sauid=81 hostname=? addr=? terminal=?'UID="dbus" AUID="unset" SAUID="dbus"
2) It just makes little sense in a kolla-world to have containers
talk to dbus/systemd and it saves us some time when a lookup triggers
the systemd module for whatever reason. Especially because the
nss-systemd module does a few things which are not useful in a container
(ensures that the root and nobody users and groups remain resolvable,
SystemD's DynamicUser= feature, provide Lookup API via Varlink)
The sed regex gives us the wanted results:
$ diff -u /etc/nsswitch.conf.orig /etc/nsswitch.conf
--- /etc/nsswitch.conf.orig 2020-06-19 07:18:10.974580755 +0000
+++ /etc/nsswitch.conf 2020-06-19 07:20:12.260230103 +0000
@@ -53,9 +53,9 @@
# group: db files
# In order of likelihood of use to accelerate lookup.
-passwd: sss files systemd
+passwd: sss files
shadow: files sss
-group: sss files systemd
+group: sss files
hosts: files dns myhostname
services: files sss
netgroup: sss
Related-Bug: #1883849
Change-Id: I81e5b7abf4571fece13a029e25911e9e4dece673
(cherry picked from commit dc2ddfa9750f94430a5b04b62fc1fa6adcb234f8)
|
|
|
|
|
|
Kolla provides log aggregation, but no solution to set retention periods
for aggregated logs. It therefore accumulates log data indefinitely unless
a user manually intervenes. This change adds Elasticsearch Curator which
provides a mechanism for automating such retention periods [1].
[1] https://www.elastic.co/guide/en/elasticsearch/client/curator/current/about.html
The container contains cron to support running Curator periodically.
Change-Id: Ief2c554a64ef6cc971635d7e2a718f63c310fbf6
(cherry picked from commit 37de8920aa7394089731c0fae75034c2c3d137ca)
|
|
|
|
In the Centos8 images, Storm fails to start since it cannot find
Python. This patch ensures that it can by setting the PYTHON env
var.
Change-Id: If6663b2f6dfadfd6a6db5e4aaca4eb782b87161f
Closes-Bug: #1876461
|
|
On Centos 8 we have Elasticsearch and Kibana 6, but Logstash (used by
Monasca) is stuck on 2 which is not compatible. See the product
compatibility matrix [1]. This adds Logstash 6, so that Monasca
can use the full ELK 6 stack on Centos 8.
Note that Monasca doesn't officially support ELK 6, but it does work.
[1] https://www.elastic.co/support/matrix#matrix_compatibility
Partial-Bug: #1884090
Change-Id: Id6f79e301cc109a029913f5628bab1848c138e7f
|
|
Pulls in this fix:
https://review.opendev.org/#/c/736610/
Partial-Bug: #1885104
Change-Id: I014c9a52b2f61a86e3a8adafdf4783c3f58148ef
|
|
|
|
Change-Id: Ic1f94f2f69aa19455093526fab2951f1d556342d
|
|
`ndctl` is a utility for managing the nvdimm subsystem required
by Nova for attaching PMEM(persistent memory) namespaces to guests [0].
[0]: https://docs.openstack.org/nova/latest/admin/virtual-persistent-memory.html
Change-Id: I8ed1b6c1d0985b2a73206bd9249a5664cd80c912
Closes-Bug: #1870455
(cherry picked from commit f9ea23d940ec4ce35d503ac62f202daaa9db502d)
|
|
|
|
RDO builds of stable/train for CentOS 8 went into phase when we do not
need to use delorean builds.
EPEL is disabled like it is in Ussuri (RDO packages may conflict with
it).
Erlang for RabbitMQ is used from CentOS 8 'rabbitmq-38' repo while
RabbitMQ is still 3.7.* from same place as usual.
'python3-pika' is dropped from openstack-base as it is EPEL only package
and is not used by any image (dropped in Ussuri in
https://review.opendev.org/#/c/710003/ patch)
AArch64 is not supported until CentOS 8.2 release.
Change-Id: I9d3ed54f4d10ef7ad2d52ed3aeca15f3968cd163
|
|
In Train kolla switched to Python 3 in Debian and Ubuntu source images.
For services running under httpd with mod_wsgi, they should install
libapache2-mod-wsgi-py3 rather than libapache2-mod-wsgi. This was done
for most images, but cyborg and monasca were omitted. This change fixes
that.
Change-Id: I25cfa62dbf20490685617b4a4185bf95ab91725e
Closes-Bug: #1873421
(cherry picked from commit f5d31da12fa66cedfc2e270b8cd3be8e2bb9254b)
|
|
Glance API image lacked nfs programs/utils to be able to mount
NFS exports.
This affected Glance usage with Cinder backend when NFS was used
as Cinder backend.
Closes-Bug: #1868574
Change-Id: I3aac1e6a7f7c5f839b036f15713af949fba206cc
(cherry picked from commit cac9d46aa6dbac6631bdcae67db125a88e1c3234)
|
|
|
|
|
|
for C8" into stable/train
|
|
|
|
|
|
The CentOS 7 rabbitmq image contains 3.7.10, and the CentOS 8 rabbitmq
image contains 3.7.24. These versions are not compatible, which prevents
a rolling migration from CentOS 7 to CentOS 8.
This change adds a rabbitmq-3.7.24 image, available for CentOS 7 images
only. This provides a path for migration, by first upgrading all CentOS
7 rabbitmq containers to use this image.
Depends-On: https://review.opendev.org/720218
Change-Id: Icff279be07c4767dd0b18c697894373f46102bf1
Partially-Implements: blueprint centos-rhel-8
|
|
This fixes the issue when keystone-manage output included
backslashes and/or double quotation marks which broke JSON string.
Change-Id: Ifae18c407210c12745d29fc4c95dca69aeafe6a8
Closes-bug: #1866017
(cherry picked from commit 33d184a731458bb45d1206bdae4fb172629a76e7)
|
|
Kolla was missing vitrage-persistor image required by Vitrage
for data storage.
Change-Id: Id5e143636f9a81e7294b775f3d8b9134bee58054
Closes-Bug: #1869319
(cherry picked from commit ad06096448a585a16ad73b610b0adc81b597ccec)
|
|
Elasticsearch and Kibana only support CentOS 8 from version 6.x, but in
Train Kolla uses ELK 5.x.
This change adds new elasticsearch6 and kibana6 images based on ELK 6.x
only for CentOS 7 and 8. This enables two things:
1. provide elasticsearch and kibana for CentOS 8 in Train
2. provide a migration path from CentOS 7 to 8 that does not involve a
simultaneous major upgrade, by first upgrading CentOS 7 to ELK 6.x.
The new images are based on the upgrade to 6.x in Ussuri in
I8b52b2630a72393bf4b9c7095fc67db6c36c27fa.
This includes a switch to start using oss tagged binaries. These
binaries contain only features that are available under the Apache 2.0
license.
Related: blueprint elasticsearch-kibana-version-upgrade
Co-authored-by: Dincer Celik <hello@dincercelik.com>
Change-Id: I8b52b2630a72393bf4b9c7095fc67db6c36c27fa
|