openstack/devstack - devstack - OpenDev: Free Software Needs Free Tools

It may be required to troubleshoot performance related bugs during
tests exection, to have ability to inspect environment processes
and theirs status during test run this patch installs atop
by default if not explicitly disabled.
Related-Bug: #2095224
Change-Id: Iedbd61f3ce3cd2255ea5f2a7a93ba2f39ad28ff2

OpenStackClient has a significant amount of startup overhead, which
adds a non-trivial amount of time to each devstack run because it makes
a lot of OSC calls. This change uses the OSC service from [0] to run
a persistent process that handles openstack calls. This removes most
of the startup overhead and in my local testing removes about three
minutes per devstack run.
Currently this is implemented as an opt-in feature. There are likely a
lot of edge cases in projects that use a devstack plugin so turning it
on universally is going to require boiling the ocean. I think getting
this in and enabled for some of the major projects should give us a lot
of the benefit without the enormous effort of making it 100% compatible
across all of OpenStack.
Depends-On: https://review.opendev.org/c/openstack/nova/+/918689
Depends-On: https://review.opendev.org/c/openstack/ironic/+/918690
Change-Id: I28e6159944746abe2d320369249b87f1c4b9e24e
0: http://lists.openstack.org/pipermail/openstack-dev/2016-April/092546.html 

stop_dstat() calls stop_process() for dstat, memory_tracker and
file_tracker respectively. Inside stop_process(), a check for the
existence of the service is performed by is_service_enabled().
So even if we apply this seemingly dangerous commit,
is_service_enabled() is respected, so it's safe.
Closes-Bug: #1998990
Change-Id: Ica58cdb1d60c4c796f582d82ed2cde0be94b1a7e
Signed-off-by: Nobuhiro MIKI <nmiki@yahoo-corp.jp>

I941ef5ea90970a0901236afe81c551aaf24ac1d8 added a sed command that
should match and delete path values but used '/' as sed separator. This
leads to error in unstack.sh runs when the path also contains '/':
+./unstack.sh:main:188 sudo sed -i '/directory=/opt/stack/ d' /etc/gitconfig
sed: -e expression #1, char 13: unknown command: `o'
So this patch replace '/' separator with '+'.
Change-Id: I06811c0d9ee7ecddf84ef1c6dd6cff5129dbf4b1

This is necessary for more consistent behavior across multiple
distro versions. Apparently somewhere along the way, git started
looking at the current user's home directory instead of $HOME.
Related-Bug: https://bugs.launchpad.net/devstack/+bug/1968798
Change-Id: I941ef5ea90970a0901236afe81c551aaf24ac1d8

We have a *ton* of stuff in devstack that is very linear, specifically
the ten-ish minutes we spend loading osc to run a single API command
against something. We also generate configs, sync databases, and other
things that use one core of our worker and make our runtime longer
than it really needs to be.
The idea in this patch is to make it super simple to run some things
in the background and then wait for them to finish before proceeding
to something that will require them to be done. This avoids the
interleaving you would expect by redirecting the async tasks to a log
file, and then cat'ing that log file synchronously during the wait
operation. The per-task log file remains so it's easier to examine
it in isolation.
Multiple people have reported between 22-30% improvement in the
time it takes to stack with this. More can be done, but what is here
already makes a significant difference.
Change-Id: I270a910b531641b023c13f75dfedca057a1f1031

Previously pyc files were only cleaned if clean.sh was run.
with this change a new clean_pyc_files function was introduced
with the logic that was previously in clean.sh but it is now
invoked from unstack.sh
With the previous behavior you could not stack with horizon
enabled then unstack and stack again due to the presence of pyc
files that were owned by root.
By moving the clean to unstack in stead of clean.sh you can
now stack, unstack and stack again without hitting the pyc issue.
since unstack is invoked by clean the existing clean.sh behavior has
not changed in practice except for the fact the pyc files are
removed sooner in the process.
This change also removes support for findutils < 4.2.3
Ubuntu 12.04 and CentOS 6 both have 4.4.2 since they were
released 8 years ago and are now EOL its fair to assume
that all modern distros have 4.2.3+
https://repology.org/project/findutils/versions
Change-Id: I13c9aad9be7e0930a0d875b7d382090caf0b8982

Currently VMs created by a previous DevStack run still exists even
after re-run stack.sh. This leads to a failure in launching a VM after
the second run of stack.sh. We need to check the reason of the failure
by nova compute log and clean up remaining VMs. It is annoying.
IIRC we cleaned up existing VMs. While I failed to identify which commit
changed this behavior, I believe it is worth recovering it.
This commit changes unstack.sh to call cleanup_nova. cleanup_cinder()
already cleans up LVM volumes and some of them may be used by VMs,
so I believe it is reasonable to clean up VMs in unstack.sh.
Change-Id: I9fcbc5105e443037fada1ef6a76a078145964256

This reverts commit 168ca7f0a4.
Removing postgresql support from devstack was unnecessary
since it's not broken and not causing maintenance issues
as far as I know. The commit being reverted said that pg
support was deprecated in Pike but nothing in the docs or
commit message refer to official deprecation of postgres
support in devstack or openstack in general. Not to mention
that there are still postgres-based jobs that will no
longer work *and* the notification to the mailing list about
doing this happened *after* it was already done [1] leaving
stakeholders with no time to reply.
[1] http://lists.openstack.org/pipermail/openstack-discuss/2019-October/010196.html
Change-Id: Ie7036d37d79e6aba462b7c97f917e2e7aed108f9

This was deprecated for removal in Pike. It's probably time to drop it.
Note that the 'postgresql-devel'/'postgresql-server-dev-all' packages
are retained since some packages still include 'psycopg2' in their
general requirements.
Change-Id: I51e8354e99972757253ce259e6c03c91da24398c
Signed-off-by: Stephen Finucane <sfinucan@redhat.com>

DEFAULT_VOLUME_GROUP_NAME volume group is LVM ephemeral storage used by
Nova. It is created by init_nova() if user sets NOVA_BACKEND to "LVM".
However, init_cinder() is also hardcoded to create it, based on the
asumption that CINDER_ENABLED_BACKENDS includes it. That assumption
doesn't hold for the current code. What's more important, even if user
wants to use DEFAULT_VOLUME_GROUP_NAME as one of cinder backends and
adds it to CINDER_ENABLED_BACKENDS, the current code in init_cinder()
are general enough and should work fine. This change removes relevant
code in init_cinder(). It also moves DEFAULT_VOLUME_GROUP_NAME clean-up
code from unstack.sh to cleanup_nova().
Change-Id: I53762f8eda6256f962cc4e1f1098406879bbcf5c

Change-Id: I36cab41fdb767e8e31073a4fa1f929090f702910
Closes-Bug: #1738938 

This tears out the alternative path of using screen, so that we only
use systemd enabled paths. This simplifies the number of ways that
devstack can be run, and provides a much more reliable process
launcher than the screen based approach.
Change-Id: I8c27182f60b0f5310b3a8bf5feb02beb7ffbb26a

In Ibbb430fb1dbf66942168e0cb52d990ab6a2eb8d7, we are adding
etcd3 as a new base service. We should drop zookeeper
and use etcd3 as the backend.
Since cinder is the first service for this tooz+etcd3 DLM
scenario and cinder uses eventlet we have cannnot use the
grpc based driver in tooz. So new CINDER_COORDINATION_URL
that defaults to the etcd3's grpc HTTP gateway based
tooz backend.
We need to hold this change until the tooz change (see
Depends-On) is available in a tooz release.
Depends-On: I6184ed193482dad9643ccb2b97133d4957485408
Change-Id: Ia187e1a86413edf25b909b6bb57e84fb4930a696

ETCD_DOWNLOAD_URL is set to github url, in our CI, we can point
ETCD_DOWNLOAD_URL to a url in tarballs.openstack.org possibly
in devstack-gate
Download the etcd binaries and drop them into /opt/stack/bin and
use it from there. Cache the tgz for subsequent use (local workflow)
daemon-reload is called twice once from inside the write_user_unit_file
and then when we adjust the entries with additional things recommended
by the etcd team. We need a better way to do this in the future.
Added a TODO to verify the downloaded artifact later. The etcd team
posts gpg signature, we could verify that or run sha256sum and hard
code that in lib/etcd3 file. We would have to update it whenever we
bump the etcd3 version.
We use the public key "CoreOS Application Signing Key <security@coreos.com>"
with ID FC8A365E to verify the integrity of the downloaded file
Any jobs that need to be run on architectures where v3.1.7 is not available
should rey the v3.2.0-rcX release candidates. We can switch to v3.2.0
when it gets released.
Initial version of this code was borrowed from the dragonflow
repo:
http://git.openstack.org/cgit/openstack/dragonflow/tree/devstack
Change-Id: Ibbb430fb1dbf66942168e0cb52d990ab6a2eb8d7

tls-proxy is the way we're now doing a standard install using https
between services. There is a lot more work to make services directly
handle https, and having python daemons do that directly is a bit of
an anti pattern. Nothing currently tests this in project-config from
my recent grepping, so in the interest of long term maintenance,
delete it all.
Change-Id: I910df4ceab6f24f3d9c484e0433c93b06f17d6e1

To avoid using legacy functions accidentially.
Depends-On: Ida1f83b6b3ef9b76be13c063c7e35a8703214078
Change-Id: I3ff136fc8330c92007cdfe91b77d7f9865eabd8d

This removes all of the heat code from the devstack tree, in favor of the
devstack plugin in Heat's tree.
Depends-On: I4bed1e5cef5afa7b049b07640086a86a3f881e13
Depends-On: Ic392bcc24bc374ee8511a94f1d8f6ac23131c7e3
Change-Id: I5b60422bf1f5fa78aa8f3383f7a222e0356d9e42

Previously the usage of neutron debug ports was removed by
5e01c47e4d but there was still call to
teardown_neutron_debug. Recently a change to devstack-gate
1d6cc0771a3399300117f488e9d71e7ea46a4d82 caused that call to be
triggered and breaking the gate-devstack-dsvm-updown job.
This patch deletes the call and comments regarding setup_neutron_debug
and teardown_neutron_debug.
Change-Id: Ifdacb0cec1307db469bd66f551474539184cf2cd

In some initialization conditions (having never ran stack.sh) the
result of unstack.sh is dependent on if the user had previously
installed lvm2 or disabled the cinder service.
This change makes all results the same with a bit of LBYL.
There's also a drive-by to put a comment back where it belongs after
being accidentally moved in the related change.
Related-Change: I09b1a7bee0785e5e1bb7dc96158a654bd3f15c83
Change-Id: I9a7e052677d60cbbbdd582877f3c6c48c387f668
Closes-Bug: #1619195 

Uses lib/placement, but relies on some functionality from
lib/nova. This leads to some weirdness since the nova has
special status in stack.sh. If/when placement is extracted
it may be good to follow the devstack plugin structure
instead.
Because the placement code is currently a part of nova, there
are dependencies in lib/placement on a some $NOVA_* variable
and, if virtenv is being used, the virtualenv used by nova.
Because placement currently runs using nova's configuration
settings, not a lot actually happens in lib/placement: apache
is configured and keystone accounts and endpoints are created.
If PLACEMENT_DB_ENABLED is true then a separate placement db
will be configured.
When complete the initial version of the placement service will
provide support for managing resource providers, inventories and
allocations.
The placement api only runs under mod-wsgi.
Change-Id: I53dd3e6b41de17387a0e179fc9ac64c143b6a9eb

This was used solely by bigswitch, and everyone else has moved over to
devstack plugins. Cleaning this out makes the core logic much simpler.
Depends-On: I8fd2ec6e651f858d0ce109fc335189796c3264b8
(grenade removal)
Change-Id: I47769fc7faae22d263ffd923165abd48f0791a2c

Background for this work can be read on the mailing list:
http://lists.openstack.org/pipermail/openstack-dev/2016-May/094063.html
Usage of the new Neutron is by setting the following in
ENABLED_SERVICES:
* neutron-api
* neutron-l3
* neutron-agent
* neutron-dhcp
* neutron-metadata-agent
For now, the new neutron library supports just the ML2 plugin, with the
Open vSwitch and Linux Bridge agents supported. All other Neutron
plugins should be creating their own DevStack plugin if they wish for
DevStack to support them. Many of them already do.
Other notable changes compared to neutron-legacy:
* Rely on the Neutron defaults, and force Neutron to make
 sane defaults instead of all kinds of knobs in DevStack.
* Default to rootwrap daemon support
* Use the security group driver by default
* interface_driver can now use NEUTRON_AGENT (linuxbridge, openvswitch), since
 they are entrypoints in neutron's setup.cfg
* Use NEUTRON_AGENT variable to determine which agent to run
 Works with NEUTRON_AGENT set to either "linuxbridge" or "openvswitch"
 Default is openvswitch for the time being.
* Set ML2 configuration for VXLAN support
* Remove Xen hypervisor stuff - it should be a plugin
* Move L3 crud into separate service file:
 There's a lot of L3 configuration that was in the main neutron file, but
 a lot of it is self contained and can be moved into its own file.
 The new l3 service file will contain all the previous L3 plumbing and
 configuration that the OpenStack Gate expects, while also eventually
 moving the whole l3 network creation step into a single hook that can be
 overridden by plugins.
* Introduce a check for a function "neutron_plugin_create_initial_networks" which
 will become the mechanism through which different topologies, and
 networking plugins can create and wire the initial networks that are
 created during a stack.sh run.
The new lib/neutron is considered experimental, and followup patches
will build upon this one. Existing users of lib/neutron-legacy should
remain unharmed.
Co-Authored-By: Hirofumi Ichihara <ichihara.hirofumi@lab.ntt.co.jp>
Co-Authored-By: Dean Troyer <dtroyer@gmail.com>
Change-Id: I31b6362c6d9992f425f2dedbbeff2568390a93da

In I33525e2b83a4497a57ec95f62880e0308c88b34f, we switched
from zookeeper to dlm. Somehow this got left behind.
Change-Id: I41d13d33c9a81271d4a9752cbe98c0028a17ab1e

Change-Id: I7382eccf96d14df223777cd43344d554b8452a77

These flags were not functioning as described. Check if UNSTACK_ALL is
set in env or -a flag is set when calling script.
Change-Id: I128d32b1e74ee46e24a9eb2e81560e58137b1553
Closes-Bug: #1546687 

After staring at a bunch of logs, try to crisp up the ps4 output for
maximum readability.
This also adds PS4 to all calling scripts by having a common PS4 in
stackrc. It should make understanding when clean fails a bit more
straight forward.
Change-Id: Ia1f8dff5c1102c32c146a020a9f033c65d2c50de

The existing GetOSVersion has a lot of unused code which is wrong in
several ways
 - the only path tested in upstream CI is with lsb_release, because
 it's pre-installed on all nodes
 - the /etc/redhat-release checking probably still works, but is
 unnecessary
 - If using lsb_release, os_UPDATE has never actually been set.
 - the /etc/SuSE-release branch checking is broken if the lsb package
 is actually installed. lsb checking does not set os_UPDATE but yet
 the SuSE DISTRO setting relies on this to set a patch level (and so
 does some of the rpm tags). SuSE 11 is up to update 3, but the rpm
 matching is stuck hard-coded to update 2. I'm guessing
 installation is actually broken there.
 - the debian checking branch is broken. The VERSION tags have been
 removed and were not supposed to be relied on anyway (see notes in
 [1])
This simplifies things:
 - remove OSX checking (moved here after discussions in
 I31d0fdd30928ecc8d959a95838b1d3affd28ac6f)
 - only use the output of lsb_release.
 - A small best-effort check to pre-install lsb packages if not
 detected (that avoids chicken-egg-problem of package-install
 wrappers relying on os_* flags).
 - The unset os_UPDATE is removed. It's only previous use was for
 setting separate suse versions in the DISTRO element for matching
 during package installs (since removed)
 - DISTRO setting is modified to use the parts of os_RELEASE it wants.
 Per-above, this is the correct place to parse out specifics.
 - Call out the is_* functions, which are a better way to detect
 platforms
 - Export the variables as read-only, since they shouldn't be reset
[1] http://sources.debian.net/src/base-files/7.5/debian/changelog/
Change-Id: I46a2c36d95327087085df07cb797eb91249a893c

This attempts to make the zookeeper installation a bit more modular
(assuming that other folks will want to add other dlms as plugins),
and addresses the service start issues with zookeeper under
ubuntu/upstart.
Zookeeper is not going to be installed by default. Services need to
ask for it with use_dlm.
Change-Id: I33525e2b83a4497a57ec95f62880e0308c88b34f

In Tokyo, there was a cross project session on distributed
key locking:
https://etherpad.openstack.org/p/mitaka-cross-project-dlm
In support of the discussion there, we'll need support for
a zookeeper service in Devstack and ability to use libraries
like Tooz for DLM functionality.
In this review, we pick up some configuration files from
monasca-api and copy the lib/template to implement the
zookeeper lifecycle. Those services that need zookeeper
need to add "zookeeper" in ENABLED_SERVICES.
Change-Id: Icef26e5cdaa930a581e27d330e47706776a7f98f

Componets => Components
pluggins => plugins
Change-Id: I82634a55fd5895599099c94817af7d8d2f602859

The ceilometer project is moving to using a devstack plugin rather
than having ceilometer in the base devstack. This is to allow
greater control and flexibility.
Change-Id: I413ab159474b7d7231ad66d3a482201f74efe8a8

On some system Cinder is not enabled so we can't assume LVM is installed. So
do not try to `sed` /etc/lvm/lvm.conf or clean LVM VG if cinder is not enabled
Change-Id: I09b1a7bee0785e5e1bb7dc96158a654bd3f15c83

Once the trove code is copied into the trove repo and it is used
as a devstack-plugin, we can remove trove-specific code from
devstack.
Change-Id: I8f9f1a015edb7ec1033e2eaf0b29ab15d89384ce
Depends-On: I3506dec0e6097f9c2e9267110fdfb768faa23c85

This avoids us leaving the filter in the global lvm config. Without cleaning
this up, we can hit some failures to run stack.sh because devices are excluded
that we need to be able to see. This resets it to what it was before when
we do a cleanup.
Also, do this before we add the line, so we don't add multiple lines on
successive runs.
Closes-bug: #1437998
Change-Id: Idbf8a06b723f79ef16a7c175ee77a8c25f813244

Fix documentation build errors and RST formatting
Change-Id: Id93153400c5b069dd9d772381558c7085f64c207

during the glusterfs integration it was seen that plugins might need
to set new defaults on projects before the project files load. Create
a new override-defaults phase for that.
Intentionally not adding to the documentation yet until we're sure
this works right in the glusterfs case.
Reported-By: Deepak C Shetty <deepakcs@redhat.com>
Change-Id: I13c961b19bdcc1a99e9a7068fe91bbaac787e948

Preparing to refactor lib/neutron to support Neutron as the default
network config. lib/neutron will be renamed internally and refined
to support a couple of specific configurations.
Change-Id: I0d3773d14c4c636a4b915734784e7241f4d15474

unstack.sh always stop dstat process, however unstack.sh show 'dstat: no process
found' when dstat is disabled.
This patch stop function of stop_dstat, when dstat is disabled.
Change-Id: If9054826bed8a7fedd4f77ef4efef2c0ccd7f16e
Closes-Bug: #1435148 

is_keystone_enabled() was calling is_service_enabled(), which is what called
is_keystone_enabled() in the first place. Make it work as designed and
also change calls to use the full service name. Note that this is all
still comptible with the prior usage of 'is_service_enabled key'.
Change-Id: I9c28377ecf074b7996461d2a4ca12d88dfc4d47e

This adds the test infrastructure for testing that unstack.sh and
clean.sh do the right thing, and actually stop what's expected. This
is designed to be used in upstream testing to make unstack and clean a
bit more certain.
It includes numerous fixes to make these pass in an errexit
environment with the gate config. The scripts still don't run under
errexit because we don't assume we've handled all possible cleanup safely.
Change-Id: I774dfb2cc934367eef2bb7ea5123197f6da7565b

Change-Id: I17cb5e5d93bfe6fc6746ee530f639e2ed42da85a

Now that we have a working external plugin mechanism stackforge
projects definitely don't need to be directly in devstack. These were
largely unused previously anyway.
Change-Id: I300686b2ac976d9b454404842b3f210fd7c239d9

DevStack currently lacks support for LVM ephemeral storage in Nova.
This support is important for testing of Nova's LVM backend. The
proposed change adds a default volume group, to be shared by Cinder
and Nova. It also adds a configuration option NOVA_BACKEND, which
must be LVM if it is set, that determines whether Nova should be
configured to use LVM ephemeral storage.
Change-Id: I4eb9afff3536fbcd563939f2d325efbb845081bb