This commit is part of a series to retire the Packaging Deb
project. Step 2 is to remove all content from the project
repos, replacing it with a README notification where to find
ongoing work, and how to recover the repo if needed at some
future point (as in
https://docs.openstack.org/infra/manual/drivers.html#retiring-a-project).
Change-Id: Ic58bebe25bdaaf22e8c26e79e426e8370f91b0ea
Copy&paste of the relevant code parts of openSUSE 42.2 without
the dot/underscore, which we dropped in other places as well.
Change-Id: I08a663fd03c8545de09d650001ab250eaf40e427
Failed uploads happen when things like clouds disappearing happen. This
then results in hundreds or even thousands of upload records all failed
while we wait for the cloud to return. THis is noise that makes it hard
to debug actual problems with image uploads as you can't easily see what
is current.
Avoid this noise in output by cleaning up Failed records after they have
been recorded.
Change-Id: Ife0cf3dd0d9af6cf7a587e9906726c67271b0d5c
(cherry picked from commit 84336a4d3c)
We currently have a race condition between our cleanup worker and our
unit test. My hope is, if we agree to disable the CleanupWorker thread
for the test, we still consider this a valid test.
Change-Id: I04b87ef044de7f99cc9cbd0c08747e53d383693b
Signed-off-by: Paul Belanger <pabelanger@redhat.com>
(cherry picked from commit 78dcd29fa3)
If DIB fails to produce local files, or if we remove a failing diskimage
from the config before the ZooKeeper node has been removed, we never
remove the znode. Work around this by considering any recorded builds
that have a 'builder' that matches the host doing the cleanup as owned
by that host.
Change-Id: Id3b1e2ae0d0f71209cd66091a7d69143b0b4b20c
Now devstack is systemd-ified, make sure this has full path to avoid
it complaining.
[/etc/systemd/system/devstack@statsd.service:6] Executable path is not absolute, ignoring: socat -u udp-recv:8125 -
Change-Id: I47ebf5abb86bc202e5bf90b0cc318764cbb0d062
Validate that glean will properly use the key-name from nodepool, and
allow a root user to ssh into the node.
Change-Id: I8e50caed4541ae31e61fb0b84c1b285f7baa4c6b
Signed-off-by: Paul Belanger <pabelanger@redhat.com>
Now that we have /etc/ci/mirror_info.sh on our images, we can source
it to learn about our mirrors.
Change-Id: I5c90d2aa51565c9f9cc10f33375b8a46967e1989
Signed-off-by: Paul Belanger <pabelanger@redhat.com>
This old config option was not plumbed though anymore because we stopped
supported snapshot builds where this was used. Nodepool doesn't want to
manage keypairs at all so no creation/addition/listing/deletion, instead
it now accepts per label key name for precreated key in cloud regions
and will boot with that. Unfortunately this old code was "cleaning" up
this new label specified key for us as the only thing it did anymore.
Fix this by removing the dead code and stop deleting keys it doesn't
know about as we aren't managing keys anymore.
Change-Id: Ifff93de1ad81315141bfe1a53dc3d8578faee8e6
In order to support putting less things into images via puppet in Infra,
we'd like to be able to pre-populate our clouds with keypairs for the
infra-root accounts and have nova add those at boot time.
Change-Id: I72cfffacdeb56e3e9743e47a107fb309a7517935
Ubuntu Precise is EOL now, and we have removed the DIBs from
openstack-infra. Remove test coverage for it.
Change-Id: I70d647da84f5f08005922ca7e9d59ca550330006
Signed-off-by: Paul Belanger <pabelanger@redhat.com>
The documentation build does not generate any module index, thus remove
the link to the page. The page
http://docs.openstack.org/infra/nodepool/py-modindex.html
does not exist.
Change-Id: I0da636955e25dadf60bc9f72f5774d2998c52612
As we look to enable opensuse-42.2 images for nodepool.o.o, add
support for jobs to build them. EG: glean and diskimage-builder.
Change-Id: I34e98f0f7693859ed05011b008334628adff612f
Signed-off-by: Paul Belanger <pabelanger@redhat.com>
Add more test coverage for debian, due to recent issues with
system-config debian-jessie jobs failing. It also give projects like
glean and diskimage-builder more coverage.
Change-Id: If5e7fa98ee379f7339148de3ea6574bbcda2b032
Signed-off-by: Paul Belanger <pabelanger@redhat.com>
So that we can trace the API calls being made, put in debug logging for
novaclient and keystoneauth.
Change-Id: Ib2c95a416131f5b7b3d7779e8b9cbf0f4c823e49
For example, it doesn't make sense to add this to centos / fedora
DIBs.
Change-Id: If7574c84e135f447d9601140dae138242162c2d2
Signed-off-by: Paul Belanger <pabelanger@redhat.com>
We don't actually mirror precise, so fall back to default settings for
APT mirrors.
Change-Id: Ifed9e8e6e22aba5a65b27d22706d4a665a7ffe0a
Signed-off-by: Paul Belanger <pabelanger@redhat.com>
Newer versions of paramiko require a client object to be explicitly
closed. Fortunately, we wrap all of our use of paramiko client
objects in our own class. Add a destructor to our class which
closes the client object.
Note, this has been tested to work (and is needed) even if a
connection is not established.
Change-Id: I5dff7ed254567968b42d053b85004769f8647ecb
In our case, we won't have zuulv3-dev setting metadata but
nodepool.o.o will. This now means, nodepool.o.o should not delete
'leaked' nodes because zuulv3-dev.o.o has launched them.
Change-Id: I9a2dbd3845928f83a9cca67082fd6b1ca247a607
Signed-off-by: Paul Belanger <pabelanger@redhat.com>
Currently, while testing zuulv3, we are wanting to share the
infracloud-chocolate provider between 2 nodepool servers. The current
issue is, if we launch nodes from zuulv3-dev.o.o, nodepool.o.o will
detect the nodes as leaked and delete them.
A way to solve this, is to create a per provider 'nodepool-id' where
an admin can configure 2 separate nodepool servers to share the same
tenant. The big reason for doing this, is so we don't have to stand
up a duplicate nodepool-builder and upload duplicate images.
Change-Id: I03a95ce7b8bf06199de7f46fd3d0f82407bec8f5
Signed-off-by: Paul Belanger <pabelanger@redhat.com>
It's possible for an active image upload to complete during the
builder cleanup phase, which would release its lock and change
the state to READY. If the cleanup thread has cached the UPLOADING
state and checks to see if the upload is in progress (just after
the lock has been release), then we could accidentally delete the
upload record because we don't have the correct state. This adds
a second check on the state to make sure it didn't change on us.
Change-Id: I60be43e999cf86d4c3c46e6ea69ecd1bcb69f533
When I added the JSON targets in
I3410a4e4efd649732747b0502d66d212f50fa1bb I also changed the text
output mime-type to be application/text, which seemed to pair with
application/json. But this means browsers prompt to download the file
rather than display it, which is not the intention. Restore to
text/plain.
Change-Id: I6fa6597b2fcd05a4c33ba4932c966101daf98ecf
At first glance, this may be causing a hang on the remote side that is
causing the nodepool test to hang. We didn't notice it because the test
was otherwise broken.
This reverts commit 44ee08c2d7.
Change-Id: Iaef2a140e33fc48f8bfa8ff4769eded37ce152c6
test-config is a devstack plugin phase that runs at the very end. That's
much more what we want, and allows us to continue running the services
using the devstack functions.
Change-Id: I1bdfb23d0ba2c988e367cd13171a59398a3b9c71
We mistakenly skipped this setting, as a results fedora-25 images are
built by default, causing un needed churn for our nodepool project
jobs.
Change-Id: Id91991a490709f9bbac5a4f6e9847e047b83ca51
Signed-off-by: Paul Belanger <pabelanger@redhat.com>