The Open Infrastructure Foundation's developers who maintain the
OpenStackID software are taking over management of the site itself,
and have deployed it on new servers. DNS records have already been
updated to the new IP address, so it's time to clean up our end in
preparation for deleting the old servers we've been running.
OpenStackID is still used by some services we run, like RefStack and
Zanata, and we're still hosting the OpenStackID Git repository and
documentation, so this does not get rid of all references to it.
This reduces the total number of git repos we need to manage as we
migrated away from puppet. Keeping this pruned is a good way of tracking
progress and should make the jobs more reliable until we can delete
In this change remove modules that have moved to base server roles
like ntp and haveged. Also removed are ircbot management,
selinux (no more centos here), haproxy (this moved into ansible with
gitea), and lodgit modules.
This reduces the scope of our puppet related testing to things that
continue to use puppet. This is probably not strictly necessary but
helps keep us up to date with our TODO list.
Previous review pointed out some additional modules we probably
aren't using any longer.
Remove the openafs::client section from openstack_project::server
because we're doing this with ansible now.
Extract eavedrop into its own service playbook and
puppet manifest. While doing that, stop using jenkinsuser
on eavesdrop in favor of zuul-user.
Add the ability to override the keys for the zuul user.
Remove openstack_project::server, it doesn't do anything.
Containerize and anisblize accessbot. The structure of
how we're doing it in puppet makes it hard to actually
run the puppet in the gate. Run the script in its own
playbook so that we can avoid running it in the gate.
* add zk* to emergency
* copy data files on each node to a safe place for DR backup
* make a json data backup: zk-shell localhost:2181 --run-once 'mirror / json://!tmp!zookeeper-backup.json/'
* manually run a modified playbook to set up the docker infra without starting containers
* rolling restart; for each node:
* stop zk
* split data and log files and move them to new locations
* remove zk packages
* start zk containers
* remove from emergency; land this change.
It doesn't seem like this is used anymore. Let's remove it before
we update the rest of this, so that we don't have to, you know,
update abandoned things.
There's a bunch in here. This is mostly big-ticket things and test
fixes. Also, change the README to rst - because why is it markdown?
Grafana has moved to a new package repo  and the apt-get update
step is failing on the current host.
The first version of puppet-grafana that has this update is 6.0.0;
this is two years later than the current version we're using so
... yeah. It does not work with puppet3, so only run apply test with
It looks like upstream has moved from camptocamp/archive to
voxpupuli/archive so the comment is no longer required.
The puppet-python module had some broken facter custom facts that they
removed in 1.9.5. After upgrading to puppet-4 we notice this now. Update
to 1.9.5 from 1.9.4 to fix the error.
Error from puppet log:
Apr 11 00:15:30 afs01 puppet-user: (Facter) error while resolving custom fact "virtualenv_version": undefined method `' for nil:NilClass
Puppet4 is a lot pickier about puppet module dependency resolution. On
some hosts with puppet4 we are getting:
ModuleLoader: module 'project_config' has unresolved dependencies - it will only see those that are resolved. Use 'puppet module list --tree' to see information about modules
Running the suggested command we see that vcsrepo is not resolved
because the version of vcsrepo we install doesn't ahve a metadata.json.
Thankfully the HEAD of vcsrepo does have metadata.json which should
allow this to be resolved.
Puppet 4 is a lot more particular about module deps and has notices that
our mysql module version depends on puppet/staging >=1.0.1 but we only
have 1.0.0 installed. Fix this dep issue by bumping to 1.0.1.
Remove the puppetry for managing nameservers as we now use ansible
configured name servers without puppet.
We will need to follow this up with deletion of the existing
ns*.openstack.org and adns1.openstack.org servers.
Bandersnatch mirroring has been disabled since
I88a838cb28fee3bd16b2b0a26e614ac5c2f23241 which is currently almost 6
months ago. Since then we have been running a reverse caching proxy.
Although bandersnatch served us well, it seems pypi has become
impractical to mirror locally. This is partially due to 2TB volume
limitations of OpenAFS and partially due to us not having a sane way
to filter large, frequently updating packages. With the reverse proxy
working there are no plans to restore our local mirror.
Retire the references to it before we clean up the AFS volumes.
stackalytics.openstack.org does not resolve and seems very dead. Remove
its node from site.pp and remove it from the docs to avoid confusion
about what servers we're really managing. We can always add it back when
the time comes to try again.
The odsreg.openstack.org server was removed from service 2018-01-11,
and should also be cleared from our global site manifest. It was the
only thing utilizing the puppet-odsreg module, so we can take that
out of the modules list as well (that will be retired separately).
This module has been deprecated for the voxpupuli version. 2.6.1 is
the last release with Puppet 3 support. It seems to mostly be a fork,
so I think it is compatible. The repo is hard-coded which is why I
think it's worth updating.
Infracloud is sadly deceased. The upside is we can delete a lot of code
we don't need anymore. This patch removes infracloud nodes from
site.pp so that the puppet-apply test no longer bothers to validate
them, removes the infracloud modules from modules.env so that we don't
bother to install those modules in puppet-apply and puppet functional
tests, and removes the infracloud-specific data from the public hiera.
Additionally stop the puppetmaster from trying to run the infracloud
ansible playbook and finally remove the chocolate region from nodepool's
clouds.yaml (vanilla was already done).
This patch leaves the run_infracloud.sh script and the
infracloud-specific ansible playbooks as well as the infracloud
manifests in the openstack_project puppet module. It's possible those
tools could come in handy in the future if we ever have another
infracloud, and leaving those tools in place doesn't add confusion about
which hosts are actually active nor does it leave cruft that gets
This is currently the last 1.x release that still support puppet 3, we
could make the jobs to puppet 4, but for now we just need a new enough
module for xenial support.
Also, seems puppetlabs-apache is only uesd by cacti.o.o.
Signed-off-by: Paul Belanger <email@example.com>
This is the lastest version of puppet-nodejs that both supports Xenial
and puppet 3 ... after this it starts using puppet 4 syntax. Update
because we need it for Xenial. (note they also changed to using a "v"
Currently puppetdb and puppetboard have been broken for some time (+1
year) and with ubuntu precise becoming EOL it is prime for deleting.
This leaves openstack-infra with a gap in reporting for non-root
users. As such, as proposal is in the works to maybe use ARA.
Signed-off-by: Paul Belanger <firstname.lastname@example.org>
This installs the ptgbot Puppet module so it will run from the
eavesdrop.openstack.org server and generate its Web content there.
Include some rudimentary operational documentation.
To get metrics on MQTT usage into graphite/grafana this commit adds
running mqtt_statsd on firehose
This is a simple first deployment of an ethercalc service. It does not
come with authenticated redis or redis backups. It will however have
We were using version 0.4.0 which supported Wildfly 9 just fine (with
the systemd sysv support hack), but now we need wildfly 10 on the
translate-dev server. Unfortunately 0.4.0 does not support wildfly 10
properly because wildfly 10 removed its built in init scripts. This
newer puppet module will install systemd unit files that should work
Note that even though the new version of the puppet module supports
wildfly 8, 9, and 10 it does not appear to do so on all ubuntu releases.
Wildfly 10 requires a current enough release to have proper systemd.
Older wildfly versions shoudl work on older ubuntu releases via the
built in init scripts.