Reduce 'internal_poll_interval' to stem the CPU usage tide

See https://github.com/ansible/ansible/pull/16560 and
https://github.com/ansible/ansible/issues/14219

Signed-off-by: Dan Mick <dan.mick@redhat.com>

Merge pull request #299 from ceph/wip-nsupdate

Add role for nsupdate_web

Add role for nsupdate_web

See https://github.com/zmc/nsupdate-web/

Signed-off-by: Zack Cerza <zack@redhat.com>

Merge pull request #306 from ceph/wip-lsbrelease

nameserver: Install redhat-lsb-core for `lsb_release`

Reviewed-by: Dan Mick <dmick@redhat.com>

nameserver: Install redhat-lsb-core for `lsb_release`

Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #305 from dmick/master

testnode/tasks/ntp: remove '/etc/network/if-up.d/ntpdate' on Xenial

testnode/tasks/ntp: remove '/etc/network/if-up.d/ntpdate' on Xenial

It's buggy (see comment in code) and we don't need it, because
teuthology does what it would do anyway.  Supposed to be fixed in
a later version of the Debian package which is in Yakkety.

Fixes: http://tracker.ceph.com/issues/18837
Signed-off-by: Dan Mick <dan.mick@redhat.com>

Merge pull request #281 from rjfd/wip-opensuse-upstream-pr

Support for running teuthology suites in OpenSUSE Leap images

Merge pull request #303 from ceph/wip-teuth-mime

teuthology: Correct MIME type spec for log files

teuthology: Correct MIME type spec for log files

Gzipped files were being served as text/plain. Instead of applying text/plain to everythincg, extend the default MIME configuration to serve *.log as text/plain.

Signed-off-by: Zack Cerza <zack@redhat.com>

Merge pull request #302 from ceph/wip-named-chown

nameserver: Fix permissions on /var/named/zones

nameserver: Fix permissions on /var/named/zones

Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #301 from ceph/wip-cloud-dns

Ensure search domain and domain are set properly in resolv.conf

Ensure domain is set in resolv.conf

Signed-off-by: Zack Cerza <zack@redhat.com>

Ensure search domain is set in resolv.conf

Signed-off-by: Zack Cerza <zack@redhat.com>

Merge pull request #300 from ceph/wip-dns-slave

nameserver: Add support for BIND slave nameservers

Reviewed-by: Dan Mick <dmick@redhat.com>

nameserver: Add support for BIND slave nameservers

Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #298 from ceph/wip-centos73

Add CentOS 7.3 to Cobbler

Add CentOS 7.3 to Cobbler

Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #297 from ceph/wip-ddns-in-front

nameserver: Dynamic DNS support

Reviewed-by: Dan Mick <dmick@redhat.com>

nameserver: Enable ntpd

Required for dynamic DNS updates to work

Signed-off-by: David Galloway <dgallowa@redhat.com>

nameserver: Configure SELinux for dynamic DNS support

Signed-off-by: David Galloway <dgallowa@redhat.com>

nameserver: Support dynamic DNS in named config

Signed-off-by: David Galloway <dgallowa@redhat.com>

nameserver: Support dynamic DNS zone in records task

Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #295 from ceph/wip-dhcp-cleanup

Cleanup cobbler network setup stuff for Ubuntu testnodes

Reviewed-by: Dan Mick <dmick@redhat.com>

roles: testnode: Support for openSUSE Leap

Signed-off-by: Ricardo Dias <rdias@suse.com>

roles: common: Support for openSUSE Leap

Signed-off-by: Ricardo Dias <rdias@suse.com>

testnode: do not configure Nagios on zypper-based systems

Nagios is only used in the Sepia lab. For zypper-based systems we are currently
targeting OpenStack only.

Signed-off-by: Nathan Cutler <ncutler@suse.com>

Add sshd_config template for openSUSE Leap

Signed-off-by: Nathan Cutler <ncutler@suse.com>

cleanup: remove fullstops in tasks/ssh.yml

Signed-off-by: Nathan Cutler <ncutler@suse.com>

testnode/templates/ssh: unify permissions to 644

Signed-off-by: Nathan Cutler <ncutler@suse.com>

testnode/tasks/main.yml: support openSUSE

Signed-off-by: Nathan Cutler <ncutler@suse.com>

tasks/main.yml: convert comments into names

Signed-off-by: Nathan Cutler <ncutler@suse.com>

Add zypper_systems.yml (straight copy from yum_systems.yml)

cobbler_profile: Concede to systemd NIC naming methods

Setting biosdevname=0 was causing issues with NICs being renamed after
cobbler reimage.  The system would reimage/install fine but the NIC would have a
different name on firstboot.

Signed-off-by: David Galloway <dgallowa@redhat.com>

cobbler: Remove cephlab_ubuntu_network snippet

Just cruft at this point.  It hasn't been used since
05267fe001432d326c30262d4b07d1775a2ad0d4

Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #294 from ceph/wip-18067

testnode: Install libtool-bin for building xfstests

testnode: Install libtool-bin for building xfstests

Fixes: http://tracker.ceph.com/issues/18067
Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #293 from ceph/wip-zap-disks

Zap all non-root disks on testnodes

cobbler: Install gdisk for use with testnodes zap ansible tag

Signed-off-by: David Galloway <dgallowa@redhat.com>

cobbler: zap disks after cobbler reimage

Added to first line to make sure it's run when using -stock profiles

Signed-off-by: David Galloway <dgallowa@redhat.com>

testnode: Zap GPT and MBR on all disks but root

Fixes: http://tracker.ceph.com/issues/15951
Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #292 from ceph/wip-dns-recurse

nameserver: Add support for enabling/disabling recursion

Reviewed-by: Dan Mick <dmick@redhat.com>

Merge pull request #291 from ceph/wip-17922

Updates to downstream_setup role

Reviewed-by: Dan Mick <dmick@redhat.com>

downstream_setup: Dynamically evaluate when statements in main.yml

See https://github.com/ansible/ansible/issues/18483

Signed-off-by: David Galloway <dgallowa@redhat.com>

nameserver: Add support for enabling/disabling recursion

Signed-off-by: David Galloway <dgallowa@redhat.com>

downstream-setup: define empty var for repos_to_remove

This var got missed when working on
https://github.com/ceph/ceph-cm-ansible/pull/287

Fixes: http://tracker.ceph.com/issues/17922
Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #290 from ceph/wip-firmware

New role for updating firmware

Reviewed-by: Dan Mick <dmick@redhat.com>

firmware: Add README for role

Signed-off-by: David Galloway <dgallowa@redhat.com>

firmware: Add NVMe firmware update support for smithi

Signed-off-by: David Galloway <dgallowa@redhat.com>

firmware: Add BMC update support for smithi

Signed-off-by: David Galloway <dgallowa@redhat.com>

firmware: Add Areca RAID controller support

Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #287 from ceph/wip-ansible22

testnode: Define a few empty variables for ansible v2.2 compatibility

firmware: New role.  Adding BIOS and BMC support for mira

Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #289 from ceph/wip-rhel73

Add RHEL7.3 support

Reviewed-by: Dan Mick <dmick@redhat.com>

Add RHEL7.3 support

Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #288 from ceph/wip-nrpe-rhel

common: Install a few additional packages for NRPE

common: Install a few additional packages for NRPE

check_load nagios plugin needs to be installed on RHEL and CentOS

Signed-off-by: David Galloway <dgallowa@redhat.com>

testnode: Define a few empty variables for ansible v2.2 compatibility

Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #286 from ceph/wip-users-become

Use "become: true" for users role

Use "become: true" for users role

This is needed if the users role is called manually.  Typically it's
called via the common role and "become: true" is inherited.

But in the case of just needing to update update the teuthology_user's pubkeys with @all.pub, just calling the users role manually is quicker.  sudo access is obviously required to modify users and groups.

Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #285 from ceph/wip-beta2

common: Change method of detecting beta distro

common: Change method of detecting beta distro

It was recently discovered that Release Candidate ISOs of RHEL don't output "Beta" or
"Alpha" in `lsb_release` output.  Therefore, RHEL RC Distros were
registering with CDN and not putting beta_repos in place.

This new method registers every RHEL system with CDN, checks if there are
any available matching repos based on ansible_distribution_version, and
either:
  - Unregisters from CDN and runs beta_distros.yml if there are no CDN repos
  - Continues with setting up desired rhsm_repos

I also added some additional working to task descriptions to hopefully
claer up what's going on in this playbook since there are a lot of when
and if statements.

Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #283 from ceph/wip-common-updates

Updates to common role

Reviewed-by: Dan Mick <dmick@redhat.com>

common: Add nagios check_mem command

Signed-off-by: David Galloway <dgallowa@redhat.com>

common: Create nagios_allowed_hosts var for lab-specific nagios IPs

Signed-off-by: David Galloway <dgallowa@redhat.com>

Add common.yml playbook file

Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #282 from ceph/wip-vmlist

Add vmhosts to vmlist.py and run virsh in read-only mode

Reviewed-by: Dan Mick <dmick@redhat.com>

Add vmhosts to vmlist.py and run virsh in read-only mode

'virsh -r' allows vmlist.py to work with RHEV hypervisor hosts

Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #280 from ceph/wip-ubuntu-groups

users: Leave managed_admin_users' other group memberships alone

Reviewed-by: Dan Mick <dmick@redhat.com>

users: Leave managed_admin_users' other group memberships alone

This change is mainly so the users role can be run on vps_hosts.

Previously, in order to update the ubuntu user's authorized_keys on
vps_hosts, the testnodes role had to be run which caused ubuntu to be
removed from the libvirtd group.

The ubuntu user is in managed_admin_users on vps_hosts so this will
ensure the user is added to sudo and left in the libvirtd group.

managed_users, on the other hand, we want to make sure are only in
groups we specify so the "Create all users without sudo access." task is
left as-is (without append: yes).

Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #279 from ceph/wip-beta-repos

defaults: beta_repos must be a dict

defaults: beta_repos must be a dict

This causes ceph-cm-ansible to fail every teuthology job due to
the bad default: 'with_dict expects a dict'

Signed-off-by: Josh Durgin <jdurgin@redhat.com>

Merge pull request #278 from ceph/wip-beta-distros

Add support for testing RHEL beta distros using internal repos

cobbler: redhat-lsb-core package is required for ansible_lsb variables

Signed-off-by: David Galloway <dgallowa@redhat.com>

testnodes: Add support for internal RHEL beta repos

Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #277 from ceph/wip-teuthology

teuthology role updates

Provide a default for teuthology_yaml_extra

Signed-off-by: Zack Cerza <zack@redhat.com>

Merge pull request #272 from ceph/wip-teuth-kill

teuthology: Create tkill group and grant tkill_users right to run `kill`

teuthology: Create test-admins group and grant test_admins right to run `kill`

Fixes: http://tracker.ceph.com/issues/16614
Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #276 from ceph/wip-teuth-fixes

A couple fixes to teuthology role

Merge pull request #274 from ceph/wip-profile-tasks

Use the profile_tasks plugin

Use the profile_tasks plugin

This will cause ansible to display timing information for tasks,
allowing us to see what's slow and potentially improve speed.

I've been using this locally for a while now. It's nice.

Signed-off-by: Zack Cerza <zack@redhat.com>

Merge pull request #275 from ceph/wip-ssh-retries

Retry SSH connections five times

Merge pull request #273 from ceph/wip-strategy-free

Use strategy: free on some playbooks

Retry SSH connections five times

This may help with some of the problems we've been seeing.

Related to http://tracker.ceph.com/issues/16826
Inspired by
https://github.com/ansible/ansible/issues/13401#issuecomment-216768025

Signed-off-by: Zack Cerza <zack@redhat.com>

Use strategy: free on some playbooks

See http://docs.ansible.com/ansible/playbooks_strategies.html#strategies

This could potentially improve performance.

Signed-off-by: Zack Cerza <zack@redhat.com>

teuthology: Ignore any errors when disabling apache2

Signed-off-by: David Galloway <dgallowa@redhat.com>

teuthology: Additional packages needed to run bootstrap now

Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #271 from ceph/wip-no-fw

testnode: Disable firewalld and iptables regardless of OS version

testnode: Disable firewalld and iptables regardless of OS version

iptables was recently found installed and running on a RHEL7 system.
Previous testnode playbook runs wouldn't catch this since it shouldn't
be installed in the first place.  This change ensures firewalld and
iptables are stopped on all RPM-based distros.

Fixes: http://tracker.ceph.com/issues/16809
Signed-off-by: David Galloway <dgallowa@redhat.com>

Merge pull request #270 from ceph/wip-keys-retry

Add a retry when cloning the keys repo

Add a retry when cloning the keys repo

Signed-off-by: Zack Cerza <zack@redhat.com>

Merge pull request #269 from ceph/wip-cobbler-gitrace

Use flock to avoid race conditions with git

Use flock to avoid race conditions with git

Signed-off-by: Zack Cerza <zack@redhat.com>

Merge pull request #265 from ceph/wip-nagios-common

Move NRPE setup to common role

Merge pull request #267 from ceph/wip-no-gperftools-devel

testnode: drop gperftools-devel

Merge pull request #268 from ceph/wip-failure-log-debug

If a YAML error is hit, log the raw object

Reviewed-by: Dan Mick <dmick@redhat.com>

If a YAML error is hit, log the raw object

We are seeing: "RepresenterError: cannot represent an object: pcp"

Signed-off-by: Zack Cerza <zack@redhat.com>

testnode: drop gperftools-devel

Stop installing the gperftools-devel package on testnodes, for the
following reasons:

1. Ceph does not require gperftools-devel at runtime, so installing it
   unnecessarily just slows everything down at this point.

2. When we want to test newer builds of gperftools (say, from RHEL 7.3),
   if the corresponding "gperftools-devel" package is not available on
   the test node, then yum will fail the transaction.

(Additionally, gperftools-devel is no longer in EPEL 7, and it has moved to
RHEL 7 Optional in RHEL 7.2 (https://bugzilla.redhat.com/1213879,
https://access.redhat.com/errata/RHEA-2015:2293). The "epel" list is no
longer the correct list for this package on redhat_7 and centos_7.)