diff options
Diffstat (limited to 'roles')
74 files changed, 1033 insertions, 386 deletions
diff --git a/roles/chrony/README.md b/roles/chrony/README.md new file mode 100644 index 000000000..bf15d9669 --- /dev/null +++ b/roles/chrony/README.md @@ -0,0 +1,31 @@ +Role Name +========= + +A role to configure chrony as the ntp client + +Requirements +------------ + + +Role Variables +-------------- + +chrony_ntp_servers: a list of ntp servers to use the chrony.conf file + +Dependencies +------------ + +roles/lib_timedatectl + +Example Playbook +---------------- + +License +------- + +Apache 2.0 + +Author Information +------------------ + +Openshift Operations diff --git a/roles/chrony/defaults/main.yml b/roles/chrony/defaults/main.yml new file mode 100644 index 000000000..95576e666 --- /dev/null +++ b/roles/chrony/defaults/main.yml @@ -0,0 +1,2 @@ +--- +# defaults file for chrony diff --git a/roles/chrony/handlers/main.yml b/roles/chrony/handlers/main.yml new file mode 100644 index 000000000..1973c79e2 --- /dev/null +++ b/roles/chrony/handlers/main.yml @@ -0,0 +1,5 @@ +--- +- name: Restart chronyd + service: + name: chronyd + state: restarted diff --git a/roles/chrony/meta/main.yml b/roles/chrony/meta/main.yml new file mode 100644 index 000000000..85595d7c3 --- /dev/null +++ b/roles/chrony/meta/main.yml @@ -0,0 +1,18 @@ +--- +galaxy_info: + author: Openshift Operations + description: Configure chrony as an ntp server + company: Red Hat + license: Apache 2.0 + min_ansible_version: 1.9.2 + platforms: + - name: EL + versions: + - 7 + - name: Fedora + versions: + - all + categories: + - system +dependencies: +- roles/lib_timedatectl diff --git a/roles/chrony/tasks/main.yml b/roles/chrony/tasks/main.yml new file mode 100644 index 000000000..fae6d8e4c --- /dev/null +++ b/roles/chrony/tasks/main.yml @@ -0,0 +1,30 @@ +--- +- name: remove ntp package + yum: + name: ntp + state: absent + +- name: ensure chrony package is installed + yum: + name: chrony + state: installed + +- name: Install /etc/chrony.conf + template: + src: chrony.conf.j2 + dest: /etc/chrony.conf + owner: root + group: root + mode: 0644 + notify: + - Restart chronyd + +- name: enabled timedatectl set-ntp yes + timedatectl: + ntp: True + +- name: + service: + name: chronyd + state: started + enabled: yes diff --git a/roles/chrony/templates/chrony.conf.j2 b/roles/chrony/templates/chrony.conf.j2 new file mode 100644 index 000000000..de43b6364 --- /dev/null +++ b/roles/chrony/templates/chrony.conf.j2 @@ -0,0 +1,45 @@ +# Use public servers from the pool.ntp.org project. +# Please consider joining the pool (http://www.pool.ntp.org/join.html). +{% for server in chrony_ntp_servers %} +server {{ server }} iburst +{% endfor %} + +# Ignore stratum in source selection. +stratumweight 0 + +# Record the rate at which the system clock gains/losses time. +driftfile /var/lib/chrony/drift + +# Enable kernel RTC synchronization. +rtcsync + +# In first three updates step the system clock instead of slew +# if the adjustment is larger than 10 seconds. +makestep 10 3 + +# Allow NTP client access from local network. +#allow 192.168/16 + +# Listen for commands only on localhost. +bindcmdaddress 127.0.0.1 +bindcmdaddress ::1 + +# Serve time even if not synchronized to any NTP server. +#local stratum 10 + +keyfile /etc/chrony.keys + +# Specify the key used as password for chronyc. +commandkey 1 + +# Generate command key if missing. +generatecommandkey + +# Disable logging of client accesses. +noclientlog + +# Send a message to syslog if a clock adjustment is larger than 0.5 seconds. +logchange 0.5 + +logdir /var/log/chrony +#log measurements statistics tracking diff --git a/roles/chrony/vars/main.yml b/roles/chrony/vars/main.yml new file mode 100644 index 000000000..061a21547 --- /dev/null +++ b/roles/chrony/vars/main.yml @@ -0,0 +1,2 @@ +--- +# vars file for chrony diff --git a/roles/cockpit/tasks/main.yml b/roles/cockpit/tasks/main.yml index 6e9f3a8bd..e83f72a3d 100644 --- a/roles/cockpit/tasks/main.yml +++ b/roles/cockpit/tasks/main.yml @@ -6,7 +6,7 @@ - cockpit-shell - cockpit-bridge - "{{ cockpit_plugins }}" - when: not openshift.common.is_containerized | bool + when: not openshift.common.is_atomic | bool - name: Enable cockpit-ws service: diff --git a/roles/etcd/tasks/main.yml b/roles/etcd/tasks/main.yml index e83cfc33c..1e97b047b 100644 --- a/roles/etcd/tasks/main.yml +++ b/roles/etcd/tasks/main.yml @@ -11,24 +11,8 @@ action: "{{ ansible_pkg_mgr }} name=etcd-2.* state=present" when: not openshift.common.is_containerized | bool -- name: Get docker images - command: docker images - changed_when: false - when: openshift.common.is_containerized | bool - register: docker_images - - name: Pull etcd container command: docker pull {{ openshift.etcd.etcd_image }} - when: openshift.common.is_containerized | bool and openshift.etcd.etcd_image not in docker_images.stdout - -- name: Wait for etcd image - command: > - docker images - register: docker_images - until: openshift.etcd.etcd_image in docker_images.stdout - retries: 30 - delay: 10 - changed_when: false when: openshift.common.is_containerized | bool - name: Install etcd container service file diff --git a/roles/etcd_ca/tasks/main.yml b/roles/etcd_ca/tasks/main.yml index d32f5e48c..cf7bc00a3 100644 --- a/roles/etcd_ca/tasks/main.yml +++ b/roles/etcd_ca/tasks/main.yml @@ -1,4 +1,8 @@ --- +- name: Install openssl + action: "{{ ansible_pkg_mgr }} name=openssl state=present" + when: not openshift.common.is_atomic | bool + - file: path: "{{ item }}" state: directory diff --git a/roles/fluentd_master/tasks/main.yml b/roles/fluentd_master/tasks/main.yml index 1c87d562a..32f972f0a 100644 --- a/roles/fluentd_master/tasks/main.yml +++ b/roles/fluentd_master/tasks/main.yml @@ -1,12 +1,12 @@ --- - fail: msg: "fluentd master is not yet supported on atomic hosts" - when: openshift.common.is_containerized | bool + when: openshift.common.is_atomic | bool # TODO: Update fluentd install and configuration when packaging is complete - name: download and install td-agent action: "{{ ansible_pkg_mgr }} name='http://packages.treasuredata.com/2/redhat/7/x86_64/td-agent-2.2.0-0.x86_64.rpm' state=present" - when: not openshift.common.is_containerized | bool + when: not openshift.common.is_atomic | bool - name: Verify fluentd plugin installed command: '/opt/td-agent/embedded/bin/gem query -i fluent-plugin-kubernetes' diff --git a/roles/fluentd_node/tasks/main.yml b/roles/fluentd_node/tasks/main.yml index 8d34c0b19..9fd908687 100644 --- a/roles/fluentd_node/tasks/main.yml +++ b/roles/fluentd_node/tasks/main.yml @@ -1,12 +1,12 @@ --- - fail: msg: "fluentd node is not yet supported on atomic hosts" - when: openshift.common.is_containerized | bool + when: openshift.common.is_atomic | bool # TODO: Update fluentd install and configuration when packaging is complete - name: download and install td-agent action: "{{ ansible_pkg_mgr }} name='http://packages.treasuredata.com/2/redhat/7/x86_64/td-agent-2.2.0-0.x86_64.rpm' state=present" - when: not openshift.common.is_containerized | bool + when: not openshift.common.is_atomic | bool - name: Verify fluentd plugin installed command: '/opt/td-agent/embedded/bin/gem query -i fluent-plugin-kubernetes' @@ -55,4 +55,3 @@ name: 'td-agent' state: started enabled: yes - diff --git a/roles/haproxy/defaults/main.yml b/roles/haproxy/defaults/main.yml index 7ba5bd485..937d94209 100644 --- a/roles/haproxy/defaults/main.yml +++ b/roles/haproxy/defaults/main.yml @@ -1,4 +1,6 @@ --- +haproxy_frontend_port: 80 + haproxy_frontends: - name: main binds: @@ -18,4 +20,4 @@ os_firewall_allow: - service: haproxy stats port: "9000/tcp" - service: haproxy balance - port: "8443/tcp" + port: "{{ haproxy_frontend_port }}/tcp" diff --git a/roles/haproxy/handlers/main.yml b/roles/haproxy/handlers/main.yml index ee60adcab..5b8691b26 100644 --- a/roles/haproxy/handlers/main.yml +++ b/roles/haproxy/handlers/main.yml @@ -3,3 +3,4 @@ service: name: haproxy state: restarted + when: not (haproxy_start_result_changed | default(false) | bool) diff --git a/roles/haproxy/tasks/main.yml b/roles/haproxy/tasks/main.yml index 97f870829..0b8370ce2 100644 --- a/roles/haproxy/tasks/main.yml +++ b/roles/haproxy/tasks/main.yml @@ -19,6 +19,5 @@ enabled: yes register: start_result -- name: Pause 30 seconds if haproxy was just started - pause: seconds=30 - when: start_result | changed +- set_fact: + haproxy_start_result_changed: "{{ start_result | changed }}" diff --git a/roles/lib_timedatectl/library/timedatectl.py b/roles/lib_timedatectl/library/timedatectl.py new file mode 100644 index 000000000..b6eab5918 --- /dev/null +++ b/roles/lib_timedatectl/library/timedatectl.py @@ -0,0 +1,74 @@ +#!/usr/bin/env python +''' + timedatectl ansible module + + This module supports setting ntp enabled +''' +import subprocess + + + + +def do_timedatectl(options=None): + ''' subprocess timedatectl ''' + + cmd = ['/usr/bin/timedatectl'] + if options: + cmd += options.split() + + proc = subprocess.Popen(cmd, stdin=None, stdout=subprocess.PIPE) + proc.wait() + return proc.stdout.read() + +def main(): + ''' Ansible module for timedatectl + ''' + + module = AnsibleModule( + argument_spec=dict( + #state=dict(default='enabled', type='str'), + ntp=dict(default=True, type='bool'), + ), + #supports_check_mode=True + ) + + # do something + ntp_enabled = False + + results = do_timedatectl() + + for line in results.split('\n'): + if 'NTP enabled' in line: + if 'yes' in line: + ntp_enabled = True + + ######## + # Enable NTP + ######## + if module.params['ntp']: + if ntp_enabled: + module.exit_json(changed=False, results="enabled", state="enabled") + + # Enable it + # Commands to enable ntp + else: + results = do_timedatectl('set-ntp yes') + module.exit_json(changed=True, results="enabled", state="enabled", cmdout=results) + + ######### + # Disable NTP + ######### + else: + if not ntp_enabled: + module.exit_json(changed=False, results="disabled", state="disabled") + + results = do_timedatectl('set-ntp no') + module.exit_json(changed=True, results="disabled", state="disabled") + + module.exit_json(failed=True, changed=False, results="Something went wrong", state="unknown") + +# Pylint is getting in the way of basic Ansible +# pylint: disable=redefined-builtin,wildcard-import,unused-wildcard-import +from ansible.module_utils.basic import * + +main() diff --git a/roles/lib_zabbix/library/zbx_action.py b/roles/lib_zabbix/library/zbx_action.py index c08bef4f7..2f9524556 100644 --- a/roles/lib_zabbix/library/zbx_action.py +++ b/roles/lib_zabbix/library/zbx_action.py @@ -81,6 +81,61 @@ def filter_differences(zabbix_filters, user_filters): return rval +def opconditions_diff(zab_val, user_val): + ''' Report whether there are differences between opconditions on + zabbix and opconditions supplied by user ''' + + if len(zab_val) != len(user_val): + return True + + for z_cond, u_cond in zip(zab_val, user_val): + if not all([str(u_cond[op_key]) == z_cond[op_key] for op_key in \ + ['conditiontype', 'operator', 'value']]): + return True + + return False + +def opmessage_diff(zab_val, user_val): + ''' Report whether there are differences between opmessage on + zabbix and opmessage supplied by user ''' + + for op_msg_key, op_msg_val in user_val.items(): + if zab_val[op_msg_key] != str(op_msg_val): + return True + + return False + +def opmessage_grp_diff(zab_val, user_val): + ''' Report whether there are differences between opmessage_grp + on zabbix and opmessage_grp supplied by user ''' + + zab_grp_ids = set([ugrp['usrgrpid'] for ugrp in zab_val]) + usr_grp_ids = set([ugrp['usrgrpid'] for ugrp in user_val]) + if usr_grp_ids != zab_grp_ids: + return True + + return False + +def opmessage_usr_diff(zab_val, user_val): + ''' Report whether there are differences between opmessage_usr + on zabbix and opmessage_usr supplied by user ''' + + zab_usr_ids = set([usr['usrid'] for usr in zab_val]) + usr_ids = set([usr['usrid'] for usr in user_val]) + if usr_ids != zab_usr_ids: + return True + + return False + +def opcommand_diff(zab_op_cmd, usr_op_cmd): + ''' Check whether user-provided opcommand matches what's already + stored in Zabbix ''' + + for usr_op_cmd_key, usr_op_cmd_val in usr_op_cmd.items(): + if zab_op_cmd[usr_op_cmd_key] != str(usr_op_cmd_val): + return True + return False + def host_in_zabbix(zab_hosts, usr_host): ''' Check whether a particular user host is already in the Zabbix list of hosts ''' @@ -106,23 +161,11 @@ def hostlist_in_zabbix(zab_hosts, usr_hosts): return True -def opcommand_diff(zab_op_cmd, usr_op_cmd): - ''' Check whether user-provided opcommand matches what's already - stored in Zabbix ''' - - for usr_op_cmd_key, usr_op_cmd_val in usr_op_cmd.items(): - if zab_op_cmd[usr_op_cmd_key] != str(usr_op_cmd_val): - return True - return False - -# This logic is quite complex. We are comparing two lists of dictionaries. -# The outer for-loops allow us to descend down into both lists at the same time -# and then walk over the key,val pairs of the incoming user dict's changes -# or updates. The if-statements are looking at different sub-object types and -# comparing them. The other suggestion on how to write this is to write a recursive -# compare function but for the time constraints and for complexity I decided to go -# this route. -# pylint: disable=too-many-branches +# We are comparing two lists of dictionaries (the one stored on zabbix and the +# one the user is providing). For each type of operation, determine whether there +# is a difference between what is stored on zabbix and what the user is providing. +# If there is a difference, we take the user-provided data for what needs to +# be stored/updated into zabbix. def operation_differences(zabbix_ops, user_ops): '''Determine the differences from user and zabbix for operations''' @@ -132,49 +175,41 @@ def operation_differences(zabbix_ops, user_ops): rval = {} for zab, user in zip(zabbix_ops, user_ops): - for key, val in user.items(): - if key == 'opconditions': - if len(zab[key]) != len(val): - rval[key] = val - break - for z_cond, u_cond in zip(zab[key], user[key]): - if not all([str(u_cond[op_key]) == z_cond[op_key] for op_key in \ - ['conditiontype', 'operator', 'value']]): - rval[key] = val - break - elif key == 'opmessage': - # Verify each passed param matches - for op_msg_key, op_msg_val in val.items(): - if zab[key][op_msg_key] != str(op_msg_val): - rval[key] = val - break - - elif key == 'opmessage_grp': - zab_grp_ids = set([ugrp['usrgrpid'] for ugrp in zab[key]]) - usr_grp_ids = set([ugrp['usrgrpid'] for ugrp in val]) - if usr_grp_ids != zab_grp_ids: - rval[key] = val - - elif key == 'opmessage_usr': - zab_usr_ids = set([usr['userid'] for usr in zab[key]]) - usr_ids = set([usr['userid'] for usr in val]) - if usr_ids != zab_usr_ids: - rval[key] = val - - elif key == 'opcommand': - if opcommand_diff(zab[key], val): - rval[key] = val - break + for oper in user.keys(): + if oper == 'opconditions' and opconditions_diff(zab[oper], \ + user[oper]): + rval[oper] = user[oper] + + elif oper == 'opmessage' and opmessage_diff(zab[oper], \ + user[oper]): + rval[oper] = user[oper] + + elif oper == 'opmessage_grp' and opmessage_grp_diff(zab[oper], \ + user[oper]): + rval[oper] = user[oper] + + elif oper == 'opmessage_usr' and opmessage_usr_diff(zab[oper], \ + user[oper]): + rval[oper] = user[oper] + + elif oper == 'opcommand' and opcommand_diff(zab[oper], \ + user[oper]): + rval[oper] = user[oper] # opcommand_grp can be treated just like opcommand_hst # as opcommand_grp[] is just a list of groups - elif key == 'opcommand_hst' or key == 'opcommand_grp': - if not hostlist_in_zabbix(zab[key], val): - rval[key] = val - break + elif oper == 'opcommand_hst' or oper == 'opcommand_grp': + if not hostlist_in_zabbix(zab[oper], user[oper]): + rval[oper] = user[oper] + + # if it's any other type of operation than the ones tested above + # just do a direct compare + elif oper not in ['opconditions', 'opmessage', 'opmessage_grp', + 'opmessage_usr', 'opcommand', 'opcommand_hst', + 'opcommand_grp'] \ + and str(zab[oper]) != str(user[oper]): + rval[oper] = user[oper] - elif zab[key] != str(val): - rval[key] = val return rval def get_users(zapi, users): diff --git a/roles/lib_zabbix/library/zbx_host.py b/roles/lib_zabbix/library/zbx_host.py index e26c9caf3..560749f07 100644 --- a/roles/lib_zabbix/library/zbx_host.py +++ b/roles/lib_zabbix/library/zbx_host.py @@ -63,6 +63,19 @@ def get_template_ids(zapi, template_names): template_ids.append({'templateid': content['result'][0]['templateid']}) return template_ids +def interfaces_equal(zbx_interfaces, user_interfaces): + ''' + compare interfaces from zabbix and interfaces from user + ''' + + for u_int in user_interfaces: + for z_int in zbx_interfaces: + for u_key, u_val in u_int.items(): + if str(z_int[u_key]) != str(u_val): + return False + + return True + def main(): ''' Ansible module for zabbix host @@ -120,8 +133,9 @@ def main(): 'dns': '', # dns for host 'port': '10050', # port for interface? 10050 }] + hostgroup_names = list(set(module.params['hostgroup_names'])) params = {'host': hname, - 'groups': get_group_ids(zapi, module.params['hostgroup_names']), + 'groups': get_group_ids(zapi, hostgroup_names), 'templates': get_template_ids(zapi, module.params['template_names']), 'interfaces': ifs, } @@ -140,6 +154,11 @@ def main(): if zab_results['parentTemplates'] != value: differences[key] = value + + elif key == "interfaces": + if not interfaces_equal(zab_results[key], value): + differences[key] = value + elif zab_results[key] != value and zab_results[key] != str(value): differences[key] = value diff --git a/roles/lib_zabbix/tasks/create_template.yml b/roles/lib_zabbix/tasks/create_template.yml index 47749389e..61344357a 100644 --- a/roles/lib_zabbix/tasks/create_template.yml +++ b/roles/lib_zabbix/tasks/create_template.yml @@ -57,6 +57,7 @@ expression: "{{ item.expression }}" priority: "{{ item.priority }}" url: "{{ item.url | default(None, True) }}" + status: "{{ item.status | default('', True) }}" with_items: template.ztriggers when: template.ztriggers is defined diff --git a/roles/nickhammond.logrotate/tasks/main.yml b/roles/nickhammond.logrotate/tasks/main.yml index 0a0cf1fae..e2c51a903 100644 --- a/roles/nickhammond.logrotate/tasks/main.yml +++ b/roles/nickhammond.logrotate/tasks/main.yml @@ -1,6 +1,7 @@ --- - name: nickhammond.logrotate | Install logrotate action: "{{ ansible_pkg_mgr }} name=logrotate state=present" + when: not openshift.common.is_atomic | bool - name: nickhammond.logrotate | Setup logrotate.d scripts template: diff --git a/roles/nuage_master/README.md b/roles/nuage_master/README.md new file mode 100644 index 000000000..de101dd19 --- /dev/null +++ b/roles/nuage_master/README.md @@ -0,0 +1,8 @@ +Nuage Master +============ +Setup Nuage Kubernetes Monitor on the Master node + + +Requirements +------------ +This role assumes it has been deployed on RHEL/Fedora diff --git a/roles/nuage_master/files/serviceaccount.sh b/roles/nuage_master/files/serviceaccount.sh new file mode 100644 index 000000000..f6fdb8a8d --- /dev/null +++ b/roles/nuage_master/files/serviceaccount.sh @@ -0,0 +1,63 @@ +#!/bin/bash +# Parse CLI options +for i in "$@"; do + case $i in + --master-cert-dir=*) + MASTER_DIR="${i#*=}" + CA_CERT=${MASTER_DIR}/ca.crt + CA_KEY=${MASTER_DIR}/ca.key + CA_SERIAL=${MASTER_DIR}/ca.serial.txt + ADMIN_FILE=${MASTER_DIR}/admin.kubeconfig + ;; + --server=*) + SERVER="${i#*=}" + ;; + --output-cert-dir=*) + OUTDIR="${i#*=}" + CONFIG_FILE=${OUTDIR}/nuage.kubeconfig + ;; + esac +done + +# If any are missing, print the usage and exit +if [ -z $SERVER ] || [ -z $OUTDIR ] || [ -z $MASTER_DIR ]; then + echo "Invalid syntax: $@" + echo "Usage:" + echo " $0 --server=<address>:<port> --output-cert-dir=/path/to/output/dir/ --master-cert-dir=/path/to/master/" + echo "--master-cert-dir: Directory where the master's configuration is held" + echo "--server: Address of Kubernetes API server (default port is 8443)" + echo "--output-cert-dir: Directory to put artifacts in" + echo "" + echo "All options are required" + exit 1 +fi + +# Login as admin so that we can create the service account +oc login -u system:admin --config=$ADMIN_FILE || exit 1 +oc project default --config=$ADMIN_FILE + +ACCOUNT_CONFIG=' +{ + "apiVersion": "v1", + "kind": "ServiceAccount", + "metadata": { + "name": "nuage" + } +} +' + +# Create the account with the included info +echo $ACCOUNT_CONFIG|oc create --config=$ADMIN_FILE -f - + +# Add the cluser-reader role, which allows this service account read access to +# everything in the cluster except secrets +oadm policy add-cluster-role-to-user cluster-reader system:serviceaccounts:default:nuage --config=$ADMIN_FILE + +# Generate certificates and a kubeconfig for the service account +oadm create-api-client-config --certificate-authority=${CA_CERT} --client-dir=${OUTDIR} --signer-cert=${CA_CERT} --signer-key=${CA_KEY} --signer-serial=${CA_SERIAL} --user=system:serviceaccounts:default:nuage --master=${SERVER} --public-master=${SERVER} --basename='nuage' + +# Verify the finalized kubeconfig +if ! [ $(oc whoami --config=$CONFIG_FILE) == 'system:serviceaccounts:default:nuage' ]; then + echo "Service account creation failed!" + exit 1 +fi diff --git a/roles/nuage_master/handlers/main.yaml b/roles/nuage_master/handlers/main.yaml new file mode 100644 index 000000000..635d8a419 --- /dev/null +++ b/roles/nuage_master/handlers/main.yaml @@ -0,0 +1,18 @@ +--- +- name: restart nuagekubemon + sudo: true + service: name=nuagekubemon state=restarted + +- name: restart master + service: name={{ openshift.common.service_type }}-master state=restarted + when: (not openshift_master_ha | bool) and (not master_service_status_changed | default(false)) + +- name: restart master api + service: name={{ openshift.common.service_type }}-master-api state=restarted + when: (openshift_master_ha | bool) and (not master_api_service_status_changed | default(false)) and openshift.master.cluster_method == 'native' + +# TODO: need to fix up ignore_errors here +- name: restart master controllers + service: name={{ openshift.common.service_type }}-master-controllers state=restarted + when: (openshift_master_ha | bool) and (not master_controllers_service_status_changed | default(false)) and openshift.master.cluster_method == 'native' + ignore_errors: yes diff --git a/roles/nuage_master/tasks/main.yaml b/roles/nuage_master/tasks/main.yaml new file mode 100644 index 000000000..a7baadc76 --- /dev/null +++ b/roles/nuage_master/tasks/main.yaml @@ -0,0 +1,34 @@ +--- +- name: Create directory /usr/share/nuagekubemon + sudo: true + file: path=/usr/share/nuagekubemon state=directory + +- name: Create the log directory + sudo: true + file: path={{ nuagekubemon_log_dir }} state=directory + +- name: Install Nuage Kubemon + sudo: true + yum: name={{ nuage_kubemon_rpm }} state=present + +- name: Run the service account creation script + sudo: true + script: serviceaccount.sh --server={{ openshift.master.api_url }} --output-cert-dir={{ cert_output_dir }} --master-cert-dir={{ openshift_master_config_dir }} + +- name: Download the certs and keys + sudo: true + fetch: src={{ cert_output_dir }}/{{ item }} dest=/tmp/{{ item }} flat=yes + with_items: + - ca.crt + - nuage.crt + - nuage.key + - nuage.kubeconfig + +- name: Create nuagekubemon.yaml + sudo: true + template: src=nuagekubemon.j2 dest=/usr/share/nuagekubemon/nuagekubemon.yaml owner=root mode=0644 + notify: + - restart master + - restart master api + - restart master controllers + - restart nuagekubemon diff --git a/roles/nuage_master/templates/nuagekubemon.j2 b/roles/nuage_master/templates/nuagekubemon.j2 new file mode 100644 index 000000000..fb586bcee --- /dev/null +++ b/roles/nuage_master/templates/nuagekubemon.j2 @@ -0,0 +1,19 @@ +# .kubeconfig that includes the nuage service account +kubeConfig: {{ kube_config }} +# name of the nuage service account, or another account with 'cluster-reader' +# permissions +# Openshift master config file +openshiftMasterConfig: {{ master_config_yaml }} +# URL of the VSD Architect +vsdApiUrl: {{ vsd_api_url }} +# API version to query against. Usually "v3_2" +vspVersion: {{ vsp_version }} +# File containing a VSP license to install. Only necessary if no license has +# been installed on the VSD Architect before, only valid for standalone vsd install +# licenseFile: "/path/to/base_vsp_license.txt" +# Name of the enterprise in which pods will reside +enterpriseName: {{ enterprise }} +# Name of the domain in which pods will reside +domainName: {{ domain }} +# Location where logs should be saved +log_dir: {{ nuagekubemon_log_dir }} diff --git a/roles/nuage_master/vars/main.yaml b/roles/nuage_master/vars/main.yaml new file mode 100644 index 000000000..db901fea6 --- /dev/null +++ b/roles/nuage_master/vars/main.yaml @@ -0,0 +1,7 @@ +openshift_master_config_dir: "{{ openshift.common.config_base }}/master" +ca_cert: "{{ openshift_master_config_dir }}/ca.crt" +admin_config: "{{ openshift.common.config_base }}/master/admin.kubeconfig" +cert_output_dir: /usr/share/nuagekubemon +kube_config: /usr/share/nuagekubemon/nuage.kubeconfig +kubemon_yaml: /usr/share/nuagekubemon/nuagekubemon.yaml +master_config_yaml: "{{ openshift_master_config_dir }}/master-config.yaml" diff --git a/roles/nuage_node/README.md b/roles/nuage_node/README.md new file mode 100644 index 000000000..02a3cbc77 --- /dev/null +++ b/roles/nuage_node/README.md @@ -0,0 +1,9 @@ +Nuage Node +========== + +Setup Nuage VRS (Virtual Routing Switching) on the Openshift Node + +Requirements +------------ + +This role assumes it has been deployed on RHEL/Fedora diff --git a/roles/nuage_node/handlers/main.yaml b/roles/nuage_node/handlers/main.yaml new file mode 100644 index 000000000..25482a845 --- /dev/null +++ b/roles/nuage_node/handlers/main.yaml @@ -0,0 +1,8 @@ +--- +- name: restart vrs + sudo: true + service: name=openvswitch state=restarted + +- name: restart node + sudo: true + service: name={{ openshift.common.service_type }}-node state=restarted diff --git a/roles/nuage_node/tasks/main.yaml b/roles/nuage_node/tasks/main.yaml new file mode 100644 index 000000000..e0117bf71 --- /dev/null +++ b/roles/nuage_node/tasks/main.yaml @@ -0,0 +1,37 @@ +--- +- name: Install Nuage VRS + sudo: true + yum: name={{ vrs_rpm }} state=present + +- name: Set the uplink interface + sudo: true + lineinfile: dest={{ vrs_config }} regexp=^NETWORK_UPLINK_INTF line='NETWORK_UPLINK_INTF={{ uplink_interface }}' + +- name: Set the Active Controller + sudo: true + lineinfile: dest={{ vrs_config }} regexp=^ACTIVE_CONTROLLER line='ACTIVE_CONTROLLER={{ vsc_active_ip }}' + +- name: Set the Standby Controller + sudo: true + lineinfile: dest={{ vrs_config }} regexp=^STANDBY_CONTROLLER line='STANDBY_CONTROLLER={{ vsc_standby_ip }}' + when: vsc_standby_ip is defined + +- name: Install plugin rpm + sudo: true + yum: name={{ plugin_rpm }} state=present + +- name: Copy the certificates and keys + sudo: true + copy: src="/tmp/{{ item }}" dest="{{ vsp_k8s_dir }}/{{ item }}" + with_items: + - ca.crt + - nuage.crt + - nuage.key + - nuage.kubeconfig + +- name: Set the vsp-k8s.yaml + sudo: true + template: src=vsp-k8s.j2 dest={{ vsp_k8s_yaml }} owner=root mode=0644 + notify: + - restart vrs + - restart node diff --git a/roles/nuage_node/templates/vsp-k8s.j2 b/roles/nuage_node/templates/vsp-k8s.j2 new file mode 100644 index 000000000..98d6c3a9c --- /dev/null +++ b/roles/nuage_node/templates/vsp-k8s.j2 @@ -0,0 +1,14 @@ +clientCert: {{ client_cert }} +# The key to the certificate in clientCert above +clientKey: {{ client_key }} +# The certificate authority's certificate for the local kubelet. Usually the +# same as the CA cert used to create the client Cert/Key pair. +CACert: {{ ca_cert }} +# Name of the enterprise in which pods will reside +enterpriseName: {{ enterprise }} +# Name of the domain in which pods will reside +domainName: {{ domain }} +# IP address and port number of master API server +masterApiServer: {{ api_server }} +# Bridge name for the docker bridge +dockerBridgeName: {{ docker_bridge }} diff --git a/roles/nuage_node/vars/main.yaml b/roles/nuage_node/vars/main.yaml new file mode 100644 index 000000000..a6b7cf997 --- /dev/null +++ b/roles/nuage_node/vars/main.yaml @@ -0,0 +1,9 @@ +--- +vrs_config: /etc/default/openvswitch +vsp_k8s_dir: /usr/share/vsp-k8s +vsp_k8s_yaml: "{{ vsp_k8s_dir }}/vsp-k8s.yaml" +client_cert: "{{ vsp_k8s_dir }}/nuage.crt" +client_key: "{{ vsp_k8s_dir }}/nuage.key" +ca_cert: "{{ vsp_k8s_dir }}/ca.crt" +api_server: "{{ openshift_node_master_api_url }}" +docker_bridge: "docker0" diff --git a/roles/openshift_cli/tasks/main.yml b/roles/openshift_cli/tasks/main.yml index 8d7686ffd..a6b6b1925 100644 --- a/roles/openshift_cli/tasks/main.yml +++ b/roles/openshift_cli/tasks/main.yml @@ -3,32 +3,17 @@ role: common local_facts: deployment_type: "{{ openshift_deployment_type }}" + cli_image: "{{ osm_image | default(None) }}" - name: Install clients - yum: pkg={{ openshift.common.service_type }}-clients state=installed + action: "{{ ansible_pkg_mgr }} name={{ openshift.common.service_type }}-clients state=present" when: not openshift.common.is_containerized | bool -- name: List Docker images - command: > - docker images - register: docker_images - - name: Pull CLI Image command: > docker pull {{ openshift.common.cli_image }} - when: openshift.common.is_containerized | bool and openshift.common.cli_image not in docker_images.stdout - -- name: Wait for CLI image - command: > - docker images - register: docker_images - until: openshift.common.cli_image in docker_images.stdout - retries: 30 - delay: 10 - changed_when: false when: openshift.common.is_containerized | bool - - name: Create /usr/local/bin/openshift cli wrapper template: src: openshift.j2 diff --git a/roles/openshift_common/tasks/main.yml b/roles/openshift_common/tasks/main.yml index 0ee873a2b..ff8c3b50f 100644 --- a/roles/openshift_common/tasks/main.yml +++ b/roles/openshift_common/tasks/main.yml @@ -4,6 +4,14 @@ when: openshift_use_openshift_sdn | default(false) | bool and openshift_use_flannel | default(false) | bool - fail: + msg: Nuage sdn can not be used with openshift sdn + when: openshift_use_openshift_sdn | default(false) | bool and openshift_use_nuage | default(false) | bool + +- fail: + msg: Nuage sdn can not be used with flannel + when: openshift_use_flannel | default(false) | bool and openshift_use_nuage | default(false) | bool + +- fail: msg: openshift_hostname must be 64 characters or less when: openshift_hostname is defined and openshift_hostname | length > 64 @@ -23,7 +31,9 @@ deployment_type: "{{ openshift_deployment_type }}" use_fluentd: "{{ openshift_use_fluentd | default(None) }}" use_flannel: "{{ openshift_use_flannel | default(None) }}" + use_nuage: "{{ openshift_use_nuage | default(None) }}" use_manageiq: "{{ openshift_use_manageiq | default(None) }}" + data_dir: "{{ openshift_data_dir | default(None) }}" - name: Install the base package for versioning action: "{{ ansible_pkg_mgr }} name={{ openshift.common.service_type }}{{ openshift_version | default('') }} state=present" @@ -38,5 +48,6 @@ set_hostname_default: "{{ not openshift.common.version_greater_than_3_1_or_1_1 }}" - name: Set hostname - hostname: name={{ openshift.common.hostname }} + command: > + hostnamectl set-hostname {{ openshift.common.hostname }} when: openshift_set_hostname | default(set_hostname_default) | bool diff --git a/roles/openshift_common/vars/main.yml b/roles/openshift_common/vars/main.yml index 50816d319..b163f8aae 100644 --- a/roles/openshift_common/vars/main.yml +++ b/roles/openshift_common/vars/main.yml @@ -5,3 +5,4 @@ # chains with the public zone (or the zone associated with the correct # interfaces) os_firewall_use_firewalld: False +openshift_version: "{{ openshift_pkg_version | default('') }}" diff --git a/roles/openshift_facts/library/openshift_facts.py b/roles/openshift_facts/library/openshift_facts.py index 2a3d4acbd..40e54d706 100755 --- a/roles/openshift_facts/library/openshift_facts.py +++ b/roles/openshift_facts/library/openshift_facts.py @@ -188,9 +188,6 @@ def normalize_gce_facts(metadata, facts): _, _, zone = metadata['instance']['zone'].rpartition('/') facts['zone'] = zone - # Default to no sdn for GCE deployments - facts['use_openshift_sdn'] = False - # GCE currently only supports a single interface facts['network']['ip'] = facts['network']['interfaces'][0]['ips'][0] pub_ip = facts['network']['interfaces'][0]['public_ips'][0] @@ -341,6 +338,23 @@ def set_flannel_facts_if_unset(facts): facts['common']['use_flannel'] = use_flannel return facts +def set_nuage_facts_if_unset(facts): + """ Set nuage facts if not already present in facts dict + dict: the facts dict updated with the nuage facts if + missing + Args: + facts (dict): existing facts + Returns: + dict: the facts dict updated with the nuage + facts if they were not already present + + """ + if 'common' in facts: + if 'use_nuage' not in facts['common']: + use_nuage = False + facts['common']['use_nuage'] = use_nuage + return facts + def set_node_schedulability(facts): """ Set schedulable facts if not already present in facts dict Args: @@ -461,52 +475,68 @@ def set_url_facts_if_unset(facts): were not already present """ if 'master' in facts: - api_use_ssl = facts['master']['api_use_ssl'] - api_port = facts['master']['api_port'] - console_use_ssl = facts['master']['console_use_ssl'] - console_port = facts['master']['console_port'] - console_path = facts['master']['console_path'] - etcd_use_ssl = facts['master']['etcd_use_ssl'] - etcd_hosts = facts['master']['etcd_hosts'] - etcd_port = facts['master']['etcd_port'] hostname = facts['common']['hostname'] - public_hostname = facts['common']['public_hostname'] cluster_hostname = facts['master'].get('cluster_hostname') cluster_public_hostname = facts['master'].get('cluster_public_hostname') + public_hostname = facts['common']['public_hostname'] + api_hostname = cluster_hostname if cluster_hostname else hostname + api_public_hostname = cluster_public_hostname if cluster_public_hostname else public_hostname + console_path = facts['master']['console_path'] + etcd_hosts = facts['master']['etcd_hosts'] + + use_ssl = dict( + api=facts['master']['api_use_ssl'], + public_api=facts['master']['api_use_ssl'], + loopback_api=facts['master']['api_use_ssl'], + console=facts['master']['console_use_ssl'], + public_console=facts['master']['console_use_ssl'], + etcd=facts['master']['etcd_use_ssl'] + ) + + ports = dict( + api=facts['master']['api_port'], + public_api=facts['master']['api_port'], + loopback_api=facts['master']['api_port'], + console=facts['master']['console_port'], + public_console=facts['master']['console_port'], + etcd=facts['master']['etcd_port'], + ) + + etcd_urls = [] + if etcd_hosts != '': + facts['master']['etcd_port'] = ports['etcd'] + facts['master']['embedded_etcd'] = False + for host in etcd_hosts: + etcd_urls.append(format_url(use_ssl['etcd'], host, + ports['etcd'])) + else: + etcd_urls = [format_url(use_ssl['etcd'], hostname, + ports['etcd'])] + + facts['master'].setdefault('etcd_urls', etcd_urls) + + prefix_hosts = [('api', api_hostname), + ('public_api', api_public_hostname), + ('loopback_api', hostname)] + + for prefix, host in prefix_hosts: + facts['master'].setdefault(prefix + '_url', format_url(use_ssl[prefix], + host, + ports[prefix])) + + + r_lhn = "{0}:{1}".format(api_hostname, ports['api']).replace('.', '-') + facts['master'].setdefault('loopback_cluster_name', r_lhn) + facts['master'].setdefault('loopback_context_name', "default/{0}/system:openshift-master".format(r_lhn)) + facts['master'].setdefault('loopback_user', "system:openshift-master/{0}".format(r_lhn)) + + prefix_hosts = [('console', api_hostname), ('public_console', api_public_hostname)] + for prefix, host in prefix_hosts: + facts['master'].setdefault(prefix + '_url', format_url(use_ssl[prefix], + host, + ports[prefix], + console_path)) - if 'etcd_urls' not in facts['master']: - etcd_urls = [] - if etcd_hosts != '': - facts['master']['etcd_port'] = etcd_port - facts['master']['embedded_etcd'] = False - for host in etcd_hosts: - etcd_urls.append(format_url(etcd_use_ssl, host, - etcd_port)) - else: - etcd_urls = [format_url(etcd_use_ssl, hostname, - etcd_port)] - facts['master']['etcd_urls'] = etcd_urls - if 'api_url' not in facts['master']: - api_hostname = cluster_hostname if cluster_hostname else hostname - facts['master']['api_url'] = format_url(api_use_ssl, api_hostname, - api_port) - if 'public_api_url' not in facts['master']: - api_public_hostname = cluster_public_hostname if cluster_public_hostname else public_hostname - facts['master']['public_api_url'] = format_url(api_use_ssl, - api_public_hostname, - api_port) - if 'console_url' not in facts['master']: - console_hostname = cluster_hostname if cluster_hostname else hostname - facts['master']['console_url'] = format_url(console_use_ssl, - console_hostname, - console_port, - console_path) - if 'public_console_url' not in facts['master']: - console_public_hostname = cluster_public_hostname if cluster_public_hostname else public_hostname - facts['master']['public_console_url'] = format_url(console_use_ssl, - console_public_hostname, - console_port, - console_path) return facts def set_aggregate_facts(facts): @@ -884,10 +914,6 @@ def apply_provider_facts(facts, provider_facts): if not provider_facts: return facts - use_openshift_sdn = provider_facts.get('use_openshift_sdn') - if isinstance(use_openshift_sdn, bool): - facts['common']['use_openshift_sdn'] = use_openshift_sdn - common_vars = [('hostname', 'ip'), ('public_hostname', 'public_ip')] for h_var, ip_var in common_vars: ip_value = provider_facts['network'].get(ip_var) @@ -1038,6 +1064,10 @@ def set_container_facts_if_unset(facts): if 'ovs_image' not in facts['node']: facts['node']['ovs_image'] = ovs_image + if facts['common']['is_containerized']: + facts['common']['admin_binary'] = '/usr/local/bin/oadm' + facts['common']['client_binary'] = '/usr/local/bin/oc' + return facts @@ -1078,7 +1108,7 @@ class OpenShiftFacts(object): Raises: OpenShiftFactsUnsupportedRoleError: """ - known_roles = ['common', 'master', 'node', 'master_sdn', 'node_sdn', 'etcd', 'nfs'] + known_roles = ['common', 'master', 'node', 'etcd', 'nfs'] def __init__(self, role, filename, local_facts, additive_facts_to_overwrite=False): self.changed = False @@ -1115,6 +1145,7 @@ class OpenShiftFacts(object): facts = set_project_cfg_facts_if_unset(facts) facts = set_fluentd_facts_if_unset(facts) facts = set_flannel_facts_if_unset(facts) + facts = set_nuage_facts_if_unset(facts) facts = set_node_schedulability(facts) facts = set_master_selectors(facts) facts = set_metrics_facts_if_unset(facts) @@ -1156,7 +1187,7 @@ class OpenShiftFacts(object): defaults['common'] = common if 'master' in roles: - master = dict(api_use_ssl=True, api_port='8443', + master = dict(api_use_ssl=True, api_port='8443', controllers_port='8444', console_use_ssl=True, console_path='/console', console_port='8443', etcd_use_ssl=True, etcd_hosts='', etcd_port='4001', portal_net='172.30.0.0/16', diff --git a/roles/openshift_facts/tasks/main.yml b/roles/openshift_facts/tasks/main.yml index 87fa99a3b..0dbac1b54 100644 --- a/roles/openshift_facts/tasks/main.yml +++ b/roles/openshift_facts/tasks/main.yml @@ -1,15 +1,14 @@ --- -- name: Verify Ansible version is greater than 1.8.0 and not 1.9.0 and not 1.9.0.1 - assert: - that: - - ansible_version | version_compare('1.8.0', 'ge') - - ansible_version | version_compare('1.9.0', 'ne') - - ansible_version | version_compare('1.9.0.1', 'ne') - +- name: Verify Ansible version is greater than or equal to 1.9.4 and less than 2.0 + fail: + msg: "Unsupported ansible version: {{ ansible_version }} found" + when: ansible_version.full | version_compare('1.9.4', 'lt') or ansible_version.full | version_compare('2.0', 'ge') + - name: Detecting Operating System shell: ls /run/ostree-booted ignore_errors: yes failed_when: false + changed_when: false register: ostree_output # Locally setup containerized facts for now diff --git a/roles/openshift_master/defaults/main.yml b/roles/openshift_master/defaults/main.yml index 9766d01ae..1f74d851a 100644 --- a/roles/openshift_master/defaults/main.yml +++ b/roles/openshift_master/defaults/main.yml @@ -6,7 +6,9 @@ os_firewall_allow: - service: etcd embedded port: 4001/tcp - service: api server https - port: 8443/tcp + port: "{{ openshift.master.api_port }}/tcp" +- service: api controllers https + port: "{{ openshift.master.controllers_port }}/tcp" - service: dns tcp port: 53/tcp - service: dns udp @@ -24,7 +26,5 @@ os_firewall_allow: os_firewall_deny: - service: api server http port: 8080/tcp -- service: former web console port - port: 8444/tcp - service: former etcd peer port port: 7001/tcp diff --git a/roles/openshift_master/handlers/main.yml b/roles/openshift_master/handlers/main.yml index e1b95eda4..6b9992eea 100644 --- a/roles/openshift_master/handlers/main.yml +++ b/roles/openshift_master/handlers/main.yml @@ -2,11 +2,24 @@ - name: restart master service: name={{ openshift.common.service_type }}-master state=restarted when: (not openshift_master_ha | bool) and (not (master_service_status_changed | default(false) | bool)) + notify: Verify API Server - name: restart master api service: name={{ openshift.common.service_type }}-master-api state=restarted when: (openshift_master_ha | bool) and (not (master_api_service_status_changed | default(false) | bool)) and openshift.master.cluster_method == 'native' + notify: Verify API Server - name: restart master controllers service: name={{ openshift.common.service_type }}-master-controllers state=restarted when: (openshift_master_ha | bool) and (not (master_controllers_service_status_changed | default(false) | bool)) and openshift.master.cluster_method == 'native' + +- name: Verify API Server + # Using curl here since the uri module requires python-httplib2 and + # wait_for port doesn't provide health information. + command: > + curl -k --head --silent {{ openshift.master.api_url }} + register: api_available_output + until: api_available_output.stdout.find("200 OK") != -1 + retries: 120 + delay: 1 + changed_when: false diff --git a/roles/openshift_master/tasks/main.yml b/roles/openshift_master/tasks/main.yml index 397122631..57b50bee4 100644 --- a/roles/openshift_master/tasks/main.yml +++ b/roles/openshift_master/tasks/main.yml @@ -42,7 +42,12 @@ public_console_url: "{{ openshift_master_public_console_url | default(None) }}" logging_public_url: "{{ openshift_master_logging_public_url | default(None) }}" metrics_public_url: "{{ openshift_master_metrics_public_url | default(None) }}" - etcd_hosts: "{{ openshift_master_etcd_hosts | default(None)}}" + logout_url: "{{ openshift_master_logout_url | default(None) }}" + extension_scripts: "{{ openshift_master_extension_scripts | default(None) }}" + extension_stylesheets: "{{ openshift_master_extension_stylesheets | default(None) }}" + extensions: "{{ openshift_master_extensions | default(None) }}" + oauth_template: "{{ openshift_master_oauth_template | default(None) }}" + etcd_hosts: "{{ openshift_master_etcd_hosts | default(None) }}" etcd_port: "{{ openshift_master_etcd_port | default(None) }}" etcd_use_ssl: "{{ openshift_master_etcd_use_ssl | default(None) }}" etcd_urls: "{{ openshift_master_etcd_urls | default(None) }}" @@ -51,6 +56,7 @@ embedded_dns: "{{ openshift_master_embedded_dns | default(None) }}" dns_port: "{{ openshift_master_dns_port | default(None) }}" bind_addr: "{{ openshift_master_bind_addr | default(None) }}" + pod_eviction_timeout: "{{ openshift_master_pod_eviction_timeout | default(None) }}" portal_net: "{{ openshift_master_portal_net | default(None) }}" session_max_seconds: "{{ openshift_master_session_max_seconds | default(None) }}" session_name: "{{ openshift_master_session_name | default(None) }}" @@ -86,36 +92,20 @@ action: "{{ ansible_pkg_mgr }} name={{ openshift.common.service_type }}-master{{ openshift_version }} state=present" when: not openshift.common.is_containerized | bool -- name: Get docker images - command: docker images - changed_when: false - when: openshift.common.is_containerized | bool - register: docker_images - - name: Pull master image command: > docker pull {{ openshift.master.master_image }} - when: openshift.common.is_containerized | bool and openshift.master.master_image not in docker_images.stdout - -- name: Wait for master image - command: > - docker images - register: docker_images - until: openshift.master.master_image in docker_images.stdout - retries: 30 - delay: 10 - changed_when: false when: openshift.common.is_containerized | bool - name: Install Master docker service file template: dest: "/etc/systemd/system/{{ openshift.common.service_type }}-master.service" - src: master.docker.service.j2 + src: docker/master.docker.service.j2 register: install_result when: openshift.common.is_containerized | bool and not openshift_master_ha | bool - + - name: Create openshift.common.data_dir - file: + file: path: "{{ openshift.common.data_dir }}" state: directory mode: 0755 @@ -190,31 +180,42 @@ when: openshift.common.is_containerized | bool # workaround for missing systemd unit files for controllers/api -- name: Create the api service file +- name: Create the systemd unit files template: - src: atomic-openshift-master-api{{ ha_suffix }}.service.j2 - dest: "{{ ha_svcdir }}/{{ openshift.common.service_type }}-master-api.service" + src: "{{ ha_svc_template_path }}/atomic-openshift-master-{{ item }}.service.j2" + dest: "{{ ha_svcdir }}/{{ openshift.common.service_type }}-master-{{ item }}.service" when: openshift_master_ha | bool and openshift_master_cluster_method == "native" -- name: Create the controllers service file - template: - src: atomic-openshift-master-controllers{{ ha_suffix }}.service.j2 - dest: "{{ ha_svcdir }}/{{ openshift.common.service_type }}-master-controllers.service" - when: openshift_master_ha | bool and openshift_master_cluster_method == "native" -- name: Create the api env file + with_items: + - api + - controllers + register: create_unit_files + +- command: systemctl daemon-reload + when: create_unit_files | changed +# end workaround for missing systemd unit files + +- name: Create the master api service env file template: - src: atomic-openshift-master-api.j2 + src: "{{ ha_svc_template_path }}/atomic-openshift-master-api.j2" dest: /etc/sysconfig/{{ openshift.common.service_type }}-master-api - force: no when: openshift_master_ha | bool and openshift_master_cluster_method == "native" -- name: Create the controllers env file + notify: + - restart master api + +- name: Create the master controllers service env file template: - src: atomic-openshift-master-controllers.j2 + src: "{{ ha_svc_template_path }}/atomic-openshift-master-controllers.j2" dest: /etc/sysconfig/{{ openshift.common.service_type }}-master-controllers - force: no when: openshift_master_ha | bool and openshift_master_cluster_method == "native" -- command: systemctl daemon-reload - when: openshift_master_ha | bool and openshift_master_cluster_method == "native" -# end workaround for missing systemd unit files + notify: + - restart master controllers + +- name: Create the master service env file + template: + src: "atomic-openshift-master.j2" + dest: /etc/sysconfig/{{ openshift.common.service_type }}-master + notify: + - restart master - name: Create session secrets file template: @@ -239,52 +240,42 @@ - restart master api - restart master controllers -- name: Configure master settings - lineinfile: - dest: /etc/sysconfig/{{ openshift.common.service_type }}-master - regexp: "{{ item.regex }}" - line: "{{ item.line }}" - create: yes - with_items: - - regex: '^OPTIONS=' - line: "OPTIONS=--loglevel={{ openshift.master.debug_level }}" - - regex: '^CONFIG_FILE=' - line: "CONFIG_FILE={{ openshift_master_config_file }}" - notify: - - restart master - -- name: Configure master api settings - lineinfile: - dest: /etc/sysconfig/{{ openshift.common.service_type }}-master-api - regexp: "{{ item.regex }}" - line: "{{ item.line }}" - with_items: - - regex: '^OPTIONS=' - line: "OPTIONS=--loglevel={{ openshift.master.debug_level }} --listen=https://0.0.0.0:8443 --master=https://{{ openshift.common.ip }}:8443" - - regex: '^CONFIG_FILE=' - line: "CONFIG_FILE={{ openshift_master_config_file }}" - when: openshift_master_ha | bool and openshift_master_cluster_method == "native" - notify: - - restart master api - -- name: Configure master controller settings - lineinfile: - dest: /etc/sysconfig/{{ openshift.common.service_type }}-master-controllers - regexp: "{{ item.regex }}" - line: "{{ item.line }}" - with_items: - - regex: '^OPTIONS=' - line: "OPTIONS=--loglevel={{ openshift.master.debug_level }} --listen=https://0.0.0.0:8444" - - regex: '^CONFIG_FILE=' - line: "CONFIG_FILE={{ openshift_master_config_file }}" - when: openshift_master_ha | bool and openshift_master_cluster_method == "native" - notify: - - restart master controllers +- name: Test local loopback context + command: > + {{ openshift.common.client_binary }} config view + --config={{ openshift_master_loopback_config }} + changed_when: false + register: loopback_config + +- command: > + {{ openshift.common.client_binary }} config set-cluster + --certificate-authority={{ openshift_master_config_dir }}/ca.crt + --embed-certs=true --server={{ openshift.master.loopback_api_url }} + {{ openshift.master.loopback_cluster_name }} + --config={{ openshift_master_loopback_config }} + when: loopback_context_string not in loopback_config.stdout + register: set_loopback_cluster + +- command: > + {{ openshift.common.client_binary }} config set-context + --cluster={{ openshift.master.loopback_cluster_name }} + --namespace=default --user={{ openshift.master.loopback_user }} + {{ openshift.master.loopback_context_name }} + --config={{ openshift_master_loopback_config }} + when: set_loopback_cluster | changed + register: set_loopback_context + +- command: > + {{ openshift.common.client_binary }} config use-context {{ openshift.master.loopback_context_name }} + --config={{ openshift_master_loopback_config }} + when: set_loopback_context | changed + register: set_current_context - name: Start and enable master service: name={{ openshift.common.service_type }}-master enabled=yes state=started when: not openshift_master_ha | bool register: start_result + notify: Verify API Server - name: Stop and disable non HA master when running HA service: name={{ openshift.common.service_type }}-master enabled=no state=stopped @@ -303,6 +294,20 @@ master_api_service_status_changed: "{{ start_result | changed }}" when: openshift_master_ha | bool and openshift.master.cluster_method == 'native' +# A separate wait is required here for native HA since notifies will +# be resolved after all tasks in the role. +- name: Wait for API to become available + # Using curl here since the uri module requires python-httplib2 and + # wait_for port doesn't provide health information. + command: > + curl -k --head --silent {{ openshift.master.api_url }} + register: api_available_output + until: api_available_output.stdout.find("200 OK") != -1 + retries: 120 + delay: 1 + changed_when: false + when: openshift_master_ha | bool and openshift.master.cluster_method == 'native' and master_api_service_status_changed | bool + - name: Start and enable master controller service: name={{ openshift.common.service_type }}-master-controllers enabled=yes state=started when: openshift_master_ha | bool and openshift.master.cluster_method == 'native' diff --git a/roles/openshift_master/templates/atomic-openshift-master-controllers.j2 b/roles/openshift_master/templates/atomic-openshift-master.j2 index 205934248..81bae5470 100644 --- a/roles/openshift_master/templates/atomic-openshift-master-controllers.j2 +++ b/roles/openshift_master/templates/atomic-openshift-master.j2 @@ -1,5 +1,5 @@ -OPTIONS= -CONFIG_FILE={{ openshift_master_config_dir }}/master-config.yaml +OPTIONS=--loglevel={{ openshift.master.debug_level }} +CONFIG_FILE={{ openshift_master_config_file }} # Proxy configuration # Origin uses standard HTTP_PROXY environment variables. Be sure to set diff --git a/roles/openshift_master/templates/docker-cluster/atomic-openshift-master-api.j2 b/roles/openshift_master/templates/docker-cluster/atomic-openshift-master-api.j2 new file mode 120000 index 000000000..4bb7095ee --- /dev/null +++ b/roles/openshift_master/templates/docker-cluster/atomic-openshift-master-api.j2 @@ -0,0 +1 @@ +../native-cluster/atomic-openshift-master-api.j2
\ No newline at end of file diff --git a/roles/openshift_master/templates/atomic-openshift-master-api.docker.service.j2 b/roles/openshift_master/templates/docker-cluster/atomic-openshift-master-api.service.j2 index 936c39edf..a935b82f6 100644 --- a/roles/openshift_master/templates/atomic-openshift-master-api.docker.service.j2 +++ b/roles/openshift_master/templates/docker-cluster/atomic-openshift-master-api.service.j2 @@ -23,4 +23,4 @@ Restart=always [Install] WantedBy=multi-user.target -WantedBy={{ openshift.common.service_type }}-node.service
\ No newline at end of file +WantedBy={{ openshift.common.service_type }}-node.service diff --git a/roles/openshift_master/templates/docker-cluster/atomic-openshift-master-controllers.j2 b/roles/openshift_master/templates/docker-cluster/atomic-openshift-master-controllers.j2 new file mode 120000 index 000000000..8714ebbae --- /dev/null +++ b/roles/openshift_master/templates/docker-cluster/atomic-openshift-master-controllers.j2 @@ -0,0 +1 @@ +../native-cluster/atomic-openshift-master-controllers.j2
\ No newline at end of file diff --git a/roles/openshift_master/templates/atomic-openshift-master-controllers.docker.service.j2 b/roles/openshift_master/templates/docker-cluster/atomic-openshift-master-controllers.service.j2 index 6ba7d6e2a..6ba7d6e2a 100644 --- a/roles/openshift_master/templates/atomic-openshift-master-controllers.docker.service.j2 +++ b/roles/openshift_master/templates/docker-cluster/atomic-openshift-master-controllers.service.j2 diff --git a/roles/openshift_master/templates/master.docker.service.j2 b/roles/openshift_master/templates/docker/master.docker.service.j2 index 23781a313..23781a313 100644 --- a/roles/openshift_master/templates/master.docker.service.j2 +++ b/roles/openshift_master/templates/docker/master.docker.service.j2 diff --git a/roles/openshift_master/templates/master.yaml.v1.j2 b/roles/openshift_master/templates/master.yaml.v1.j2 index 647476b7f..1eeab46fe 100644 --- a/roles/openshift_master/templates/master.yaml.v1.j2 +++ b/roles/openshift_master/templates/master.yaml.v1.j2 @@ -5,7 +5,7 @@ apiLevels: - v1 apiVersion: v1 assetConfig: - logoutURL: "" + logoutURL: "{{ openshift.master.logout_url | default('') }}" masterPublicURL: {{ openshift.master.public_api_url }} publicURL: {{ openshift.master.public_console_url }}/ {% if 'logging_public_url' in openshift.master %} @@ -14,6 +14,15 @@ assetConfig: {% if 'metrics_public_url' in openshift.master %} metricsPublicURL: {{ openshift.master.metrics_public_url }} {% endif %} +{% if 'extension_scripts' in openshift.master %} + extensionScripts: {{ openshift.master.extension_scripts | to_padded_yaml(1, 2) }} +{% endif %} +{% if 'extension_stylesheets' in openshift.master %} + extensionStylesheets: {{ openshift.master.extension_stylesheets | to_padded_yaml(1, 2) }} +{% endif %} +{% if 'extensions' in openshift.master %} + extensions: {{ openshift.master.extensions | to_padded_yaml(1, 2) }} +{% endif %} servingInfo: bindAddress: {{ openshift.master.bind_addr }}:{{ openshift.master.console_port }} bindNetwork: tcp4 @@ -87,11 +96,11 @@ kubernetesMasterConfig: - v1beta3 - v1 {% endif %} - apiServerArguments: {{ openshift.master.api_server_args | default(None) | to_json }} - controllerArguments: {{ openshift.master.controller_args | default(None) | to_json }} + apiServerArguments: {{ openshift.master.api_server_args | default(None) | to_padded_yaml( level=2 ) }} + controllerArguments: {{ openshift.master.controller_args | default(None) | to_padded_yaml( level=2 ) }} masterCount: {{ openshift.master.master_count if openshift.master.cluster_method | default(None) == 'native' else 1 }} masterIP: {{ openshift.common.ip }} - podEvictionTimeout: "" + podEvictionTimeout: {{ openshift.master.pod_eviction_timeout | default("") }} proxyClientInfo: certFile: master.proxy-client.crt keyFile: master.proxy-client.key @@ -108,12 +117,16 @@ masterPublicURL: {{ openshift.master.public_api_url }} networkConfig: clusterNetworkCIDR: {{ openshift.master.sdn_cluster_network_cidr }} hostSubnetLength: {{ openshift.master.sdn_host_subnet_length }} -{% if openshift.common.use_openshift_sdn %} +{% if openshift.common.use_openshift_sdn or openshift.common.use_nuage %} networkPluginName: {{ openshift.common.sdn_network_plugin_name }} {% endif %} # serviceNetworkCIDR must match kubernetesMasterConfig.servicesSubnet serviceNetworkCIDR: {{ openshift.master.portal_net }} oauthConfig: +{% if 'oauth_template' in openshift.master %} + templates: + login: {{ openshift.master.oauth_template }} +{% endif %} assetPublicURL: {{ openshift.master.public_console_url }}/ grantConfig: method: {{ openshift.master.oauth_grant_method }} diff --git a/roles/openshift_master/templates/native-cluster/atomic-openshift-master-api.j2 b/roles/openshift_master/templates/native-cluster/atomic-openshift-master-api.j2 new file mode 100644 index 000000000..48bfa5f04 --- /dev/null +++ b/roles/openshift_master/templates/native-cluster/atomic-openshift-master-api.j2 @@ -0,0 +1,9 @@ +OPTIONS=--loglevel={{ openshift.master.debug_level }} --listen={{ 'https' if openshift.master.api_use_ssl else 'http' }}://{{ openshift.master.bind_addr }}:{{ openshift.master.api_port }} --master={{ openshift.master.loopback_api_url }} +CONFIG_FILE={{ openshift_master_config_file }} + +# Proxy configuration +# Origin uses standard HTTP_PROXY environment variables. Be sure to set +# NO_PROXY for your master +#NO_PROXY=master.example.com +#HTTP_PROXY=http://USER:PASSWORD@IPADDR:PORT +#HTTPS_PROXY=https://USER:PASSWORD@IPADDR:PORT diff --git a/roles/openshift_master/templates/atomic-openshift-master-api.service.j2 b/roles/openshift_master/templates/native-cluster/atomic-openshift-master-api.service.j2 index ba19fb348..ba19fb348 100644 --- a/roles/openshift_master/templates/atomic-openshift-master-api.service.j2 +++ b/roles/openshift_master/templates/native-cluster/atomic-openshift-master-api.service.j2 diff --git a/roles/openshift_master/templates/atomic-openshift-master-api.j2 b/roles/openshift_master/templates/native-cluster/atomic-openshift-master-controllers.j2 index 205934248..cdc56eece 100644 --- a/roles/openshift_master/templates/atomic-openshift-master-api.j2 +++ b/roles/openshift_master/templates/native-cluster/atomic-openshift-master-controllers.j2 @@ -1,5 +1,5 @@ -OPTIONS= -CONFIG_FILE={{ openshift_master_config_dir }}/master-config.yaml +OPTIONS=--loglevel={{ openshift.master.debug_level }} --listen={{ 'https' if openshift.master.api_use_ssl else 'http' }}://{{ openshift.master.bind_addr }}:{{ openshift.master.controllers_port }} +CONFIG_FILE={{ openshift_master_config_file }} # Proxy configuration # Origin uses standard HTTP_PROXY environment variables. Be sure to set diff --git a/roles/openshift_master/templates/atomic-openshift-master-controllers.service.j2 b/roles/openshift_master/templates/native-cluster/atomic-openshift-master-controllers.service.j2 index e6e97b24f..e6e97b24f 100644 --- a/roles/openshift_master/templates/atomic-openshift-master-controllers.service.j2 +++ b/roles/openshift_master/templates/native-cluster/atomic-openshift-master-controllers.service.j2 diff --git a/roles/openshift_master/vars/main.yml b/roles/openshift_master/vars/main.yml index 534465451..fe88c3c16 100644 --- a/roles/openshift_master/vars/main.yml +++ b/roles/openshift_master/vars/main.yml @@ -1,11 +1,16 @@ --- openshift_master_config_dir: "{{ openshift.common.config_base }}/master" openshift_master_config_file: "{{ openshift_master_config_dir }}/master-config.yaml" +openshift_master_loopback_config: "{{ openshift_master_config_dir }}/openshift-master.kubeconfig" +loopback_context_string: "current-context: {{ openshift.master.loopback_context_name }}" openshift_master_scheduler_conf: "{{ openshift_master_config_dir }}/scheduler.json" openshift_master_session_secrets_file: "{{ openshift_master_config_dir }}/session-secrets.yaml" openshift_master_policy: "{{ openshift_master_config_dir }}/policy.json" openshift_version: "{{ openshift_pkg_version | default('') }}" +ha_svc_template_path: "{{ 'docker-cluster' if openshift.common.is_containerized | bool else 'native-cluster' }}" +ha_svc_svc_dir: "{{ '/etc/systemd/system' if openshift.common.is_containerized | bool else '/usr/lib/systemd/system' }}" + openshift_master_valid_grant_methods: - auto - prompt diff --git a/roles/openshift_master_ca/tasks/main.yml b/roles/openshift_master_ca/tasks/main.yml index 5b4c92f2b..6d9be81c0 100644 --- a/roles/openshift_master_ca/tasks/main.yml +++ b/roles/openshift_master_ca/tasks/main.yml @@ -13,16 +13,10 @@ path: "{{ openshift_master_config_dir }}" state: directory -- name: Get docker images - command: docker images - changed_when: false - when: openshift.common.is_containerized | bool - register: docker_images - -- name: Pull required docker image +- name: Pull master docker image command: > docker pull {{ openshift.common.cli_image }} - when: openshift.common.is_containerized | bool and openshift.common.cli_image not in docker_images.stdout + when: openshift.common.is_containerized | bool - name: Create the master certificates if they do not already exist command: > diff --git a/roles/openshift_master_cluster/tasks/configure.yml b/roles/openshift_master_cluster/tasks/configure.yml index 7ab9afb51..1b94598dd 100644 --- a/roles/openshift_master_cluster/tasks/configure.yml +++ b/roles/openshift_master_cluster/tasks/configure.yml @@ -34,11 +34,10 @@ - name: Disable stonith command: pcs property set stonith-enabled=false -# TODO: handle case where api port is not 8443 - name: Wait for the clustered master service to be available wait_for: host: "{{ openshift_master_cluster_vip }}" - port: 8443 + port: "{{ openshift.master.api_port }}" state: started timeout: 180 delay: 90 diff --git a/roles/openshift_node/tasks/main.yml b/roles/openshift_node/tasks/main.yml index 33852d7f8..9035248f9 100644 --- a/roles/openshift_node/tasks/main.yml +++ b/roles/openshift_node/tasks/main.yml @@ -44,41 +44,14 @@ action: "{{ ansible_pkg_mgr }} name={{ openshift.common.service_type }}-sdn-ovs{{ openshift_version }} state=present" when: openshift.common.use_openshift_sdn and not openshift.common.is_containerized | bool -- name: Get docker images - command: docker images - changed_when: false - when: openshift.common.is_containerized | bool - register: docker_images - - name: Pull node image command: > docker pull {{ openshift.node.node_image }} - when: openshift.common.is_containerized | bool and openshift.node.node_image not in docker_images.stdout - -- name: Wait for node image - command: > - docker images - register: docker_images - until: openshift.node.node_image in docker_images.stdout - retries: 30 - delay: 10 - changed_when: false when: openshift.common.is_containerized | bool - + - name: Pull OpenVSwitch image command: > docker pull {{ openshift.node.ovs_image }} - when: openshift.common.is_containerized | bool and openshift.node.ovs_image not in docker_images.stdout - and openshift.common.use_openshift_sdn | bool - -- name: Wait for OpenVSwitch image - command: > - docker images - register: docker_images - until: openshift.node.ovs_image in docker_images.stdout - retries: 30 - delay: 10 - changed_when: false when: openshift.common.is_containerized | bool and openshift.common.use_openshift_sdn | bool - name: Install Node docker service file @@ -130,6 +103,21 @@ - name: Additional storage plugin configuration include: storage_plugins/main.yml +# Necessary because when you're on a node that's also a master the master will be +# restarted after the node restarts docker and it will take up to 60 seconds for +# systemd to start the master again +- name: Wait for master API to become available before proceeding + # Using curl here since the uri module requires python-httplib2 and + # wait_for port doesn't provide health information. + command: > + curl -k --head --silent {{ openshift_node_master_api_url }} + register: api_available_output + until: api_available_output.stdout.find("200 OK") != -1 + retries: 120 + delay: 1 + changed_when: false + when: openshift.common.is_containerized | bool + - name: Start and enable node service: name={{ openshift.common.service_type }}-node enabled=yes state=started register: start_result diff --git a/roles/openshift_node/tasks/storage_plugins/nfs.yml b/roles/openshift_node/tasks/storage_plugins/nfs.yml index 1edf21d9b..14a613786 100644 --- a/roles/openshift_node/tasks/storage_plugins/nfs.yml +++ b/roles/openshift_node/tasks/storage_plugins/nfs.yml @@ -1,4 +1,8 @@ --- +- name: Install NFS storage plugin dependencies + action: "{{ ansible_pkg_mgr }} name=nfs-utils state=present" + when: not openshift.common.is_atomic | bool + - name: Set seboolean to allow nfs storage plugin access from containers seboolean: name: virt_use_nfs diff --git a/roles/openshift_node/templates/node.yaml.v1.j2 b/roles/openshift_node/templates/node.yaml.v1.j2 index 23bd81f91..44065f4bd 100644 --- a/roles/openshift_node/templates/node.yaml.v1.j2 +++ b/roles/openshift_node/templates/node.yaml.v1.j2 @@ -11,9 +11,7 @@ imageConfig: format: {{ openshift.node.registry_url }} latest: false kind: NodeConfig -{% if openshift.node.kubelet_args is defined and openshift.node.kubelet_args %} -kubeletArguments: {{ openshift.node.kubelet_args | to_json }} -{% endif %} +kubeletArguments: {{ openshift.node.kubelet_args | default(None) | to_padded_yaml(level=1) }} masterKubeConfig: system:node:{{ openshift.common.hostname }}.kubeconfig {% if openshift.common.use_openshift_sdn %} networkPluginName: {{ openshift.common.sdn_network_plugin_name }} @@ -22,7 +20,7 @@ networkPluginName: {{ openshift.common.sdn_network_plugin_name }} # deprecates networkPluginName above. The two should match. networkConfig: mtu: {{ openshift.node.sdn_mtu }} -{% if openshift.common.use_openshift_sdn %} +{% if openshift.common.use_openshift_sdn or openshift.common.use_nuage %} networkPluginName: {{ openshift.common.sdn_network_plugin_name }} {% endif %} {% if openshift.node.set_node_ip | bool %} diff --git a/roles/os_firewall/defaults/main.yml b/roles/os_firewall/defaults/main.yml index bcf1d9a34..e3176e611 100644 --- a/roles/os_firewall/defaults/main.yml +++ b/roles/os_firewall/defaults/main.yml @@ -1,2 +1,3 @@ --- +os_firewall_enabled: True os_firewall_use_firewalld: True diff --git a/roles/os_firewall/tasks/main.yml b/roles/os_firewall/tasks/main.yml index ad89ef97c..076e5e311 100644 --- a/roles/os_firewall/tasks/main.yml +++ b/roles/os_firewall/tasks/main.yml @@ -1,6 +1,6 @@ --- - include: firewall/firewalld.yml - when: os_firewall_use_firewalld + when: os_firewall_enabled | bool and os_firewall_use_firewalld | bool - include: firewall/iptables.yml - when: not os_firewall_use_firewalld + when: os_firewall_enabled | bool and not os_firewall_use_firewalld | bool diff --git a/roles/os_zabbix/tasks/main.yml b/roles/os_zabbix/tasks/main.yml index a8b65dd56..1c8d88854 100644 --- a/roles/os_zabbix/tasks/main.yml +++ b/roles/os_zabbix/tasks/main.yml @@ -1,8 +1,4 @@ --- -- fail: - msg: "Zabbix config is not yet supported on atomic hosts" - when: openshift.common.is_containerized | bool - - name: Main List all templates zbx_template: zbx_server: "{{ ozb_server }}" @@ -45,6 +41,10 @@ tags: - zagg_server +- include_vars: template_config_loop.yml + tags: + - config_loop + - name: Include Template Heartbeat include: ../../lib_zabbix/tasks/create_template.yml vars: @@ -154,3 +154,13 @@ password: "{{ ozb_password }}" tags: - zagg_server + +- name: Include Template Config Loop + include: ../../lib_zabbix/tasks/create_template.yml + vars: + template: "{{ g_template_config_loop }}" + server: "{{ ozb_server }}" + user: "{{ ozb_user }}" + password: "{{ ozb_password }}" + tags: + - config_loop diff --git a/roles/os_zabbix/vars/template_config_loop.yml b/roles/os_zabbix/vars/template_config_loop.yml new file mode 100644 index 000000000..823da1868 --- /dev/null +++ b/roles/os_zabbix/vars/template_config_loop.yml @@ -0,0 +1,14 @@ +--- +g_template_config_loop: + name: Template Config Loop + zitems: + - key: config_loop.run.exit_code + applications: + - Config Loop + value_type: int + + ztriggers: + - name: 'config_loop.run.exit_code not zero on {HOST.NAME}' + expression: '{Template Config Loop:config_loop.run.exit_code.min(#2)}>0' + url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_config_loop.asciidoc' + priority: average diff --git a/roles/os_zabbix/vars/template_docker.yml b/roles/os_zabbix/vars/template_docker.yml index a05e552e3..dd13e76f7 100644 --- a/roles/os_zabbix/vars/template_docker.yml +++ b/roles/os_zabbix/vars/template_docker.yml @@ -72,10 +72,12 @@ g_template_docker: url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_docker_ping.asciidoc' priority: high + # Re-enable for OpenShift 3.1.1 (https://bugzilla.redhat.com/show_bug.cgi?id=1292971#c6) - name: 'docker.container.dns.resolution failed on {HOST.NAME}' expression: '{Template Docker:docker.container.dns.resolution.min(#3)}>0' url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_docker_dns.asciidoc' priority: average + status: disabled - name: 'docker.container.existing.dns.resolution.failed on {HOST.NAME}' expression: '{Template Docker:docker.container.existing.dns.resolution.failed.min(#3)}>0' diff --git a/roles/os_zabbix/vars/template_openshift_master.yml b/roles/os_zabbix/vars/template_openshift_master.yml index a0ba8d104..12ea36c8b 100644 --- a/roles/os_zabbix/vars/template_openshift_master.yml +++ b/roles/os_zabbix/vars/template_openshift_master.yml @@ -98,6 +98,18 @@ g_template_openshift_master: applications: - Openshift Master + - key: openshift.master.skydns.port.open + description: State of the SkyDNS port open and listening + type: int + applications: + - Openshift Master + + - key: openshift.master.skydns.query + description: SkyDNS can be queried or not + type: int + applications: + - Openshift Master + - key: openshift.master.etcd.create.success description: Show number of successful create actions type: int @@ -305,6 +317,20 @@ g_template_openshift_master: - 'Openshift Master process not running on {HOST.NAME}' priority: high + - name: 'SkyDNS port not listening on {HOST.NAME}' + expression: '{Template Openshift Master:openshift.master.skydns.port.open.max(#3)}<1' + url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/openshift_master.asciidoc' + dependencies: + - 'Openshift Master process not running on {HOST.NAME}' + priority: high + + - name: 'SkyDNS query failed on {HOST.NAME}' + expression: '{Template Openshift Master:openshift.master.skydns.query.max(#3)}<1' + url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/openshift_master.asciidoc' + dependencies: + - 'Openshift Master API health check is failing on {HOST.NAME}' + priority: high + zgraphs: - name: Openshift Master API Server Latency Pods LIST Quantiles width: 900 diff --git a/roles/oso_host_monitoring/handlers/main.yml b/roles/oso_host_monitoring/handlers/main.yml index 7863ad15b..3a5d8024c 100644 --- a/roles/oso_host_monitoring/handlers/main.yml +++ b/roles/oso_host_monitoring/handlers/main.yml @@ -4,9 +4,3 @@ name: "{{ osohm_host_monitoring }}" state: restarted enabled: yes - -- name: "Restart the {{ osohm_zagg_client }} service" - service: - name: "{{ osohm_zagg_client }}" - state: restarted - enabled: yes diff --git a/roles/oso_host_monitoring/tasks/main.yml b/roles/oso_host_monitoring/tasks/main.yml index 6ddfa3dcb..a0a453416 100644 --- a/roles/oso_host_monitoring/tasks/main.yml +++ b/roles/oso_host_monitoring/tasks/main.yml @@ -5,7 +5,6 @@ with_items: - osohm_zagg_web_url - osohm_host_monitoring - - osohm_zagg_client - osohm_docker_registry_url - osohm_default_zagg_server_user - osohm_default_zagg_server_password @@ -37,29 +36,12 @@ - "Restart the {{ osohm_host_monitoring }} service" register: systemd_host_monitoring -- name: "Copy {{ osohm_zagg_client }} systemd file" - template: - src: "{{ osohm_zagg_client }}.service.j2" - dest: "/etc/systemd/system/{{ osohm_zagg_client }}.service" - owner: root - group: root - mode: 0644 - notify: - - "Restart the {{ osohm_zagg_client }} service" - register: zagg_systemd - - name: reload systemd command: /usr/bin/systemctl --system daemon-reload - when: systemd_host_monitoring | changed or zagg_systemd | changed + when: systemd_host_monitoring | changed - name: "Start the {{ osohm_host_monitoring }} service" service: name: "{{ osohm_host_monitoring }}" state: started enabled: yes - -- name: "Start the {{ osohm_zagg_client }} service" - service: - name: "{{ osohm_zagg_client }}" - state: started - enabled: yes diff --git a/roles/oso_host_monitoring/templates/oso-f22-host-monitoring.service.j2 b/roles/oso_host_monitoring/templates/oso-f22-host-monitoring.service.j2 deleted file mode 100644 index d18ad90fe..000000000 --- a/roles/oso_host_monitoring/templates/oso-f22-host-monitoring.service.j2 +++ /dev/null @@ -1,43 +0,0 @@ -# This is a systemd file to run this docker container under systemd. -# To make this work: -# * pull the image (probably from ops docker registry) -# * place this file in /etc/systemd/system without the .systemd extension -# * run the commands: -# systemctl daemon-reload -# systemctl enable pcp-docker -# systemctl start pcp-docker -# -# -[Unit] -Description=PCP Collector Contatainer -Requires=docker.service -After=docker.service - - -[Service] -Type=simple -TimeoutStartSec=5m -Environment=HOME=/etc/docker/ops -#Slice=container-small.slice - -# systemd syntax '=-' ignore errors from return codes. -ExecStartPre=-/usr/bin/docker kill "{{ osohm_host_monitoring }}" -ExecStartPre=-/usr/bin/docker rm "{{ osohm_host_monitoring }}" -ExecStartPre=-/usr/bin/docker pull "{{ osohm_docker_registry_url }}{{ osohm_host_monitoring }}" - - -ExecStart=/usr/bin/docker run --rm --name="{{ osohm_host_monitoring }}" \ - --privileged --net=host --pid=host --ipc=host \ - -v /sys:/sys:ro -v /etc/localtime:/etc/localtime:ro \ - -v /var/lib/docker:/var/lib/docker:ro -v /run:/run \ - -v /var/log:/var/log \ - {{ osohm_docker_registry_url }}{{ osohm_host_monitoring }} - -ExecReload=-/usr/bin/docker stop "{{ osohm_host_monitoring }}" -ExecReload=-/usr/bin/docker rm "{{ osohm_host_monitoring }}" -ExecStop=-/usr/bin/docker stop "{{ osohm_host_monitoring }}" -Restart=always -RestartSec=30 - -[Install] -WantedBy=default.target diff --git a/roles/oso_host_monitoring/templates/oso-rhel7-zagg-client.service.j2 b/roles/oso_host_monitoring/templates/oso-rhel7-host-monitoring.service.j2 index bcc8a5e03..ac950b4e5 100644 --- a/roles/oso_host_monitoring/templates/oso-rhel7-zagg-client.service.j2 +++ b/roles/oso_host_monitoring/templates/oso-rhel7-host-monitoring.service.j2 @@ -4,12 +4,12 @@ # * place this file in /etc/systemd/system without the .systemd extension # * run the commands: # systemctl daemon-reload -# systemctl enable zagg-client-docker -# systemctl start zagg-client-docker +# systemctl enable oso-rhel7-host-monitoring +# systemctl start oso-rhel7-host-monitoring # # [Unit] -Description=Zagg Client Contatainer +Description=Openshift Host Monitoring Container Requires=docker.service After=docker.service @@ -21,40 +21,54 @@ Environment=HOME=/etc/docker/ops #Slice=container-small.slice # systemd syntax '=-' ignore errors from return codes. -ExecStartPre=-/usr/bin/docker kill "{{ osohm_zagg_client }}" -ExecStartPre=-/usr/bin/docker rm "{{ osohm_zagg_client }}" -ExecStartPre=-/usr/bin/docker pull "{{ osohm_docker_registry_url }}{{ osohm_zagg_client }}" +ExecStartPre=-/usr/bin/docker kill "{{ osohm_host_monitoring }}" +ExecStartPre=-/usr/bin/docker rm "{{ osohm_host_monitoring }}" +ExecStartPre=-/usr/bin/docker pull "{{ osohm_docker_registry_url }}{{ osohm_host_monitoring }}" +# mwoodson note 1-7-16: +# pcp recommends mounting /run in their Dockerfile +# /run conflicts with cron which also runs in this container. +# I am leaving /run out for now. the guys in #pcp said that they mounted /run +# to shared the pcp socket that is created in /run. We are not using this, +# as far as I know. +# This problem goes away with systemd being run in the containers and not using +# cron but using systemd timers +# -v /run:/run \ -ExecStart=/usr/bin/docker run --name {{ osohm_zagg_client }} \ +ExecStart=/usr/bin/docker run --name {{ osohm_host_monitoring }} \ --privileged \ --pid=host \ --net=host \ - -e ZAGG_URL={{ osohm_zagg_web_url }} \ - -e ZAGG_USER={{ osohm_default_zagg_server_user }} \ - -e ZAGG_PASSWORD={{ osohm_default_zagg_server_password }} \ + --ipc=host \ + -e ZAGG_URL={{ osohm_zagg_web_url }} \ + -e ZAGG_USER={{ osohm_default_zagg_server_user }} \ + -e ZAGG_PASSWORD={{ osohm_default_zagg_server_password }} \ -e ZAGG_CLIENT_HOSTNAME={{ ec2_tag_Name }} \ - -e ZAGG_SSL_VERIFY={{ osohm_zagg_verify_ssl }} \ + -e ZAGG_SSL_VERIFY={{ osohm_zagg_verify_ssl }} \ -e OSO_CLUSTER_GROUP={{ cluster_group }} \ - -e OSO_CLUSTER_ID={{ oo_clusterid }} \ + -e OSO_CLUSTER_ID={{ oo_clusterid }} \ + -e OSO_ENVIRONMENT={{ oo_environment }} \ -e OSO_HOST_TYPE={{ hostvars[inventory_hostname]['ec2_tag_host-type'] }} \ -e OSO_SUB_HOST_TYPE={{ hostvars[inventory_hostname]['ec2_tag_sub-host-type'] }} \ + -e OSO_MASTER_HA={{ osohm_master_ha }} \ -v /etc/localtime:/etc/localtime \ - -v /run/pcp:/run/pcp \ + -v /sys:/sys:ro \ + -v /sys/fs/selinux \ + -v /var/lib/docker:/var/lib/docker:ro \ -v /var/run/docker.sock:/var/run/docker.sock \ - -v /var/run/openvswitch:/var/run/openvswitch \ + -v /var/run/openvswitch:/var/run/openvswitch \ {% if hostvars[inventory_hostname]['ec2_tag_host-type'] == 'master' %} -v /etc/openshift/master/admin.kubeconfig:/etc/openshift/master/admin.kubeconfig \ -v /etc/openshift/master/master.etcd-client.crt:/etc/openshift/master/master.etcd-client.crt \ -v /etc/openshift/master/master.etcd-client.key:/etc/openshift/master/master.etcd-client.key \ -v /etc/openshift/master/master-config.yaml:/etc/openshift/master/master-config.yaml \ {% endif %} - {{ osohm_docker_registry_url }}{{ osohm_zagg_client }} + {{ osohm_docker_registry_url }}{{ osohm_host_monitoring }} -ExecReload=-/usr/bin/docker stop "{{ osohm_zagg_client }}" -ExecReload=-/usr/bin/docker rm "{{ osohm_zagg_client }}" -ExecStop=-/usr/bin/docker stop "{{ osohm_zagg_client }}" +ExecReload=-/usr/bin/docker stop "{{ osohm_host_monitoring }}" +ExecReload=-/usr/bin/docker rm "{{ osohm_host_monitoring }}" +ExecStop=-/usr/bin/docker stop "{{ osohm_host_monitoring }}" Restart=always RestartSec=30 diff --git a/roles/oso_monitoring_tools/README.md b/roles/oso_monitoring_tools/README.md new file mode 100644 index 000000000..4215f9eeb --- /dev/null +++ b/roles/oso_monitoring_tools/README.md @@ -0,0 +1,54 @@ +Role Name +========= + +This role will install the Openshift Monitoring Utilities + +Requirements +------------ + +Any pre-requisites that may not be covered by Ansible itself or the role should be mentioned here. For instance, if the role uses the EC2 module, it may be a good idea to mention in this section that the boto package is required. + +Role Variables +-------------- + +osomt_zagg_client_config + +from vars/main.yml: + +osomt_zagg_client_config: + host: + name: "{{ osomt_host_name }}" + zagg: + url: "{{ osomt_zagg_url }}" + user: "{{ osomt_zagg_user }}" + pass: "{{ osomt_zagg_password }}" + ssl_verify: "{{ osomt_zagg_ssl_verify }}" + verbose: "{{ osomt_zagg_verbose }}" + debug: "{{ osomt_zagg_debug }}" + +Dependencies +------------ + +None + +Example Playbook +---------------- + +- role: "oso_monitoring_tools" + osomt_host_name: hostname + osomt_zagg_url: http://path.to/zagg_web + osomt_zagg_user: admin + osomt_zagg_password: password + osomt_zagg_ssl_verify: True + osomt_zagg_verbose: False + osomt_zagg_debug: False + +License +------- + +BSD + +Author Information +------------------ + +Openshift Operations diff --git a/roles/oso_monitoring_tools/defaults/main.yml b/roles/oso_monitoring_tools/defaults/main.yml new file mode 100644 index 000000000..a17424f25 --- /dev/null +++ b/roles/oso_monitoring_tools/defaults/main.yml @@ -0,0 +1,2 @@ +--- +# defaults file for oso_monitoring_tools diff --git a/roles/oso_monitoring_tools/handlers/main.yml b/roles/oso_monitoring_tools/handlers/main.yml new file mode 100644 index 000000000..cefa780ab --- /dev/null +++ b/roles/oso_monitoring_tools/handlers/main.yml @@ -0,0 +1,2 @@ +--- +# handlers file for oso_monitoring_tools diff --git a/roles/oso_monitoring_tools/meta/main.yml b/roles/oso_monitoring_tools/meta/main.yml new file mode 100644 index 000000000..9c42b68dc --- /dev/null +++ b/roles/oso_monitoring_tools/meta/main.yml @@ -0,0 +1,8 @@ +--- +galaxy_info: + author: OpenShift Operations + description: Install Openshift Monitoring tools + company: Red Hat, Inc + license: ASL 2.0 + min_ansible_version: 1.2 +dependencies: [] diff --git a/roles/oso_monitoring_tools/tasks/main.yml b/roles/oso_monitoring_tools/tasks/main.yml new file mode 100644 index 000000000..c90fc56e2 --- /dev/null +++ b/roles/oso_monitoring_tools/tasks/main.yml @@ -0,0 +1,18 @@ +--- +# tasks file for oso_monitoring_tools +- name: Install the Openshift Tools RPMS + yum: + name: "{{ item }}" + state: latest + with_items: + - openshift-tools-scripts-monitoring-zagg-client + - python-openshift-tools-monitoring-zagg + - python-openshift-tools-monitoring-zabbix + +- debug: var=g_zagg_client_config + +- name: Generate the /etc/openshift_tools/zagg_client.yaml config file + copy: + content: "{{ osomt_zagg_client_config | to_nice_yaml }}" + dest: /etc/openshift_tools/zagg_client.yaml + mode: "644" diff --git a/roles/oso_monitoring_tools/vars/main.yml b/roles/oso_monitoring_tools/vars/main.yml new file mode 100644 index 000000000..3538ba30b --- /dev/null +++ b/roles/oso_monitoring_tools/vars/main.yml @@ -0,0 +1,12 @@ +--- +# vars file for oso_monitoring_tools +osomt_zagg_client_config: + host: + name: "{{ osomt_host_name }}" + zagg: + url: "{{ osomt_zagg_url }}" + user: "{{ osomt_zagg_user }}" + pass: "{{ osomt_zagg_password }}" + ssl_verify: "{{ osomt_zagg_ssl_verify }}" + verbose: "{{ osomt_zagg_verbose }}" + debug: "{{ osomt_zagg_debug }}" diff --git a/roles/rhel_subscribe/tasks/enterprise.yml b/roles/rhel_subscribe/tasks/enterprise.yml index e9e6e4bd4..08540f440 100644 --- a/roles/rhel_subscribe/tasks/enterprise.yml +++ b/roles/rhel_subscribe/tasks/enterprise.yml @@ -2,8 +2,24 @@ - name: Disable all repositories command: subscription-manager repos --disable="*" +- set_fact: + default_ose_version: '3.0' + when: deployment_type == 'enterprise' + +- set_fact: + default_ose_version: '3.1' + when: deployment_type in ['atomic-enterprise', 'openshift-enterprise'] + +- set_fact: + ose_version: "{{ lookup('oo_option', 'ose_version') | default(default_ose_version, True) }}" + +- fail: + msg: "{{ ose_version }} is not a valid version for {{ deployment_type }} deployment type" + when: ( deployment_type == 'enterprise' and ose_version not in ['3.0'] ) or + ( deployment_type in ['atomic-enterprise', 'openshift-enterprise'] and ose_version not in ['3.1'] ) + - name: Enable RHEL repositories command: subscription-manager repos \ --enable="rhel-7-server-rpms" \ --enable="rhel-7-server-extras-rpms" \ - --enable="rhel-7-server-ose-3.0-rpms" + --enable="rhel-7-server-ose-{{ ose_version }}-rpms" diff --git a/roles/rhel_subscribe/tasks/main.yml b/roles/rhel_subscribe/tasks/main.yml index c160ea4e9..eecfd04a0 100644 --- a/roles/rhel_subscribe/tasks/main.yml +++ b/roles/rhel_subscribe/tasks/main.yml @@ -41,4 +41,4 @@ command: subscription-manager subscribe --pool {{ openshift_pool_id.stdout_lines[0] }} - include: enterprise.yml - when: deployment_type == 'enterprise' + when: deployment_type in [ 'enterprise', 'atomic-enterprise', 'openshift-enterprise' ] |