ansible/cheaha.yml → ansible/cluster.yml

@@ -3,8 +3,14 @@
   hosts: all
   become: true
   roles:
-    - { name: 'fix_centos_repo', tags: 'fix_centos_repo' }
     - { name: 'cheaha.node', tags: 'cheaha.node' }
-    - { name: 'nfs_mounts', tags: 'nfs_mounts' }
+    - { name: 'nfs_mounts', tags: 'nfs_mounts', when: enable_nfs_mounts }
     - { name: 'ldap_config', tags: 'ldap_config' }
     - { name: 'slurm_client', tags: 'slurm_client', when: enable_slurm_client }
+    - { name: 'ssh_host_keys', tags: 'ssh_host_keys' }
+    - { name: 'ssh_proxy_config', tags: 'ssh_proxy_config', when: enable_ssh_proxy_config }
+    - { name: 'ssl_cert', tags: 'ssl_cert', when: enable_ssl_certs }
+    - { name: 'rsyslog_config', tags: 'rsyslog_config', when: enable_rsyslog_config }
+    - { name: 'rewrite_map', tags: 'rewrite_map', when: enable_rewrite_map }
+    - { name: 'fail2ban', tags: 'fail2ban', when: enable_fail2ban }
+    - { name: 'install_node_exporter', tags: 'install_node_exporter', when: enable_node_exporter }

ansible/group_vars/all

@@ -17,14 +17,75 @@
 # cheaha.node related
   hostname_lookup_table:
     - "10.141.255.254 master.cm.cluster master localmaster.cm.cluster localmaster ldapserver.cm.cluster ldapserver"
+  domain_search_list:
+    - openstack.internal
+    - cm.cluster
+  nameserver_list:
+    - 10.141.255.254
 
 # ldap_config related
   ldap_cert_path: "/etc/openldap/certs"
   ldap_uri: "ldap://ldapserver"
 
 # nfs_mounts related
+  enable_nfs_mounts: true
   use_autofs: false
+  use_fstab: false
   mount_points:
-    - /gpfs4
-    - /gpfs5
+    - { "src": "master:/gpfs4", "path": "/gpfs4", "opts": "ro,sync,hard", "mode": "0755" }
+    - { "src": "master:/gpfs5", "path": "/gpfs5", "opts": "ro,sync,hard", "mode": "0755" }
+  autofs_mounts:
+    - { "src": "master:/gpfs4/&", "path": "/gpfs4", "opts": "fstype=nfs,vers=3,_netdev,default", "mode": '0755', "mount_point": "/gpfs4", "map_name": "gpfs4", key: "*" }
+    - { "src": "master:/gpfs5/&", "path": "/gpfs5", "opts": "fstype=nfs,vers=3,_netdev,default", "mode": '0755', "mount_point": "/gpfs5", "map_name": "gpfs5", key: "*" }
 
+#SSH Host Keys
+  S3_ENDPOINT: ""
+  SSH_HOST_KEYS_S3_BUCKET: ""
+  SSH_HOST_KEYS_S3_OBJECT: ""
+
+# AWS credentials
+  LTS_ACCESS_KEY: ""
+  LTS_SECRET_KEY: ""
+
+# ssh proxy
+  enable_ssh_proxy_config: false
+  sshpiper_dest_dir: "/opt/sshpiper"
+
+# rsyslog
+  enable_rsyslog_config: true
+  rsyslog_target: "*.* @master:514"
+
+# ssl certs
+  enable_ssl_certs: false
+  ssl_cert_s3_bucket: ""
+  ssl_cert_key_location: "/etc/pki/tls/private"
+  ssl_cert_file_location: "/etc/pki/tls/certs"
+  ssl_cert_key: ""
+  ssl_cert_file: ""
+  ssl_cert_chain_file: ""
+  ssl_apache_config: ""
+  apache_service: "httpd"
+
+# rewrite map
+  enable_rewrite_map: false
+  target_groups:
+    - {"name": "gpfs4", "host": "login001", "default": True }
+    - {"name": "gpfs5", "host": "login002", "default": False }
+
+# account app
+  account_app_port: 8000
+
+# fail2ban
+  enable_fail2ban: false
+  maxretry: 1
+  findtime: 600
+  bantime: 1200
+  fail2ban_white_list: "127.0.0.1/8"
+
+# Node Exporter
+  enable_node_exporter: false
+  node_exporter_ver: "1.8.2"
+  node_exporter_filename: "node_exporter-{{ node_exporter_ver }}.linux-amd64"
+  node_exporter_user: node_exporter
+  node_exporter_group: node_exporter
+  node_exporter_port: 9100

ansible/group_vars/prod

 ---
+  # cheaha.node related
   hostname_lookup_table:
     - "172.20.0.24 cheaha-master02.cm.cluster cheaha-master02"
     - "172.20.0.22 cheaha-master01.cm.cluster cheaha-master01"
     - "172.20.0.25 master.cm.cluster master localmaster.cm.cluster localmaster ldapserver.cm.cluster ldapserver"
+  domain_search_list:
+    - cm.cluster
+    - rc.uab.edu
+    - ib.cluster
+    - drac.cluster
+    - eth.cluster
+    - ib-hdr.cluster
+  nameserver_list:
+    - 172.20.0.25
 
   bright_openldap_path: "/cm/local/apps/openldap"
   ldap_cert_path: "{{bright_openldap_path}}/etc/certs"
   ldap_uri: "ldaps://ldapserver"
+
+  # proxy_config
+  target_groups:
+    - {"name": "gpfs5", "host": "login002", "default": False, "authorized_keys":"/gpfs5/data/user/home/$DOWNSTREAM_USER/.ssh/authorized_keys", "private_key":"/gpfs5/data/user/home/$DOWNSTREAM_USER/.ssh/id_ecdsa"}
+    - {"name": "gpfs4", "host": "login001", "default": True, "authorized_keys":"/gpfs4/data/user/home/$DOWNSTREAM_USER/.ssh/authorized_keys", "private_key":"/gpfs4/data/user/home/$DOWNSTREAM_USER/.ssh/id_ecdsa"}

ansible/compute.yml → ansible/login.yml

@@ -5,8 +5,5 @@
   roles:
     - { name: 'fix_centos_repo', tags: 'fix_centos_repo' }
     - { name: 'install_packages', tags: 'install_packages' }
-    - { name: 'pam_slurm_adopt', tags: 'pam_slurm_adopt' }
     - { name: 'install_nhc', tags: 'install_nhc'}
 
-- name: Setup node for use as a virtual cheaha node
-  ansible.builtin.import_playbook: cheaha.yml

ansible/ood.yml

@@ -6,6 +6,3 @@
     - { name: 'fix_centos_repo', tags: 'fix_centos_repo' }
     - { name: 'install_packages', tags: 'install_packages' }
     - { name: 'install_zsh', tags: 'install_zsh' }
-
-- name: Setup node for use as a virtual cheaha node
-  ansible.builtin.import_playbook: cheaha.yml

ansible/roles/cheaha.node/tasks/main.yml

@@ -11,6 +11,17 @@
     path: /etc/dhcp/dhclient.conf
     insertbefore: BOF
     line: 'append domain-name " cm.cluster rc.uab.edu ib.cluster drac.cluster eth.cluster ib-hdr.cluster";'
+    create: true
+    state: present
+
+- name: Template resolv.conf
+  ansible.builtin.template:
+    src: resolv.conf.j2
+    dest: /etc/resolv.conf
+    owner: root
+    group: root
+    mode: 0644
+    backup: true
 
 - name: Disable SELinux
   ansible.posix.selinux:

ansible/roles/cheaha.node/templates/resolv.conf.j2

+search {{ domain_search_list | join(' ') }}
+{% for name_server in nameserver_list %}
+nameserver {{ name_server }}
+{% endfor %}

ansible/roles/fail2ban/tasks/main.yml

+---
+
+- name: Install fail2ban
+  ansible.builtin.package:
+    name: "{{ item }}"
+    state: present
+  loop:
+    - fail2ban
+    - fail2ban-firewalld
+
+- name: Configure fail2ban
+  ansible.builtin.template:
+    src: "{{ item.src }}"
+    dest: "{{ item.dest }}"
+    backup: true
+  loop:
+    - { src: 'jail.local.j2', dest: '/etc/fail2ban/jail.local' }
+    - { src: 'sshpiperd_filter.local.j2', dest: '/etc/fail2ban/filter.d/sshpiperd.local' }
+    - { src: 'sshpiperd_jail.local.j2', dest: '/etc/fail2ban/jail.d/sshpiperd.local' }
+
+- name: Activate the firewalld support for fail2ban
+  ansible.builtin.command:
+    cmd: mv /etc/fail2ban/jail.d/00-firewalld.conf /etc/fail2ban/jail.d/00-firewalld.local
+
+- name: Configure firewalld to allow ssh and sshpiper traffic
+  ansible.posix.firewalld:
+    port: "{{ item }}"
+    zone: public
+    state: enabled
+    permanent: true
+  loop:
+    - 2222/tcp
+    - 22/tcp
+
+- name: Enable and start firewalld
+  ansible.builtin.service:
+    name: firewalld
+    enabled: true
+    state: restarted
+
+- name: Enable and start fail2ban
+  ansible.builtin.service:
+    name: fail2ban
+    enabled: true
+    state: restarted
+

ansible/roles/fail2ban/templates/jail.local.j2

+[DEFAULT]
+banaction = firewalld
+bantime  = {{ bantime }}
+ignoreip = {{ fail2ban_white_list }}
+
+[sshd]
+enabled = true

ansible/roles/fail2ban/templates/sshpiperd_filter.local.j2

+# Refer to https://github.com/fail2ban/fail2ban/wiki/Developing-Regex-in-Fail2ban for developing regex using fail2ban
+#
+[INCLUDES]
+before = common.conf
+
+[DEFAULT]
+_daemon = sshpiperd
+__iso_datetime = "\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}(?:[+-]\d{2}:\d{2}|Z)"
+__pref = time=%(__iso_datetime)s level=(?:debug|error)
+
+[Definition]
+# Define the prefix regex for the log lines
+prefregex = ^<F-MLFID>%(__prefix_line)s%(__pref)s</F-MLFID>\s+<F-CONTENT>.+</F-CONTENT>$
+
+# Failregex to match the specific failure log lines (prefregex is automatically included)
+failregex = ^msg="connection from .*failtoban: ip <HOST> too auth many failures"$
+
+ignoreregex =
+
+mode = normal
+
+maxlines = 1

ansible/roles/fail2ban/templates/sshpiperd_jail.local.j2

+# This configuration will block the remote host after {{maxretry}} failed SSH login attempts.
+[sshpiperd]
+enabled  = true
+filter   = sshpiperd
+logpath  = /var/log/messages
+port     = 22
+maxretry = {{ maxretry }}
+backend  = auto
+findtime = {{ findtime }}

ansible/roles/install_node_exporter/tasks/main.yaml

+---
+- name: Download node_exporter binary
+  ansible.builtin.get_url:
+    url: "https://github.com/prometheus/node_exporter/releases/download/v{{ node_exporter_ver }}/{{ node_exporter_filename }}.tar.gz"
+    dest: "/tmp/{{ node_exporter_filename }}.tar.gz"
+
+- name: Extract node_exporter
+  ansible.builtin.unarchive:
+    src: "/tmp/{{ node_exporter_filename }}.tar.gz"
+    dest: "/tmp"
+    remote_src: yes
+
+- name: Create system group for user account {{ node_exporter_group }}
+  ansible.builtin.group:
+    name: "{{ node_exporter_group }}"
+    system: true
+    state: present
+
+- name: Create system user account {{ node_exporter_user }}
+  ansible.builtin.user:
+    name: "{{ node_exporter_user }}"
+    comment: Prometheus node_exporter system account
+    group: "{{ node_exporter_group }}"
+    system: true
+    home: /var/lib/node_exporter
+    create_home: false
+    shell: /sbin/nologin
+    state: present
+
+- name: Copy node_exporter binary
+  ansible.builtin.copy:
+    src: "/tmp/{{ node_exporter_filename }}/node_exporter"
+    dest: /usr/local/bin/node_exporter
+    remote_src: yes
+    owner: root
+    group: root
+    mode: 0755
+
+- name: Copy systemd unit file
+  ansible.builtin.template:
+    src: node_exporter.service.j2
+    dest: /etc/systemd/system/node_exporter.service
+    owner: root
+    group: root
+    mode: '0644'
+
+- name: Clean up /tmp
+  ansible.builtin.file:
+    path: "/tmp/{{ item }}"
+    state: absent
+  loop:
+    - "{{ node_exporter_filename }}.tar.gz"
+    - "{{ node_exporter_filename }}"
+
+- name: Restart node_exporter service
+  ansible.builtin.systemd:
+    daemon_reload: yes
+    name: node_exporter
+    state: restarted
+    enabled: true
+
+- name: Collect facts about system services
+  ansible.builtin.service_facts:
+
+- name: Configure firewalld to allow prometheus
+  ansible.posix.firewalld:
+    port: "{{ node_exporter_port }}/tcp"
+    zone: public
+    state: enabled
+    permanent: true
+  when:
+    - "'firewalld.service' in ansible_facts.services"
+    - ansible_facts.services["firewalld.service"].state == "running"
+
+- name: Enable and start firewalld
+  ansible.builtin.service:
+    name: firewalld
+    enabled: true
+    state: restarted
+  when:
+    - "'firewalld.service' in ansible_facts.services"
+    - ansible_facts.services["firewalld.service"].state == "running"

ansible/roles/install_node_exporter/templates/node_exporter.service.j2

+[Unit]
+Description=Node Exporter
+After=network.target
+
+[Service]
+User={{ node_exporter_user }}
+Group={{ node_exporter_group }}
+Type=simple
+ExecStart=/usr/local/bin/node_exporter --web.listen-address=:{{ node_exporter_port }} --collector.filesystem.mount-points-exclude "^/(dev|proc|run/user/.+|run/credentials/.+|sys|var/lib/docker/.+)($|/)" --collector.filesystem.fs-types-exclude "^(autofs|binfmt_misc|bpf|cgroup|tmpfs|sunrpc|cgroup2?|configfs|debugfs|devpts|devtmpfs|fusectl|hugetlbfs|iso9660|mqueue|nsfs|overlay|proc|procfs|pstore|rpc_pipefs|securityfs|selinuxfs|squashfs|sysfs|tracefs)$"
+
+[Install]
+WantedBy=multi-user.target

ansible/roles/ldap_config/tasks/main.yml

@@ -7,7 +7,6 @@
     - nss-pam-ldapd
     - openldap
     - openldap-clients
-    - openldap-servers
     - sssd-ldap
 
 - name: Update nsswitch.conf to look for ldap

ansible/roles/nfs_mounts/tasks/autofs.yml

 ---
 - name: Create base directories
   ansible.builtin.file:
-    path: "{{ item.dir }}"
+    path: "{{ item.path }}"
     state: directory
     mode: "{{ item.mode }}"
   loop:
-    - { dir: /local, mode: '0777' }
-    - { dir: /scratch, mode: '0755' }
-    - { dir: /share, mode: '0755' }
-    - { dir: /data/rc/apps, mode: '0755' } # this is only required for the symlink to be happy
-    - { dir: /data/user, mode: '0755' }
-    - { dir: /data/project, mode: '0755' }
+    - { path: /local, mode: '0777' }
+    - { path: /share, mode: '0755' }
+
+- name: Create mountpoint dirs
+  ansible.builtin.file:
+    path: "{{ item.path }}"
+    state: directory
+    mode: "{{ item.mode }}"
+  loop:
+    "{{ autofs_mounts }}"
 
 - name: Remove unused entry in master map
   ansible.builtin.replace:
@@ -29,12 +33,7 @@
     line: "{{ item.mount_point }} /etc/auto.{{ item.map_name }}"
     create: yes
   loop:
-    - { mount_point: "/cm/shared", map_name: "cm-share" }
-    - { mount_point: "/data/project", map_name: "data-project" }
-    - { mount_point: "/data/user", map_name: "data-user" }
-    - { mount_point: "/data/rc/apps", map_name: "data-rc-apps" }
-    - { mount_point: "/-", map_name: "scratch" }
-    - { mount_point: "/home", map_name: "home" }
+    "{{ autofs_mounts }}"
 
 - name: Set up autofs map files
   ansible.builtin.lineinfile:
@@ -42,12 +41,7 @@
     line: "{{ item.key }} -{{ item.opts }} {{ item.src }}"
     create: true
   loop:
-    - { map_name: "cm-share", key: "*", src: "gpfs.rc.uab.edu:/data/cm/shared-8.2/&", opts: "fstype=nfs,vers=3,_netdev,defaults" }
-    - { map_name: "data-project", key: "*", src: "gpfs.rc.uab.edu:/data/project/&", opts: "fstype=nfs,vers=3,_netdev,defaults" }
-    - { map_name: "data-user", key: "*", src: "gpfs.rc.uab.edu:/data/user/&", opts: "fstype=nfs,vers=3,_netdev,local_lock=posix,defaults" }
-    - { map_name: "data-rc-apps", key: "*", src: "gpfs.rc.uab.edu:/data/rc/apps/&", opts: "fstype=nfs,vers=3,_netdev,defaults" }
-    - { map_name: "scratch", key: "/scratch", src: "gpfs.rc.uab.edu:/scratch", opts: "fstype=nfs,vers=3,_netdev,local_lock=posix,defaults" }
-    - { map_name: "home", key: "*", src: ":/data/user/home/&", opts: 'fstype=bind' }
+    "{{ autofs_mounts }}"
 
 - name: Create symbolic links
   ansible.builtin.file:
@@ -60,7 +54,8 @@
   loop:
     - { src: /data/rc/apps, dest: /share/apps }
 
-- name: Enable autofs service
+- name: Enable and start autofs service
   ansible.builtin.service:
     name: autofs
     enabled: true
+    state: restarted

ansible/roles/nfs_mounts/tasks/fstab.yml

 ---
 - name: Create base directories
   ansible.builtin.file:
-    path: "{{ item }}"
+    path: "{{ item.path }}"
     state: directory
-    mode: '0755'
+    mode: "{{ item.mode }}"
   loop:
     "{{ mount_points }}"
 
-- name: Make an entry in the fstab
+- name: Mount the directories
   ansible.posix.mount:
-    src: "master:{{ item }}"
-    path: "{{ item }}"
-    opts: rw,sync,hard
-    state: present
+    src: "{{ item.src }}"
+    path: "{{ item.path }}"
+    opts: "{{ item.opts }}"
+    state: mounted
     fstype: nfs
   loop:
     "{{ mount_points }}"

ansible/roles/nfs_mounts/tasks/main.yml

 ---
 - name: nfs_mounts using fstab
   include_tasks: fstab.yml
-  when: not use_autofs
+  when: use_fstab
 
 - name: nfs_mounts using autofs
   include_tasks: autofs.yml

ansible/roles/rewrite_map/tasks/main.yaml

+---
+- name: Add apache rewritemap script config
+  ansible.builtin.template:
+    src: rewrite_map_config_py.j2
+    mode: '600'
+    owner: root
+    group: root
+    dest: /var/www/rewrite_map_config.py
+
+- name: Replace OOD rewrite condition regex in Apache configuration
+  ansible.builtin.replace:
+    path: /etc/httpd/conf.d/front-end.conf
+    regexp: "RewriteCond %{HTTP:REMOTE_USER} '\\^\\(\\.\\+\\)\\$'"
+    replace: |
+      RewriteCond %{HTTP:REMOTE_USER} '([a-zA-Z0-9_.+-]+)@uab.edu$' [OR]
+          RewriteCond %{HTTP:REMOTE_USER} 'urn:mace:incommon:uab.edu!https://uabgrid.uab.edu/shibboleth!(.+)$'
+
+- name: Replace account app port in Apache configuration
+  ansible.builtin.replace:
+    path: /etc/httpd/conf.d/front-end.conf
+    regexp: "account-app:8000"
+    replace: "account-app:{{ account_app_port }}"
+
+- name: Restart httpd services
+  ansible.builtin.service:
+    name: httpd
+    enabled: true
+    state: restarted

ansible/roles/rewrite_map/templates/rewrite_map_config_py.j2

+DEBUG = False
+target_groups = {
+    {% for group in target_groups %}
+    "{{ group.name }}": "{{ group.host }}",
+    {% endfor %}
+}
+{% for group in target_groups %}
+{% if group.default %}
+default_hostname = "{{ group.host }}"
+{% endif %}
+{% endfor %}