Change setup to my local centos7 VM

inv.yml

@@ -1,15 +1,12 @@
 all:
   hosts:
-    ed-c7-1:
+    centos7VM:
       ansible_user: root
-      ansible_host: ed-c7-1.virt.magni.thoto.net
+      ansible_host: localhost
+      ansible_port: 19322
       ssh_args: -o ControlMaster=auto -o ControlPersist=60s
-      # ansible_host: 192.168.122.139
+      unpriv_user: centos7 # gains privileges to use docker
-      unpriv_user: thoto
   children:
-    htcondor:
-      hosts:
-        ed-c7-1:
     slurm:
       hosts:
-        ed-c7-1:
+        centos7VM

play.yml

@@ -1,46 +1,15 @@
 ---
 - hosts: all
   tasks:
+  - name: "install epel-release"
+    yum:
+      name: [ epel-release ] # necessary to find htop => separate step
+      state: present
   - name: "install tools"
     yum:
-      name: [ vim-enhanced, htop, screen, bind-utils, nmap-ncat, net-tools ]
+      name: [ vim-enhanced, htop, screen, tmux, bind-utils, nmap-ncat, net-tools ]
       state: present
 
-- hosts: htcondor
-  pre_tasks:
-  - name: "install htcondor repo"
-    yum:
-      name: https://research.cs.wisc.edu/htcondor/repo/8.9/htcondor-release-current.el7.noarch.rpm
-      state: present
-    tags: htcondor
-
-  - name: "install htcondor software "
-    yum:
-      name: htcondor-ce
-      state: present
-    tags: htcondor
-
-  - name: "remove minicondor configuration"
-    yum:
-      name: minicondor
-      state: absent
-    tags: htcondor
-
-  - name: "setup singularity"
-    import_tasks: "singularity.yml"
-    tags: singularity
-
-  roles:
-  - name: "setup docker"
-    role: docker
-    tags: docker
-
-  - name: "setup htcondor test environment in docker containers"
-    role: docker-htcondor
-    tags:
-      - htcondor-containered
-      - htcondor
-
 - hosts: slurm
   vars:
     container_privileged: True
@@ -51,6 +20,4 @@
     tags: docker
   - name: "setup slurm test environment in docker containers"
     role: slurm
-    vars:
-      slurm_user: slurm # or root
     tags: slurm

roles/docker-htcondor/files/conf/cm/config.d/50-condor-centralmanager

@@ -1,3 +0,0 @@
-use ROLE:centralmanager
-# ALLOW_WRITE per default on * -> following has no effect yet
-ALLOW_WRITE_COLLECTOR=$(ALLOW_WRITE) condor-exec.htc.local condor-sub.htc.local

roles/docker-htcondor/files/conf/common/condor_config

@@ -1,92 +0,0 @@
-######################################################################
-##
-##  condor_config
-##
-##  This is the global configuration file for condor. This is where
-##  you define where the local config file is. Any settings
-##  made here may potentially be overridden in the local configuration
-##  file.  KEEP THAT IN MIND!  To double-check that a variable is
-##  getting set from the configuration file that you expect, use
-##  condor_config_val -v <variable name>
-##
-##  condor_config.annotated is a more detailed sample config file
-##
-##  Unless otherwise specified, settings that are commented out show
-##  the defaults that are used if you don't define a value.  Settings
-##  that are defined here MUST BE DEFINED since they have no default
-##  value.
-##
-######################################################################
-
-##  Where have you installed the bin, sbin and lib condor directories?
-RELEASE_DIR = /usr
-
-##  Where is the local condor directory for each host?  This is where the local config file(s), logs and
-##  spool/execute directories are located. this is the default for Linux and Unix systems.
-LOCAL_DIR = /var
-
-##  Where is the machine-specific local config file for each host?
-# LOCAL_CONFIG_FILE = /etc/condor/condor_config.local
-LOCAL_CONFIG_FILE = /etc/condor/condor_config_$(HOSTNAME).local
-##  If your configuration is on a shared file system, then this might be a better default
-#LOCAL_CONFIG_FILE = $(RELEASE_DIR)/etc/$(HOSTNAME).local
-##  If the local config file is not present, is it an error? (WARNING: This is a potential security issue.)
-REQUIRE_LOCAL_CONFIG_FILE = false
-
-##  The normal way to do configuration with RPMs is to read all of the
-##  files in a given directory that don't match a regex as configuration files.
-##  Config files are read in lexicographic order.
-LOCAL_CONFIG_DIR = /etc/condor/config.d
-#LOCAL_CONFIG_DIR_EXCLUDE_REGEXP = ^((\..*)|(.*~)|(#.*)|(.*\.rpmsave)|(.*\.rpmnew))$
-
-##
-## Do NOT use host-based security by default.
-##
-## This was the default for the 8.8 series (and earlier), but it is
-## intrinsically insecure.  To make the 9.0 series secure by default, we
-## commented it out.
-##
-## You should seriously consider improving your security configuration.
-##
-## To continue to use your old security configuration, knowing that it is
-## insecure, add the line 'use SECURITY : HOST_BASED' to your local
-## configuration directory.  Don't just uncomment the final line in this
-## comment block; changes in this file may be lost during your next upgrade.
-## The following shell command will make the change on most Linux systems.
-##
-## echo 'use SECURITY : HOST_BASED' >> $(condor_config_val LOCAL_CONFIG_DIR)/00-insecure.config
-##
-
-##  To expand your condor pool beyond a single host, set ALLOW_WRITE to match all of the hosts
-#ALLOW_WRITE = *.cs.wisc.edu
-##  FLOCK_FROM defines the machines that grant access to your pool via flocking. (i.e. these machines can join your pool).
-#FLOCK_FROM =
-##  FLOCK_TO defines the central managers that your schedd will advertise itself to (i.e. these pools will give matches to your schedd).
-#FLOCK_TO = condor.cs.wisc.edu, cm.example.edu
-
-##--------------------------------------------------------------------
-## Values set by the rpm patch script:
-##--------------------------------------------------------------------
-
-## For Unix machines, the path and file name of the file containing
-## the pool password for password authentication.
-#SEC_PASSWORD_FILE = $(LOCAL_DIR)/lib/condor/pool_password
-
-##  Pathnames
-RUN     = $(LOCAL_DIR)/run/condor
-LOG     = $(LOCAL_DIR)/log/condor
-LOCK    = $(LOCAL_DIR)/lock/condor
-SPOOL   = $(LOCAL_DIR)/lib/condor/spool
-EXECUTE = $(LOCAL_DIR)/lib/condor/execute
-BIN     = $(RELEASE_DIR)/bin
-LIB     = $(RELEASE_DIR)/lib64/condor
-INCLUDE = $(RELEASE_DIR)/include/condor
-SBIN    = $(RELEASE_DIR)/sbin
-LIBEXEC = $(RELEASE_DIR)/libexec/condor
-SHARE   = $(RELEASE_DIR)/share/condor
-
-PROCD_ADDRESS = $(RUN)/procd_pipe
-
-JAVA_CLASSPATH_DEFAULT = $(SHARE) .
-
-##  Install the minicondor package to run HTCondor on a single node

roles/docker-htcondor/files/conf/common/condor_ssh_to_job_sshd_config_template

@@ -1,64 +0,0 @@
-
-HostKey _INSERT_HOST_KEY_
-AuthorizedKeysFile _INSERT_AUTHORIZED_KEYS_FILE_
-
-# The following option is not supported by all recent versions of OpenSSH,
-# so instead we rely on injection of the shell setup command in the authorized
-# keys file.
-#ForceCommand _INSERT_FORCE_COMMAND_
-
-# as a convenience to users, allow remote setting of environment
-# since sshd is running as the job uid, there isn't really a security concern
-AcceptEnv *
-
-Subsystem       sftp    /usr/libexec/openssh/sftp-server
-
-# prevent sshd from freaking out about reading files from inside
-# a tmp-like directory
-StrictModes no
-
-# Rejection by tcp wrappers is not logged at INFO or VERBOSE log levels,
-# so to make diagnosis of problems easier, we use DEBUG.
-LogLevel DEBUG
-
-X11Forwarding yes
-
-# By default, OpenSSH uses the ipv6 loopback even if ipv6 is disabled in the
-# kernel.  This forces OpenSSH to use the "local" network instead
-X11UseLocalhost no
-
-# The following settings are recommended for good security.
-# In particular, only FIPS 140-2 algorithms are used.
-# URLs for extra information re FIPS security compliance:
-#    https://rhel7stig.readthedocs.io/en/latest/
-#    https://www.stigviewer.com/stig/red_hat_enterprise_linux_7/
-#    https://people.redhat.com/swells/scap-security-guide/tables/table-rhel7-stig-manual.html
-
-#-FIPS High Severity
-Protocol 2
-PermitEmptyPasswords no
-
-#-FIPS Medium Severity
-# Note: Ciphers and MACs below will be incompatible with RHEL5 or earlier.
-Ciphers aes128-ctr,aes192-ctr,aes256-ctr
-MACs hmac-sha2-256,hmac-sha2-512
-PermitRootLogin no
-PermitUserEnvironment no
-GSSAPIAuthentication no
-KerberosAuthentication no
-HostbasedAuthentication no
-IgnoreRhosts yes
-IgnoreUserKnownHosts yes
-PrintLastLog yes
-UsePrivilegeSeparation sandbox
-Compression delayed
-
-#-Recommended for security, but left out ssh_to_job config
-# because they provide minimal value and are likely to annoy
-# users or generate needless warnings in the ssh_to_job setting.
-#
-# ClientAliveInterval 600  # Note: condor_submit -i sets TMOUT
-# ClientAliveCountMax 0
-# banner=/etc/issue # Set to your warning banner
-# StrictModes yes # Can't set due to tmp-like directory
-# RhostsRSAAuthentication no # Obsolete Protocol version 1 option

roles/docker-htcondor/files/conf/common/config.d/00-htcondor-9.0.config

@@ -1,26 +0,0 @@
-##
-## Default security settings
-##
-## Host-based security was the default for the 8.8 series (and earlier).
-##
-## Host-based security assumes that all users on a machine are trusted.
-## For example, if host-based security trusts that a given machine can 
-## run jobs, then any user who can start a process on that machine can 
-## start a startd that can "steal" jobs from the system.
-#
-## To help make HTCondor secure by default, we removed host-based security
-## from the default configuration file
-## (/etc/condor/condor_config).
-##
-## New installations of HTCondor should be made using the get_htcondor tool,
-## which can automatically establish IDTOKENS-based security across a multi-
-## node pool.  For existing installations, we recommend you
-## consider improving your security configuration.
-##
-## To continue to use your old security configuration, 
-## comment out the 'recommended' line below, and uncomment the
-## 'host_based' line.
-##
-
-# use security : host_based
-use security : recommended_v9_0

roles/docker-htcondor/files/conf/common/config.d/20-common.config

@@ -1 +0,0 @@
-CONDOR_HOST = condor-cm.htc.local

roles/docker-htcondor/files/conf/exec/config.d/50-condor-exec

@@ -1 +0,0 @@
-use ROLE:execute

roles/docker-htcondor/files/conf/sub/config.d/50-condor-submit

@@ -1 +0,0 @@
-use ROLE:submit

roles/docker-htcondor/files/htcondor.Dockerfile

@@ -1,11 +0,0 @@
-FROM docker.io/library/centos:7
-
-RUN yum install -y https://research.cs.wisc.edu/htcondor/repo/8.9/htcondor-release-current.el7.noarch.rpm && \
-	yum install --nogpgcheck -y condor && \
-	yum install -y less && \
-	yum clean all
-
-RUN	yum install -y iproute bind-utils nmap-ncat net-tools && \
-	yum clean all
-
-CMD bash -c 'cat <({ condor_master -f & tail --retry --pid $! -f /var/log/condor/MasterLog & })'

roles/docker-htcondor/tasks/main.yml

@@ -1,142 +0,0 @@
-- name: "htcondor docker image"
-  file:
-    path: "/container/docker-images/htcondor"
-    state: directory
-    owner: "{{unpriv_user}}"
-    group: docker
-    mode: "u=rwx,g=rwx,o=rx"
-
-- copy:
-    dest: "/container/docker-images/htcondor/Dockerfile"
-    src: "htcondor.Dockerfile"
-    owner: "{{unpriv_user}}"
-    group: docker
-  register: cp_dockerfile
-
-- docker_image:
-    name: "htcondor"
-#    pull: False
-    build:
-      pull: False
-      path: "/container/docker-images/htcondor"
-    source: build
-    force_source: "{{cp_dockerfile.changed}}"
-
-- name: "copy htcondor container configuration"
-  copy:
-    src: "conf/{{item}}/"
-    dest: "/container/volumes/{{item}}/"
-    owner: "{{unpriv_user}}"
-    group: docker
-    mode: "u=rwx,g=rwx"
-  with_items: [ "cm", "exec", "sub", "common"]
-
-- name: "check if pool shared secret exists"
-  stat:
-    path: "/container/volumes/common/passwords.d/POOL"
-  register: pool_pw
-
-- block:
-  - name: "create temporary password store"
-    tempfile:
-      state: directory
-    register: pool_pw_tmp
-
-  - name: "generate pool password"
-    copy:
-      dest: "{{pool_pw_tmp.path}}/poolpw"
-      content: "{{lookup('password','/dev/null')}}"
-    no_log: True
-
-  - name: "install pool password"
-    docker_container:
-      name: "condor-common"
-      image: htcondor
-      state: started
-      volumes:
-      - "/container/volumes/common/:/etc/condor/:rw"
-      - "{{pool_pw_tmp.path}}:/tmp/poolpw:ro"
-      detach: False
-      cleanup: True
-      command: "condor_store_cred add -c -i /tmp/poolpw/poolpw"
-
-  - name: "remove tokens since pool password (cert) changed"
-    file:
-      path: "/container/volumes/{{item}}/tokens.d/condor@htc.local"
-      state: absent
-    with_items: [ "cm", "exec", "sub" ]
-
-  always:
-  - name: "remove temporary password store"
-    file:
-      path: "{{pool_pw_tmp.path}}"
-      state: absent
-    when: pool_pw_tmp is defined and pool_pw_tmp.path
-  when: not pool_pw.stat.exists
-
-- name: "sync common files to individual containers"
-  copy_2:
-    remote_src: True
-    force: True
-    directory_mode: preserve
-    mode: preserve
-    src: "/container/volumes/common/"
-    dest: "/container/volumes/{{item}}/"
-  with_items: [ "cm", "exec", "sub"]
-
-
-- name: "collect tokens to generate"
-  stat:
-    path: "/container/volumes/{{item}}/tokens.d/condor@htc.local"
-  with_items: [ "cm", "exec", "sub" ]
-  register: tokens_state
-
-- name: "generate tokens"
-  docker_container:
-    name: "condor-common"
-    image: htcondor
-    state: started
-    volumes:
-    - "/container/volumes/{{item}}/:/etc/condor/:rw"
-    detach: False
-    cleanup: True
-    command: "condor_token_create -identity condor@$(domainname) -token /etc/condor/tokens.d/condor@htc.local"
-  with_items: "{{tokens_state.results | rejectattr('stat.exists') | map(attribute='item') | list }}"
-
-
-- name: "create docker network to make service discovery work"
-  docker_network:
-    name: condor
-    state: present
-
-# TODO: reserve some address using docker_network_info and assign as aux
-# address to enable cm to get a static address in order to be reachable from
-# htcondor running on docker host to enable submitting jobs.
-
-- name: "run htcondor containers"
-  docker_container:
-    name: "condor-{{item}}"
-    hostname: "condor-{{item}}"
-    domainname: "htc.local"
-    image: htcondor
-    state: started
-    detach: True
-    cleanup: True
-    networks_cli_compatible: True
-    networks:
-    - name: "condor"
-      aliases: [ "condor-{{item}}.htc.local" ]
-    volumes:
-    - "/container/volumes/{{item}}/:/etc/condor/:rw"
-  with_items: [ "cm", "exec", "sub"]
-#    auto_remove: True
-#    mounts:
-#      src: /container/volumes/cm/
-#      dest: /etc/condor/
-
-
-#- add_host:
-#    hostname: foo
-#    ansible_connection: docker_api
-#    docker_host: ssh://ed-c7-1.virt.magni.thoto.net
-

roles/docker/tasks/main.yml

@@ -8,41 +8,6 @@
     name: [ "docker-ce", "python-docker-py" ] # latter for ansible modules
     state: present
 
-- name: "partition container image disk"
-  parted:
-    device: /dev/vdb
-    number: 1
-    state: present
-#    fs_type: xfs
-
-- filesystem:
-    dev: /dev/vdb1
-    fstype: xfs
-    opts: "-L image-store"
-
-- mount: 
-    path: "/container"
-    src: "/dev/vdb1"
-    fstype: xfs
-    opts: "noatime"
-    state: mounted
-
-- file:
-    path: "/container/docker"
-    state: directory
-    owner: root
-    group: root
-    mode: "u=rwx,g=x,o=x"
-
-- name: "link docker configuration to new container partition"
-  file:
-    path: "/var/lib/docker"
-    src: "/container/docker"
-    state: link
-    owner: root
-    group: root
-    mode: "u=rwx,g=x,o=x"
-    
 - name: "enable docker service in systemd"
   service:
     name: docker

roles/slurm/defaults/main.yml

@@ -1,5 +1 @@
 container_privileged: False
-slurm_user: slurm
-slurm_log_path_ctld: /var/log/slurm/slurmctld.log
-slurm_log_path_d: /var/log/slurm/slurmd.log
-slurm_log_path_sched: /var/log/slurm/slurmsched.log

roles/slurm/files/entrypoint.sh

@@ -1,9 +1,8 @@
 #!/usr/bin/env bash
 set -e
 
-if [ -f "/etc/munge/munge.key" ] ; then
+chown munge:munge /etc/munge/munge.key
-	chown munge:munge /etc/munge/munge.key
+
-	chmod 600 /etc/munge/munge.key
+chmod 600 /etc/munge/munge.key
-fi
 
 exec "$@"

roles/slurm/files/slurm.Dockerfile

@@ -2,42 +2,25 @@ FROM docker.io/library/centos:7 as base
 
 RUN yum install -y epel-release && \
 	yum install -y slurm && \
-	yum clean all && rm -rf /var/cache/yum
+	yum clean all
 
 RUN	yum install -y less iproute bind-utils nmap-ncat net-tools && \
-	yum clean all && rm -rf /var/cache/yum
+	yum clean all
 
 COPY entrypoint.sh /usr/local/sbin/entrypoint.sh
 
-RUN chown root:root /usr/local/sbin/entrypoint.sh && \
+RUN chown root:root /usr/local/sbin/entrypoint.sh && chmod 755 /usr/local/sbin/entrypoint.sh
-	chmod 755 /usr/local/sbin/entrypoint.sh
 
 ENTRYPOINT [ "/usr/local/sbin/entrypoint.sh" ]
 
-ARG slurmuser=slurm
-ENV slurmuser=${slurmuser}
-
-RUN useradd -d /var/lib/slurm -m --no-log-init --system $slurmuser &&\
-	slurm-setuser -u $slurmuser -g $slurmuser -y
-
-ENV SLURMCTLD_LOG_PATH="/var/log/slurm/slurmctld.log"
-ENV SLURMD_LOG_PATH="/var/log/slurm/slurmd.log"
-ENV SLURM_SCHED_LOG_PATH="/var/log/slurm/slurmsched.log"
-
 FROM base as slurmd
 
 RUN	yum install -y slurm-slurmd && \
-	yum clean all && rm -rf /var/cache/yum
+	yum clean all
-
-CMD bash -c 'cat <({ su -s /bin/sh -c "munged -F" munge & \
-	slurmd -D 2>/dev/null 1>/dev/null & \
-	tail --retry --pid $! -f ${SLURMD_LOG_PATH} ${SLURM_SCHED_LOG_PATH} & })'
 
 FROM base as slurmctld
 
 RUN	yum install -y slurm-slurmctld && \
-	yum clean all && rm -rf /var/cache/yum
+	yum clean all
 
-CMD bash -c 'cat <({ su -s /bin/sh -c "munged -F" munge & \
+# CMD bash -c 'cat <({ condor_master -f & tail --retry --pid $! -f /var/log/condor/MasterLog & })'
-	su -s /bin/sh -c "slurmctld -D" ${slurmuser} 2>/dev/null 1>/dev/null & \
-	tail --retry --pid $! -f ${SLURMCTLD_LOG_PATH} ${SLURM_SCHED_LOG_PATH} & })'

roles/slurm/files/slurmctld.Dockerfile

@@ -1,32 +1,20 @@
 FROM docker.io/library/centos:7 as base
 
 RUN yum install -y epel-release && \
-	yum install -y slurm && \
+	yum install -y slurm slurm-slurmctld && \
-	yum clean all && rm -rf /var/cache/yum
+	yum clean all
 
 RUN	yum install -y less iproute bind-utils nmap-ncat net-tools && \
-	yum clean all && rm -rf /var/cache/yum
+	yum clean all
+
+RUN	yum install -y slurm-slurmctld && \
+	yum clean all
 
 COPY entrypoint.sh /usr/local/sbin/entrypoint.sh
 
-RUN chown root:root /usr/local/sbin/entrypoint.sh && \
+RUN chown root:root /usr/local/sbin/entrypoint.sh && chmod 755 /usr/local/sbin/entrypoint.sh
-	chmod 755 /usr/local/sbin/entrypoint.sh
 
 ENTRYPOINT [ "/usr/local/sbin/entrypoint.sh" ]
 
-ARG slurmuser=slurm
+CMD bash -c 'cat <({ su -s /bin/sh -c "munged -F" munge & slurmctld -D & })'
-ENV slurmuser=${slurmuser}
+# ... & tail --retry --pid $! -f /var/log/condor/MasterLog & })'
-
-RUN useradd -d /var/lib/slurm -m --no-log-init --system $slurmuser &&\
-	slurm-setuser -u $slurmuser -g $slurmuser -y
-
-RUN	yum install -y slurm-slurmctld && \
-	yum clean all && rm -rf /var/cache/yum
-
-ENV SLURMCTLD_LOG_PATH="/var/log/slurm/slurmctld.log"
-ENV SLURMD_LOG_PATH="/var/log/slurm/slurmd.log"
-ENV SLURM_SCHED_LOG_PATH="/var/log/slurm/slurmsched.log"
-
-CMD bash -c 'cat <({ su -s /bin/sh -c "munged -F" munge & \
-	su -s /bin/sh -c "slurmctld -D" ${slurmuser} 2>/dev/null 1>/dev/null & \
-	tail --retry --pid $! -f ${SLURMCTLD_LOG_PATH} ${SLURM_SCHED_LOG_PATH} & })'

roles/slurm/files/slurmd.Dockerfile

@@ -1,32 +1,17 @@
 FROM docker.io/library/centos:7
 
 RUN yum install -y epel-release && \
-	yum install -y slurm && \
+	yum install -y slurm slurm-slurmd && \
-	yum clean all && rm -rf /var/cache/yum
+	yum clean all
 
 RUN	yum install -y less iproute bind-utils nmap-ncat net-tools && \
-	yum clean all && rm -rf /var/cache/yum
+	yum clean all
 
 COPY entrypoint.sh /usr/local/sbin/entrypoint.sh
 
-RUN chown root:root /usr/local/sbin/entrypoint.sh && \
+RUN chown root:root /usr/local/sbin/entrypoint.sh && chmod 755 /usr/local/sbin/entrypoint.sh
-	chmod 755 /usr/local/sbin/entrypoint.sh
 
 ENTRYPOINT [ "/usr/local/sbin/entrypoint.sh" ]
 
-ARG slurmuser=slurm
+# CMD bash -c 'cat <({ condor_master -f & tail --retry --pid $! -f /var/log/condor/MasterLog & })'
-ENV slurmuser=${slurmuser}
+CMD bash -c 'cat <({ su -s /bin/sh -c "munged -F" munge & slurmd -D & })'
-
-RUN useradd -d /var/lib/slurm -m --no-log-init --system $slurmuser &&\
-	slurm-setuser -u $slurmuser -g $slurmuser -y
-
-RUN yum install -y slurm-slurmd && \
-	yum clean all && rm -rf /var/cache/yum
-
-ENV SLURMCTLD_LOG_PATH="/var/log/slurm/slurmctld.log"
-ENV SLURMD_LOG_PATH="/var/log/slurm/slurmd.log"
-ENV SLURM_SCHED_LOG_PATH="/var/log/slurm/slurmsched.log"
-
-CMD bash -c 'cat <({ su -s /bin/sh -c "munged -F" munge & \
-	slurmd -D 2>/dev/null 1>/dev/null & \
-	tail --retry --pid $! -f ${SLURMD_LOG_PATH} ${SLURM_SCHED_LOG_PATH} & })'

roles/slurm/tasks/dockerimage.yml

@@ -1,19 +1,19 @@
 - file:
-    path: "/container/docker-images/{{item}}"
+    path: "/home/centos7/docker-images/{{item}}"
     state: directory
     owner: "{{unpriv_user}}"
     group: docker
 
 - copy:
     src: "{{item}}.Dockerfile"
-    dest: "/container/docker-images/{{item}}/Dockerfile"
+    dest: "/home/centos7/docker-images/{{item}}/Dockerfile"
     owner: "{{unpriv_user}}"
     group: docker
   register: slurm_cp_dockerfile
 
 - copy:
     src: "entrypoint.sh"
-    dest: "/container/docker-images/{{item}}/entrypoint.sh"
+    dest: "/home/centos7/docker-images/{{item}}/entrypoint.sh"
     owner: root
     group: root
     mode: u=rwx,g=rx,o=rx
@@ -24,7 +24,7 @@
 #    pull: False
     build:
       pull: False
-      path: "/container/docker-images/{{item}}"
+      path: "/home/centos7/docker-images/{{item}}"
 #      target: "{{item}}" # unsupported on old docker-py versions as in el7 
     source: build
     force_source: "{{slurm_cp_dockerfile.changed or slurm_cp_entrypt.changed}}"

roles/slurm/tasks/main.yml

@@ -21,7 +21,7 @@
     mode: u=rw,g=,o=
 
 - file:
-    path: /container/volumes/munge
+    path: /home/centos7/volumes/munge
     state: directory
     owner: munge
     group: munge
@@ -33,17 +33,17 @@
     force: true
     mode: preserve
     src: /etc/munge/munge.key
-    dest: /container/volumes/munge/munge.key
+    dest: /home/centos7/volumes/munge/munge.key
 
 - file:
-    path: /container/volumes/slurm/
+    path: /home/centos7/volumes/slurm/
     state: directory
 
 - name: upload slurm config
   template:
     force: true
     src: "{{item}}.j2"
-    dest: "/container/volumes/slurm/{{item}}"
+    dest: "/home/centos7/volumes/slurm/{{item}}"
   loop:
     - slurm.conf
     - cgroup.conf
@@ -82,8 +82,6 @@
     volumes: "{{default_mounts + ( item.extra_mounts | default([]) ) }}"
     networks:
     - name: "slurm"
-    env:
-      slurmuser: "{{slurm_user}}"
     image: "{{item.image}}"
     state: started
     detach: True
@@ -92,8 +90,8 @@
     networks_cli_compatible: True
   vars:
     default_mounts:
-    - /container/volumes/slurm/:/etc/slurm/:rw
+    - /home/centos7/volumes/slurm/:/etc/slurm/:rw
-    - /container/volumes/munge/munge.key:/etc/munge/munge.key:rw
+    - /home/centos7/volumes/munge/munge.key:/etc/munge/munge.key:rw
     - slurm-shared:/shared/:rw
     slurm_nodes_all: | # add execute nodes
       {% for i in range(1, 4) -%}

roles/slurm/templates/slurm.conf.j2

@@ -32,7 +32,6 @@ CryptoType=crypto/munge
 #MaxStepCount=40000
 #MaxTasksPerNode=128
 MpiDefault=pmix
-# when running slurmd as user change to: MpiDefault=none
 #MpiParams=ports=#-#
 #PluginDir=
 #PlugStackConfig=
@@ -58,9 +57,8 @@ SlurmctldPort=6817
 SlurmdPidFile=/var/run/slurm/slurmd.pid
 SlurmdPort=6818
 SlurmdSpoolDir=/var/spool/slurm/d
-SlurmUser={{slurm_user}}
+SlurmUser=root
-SlurmdUser=root
+#SlurmdUser=root
-# SlurmdUser=slurm -> sbatch does not work
 #SrunEpilog=
 #SrunProlog=
 StateSaveLocation=/var/spool/slurm/ctld
@@ -132,7 +130,7 @@ AccountingStorageType=accounting_storage/none
 #AccountingStorageUser=
 AccountingStoreJobComment=YES
 ClusterName=cluster
-#DebugFlags=Steps,TraceJobs
+#DebugFlags=
 #JobCompHost=
 JobCompLoc=/tmp/jobcomp
 #JobCompPass=
@@ -143,10 +141,10 @@ JobCompType=jobcomp/filetxt
 JobAcctGatherFrequency=30
 JobAcctGatherType=jobacct_gather/none
 SlurmctldDebug=verbose
-SlurmctldLogFile={{slurm_log_path_ctld}}
+#SlurmctldLogFile=
 SlurmdDebug=verbose
-SlurmdLogFile={{slurm_log_path_d}}
+#SlurmdLogFile=
-SlurmSchedLogFile={{slurm_log_path_sched}}
+#SlurmSchedLogFile=
 #SlurmSchedLogLevel=
 #
 #
@@ -165,4 +163,4 @@ SlurmSchedLogFile={{slurm_log_path_sched}}
 # COMPUTE NODES
 NodeName=slurm-exec[1-{{num_nodes}}] CPUs=2 CoresPerSocket=2 State=UNKNOWN
 NodeName=slurm-submit1 CPUs=1 State=UNKNOWN
-PartitionName=debug Nodes=slurm-exec[1-{{num_nodes}}] AllocNodes=slurm-submit1 Default=YES MaxTime=INFINITE State=UP
+PartitionName=debug Nodes=slurm-exec[1-{{num_nodes}}] Default=YES MaxTime=INFINITE State=UP