Slurm及OpenLDAP部署

Slurm及OpenLDAP部署

概述

Slurm是一个开源、容错且高度可扩展性的集群管理和作业调度系统，用于大型和小型Linux集群。

Slurm提供三种关键功能：

  • 分配对资源的排他和/非排他访问
  • 提供一个用于在分配的节点集上启动、执行和监视作业的框架
  • 通过管理一个未完成作业队列来解决对资源的争用

一、构建拓扑结构

  • 搭建4台Linux服务器

  • 配置IP地址及主机名

    [root@localhost ~]# vi /etc/sysconfig/network-scripts/ifcfg-ens33
# 修改网卡参数，没有添加即可
IPADDR=192.168.100.100
NETMASK=255.255.255.0
GATEWAY=192.168.100.2
DNS1=114.114.114.114
ONBOOT=yes
BOOTPROTO=static

[root@localhost ~]# systemctl restart network
[root@localhost ~]# hostnamectl set-hostname slurm
[root@localhost ~]# bash
[root@slurm ~]#

#-------------------Node01----------------

[root@localhost ~]# vi /etc/sysconfig/network-scripts/ifcfg-ens33
# 修改网卡参数，没有添加即可
IPADDR=192.168.100.101
NETMASK=255.255.255.0
GATEWAY=192.168.100.2
DNS1=114.114.114.114
ONBOOT=yes
BOOTPROTO=static

[root@localhost ~]# systemctl restart network
[root@localhost ~]# hostnamectl set-hostname node01
[root@localhost ~]# bash
[root@node01 ~]#

#-------------------Node02----------------

[root@localhost ~]# vi /etc/sysconfig/network-scripts/ifcfg-ens33
# 修改网卡参数，没有添加即可
IPADDR=192.168.100.102
NETMASK=255.255.255.0
GATEWAY=192.168.100.2
DNS1=114.114.114.114
ONBOOT=yes
BOOTPROTO=static

[root@localhost ~]# systemctl restart network
[root@localhost ~]# hostnamectl set-hostname node02
[root@localhost ~]# bash
[root@node02 ~]#

#-------------------Node03----------------

[root@localhost ~]# vi /etc/sysconfig/network-scripts/ifcfg-ens33
# 修改网卡参数，没有添加即可
IPADDR=192.168.100.103
NETMASK=255.255.255.0
GATEWAY=192.168.100.2
DNS1=114.114.114.114
ONBOOT=yes
BOOTPROTO=static

[root@localhost ~]# systemctl restart network
[root@localhost ~]# hostnamectl set-hostname node03
[root@localhost ~]# bash
[root@node03 ~]#

  • 修改Hosts文件(4台服务器一致)

    [root@slurm ~]# vi /etc/hosts
# 新增
192.168.100.100 slurm
192.168.100.101 node01
192.168.100.102 node02
192.168.100.103 node03

# node01-03同样需要修改参照上面配置即可

二、安装Slurm应用

  • 上传slurm-EL7.5GUI-install.tar.gz文件至4台服务器的root目录下

    注：4台机器都需安装

    # 方法不限
E:\cmp>scp slurm19-EL7.5GUI-installer-1210.tar.gz root@192.168.100.100:/root
E:\cmp>scp slurm19-EL7.5GUI-installer-1210.tar.gz root@192.168.100.101:/root
E:\cmp>scp slurm19-EL7.5GUI-installer-1210.tar.gz root@192.168.100.102:/root
E:\cmp>scp slurm19-EL7.5GUI-installer-1210.tar.gz root@192.168.100.103:/root

  • 解压并执行

    [root@slurm ~]# yum -y install librrd* libhwloc*
[root@slurm ~]# tar zxvf slurm19-EL7.5GUI-installer-1210.tar.gz
[root@slurm ~]# cd slurm19-EL7.5GUI-installer
[root@slurm slurm19-EL7.5GUI-installer]# ./install_slurm.sh 
[root@slurm ~]# systemctl stop firewalld.service
[root@slurm ~]# systemctl disable firewalld.service

三、配置NTP时间同步

  • 修改配置文件将Slurm调度系统作为本地时间服务器

    [root@slurm etc]# yum -y install ntp
[root@slurm etc]# vi /etc/ntp.conf 
#server 0.centos.pool.ntp.org iburst
#server 1.centos.pool.ntp.org iburst
#server 2.centos.pool.ntp.org iburst
#server 3.centos.pool.ntp.org iburst
server 127.127.1.0
fudge 127.127.1.0 stratum 10

[root@slurm etc]# systemctl restart ntpd

  • 配置3台slurm计算节点的NTP

    [root@node01 etc]# vi /etc/ntp.conf
#server 0.centos.pool.ntp.org iburst
#server 1.centos.pool.ntp.org iburst
#server 2.centos.pool.ntp.org iburst
#server 3.centos.pool.ntp.org iburst
server slurm iburst

[root@node01 etc]# ntpdate -u slurm
 3 Jun 14:34:21 ntpdate[1944]: adjust time server 192.168.100.100 offset -0.034998 sec
[root@slurm etc]# systemctl restart ntpd
[root@node01 etc]# ntpq -p
     remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
*slurm           LOCAL(0)        11 u    2   64    1    0.237  -31.197   3.549
[root@node01 etc]# ntpstat 	# 显示如下即成功
synchronised to NTP server (192.168.100.100) at stratum 12
   time correct to within 983 ms
   polling server every 64 s
[root@node01 etc]#

四、修改Slurm配置文件

  • 所有Slurm集群结点均使用相同的配置文件

    [root@slurm ~]# cd slurm19-EL7.5GUI-installer/slurm-config/slurm.conf
# 将所有的ctl01修改为slurm
# 12 34 159
#
# 并修改文件最后的NODES和PARTITIONS

################################################
#                    NODES                     #
################################################
NodeName=node[01-10] CPUs=1 Boards=1 SocketsPerBoard=1 CoresPerSocket=1 ThreadsPerCore=1 RealMemory=972
#
################################################
#                  PARTITIONS                  #
################################################
PartitionName=computerPartiton Default=YES MinNodes=0 Nodes=cmp[001-101] State=UP
#
#  NODES的参数设置为主机名[节点数] 后面的参数请使用slumpd -C命令查看
[root@node01 slurm-config]# slurmd -C
NodeName=node01 CPUs=1 Boards=1 SocketsPerBoard=1 CoresPerSocket=1 ThreadsPerCore=1 RealMemory=972

  • 将slurm.conf文件复制到各个节点包括主节点

    [root@slurm ~]# cp /root/slurm19-EL7.5GUI-installer/slurm-config/slurm.conf /etc/slurm/
[root@slurm ~]# scp /root/slurm19-EL7.5GUI-installer/slurm-config/slurm.conf root@192.168.100.101:/etc/slurm/
[root@slurm ~]# scp /root/slurm19-EL7.5GUI-installer/slurm-config/slurm.conf root@192.168.100.102:/etc/slurm/
[root@slurm ~]# scp /root/slurm19-EL7.5GUI-installer/slurm-config/slurm.conf root@192.168.100.103:/etc/slurm/

  • 配置主控节点

    [root@slurm slurm19-EL7.5GUI-installer]# ./slurm_init_ctld.sh 
# 自动安装数据库并初始化
# user = root
# passwd = 123456a?

[root@slurm slurm19-EL7.5GUI-installer]# cd
[root@slurm ~]# systemctl restart slurmdbd
[root@slurm ~]# systemctl status slurmdbd
● slurmdbd.service - Slurm DBD accounting daemon
   Loaded: loaded (/usr/lib/systemd/system/slurmdbd.service; enabled; vendor preset: disabled)
   Active: active (running) since Thu 2021-06-03 17:00:20 CST; 6s ago
  Process: 13388 ExecStart=/usr/sbin/slurmdbd $SLURMDBD_OPTIONS (code=exited, status=0/SUCCESS)
   CGroup: /system.slice/slurmdbd.service
           └─13391 /usr/sbin/slurmdbd
[root@slurm ~]# systemctl enable slurmctld.service
[root@slurm ~]# systemctl start slurmctld.service

  • 配置4台计算节点

    [root@node01 slurm19-EL7.5GUI-installer]# ./cmp_slurm_init.sh 

Slurm computer node installed, configuration successfully!.

[root@node01 slurm19-EL7.5GUI-installer]# sytemctl restart slurmd.service
[root@node01 slurm19-EL7.5GUI-installer]# sytemctl enable slurmd.service

五、运行Slurm调度

[root@slurm ~]# sinfo
PARTITION         AVAIL  TIMELIMIT  NODES  STATE NODELIST
computerPartiton*    up   infinite      3   idle node[01-03]
[root@slurm ~]# srun -n 3 hostname
node01
node02
node03

六、部署OpenLDAP

安装前准备

# 关闭selinux、关闭防火墙
[root@slurm ~]# vi /etc/sysconfig/selinux
# This file controls the state of SELinux on the system.
# SELINUX= can take one of these three values:
#     enforcing - SELinux security policy is enforced.
#     permissive - SELinux prints warnings instead of enforcing.
#     disabled - No SELinux policy is loaded.
SELINUX=disabled
# SELINUXTYPE= can take one of three values:
#     targeted - Targeted processes are protected,
#     minimum - Modification of targeted policy. Only selected processes are protected.
#     mls - Multi Level Security protection.
SELINUXTYPE=targeted

[root@slurm ~]# systemctl stop NetworkManager
[root@slurm ~]# systemctl disable NetworkManager
Removed symlink /etc/systemd/system/multi-user.target.wants/NetworkManager.service.
Removed symlink /etc/systemd/system/dbus-org.freedesktop.nm-dispatcher.service.
Removed symlink /etc/systemd/system/network-online.target.wants/NetworkManager-wait-online.service.
[root@slurm ~]#

  • 安装OpenLDAP服务端（Slurm主控节点上安装）

    [root@slurm ~]# yum -y install openldap openldap-servers openldap-clients migrationtools
[root@slurm ~]# sed -i -e 's/olcSuffix:.*/olcSuffix: dc=xiaowangc,dc=com/g' /etc/openldap/slapd.d/cn\=config/olcDatabase\=\{2\}hdb.ldif
[root@slurm ~]# sed -i -e 's/olcRootDN:.*/olcRootDN: cn=admin,dc=xiaowangc,dc=com/g' /etc/openldap/slapd.d/cn\=config/olcDatabase\=\{2\}hdb.ldif
[root@slurm ~]# echo 'olcRootPw: 123456a?' >> /etc/openldap/slapd.d/cn\=config/olcDatabase\=\{2\}hdb.ldif
[root@slurm ~]# sed -i -e 's/dn.base="cn=.*"/dn.base="cn=admin,dc=xiaowangc,dc=com"/g' /etc/openldap/slapd.d/cn\=config/olcDatabase\=\{1\}monitor.ldif
[root@slurm ~]# cp /usr/share/openldap-servers/DB_CONFIG.example /var/lib/ldap/DB_CONFIG
[root@slurm ~]# chown -R ldap.ldap /var/lib/ldap

# 超级管理员为cn=admin,dc=xiaowangc,dc=com  密码为:123456a?
[root@slurm ~]# systemctl restart slapd
[root@slurm ~]# systemctl enable slapd
Created symlink from /etc/systemd/system/multi-user.target.wants/slapd.service to /usr/lib/systemd/system/slapd.service.
[root@slurm ~]# cd /etc/openldap/schema/
[root@slurm ~]# find . -name '*.ldif' -exec ldapadd -Y EXTERNAL -H ldapi:/// -D "cn=config" -f {} \;
[root@slurm ~]# cd /usr/share/migrationtools/
[root@slurm ~]# sed -i -e 's/"ou=Group"/"ou=Groups"/g' migrate_common.ph
[root@slurm ~]# sed -i -e 's/$DEFAULT_MAIL_DOMAIN = .*/$DEFAULT_MAIL_DOMAIN = "hpcce.com";/g' migrate_common.ph
[root@slurm ~]# sed -i -e 's/$DEFAULT_BASE = .*/$DEFAULT_BASE = "dc=xiaowangc,dc=com";/g' migrate_common.ph
[root@slurm ~]# sed -i -e 's/$EXTENDED_SCHEMA = 0;/$EXTENDED_SCHEMA = 1;/g' migrate_common.ph
./migrate_base.pl > /root/base.ldif
[root@slurm ~]# ldapadd -x -w 123456a? -D "cn=admin,dc=xiaowangc,dc=com" -f /root/base.ldif

  • 部署OpenLDAP客户端（3个计算节点）

    [root@node01 ~]# yum -y install nss-pam-ldapd
[root@node01 ~]# authconfig --enableldap --enableldapauth --ldapserver="ldap://192.168.100.100:389" --ldapbasedn=" dc=xiaowangc,dc=com " --update
[root@node01 ~]# authconfig --enablemkhomedir --update
[root@node01 ~]# authconfig --updateall


​```
#确保一下三个文件配置一致
​```

[root@node01 ~]# vi /etc/nsswitch.conf
# looked up first in the databases
#
# Example:
#passwd:    db files nisplus nis
#shadow:    db files nisplus nis
#group:     db files nisplus nis

passwd:     files sss ldap		# 修改这
shadow:     files sss ldap		# 修改这
group:      files sss ldap		# 修改这
#initgroups: files sss

#hosts:     db files nisplus nis dns
hosts:      files dns myhostname

# Example - obey only what nisplus tells us...
#services:   nisplus [NOTFOUND=return] files
#networks:   nisplus [NOTFOUND=return] files
#protocols:  nisplus [NOTFOUND=return] files
#rpc:        nisplus [NOTFOUND=return] files
#ethers:     nisplus [NOTFOUND=return] files
#netmasks:   nisplus [NOTFOUND=return] files

bootparams: nisplus [NOTFOUND=return] files

ethers:     files
netmasks:   files
networks:   files
protocols:  files
rpc:        files
services:   files sss

netgroup:   files sss ldap		# 修改这

publickey:  nisplus

automount:  files ldap		# 修改这
aliases:    files nisplus

[root@node01 ~]# vi /etc/pam.d/system-auth
#%PAM-1.0
# This file is auto-generated.
# User changes will be destroyed the next time authconfig is run.
auth        required      pam_env.so
auth        required      pam_faildelay.so delay=2000000
auth        sufficient    pam_unix.so nullok try_first_pass
auth        requisite     pam_succeed_if.so uid >= 1000 quiet_success
auth        sufficient    pam_ldap.so use_first_pass		# 修改这
auth        required      pam_deny.so

account     required      pam_unix.so broken_shadow
account     sufficient    pam_localuser.so
account     sufficient    pam_succeed_if.so uid < 1000 quiet
account     [default=bad success=ok user_unknown=ignore] pam_ldap.so		# 修改这
account     required      pam_permit.so

password    requisite     pam_pwquality.so try_first_pass local_users_only retry=3 authtok_type=
password    sufficient    pam_unix.so sha512 shadow nullok try_first_pass use_authtok
password    sufficient    pam_ldap.so use_authtok		# 修改这
password    required      pam_deny.so

session     optional      pam_keyinit.so revoke
session     required      pam_limits.so
-session     optional      pam_systemd.so
session     optional      pam_mkhomedir.so umask=0077		# 修改这
session     [success=1 default=ignore] pam_succeed_if.so service in crond quiet use_uid
session     required      pam_unix.so
session     optional      pam_ldap.so		# 修改这

[root@node01 ~]# vi /etc/pam.d/password-auth
#%PAM-1.0
# This file is auto-generated.
# User changes will be destroyed the next time authconfig is run.
auth        required      pam_env.so
auth        required      pam_faildelay.so delay=2000000
auth        sufficient    pam_unix.so nullok try_first_pass
auth        requisite     pam_succeed_if.so uid >= 1000 quiet_success
auth        sufficient    pam_ldap.so use_first_pass	# 修改这
auth        required      pam_deny.so

account     required      pam_unix.so broken_shadow
account     sufficient    pam_localuser.so
account     sufficient    pam_succeed_if.so uid < 1000 quiet
account     [default=bad success=ok user_unknown=ignore] pam_ldap.so		# 修改这
account     required      pam_permit.so

password    requisite     pam_pwquality.so try_first_pass local_users_only retry=3 authtok_type=
password    sufficient    pam_unix.so sha512 shadow nullok try_first_pass use_authtok
password    sufficient    pam_ldap.so use_authtok		# 修改这


password    required      pam_deny.so

session     optional      pam_keyinit.so revoke
session     required      pam_limits.so
-session     optional      pam_systemd.so
session     optional      pam_mkhomedir.so umask=0077		# 修改这
session     [success=1 default=ignore] pam_succeed_if.so service in crond quiet use_uid
session     required      pam_unix.so
session     optional      pam_ldap.so		# 修改这


[root@node01 ~]# systemctl restart nslcd
[root@node01 ~]# systemctl restart sshd
[root@node01 ~]# systemctl enable nslcd
[root@node01 ~]#

八、测试OpenLDAP

使用LDAP Admin软件测试

  • 随意在一个计算节点验证

    [root@node01 ~]# id testuser01
uid=44317(testuser01) gid=0(root) groups=0(root)
[root@node01 ~]# cat /etc/passwd | grep testuser01
[root@node01 ~]#

posted @ 2021-06-03 17:51  尤妤  阅读(1857)  评论(1)  编辑  收藏  举报