Linux – timedatectl & chronyd (systemd)

測試作業系統 CentOS 8 x86_64 (虛擬機)

目前 Linux 上的時間同步 NTP ( Network Time Protocol , 透過這個協定跟 NTP Server 的時間來同步) 是透過 chronyd 這個服務,下面來看一下用法.

timedatectl

先來看一下 timedatectl 這個指令,之前通常會使用 date 來修改時間,不過在 CentOS 7 / CentOS 8 使用 systemd 的系統,改使用 timedatectl 來設定與檢視日期時間 ( 該指令是由 systemd 套件提供).

[root@localhost ~]# rpm -qf /usr/bin/timedatectl
systemd-239-39.el8.x86_64

設定日期時間.

[root@localhost ~]# timedatectl set-time "2019-08-08 11:50:00"

設定時區 (TimeZone).

[root@localhost ~]# timedatectl list-timezones | grep -i taipei
Asia/Taipei

[root@localhost ~]# timedatectl set-timezone "Asia/Taipei"
[root@localhost ~]# timedatectl 
      Local time: Thu 2019-08-08 11:55:49 CST
  Universal time: Thu 2019-08-08 03:55:49 UTC
        RTC time: Thu 2019-08-08 03:55:49
       Time zone: Asia/Taipei (CST, +0800)
     NTP enabled: yes
NTP synchronized: yes
 RTC in local TZ: no
      DST active: n/a

Note: 改時區需把 rsyslog – https://benjr.tw/34103 日誌 log 服務重啟,避免時間與 log 紀錄的時間不相同.

[root@localhost ~]# systemctl restart rsyslog

chronyd (NTP Client)

如需與 NTP Server 做時間同步, 從 RHEL / CentOS 7 之後使用 Chrony 為預設校時服務 (NTP :Network Time Protocol ) ,預設應該都有安裝, chrony 套件如下.

[root@localhost ~]# yum install chrony

設定 NTP 可以透過 timedatectl 指令來啟動 NTP 的網路校時.

[root@localhost ~]# timedatectl set-ntp yes 
[root@localhost ~]# timedatectl 
      Local time: Thu 2019-08-08 11:54:53 CST
  Universal time: Thu 2019-08-08 03:54:53 UTC
        RTC time: Thu 2019-08-08 03:54:00
       Time zone: Asia/Taipei (CST, +0800)
     NTP enabled: yes
NTP synchronized: no
 RTC in local TZ: no
      DST active: n/a
[root@localhost ~]# systemctl status chronyd.service
● chronyd.service - NTP client/server
   Loaded: loaded (/usr/lib/systemd/system/chronyd.service; enabled; vendor preset: enabled)
   Active: active (running) since Tue 2024-05-28 10:57:40 CST; 12min ago
     Docs: man:chronyd(8)
           man:chrony.conf(5)
  Process: 3741 ExecStartPost=/usr/libexec/chrony-helper update-daemon (code=exited, status=0/SUCCESS)
  Process: 3709 ExecStart=/usr/sbin/chronyd $OPTIONS (code=exited, status=0/SUCCESS)
 Main PID: 3739 (chronyd)
    Tasks: 1 (limit: 22887)
   Memory: 924.0K
   CGroup: /system.slice/chronyd.service
           └─3739 /usr/sbin/chronyd

May 28 10:57:40 localhost.localdomain systemd[1]: Starting NTP client/server...
May 28 10:57:40 localhost.localdomain chronyd[3739]: chronyd version 4.5 starting (+CMDMON +NTP +REFCLOCK +RTC +PRIVDRO>
May 28 10:57:40 localhost.localdomain chronyd[3739]: Loaded 0 symmetric keys
May 28 10:57:40 localhost.localdomain chronyd[3739]: Frequency 0.000 +/- 1000000.000 ppm read from /var/lib/chrony/drift
May 28 10:57:40 localhost.localdomain chronyd[3739]: Using right/UTC timezone to obtain leap second data
May 28 10:57:40 localhost.localdomain systemd[1]: Started NTP client/server.
May 28 10:57:46 localhost.localdomain chronyd[3739]: Selected source 183.177.72.201 (2.centos.pool.ntp.org)
May 28 10:57:46 localhost.localdomain chronyd[3739]: System clock TAI offset set to 37 seconds

[root@localhost ~]# systemctl enable chronyd.service

當我們設定 NTP (Network Time Protocol) 為 Yes 時,會把 chronyd 這個 NTP 服務打開.

[root@localhost ~]# cat /var/log/messages
May 28 11:10:33 localhost systemd[1]: Starting NTP client/server...
May 28 11:10:33 localhost chronyd[3940]: chronyd version 4.5 starting (+CMDMON +NTP +REFCLOCK +RTC +PRIVDROP +SCFILTER +SIGND +ASYNCDNS +NTS +SECHASH +IPV6 +DEBUG)
May 28 11:10:33 localhost chronyd[3940]: Loaded 0 symmetric keys
May 28 11:10:33 localhost chronyd[3940]: Frequency -2.622 +/- 15.251 ppm read from /var/lib/chrony/drift
May 28 11:10:33 localhost chronyd[3940]: Using right/UTC timezone to obtain leap second data
May 28 11:10:33 localhost systemd[1]: Started NTP client/server.
May 28 11:10:41 localhost chronyd[3940]: Selected source 114.34.171.136 (2.centos.pool.ntp.org)
May 28 11:10:41 localhost chronyd[3940]: System clock TAI offset set to 37 seconds
May 28 11:10:42 localhost chronyd[3940]: Selected source 114.33.15.129 (2.centos.pool.ntp.org)
May 28 11:11:05 localhost systemd[1]: timedatex.service: Succeeded.
May 28 11:11:45 localhost chronyd[3940]: Selected source 118.163.81.61 (2.centos.pool.ntp.org)

或是透過啟動 chrony 服務來開啟 NTP.

[root@localhost ~]# timedatectl set-ntp no
[root@localhost ~]# timedatectl
               Local time: Tue 2024-05-28 11:10:18 CST
           Universal time: Tue 2024-05-28 03:10:18 UTC
                 RTC time: Tue 2024-05-28 03:10:18
                Time zone: Asia/Taipei (CST, +0800)
System clock synchronized: yes
              NTP service: inactive
          RTC in local TZ: no
[root@localhost ~]# systemctl status chronyd.service
● chronyd.service - NTP client/server
   Loaded: loaded (/usr/lib/systemd/system/chronyd.service; disabled; vendor preset: enabled)
   Active: inactive (dead)
     Docs: man:chronyd(8)
           man:chrony.conf(5)

May 28 10:57:40 localhost.localdomain chronyd[3739]: Loaded 0 symmetric keys
May 28 10:57:40 localhost.localdomain chronyd[3739]: Frequency 0.000 +/- 1000000.000 ppm read from /var/lib/chrony/drift
May 28 10:57:40 localhost.localdomain chronyd[3739]: Using right/UTC timezone to obtain leap second data
May 28 10:57:40 localhost.localdomain systemd[1]: Started NTP client/server.
May 28 10:57:46 localhost.localdomain chronyd[3739]: Selected source 183.177.72.201 (2.centos.pool.ntp.org)
May 28 10:57:46 localhost.localdomain chronyd[3739]: System clock TAI offset set to 37 seconds
May 28 11:10:16 localhost.localdomain systemd[1]: Stopping NTP client/server...
May 28 11:10:16 localhost.localdomain chronyd[3739]: chronyd exiting
May 28 11:10:16 localhost.localdomain systemd[1]: chronyd.service: Succeeded.
May 28 11:10:16 localhost.localdomain systemd[1]: Stopped NTP client/server.
[root@localhost ~]# systemctl start chronyd.service
[root@localhost ~]# timedatectl
               Local time: Tue 2024-05-28 11:10:35 CST
           Universal time: Tue 2024-05-28 03:10:35 UTC
                 RTC time: Tue 2024-05-28 03:10:35
                Time zone: Asia/Taipei (CST, +0800)
System clock synchronized: no
              NTP service: active
          RTC in local TZ: no

設定檔位於 /etc/chrony.conf 預設使用 pool 2.centos.pool.ntp.org ,在官方的說明 – https://www.ntppool.org/zh/use.html 其中 0, 1, 2 and 3.pool.ntp.org 的網域名稱解析會每一個小時更新,隨機指向一個離你最近的伺服器.

我們可以透過下面指令來檢視其 NTP 是哪一個.

tracking : The tracking command displays parameters about the system’s clock performance.

[root@localhost ~]# chronyc tracking
Reference ID    : B7B148CA (t2.time.tw1.yahoo.com)
Stratum         : 3
Ref time (UTC)  : Thu Mar 12 03:43:18 2020
System time     : 0.118893698 seconds slow of NTP time
Last offset     : +0.091888450 seconds
RMS offset      : 1.505703568 seconds
Frequency       : 100000.000 ppm fast
Residual freq   : -17442.777 ppm
Skew            : 1000000.000 ppm
Root delay      : 0.048257317 seconds
Root dispersion : 19.612304688 seconds
Update interval : 68.3 seconds
Leap status     : Normal

sources : This command displays information about the current time sources that chronyd is accessing.

[root@localhost ~]# chronyc sources
210 Number of sources = 4
MS Name/IP address         Stratum Poll Reach LastRx Last sample               
===============================================================================
^* t2.time.tw1.yahoo.com         2   6   377    40    +98ms[ +190ms] +/-   35ms
^- zimbra.alqualonde.org         3   6   377    40   +102ms[ +102ms] +/-  155ms
^- 103.122.188.3                 2   6   377    40    +99ms[  +99ms] +/-   51ms
^- 106-104-162-193.adsl.sta>     2   6   375    45    +95ms[  +95ms] +/-   84ms

chronyd (NTP Server)

chronyd Server (IP: 192.168.31.132)
當 chronyd.service 服務啟動後,其他電腦就可以把這一台當作 NTP(Network Time Protocol) 伺服器,需調整 chrony.conf 設定檔(須設定哪些網段可以使用).
```
[root@localhost ~]# vi /etc/chrony.conf
# Allow NTP client access from local network.
allow 192.168.0.0/16
[root@localhost ~]# systemctl restart chronyd.service
```
allow 這一行設定允許特定網段的使用者把你當做時間伺服器,如果只寫 allow 就是允許全部.

NTP Client (Internal chronyd)
要使用剛剛設定好的 NTP Server (chronyd IP: 192.168.31.132 ), Client 須把設定檔 /etc/chrony.conf 指定為這一台 IP.

[root@localhost ~]# vi /etc/chrony.conf
# Use public servers from the pool.ntp.org project.
# Please consider joining the pool (http://www.pool.ntp.org/join.html).
# pool 2.centos.pool.ntp.org iburst
server 192.168.31.132 iburst
[root@localhost ~]# systemctl restart chronyd

測試一下 NTP Server 是否正常.

[root@localhost ~]# chronyc activity
200 OK
1 sources online
0 sources offline
0 sources doing burst (return to online)
0 sources doing burst (return to offline)
0 sources with unknown address

測試時 Source state 必須為 ‘*’ ( current best )

[root@localhost ~]# chronyc sources -v

  .-- Source mode  '^' = server, '=' = peer, '#' = local clock.
 / .- Source state '*' = current best, '+' = combined, '-' = not combined,
| /             'x' = may be in error, '~' = too variable, '?' = unusable.
||                                                 .- xxxx [ yyyy ] +/- zzzz
||      Reachability register (octal) -.           |  xxxx = adjusted offset,
||      Log2(Polling interval) --.      |          |  yyyy = measured offset,
||                                \     |          |  zzzz = estimated error.
||                                 |    |           \
MS Name/IP address         Stratum Poll Reach LastRx Last sample
===============================================================================
^* 192.168.31.132            3   6   377    52   +132us[ +143us] +/-   36ms

手動校正時間.

[root@localhost ~]# chronyc -a makestep
200 OK

chronyd (Internal NTP Server)

如果你的 Chrony (NTP Server) 是在內部無法對外,這樣還能當 NTP Server 嗎? 找了一下資料 https://serverfault.com/questions/995101/how-to-synchronize-clocks-on-servers-with-no-internet-access 設定如下.

[root@localhost ~]# vi /etc/chrony.conf
# Allow NTP client access from local network.
allow 
# Record the rate at which the system clock gains/losses time.
driftfile /var/lib/chrony/drift
# Serve time even if not synchronized to a time source.
local stratum 10
manual
[root@localhost ~]# systemctl restart chronyd.service

說明:

driftfile /var/lib/chrony/drift
儲存本地端主機與上層 NTP Server 的誤差 (單位 : PPM – Parts Per Million , 1 PPM (Part Per Million) -> 1 microsecond per second -> 1*60*60/1000 (m) = 3.6ms per hour -> 3.6ms*24= 86.4ms per day)
local stratum 8
NTP是階層架構 (stratum) ,把服務設定在 local 階層架構 (stratum) 8 .
manual
?

遇過的問題

發現 NTP Client 要去連線自行架設的 NTP Server 狀態是 ‘?’ = unusable.

[root@localhost ~]# chronyc sources -v

  .-- Source mode  '^' = server, '=' = peer, '#' = local clock.
 / .- Source state '*' = current best, '+' = combined, '-' = not combined,
| /             'x' = may be in error, '~' = too variable, '?' = unusable.
||                                                 .- xxxx [ yyyy ] +/- zzzz
||      Reachability register (octal) -.           |  xxxx = adjusted offset,
||      Log2(Polling interval) --.      |          |  yyyy = measured offset,
||                                \     |          |  zzzz = estimated error.
||                                 |    |           \
MS Name/IP address         Stratum Poll Reach LastRx Last sample
===============================================================================
^? 192.168.31.129                0  10     0     -     +0ns[   +0ns] +/-    0ns

[root@localhost ~]# ntpstat
unsynchronised
poll interval unknown

我的 Server 都是架設在 VMware Workwork 的 VM , 有查到是可能跟 open-vm-tools enable time sync 有關 – https://serverfault.com/questions/1073299/ntp-chrony-not-keeping-time-synchronized-on-centos-7-9-vm-running-on-vmware-esx

後來是先把 Firewalld 與 SELinux 先關閉,再把 VM 與 Host 時間同步關閉,方式有2種.

指令 disable

[root@localhost ~]# vmware-toolbox-cmd timesync disable
Disabled
[root@localhost ~]# vmware-toolbox-cmd timesync status
Disabled

設定

這樣再連線就沒遇過問題了.

沒有解決問題,試試搜尋本站其他內容

timedatectl

chronyd (NTP Client)

chronyd (NTP Server)

chronyd (Internal NTP Server)

遇過的問題

發佈留言 取消回覆

發佈留言取消回覆