Programátoři Blog (nejen) o programování

jak rychle a jednoduse nastavit synchronizaci casu na strojich v ramci site oproti jednomu stroji ktery se synchronizuje pool.ntpd.org

zapis nevychazi z hlubsiho studia dokumentaco, slo jen o to nejak rychle neco zprovoznit a tudiz je nize popsane reseni zrejme ne prilis dobre.

cilem tohoto zapisu je na zaklade diskuze dojit k idealnimu reseni/konfiguraci.

mejme stroj A ktery bude synchronizovan se poolem pripojeny do lokalni site 10.0.0.0/24 ipadresou 10.0.0.1:

# cat /etc/ntp.conf
restrict default 
restrict 127.0.0.1
restrict 10.0.0.0 mask 255.255.255.0 nomodify notrap
server 1.sk.pool.ntp.org
server 3.europe.pool.ntp.org
server 2.europe.pool.ntp.org

# cat  /etc/ntp/step-tickers
1.sk.pool.ntp.org
3.europe.pool.ntp.org
2.europe.pool.ntp.org

po service ntpd restart a po nejakem case - cca pul hodiny a vic 0 ktery ntp server potrebuje k tomu aby si byl synchronizaci jisty se dostaneme do stavu:

# /usr/sbin/ntpq -pn
     remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
+85.248.115.102  192.36.143.150   2 u   73  256  377    0.495   -1.938   0.200
+217.147.223.78  129.132.2.21     3 u   74  256  377   16.302   -1.909   0.181
*217.172.189.31  91.66.35.100     2 u   74  256  377   24.853    7.246   1.143

timto by mel byt hlavni stroj vyresen.

na klientskych stanicich ze site 10.0.0.0/24 ktere se budou synchronizovat se strojem A je nasledujici konfigurace

# cat /etc/ntp.conf  | grep -vE '^#'
restrict default nomodify notrap noquery
restrict 127.0.0.1
server 10.0.0.1
server  127.127.1.0     # local clock
fudge   127.127.1.0 stratum 10
driftfile /var/lib/ntp/drift
broadcastdelay  0.008

# cat /etc/ntp/step-tickers
10.0.0.1
1.sk.pool.ntp.org

a opet po service ntpd restart se na klientech dostaneme do stavu:

# /usr/sbin/ntpq -pn
     remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
*10.0.0.1        217.172.189.31   3 u  430  512  377    0.321    1.546   0.672
 127.127.1.0     LOCAL(0)        10 l   45   64  377    0.000    0.000   0.001

nasledne pro kontrolu na hlavnim stroji 10.0.0.1 muzeme vypsat monitorovaci udaje:

# /usr/sbin/ntpdc -n -c monlist
remote address          port local address      count m ver code avgint  lstint
===============================================================================
127.0.0.1              39909 127.0.0.1             19 7 2      0   1805       0
10.0.0.200               123 10.0.0.1              59 3 4    180    733      24
10.0.0.198               123 10.0.0.1             229 3 4    180    290      43
10.0.0.197               123 10.0.0.1             229 3 4    180    417      95
10.0.0.194               123 10.0.0.1             158 3 4    180    231     149
217.172.189.31           123 217.123.27.181        210 4 4    180    305     200
85.248.115.102           123 217.123.27.181        207 4 4    180    313     200
217.147.223.78           123 217.123.27.181        208 4 4    180    313     201
10.0.0.199               123 10.0.0.1             151 3 4    180    330     478
10.0.0.2                 123 10.0.0.1             166 3 4    180    714     770