ntpd sucks


Today we found this on a production VM

$ sudo ntpdate -q asia.pool.ntp.org
server 157.7.154.23, stratum 2, offset 58.792962, delay 0.11925
server 202.65.114.202, stratum 2, offset 58.827350, delay 0.15126
server 202.118.1.130, stratum 2, offset 58.816456, delay 0.08154
server 60.56.214.78, stratum 1, offset 58.801430, delay 0.09795
17 Apr 16:34:00 ntpdate[38518]: step time server 60.56.214.78 offset 58.801430 sec

$ sudo netstat  -putan |grep ntp
udp        0      0 127.0.0.1:123           0.0.0.0:*                           1161/ntpd
udp6       0      0 ::1:123                 :::*                                1161/ntpd

# ntpdc
ntpdc> peers
remote           local      st poll reach  delay   offset    disp
=======================================================================
=dns2.synet.edu. 0.0.0.0         16 1024    0 0.00000  0.000000 3.99217
=ts0.itsc.cuhk.e 0.0.0.0         16 1024    0 0.00000  0.000000 3.99217
=27.114.150.11   0.0.0.0         16 1024    0 0.00000  0.000000 3.99217
=dns1.synet.edu. 0.0.0.0         16 1024    0 0.00000  0.000000 3.99217

The thing lost internet connection and it won't reconnect, doh!

How to fix it:

$ sudo apt-get install openntpd

First thing Monday I will issue new images, damn.
Its hard to depend on services from the internet (not my decision).
In my previous jobs I always had private NTP servers on my network.