Remember the clock sync problem with an egg ache

The company deployed three virtual machines to run CDH's kit, but suddenly CDH after a reboot. View Zookeeper log file, is connection refused, feel very momin.

Later viewed Slave2 's log found:

The host'S NTP Service is not a synchronized to any remote server.

It's a clock synchronization problem!

Run:

[[Email protected] log]# service ntpd stop[[email protected] log]# ntpdate slave1[[email protected] log]# service NTPD STA Rt

Solve the problem ...

This problem caused by the thinking, in the future in the management of this cluster, it is important to pay attention to clock synchronization problems.

Remember the clock sync problem with an egg ache