Chrony 报告系统时间比 NTP 慢且不断增长

Question

我的几个实例上运行着 chrony。不幸的是，其中一个实例不断报告系统时间比 NTP 时间慢。这个秒数以每分钟约 10 秒的速度不断增长。以下是其中一个读数：

Reference ID : D8E50445 (triton.ellipse.net)
Stratum : 3
Ref time (UTC) : Mon Dec 18 21:09:33 2017
System time : 10.736123085 seconds slow of NTP time
Last offset : +0.000084658 seconds
RMS offset : 0.001023541 seconds
Frequency : 217553.062 ppm slow
Residual freq : +0.017 ppm
Skew : 0.698 ppm
Root delay : 0.050225362 seconds
Root dispersion : 0.009489730 seconds
Update interval : 71.6 seconds
Leap status : Normal

几分钟后：

Reference ID : D8E50445 (triton.ellipse.net)
Stratum : 3
Ref time (UTC) : Mon Dec 18 21:17:55 2017
System time : 49.262691498 seconds slow of NTP time
Last offset : +0.000187908 seconds
RMS offset : 0.000752060 seconds
Frequency : 217553.109 ppm slow
Residual freq : +0.007 ppm
Skew : 0.306 ppm
Root delay : 0.050148461 seconds
Root dispersion : 0.016217658 seconds
Update interval : 71.3 seconds
Leap status : Normal

这是我的配置文件：

# use the local instance NTP service, if available
#server 169.254.169.123 prefer iburst
server 0.amazon.pool.ntp.org  prefer iburst
server 1.amazon.pool.ntp.org  prefer iburst
server 2.amazon.pool.ntp.org  prefer iburst
server 3.amazon.pool.ntp.org  prefer iburst

# Use public servers from the pool.ntp.org project.
# Please consider joining the pool (http://www.pool.ntp.org/join.html).
pool 2.amazon.pool.ntp.org iburst

# Record the rate at which the system clock gains/losses time.
driftfile /var/lib/chrony/drift

# Allow the system clock to be stepped in the first three updates
# if its offset is larger than 1 second.
#makestep 1.0 3
makestep 1000 10

# Enable hardware timestamping on all interfaces that support it.
#hwtimestamp *

# Increase the minimum number of selectable sources required to adjust
# the system clock.
minsources 2

# Allow NTP client access from local network.
#allow 192.168.0.0/16

# Serve time even if not synchronized to a time source.
#local stratum 10

# Specify file containing keys for NTP authentication.
#keyfile /etc/chrony.keys

## Get TAI-UTC offset and leap seconds from the system tz database.
#leapsectz right/UTC

# Specify directory for log files.
logdir /var/log/chrony

# Select which information is logged.
log measurements statistics tracking

# save data between restarts for fast re-load
dumponexit
dumpdir /var/run/chrony

我不知道该进一步研究什么。有什么想法吗？