Log an NTP server for troubleshooting

The network model is probably like this.

The IP of the core layer two devices is 3.3.20.2 and 3.3.20.3

The IP of the server area switch is 3.3.20.252 and 3.3.20.253

The IP address of the NTP server is 3.3.100.103

Initial configuration:

Primary Core Switch

NTP Logging

NTP Master 2

NTP Server 3.3.100.103

Standby Core Switch

NTP Logging

NTP Server 3.3.20.2

Server area Switches

NTP Logging

NTP Server 3.3.20.2

The NTP server uses the Windows 2003 self-win32time service as the NTP service.

Problem phenomenon:

The server zone switch is capable of synchronizing clocks from the primary core switch, such as changing the clock of the primary core switch and changing the server area switch, but the primary core switch is not able to synchronize clocks from the NTP server.

Idea One:

Use show NTP associations detail to view the status of the NTP service synchronization, show 127.127.7.1 configured, Our_master, sane, valid, stratum 1, synchronization server,

3.3.100.103 configured, insane, invalid, stratum 1, cannot be synchronized.

The core switch is judged to be dominant, so it does not sync from 3.3.100.103.

workaround : No NTP master on the primary core switch, so that it synchronizes the clock only from the 3.3.100.103.

results : The problem is still, 3.3.100.103 configured, insane, invalid, stratum 1, can not be synchronized;

Idea two:

Check the Cisco official documentation that W32Time is using the SNTP service, and Cisco IOS uses the NTP service, which requires the installation of NTP services on Windows;

Original:

Windows W32Time shows that it's an SNTP implementation inside (rather claiming itself NTP). Cisco IOS-NTP, which tries to sync with W32Time, gets their own root-dispersion value that it sends to the W32Time and this proves costly for Cisco IOS-NTP to synchronize. Because the root-dispersion value of Cisco IOS-NTP goes higher than MS, it unsynchronizes itself (Clock-select proced URE). Since the Cisco IOS based routers run the full RFC implementation of NTP they does not sync to an SNTP server. In this case the output of the show NTP Associations Detail command shows that the server is flagged as insane, invalid. The root dispersion value is in excess of $ MS, which causes the Cisco IOS NTP implementation to reject the association . Routers that run Cisco IOS can being unable to synchronize to an NTP server if it's a Windows system that runs the W32Time s Ervice. If the server is not a synchronized, the routers is not a able to transmit to and receive packets from the server.

In order to workaround this issue and sync a Cisco IOS based router, use an authoritative NTP server on the Internet, a UN IX box that runs NTPD or a GPS on certain platforms. As an alternative, you can choose not to run the W32Time service on the Windows system. Instead, you can use NTP 4.x. All versions of Windows $ and later can serve as an NTP server. Other machines on the network can then use the NTP server to synchronize their time.

From

workaround : Download NTP 4.6, install it;

Under the installation directory, locate the ntp.conf file, modify the configuration

Server 127.127.1.0

# Local Clock host clocks are also treated as external clock sources, and the assigned address is 127.127.1.0

Fudge 127.127.1.0 Stratum 10

#设置本地时钟源的层次为10 so that if the NTPD service gets time from the local clock source, NTPD announces a time level of 11

#如果将其设置为0, the external claim is 1, the core switch configuration NTP Master 3, is not the priority to synchronize from the NTP server?

#明天验证!

Driftfile "C:/Program Files\ntp\etc\ntp.drift"

After that, restart the NTP service

After the configuration is correct, enter the bin directory and the ntpd-p has normal results to display

PS: I started not configured correctly, error no association ID ' s returned;

result : The core switch can synchronize the clock from the NTP server.

Other Knowledge :

The CISCO 4506E switch has a battery clock, and the command is the calendar set XX:XX:XX

Stratum represents the level of the NTP clock:

Top-level NTP server: refers to a level 0 NTP server, which provides a time synchronization service down (Stratum-1).

Middle-tier NTP server: Refers to Level 1, Level 2, etc. from the previous NTP server to obtain time, for the next level to provide time synchronization services.

NTP client: Refers to the time taken from the previous NTP server and does not provide a time synchronization service.

One host can obtain time information from multiple NTP servers; an NTP server can also provide time information to multiple hosts, and the same level of hosts can provide time information to each other. The NTP protocol supports up to 15 levels of clients.

The NTP protocol occupies port 123 and communicates using the UDP (User Datagram Protocol) protocol.

Log an NTP server for troubleshooting