Forum Discussion

Nimbostratus

Feb 19, 2014

NTP not sync'ing - what interface is used?

Hi all,

I'm currently going through a complete reconfigure of our internal loadbalancers and at the moment wI'm stuck on getting NTP working.

Configuring of the internal/external/HA VLANs and assigning all self/float IP's etc is complete, Config sync and failover all work fine.

I've got a list of public NTP servers configured but they all sit at the state of INIT.

This is what I'm getting on all of the servers:

ind assID  status  conf reach auth  condition  last_event cnt
===========================================================
1   12174  8000    yes   yes  none    reject

Runnging ntpq -c "rv 12172" give:

[root@LB1:Active:In Sync] config  ntpq -c "rv 12174"
assID=12174 status=8000 unreach, conf, no events,
srcadr=ntppub.le.ac.uk, srcport=123, dstadr=192.168.74.1, dstport=123,
leap=11, stratum=16, precision=-20, rootdelay=0.000,
rootdispersion=0.000, refid=INIT, reach=000, unreach=37, hmode=3,
pmode=0, hpoll=10, ppoll=10, flash=00 ok, keyid=0, ttl=0, offset=0.000,
delay=0.000, dispersion=15937.500, jitter=0.000,
reftime=00000000.00000000  Thu, Feb  7 2036  6:28:16.000,
org=00000000.00000000  Thu, Feb  7 2036  6:28:16.000,
rec=00000000.00000000  Thu, Feb  7 2036  6:28:16.000,
xmt=d6af521e.ab559847  Wed, Feb 19 2014 15:54:06.669,
filtdelay=     0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00,
filtoffset=    0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00,
filtdisp=   16000.0 16000.0 16000.0 16000.0 16000.0 16000.0 16000.0 16000.0

The firewall all seems to be clear of any blocks on the IP adresses that have been configured, so what else could be the likely cause of something like this?

Thanks in advance for any advice.

Anthony

14 Replies

JPV_131616
Cirrus
Feb 21, 2014
remove the mgmt route to NTP server and create a host route in the GUI to ntp, then it will use def GW addr to get there, or whatever you force it to use.

thx
Anthony
Nimbostratus
Feb 28, 2014
Thanks for the reply. This is the first time I've had a chance to get back on here.

So if I remove the management-route (tmsh delete /sys management-route default) and then leave the external default gateway in place that should send traffic out that way? Or do I need to create a specific route to the NTP server address as the destination? This is the external default gateway I have set up at the moment:

[root@LB1:Active:Changes Pending] config tmsh show /net route ------------------------------------------------------------------------------------ Net::Routes Name Destination Type NextHop Origin ------------------------------------------------------------------------------------ external_default_gateway default gw 192.168.74.126 static

Many thanks,

Anthony
Anthony
Nimbostratus
Mar 06, 2014
Well I took the dive and removed the management route as explained above, and with a bit more tweaking to the firewall it all came together.

Thanks all for your assistance on this.

Anthony
mmd_2014_144687
Nimbostratus
Sep 09, 2014
Hi guys, I faced the same issue, my machines were not sending any ntp traffic via the management interface to the required servers. Crawling the askF5 solutions I found this SOL13284. To sum it up: it explains that outbound ntpd traffic is considered unsolicited! That means you have to explicitly set a host or net route to your ntp servers (in my case internal ntp servers) on the management interface. here is an example of how to do it (assuming the ntp servers are on the net 10.0.0.0/8):

tmsh create /sys management-route lognet network 10.0.0.0/255.0.0.0 gateway 172.1.1.254 tmsh save /sys config

hope that helps :) martin