Controllerless Networks

 View Only
  • 1.  NTP problem

    Posted Feb 12, 2022 10:41 AM
    Edited by pos42 Feb 12, 2022 11:45 AM
    Hi

    I have a working NTP server set in the instant web ui. Not an IP, but a fqdn hostname.

    I could see that time on one of my APs was off and had drifted in time. Now.... If running "show ntp status" in all 4 APs, only two show an NTP server. Running firmware 8.6.0.16.

    AP1 225
    AP2 225
    AP3 535 set as preferred master and holds the virtual controller
    AP4 535 

    The two newer 535 AP:s show the ntp server When running "show ntp status", but the older 225:s don't. Is this a bug or am I missing something here?  But when running "show ntp debug" it seems though that the "ntpdate" command is run on all 4 AP:s.

    Is it maybe the case that older APs just use ntpdate do set time, and newer APs also enables a local own ntpd ?



    Tnx in advance

    ------------------------------
    Per-Olov Sj�holm
    ------------------------------


  • 2.  RE: NTP problem

    Posted Feb 12, 2022 12:30 PM
    Have you tried using an ip address?  Are you using IAP-VPN?

    ------------------------------
    Any opinions expressed here are solely my own and not necessarily that of Hewlett Packard Enterprise or Aruba Networks.
    ------------------------------



  • 3.  RE: NTP problem

    Posted Feb 12, 2022 01:00 PM
    Edited by pos42 Feb 12, 2022 01:03 PM
    Hi

    I just tried using the ntp server IP instead of fqdn. Behaves the same. I use no VPN.

    As said, it seems to sync time in all 4 APs by ntp date. I can see that with the "show ntp debug" command. But nothing if checking with "show ntp status" in the old 225:s.


    An old 225 in the cluster
    ---snip--

    ap2.incedo.org (upstairs)# show clock 

    Current Time     :2022-02-12 18:55:19

    ap2.incedo.org (upstairs)# show ntp status 

    * master (synced), # master (unsynced), + selected, - candidate, x candidate, ~ configured

    ap2.incedo.org (upstairs)# 

    ap2.incedo.org (upstairs)# show ntp debug  

    12 Feb 18:52:01 ntpdate[14350]: ntpdate 4.2.8p9@1.3265 Tue Sep  5 07:19:17 UTC 2017 (4)

    Looking for host ntp.incedo.org and service ntp

    2001:470:dc5d:1::1 reversed to fw-lan.incedo.org

    host found : fw-lan.incedo.org

    transmit(192.168.0.1)

    receive(192.168.0.1)

    transmit(192.168.0.1)

    receive(192.168.0.1)

    transmit(192.168.0.1)

    receive(192.168.0.1)

    transmit(192.168.0.1)

    receive(192.168.0.1)

    server 192.168.0.1, port 123

    stratum 2, precision -24, leap 00, trust 000

    refid [192.168.0.1], delay 0.02641, dispersion 0.00000

    transmitted 4, in filter 4

    reference time:    e5b271e0.23b47fcf  Sat, Feb 12 2022 18:50:24.139

    originate timestamp: e5b27247.379990e9  Sat, Feb 12 2022 18:52:07.217

    transmit timestamp:  e5b27247.2a6687f4  Sat, Feb 12 2022 18:52:07.165

    filter delay:  0.02649  0.02641  0.02644  0.02646 

             0.00000  0.00000  0.00000  0.00000 

    filter offset: 0.051036 0.051017 0.051035 0.051048

             0.000000 0.000000 0.000000 0.000000

    delay 0.02641, dispersion 0.00000

    offset 0.051017

    12 Feb 18:52:07 ntpdate[14350]: adjust time server 192.168.0.1 offset 0.051017 sec

    ap2.incedo.org (upstairs)# 

    --snip--




    A new 535 in the cluster
    ---snip--

    ap3.incedo.org (hallway)# show clock 

    Current Time     :2022-02-12 18:55:20

    ap3.incedo.org (hallway)# show ntp status 

         remote           refid      st t when poll reach   delay   offset  jitter

    ==============================================================================

     fw-lan.incedo.o .INIT.          16 u    -   64    0    0.000    0.000   0.000

    * master (synced), # master (unsynced), + selected, - candidate, x candidate, ~ configured

    ap3.incedo.org (hallway)# 

    ap3.incedo.org (hallway)# show ntp debug

    12 Feb 18:55:58 ntpdate[6597]: ntpdate 4.2.8p9@1.3265 Thu Mar  1 07:22:25 UTC 2018 (1)

    Looking for host ntp.incedo.org and service ntp

    2001:470:dc5d:1::1 reversed to fw-lan.incedo.org

    host found : fw-lan.incedo.org

    transmit(192.168.0.1)

    receive(192.168.0.1)

    transmit(192.168.0.1)

    receive(192.168.0.1)

    transmit(192.168.0.1)

    receive(192.168.0.1)

    transmit(192.168.0.1)

    receive(192.168.0.1)

    server 192.168.0.1, port 123

    stratum 2, precision -24, leap 00, trust 000

    refid [192.168.0.1], delay 0.02603, dispersion 0.00006

    transmitted 4, in filter 4

    reference time:    e5b271e0.23b47fcf  Sat, Feb 12 2022 18:50:24.139

    originate timestamp: e5b27334.4c57a0a5  Sat, Feb 12 2022 18:56:04.298

    transmit timestamp:  e5b27334.460d971c  Sat, Feb 12 2022 18:56:04.273

    filter delay:  0.02609  0.02615  0.02603  0.02609 

             0.00000  0.00000  0.00000  0.00000 

    filter offset: 0.024448 0.024341 0.024387 0.024275

             0.000000 0.000000 0.000000 0.000000

    delay 0.02603, dispersion 0.00006

    offset 0.024387

    12 Feb 18:56:04 ntpdate[6597]: adjust time server 192.168.0.1 offset 0.024387 sec

    ap3.incedo.org (hallway)# 

    --snip--

    ------------------------------
    Per-Olov Sj�holm
    ------------------------------



  • 4.  RE: NTP problem

    Posted Feb 13, 2022 01:30 AM
    ensure you have configured Time Zone. Here is the output of my NTP status

    AP-Lab# sh run | include ntp
    ntp-server au.pool.ntp.org


    AP-Lab# sh ntp status
    remote refid st t when poll reach delay offset jitter
    ==============================================================================
    *time.cloudflare 10.47.8.8 3 u 192 512 377 8.297 2.161 0.514

    * master (synced), # master (unsynced), + selected, - candidate, x candidate, ~ configured
    AP-Lab#



    ------------------------------
    Any opinions expressed here are solely my own and not necessarily that of Hewlett Packard Enterprise or Aruba.
    ------------------------------



  • 5.  RE: NTP problem

    Posted Feb 14, 2022 03:17 AM
    Hi

    Show ntp status won't work (won't show anything) on the older 225:s. Only on the 535:s. 8.6.0.16 and all APs in the same cluster. I can note though that even on the 535:s, it says stratum 16 and I see no jitter data or so. So it seems it show the ntp server but won't communicate. But 225:s show nothing at all. The APs in the cluster are all wired. All APs and the virtual controller IP have free access to the internet. But when running the "show ntp debug" I can see the AP runs "ntpdate" against the ntp server (which is the inside interface of the firewall). So that works (i.e communication to the ntp server verified as working)....

    A possible bug when running "show ntp status" ?

    Am I missing something important maybe?



    Output from an non cluster master (IAP225)

    ap2.incedo.org (upstairs)# sh run | include ntp

    ntp-server ntp.incedo.org

    ap2.incedo.org (upstairs)# sh ntp status

    * master (synced), # master (unsynced), + selected, - candidate, x candidate, ~ configured

    ap2.incedo.org (upstairs)# 

    ap2.incedo.org (upstairs)# sh run | include clock

    clock timezone Stockholm 01 00

    clock summer-time CEST recurring last sunday march 02:00 last sunday october 03:00


    Output from an non cluster master (IAP535)

    ap4.incedo.org (livingroom)# sh run | include ntp

    ntp-server ntp.incedo.org

    ap4.incedo.org (livingroom)# sh ntp status

         remote           refid      st t when poll reach   delay   offset  jitter

    ==============================================================================

     fw-lan.incedo.o .INIT.          16 u    - 1024    0    0.000    0.000   0.000

    * master (synced), # master (unsynced), + selected, - candidate, x candidate, ~ configured

    ap4.incedo.org (livingroom)# 


    ap4.incedo.org (livingroom)# sh run | include clock

    clock timezone Stockholm 01 00

    clock summer-time CEST recurring last sunday march 02:00 last sunday october 03:00

    ap4.incedo.org (livingroom)# 



    ------------------------------
    Per-Olov Sj�holm
    ------------------------------