[ntp:questions] Re: xntpd crash on solaris 8

Richard B. Gilbert rgilbert88 at comcast.net
Tue Jul 25 05:12:46 UTC 2006


kk at girard.org wrote:

> Hi,all:
> 
> 
> There are 3 sun solaris 8 boxes in my lab, resently I tried  to open xntpd
> service on them---2 as servers,one as client, but met a strange thing.
> 
> When startup the xntpd on 3 boxes, all time are syncronized. If I adjusted
> the time of client 10 min forward, xntpd could sync it by about 4 poll
> interval, but when I adjusted the client time about 2 hours forward, the
> xntpd daemon crashed after about 4 poll interval!
> 
> I checked the process by "ntpq -c peers", the normal log is as following:
> 
>  BJWSS_ccm_1_14_1:> while [ 1 ]; do ntpq -c peers; sleep 10; done
>      remote           refid      st t when poll reach   delay   offset
> disp
> ============================================================================
> ==
> *BJWSS_sam_1_5_1 .LCL.            1 u   25  128  377     0.53   -0.041
> 0.03
> +BJWSS_sam_1_6_1 BJWSS_sam_1_5_1  2 u   35  128  377     0.49   -0.011
> 0.03
>      remote           refid      st t when poll reach   delay   offset
> disp
> ============================================================================
> ==
> *BJWSS_sam_1_5_1 .LCL.            1 u   35  128  377     0.53   -0.041
> 0.03
> +BJWSS_sam_1_6_1 BJWSS_sam_1_5_1  2 u   45  128  377     0.49   -0.011
> 0.03
> 
> 
> but if I adjusted 2 hours forward of client, xntpd couldn't select a new
> primary ntp server,after about 4 poll interval, xntpd daemon will crash
> automatically :((
> 
> ============================================================================
> ==
> *BJWSS_sam_1_5_1 .LCL.            1 u 140m  128  377     0.53   -0.024
> 0.03
> +BJWSS_sam_1_6_1 BJWSS_sam_1_5_1  2 u 141m  128  377     0.41    0.026
> 0.05
>      remote           refid      st t when poll reach   delay   offset
> disp
> ============================================================================
> ==
> *BJWSS_sam_1_5_1 .LCL.            1 u 141m  128  377     0.53   -0.024
> 0.03
>  BJWSS_sam_1_6_1 BJWSS_sam_1_5_1  2 u    8  128  377     0.46  -830538
> 15875.0
>      remote           refid      st t when poll reach   delay   offset
> disp
> ============================================================================
> ==
>  BJWSS_sam_1_5_1 .LCL.            1 u    8  128  377     0.34  -830538
> 15875.0
>  BJWSS_sam_1_6_1 BJWSS_sam_1_5_1  2 u   18  128  377     0.46  -830538
> 15875.0
>      remote           refid      st t when poll reach   delay   offset
> disp
> ============================================================================
> ==
>  BJWSS_sam_1_5_1 .LCL.            1 u   18  128  377     0.34  -830538
> 15875.0
>  BJWSS_sam_1_6_1 BJWSS_sam_1_5_1  2 u   28  128  377     0.46  -830538
> 15875.0
> 
> ......
> 
> Any one met similar problem before? The OS version of the boxes is "SunOS
> BJWSS_ccm_1_14_1 5.8 Generic_117350-12 sun4u sparc SUNW,Netra-240"...
> 
> 
> 
> 
> 
> 
> The ntp.conf files are as following:
> 
> server1:
> BJWSS_sam_1_5_1
> 
> # Either a peer or server.  Replace "XType" with a value from the
> # table above.
> server 127.127.1.0 prefer
> fudge 127.127.1.0 stratum 0
> peer BJWSS_sam_1_6_1
> 
> enable auth monitor
> driftfile /var/ntp/ntp.drift
> statsdir /var/ntp/ntpstats/
> filegen peerstats file peerstats type day enable
> filegen loopstats file loopstats type day enable
> filegen clockstats file clockstats type day enable
> 
> ============================================================================
> 
> server2:
> 
> BJWSS_sam_1_6_1
> # Either a peer or server.  Replace "XType" with a value from the
> # table above.
> server BJWSS_sam_1_5_1  prefer
> #server 127.127.1.0
> fudge 127.127.1.0 stratum 0
> peer BJWSS_sam_1_5_1 minpoll 6
> 
> enable auth monitor
> driftfile /var/ntp/ntp.drift
> statsdir /var/ntp/ntpstats/
> filegen peerstats file peerstats type day enable
> filegen loopstats file loopstats type day enable
> filegen clockstats file clockstats type day enable
> 
> 
> ============================================================================
> ================
> 
> Client
> 
> # /etc/inet/ntp.conf generated on Sat Jun  3 23:23:22 CST 2006
> # @(#)ntp.client        1.2     96/11/06 SMI
> #
> # /etc/inet/ntp.client
> #
> # An example file that could be copied over to /etc/inet/ntp.conf; it
> # provides a configuration for a host that passively waits for a server
> # to provide NTP packets on the ntp multicast net.
> #
> 
> server BJWSS_sam_1_5_1 prefer minpoll 6 maxpoll 7
> server BJWSS_sam_1_6_1 minpoll 6 maxpoll 7
> driftfile /var/ntp/ntp.drift
> 
> b/r
> 
> oxy
> 
> 

I believe you will find that ntpd is working as designed!  The "panic 
threshold" is 1024 seconds or about 17 minutes.  If the clock is off by 
more than the panic threshold, ntpd does a panic exit!

The same thing will happen if you start ntpd with your local clock more 
than 1024 seconds off and do not use the -g option.




More information about the questions mailing list