[ntp:questions] Peridoc anomalies in loopstats file ...

unruh unruh at wormhole.physics.ubc.ca
Wed Jan 12 17:54:42 UTC 2011


On 2011-01-12, Heiko Gerstung <heiko.removethistext.gerstung at meinberg.de> wrote:
> Just an addition:
>
> The peerstats file shows the nearly the same time jumps at exactly the same time. I am still wondering if this is NTP or the
> Kernel ...

Probably the hardware. But what kind of kernel are you running?
Tickless? Or is this virtual OS? Remember, all the kernel has to figure
out the time is its own interenal counters/oscillators. If they go out,
or skip, then suddenly your time is out. The time does not seem to be
jumping (ie is not going from 2us to 2ms between one reading and the
next) but rather seems suddenly to be drifting fast-- as if the CPU
suddenly got hot. Ie, is this correlated with a sudden burst of activity
of the computer?


>
> Best Regards,
>   Heiko
>
>
> Am 10.01.2011 11:15, schrieb Heiko Gerstung:
>> Hi Everybody!
>> 
>> Did anybody ever see this kind of anomaly:
>> 
>> I see a periodic "jump" of time on my Geode LX800 based embedded CPU
>> board running Linux. This looks like an overflow or something, but the
>> time jumps recorded in loopstats are up to -400 microseconds and appear
>> to be happening exactly every 12264 seconds (3 hours, 24 minutes, 24
>> seconds). In between those jumps, the loopstats file records a maxmimum
>> offset of +/- 2 microseconds to the refclock.
>> 
>> Here are the three occurrences from my latest loopstats file:
>> 
>> [... lots of records with a max. of +/-2 microseconds offset]
>> 55571 1278.037 -0.000001000 -85.118 0.000001924 0.000352 3
>> 55571 1286.037 0.000000000 -85.118 0.000001922 0.000329 3
>> 55571 1294.037 -0.000228000 -85.230 0.000080528 0.039362 3
>> 55571 1302.037 0.000035000 -85.213 0.000119531 0.037312 3
>> 55571 1310.037 0.000047000 -85.190 0.000111893 0.035833 3
>> 55571 1318.037 0.000040000 -85.170 0.000104692 0.034222 3
>> 55571 1326.037 0.000032000 -85.154 0.000097979 0.032485 3
>> 55571 1334.037 0.000021000 -85.144 0.000091723 0.030603 3
>> 55571 1342.037 0.000016000 -85.136 0.000085819 0.028759 3
>> 55571 1350.037 0.000012000 -85.131 0.000080291 0.026981 3
>> 55571 1358.037 0.000009000 -85.126 0.000075111 0.025287 3
>> 55571 1366.037 0.000008000 -85.122 0.000070263 0.023694 3
>> 55571 1374.037 0.000007000 -85.119 0.000065729 0.022196 3
>> 55571 1382.037 0.000004000 -85.117 0.000061491 0.020774 3
>> 55571 1390.037 0.000001000 -85.116 0.000057527 0.019433 3
>> 55571 1398.037 0.000002000 -85.115 0.000053816 0.018181 3
>> 55571 1406.037 0.000001000 -85.115 0.000050344 0.017008 3
>> 55571 1414.037 0.000001000 -85.114 0.000047098 0.015911 3
>> 55571 1422.037 0.000002000 -85.113 0.000044061 0.014887 3
>> [... lots of records with a max. of +/-2 microseconds offset]
>> 55571 13526.037 -0.000001000 -85.112 0.000001907 0.000348 3
>> 55571 13534.037 0.000000000 -85.112 0.000001907 0.000325 3
>> 55571 13542.037 0.000000000 -85.112 0.000001907 0.000304 3
>> 55571 13550.037 0.000000000 -85.112 0.000001907 0.000285 3
>> 55571 13558.037 -0.000400000 -85.307 0.000141482 0.069054 3
>> 55571 13566.037 0.000063000 -85.276 0.000210342 0.065503 3
>> 55571 13574.037 0.000087000 -85.234 0.000196946 0.063087 3
>> 55571 13582.037 0.000074000 -85.197 0.000184281 0.060379 3
>> 55571 13590.037 0.000056000 -85.170 0.000172496 0.057301 3
>> 55571 13598.037 0.000040000 -85.151 0.000161456 0.054043 3
>> 55571 13606.037 0.000028000 -85.137 0.000151090 0.050783 3
>> 55571 13614.037 0.000022000 -85.126 0.000141345 0.047655 3
>> 55571 13622.037 0.000016000 -85.118 0.000132238 0.044663 3
>> 55571 13630.037 0.000012000 -85.112 0.000123703 0.041829 3
>> 55571 13638.037 0.000010000 -85.108 0.000115716 0.039166 3
>> 55571 13646.037 0.000007000 -85.104 0.000108248 0.036656 3
>> 55571 13654.037 0.000004000 -85.102 0.000101261 0.034296 3
>> 55571 13662.037 0.000003000 -85.101 0.000094723 0.032085 3
>> 55571 13670.037 0.000003000 -85.099 0.000088608 0.030017 3
>> 55571 13678.037 0.000001000 -85.099 0.000082888 0.028079 3
>> 55571 13686.037 0.000002000 -85.098 0.000077537 0.026268 3
>> 55571 13694.037 0.000001000 -85.097 0.000072533 0.024572 3
>> 55571 13702.037 0.000001000 -85.097 0.000067851 0.022985 3
>> 55571 13710.037 -0.000001000 -85.097 0.000063474 0.021502 3
>> 55571 13718.037 0.000000000 -85.097 0.000059378 0.020113 3
>> [... lots of records with a max. of +/-2 microseconds offset]
>> 55571 25806.037 -0.000001000 -85.065 0.000001931 0.000281 3
>> 55571 25814.037 -0.000152000 -85.139 0.000053561 0.026242 3
>> 55571 25822.037 0.000024000 -85.127 0.000079866 0.024894 3
>> 55571 25830.037 0.000032000 -85.112 0.000074760 0.023932 3
>> 55571 25838.037 0.000027000 -85.098 0.000069950 0.022867 3
>> 55571 25846.037 0.000020000 -85.089 0.000065476 0.021667 3
>> 55571 25854.037 0.000014000 -85.082 0.000061292 0.020411 3
>> 55571 25862.037 0.000011000 -85.076 0.000057339 0.019187 3
>> 55571 25870.037 0.000009000 -85.072 0.000053645 0.018015 3
>> 55571 25878.037 0.000008000 -85.068 0.000050185 0.016908 3
>> 55571 25886.037 0.000006000 -85.065 0.000046951 0.015850 3
>> 55571 25894.037 0.000004000 -85.063 0.000043925 0.014842 3
>> 55571 25902.037 0.000002000 -85.062 0.000041093 0.013888 3
>> 55571 25910.037 0.000002000 -85.061 0.000038445 0.012996 3
>> 55571 25918.037 0.000001000 -85.061 0.000035968 0.012157 3
>> 55571 25926.037 0.000000000 -85.061 0.000033652 0.011372 3
>> 55571 25934.037 0.000001000 -85.060 0.000031486 0.010639 3
>> 55571 25942.037 0.000001000 -85.060 0.000029460 0.009954 3
>> 55571 25950.037 0.000000000 -85.060 0.000027566 0.009311 3
>> [... lots of records with a max. of +/-2 microseconds offset]
>> 
>> I would be grateful for *any* suggestions/explanations. I already tried
>> using a different clocksource, to no avail. The same hardware did not
>> show this behavior with an older kernel/ntp combination (2.6.15.1 and
>> ntp 4.2.0b), unfortunately it is not easily possible to replace only the
>> kernel OR ntp.
>> 
>> Some information: ntp4.2.6p3 on Linux 32bit using tsc clocksource,
>> Embedded device, synchronized to GPS receiver with PPS support, refclock
>> driver 8, PPSAPI patches for kernel installed and working
>> 
>> Best Regards,
>>   Heiko
>




More information about the questions mailing list