Re: [chrony-dev] SW/HW timestamping on Linux

[Denny Page <dennypage@xxxxxx>]

Miroslav,

There is an additional issue, which is perhaps unrelated to the other issues discussed.

Background: Each target IP is a hardware based NTP server. IPs 240 and 244 are across the switch. Chrony is synchronizing with these two IPs. Incremental latency from the switch is approximately ~1.8 microseconds each way. IP 245 is directly attached, and is marked as noselect.

With software time stamping, the directly connected unit (IP 245) shows an average offset of approximately +250ns from the reference units.

MS Name/IP address         Stratum Poll Reach LastRx Last sample               

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

^+ 192.168.230.240               1   0   377     1    +32ns[  +32ns] +/-   19us

^* 192.168.230.244               1   0   377     1    -47ns[  -76ns] +/-   19us

^? 192.168.230.245               1   0   377     1   +310ns[ +310ns] +/-   16us

Name/IP Address            NP  NR  Span  Frequency  Freq Skew  Offset  Std Dev

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

192.168.230.240            58  36    78     -0.000      0.002    -33ns    80ns

192.168.230.244            30  18    41     +0.000      0.005    +25ns   111ns

192.168.230.245            20  16    24     -0.005      0.011   +188ns    87ns

With hardware time stamping enabled, this offset just to an average of approximate -2200ns from the reference units.

MS Name/IP address         Stratum Poll Reach LastRx Last sample               

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

^* 192.168.230.240               1   0   377     0   -515ns[ -547ns] +/-   13us

^+ 192.168.230.244               1   0   377     0   -454ns[ -486ns] +/-   13us

^? 192.168.230.245               1   0   377     0  -1800ns[-1832ns] +/- 6546ns

Name/IP Address            NP  NR  Span  Frequency  Freq Skew  Offset  Std Dev

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

192.168.230.240            64  30    76     -0.002      0.005    -66ns   264ns

192.168.230.244            64  32    76     +0.001      0.005    +66ns   231ns

192.168.230.245            57  33    68     +0.003      0.004  -2036ns   183ns

While I can get my head around a differential of 250ns resulting from the switch, I’m finding it very difficult to believe the  almost 2500ns differential that appears when hardware time stamping is enabled.

Any thoughts on this?

Thanks,

Denny