tcpdump mailing list archives
Re: Problems with ns Timestamps
From: Guy Harris <guy () alum mit edu>
Date: Thu, 28 Jul 2016 11:50:32 -0700
On Jul 28, 2016, at 7:12 AM, Christian <christian.rupp.stuttgart () freenet de> wrote:
I have encountered a problem. I am capturing Packages with hardware timestamps and nanosecond precision...but the timestamps are bugged. sadly I'm not at the place where I captured the data, so the used used command might be slightly different, as I have to work from my memory: tcpdump -i eth4 port 33330 -v -tt -j adapter_unsynced --time-stamp-precision=nano > dumpfile The problem is, that if the time is at below 100ms, the 0.1 field is not 0 as it should be, but doesn't exist.
So by "the 0.1 field" you mean "the tenths digit", so that, for example, a time stamp of 1.05 seconds would appear as 1.5? And are you saying that libpcap is providing bad time stamps, with an incorrect nanoseconds field, or are you saying that tcpdump is incorrectly printing a good time stamp? _______________________________________________ tcpdump-workers mailing list tcpdump-workers () lists tcpdump org https://lists.sandelman.ca/mailman/listinfo/tcpdump-workers
Current thread:
- Problems with ns Timestamps Christian (Jul 28)
- Re: Problems with ns Timestamps Guy Harris (Jul 28)
- Re: Problems with ns Timestamps Christian Rupp (Jul 28)
- Re: Problems with ns Timestamps Guy Harris (Jul 28)
- Re: Problems with ns Timestamps Christian Rupp (Jul 28)
- Re: Problems with ns Timestamps Guy Harris (Jul 28)