There was an interesting discussion on a software-defined networks list
about the amount of CPU and memory required to handle larger and faster UDP
data streams. Of course, they're talking about 10gigabit and 40gigabit
datastreams, but if your ingest machine is a little older and might be a
little light on memory, this could be an element.
gerry
On Thu, Aug 21, 2014 at 8:59 AM, Daniel Forrest <dan.forrest@xxxxxxxxxxxxx>
wrote:
> Gilbert,
>
> > OK, everything is good with the new NOAAport, right? C/N of 18. Signal
> > strength: 76. VBER: 0. Yay! But wait...
> >
> > I am seeing a ton of these in our ldmd.log:
> >
> > Aug 21 13:00:32 bird01 noaaportIngester[24645] WARN: Gap in packet
> sequence:
> > 13452198 to 13452266 [skipped 67]
> > Aug 21 13:00:32 bird01 noaaportIngester[24645] WARN: Retrograde packet
> > number: previous=13452270, latest=13452230, difference=4294967336
> > Aug 21 13:00:32 bird01 noaaportIngester[24645] WARN: Gap in packet
> sequence:
> > 13452230 to 13452232 [skipped 1]
> > Aug 21 13:00:32 bird01 noaaportIngester[24645] WARN: Gap in packet
> sequence:
> > 13452232 to 13452245 [skipped 12]
> > Aug 21 13:00:43 bird01 noaaportIngester[24645] WARN: Gap in packet
> sequence:
> > 13452515 to 13452524 [skipped 8]
> > Aug 21 13:00:43 bird01 noaaportIngester[24645] ERROR: Missing GOES
> fragment
> > in sequence, last 37/86751 this 46/86751
> > Aug 21 13:00:43 bird01 noaaportIngester[24645] WARN: Gap in packet
> sequence:
> > 13452524 to 13452598 [skipped 73]
> > Aug 21 13:00:43 bird01 noaaportIngester[24645] WARN: Retrograde packet
> > number: previous=13452598, latest=13452540, difference=4294967354
> > Aug 21 13:01:07 bird01 noaaportIngester[24645] WARN: Gap in packet
> sequence:
> > 13453270 to 13453274 [skipped 3]
> > Aug 21 13:01:07 bird01 noaaportIngester[24645] ERROR: Missing GOES
> fragment
> > in sequence, last 115/86753 this 119/86753
> > Aug 21 13:01:07 bird01 noaaportIngester[24645] WARN: Gap in packet
> sequence:
> > 13453279 to 13453283 [skipped 3]
> >
> > Is anyone else seeing this?
>
> I recently had a similar problem using the GOES-R simulator.
>
> Just a guess, but if the data is coming in via UDP and the data rate is
> higher than it was before then it may be a case of too little kernel
> buffer space. You can try this (assuming this is Linux):
>
> cat /proc/sys/net/core/rmem_max
> cat /proc/sys/net/core/wmem_max
> cat /proc/sys/net/core/rmem_default
> cat /proc/sys/net/core/wmem_default
>
> Save these values, if this doesn't help you can reset them.
>
> echo 16777216 >/proc/sys/net/core/rmem_max
> echo 16777216 >/proc/sys/net/core/wmem_max
> echo 16777216 >/proc/sys/net/core/rmem_default
> echo 16777216 >/proc/sys/net/core/wmem_default
>
> And restart LDM.
>
> If this works you can add these values to /etc/sysctl.conf.
>
> --
> Dan
>
> _______________________________________________
> ldm-users mailing list
> ldm-users@xxxxxxxxxxxxxxxx
> For list information or to unsubscribe, visit:
> http://www.unidata.ucar.edu/mailing_lists/
>
--
Gerry Creager
NSSL/CIMMS
405.325.6371
++++++++++++++++++++++
“Big whorls have little whorls,
That feed on their velocity;
And little whorls have lesser whorls,
And so on to viscosity.”
Lewis Fry Richardson (1881-1953)
