[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

20040825: ldmfail

Luis,

> Date: Wed, 25 Aug 2004 14:07:57 -0400
> From: "Luis Cano" <address@hidden>
> Organization: DOC/NOAA/NWS/NCEP
> To: address@hidden
> Cc: Scott Jacobs <address@hidden>
> Subject: ldmfail
> Keywords: 200408251807.i7PI7RXn018684

The above message contained the following:

> I was wondering if you knew of an issue with ldmping with a busy ldmd
> server. We use ldmfail with SPC and intermittently, SPC is determined
> down when it is not. I captured a failed ldmfail with a tcpdump, its
> attached. It looks like the server procedure never breaks out to give
> time to the network code to respond to the sync (line 243, 277, 331). I
> think spc is running a version of ldm6 and we are running a version of
> ldm5.

It is definitely possible for the ldmping(1) utility to return a false
negative if the network is too congested or the remote LDM host is too
busy.  As far as I know, there is no way to avoid this.

If you have the bandwidth, then I would configure your LDM system to
receive duplicate feeds of the same data from different upstream hosts.
That way if one goes down, you still get the data.

> Thanks,
> 
> Lou
...
> 13:24:14.611941 ncosrv5.ncep.noaa.gov.35046 > alice.spc.noaa.gov.ldm: P 
> 1505:1537(32) ack 116340 win 35040 (DF)
> 13:24:14.632407 ncosrv5.ncep.noaa.gov.35107 > alice.spc.noaa.gov.ldm: S 
> 3870427560:3870427560(0) win 5840 <mss 1460,sackOK,timestamp 8689395 
> 0,nop,wscale 0> (DF)
> 13:24:14.654639 alice.spc.noaa.gov.ldm > ncosrv5.ncep.noaa.gov.35046: . ack 
> 1537 win 65535 (DF)
> 13:24:15.273414 alice.spc.noaa.gov.ldm > ncosrv5.ncep.noaa.gov.35046: P 
> 116340:116604(264) ack 1537 win 65535 (DF)
...
> 13:24:17.235982 ncosrv5.ncep.noaa.gov.35046 > alice.spc.noaa.gov.ldm: P 
> 1825:1857(32) ack 120612 win 35040 (DF)
> 13:24:17.278418 alice.spc.noaa.gov.ldm > ncosrv5.ncep.noaa.gov.35046: . ack 
> 1857 win 65535 (DF)
> 13:24:17.629126 ncosrv5.ncep.noaa.gov.35107 > alice.spc.noaa.gov.ldm: S 
> 3870427560:3870427560(0) win 5840 <mss 1460,sackOK,timestamp 8689695 
> 0,nop,wscale 0> (DF)
> 13:24:19.470711 alice.spc.noaa.gov.ldm > ncosrv5.ncep.noaa.gov.35046: P 
> 120612:120896(284) ack 1857 win 65535 (DF)
...
> 13:24:23.462998 alice.spc.noaa.gov.ldm > ncosrv5.ncep.noaa.gov.35046: P 
> 155232:155680(448) ack 1985 win 65535 (DF)
13:24:23.463528 ncosrv5.ncep.noaa.gov.35046 > alice.spc.noaa.gov.ldm: P 
1985:2017(32) ack 155680 win 35040 (DF)
> 13:24:23.508991 alice.spc.noaa.gov.ldm > ncosrv5.ncep.noaa.gov.35046: . ack 
> 2017 win 65535 (DF)
> 13:24:23.630552 ncosrv5.ncep.noaa.gov.35107 > alice.spc.noaa.gov.ldm: S 
> 3870427560:3870427560(0) win 5840 <mss 1460,sackOK,timestamp 8690295 
> 0,nop,wscale 0> (DF)
> 13:24:25.454202 alice.spc.noaa.gov.ldm > ncosrv5.ncep.noaa.gov.35046: . 
> 155680:157140(1460) ack 2017 win 65535 (DF)
...

Regards,
Steve Emmerson
LDM Developer