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

20011008: motherlode feed slow? (cont.)



>From: "Arthur A. Person" <address@hidden>
>Organization: Penn State
>Keywords: 200110061836.f96IaT125601 IDD

Art,

>Our latencies have quickly popped back up to about 55 minutes now from
>motherlode.  SSEC NEXRAD is still running within seconds.  I'm working
>with our networking people now to see if they can discover anything.

We see a jump in traceroute times at our abilene connection right at the
moment, but I don't know if this is enough to slow the delivery from
motherlode to downstream sites to the extent you are seeing:

motherlode.ucar.edu% traceroute sunny.atmos.washington.edu
traceroute to sunny.atmos.washington.edu (128.95.89.38), 30 hops max, 40 byte 
packets
 1  mlra-n13 (128.117.13.253)  0.511 ms  0.542 ms  0.310 ms
 2  vbnsr-n2 (128.117.2.252)  0.463 ms  0.591 ms  0.335 ms
 3  internetr-n243-104 (128.117.243.106)  1.107 ms  1.042 ms  0.873 ms
 4  denv-abilene (128.117.243.126)  2.196 ms  2.739 ms  2.150 ms
 5  sttl-dnvr.abilene.ucaid.edu (198.32.8.49)  29.941 ms  30.317 ms  33.031 ms
 6  hnsp1-wes-S5-0-0.pnw-gigapop.net (198.48.91.77)  30.270 ms  30.288 ms  
29.983 ms
 7  uwbr1-GE5-0.cac.washington.edu (198.107.151.51)  29.922 ms  29.917 ms  
30.412 ms
 8  dustbuster-V49.cac.washington.edu (140.142.150.2)  30.968 ms  30.772 ms  
30.594 ms
 9  sunny.atmos.washington.edu (128.95.89.38)  30.325 ms  30.527 ms  30.436 ms

I am CCing this to Anne for further comment/action.

Tom

>On Mon, 8 Oct 2001, Arthur A. Person wrote:
>
>> On Sun, 7 Oct 2001, Unidata Support wrote:
>>
>> > >From: "Arthur A. Person" <address@hidden>
>> > >Organization: UCAR/Unidata
>> > >Keywords: 200110061836.f96IaT125601
>> >
>> > Art,
>> >
>> > Sorry I didn't see this before this morning.
>> >
>> > >Our feed from motherlode is about 40 minutes behind...  this has been
>> > >happening for several hours according to our users.  Is motherlode having
>> > >a problem, or some networking issue?  Our feed from SSEC and WSI (for
>> > >example) are on-time.
>> >
>> > 'traceroute's from motherlode to three different machines show a significa
> nt
>> > increase at the abilene connection in Denver, but these are not (by an
>> > order of magnitude) as bad as what we were seeing last week:
>> >
>> > motherlode.ucar.edu% windfall.evsc.virginia.edu (128.143.142.246), 30 hops
>  max, 40 byte packets
>> >  1  mlra-n13 (128.117.13.253)  0.393 ms  0.406 ms  0.321 ms
>> >  2  vbnsr-n2 (128.117.2.252)  0.719 ms  0.808 ms  0.462 ms
>> >  3  internetr-n243-104 (128.117.243.106)  0.835 ms  0.750 ms  0.687 ms
>> >  4  denv-abilene (128.117.243.126)  1.745 ms  1.435 ms  2.519 ms
>> >  5  kscy-dnvr.abilene.ucaid.edu (198.32.8.14)  12.757 ms  12.745 ms  12.61
> 0 ms
>> >  6  ipls-kscy.abilene.ucaid.edu (198.32.8.6)  21.673 ms  21.931 ms  22.044
>  ms
>> >  7  clev-ipls.abilene.ucaid.edu (198.32.8.26)  28.459 ms  27.756 ms  28.04
> 2 ms
>> >  8  nycm-clev.abilene.ucaid.edu (198.32.8.30)  39.994 ms  42.397 ms  40.91
> 6 ms
>> >  9  uva-internet.misc.Virginia.EDU (192.35.48.41)  48.781 ms  48.051 ms  4
> 8.335 ms
>> > 10  carruthers1-all-router.acc.Virginia.EDU (128.143.99.2)  48.793 ms  50.
> 587 ms  48.993 ms
>> > 11  garrett-router.acc.Virginia.EDU (128.143.227.55)  48.427 ms  55.669 ms
>   56.705 ms
>> > 12  windfall.evsc.Virginia.EDU (128.143.142.246)  49.113 ms *  51.171 ms
>> > motherlode.ucar.edu% traceroute ldm.meteo.psu.edu
>> > traceroute to ldm.meteo.psu.edu (128.118.28.12), 30 hops max, 40 byte pack
> ets
>> >  1  mlra-n13 (128.117.13.253)  0.397 ms  0.293 ms  0.260 ms
>> >  2  vbnsr-n2 (128.117.2.252)  0.497 ms  0.723 ms  0.714 ms
>> >  3  internetr-n243-104 (128.117.243.106)  0.985 ms  0.655 ms  0.593 ms
>> >  4  denv-abilene (128.117.243.126)  2.250 ms  1.694 ms  1.895 ms
>> >  5  kscy-dnvr.abilene.ucaid.edu (198.32.8.14)  13.265 ms  12.135 ms  12.27
> 0 ms
>> >  6  ipls-kscy.abilene.ucaid.edu (198.32.8.6)  21.351 ms  21.388 ms  21.532
>  ms
>> >  7  clev-ipls.abilene.ucaid.edu (198.32.8.26)  27.697 ms  27.944 ms  27.63
> 4 ms
>> >  8  abilene.psc.net (192.88.115.122)  31.044 ms  31.789 ms  30.856 ms
>> >  9  penn-state.psc.net (198.32.224.66)  34.750 ms  35.134 ms  35.304 ms
>> > 10  * * *
>> > 11  * * *
>> > 12  * * *
>> > 13 ^C
>> >
>> > motherlode.ucar.edu% traceroute sunny.atmos.washington.edu
>> > traceroute to sunny.atmos.washington.edu (128.95.89.38), 30 hops max, 40 b
> yte packets
>> >  1  mlra-n13 (128.117.13.253)  0.531 ms  0.237 ms  0.376 ms
>> >  2  vbnsr-n2 (128.117.2.252)  0.525 ms  0.578 ms  0.611 ms
>> >  3  internetr-n243-104 (128.117.243.106)  1.091 ms  0.748 ms  0.603 ms
>> >  4  denv-abilene (128.117.243.126)  1.559 ms  1.391 ms  1.361 ms
>> >  5  sttl-dnvr.abilene.ucaid.edu (198.32.8.49)  30.196 ms  30.232 ms  30.05
> 6 ms
>> >  6  hnsp1-wes-S5-0-0.pnw-gigapop.net (198.48.91.77)  29.871 ms  30.391 ms 
>  30.313 ms
>> >  7  uwbr1-GE5-0.cac.washington.edu (198.107.151.51)  30.106 ms  30.904 ms 
>  30.213 ms
>> >  8  dustbuster-V49.cac.washington.edu (140.142.150.2)  31.096 ms  31.053 m
> s  30.981 ms
>> >  9  sunny.atmos.washington.edu (128.95.89.38)  30.621 ms  31.133 ms  30.48
> 3 ms
>> >
>> > Are you still seeing the slowness today?
>>
>> Our traceroutes this morning (Monday) to motherlode don't look bad (except
>> for a few blips now-and-then) but we've seen data slowness all weekend I'm
>> told.  Looking at things this morning, the DDPLUS/HDS stuff is running
>> about 3-4 minutes behind which isn't bad, but normally this stuff is
>> within a minute, usually less.  The NEXRAD stuff from SSEC, for
>> comparison, is within a few seconds of transmission.  Since we're in a
>> non-peak load time, I'm guessing delays from motherlode will increase as
>> the data rate increases... we'll see.  Since the traceroutes look pretty
>> good, it would appear that the data are either being throttled or hitting
>> severe bottlenecks.  I'll continue to monitor it today.
>>
>>                                      Art.
>>
>> >
>> > Tom Yoksas
>
>Arthur A. Person
>Research Assistant, System Administrator
>Penn State Department of Meteorology
>email:  address@hidden, phone:  814-863-1563



NOTE: All email exchanges with Unidata User Support are recorded in the Unidata inquiry tracking system and then made publicly available through the web. If you do not want to have your interactions made available in this way, you must let us know in each email you send to us.