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

RE: ERCWC update



Hi all,


In regard to:

> The queue size on our LDM is fairly large, and I suspect that unless you
> have added the "-o 0" option into the perl script ldmadmin within the
> start_ldm function, on the line "$cmd_line .= "-o 0 -q $pq_path
$ldmd_conf > $pid_file";
> This will tell them LDM to request only current products.


the offset request flag will only spike latencies if and when the LDM is
stopped and then started again..as it will slurp data back to the -o
flag..default is 1 hour. Once the LDM is running in a stable mode, the -o
flag should be a non-issue (unless radar clocks are off..which we know
some are, but again that would only be a site or two jumping around the
high latency realm)


Cheers,


Jeff
---------------------------------------------------------------------
Jeff Weber                                    address@hidden        :
Unidata Program Center                        PH:303-497-8676        :
University Corp for Atmospheric Research      3300 Mitchell Ln       :
http://www.unidata.ucar.edu/staff/jweber      Boulder,Co 80307-3000  :
---------------------------------------------------------------------

On Tue, 26 Oct 2004, Timothy Sampair wrote:

> Hi Scott,
>
> Are network folks are working on updating/replacing a router we have to I2, so
> we should see improvements when they have completed those upgrades.
>
> The queue size on our LDM is fairly large, and I suspect that unless you
> have added the "-o 0" option into the perl script ldmadmin within the
> start_ldm function, on the line "$cmd_line .= "-o 0 -q $pq_path $ldmd_conf > 
> $pid_file";
> This will tell them LDM to request only current products.
>
> -tim
>
>       -----Original Message-----
>       From: Scott Podgorny [mailto:address@hidden]
>       Sent: Tue 10/26/2004 9:38 AM
>       To: Timothy Sampair
>       Cc: Jeff Weber
>       Subject: ERCWC update
>
>
>
>       Hi Tim,
>
>       An update for you on your feed to me.  We are still experiencing
>       significant lantencies of up to 55minutes.  Any chance your IT team has
>       anymore adjustments they can make for I2 connectivity?  Here is what my
>       traceroute looks like now.  Between routers 13 and 14 there seems to a
>       significant jump in packet latency.
>
>       For now I'm going to move back to the NSF machine.
>
>       thanks
>
>       scott
>
>       traceroute to ldm.erswc.org (152.53.28.18), 30 hops max, 38 byte packets
>        1  192.67.108.253 (192.67.108.253)  0.427 ms  0.347 ms  0.364 ms
>        2  198.116.144.78 (198.116.144.78)  0.942 ms  0.938 ms  0.946 ms
>        3  128.161.3.78 (128.161.3.78)  1.325 ms  1.320 ms  0.755 ms
>        4  128.161.3.110 (128.161.3.110)  1.325 ms  0.939 ms  0.948 ms
>        5  128.161.3.108 (128.161.3.108)  0.943 ms  1.320 ms  1.329 ms
>        6  128.161.1.22 (128.161.1.22)  134.268 ms  284.970 ms  286.896 ms
>        7  192.150.29.6 (192.150.29.6)  35.509 ms  35.525 ms  36.067 ms
>        8  nycmng-chinng.abilene.ucaid.edu (198.32.8.83)  35.712 ms  42.550 ms
>       35.911 ms
>        9  washng-nycmng.abilene.ucaid.edu (198.32.8.85)  35.980 ms  35.710 ms
>       50.954 ms
>       10  rlgh1-gw-abilene-oc48.ncren.net (198.86.17.65)  42.717 ms  42.704
>       ms  42.724 ms
>       11  chltgsr-gw-to-rpop-oc48.ncren.net (128.109.52.7)  46.970 ms  46.959
>       ms  46.724 ms
>       12  chlt7600-gw-to-chltgsr-gw.ncren.net (128.109.246.18)  47.221 ms
>       46.959 ms  46.981 ms
>       13  ercwc-gw-to-chlt7600-gw.ncren.net (128.109.246.58)  47.452 ms
>       47.202 ms  46.972 ms
>       14  152.53.4.1 (152.53.4.1)  144.449 ms  127.923 ms  135.687 ms
>       15  152.53.6.34 (152.53.6.34)  123.439 ms  119.684 ms  120.447 ms
>       16  152.53.28.18 (152.53.28.18)  121.691 ms  143.678 ms  128.951 ms
>
>       Just for comparison, here's the traceroute to atmos.geo.nsf.gov
>
>       traceroute to atm.geo.nsf.gov (198.181.231.53), 30 hops max, 38 byte 
> packets
>        1  192.67.108.253 (192.67.108.253)  0.493 ms  0.347 ms  0.380 ms
>        2  198.116.144.78 (198.116.144.78)  0.899 ms  0.893 ms  0.901 ms
>        3  128.161.3.78 (128.161.3.78)  1.292 ms  1.152 ms  1.162 ms
>        4  128.161.3.110 (128.161.3.110)  1.423 ms  1.283 ms  0.774 ms
>        5  128.161.3.108 (128.161.3.108)  1.681 ms  1.153 ms  1.033 ms
>        6  128.161.1.22 (128.161.1.22)  18.911 ms  18.792 ms  19.470 ms
>        7  192.150.29.6 (192.150.29.6)  35.670 ms  35.382 ms  37.106 ms
>        8  nycmng-chinng.abilene.ucaid.edu (198.32.8.83)  35.786 ms  40.502 ms
>       35.393 ms
>        9  washng-nycmng.abilene.ucaid.edu (198.32.8.85)  35.921 ms  35.903 ms
>       35.526 ms
>       10  dcne-abilene-oc48.maxgigapop.net (206.196.177.1)  35.392 ms  35.506
>       ms  35.790 ms
>       11  arlg-so3-1-0.maxgigapop.net (206.196.178.41)  35.787 ms  35.615 ms
>       35.657 ms
>       12  nsf-rtr.maxgigapop.net (206.196.177.138)  36.841 ms  36.301 ms
>       36.442 ms
>       13  atm.geo.nsf.gov (198.181.231.53)  36.706 ms  36.824 ms  37.368 ms
>
>
>       A significant diference in packet latency
>
>
>
>
>
>
>       --
>       _______________________________________________________________________
>       Scott Podgorny                             Phone   (256) 961-7819
>       Research Engineer                          Fax     (256) 961-7755
>       University of Alabama in Huntsville/ESSC   Email   address@hidden
>       NASA Marshall Space Flight Center/XD10     Web     http://nsstc.uah.edu
>
>       National Space Science & Technology Center
>       320 Sparkman Drive
>       Huntsville, AL  35805
>
>
>
>
>