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

Re: 20011004: stcloudstate reclass

"Anderson, Alan C." wrote:
> 
> -----Original Message-----
> From: anne [mailto:address@hidden]
> Sent: Thursday, October 04, 2001 10:28 PM
> To: Anderson, Alan C.; Pete Pokrandt
> Cc: address@hidden; address@hidden
> Subject: Re: 20011004: stcloudstate reclass
> 
> Unidata Support wrote:
> 
> Alan,
> 
> If your site is dropping products because they're arriving too late, you
> should see evidence of this in your log.  Look for 'skipped' entries.
> 
> If you need help diagnosing the problem, let me know.
> 
> Anne
> 
> --
> ***************************************************
> Anne Wilson                     UCAR Unidata Program
> address@hidden                  P.O. Box 3000
>                                   Boulder, CO  80307
> ----------------------------------------------------
> Unidata WWW server       http://www.unidata.ucar.edu/
> ****************************************************
> 
> Anne
> 
> Thanks for writing, I appreciate your interest.  I looked at our
> ldm log files this morning and can see some skipped entries, but
> I don't feel competent to judge whether we still have a problem.
> 
> We do seem to have a full stack of IR images, but I only get
> one Goes East Vis image even though my directory shows a full
> set of AREA014* files, which is where I think the Goes E vis
> files are supposed to be.
> 
> I do see what looks like a lot of startup messages with a file
> decoding line, and then an exit within just a second or so later.
> This was for Area files, which I would have expected to take
> somewhat longer.
> I have pasted a couple of exerpts from our log file below.
> I would appreciate your comments about the meaning.
> 
> Our ldm ingest machine is host waldo.stcloudstate.edu
> (ip # 199.17.31.240) 
> if you want to take a look.
> 
> Thanks
> 
> Alan Anderson
> 
> Oct 05 15:30:35 waldo pnga2area[24631]: output file pathname:
> /var/data/mcidas/A
> REA0169
> Oct 05 15:30:35 waldo pnga2area[24631]: unPNG::   292621    607776
> 2.0770
> Oct 05 15:30:35 waldo pnga2area[24631]: Exiting
> Oct 05 15:32:30 waldo pqexpire[424]: > Recycled   2185.062 kb/hr (
> 5931.031 pro
> ds per hour)
> Oct 05 15:32:33 waldo pnga2area[24635]: Starting Up
> Oct 05 15:32:33 waldo pnga2area[24635]: output file pathname:
> /var/data/mcidas/A
> REA1133
> Oct 05 15:32:34 waldo pnga2area[24635]: unPNG::    78000    242720
> 3.1118
> Oct 05 15:32:34 waldo pnga2area[24635]: Exiting
> Oct 05 15:34:47 waldo pnga2area[24651]: Starting Up
> Oct 05 15:34:47 waldo pnga2area[24651]: output file pathname:
> /var/data/mcidas/A
> REA0066
> Oct 05 15:34:47 waldo pnga2area[24651]: unPNG::   317273    607776
> 1.9156
> Oct 05 15:34:47 waldo pnga2area[24651]: Exiting
> Oct 05 15:35:19 waldo nldn2md[24669]: NLDN2MD -- BEGIN
> Oct 05 15:35:19 waldo nldn2md[24669]: PRODUCT CODE=LD          2001278
> 15240
> 0
> 
> Oct 05 15:57:30 waldo pqexpire[424]: > Recycled   2183.065 kb/hr (
> 5925.494 pro
> ds per hour)
> Oct 05 15:59:03 waldo proftomd[24743]: Starting up
> Oct 05 15:59:03 waldo proftomd[24743]: Decoding 2001278.1454 data into
> /var/data
> /mcidas/MDXX0098
> Oct 05 15:59:03 waldo proftomd[24743]: Exiting
> Oct 05 15:59:21 waldo nldn2md[24744]: NLDN2MD -- BEGIN
> Oct 05 15:59:21 waldo nldn2md[24744]: PRODUCT CODE=LD          2001278
> 15480
> 0
> Oct 05 15:59:21 waldo nldn2md[24744]: Appending to MD file: 78
> Oct 05 15:59:21 waldo nldn2md[24744]: NLDN2MD - Flash events in file:
> 56256
> Oct 05 15:59:21 waldo nldn2md[24744]: NLDN2MD -- DONE
> Oct 05 16:00:34 waldo pnga2area[24798]: Starting Up
> Oct 05 16:00:34 waldo pnga2area[24798]: output file pathname:
> /var/data/mcidas/A
> REA0202
> Oct 05 16:00:34 waldo pnga2area[24798]: unPNG::     6982    225088
> 32.2383
> Oct 05 16:00:34 waldo pnga2area[24798]: Exiting
> Oct 05 16:02:30 waldo pqexpire[424]: > Recycled   2182.576 kb/hr (
> 5924.132 pro
> ds per hour)
> Oct 05 16:04:13 waldo pnga2area[24830]: Starting Up
> Oct 05 16:04:13 waldo pnga2area[24830]: output file pathname:
> /var/data/mcidas/A
> REA0190
> Oct 05 16:04:14 waldo pnga2area[24830]: unPNG::   124315    308448
> 2.4812
> Oct 05 16:04:14 waldo pnga2area[24830]: Exiting
> /usr/local/ldm/logs%



Hi Alan,

I logged on to your machine.  First, your products are coming in rather
slow - there's about a 40 minute latency.  I did a traceroute to sunset
to check the connectivity:

/usr/local/ldm/etc% traceroute sunset.meteor.wisc.edu
traceroute to sunset.meteor.wisc.edu (144.92.130.6), 30 hops max, 40
byte packets
 1  router (199.17.31.254)  0.794 ms  1.315 ms  1.463 ms
 2  * * *
 3  199.17.55.253 (199.17.55.253)  3.181 ms  12.886 ms  15.104 ms
 4  umn-lena-ATM3-0-0-940.r.state.mn.us (207.171.90.122)  5.692 ms 
5.741 ms  7.519 ms
 5  otrx.northernlights.gigapop.net (192.42.152.22)  8.496 ms  5.986 ms 
7.621 ms
 6  otr-onvoy.northernlights.gigapop.net (192.42.152.14)  40.917 ms 
38.025 ms  39.606 ms
 7  core1-ge1-1-0.msc.mr.net (137.192.3.254)  47.190 ms  48.657 ms 
50.747 ms
 8  core1-so1-0-1.ply.mr.net (137.192.5.9)  40.182 ms  25.065 ms  21.099
ms
 9  206.220.243.118 (206.220.243.118)  26.067 ms  34.686 ms  27.808 ms
10  r-peer-WNMadison-gw.net.wisc.edu (216.56.1.18)  26.100 ms  29.289
ms  30.624 ms
11  144.92.128.206 (144.92.128.206)  30.048 ms  28.222 ms  26.280 ms
12  sunset.meteor.wisc.edu (144.92.130.6)  30.847 ms  26.034 ms  33.446
ms

The second hop may indicate a firewall, I'm not sure. Or, there's
congestion on your campus, as the third hop is still in your 199.17
domain.

In looking at the logs, you are definately loosing some products.  So
far today you've lost 31.  Here are loses for three previous days:
        Oct 4:  81
        Oct 3: 292
        Oct 2:  34

The products appear to be running between a few fractions of seconds too
old and over 10 minutes to old.  I think this is because of the quality
of your connection to sunset.

Otherwise, I don't see any other problems in the log.  I think the
products that are being decoded are probably complete, otherwise I
believe an error would be reported in the log.  Are they working in
Mcidas for you?

One way to avoid losing products due to latencies is to change the
default setting from an hour old to something longer.  I can do this for
you if you'd like.

However, I have one other recommendation for you.  You are running a
very old version of the LDM.  I highly recommend that you upgrade.  You
are missing the advantages of significantly improved algorithms, plus
some new feed types including NEXRAD.  What do you think?

Anne

-- 
***************************************************
Anne Wilson                     UCAR Unidata Program            
address@hidden                 P.O. Box 3000
                                  Boulder, CO  80307
----------------------------------------------------
Unidata WWW server       http://www.unidata.ucar.edu/
****************************************************