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

20030128: LDM 6 release candidate available for testing (cont.)

>From: Harry Edmon <address@hidden>
>Organization: University of Washington
>Keywords: 200212271733.gBRHXAt25394 LDM 6 test

Hi Harry,

>Okay - I grabbed 6.0.0.7, built it, and I am now running it on marzipan using
>the same ldmd.conf file I had before.  It appears to be working.

I was going to test ldm-6.0.0.7 on our internal machines to make sure
that the new logic in a receiving LDM that keeps track of the time
associated with the most recently received product _before_ sending you
an email to grab the distribution, but your download beat me to the
punch.

We are just in the process of testing the new logic, so we would
appreciate it if you would keep an eye out for reconnections by
marzipan to thelma.  I will send you any warnings if/when we run into
problems here.

There is one modification that it would be nice for you to make in
your ldmd.conf file:

- change five request lines you have for your split NMC2 feed:
  
from:

request NMC2    ".*[09]$"   nmc2_0
request NMC2    ".*[18]$"   nmc2_1
request NMC2    ".*[27]$"   nmc2_2
request NMC2    ".*[36]$"   nmc2_3
request NMC2    ".*[45]$"   nmc2_4

to:

request NMC2    ".*[09]$"   thelma.ucar.edu
request NMC2    ".*[18]$"   thelma.ucar.edu
request NMC2    ".*[27]$"   thelma.ucar.edu
request NMC2    ".*[36]$"   thelma.ucar.edu
request NMC2    ".*[45]$"   thelma.ucar.edu

(assuming, of course that nmc2_* are your aliases for thelma.ucar.edu).

This will make the statistics plots look nicer, and the calculations
for differential latencies run faster.

Since you may want to bail out of ldm-6.0,0.7 back to the LDM 5 you
were using, the best way to make the change to ldmd.conf would be to
add a new section and comment out the old.

LDM 6 will _not_ collect these separate feed requests into a single
process, so your ingestion will continue to be split 5 ways.

You can keep an eye on your real time ingestion stats for marzipan at:

http://www.unidata.ucar.edu/staff/chiz/rtstats

Click on stats by host and then marzipan.  At some point, your latencies
for CONDUIT should exactly match those on thelma.

Tom

>From address@hidden Tue Jan 28 15:42:20 2003
>Subject: Re: 20030128: LDM 6 release candidate available for testing (cont.)

Okay - I did the restart with the ldmd.conf changes.  I did not see
reconnections.

-- 
 Dr. Harry Edmon                        E-MAIL: address@hidden
 206-543-0547                                   address@hidden
 Dept of Atmospheric Sciences           FAX:    206-543-0308
 University of Washington, Box 351640, Seattle, WA 98195-1640