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

Re:Meteora

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: Re:Meteora
  • Date: Mon, 3 Dec 2001 09:34:05 -0800 (PST)
Hi Larry,

Is Meteora alright. Our server switched to the back
up feed (U.of Washington) last night (0400 UTC) and
hasn't been there since. Latencies on the back up
line are at least an hour.

This has been a recurring matter, but I only contact
you when latencies get out of hand.

James

--------------------------------------
James Murakami
Staff Meteorologist/Student Affairs
Department of Atmospheric Sciences
University of California, Los Angeles
405 Hilgard Ave.
Los Angeles, CA  90095-1565


   e-mail:  address@hidden
telephone:  310-825-2418
      Fax:  310-206-5219
---------------------------------------

>From address@hidden Mon Dec  3 11:40:33 2001
>To: James Murakami <address@hidden>
>From: Larry Riddle <address@hidden>
>Subject: Re:Meteora
>Cc: Larry Riddle <address@hidden>, address@hidden,
>   Mike Leuthold <address@hidden>

>Is Meteora alright. Our server switched to the back
>up feed (U.of Washington) last night (0400 UTC) and
>hasn't been there since. Latencies on the back up
>line are at least an hour.
>
>This has been a recurring matter, but I only contact
>you when latencies get out of hand.

James:

My files sizes look good, so the problem doesn't appear to be between 
motherlode.ucar.edu and meteora.ucsd.edu.  Mike Leuthold has been having 
the same problem at Arizona, however, so you're not alone.

I've just stopped and restarted the ldm.  Maybe that will make a 
difference.  But, I doubt it.

We should have an additional host running soon that will be an ldm 
dedicated machine.  The hostname is aeolus.ucsd.edu.  It will do nothing 
else but think about the ldm.  The current host, meteora.ucsd.edu, is also 
a file server and has "other duties" (but will remain as a back-up).

As soon as aeolus is up and running (with the same ldmd.conf as meteora), 
I'll send out email and let everyone know.

Larry

      ---===---=-=-=-=-=-=-=-=-=-=-=====[\/]=====-=-=-=-=-=-=-=-=-=-=---===---
    -----===(*  Climate's what we expect, but weather's what we 
get.  *)===-----
   Larry Riddle : Climate Research Division : Scripps Institution of 
Oceanography
       University of California, San Diego : La Jolla, California  92093-0224
       Phone: (858) 534-1869 : Fax: (858) 534-8561 : E-Mail: address@hidden

>From address@hidden Mon Dec  3 13:06:28 2001
>To: Larry Riddle <address@hidden>
>cc: James Murakami <address@hidden>, address@hidden
>Subject: Re:Meteora

Larry, James:
        Our problem seems to be related to some filtering that was done on
Arizona's I2 link by the I2 people, starting around Thanksgiving.   This
additional filtering added 50% more load to the cpu on the router and I
found that it would occasionally drop packets under load.  It 'appears'
that the I2 people did fix this problem as of this morning.  Perhaps the
I2 people did the same to the ucla->I2 link?  I saw 1-5% packet loss and
this is enough to cause high latencies with HDS. 

Mike