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

[LDM #IIX-893121]: Question...

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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: [LDM #IIX-893121]: Question...
  • Date: Wed, 31 Oct 2007 12:07:54 -0600
Hi Gerry,

re:
> Update time.
> 
> 1.  Upgraded kernel.  Errors persisted.  rmmod'd the k8 ecc testing
> module.  Errors went away.

Very interesting!

> 2.  Rebooted into memtest86.  6% complete, test 4 thru 10 GB and there
> are no errors so far.  I'll run to completion.  That could take awhile.

Yup, full memory testing can take LOTS of hours.  We know; we (Mike Schmidt)
have done this on several machines here at the UPC.

> Looks like the kernel issue is still real.  This is on RHEL4/CentOS4
> with, now, the latest kernel.

Curious.  I have a comment/observation:

- the latencies that are being seen in data transfers from sasquatch to
  its SCOOP downstreams are real, AND very different from latencies
  being seen in feeds from bigbird.  Case in point is the two different
  feeds to weather.renci.org: one from bigbird (NEXRAD2) and one from
  sasquatch (limited EXP).  The feed from bigbird to weather.renci.org
  showed very low latencies while the latencies from sasquatch to
  weather.renci.org were at their maximum value ** even though the volume
  of data in the EXP feed was roughly half of the volume in the NEXRAD2
  feed **.

  What I am driving at is the problems on sasquatch are causing more
  than a logging load; they are affecting the delivery of data in a timely
  manner.  We first thought that there may be a problem in with the
  network connection from sasquatch to weather.renci.org (e.g., a network
  card that was choked down from 1000Mbps to 100Mbps; a bad network card;
  a problem in a router; etc).  Further investigation did not reveal
  a network problem (e.g., card was running at 1000 Mbps; the card was
  not showing errors; etc.), so we went looking elsewhere.

Cheers,

Tom
****************************************************************************
Unidata User Support                                    UCAR Unidata Program
(303) 497-8642                                                 P.O. Box 3000
address@hidden                                   Boulder, CO 80307
----------------------------------------------------------------------------
Unidata HomePage                       http://www.unidata.ucar.edu
****************************************************************************


Ticket Details
===================
Ticket ID: IIX-893121
Department: Support LDM
Priority: Normal
Status: Closed