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

20050902: Think we can look at sasquatch sometime?



>From: Gerry Creager N5JXS <address@hidden>
>Organization: Texas A&M University -- AATLT
>Keywords: 200509021505.j82F56jo020896 IDD SCOOP

Hi Gerry,

>Still looks like sasquatch is unhappy.  Is there a chance we could get 
>on the phone and look at him again?
>
>http://my.unidata.ucar.edu/cgi-bin/rtstats/iddstats_nc?EXP+sura-uf-pe6600-1.co
> astal.ufl.edu

I jumped onto sasquatch a few minutes ago and have the following comments:

- the machine configuration looks very good:

  dual Opteron
  4 GB queue
  8 GB RAM

- please load Tcl/Tk and iostat (like on bigbird) so we can run the uptime.tcl
  monitoring

- the overall ingest and relay configuration looked good, but I decided
  to tweek the request lines in ~ldm/etc/ldmd.conf:

  - combine all feed requests from bigbird into a single request line

- there is something amiss with processing of MADIS data:

Sep 02 15:52:01 sasquatch pqact[2619] ERROR: pbuf_flush (3) write: Broken pipe
Sep 02 15:52:01 sasquatch pqact[2619] ERROR: pipe_put: 
-closebin/store_MADIS.shdata/LDAD/mesonet/netCDF20050902_1400.gz write error
Sep 02 15:52:01 sasquatch pqact[2619] ERROR: pipe_prodput: trying again
Sep 02 15:52:01 sasquatch pqact[2619] ERROR: pbuf_flush (3) write: Broken pipe
Sep 02 15:52:01 sasquatch pqact[2619] ERROR: pipe_put: 
-closebin/store_MADIS.shdata/LDAD/mesonet/netCDF20050902_1400.gz write error

- mesodata3 is trying to request data from sasquatch but is being continuously
  denied:

Sep 02 16:08:49 sasquatch rpc.ldmd[12101] NOTE: Denying connection from 
"mesodata3.TAMU.EDU" [165.91.140.24]
Sep 02 16:08:54 sasquatch sura-uf-pe6600-1[12110] NOTE: Switching data-product 
transfer-mode to primary

At the current time, I have no explanation for the sawtooth pattern for
the EXP latency from sasquatch on sura-uf-pe6600-1.coastal.ufl.edu.
BTW, I see the same sawtooth latency pattern on products received from
sasquatch -- but much worse -- on ldm.itsc.uah.edu.

My first inclination is to downgrade the LDM on sasquatch to LDM-6.3.0
and see if that makes any difference...  Before downgrading, and since
the LDM-6.3.0 code was built back in may, I wanted to rebuild LDM-6.3.0
from source:

<as 'ldm'>
cd ~/ldm-6.3.0/src
make distclean
./configure

At this point I am seeing some flags being set for the eventual make
that seem to indicate that the build environment may not be quite
correct.  I will need to discuss this with Steve Emmerson to see if I
am misinterpreting something (I'll get back to you).

Please let me know when Tcl/Tk and iostat have been installed on
sasquatch so I can turn on performance monitoring (like I did on
bigbird early this morning).  I am at home at the moment, but
will be heading into work momentarily.

Cheers,

Tom
--
NOTE: All email exchanges with Unidata User Support are recorded in the
Unidata inquiry tracking system and then made publicly available
through the web.  If you do not want to have your interactions made
available in this way, you must let us know in each email you send to us.