Steven,
>Date: Sun, 21 Mar 2004 11:27:43 -0500
>From: Steven Lazarus <slazarus@xxxxxxx>
>Organization: Florida Institute of Technology
>To: ldm-users@xxxxxxxxxxxxxxxx
>Subject: data feed problems
The above message contained the following:
> Can anyone assist with this problem? I'm sure others have seen this. We
> are trying to get ldm running on a linux box here at Florida Tech but are
> have problems grabbing the data for some reason. We seem to able to see
> the feed from FSU...but can't seem to grab it!
>
> Thanks ahead of time for any help on this one!
>
> Steven
>
> Dr. Steven M. Lazarus
> Florida Institute of Technology
> Department of Marine and Environmental Systems
>
> ************************************************************
>
> Note that the following gives results:
>
> [ldm@cumulus ~/bin]$ ./notifyme -vl - -h pluto.met.fsu.edu
> Mar 21 16:00:56 notifyme[14454]: Starting Up: pluto.met.fsu.edu:
> 20040321160056.730 TS_ENDT {{ANY, ".*"}}
> Mar 21 16:00:56 notifyme[14454]: Connected to upstream LDM-5
> Mar 21 16:00:56 notifyme[14454]: NOTIFYME(pluto.met.fsu.edu): reclass:
> 20040321160056.730 TS_ENDT {{FSL2|UNIDATA, ".*"}}
> Mar 21 16:00:56 notifyme[14454]: NOTIFYME(pluto.met.fsu.edu): OK
> Mar 21 16:00:56 notifyme[14454]: 3282 20040321160056.737 HDS
> 71367571 HRKB99 KWBC 211200 /mAVN
> !grib/ncep/AVN/#039/200403211200/F006/RH/1000 mb/
> Mar 21 16:00:56 notifyme[14454]: 3282 20040321160056.754 HDS
> 71367572 HRJB99 KWBC 211200 /mAVN
> !grib/ncep/AVN/#038/200403211200/F006/RH/1000 mb/
> Mar 21 16:00:56 notifyme[14454]: 3282 20040321160056.770 HDS
> 71367573 HRLB99 KWBC 211200 /mAVN
> !grib/ncep/AVN/#040/200403211200/F006/RH/1000 mb/
The above indicates that Cumulus is allowed to receive all data-products
from Pluto that have feedtype FSL2 or UNIDATA.
> Below are comments from FSU:
>
>
> Subject: Re: data feed problems
> Date: Fri, 19 Mar 2004 14:19:01 -0500
> From: Bret Whissel
> To: Steven Lazarus
> References: <405B1E0C.4010404@xxxxxxx>
> <1079715482.4370.1.camel@xxxxxxxxxxxxxxxxxxxxx> <405B3A64.2040405@xxxxxxx>
>
>
>
> Steven,
>
> Below is a trace of the ldmd.log file pertaining to cumulus. I don't
> know what would cause a 'pq_sequence failed: errno=5', but it's just a
> generic I/O error. I'm not sure what the "Unable to receive" error
> means. Is your queue file writable? Is the disk partition mounted r/w?
>
> Bret
>
> Mar 19 14:06:58 pluto cumulus(noti)[13998]: Starting Up(6.0.14/5):
> 20040319140541.299 TS_ENDT {{FSL2|UNIDATA, ".*"}}
> Mar 19 14:06:58 pluto cumulus(noti)[13998]: topo: cumulus.dmes.fit.edu
> FSL2|UNIDATA
The above indicates that Pluto received a NOTIFYME request from Cumulus
for data-products of type FSL2 or UNIDATA. A NOTIFYME request results
in metadata being sent -- but not data.
> Mar 19 14:07:03 pluto cumulus(noti)[13998]: YVNA50 KWBE 191200 /mETA_84
> !grib/ncep/ETA_84/#207/200403191200/F000/VGRD/500 mb/ : RPC: Unable to
> receive
> Mar 19 14:07:03 pluto cumulus(noti)[13998]: pq_sequence failed:
> Input/output error (errno = 5)
The above indicates that the upstream LDM on Pluto that was created by
the previous NOTIFYME request did not receive an acknowledgement of an
RPC NOTIFICATION message that was sent to the downstream LDM on Cumulus.
The reason for the failure is unknown but might be found in the LDM
logfile on Cumulus.
> Mar 19 14:35:38 pluto cumulus(noti)[14018]: Starting Up(6.0.14/5):
> 20040319143421.143 TS_ENDT {{FSL2|UNIDATA, ".*"}}
> Mar 19 14:35:38 pluto cumulus(noti)[14018]: topo: cumulus.dmes.fit.edu
> FSL2|UNIDATA
> Mar 19 14:36:02 pluto cumulus(noti)[14018]: SXUS67 KWOH 191431 /pRRSVEF:
> RPC: Unable to receive
> Mar 19 14:36:02 pluto cumulus(noti)[14018]: pq_sequence failed:
> Input/output error (errno = 5)
> Mar 19 14:44:44 pluto cumulus(noti)[14026]: Starting Up(6.0.14/5):
> 20040319144327.242 TS_ENDT {{FSL2|UNIDATA, ".*"}}
> Mar 19 14:44:44 pluto cumulus(noti)[14026]: topo: cumulus.dmes.fit.edu
> FSL2|UNIDATA
> Mar 19 14:44:45 pluto cumulus(noti)[14026]: SDUS81 KLWX 191440 /pDPALWX:
> RPC: Unable to receive
> Mar 19 14:44:45 pluto cumulus(noti)[14026]: pq_sequence failed:
> Input/output error (errno = 5)
The above indicates that the previously-mentioned failure mode continues
to exist.
I suggest looking at corresponding entries in the LDM logfile on Cumulus.
Regards,
Steve Emmerson