Hey Robb,
I noticed some unusual behavior out of metar2nc.new on lead1 recently. First I noted these in ldmd.log: Aug 09 10:10:43 lead1 pqact[4803]: pbuf_flush 7: time elapsed 3.919847
Next I noted that 4803 is the process id associated with surface metar. It is still running both the old and the new metar decoders. The metar.new log file showed a bunch of the following messages:
ncvarget: ncid 3; varid 1: Edge+start exceeds dimension bound NetCDF::recget status = -1
Perhaps these two are correlated...
Thanks for looking into it. Tom.
Tom Baltzer Software Engineer UCAR Office of Programs
E-mail: address@hidden Phone: 303-497-8637
|