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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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.
Hi Mark, re: > Tom, I just received work from the NWS that they are running v6.6.3 of > LDM, OK, so there should not be a version-related problem. > so i don't know where that leaves us with finding a reason for the > files being dropped between here and there. > > Any ideas? Since we are not receiving statistics from the NWS site who is indicating a loss of products, we can not verify that they are, in fact, not getting all of the products that they should. The reasons that I emphasize this are: - there is no end-to-end comparisons possible for number of products and volume of data being delivered - we have seen numerous instances of sites reporting that they are not receiving all of the products that they should when they are, in fact, receiving all of the products. In a number of cases, the perception that not all of the products are being received is caused by the pqact.conf processing of products out of the downstream queue not being fast enough to process products before they are overwritten by newly received products. This can happen when the downstream site is using a single 'pqact' to process lots of products (the processing of pqact.conf actions is serialized) and/or when the downstream machine's LDM queue is too small to hold the received products for a period long enough for pqact.conf processing to occur fully So, we must ask the following question again: - How is the NWS site determining that they are not receiving all of the products that they believe that they should be getting? If the answer is that they are not seeing all of the products being processed out of the queue, then the solution might be that they need to breakup their pqact.conf processing into mutually-exclusive pieces that are each handled by a separate instance of 'pqact'. 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: ZOC-298338 Department: Support LDM Priority: Emergency Status: Closed