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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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.
Patrick, > Yes those are two separate errors. I guess we hit the lottery! Since we > have two errors with the same signature, and it's astronomically > improbable to get two different products with the same signature, it would > follow that it's the same product, no? Or it might have > something to do with the odd-looking checksum, maybe it's lost or garbled at > some point. The two data-products either have identical data (unlikely) or the same MD5 checksum (unbelievably unlikely for large files). In either case, the product-queue appears to be corrupt to the code. The likely culprits are disk error and memory error. > It doesn't appear we are using the -i option of pqinsert. The files vary in > size widely, from a few hundred KB to 20 MB or so. I spot > checked a few hours and it looks like we insert about 2500 products per hour > into the queue. > > Patrick Regards, Steve Emmerson Ticket Details =================== Ticket ID: GRA-463091 Department: Support LDM Priority: Normal Status: Open