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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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.
Mike, I'm convinced that your product-queue was corrupted somehow. Please keep me apprised. > Steve, The same thing happened again yesterday morning but this time I have > log captures to demonstrate what happened. Pqinsert fails when the max > number of slots is reached. I estimated a file rate for the queue and set a > value along with a size in the registry.xml file. I figured it might make > the file management effective to have a product counter that wasn’t much > larger than what I expected for the number of products it would take to fill > the queue. > > In yesterdays case we were receive a larger ingest of smaller files so the > product queue had grown to about 16GB but the count of 65536 slot was reached > and the log file I’m attaching shows that pqinsert failed, then the process > hung up backed up all the other pqinserts and corrupted the queue. This was > the behavior I was expected. I was under the presumption that if the max > queue slots was reached the oldest product was deleted to make room for the > next one. > > To mitigate the problem in the mean time I quadrupled the number of slots in > the registry.xml configuration, rebuilt the queue and will watch and see in > about 4 more days when the queue fills up how LDM does. > > -Mike > > This is line 67486 of the attached log file: > pqinsert ERROR: fb layer ran out of product slots, too many products in queue > > > > > Just an FYI for you. > -Mike Regards, Steve Emmerson Ticket Details =================== Ticket ID: HCO-891524 Department: Support LDM Priority: Normal Status: Closed