Good morning, all! Under the category of "Well, that didn't take long".....I'm hoping someone might be able to provide insight into this issue! CONFIGURATION: koha 3.20.04 on Debian 8.2 Using the "new indexer" (not quite sure what to call it--my apologies!) SCENARIO: I'm loading records through the staged MARC records process in Tools. Many of the records loading find matches and are successfully overlaid as reported in the final "Manage staged MARC records" report. New records are being created, as needed, based on the failure of the matching rule to find a match. There are links via the new record numbers to these records. However...none of these new records are indexing. The only way to get to them is via the record number link in the staged MARC record results. Attempts to call them up in any other manner in either the staff or public interfaces fails. COMMENTS: I'm working with files of 10,000 records. Over 99% are overlays--in a file this size, I might have 50 new records. I'm applying a different framework (one I created based on the default) to the new records than to the ones being overlaid, The bibliographic framework test does still pass. The new framework is being applied to the new records (and the old ones that are just overlaid are keeping the framework selected at the time of their creation) I'm confident the "new indexer" is configured correctly, as I changed the indexing process midway through the first load of files (in which all records were new and no matching algorithm was selected) and those records were all properly indexed at the point of creation. At first I thought I just wasn't being patient enough, so I monitored top until it reported that most system resources were again idle (this being an indication to me, right or wrong, that indexing had basically caught up). Nope. So then I waited overnight, because "why not??" Nope. I tried editing one of the new records (specifically an indexed field--the 245) to see if making this change might force indexing to take place. No such luck. The change is reflected, though, when continuing to access the record via the record number link in the staged MARC record results list. I'm not working in a production environment...I've got the server running off of a VM at home with public access over port 8080 and staff access over port 8081, using a single-server configuration (no separate machine for mySQL). Because there is absolutely no personal data on this server and never will be, I have no problems opening it up to anyone that might want to take a look. I'm hardly wed to the new way of indexing. Given the eventual use of this demonstration server, the "old indexing" method will work just fine for my purposes--but if this happens to be representative of a larger problem that does need to be reported for investigation I wanted to post this to the mailing list. I hope I've provided enough information for someone to weigh in with what might be happening, and a suggestion for how to address it. I briefly searched the bug list and didn't really see anything there that jumped out at me. Thanks in advance to all that might weigh in with things to investigate! John -- -=<*>=- -=<*>=- -=<*>=- -=<*>=- -=<*>=- -=<*>=- -=<*>=- -=<*>=- -=<*>=- John E. Sterbenz, Jr. Associate Librarian Manager, Technical Services, Collections, and Library Automation 701 Tappan Street, Rm. K3359 Kresge Business Administration Library The University of Michigan Ann Arbor, Michigan, USA 48109-1234 TEL: (734) 764-5746 FAX: (734) 764-4162