So, here's the scoop: I exported only the bib records for a range of entries in my catalog so I could change their encoding to UTF-8 and fix my diacritics (they were showing up as a square with a question mark in the middle). I set the matching rules to biblionumber and to replace if there was a match. The matches were all made, but rather than replace the records, they duplicated without items (which I told koha not to export item data and not to process item data, so I'm not terribly upset about that other than the fact that they didn't "replace" the existing records like I thought they would). My problem is, now when I try to undo the import into the catalog in Koha>Tools>Manage Staged MARCRecords>Batch# -- it stalls at 0%. The .mrc file contains 1947 records, but I've been routinely importing and unimporting files with 5000 or so, so I don't think it's filesize. Is there something obvious I'm missing? Could it have something to do with not having cleaned out my reservoir? Please help! Here's the crunchy bits of our koha installation: Koha version:3.12.03.000 MySQL version:mysql Ver 14.14 Distrib 5.1.66, for debian-linux-gnu (x86_64) using readline 6.1Apache version:Server version: Apache/2.2.16 (Debian)Zebra version:Zebra 2.0.44