For the query sequence which is throwing an exception, do they use the same connection, or does each iteration create its own connection to the database? If the latter, does the connection string specifify a database name or a file name? In other words, is it possible that the second update could be fouling when it tries to write to a still-locked file or file segment?
(Caveat: I haven't worked with Ultralite on Android yet, but I'm presuming that a file-based connection may not automatically detect whether it or not it can really write to the file -- I just have vague recollections of Linux growing fur and fangs when it came to mixing file locks and multi-threaded processes.)
Bluesky
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.