> >Can you provide the schema (the CREATE TABLE and any CREATE INDEX commands) >for that table ?
CREATE TABLE preference_table ( preferenceSet text, /* name of this preference group */ preferenceName text, /* a preference in this group */ preferenceValue text /* sort order of this k...; CREATE UNIQUE INDEX preferenceindex on preference_table(preferenceSet,preferenceName); >Do you have any multi-access things going on ? Two or more computers, >applications, processes or threads trying to access the database at the same >time ? No, but it would be normal for the database to be on a different computer than the sqlite client, and be using whatever networked file system is common. The culprit clients seem to be macs, we're still seeking more information about the specifics. >Does your application check the result code returned from all sqlite3_ calls ? > Not just the ones inserting rows, but also those opening and closing the >file, setting PRAGMAs, or anything else it does with the sqlite3_ library ? Yes. It all goes through a common interface function which is careful about checking. As I said in the original message, this is something that has been working without problems for a few years, the only thing that's changing is the network and OS environment it's deployed in. My hypothesis is that a new failure mode in the file system is tickling a sqlite bug. Based on the evidence available now, a transaction that is trying to insert 4 records fails, and is retried, resulting in 8 records which can't be indexed.
