On 10/29/2013 10:32 PM, Normand Mongeau wrote:
Hmm really?
Odd thing is that although I do have a 10 second timeout as soon as C goes
into a begin transaction A receives the SQLITE_BUSY error, in other words I
don't see any 10 second delay.
This, too, is consistent with my diagnosis. When SQLite detects the
deadlock situation I described, it starts returning SQLITE_BUSY right
away, without waiting for a busy timeout or calling a busy handler.
SQLite knows that in this situation it's pointless to wait - no progress
will be made. Only a rollback helps.
--
Igor Tandetnik
_______________________________________________
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
