Hello,

I am using embedded firebird version 2.0.3.12981 in my windows application. I have found that if multiple deletes e.g. 50,000 from 100,000 rows is performed then there is a long delay processing a subsequent select statement. This delay may stretch to 10mins or so. Is this delay due to the embedded firebird server performing inline garbage collection therefore the select statement request is waiting for the garbage collection to complete before being requested?? It does seem that the delay is dependent on the number of rows being deleted and the size of the current database. Has anyone else experienced such behaviour. I am wondering whether or not to change to an non embedded version that I would suspect would remove such an issue. It would be good to hear if this is an embedded server related problem before switching to a superserver.

Thanks for any help,

Kindest Regards,

Matt Finch 




[Non-text portions of this message have been removed]