I got the message that follows and am sending it to this list to ask
whether anyone has seen a sweep corrupt the database since
InterBase V5.5.

Thanks,

Ann



Hi Ann,

I am a novice Firebird user. I first started with Interbase in version
4. When my company started using IB 5.5 about 2 years ago we ran into the
sweep bug that is documented in posts all over the newsgroups. We were
able to restore our database with the steps provided. This was a very long
time ago and we made the move to open source v6 and then finally to
Firebird v1. We have been very happy with the performance of Firebird and
are really looking forward to v2 & SMP support.

In doing some rewsearch on why my database was becoming sluggish after many
weeks of constant operation, I ran into a post that indicated that a sweep
interval set to 0 would cause this sluggishness to happen. The interval
got set to zero when we recovered our database from the bug mentioned
above. It was documented that this bug was fixed in v6, so I assumed it
was fixed in Firebird v1. So, I set my sweep interval to 20,000 and did a
backup and restore on my database and expected it to do the automatic
sweep. It did, and when it did it corrupted my database. I had to drop
the table that had the page with the bad checksum and backup and restore
the database, then pump the records from the corrupt table into the new one
up to the bad page. I am running XP on a 2.4 Ghz P4. I would really
appreciate it if you could recommend a solution for me in setting my sweep
interval or provide me with an answer as to why this bug still seems to be
an issue.

Thanks in advance,

Phillip Day