Subject Deadlocks + DB Corruption
Author Thomas Ellis
Hi,

We're having some strange corruptions that I'm hoping someone can help with.
This is the second corruption that the client has had in a month the
previous one we wrote off to a crash. We did and backup and restore and
things have been running fine for 2 weeks. On the 11th the connections
started getting "internal gds software consistency check (can't continue
after bug check)" so we had a look at the log file and found that the
previous day there was a strange "deadlock" message in the Firebird log
file.

Questions:
What does "deadlock" without any additional info mean?
Does anyone have any ideas what could be causing this?
Where do I start to try and analyze/debug this?

OS: Windows Server 2008 Standard FE 64bit
FB: 2.1.3 32bit Superserver
Server & firebird service has had 100% uptime since last backup/restore on
25 Nov.
+- 300GB free on both HDDs
Forced writes is on (although the server/service has not crashed so...)
OS drive cache is on
(DB has since been backed up and restored and everything is fine)

Log file extracts:
SERVER1 (Server) Fri Dec 10 13:29:44 2010
Database: C:\PROGRAM FILES (X86)\PROGRAM\DATABASE.FDB
Deadlock

.....

SERVER1 (Server) Sat Dec 11 08:39:37 2010
Database: C:\PROGRAM FILES (X86)\PROGRAM\DATABASE.FDB
deadlock
internal gds software consistency check (error during savepoint
backout (290), file: exe.cpp line: 4043)

SERVER1 (Server) Sat Dec 11 08:39:37 2010
Database: C:\PROGRAM FILES (X86)\PROGRAM\DATABASE.FDB
deadlock

SERVER1 (Server) Sat Dec 11 08:39:37 2010
Database: C:\PROGRAM FILES (X86)\PROGRAM\DATABASE.FDB
deadlock

SERVER1 (Server) Sat Dec 11 08:39:37 2010
Database: C:\PROGRAM FILES (X86)\PROGRAM\DATABASE.FDB
Deadlock

Thanks
Thomas Ellis