Trace files

[Fred Bone]

Running v2.2.1 for Windows (build 16462) on WinXPPro/SP2.

Once in a while the program produces a huge Trace file. Just now it did 
so. The last three runs generated trace files of 24kB, 19kB, and 32GB. 
The last time this happened it had filled the hard disk before I spotted 
and stopped it (and caused me to lose data in another application).

The earlier trace files start like this:
*   WARN <qof.engine> [guid_init()] only got 1591 bytes.
The identifiers might not be very random.
*   WARN <gnc.backend.file> Invalid timestamp in data file.  Look for a 'trn:date-posted' entry with a date of 1969-12-31 or 1970-01-01.
*   WARN <gnc.backend.file> Invalid timestamp in data file.  Look for a 'trn:date-posted' entry with a date of 1969-12-31 or 1970-01-01.
(there are about 20 of this last)

The huge one starts like this:
*   WARN <qof.engine> [guid_init()] only got 1595 bytes.
The identifiers might not be very random.
*   CRIT <gnc.app-util> gnc_detach_process: assertion `proc && proc->pid' failed
*   WARN <GLib> gmain.c:436: MsgWaitForMultipleObjects() failed: The handle is invalid.
*   WARN <GLib> gmain.c:436: MsgWaitForMultipleObjects() failed: The handle is invalid.
and for all I know to the contrary, this line is repeated throughout the 
file. It doesn't occur at all in the other trace files.

Any thoughts on what I might do to diagnose the problem and prevent a re-
occurrence?

I'd gladly forward the trace file, but even if it compresses by 99.9% I 
don't think my ISP will let it through ...