Long operations tend to kill the postmaster

Hello!

I'd like to ask if there are some known issues with time consuming
operations that lead to killing the postmaster service? I am trying to
migrate a database from oracle to postgres via odbc and pgMigration and
everything seems to go along fine, the first few tables coming being
migrated successfully, until I get to a table with some 200k+ rows. It's
taking ages and at some point an error message appears stating that it
couldn't communicate with the postmaster. So I take a look at the
services manager and right enough, postmaster is not running any more.

The very same thing applies to some import jobs we are running in
ColdFusion. Everything seems to work fine, but then on uncertain
occasions when there are very time consuming operations running, the
postmaster just seems to quit. It's not becauso of some certain query,
it seems to be related to the running time of an operation.

Are there any known issues? Could it be that I have misconfigured
anything which could lead to this behaviour? I did a standard full
install of Cygwin's latest snapshot and followed the instructions in
order to get the database running - and it does indeed run for the most
part. But for a production database even seldom failures can be quite
inacceptable.

Any ideas?

Thanks,

Markus