We are seeing intermittent problems with the calls to threaddb in solaris failing. We are not sure of the reason for the failure as it is not being logged as of now. We need to add failure code to the exception messages when there is a failure. Next time when it fails, we should be able to know the reason.