Hmmm... I see from the logs that (somehow) it is/was the syslog changes that caused the ssp tests to start timing out - even though the test run after those changes completed (after a very long time - I guess the timeout was increased) those tests all timed out in that run.
Very weird. How could syslog changes affect tests that don't use syslog? That is, unless it is just changes to the layout of libc affecting the ssp stuff ?? kre
