-when no pidfile is found by PidfileRunMonitor, just wait, and after a timeout, send email and act as if process failed
-when requeuing a job during recovery, don't just delete old results, instead move them to a temp dir and print it
-always reverify "Running" hosts (not just in host recover mode)
-reorganized recovery code a bit
-remove unnecessary check for queue.log files in clear_results_dir (was leftover from when we had ReverifyTask)
Signed-off-by: Steve Howard <showard@google.com>
git-svn-id: http://test.kernel.org/svn/autotest/trunk@1362 592f7852-d20e-0410-864c-8624ca9c26a4
1 file changed