Hi there, There is a good note on mos about troubleshooting grid infrastructure startup- the title is very similar to this. Ohas uses the init facility and does not immediately branch into your grid home. You might find something in /var/log/messages, have a look at the messages post the restart. Is the file system where the grid binaries are located mounted read write, and do permissions on the files match what they were before? Again the note is where you should start looking-it is useable for a change. Martin Bach Oracle Certified Master 10g http://martincarstenbach.wordpress.com http://www.linkedin.com/in/martincarstenbach ----- Reply message ----- From: "Stefano Cislaghi" <s.cislaghi@xxxxxxxxx> Date: Wed, Aug 25, 2010 20:51 Subject: Debug CRS-4000 failed start with ohasd To: "Oracle L" <oracle-l@xxxxxxxxxxxxx> Hi all, today when the time to go home was coming one development system crashed. Not so bad, a reboot should solve the kernel panic caused from QLogic module. System have been restarted fine but oracle haven't been able to come up. On console there was error *CRS-4000: Command Start failed, or completed with errors.* My environment is Oracle Linux 64bit 5.4 and oracle grid infrastructure 11.2.0.1. I've done some search without finding any interesting docs. Now I'm at home and I'm trying to figure out what I've to check tomorrow... - Log file ohasd.log is empty. Do not contain any information regarding a failed startup. - Suggested action from oracle to run 'rm -rf /var/tmp/.oracle /tmp/.oracle /usr/tmp/.oracle' was performed without solving anything. - Suggested action from Metalink to check '/usr/local/bin/perl' was done, but didn't solved I can check note: *1083982.1 **Incorrect Ownership and Permission after Relinking or Patching 11gR2 Grid Infrastructure* but this was a fresh install, not an upgrade and never patched. AFAIK There's no way to debug the init.ohasd script Oracle support refers many time to *How to Proceed from Failed 11gR2 Grid Infrastructure (CRS) Installation [ID 942166.1] *but there's no way to debug? seems bad! Any debug hint for this trouble? I've also found patch 8740030 that should fix a problem with 'ohasd stop' and is 280mb! Seems to be too big...how many hidden problem are solved with this patch?! Thanks all Stefano -- http://www.stefanocislaghi.eu