[Debian-ha-maintainers] Bug#596694: Workaround?

[Dr. Frank Albrecht]

Cite raoul:

"i ended up adding $syslog too and until now, i did not have any  
problem."

I spent a lot of time to fix/workaround the corosync start problem but  
with no luck. I've found your bug report (the shorewell issue) and  
tried to manage the order of init.d starter scripts (adding $syslog  
in .depend.boot resp. changing the order of the rc2.d-symlinks).

But I'm not able to get a consistent behaviour for the corosync start.  
Sometimes it starts correct, sometimes it stucks in a zombie state and  
the only solution is a "killall -s 9 corosync".

The logs are not so helpful. I will link them later here in this  
bugreport thread.


Do you have a workaround? Where did you add $syslog?