I ran into this situation with a new cluster just a couple of days ago. At first I just commented it out but that didn't work for me in vipca. I looked at the MOS (Metalink) node and it says to add a line after the if/then structure to unset it. I couldn't see what the difference was but I was stuck. Sure enough, that worked and my cluster is running fine now. On Sat, Dec 4, 2010 at 6:31 AM, Harel Safra <harel.safra@xxxxxxxxx> wrote: > Mostafa, > vipca and srvctl set the ld_assume_kernel in the script itself. You need to > comment out the lines that set the variable in the script. > You can find details here: > http://mpoojari.wordpress.com/?s=vipca > > Harel Safra > > > On 04/12/2010 12:14, Mostafa Eletriby wrote: > > Hello DBAs, > > I have a problem at running VIPCA manually at installing RAC 10g > environment > > Platform: OEL 5.2 > RAC: 10gR2 > > [root@rac2 bin]# pwd > /u01/crs/oracle/product/10.2.0/crs/bin > [root@rac2 bin]# ./vipca > > /u01/crs/oracle/product/10.2.0/crs/jdk/jre//bin/java: error while loading > shared libraries: libpthread.so.0: cannot open shared object file: No such > file or directory > > [root@rac2 bin]# > > I have checked existence of LD_ASSUME_KERNEL variable in my environment > (after unsetting it): > echo $LD_ASSUME_KERNEL > > And output is black which assure that it is unset. > > Also I followed this document that may solve the problem but no solution. > > http://forums.oracle.com/forums/thread.jspa?messageID=1615978� > > It is a bug at oracle & has many workarounds as I see. > Please try to solve my problem, & advice me how to undo any changes I made > at environment files. > > Thanks a lot. > >