On Tue, May 25, 2010 at 11:49:09AM +0100, L A Hurst wrote: > Hi Guys and Gals, > A couple of days ago I setup Spacewalk 1.0 on a brand new 32-bit CentOS 5.5 > install following the instructions on the wiki (including installing the > exact version of Oracle described). I have not yet had chance to register any > machines in it or setup any configuration channels etc. > > This morning I discovered that the machine had Kernel panicked, caused by one > 'oracle' process. Nothing was logged, unfortunately. >
Hello Laurence, have you got a photo of the panic screen uploaded something which we could examine for more details? The fact that oracle was scheduled on that CPU does not necessarily mean that the panic was caused by that process. Was there something related to the OOM killer in the log? Is this a single core or multi core system? > SELinux is enabled and I installed all the SELinux policy rpms as described > on the wiki, but could this be the root cause? > Well, obviously having our SELinux modules could change the behaviour of the underlying system for the Oracle software, so it could have taken path that it would normally not get to. So while troubleshooting, one option is definitely to try disable various systems on the machine, including SELinux. Are you able to reproduce the situation? -- Jan Pazdziora Principal Software Engineer, Satellite Engineering, Red Hat _______________________________________________ Spacewalk-list mailing list [email protected] https://www.redhat.com/mailman/listinfo/spacewalk-list
