Re: [PATCH] LSPP audit enablement: storing selinux ocontext and scontext
On Thursday 28 July 2005 10:49, Steve Grubb wrote:
On Thursday 28 July 2005 11:27, Timothy R. Chavez wrote:
> But audit_panic() doesn't just panic the system or it doesn't have to at
> least. You're able to set the 'audit_failure' such that when
audit_panic()
> is called it can fail silently, print to syslog, or panic the system.
Right, but audit_panic is reserved for use when the backlog overflows or rate
limit is too high. When you wrote the fs watch code, did you call audit_panic
when a buffer alloc failed? No, you failed the syscall. The behavior should
be consistent.
Yeah, I'm thinking I should be calling audit_panic() instead. I think that the
correct way to handle such errors that occur _in_ the audit sub system. Also,
I fail silently if the audit_watch_notify() hook is called and it can not allocate
an memory to place on the audit context... the admin has no control over this.
At least with audit_panic they can specify how they want the system to fail.
Let's say our system is extremely stressed out and we fail a memory
allocation that's nonfatal... that record is missed but the system is still running
when really the system should probably panic (in a CAPP environment at least).
That's not presently doable with the fs watch code (should I be "eeking"
right
now).
I suppose in a CAPP environment if a record cannot be generate the system
should be halted, right? Well with audit_panic the administrator has control
over that behavior.
-tim
-Steve
--
Linux-audit mailing list
Linux-audit(a)redhat.com
http://www.redhat.com/mailman/listinfo/linux-audit