lilo and initrd
Hi,
today I read in the SELinux README file, I should not use the initrd= option with the selinux entry. Knows anybody why?
Thanks
Carsten
--
This message was distributed to subscribers of the selinux mailing list. If you no longer wish to subscribe, send mail to [email protected] with the words "unsubscribe selinux" without quotes as the message.
Quoted 2 lines · Carsten Grohmann
today I read in the SELinux README file, I should not use the initrd= option with the selinux entry. Knows anybody why?
initrd creates extra complications that you need to think about order into get a working system, eg. having a working policy on the itrd infilesystem.
Unless you enjoy or especially want to debug initrd filesystems, etc, it might be wise to avoid it. ;-).
I don't think there are any inherent problems though...
Signature · 6 lines
Brian May <[email protected]> -- This message was distributed to subscribers of the selinux mailing list. If you no longer wish to subscribe, send mail to [email protected] with the words "unsubscribe selinux" without quotes as the message.
Quoted 2 lines · Carsten Grohmann
today I read in the SELinux README file, I should not use the initrd= option with the selinux entry. Knows anybody why?
The initrd has always worked well for me. I use initrd on all my systems except Cobalt machines (because the Qube BIOS has no method for loading the initrd).
I have had the policy be loaded from the initrd for old-style SE Linux, and apparently this may be required for 2.6.x. In any case there are definately no problems.
Signature · 10 lines
http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page -- This message was distributed to subscribers of the selinux mailing list. If you no longer wish to subscribe, send mail to [email protected] with the words "unsubscribe selinux" without quotes as the message.
Quoted 2 lines · Carsten Grohmann
today I read in the SELinux README file, I should not use the initrd= option with the selinux entry. Knows anybody why?
That is an old note that is no longer accurate, even for the old 2.4-based SELinux. Prior to 2.4.19, if you booted a SELinux kernel with an initrd, the kernel would try to perform the initial policy load from the initrd itself, which would fail unless you had installed a policy onto the initrd. The old 2.4-based SELinux should work fine now with an initrd; the SELinux kernel will not try to load the policy until the real root filesystem has been mounted.
In the 2.6-based SELinux and the new 2.4-based SELinux (back port of the 2.6-based SELinux), the initial policy load has been pushed into userspace, and an initrd is specifically used for this purpose. We have patched our mkinitrd script to copy load_policy and the binary policy file onto the initrd and to include commands in the /linuxrc script to load the policy from the initrd before the real root filesystem has been mounted. See the README in selinux-doc.
Signature · 8 lines
Stephen Smalley <[email protected]> National Security Agency -- This message was distributed to subscribers of the selinux mailing list. If you no longer wish to subscribe, send mail to [email protected] with the words "unsubscribe selinux" without quotes as the message.