On 03/02/2015 06:43 PM, Daniel P. Berrange wrote:
When we start a vm which have rawio = 'yes' settings without any file caps settings for qemu, qemu process still cannot use this caps (CAP_SYS_RAWIO) and the /proc/pidofqemu/status like this:
CapInh: 0000000000020000 CapPrm: 0000000000000000 CapEff: 0000000000000000 CapBnd: 0000001fffffffff
this is because we do not set file caps for qemu (see man 7 capabilities), although laine have mentioned this in commit e11451, i think it will be good if we add this in docs. This is only true if you are starting the guest under the qemu:///session URI. In such a case I think it is expected
On Mon, Mar 02, 2015 at 06:04:44PM +0800, Luyao Huang wrote: that the QEMU lacks rawio capabilities, because the whole point of qemu:///session is that the VM has no elevated privileges.
In the case of qemu:///system libvirt should ensure that it does the right thing with passing on raw io capability flag. If it does not, then we must fix that in the code, not the docs.
Hmm, what i show is the test result in qemu:///system, and we already set the right cap flag before we do execv() or execve(), however we run qemu process in qemu(107) not root(0) in most case, so only set this cap flags cannot make qemu to use this flag, because from capabilities(7): Transformation of capabilities during execve() During an execve(2), the kernel calculates the new capabilities of the process using the following algorithm: P'(permitted) = (P(inheritable) & F(inheritable)) | (F(permitted) & cap_bset) P'(effective) = F(effective) ? P'(permitted) : 0 P'(inheritable) = P(inheritable) [i.e., unchanged] where: P denotes the value of a thread capability set before the execve(2) P' denotes the value of a capability set after the execve(2) F denotes a file capability set cap_bset is the value of the capability bounding set (described below). So if not set any file cap to qemu program (/usr/libexec/qemu-kvm), the qemu process will get this cap flags: Uid: 107 107 107 107 Gid: 107 107 107 107 ... CapInh: 0000000000020000 CapPrm: 0000000000000000 CapEff: 0000000000000000 CapBnd: 0000001fffffffff and qemu process do not have this cap as the CapEff is for kernel do permission check. I think libvirt already do the right things here although running qemu process do not have rawio capability flag in this case, because i think it is not a good idea for libvirt to set file cap to qemu program, libvirt is not the only user which use or call qemu program, set a file cap to qemu program will affect other callers (although set a small file cap will not be a big deal :) ), so i guess maybe it is good to make the users to set this instead of libvirt use cap_set_file() to do this. BTW, if we make qemu process run with root(0) uid and gid, the cap flags will like this: ... Uid: 0 0 0 0 Gid: 0 0 0 0 ... CapInh: 0000000000020000 CapPrm: 0000000000020000 CapEff: 0000000000020000 CapBnd: 0000000000020000
Regards, Daniel
Thanks, Luyao