Re: [PATCH] Reject duplicate disk targets

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 07/11/2013 11:47 AM, Daniel P. Berrange wrote:
[...]
> Huh, that doesn't make any sense. The 'alias' data is not user specified.
> If they attempt to set it in the XML, it'll be dropped by the parser. The
> alias names are assigned by the QEMU driver unconditionally and unless
> that code is broken in some way, you'll never get a duplicate alias
> 

It is.  And that's what i was trying to fix.  But when trying to explain
myself properly in this email, I went through the code few times and
found out this is deeper than I thought.  I thought the addresses were
checked but it looks like it doesn't apply for other than PCI and CCW
addresses.  We should definitely check "drive" and "virtio" addresses as
well and keep them for hot(un)plug and other purposes (probably the same
way as we already do with PCI addresses).

I'll rewrite it from scratch, but just want to point out that for
example for virtio disks, the alias is based just on the 'target'
attribute, which can be duplicated in the XML and that's why I was
trying to add check for the alias as well.

Have a nice day,
Martin

--
libvir-list mailing list
libvir-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/libvir-list




[Index of Archives]     [Virt Tools]     [Libvirt Users]     [Lib OS Info]     [Fedora Users]     [Fedora Desktop]     [Fedora SELinux]     [Big List of Linux Books]     [Yosemite News]     [KDE Users]     [Fedora Tools]