On Tue, 07 Oct 2008 17:35:09 +0200 Doron Shoham <dorons@xxxxxxxxxxxx> wrote: > FUJITA Tomonori wrote: > > On Mon, 06 Oct 2008 16:59:08 +0200 > > Tomasz Chmielewski <mangoo@xxxxxxxx> wrote: > > > >> Right now, tgt-admin looks into /etc/mtab, /proc/swaps etc. to determine > >> if the device it wants to allocate is in use. This is a good way to > >> prevent the user from corrupting filesystems (i.e., mounting the device > >> locally to correct a config file, forgetting to unmount it, and > >> allocating it as a target), but unfortunately, it doesn't check all > >> kinds of "device-in-use" scenarios. > >> > >> The following patches add a full support for detecting "used devices": > >> - used by the system: mounted, swap devices, parts of RAID array, LVM, > >> dm-crypt, etc. > >> - used by userspace, i.e. dd, e2fsck, existing tgtd target etc. > >> > >> For "userspace" tests, lsof is needed. If it's not present, a warning > >> will be shown, informing the user that we can't check reliably if the > >> device is in use. > >> > >> > >> The check can be overridden by using --force command line option or > >> "allow-in-use yes" config option. > > > > Sounds a reasonable option. > > > > Doron, you wrote the patch to check device in use? Any comments on > > this? > > -- > > This sounds very reasonable but I didn't managed to test it yet. Thanks, I just wanted to know if you have objections against it (like the fundamental idea is wrong, or there is a better name than 'allow-in-use', etc) Working or not working is a different topic. If you find it's broken, please send a patch to fix it. :) Thanks, -- To unsubscribe from this list: send the line "unsubscribe stgt" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
