Lukas Ertl wrote: > how hard would it be to implement resizing of mounted filesystems? > Currently, growfs requires the filesystem to be unmounted, and this is > definitely a showstopper for FreeBSD when it comes to production use. > > I'd really like to promote FreeBSD more in my organisation, where we > currently use mostly AIX, and I often hear (and have to say that it's > true) that the AIX LVM is so robust, stable and quite easy to use. > > Could this feature be implemented once FreeBSD 5.0 is out with its > filesystem snapshot?
Nearly impossible, without a JFS. You would need to be able to add new PP's to an LP, as you can do on AIX, or assign PP's to a "hog" partition, and them provide each LP with "hog limits", so that they can allocate PP's to themselves automatically, as needed, up to some high watermark. While it should be technically possible to modify Vinum/ccd/GEOM so that, if you start with a logical instead of a physical partition that is intermediated by one of those technologies, you could grow the size of the logical partition while the system is active with a few small code changes (or a lot of them, in the GEOM case), you would still need to inform the FS of the additional space, and deal with the consequences of a size change on the FS (e.g. "defrag" FFS after you are done growing it). The problem is that the allocation space is spread over all cylinder groups, effectively as a hash. This is the same reason it is recommended that you backup and restore to "defrag" when you run "growfs". -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
