Yes, this is possible. There are multiple approaches, for example: 1. Creating a snapshot (or outright copy) of a filesystem, then disposing of it when done. (1a) Manually creating a full copy (1b) Using a blockstore-provided snapshot facility such as LVM2 2. Using a snapshot-capable filesystem, and using a snapshot provided by the filesystem itself (ZFS, BTRFS, etc.) 3. Mounting read-only, with either of two solutions for writable portions of the filesystem. This class of solution is very similar to NFS based root situations (ie. modern PXE-driven diskless network boot). (3a) 'tmpfs' or some other in-memory based write solution where required. (3b) Union-mounts.
My advice would be as follows. == simplest == (1a) and (1b) are easiest *and* allow the use of arbitrary filesystems. == medium hassle== (2) is become somewhat common but is more difficult. (3b) are more difficult == more hassle == (3a) is more hassle up front but is perhaps the neatest solution overall. (3b) i have never got working, but should be neat.. it's just not going to be as widely supported by various kernels out there as (3a) or (1b). Personally I use (3a) and (1b). - Walter ------------------------------------------------------------------------------ See everything from the browser to the database with AppDynamics Get end-to-end visibility with application monitoring from AppDynamics Isolate bottlenecks and diagnose root cause in seconds. Start your free trial of AppDynamics Pro today! http://pubads.g.doubleclick.net/gampad/clk?id=48808831&iu=/4140/ostg.clktrk _______________________________________________ Lxc-users mailing list Lxc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/lxc-users
