On Thu, Oct 31, 2002 at 09:19:47AM -0500, Dave wrote: > On Wed, 2002-10-30 at 21:03, Herbert Poetzl wrote: > > > > >> IMHO, It should not be possible for a context to exceed it's quota when > > >> some users have not. This is the point of quota mechanism. Guarantee > > >> space on the disk and not allow for over-booking. Allocated user quota > > >> should be subtracted from the total context quota, so that any users > > >> with no quota should not be able to use that space. So in a way, users > > > > hmm, that might be the original idea of quota, but > > all current implementations do not guarantee, but only > > limit the maximum available resources ... > > > > if you want to guarantee, you then simply must do the > > math an make sure that enough physical disk space is > > available (or in the context case, the context quota > > lies above the sum of all user quotas) > > If you do this, then the root user of the context will be able to raise > its allocated quota as much as he likes, by just increasing the quota of > any user he likes. The context quota will always be added to that.
as for now, I can ensure you, that the context quota is a limit like the group or user quota, which will not be added to some other quota information. > The correct behavior should be to subtract any allocated quota from the > allocated context quota. If you want to see it this way. I prefer to see it as being accounted independently (as the current implementation does) so if an inode with uid/gid/ctx is allocated or some inode is changed to another uid/gid/ctx setting or an allocated inode is freed the user/group/context quotas are adjusted accordingly (not more not less) > If you run a quotarep statistics on all users, you'll see at the bottom > some summary showing you > > -) Total Available Stace > -) Allocated Quota > -) Quota left for allocation > > When you allocate quota to a user, that is subtracted from the quota > left for allocation value. Am I missing something? > > So for Vserver to support quota "properly", you need to guarantee the > space allocated to each context or accept the consequences (like I > explained on another reply). agreed again. > Dave. > best, Herbert
