Date: Thu, 11 Feb 1999 21:57:03 +0000 (GMT) From: Terry Lambert <tlambert@primenet.com> To: casper@acc.am (Casper) Cc: freebsd-hackers@FreeBSD.ORG Subject: Re: Quotas implementation Message-ID: <199902112157.OAA14970@usr06.primenet.com> In-Reply-To: <36C2A5E9.675D0B71@acc.am> from "Casper" at Feb 11, 99 01:42:01 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> Are there another ways to implement quota mechanism except that we > use now - quota checks on low-level IO operations (when manipulating > with disk inodes)? > > It's impossible to use (or it will be very nasty hack) existing > FFS/UFS code for my purposes , because no real uids will be created > and stored on disk ... The only way i see , is to patch every > daemon that will originate this IO operations. > > Any ideas ? It would help to know the application. It's a very strange request; quotas are enforced against user ids because they are accounted by credential. If you don't have credentials, then, by definition, really you can't distinguish entities using disk space. If you can't distinguish entities, then everyone is the same. If they are all the same, then there is only one resource consumer, and quotas are irrelevent. So without uids, you will need to invent them anyway, to distinguish enforcement zones. You realize that you can enforce quotas against groups instead of uids, right? There are many other (better) ways to implement quotas, rather than wiring them into the bowels of UFS, but *all* of them depend on some type of credential being associated with VOP's that allocate or deallocate disk space on the quota-ed area. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199902112157.OAA14970>