> I am going to be running a large array.
>
> I will have quotas in the kernel and enabled BUT all users I set quotas
> on will be nowhere near the 2TB barrier I see people talking about
> recently.
>
> HOWEVER, at some point in the future, root or www (or both) users will
> _own more than_ 2 TB of files. They will not have a quota set on them,
> but they will in fact own >2 TB of files.
>
> Is this also a problem ? Or is the only problem actually _setting_ a
> quota larger than 2TB ?
>
> I assume the output in "repquota /my/fs" will be broken, and that is
> fine with me - I just don't want to corrupt or damage my filesystem (or
> existing quotas) the day that my www user goes over 2TB of owned files.
>
> Also, I am distrustful of merely testing this - just because things run
> fine for a day with quotas turned on and some user owning more than 2 TB
> of files doesn't mean it won't blow up at some future date in some
> interesting scenario - and that is why I am asking for opinions here
> rather than just creating >2 TB of files and turning on quotas.
>
> Does anyone out there already do this and can reassure me ?




I haven't heard anything - which is not surprising, since it doesn't sound like many people are using quotas these days.

Does anyone have any general thoughts as to whether this will be dangerous or not ? I know (I assume) that repquota output for my root and www users will be broken, and that's fine - I just want to make sure that as soon as one user goes over 2TB of owned files the filesystem doesn't trash itself.

Can the quota subsystem failing in some way cause data loss / filesystem inconsistencies ?




_______________________________________________
freebsd-fs@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-fs
To unsubscribe, send any mail to "freebsd-fs-unsubscribe@freebsd.org"