Re: Large File Systems - Enough inodes?

[Richard Hector <richard@walnut.gen.nz>]

On 21/05/14 04:24, Sven Hartge wrote:
> Kenneth Jacker <khj@be.cs.appstate.edu> wrote:
> 
>> I am buying two new SATA hard drives:   1TB and  2TB.
> 
>> I'd like to use the 2TB unit for backups (typical Linux directories
>> and files) ... with just a single file system (ext4 most likely).
> 
>> Will 'mkfs' create "enough" inodes?  Or, would it be better to, say,
>> split the 2TB into four 500GB file systems.  Or, some other approach?
> 
> I have in my 15 years as Linux admin only run out if inodes in two
> cases:
> 
>  a) INN2 usenet server with traditional spool which contained a metric
>     sh*t ton of very very small files. Needed to recreate the filesystem
>     with a bytes-per-inode size of 1024.
> 
>  b) squid2 spool directory. Also a motherlode of very small files.
> 
> In all other cases the defaults of mke2fs were sane and no need for
> further tuning was needed. Just look at the inode/byte ratio of the
> filesystems you want to backup. Your destination will show the same
> ratio.

There's another way I've run out; it may mean I've been doing the wrong
thing.

I like to create filesystems relatively small, on LVM, so that any of
them can be grown later, when I find out where the space is needed. But
extending an ext(2|3|4) filesystem doesn't create new inodes, so the
ratio of inodes to space drops, and eventually this is a problem.

> And if you really want to be on the safe side: use XFS.

And that's my solution.

Richard