[Nottingham] *Anyone* _ever_ use "atime"?...

[Martin]

Folks,

Does /anyone/ or /anything/ make use of "atime"?

There's the old one special example of Mutt for only when using mbox for
your mail. (Mutt does not need atime when using maildirs.) Other than
that, is there any need or good use for atime?

So... No problem for always setting all filesystem mounts to use noatime?...



Background:

I've just stumbled across the old detail of "atime" again for
filsesystem mounts...

For a few years now, the less strict "reltime" is commonly in use
whereby the "access time" noted for a file or directory is only updated
if the existing atime is over 24 hours old. That sort of keeps the atime
alive but reduces disk updates (writes) for repeat file accesses.

To reduce SSD wear, usually "noatime" is set so as to stop all updates
to the SSD for when merely reading data.

I've held onto reltime only for "just-in-case" and also just in case I
might ever want to survey file use for the sake of archiving. However,
with the ever rapidly increasing size of disk capacities, and my
not-so-quickly increasing data use, I've never needed to worry about
archiving!

So... Is atime an old anachronism that never gained useful use?

http://en.wikipedia.org/wiki/Atime_%28Unix%29#Criticism_of_atime


Use "noatime" always?

Cheers,
Martin

-- 
- ------------------ - ----------------------------------------
-    Martin Lomas    - OpenPGP (GPG/PGP) Public Key: 0xCEE1D3B7
- martin @ ml1 co uk - Import from   hkp://subkeys.pgp.net   or
- ------------------ - http:// ml1 .co .uk/martin_ml1_co_uk.gpg