[PLUG] Logrorate: checking whether a file is open with prerorate?

[Eli Stair]

This makes a blind assumption that the next run will be safe to
compress, and that this one isn't... kind of a big oversimplification
that won't always be right.

If for instance this was a monthly archive, all the archives would
always be a month behind, no?  Not to mention that there's the chance
(more likely on shorter-interval rotates), that the file will STILL be
open and being written to by some application... it just puts off the
incorrect behaviour until the next run.

This didn't seem to fit the bill for me, in fact I've never used the
delaycompress option because it doesn't seem to make much sense :)

/eli

On 7/29/05, Elliott Mitchell <ehem at m5p.com> wrote:
> >From: Eli Stair <eli.stair at gmail.com>
> > I've gotta support a bad-bad-bad application that doesn't simply
> > append data to its log, it keeps the file open in memory, and writes
> > to it "from the last known offset into the file, according to its last
> > write"... according the the brilliant developer.
> >
> > The issue is that I cannot use the create (move and create new
> > logfile) OR truncatecompress options... can't move it because it's
> > open for writing constantly (canNOT bring down the app), and when the
> > truncatecompress option is used, on the next write by the daemon the
> > "OLD" data that was removed becomes zero bytes up to the next write
> > offset:
> 
> The obvious one would be to use "delaycompress".
> 
> 
> --
> (\___(\___(\______          --=> 8-) EHM <=--          ______/)___/)___/)
>  \BS (    |         EHeM at gremlin.m5p.com PGP 8881EF59         |    )   /
>   \_CS\   |  _____  -O #include <stddisclaimer.h> O-   _____  |   /  _/
>     \___\_|_/82 04 A1 3C C7 B1 37 2A*E3 6E 84 DA 97 4C 40 E6\_|_/___/
> 
> 
> _______________________________________________
> PLUG mailing list
> PLUG at lists.pdxlinux.org
> http://lists.pdxlinux.org/mailman/listinfo/plug
> 


-- 
--I'm on vacation, and so is my .sig!