> Here is fsck output from my large volume (84TB/76.4 TiB).

I think that volumes larger than 5-10 drives or around 10-20TB
are not a good idea, but at least yours is not full.

> I suffered a power outage and this volume was actually almost
> never accessed and hadn't been accessed in many hours as the
> disks were even asleep when it occured

The defaults used by most storage systems, including the
GNU/Linux one, tend to result in excessive caching for too long.

I prefer to ensure that the page cache flusher runs often and
flushes unwritten blocks very often, ensuring a steady stream of
writes to disk.

The same for RAID HAs, and so on. Ensuring that updated data
does get to persistent storage is often difficult, as various
layers of storage (including drive firmware sometimes) cheat to
look better in benchmarks.

> but with a small number of files on it even and only 5 TiB
> used.

That's not too bad.

------------------------------------------------------------------------------
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
_______________________________________________
Jfs-discussion mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/jfs-discussion

Reply via email to