SCIENTIFIC-LINUX-USERS Archives

August 2007

SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Konstantin Olchanski <[log in to unmask]>
Reply To:
Konstantin Olchanski <[log in to unmask]>
Date:
Tue, 7 Aug 2007 22:23:32 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (25 lines)
On Mon, Aug 06, 2007 at 07:20:08PM -0700, Michael Hannon wrote:
> The wikipedia article about ext3 claims that there are size limitations
> but that with a block size of 4096 (the default), the limits are 2TB per
> file and 16 TB per file system.

I looked into this a while ago while building an 11TB raid5 md device.

The bottom line is that 8 TB is the hard limit set in the sources
for mke2fs. (See e2fsprogs-1.39/misc/mke2fs.c, look for
text "Filesystem too large"). There is a "force" switch to defeat this
check and I tried it - the resulting filesystem is unusable.

Note that the documentation is contradictory - some docs
claim 8 TB limit, others claim 16 TB limit. Whatever the case,
given the hardcoded limit in mke2fs, there is no supported way
to build ext3 filesystems larger than 8 TB.

We are now using "xfs" with 64-bit SL4.4 and it is working just fine.

-- 
Konstantin Olchanski
Data Acquisition Systems: The Bytes Must Flow!
Email: olchansk-at-triumf-dot-ca
Snail mail: 4004 Wesbrook Mall, TRIUMF, Vancouver, B.C., V6T 2A3, Canada

ATOM RSS1 RSS2