Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×
Linux Software

Linux Kernel Partition and FS Physical Limits? 7

Posted by Cliff from the how-much-can-it-handle dept.
dmelomed asks: "I would like to Ask Slashdot about Linux's PHYSICAL limits of partition and FS size on i386 and 64 bit chips. What are the maximum real-life possible partition sizes on big RAIDs (several TB) that the kernel can work with reliably? What are better filesystems to use that handle LARGE RAIDS?"
This discussion has been archived. No new comments can be posted.

Linux Kernel Partition and FS Physical Limits? More

Linux Kernel Partition and FS Physical Limits?

Comments Filter:
  • The 2Gb size limit for single files is more of an issue. This was the bane of my life when dealing with CAE on Solaris 2.5 / HPUX 10.20

    If the beloved x86 takes any longer to die peacefully it may need to be addressed.
  • I would define a physical limitation as something beyond which the _hardware cannot support. In fact I think IDE cannot handle disks larger than some limit that's not that big (IIRC 120 Gig). Hey, maybe that will knock up the volume and drop the price of SCSI! Sounds good to me... Anything that actually limited in the kernel can be gotten around (using 64 bit numbers, or even pairs of them, to represent the needed values).

    That said, I don't know what the current limits are for partition sizes in the kernel. Though I've seen a multiple-terrabyte database (holding star maps), that I think may have been running Linux (though it could have been Solaris).
  • Have you read the RAID-howto and related material yet?

  • WTF? (Score:2)

    by rbf ( 2305 )
    WTF? Doesn't anyone RTFM anymore? Ohh, and you never said which filesystem! I'll assume (there's that evil word again!) you mean EXT2...


    How about The Filesystems-HOWTO [linuxdoc.org]? A quote from section 6.2:

    ``Standard'' Ext2fs features

    The Ext2fs supports standard Unix file types: regular files, directories, device special files and symbolic links.

    Ext2fs is able to manage filesystems created on really big partitions. While the original kernel code restricted the maximal filesystem size to 2 GB, recent work in the VFS layer have raised this limit to 4 TB. Thus, it is now possible to use big disks without the need of creating many partitions.

    Ext2fs provides long file names. It uses variable length directory entries. The maximal file name size is 255 characters. This limit could be extended to 1012 if needed.

    Ext2fs reserves some blocks for the super user (root). Normally, 5% of the blocks are reserved. This allows the administrator to recover easily from situations where user processes fill up filesystems.


    Quote from section 6.3:

    Ext3 support the same features as Ext2, but includes also Journaling.


    Maybe you should RTFM and be a little more specific next time! ;-)
  • Hmm... that's strange. I think they meant that the file size limit was 2GB, and was then raised to 4GB. However, filesystems can be much bigger.
  • No, the HOWT-TO is correct. The current limit on the filesize is 2GB (however, Large File Support is in the 2.3.x kernels). The limit on filesystems was originally 2GB but is currently 4TB.
  • Ahh... I misread the TB for a GB... big mistake... but, it was late and I was tired.

Slashdot Top Deals

THEGODDESSOFTHENETHASTWISTINGFINGERSANDHERVOICEISLIKEAJAVELININTHENIGHTDUDE

Close