[sf-lug] What are the best practices for Linux partitioning & Mount points for Production systems

[jim]

(clipped from below) 
"My URL #2 (above) includes a brief schema of filesystems on the 
server that runs this mailing list -- and some of the reasons.  If
anyone's interested, I'd be glad to elaborate more about that."
JS: I'm interested. 


On Fri, 2012-03-02 at 16:00 -0800, Rick Moen wrote:
> Quoting nk oorda (nk.oorda at gmail.com):
> 
> > i need some suggestion for defining the partition size for my production
> > systems.
> 
> Your partitioning is logically dictated by what you're trying to
> achieve, including what threat modes you're attempting to protect
> against.
> 
> Some of the concerns that might drive partitioning design for a server
> are laid out here:
> http://linuxmafia.com/pipermail/conspire/2012-February/006925.html
> http://linuxmafia.com/pipermail/conspire/2012-February/006970.html
> http://linuxmafia.com/pipermail/conspire/2012-February/006921.html
> 
> Themes mentioned:
> 1.  Partitions carved out in order to use ext2 for high performance.
> 2.  Partitions carved out to enable use of custom mount options, 
>     e.g., noatime, nodev, nosuid
> 3.  Partitions carved out to cluster most-accessed parts of the file
>     tree around the swap partition for minimum average seek
>     distince/time within a spindle (where spinning media is used).
> 4.  Partitions carved out to keep them normally read-only as a 
>     protection against sysadmin error.
> 
> One might add: 
> 5.  Partitions made be NOT part of the root filesystem to better protect 
>     the root FS from getting overfilled or damaged.
> 6.  Partitions kept separate because they're network-shared e.g., via NFS
> 
> Poster rgmoore on LWN posted (https://lwn.net/Articles/484332/)
> 
>   The idea is that you should be able to have a separate partition for
>   each different kind of data. It should be possible to keep read-only
>   data (or data that is only supposed to be written by a sysadmin) on a
>   separate partition from data that's frequently written, data that's
>   specific to a particular machine separate from data that can be shared
>   across multiple machines, and data that is volatile across a reboot
>   separate from data that needs to be preserved across reboots. So the
>   idea is that standard partitions are supposed to be:
> 
>   / Machine specific, read-only
>   /var Machine specific, read-write, stable across reboots
>   /tmp Machine specific, read-write, volatile across reboots
>   /usr Shared, read-only
>   /home Shared, read-write 
> 
> Exactly so.
> 
> > What i am able to get from the google search is:
> 
> What you should be concentrating on finding is _why_ a particular
> division was used, i.e., towards what purpose or benefit.
> 
> My URL #2 (above) includes a brief schema of filesystems on the 
> server that runs this mailing list -- and some of the reasons.  If
> anyone's interested, I'd be glad to elaborate more about that.
> 
> 
> _______________________________________________
> sf-lug mailing list
> sf-lug at linuxmafia.com
> http://linuxmafia.com/mailman/listinfo/sf-lug
> Information about SF-LUG is at http://www.sf-lug.org/