[OpenIndiana-discuss] Is this kind of nfs speed just tiny bit outrageous?

James Carlson carlsonj at workingcode.com
Sat Jan 3 15:15:58 UTC 2015


On 1/3/2015 7:32 AM, Harry Putnam wrote:
> Server OS: OI 115_a9 client OS: Gentoo linux (up to date)
> 
> On client: I recently decided I'd like to write rsnapshot (using
> rsync) backups to an nfs mounted zfs share on OI nfs server.
> 
> The first trial seem pretty outrageous in terms of time taken to 
> complete.

What does your invocation line look like?  Is it like this:

   ... /path/on/linux/ /net/server/path/on/OI/

or like this:

   ... /path/on/linux/ server:/path/on/OI/

I assume it's the former, and you're trying to write zillions of tiny
files via NFS.  If so, I suspect you're seeing the action of NFS COMMIT:

https://blogs.oracle.com/roch/entry/nfs_and_zfs_a_fine

Some implementations of NFS ignore or simply mishandle COMMIT, but
OI's does not.

The simple answer is "don't do that."  You can serialize the stream,
transfer the serialized stream over the network (via ssh or rsync's
own protocol), and then write locally.  This is what I do with my
rsync jobs, regardless of whether the target server is Solaris, Linux,
AIX, or something else.

A harder answer is to make sure you have your ZFS system set up to run
faster with synchronous operations.  To do that, you need a separate
intent log on a fast device.  See:

https://blogs.oracle.com/perrin/entry/slog_blog_or_blogging_on

-- 
James Carlson         42.703N 71.076W         <carlsonj at workingcode.com>



More information about the openindiana-discuss mailing list