Hi, Krishna

On 12/13/2016 09:35 PM, Krishna Harathi wrote:
> We are planning to move to Ganesha 2.4.1 from 2.3.2 and have the
> following question.
>
> 1. Who is using 2.4.1 is their product.

Red Hat is using it for it's next Ceph (RHCS) and Gluster (RHGS) products.

> 2. Now that cache_inode is replaced by the MDCache stackable FSAL -
>     a. Is it possible to cache only the open handles and nothing else.

Sort of.  If you set the Entries_HWMark to low enough, then any handle 
with no state (open files, locks, delegations, etc.) associated with it 
should be re-used as soon as possible.

>     b. Is it possible to configure it out.

Not currently, no.  This is because the protocol layers and SAL depend 
on having stable handle pointers across calls as long as there is state 
associated with a handle.  Currently, only MDCACHE provides a handle 
cache.  It's possible for a FSAL to provide it's own handle cache, and 
to then not stack MDCACHE, but no FSAL currently does this, so code 
would need to be written.

>     c. What other configuration/tuning is available.

Configuration is unchanged from cache_inode, and still lives in the 
CACHEINODE config block.

Daniel



------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most 
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
_______________________________________________
Nfs-ganesha-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/nfs-ganesha-devel

Reply via email to