Re: [nfsv4] Linux pNFS status meeting 07/08

[sfaibish <[e-mail]>]

I posted a preliminary web page announcement for the Fall Bake-a-Thon in  
Boston
on CITI website:

  http://www.citi.umich.edu/projects/nfsv4/citi_bakeathon/15th_nfsv4_bakeathon.html


On Thu, 08 Jul 2010 13:45:58 -0400, J. Bruce Fields <[e-mail]>  
wrote:

> Also I'd like to hand off running and taking notes on this meeting to
> someone else for a while.  (Benny, maybe?)  Or we can go back to making
> the meeting more informal if that's preferred.
>
> --b.
>
> 1. upstream status/merge plans (trond, bfields)
> 	- upstream at 2.6.35-rc4.
> 	- server side: some bugfixes merged to callback code; allocation
> 	  failures changed to return DELAY.
> 	- discussion of error return on client callback-parsing error.
> 	- trond: everything from before bakeathon is in nfs-2.6.36
> 	  branch or upstream.  Also some file locking bugfixes
> 	  (bugzilla.kernel.org #14501).  Also sending release_lockowner
> 	  more often to keep server resources down.
>
> 2. State of pNFS tree (bhalevy)
> 	- working on patches from boaz and andy for release branch.
> 	  Anyone else have big pending patchsets?
> 		- Jim is getting complex-volume block patches tested,
> 		  will send them out soon.
> 		- Fred has some pending stuff ("minor cleanup")
> 	- next week would like to start cleaning up and squashing
> 	  together patches.  Also will truncate some of the history.
>
> 3. Client
> 	3.1. Client pNFS status
> 		- Andy needs to look at state reclaim stuff?
> 		- Andy still wondering what to do with the nfs inode
> 		  layout pointer.  What's the lifetime of the thing that
> 		  points to?
> 	3.2. Client state management design documentation (trond)
> 		- Also, on docs: Andy will check that wiki docs are
> 		  uptodate.  And will do some more documentation of
> 		  refcnts, as will Fred.
>
> 4. Server
> 	4.1. Todo's for minimal 4.1 server (bfields)
> 		- still working on callback changes.  Trond notes
> 		  ongoing rpc code improvements may allow us to do much
> 		  of this by keeping a single client and modifying its
> 		  transports, in which case I may be able to replace my
> 		  callbacks with single rpc tasks.
> 		- also fixing 4.0/4.1 bug that prevents exclusive locks
> 		  on upgraded open.
> 	4.2. pnfs
> 		- pnfs/gfs2
> 			- I'll ask Eric for a report on gfs2 bakeathon
> 			  testing.
> 		- exofs
> 			- I'll ask Brian about citi exofs setup status.
>
> 5. boston bakeathon
> 	- October 4-8.
> 	- Details are finalized, to be posted soon.
> 	- If your company will be bringing large equipment, let Sorin
> 	  know.
> 	- If anyone would be interested in sponsoring a reception or
> 	  other addition to the bakeathon, contact Sorin.  (Or if you
> 	  know who at your company would handle such a thing, let Sorin
> 	  know.)
>
> 6. new business
> 	- Sorin asked Trond for summary of layout-commit discussion.
> 	  Discussion ensued.
>
> 7. next meeting
> 	- next week as usual.
> --
> To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> the body of a message to [e-mail]
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
>



-- 
Best Regards

Sorin Faibish
Corporate Distinguished Engineer
Network Storage Group

         EMC²
where information lives

Phone: [phone] x 48545
Cellphone: 617-510-0422
Email : [e-mail]