linux-nfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Alex Lyakas" <alex@zadara.com>
To: "J. Bruce Fields" <bfields@fieldses.org>
Cc: <linux-nfs@vger.kernel.org>, "Shyam Kaushik" <shyam@zadara.com>
Subject: Re: [RFC-PATCH] nfsd: provide a procfs entry to release stateids of a particular local filesystem
Date: Sun, 22 Sep 2019 09:52:36 +0300	[thread overview]
Message-ID: <8F0FAB980E6F4594A8C61D927FE022E5@alyakaslap> (raw)
In-Reply-To: <20190910202533.GC26695@fieldses.org>

Hi Bruce,

I do see in the code that a delegation stateid also holds an open file on 
the file system. In my experiments, however, the nfs4_client::cl_delegations 
list was always empty. I put an extra print to print a warning if it's not, 
but did not hit this.

Thanks,
Alex.



-----Original Message----- 
From: J. Bruce Fields
Sent: Tuesday, September 10, 2019 11:25 PM
To: Alex Lyakas
Cc: linux-nfs@vger.kernel.org ; Shyam Kaushik
Subject: Re: [RFC-PATCH] nfsd: provide a procfs entry to release stateids of 
a particular local filesystem

On Tue, Sep 10, 2019 at 10:00:24PM +0300, Alex Lyakas wrote:
> I addressed your comments, and ran the patch through checkpatch.pl.
> Patch v2 is on its way.

Thanks for the revision!  I need to spend the next week or so catching
up on some other review and then I'll get back to this.

For now:

> On Fri, Sep 6, 2019 at 7:12 PM J. Bruce Fields <bfields@fieldses.org> 
> wrote:
> > You'll want to cover delegations as well.  And probably pNFS layouts.
> > It'd be OK to do that incrementally in followup patches.
> Unfortunately, I don't have much understanding of what these are, and
> how to cover them)

Delegations are give the client the right to cache files across opens.
I'm a little surprised your patches are working for you without handling
delegations.  There may be something about your environment that's
preventing delegations from being given out.  In the NFSv4.0 case they
require the server to make a tcp connection back the client, which is
easy blocked by firewalls or NAT.  Might be worth testing with v4.1 or
4.2.

Anyway, so we probably also want to walk the client's dl_perclnt list
and look for matching files.

--b. 


  reply	other threads:[~2019-09-22  6:52 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-09-03 13:55 [RFC-PATCH] nfsd: provide a procfs entry to release stateids of a particular local filesystem Alex Lyakas
2019-09-06 16:12 ` J. Bruce Fields
2019-09-10 19:00   ` Alex Lyakas
2019-09-10 20:25     ` J. Bruce Fields
2019-09-22  6:52       ` Alex Lyakas [this message]
2019-09-23 16:25         ` J. Bruce Fields
2019-09-24 14:35           ` Alex Lyakas
2019-12-05 11:47       ` Alex Lyakas
2019-12-05 15:03         ` J. Bruce Fields
2019-09-16 22:28 ` John Gallagher
2019-09-17 14:16   ` Chuck Lever

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=8F0FAB980E6F4594A8C61D927FE022E5@alyakaslap \
    --to=alex@zadara.com \
    --cc=bfields@fieldses.org \
    --cc=linux-nfs@vger.kernel.org \
    --cc=shyam@zadara.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).