linux-nfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Chuck Lever <chuck.lever@oracle.com>
To: Bruce Fields <bfields@redhat.com>
Cc: Jeff Layton <jlayton@redhat.com>,
	Bruce Fields <bfields@fieldses.org>,
	Trond Myklebust <trondmy@hammerspace.com>,
	Linux NFS Mailing List <linux-nfs@vger.kernel.org>
Subject: Re: [PATCH 0/3] Handling NFSv3 I/O errors in knfsd
Date: Wed, 28 Aug 2019 10:03:01 -0400	[thread overview]
Message-ID: <990B7B57-53B8-4ECB-A08B-1AFD2FCE13A6@oracle.com> (raw)
In-Reply-To: <20190828140044.GA14249@parsley.fieldses.org>



> On Aug 28, 2019, at 10:00 AM, J. Bruce Fields <bfields@redhat.com> wrote:
> 
> On Wed, Aug 28, 2019 at 09:57:25AM -0400, Chuck Lever wrote:
>> 
>> 
>>> On Aug 28, 2019, at 9:51 AM, Jeff Layton <jlayton@redhat.com> wrote:
>>> 
>>> On Wed, 2019-08-28 at 09:48 -0400, bfields@fieldses.org wrote:
>>>> On Tue, Aug 27, 2019 at 03:15:35PM +0000, Trond Myklebust wrote:
>>>>> I'm open to other suggestions, but I'm having trouble finding one that
>>>>> can scale correctly (i.e. not require per-client tracking), prevent
>>>>> silent corruption (by causing clients to miss errors), while not
>>>>> relying on optional features that may not be implemented by all NFSv3
>>>>> clients (e.g. per-file write verifiers are not implemented by *BSD).
>>>>> 
>>>>> That said, it seems to me that to do nothing should not be an option,
>>>>> as that would imply tolerating silent corruption of file data.
>>>> 
>>>> So should we increment the boot verifier every time we discover an error
>>>> on an asynchronous write?
>>>> 
>>> 
>>> I think so. Otherwise, only one client will ever see that error.
>> 
>> +1
>> 
>> I'm not familiar with the details of how the Linux NFS server
>> implements the boot verifier: Will a verifier bump be effective
>> for all file systems that server exports?
> 
> Yes.  It will be per network namespace, but that's the only limit.
> 
>> If so, is that an acceptable cost?
> 
> It means clients will resend all their uncommitted writes.  That could
> certainly make write errors more expensive.  But maybe you've already
> got bigger problems if you've got a full or failing disk?

One full filesystem will impact the behavior of all other exported
filesystems. That might be surprising behavior to a server administrator.
I don't have any suggestions other than maintaining a separate verifier
for each exported filesystem in each net namespace.


--
Chuck Lever




  reply	other threads:[~2019-08-28 14:03 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-08-26 16:50 [PATCH 0/3] Handling NFSv3 I/O errors in knfsd Trond Myklebust
2019-08-26 16:50 ` [PATCH 1/3] nfsd: nfsd_file cache entries should be per net namespace Trond Myklebust
2019-08-26 16:50   ` [PATCH 2/3] nfsd: Support the server resetting the boot verifier Trond Myklebust
2019-08-26 16:50     ` [PATCH 3/3] nfsd: Don't garbage collect files that might contain write errors Trond Myklebust
2019-08-27  7:58     ` [PATCH 2/3] nfsd: Support the server resetting the boot verifier kbuild test robot
2019-08-26 20:51 ` [PATCH 0/3] Handling NFSv3 I/O errors in knfsd J. Bruce Fields
2019-08-26 21:02   ` Trond Myklebust
2019-08-27  0:48     ` bfields
2019-08-27  0:56       ` Trond Myklebust
2019-08-27  1:13         ` bfields
2019-08-27  1:28           ` Trond Myklebust
2019-08-27 13:59     ` Chuck Lever
2019-08-27 14:53       ` Trond Myklebust
2019-08-27 14:58         ` bfields
2019-08-27 14:59           ` bfields
2019-08-27 15:15             ` Trond Myklebust
2019-08-27 15:20               ` Chuck Lever
2019-08-28 13:48               ` bfields
2019-08-28 13:51                 ` Jeff Layton
2019-08-28 13:57                   ` Chuck Lever
2019-08-28 14:00                     ` J. Bruce Fields
2019-08-28 14:03                       ` Chuck Lever [this message]
2019-08-28 14:16                         ` Jeff Layton
2019-08-28 14:21                           ` Chuck Lever
2019-08-28 14:40                           ` J. Bruce Fields
2019-08-28 14:48                             ` Bruce Fields
2019-08-28 14:50                               ` Chuck Lever
2019-08-28 17:07                                 ` Bruce Fields
2019-08-28 15:09                             ` Jeff Layton
2019-08-28 15:12                             ` Rick Macklem
2019-08-28 15:37                               ` Trond Myklebust
2019-08-28 15:46                               ` Bruce Fields
2019-08-27 14:54       ` Bruce Fields
2019-08-27 14:59         ` Trond Myklebust
2019-08-27 15:00           ` bfields
2019-08-27 15:17       ` Jeff Layton

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=990B7B57-53B8-4ECB-A08B-1AFD2FCE13A6@oracle.com \
    --to=chuck.lever@oracle.com \
    --cc=bfields@fieldses.org \
    --cc=bfields@redhat.com \
    --cc=jlayton@redhat.com \
    --cc=linux-nfs@vger.kernel.org \
    --cc=trondmy@hammerspace.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).