linux-nfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Steve Dickson <SteveD@redhat.com>
To: Chuck Lever <chuck.lever@oracle.com>
Cc: linux-nfs@vger.kernel.org
Subject: Re: [PATCH 2/2] nfs-iostat: Fix attribute cache statistics
Date: Wed, 30 Apr 2014 12:31:13 -0400	[thread overview]
Message-ID: <53612551.8070509@RedHat.com> (raw)
In-Reply-To: <20140425165220.30699.79542.stgit@manet.1015granger.net>



On 04/25/2014 12:52 PM, Chuck Lever wrote:
> "nfs-iostat.py --attr" was displaying nonsense (like negative
> counts and percentages).
> 
> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Committed... 

steved.
> ---
> 
>  tools/nfs-iostat/nfs-iostat.py |   28 ++++++++--------------------
>  1 files changed, 8 insertions(+), 20 deletions(-)
> 
> diff --git a/tools/nfs-iostat/nfs-iostat.py b/tools/nfs-iostat/nfs-iostat.py
> index 6831c12..b324cd8 100644
> --- a/tools/nfs-iostat/nfs-iostat.py
> +++ b/tools/nfs-iostat/nfs-iostat.py
> @@ -243,27 +243,15 @@ class DeviceData:
>          """Print attribute cache efficiency stats
>          """
>          nfs_stats = self.__nfs_data
> -        getattr_stats = self.__rpc_data['GETATTR']
> -
> -        if nfs_stats['inoderevalidates'] != 0:
> -            getattr_ops = float(getattr_stats[1])
> -            opens = float(nfs_stats['vfsopen'])
> -            revalidates = float(nfs_stats['inoderevalidates']) - opens
> -            if revalidates != 0:
> -                ratio = ((revalidates - getattr_ops) * 100) / revalidates
> -            else:
> -                ratio = 0.0
> -
> -            data_invalidates = float(nfs_stats['datainvalidates'])
> -            attr_invalidates = float(nfs_stats['attrinvalidates'])
>  
> -            print()
> -            print('%d inode revalidations, hitting in cache %4.2f%% of the time' % \
> -                (revalidates, ratio))
> -            print('%d open operations (mandatory GETATTR requests)' % opens)
> -            if getattr_ops != 0:
> -                print('%4.2f%% of GETATTRs resulted in data cache invalidations' % \
> -                   ((data_invalidates * 100) / getattr_ops))
> +        print()
> +        print('%d VFS opens' % (nfs_stats['vfsopen']))
> +        print('%d inoderevalidates (forced GETATTRs)' % \
> +            (nfs_stats['inoderevalidates']))
> +        print('%d page cache invalidations' % \
> +            (nfs_stats['datainvalidates']))
> +        print('%d attribute cache invalidations' % \
> +            (nfs_stats['attrinvalidates']))
>  
>      def __print_dir_cache_stats(self, sample_time):
>          """Print directory stats
> 

  reply	other threads:[~2014-04-30 16:31 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-04-25 16:52 [PATCH 1/2] nfs-iostat: Fix columnarization of RPC statistics Chuck Lever
2014-04-25 16:52 ` [PATCH 2/2] nfs-iostat: Fix attribute cache statistics Chuck Lever
2014-04-30 16:31   ` Steve Dickson [this message]
2014-04-30 16:30 ` [PATCH 1/2] nfs-iostat: Fix columnarization of RPC statistics Steve Dickson

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=53612551.8070509@RedHat.com \
    --to=steved@redhat.com \
    --cc=chuck.lever@oracle.com \
    --cc=linux-nfs@vger.kernel.org \
    /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).