linux-btrfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Ken D'Ambrosio" <ken@jots.org>
To: Roman Mamedov <rm@romanrm.ru>,
	linux-btrfs <linux-btrfs@vger.kernel.org>,
	Jim <jim@webstarts.com>
Subject: Re: too many files open
Date: Wed, 05 Oct 2011 12:07:58 -0400	[thread overview]
Message-ID: <5dc5c2e75fcd85e5da1ae83fcc7eba05@www.jots.org> (raw)
In-Reply-To: <4E8C7DBB.20901@webstarts.com>

Well, I hate to grasp for a flyswatter when a hammer might be better, but
what's /proc/sys/fs/file-nr show?  The first number is your currently opened
files, the last one is your maximum files (as dictated by
/proc/sys/fs/file-max), and the middle one's allocated-but-unused file handles.
 If it's showing a number anything near your max files, it's probably a fine
time to check out lsof.  Looking for where the disparity lies will probably
offer some insights, I imagine.

$.02,

-Ken


On Wed, 05 Oct 2011 11:54:35 -0400 Jim <jim@webstarts.com> wrote

> Checked ulimit and processes are not the issue here.  Rsync never has 
> more than 15 instances running and even accounting for children and 
> other processes they wouldnt approach the process limit.  The error 
> ddoes seem to be with btrfs as I cant ls the file system while this 
> condition exists.  Ls also returns "too many files open".  Btrfs sub 
> list also shows the same too many files open condition.  Actually, there 
> should be no files open after the script has failed (the script runs, 
> just reports the errors).  Something either reports files as being open 
> or is holding them open, and a remount flushes this and the fs is back 
> to normal.  Very confusing.
> Jim
> 
> On 10/05/2011 11:32 AM, Jim wrote:
> > Thanks very much for the idea.  I will check and get back.
> > Jim
> >
> >
> > On 10/05/2011 11:31 AM, Roman Mamedov wrote:
> >> On Wed, 05 Oct 2011 11:24:27 -0400
> >> Jim<jim@webstarts.com>  wrote:
> >>
> >>> Good morning Btrfs list,
> >>> I have been loading a btrfs file system via a script rsyncing data 
> >>> files
> >>> from an nfs mounted directory.  The script runs well but after several
> >>> days (moving about 10TB) rsync reports that it is sending the file list
> >>> but stops moving data because btrfs balks saying too many files 
> >>> open.  A
> >>> simple umount/mount fixes the problem.  What am I flushing when I
> >>> remount that would affect this, and is there a way to do this without a
> >>> remount.  Once again thanks for any assistance.
> >> Are you sure it's a btrfs problem? Check "ulimit -n", see "help 
> >> ulimit" (assuming you use bash).
> >>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html






  reply	other threads:[~2011-10-05 16:07 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-10-05 15:24 too many files open Jim
2011-10-05 15:31 ` Roman Mamedov
     [not found]   ` <4E8C7885.50205@webstarts.com>
2011-10-05 15:54     ` Jim
2011-10-05 16:07       ` Ken D'Ambrosio [this message]
2011-10-05 16:15         ` Jim
2011-10-05 17:18         ` Jim

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=5dc5c2e75fcd85e5da1ae83fcc7eba05@www.jots.org \
    --to=ken@jots.org \
    --cc=jim@webstarts.com \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=rm@romanrm.ru \
    /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).