All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jeff Layton <jlayton-vpEMnDpepFuMZCB2o+C8xQ@public.gmane.org>
To: "J. Bruce Fields"
	<bfields-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org>,
	"Eric W. Biederman"
	<ebiederm-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>
Cc: serge.hallyn-Z7WLFzj8eWMS+FvcfC7Uqw@public.gmane.org,
	containers-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org,
	linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	Nikolay Borisov <kernel-6AxghH7DbtA@public.gmane.org>,
	viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.org,
	linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: [RFC PATCH] locks: Show only file_locks created in the same pidns as current process
Date: Tue, 02 Aug 2016 16:01:22 -0400	[thread overview]
Message-ID: <1470168082.15226.14.camel@poochiereds.net> (raw)
In-Reply-To: <20160802194437.GD15324-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org>

On Tue, 2016-08-02 at 15:44 -0400, J. Bruce Fields wrote:
> On Tue, Aug 02, 2016 at 02:09:22PM -0500, Eric W. Biederman wrote:
> > 
> > > > "J. Bruce Fields" <bfields@fieldses.org> writes:
> > 
> > > 
> > > On Tue, Aug 02, 2016 at 11:00:39AM -0500, Eric W. Biederman wrote:
> > > > 
> > > > > > > > Nikolay Borisov <kernel@kyup.com> writes:
> > > > 
> > > > > 
> > > > > Currently when /proc/locks is read it will show all the file locks
> > > > > which are currently created on the machine. On containers, hosted
> > > > > on busy servers this means that doing lsof can be very slow. I
> > > > > observed up to 5 seconds stalls reading 50k locks, while the container
> > > > > itself had only a small number of relevant entries. Fix it by
> > > > > filtering the locks listed by the pidns of the current process
> > > > > and the process which created the lock.
> > > > 
> > > > The locks always confuse me so I am not 100% connecting locks
> > > > to a pid namespace is appropriate.
> > > > 
> > > > That said if you are going to filter by pid namespace please use the pid
> > > > namespace of proc, not the pid namespace of the process reading the
> > > > file.
> > > 
> > > Oh, that makes sense, thanks.
> > > 
> > > What does /proc/mounts use, out of curiosity?  The mount namespace that
> > > /proc was originally mounted in?
> > 
> > /proc/mounts -> /proc/self/mounts
> 
> D'oh, I knew that.
> 
> > 
> > /proc/[pid]/mounts lists mounts from the mount namespace of the
> > appropriate process.
> > 
> > That is another way to go but it is a tread carefully thing as changing
> > things that way it is easy to surprise apparmor or selinux rules and be
> > surprised you broke someones userspace in a way that prevents booting.
> > Although I suspect /proc/locks isn't too bad.
> 
> OK, thanks.
> 
> /proc/[pid]/locks might be confusing.  I'd expect it to be "all the
> locks owned by this task", rather than "all the locks owned by pid's in
> the same pid namespace", or whatever criterion we choose.
> 
> Uh, I'm still trying to think of the Obviously Right solution here, and
> it's not coming.
> 
> --b.


I'm a little leery of changing how this works. It has always been
maintained as a legacy interface, so do we run the risk of breaking
something if we turn it into a per-namespace thing? This also doesn't
solve the problem of slow traversal in the init_pid_ns -- only in a
container.

I also can't help but feel that /proc/locks is just showing its age. It
was fine in the late 90's, but its limitations are just becoming more
apparent as things get more complex. It was never designed for
performance as you end up thrashing several spinlocks when reading it.

Maybe it's time to think about presenting this info in another way? A
global view of all locks on the system is interesting but maybe it
would be better to present it more granularly somehow?

I guess I should go look at what lsof actually does with this info...

-- 
Jeff Layton <jlayton@poochiereds.net>
_______________________________________________
Containers mailing list
Containers@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/containers

WARNING: multiple messages have this Message-ID (diff)
From: Jeff Layton <jlayton@poochiereds.net>
To: "J. Bruce Fields" <bfields@fieldses.org>,
	"Eric W. Biederman" <ebiederm@xmission.com>
Cc: Nikolay Borisov <kernel@kyup.com>,
	viro@zeniv.linux.org.uk, linux-kernel@vger.kernel.org,
	linux-fsdevel@vger.kernel.org,
	containers@lists.linux-foundation.org,
	serge.hallyn@canonical.com
Subject: Re: [RFC PATCH] locks: Show only file_locks created in the same pidns as current process
Date: Tue, 02 Aug 2016 16:01:22 -0400	[thread overview]
Message-ID: <1470168082.15226.14.camel@poochiereds.net> (raw)
In-Reply-To: <20160802194437.GD15324@fieldses.org>

On Tue, 2016-08-02 at 15:44 -0400, J. Bruce Fields wrote:
> On Tue, Aug 02, 2016 at 02:09:22PM -0500, Eric W. Biederman wrote:
> > 
> > > > "J. Bruce Fields" <bfields@fieldses.org> writes:
> > 
> > > 
> > > On Tue, Aug 02, 2016 at 11:00:39AM -0500, Eric W. Biederman wrote:
> > > > 
> > > > > > > > Nikolay Borisov <kernel@kyup.com> writes:
> > > > 
> > > > > 
> > > > > Currently when /proc/locks is read it will show all the file locks
> > > > > which are currently created on the machine. On containers, hosted
> > > > > on busy servers this means that doing lsof can be very slow. I
> > > > > observed up to 5 seconds stalls reading 50k locks, while the container
> > > > > itself had only a small number of relevant entries. Fix it by
> > > > > filtering the locks listed by the pidns of the current process
> > > > > and the process which created the lock.
> > > > 
> > > > The locks always confuse me so I am not 100% connecting locks
> > > > to a pid namespace is appropriate.
> > > > 
> > > > That said if you are going to filter by pid namespace please use the pid
> > > > namespace of proc, not the pid namespace of the process reading the
> > > > file.
> > > 
> > > Oh, that makes sense, thanks.
> > > 
> > > What does /proc/mounts use, out of curiosity?  The mount namespace that
> > > /proc was originally mounted in?
> > 
> > /proc/mounts -> /proc/self/mounts
> 
> D'oh, I knew that.
> 
> > 
> > /proc/[pid]/mounts lists mounts from the mount namespace of the
> > appropriate process.
> > 
> > That is another way to go but it is a tread carefully thing as changing
> > things that way it is easy to surprise apparmor or selinux rules and be
> > surprised you broke someones userspace in a way that prevents booting.
> > Although I suspect /proc/locks isn't too bad.
> 
> OK, thanks.
> 
> /proc/[pid]/locks might be confusing.  I'd expect it to be "all the
> locks owned by this task", rather than "all the locks owned by pid's in
> the same pid namespace", or whatever criterion we choose.
> 
> Uh, I'm still trying to think of the Obviously Right solution here, and
> it's not coming.
> 
> --b.


I'm a little leery of changing how this works. It has always been
maintained as a legacy interface, so do we run the risk of breaking
something if we turn it into a per-namespace thing? This also doesn't
solve the problem of slow traversal in the init_pid_ns -- only in a
container.

I also can't help but feel that /proc/locks is just showing its age. It
was fine in the late 90's, but its limitations are just becoming more
apparent as things get more complex. It was never designed for
performance as you end up thrashing several spinlocks when reading it.

Maybe it's time to think about presenting this info in another way? A
global view of all locks on the system is interesting but maybe it
would be better to present it more granularly somehow?

I guess I should go look at what lsof actually does with this info...

-- 
Jeff Layton <jlayton@poochiereds.net>

  parent reply	other threads:[~2016-08-02 20:01 UTC|newest]

Thread overview: 62+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-08-02 14:42 [RFC PATCH] locks: Show only file_locks created in the same pidns as current process Nikolay Borisov
2016-08-02 14:42 ` Nikolay Borisov
     [not found] ` <1470148943-21835-1-git-send-email-kernel-6AxghH7DbtA@public.gmane.org>
2016-08-02 14:45   ` Nikolay Borisov
2016-08-02 14:45     ` Nikolay Borisov
2016-08-02 15:05   ` J. Bruce Fields
2016-08-02 16:00   ` Eric W. Biederman
2016-08-02 16:00     ` Eric W. Biederman
     [not found]     ` <87r3a7qhy0.fsf-JOvCrm2gF+uungPnsOpG7nhyD016LWXt@public.gmane.org>
2016-08-02 17:40       ` J. Bruce Fields
2016-08-02 17:40         ` J. Bruce Fields
     [not found]         ` <20160802174003.GD11767-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org>
2016-08-02 19:09           ` Eric W. Biederman
2016-08-02 19:09             ` Eric W. Biederman
     [not found]             ` <87invjq97h.fsf-JOvCrm2gF+uungPnsOpG7nhyD016LWXt@public.gmane.org>
2016-08-02 19:44               ` J. Bruce Fields
2016-08-02 19:44                 ` J. Bruce Fields
     [not found]                 ` <20160802194437.GD15324-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org>
2016-08-02 20:01                   ` Jeff Layton [this message]
2016-08-02 20:01                     ` Jeff Layton
     [not found]                     ` <1470168082.15226.14.camel-vpEMnDpepFuMZCB2o+C8xQ@public.gmane.org>
2016-08-02 20:11                       ` Nikolay Borisov
2016-08-02 20:11                         ` Nikolay Borisov
2016-08-02 20:34                       ` J. Bruce Fields
2016-08-02 20:34                         ` J. Bruce Fields
2016-08-03  7:35   ` [PATCH v2] locks: Filter /proc/locks output on proc pid ns Nikolay Borisov
2016-08-03  7:35     ` Nikolay Borisov
     [not found]     ` <1470209710-30022-1-git-send-email-kernel-6AxghH7DbtA@public.gmane.org>
2016-08-03 13:46       ` Jeff Layton
2016-08-03 13:46         ` Jeff Layton
     [not found]         ` <1470232012.18285.4.camel-vpEMnDpepFuMZCB2o+C8xQ@public.gmane.org>
2016-08-03 14:17           ` Nikolay Borisov
2016-08-03 14:17             ` Nikolay Borisov
2016-08-03 14:28             ` J. Bruce Fields
     [not found]               ` <20160803142850.GA27072-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org>
2016-08-03 14:33                 ` Nikolay Borisov
2016-08-03 14:33                   ` Nikolay Borisov
     [not found]             ` <57A1FCE5.3040206-6AxghH7DbtA@public.gmane.org>
2016-08-03 14:28               ` J. Bruce Fields
2016-08-03 14:54               ` Pavel Emelyanov
2016-08-03 14:54                 ` Pavel Emelyanov
     [not found]                 ` <57A205BE.3070202-5HdwGun5lf+gSpxsJD1C4w@public.gmane.org>
2016-08-03 15:00                   ` Nikolay Borisov
2016-08-03 15:00                 ` Nikolay Borisov
     [not found]                   ` <57A20702.3040805-6AxghH7DbtA@public.gmane.org>
2016-08-03 15:06                     ` J. Bruce Fields
2016-08-03 15:06                       ` J. Bruce Fields
     [not found]                       ` <20160803150631.GA3789-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org>
2016-08-03 15:10                         ` Nikolay Borisov
2016-08-03 15:10                           ` Nikolay Borisov
     [not found]                           ` <57A2097C.7060206-6AxghH7DbtA@public.gmane.org>
2016-08-03 17:35                             ` Eric W. Biederman
2016-08-03 17:35                               ` Eric W. Biederman
2016-08-03 14:54   ` [PATCHv3] " Nikolay Borisov
     [not found]     ` <1470236078-2389-1-git-send-email-kernel-6AxghH7DbtA@public.gmane.org>
2016-08-03 15:24       ` Jeff Layton
2016-08-03 16:23       ` Eric W. Biederman
     [not found]         ` <87k2fxom8a.fsf-JOvCrm2gF+uungPnsOpG7nhyD016LWXt@public.gmane.org>
2016-08-03 16:50           ` Jeff Layton
     [not found]             ` <1470243015.13804.7.camel-vpEMnDpepFuMZCB2o+C8xQ@public.gmane.org>
2016-08-03 21:09               ` Eric W. Biederman
     [not found]                 ` <87twf1ftk9.fsf-JOvCrm2gF+uungPnsOpG7nhyD016LWXt@public.gmane.org>
2016-08-03 21:26                   ` Nikolay Borisov
     [not found]                     ` <a0a58f75-0e40-c14f-d8e3-8f094e9fc62c-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2016-08-04  4:18                       ` Eric W. Biederman
     [not found]                         ` <87eg659ngh.fsf-JOvCrm2gF+uungPnsOpG7nhyD016LWXt@public.gmane.org>
2016-08-04  5:07                           ` Eric W. Biederman
2016-08-03 17:40       ` Eric W. Biederman
2016-08-04  7:26   ` [PATCHv4] " Nikolay Borisov
     [not found]     ` <1470295588-9803-1-git-send-email-kernel-6AxghH7DbtA@public.gmane.org>
2016-08-04 11:29       ` Jeff Layton
     [not found]         ` <1470310175.22052.3.camel-vpEMnDpepFuMZCB2o+C8xQ@public.gmane.org>
2016-08-04 14:09           ` Eric W. Biederman
     [not found]             ` <874m707hhm.fsf-JOvCrm2gF+uungPnsOpG7nhyD016LWXt@public.gmane.org>
2016-08-04 14:34               ` Nikolay Borisov
2016-08-04 15:09               ` Nikolay Borisov
     [not found]                 ` <57A35AC7.7040105-6AxghH7DbtA@public.gmane.org>
2016-08-04 15:21                   ` Eric W. Biederman
2016-08-05  7:30   ` [PATCHv5] " Nikolay Borisov
     [not found]     ` <1470382204-21480-1-git-send-email-kernel-6AxghH7DbtA@public.gmane.org>
2016-08-05 10:47       ` Jeff Layton
     [not found]         ` <1470394036.8100.2.camel-vpEMnDpepFuMZCB2o+C8xQ@public.gmane.org>
2016-08-05 14:58           ` J. Bruce Fields
2016-08-02 15:05 ` [RFC PATCH] locks: Show only file_locks created in the same pidns as current process J. Bruce Fields
     [not found]   ` <20160802150521.GB11767-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org>
2016-08-02 15:20     ` Nikolay Borisov
2016-08-02 15:20       ` Nikolay Borisov
     [not found]       ` <57A0BA40.5010406-6AxghH7DbtA@public.gmane.org>
2016-08-02 15:43         ` J. Bruce Fields
2016-08-02 15:43           ` J. Bruce Fields

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=1470168082.15226.14.camel@poochiereds.net \
    --to=jlayton-vpemndpepfumzcb2o+c8xq@public.gmane.org \
    --cc=bfields-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org \
    --cc=containers-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org \
    --cc=ebiederm-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org \
    --cc=kernel-6AxghH7DbtA@public.gmane.org \
    --cc=linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=serge.hallyn-Z7WLFzj8eWMS+FvcfC7Uqw@public.gmane.org \
    --cc=viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.