[Top] [All Lists]

Re: deep chmod|chown -R begin to start OOMkiller

To: Andi Kleen <ak@xxxxxxx>
Subject: Re: deep chmod|chown -R begin to start OOMkiller
From: David Chinner <dgc@xxxxxxx>
Date: Tue, 13 Dec 2005 12:32:04 +1100
Cc: David Chinner <dgc@xxxxxxx>, linux-xfs@xxxxxxxxxxx
In-reply-to: <p73bqzmpx2f.fsf@xxxxxxxxxxxxx>
References: <20051207183531.5c13e8c5.masaki-c@xxxxxxxxxx> <20051208070841.GJ501696@xxxxxxxxxxxxxxxxx> <20051209104148.346f2ff5.masaki-c@xxxxxxxxxx> <20051212014633.GC19154461@xxxxxxxxxxxxxxxxx> <p73bqzmpx2f.fsf@xxxxxxxxxxxxx>
Sender: linux-xfs-bounce@xxxxxxxxxxx
User-agent: Mutt/
On Mon, Dec 12, 2005 at 12:57:12AM -0700, Andi Kleen wrote:
> David Chinner <dgc@xxxxxxx> writes:
> > 
> > There are two immediate solutions that I can see to your problem:
> > 
> >     1. Buy more RAM. If you can afford 10TB of disk, then you can
> >        afford to buy at least a couple of GB of RAM to go with it.
> > 
> >     2. Remake your filesystem with a smaller log so that
> >            it can't hold as many active items.
> It is nasty that XFS can get into this state though. Would it make
> sense to limit the in memory log based on available memory? 

Well, if you can define "available memory" in any sane way in
the context we are operating in then that would work.

> With such a limit it would still work, but slower, right?


However, it would be far better to limit the number of active inode
items, as we do need to write the inodes back to the filesystem at some
point and it's better to trickle them out than have to write
them all out in one hit....

Besides, holding several hundred thousand active items has
scalability issues associated with it as well....


Dave Chinner
R&D Software Enginner
SGI Australian Software Group

<Prev in Thread] Current Thread [Next in Thread>