From: Johann Lombardi <johann@sun.com>
To: lustre-devel@lists.lustre.org
Subject: [Lustre-devel] Moving forward on Quotas
Date: Wed, 28 May 2008 10:06:13 +0200 [thread overview]
Message-ID: <20080528080613.GN3582@lore> (raw)
In-Reply-To: <C461688A.5460%peter.braam@sun.com>
Hello Peter,
On Tue, May 27, 2008 at 07:28:10AM +0800, Peter Braam wrote:
> >> When a slave runs out of its local quota, it sends an acquire request to the
> >> quota master. As I said earlier, the quota master is the only one having a
> >> global overview of what has been granted to slaves. If the master can satisfy
> >> the request, it grants a qunit (can be a number of blocks or inodes) to the
> >> slave. The problem is that an OST can return "quota exceeded" (=EDQUOT)
> >> whereas
> >> another OST is still having quotas. There is currently no callback to claim
> >> back the quota space that has been granted to a slave.
>
> Hmm - the slave should release quota.
I don't think that the slave can make such a decision by itself since it does
not know that we are getting closer to the global quota limit. Only the master
is aware of this.
Actually, the scenario I described above can no longer happen - with recent
lustre versions at least - thanks to the dynamic qunit patch because the
master broadcasts to all the slaves the new qunit size when it is shrunk.
Cheers,
Johann
next prev parent reply other threads:[~2008-05-28 8:06 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <18490.63940.619731.992500@gargle.gargle.HOWL>
2008-05-26 23:28 ` [Lustre-devel] Moving forward on Quotas Peter Braam
2008-05-28 8:06 ` Johann Lombardi [this message]
2008-06-01 2:32 ` Peter Braam
2008-06-02 12:22 ` Johann Lombardi
2008-06-02 23:24 ` Andreas Dilger
2008-06-03 8:49 ` Landen tian
2008-06-04 1:24 ` Peter Braam
2008-06-04 7:05 ` Landen tian
2008-06-04 8:26 ` Johann Lombardi
2008-05-28 14:29 ` Ricardo M. Correia
2008-05-28 14:54 ` Nikita Danilov
2008-05-28 15:14 ` Ricardo M. Correia
2008-05-28 16:22 ` Nikita Danilov
2008-05-28 17:05 ` Ricardo M. Correia
2008-05-28 20:06 ` Nikita Danilov
2008-05-28 21:07 ` Ricardo M. Correia
2008-05-28 21:11 ` Nikita Danilov
2008-05-28 21:33 ` Ricardo M. Correia
2008-05-29 8:39 ` Nikita Danilov
[not found] ` <18496.11672.844774.815457@gargle.gargle.HOWL>
2008-05-31 15:31 ` Ricardo M. Correia
2008-05-31 15:49 ` Ricardo M. Correia
[not found] ` <1212247447.21348.70.camel@localhost>
2008-05-31 16:19 ` Nikita Danilov
2008-05-31 17:19 ` Ricardo M. Correia
2008-05-31 19:11 ` Nikita Danilov
2008-06-01 2:36 ` Peter Braam
2008-06-01 3:17 ` Mike Shapiro
2008-06-01 2:26 ` Peter Braam
2008-06-01 4:53 ` Jeff Bonwick
2008-06-01 13:58 ` Nikita Danilov
2008-06-03 0:50 ` Matthew Ahrens
2008-06-03 7:49 ` Nikita Danilov
2008-06-04 23:50 ` Matthew Ahrens
2008-05-28 15:24 ` Nikita Danilov
2008-05-31 10:25 ` Peter Braam
[not found] <92825021-D566-4805-9297-5EFBD3260D73@Sun.COM>
2008-06-01 2:44 ` Peter Braam
[not found] <20080605083957.GQ6283@lore>
2008-06-05 11:09 ` Peter Braam
2008-06-05 12:27 ` Johann Lombardi
2008-06-05 13:45 ` Peter Braam
2008-06-06 7:33 ` Johann Lombardi
2008-06-06 12:21 ` Peter Braam
2008-06-09 8:52 ` Yong Fan
2008-06-09 15:37 ` Peter Braam
2008-06-09 16:09 ` Yong Fan
2008-06-10 13:54 ` Yong Fan
2008-06-10 16:51 ` Peter Braam
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=20080528080613.GN3582@lore \
--to=johann@sun.com \
--cc=lustre-devel@lists.lustre.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.