* Re: v0.19.1 stable updates
@ 2010-03-09 11:48 Anton
2010-03-09 13:29 ` Roland Rabben
2010-03-10 13:33 ` Anton
0 siblings, 2 replies; 6+ messages in thread
From: Anton @ 2010-03-09 11:48 UTC (permalink / raw)
To: ceph-devel; +Cc: Sage Weil, ceph-devel
Sage, do you think CEPH now is ok (stable enough) to be
atleast sure that uploaded data will not be corrupted, (and
could be recovered in the case of fault) - for use as simple
file store - say in the scenario of "upload/modify rarely,
read often"?. Right now I use Gluster as clustered FS - but
overhead of 1+1 replication in Gluster is just HUGE
Regards,
Anton.
On Wednesday 03 March 2010, Sage Weil wrote:
> Hi all,
>
> We've pushed out some stable updates to master granches
> of the git trees and tagged a 0.19.1 stable release.
> For the server side, the fixes are minor: mainly a
> problem with /etc/init.d/ceph when /bin/sh is dash
> (ubuntu) and a rare mds crash. The kernel client fixes
> are a bit more significant: mainly a bad write(2) return
> value and a crash when snapshots were deleted.
>
> sage
>
> -- server side ------------------
>
> Sage Weil (8):
> debian: mount.ceph in /sbin, not /usr/sbin
> Makefile: include debian/
> Makefile: fix /sbin hack
> mds: fix sessionmap decoding
> init-ceph: don't barf on dash when no command
> cauthtool: --caps fn alone is a command
> debian: new release, push, build, publish scripts
> 0.19.1
>
> Yehuda Sadeh (1):
> automake: fix mount sbin dir when configured with
> prefix
>
>
> -- kernel client -----------------
>
> Alexander Beregalov (1):
> ceph: move dereference after NULL test
>
> Sage Weil (5):
> ceph: fix handle_forward parsing
> ceph: reset front len on return to msgpool; BUG on
> mismatched front iov ceph: set osd request message front
> length correctly ceph: fix osdmap decoding when pools
> include (removed) snaps ceph: v0.19.1 stable release
>
> Yehuda Sadeh (1):
> ceph: don't clobber write return value when using
> O_SYNC
>
>
> ---------------------------------------------------------
> --------------------- Download Intel® Parallel
> Studio Eval
> Try the new software tools for yourself. Speed compiling,
> find bugs proactively, and fine-tune applications for
> parallel performance. See why Intel Parallel Studio got
> high marks during beta. http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Ceph-devel mailing list
> Ceph-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/ceph-devel
>
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: v0.19.1 stable updates
2010-03-09 11:48 v0.19.1 stable updates Anton
@ 2010-03-09 13:29 ` Roland Rabben
2010-03-10 13:33 ` Anton
1 sibling, 0 replies; 6+ messages in thread
From: Roland Rabben @ 2010-03-09 13:29 UTC (permalink / raw)
To: Anton; +Cc: Sage Weil, ceph-devel, ceph-devel
[-- Attachment #1.1: Type: text/plain, Size: 3330 bytes --]
Same story with me. Currently a frustrated Gluster user waiting for that
message saying things are safe to store data with CEPH - in production.
Regards,
Roland
2010/3/9 Anton <anton.vazir@gmail.com>
> Sage, do you think CEPH now is ok (stable enough) to be
> atleast sure that uploaded data will not be corrupted, (and
> could be recovered in the case of fault) - for use as simple
> file store - say in the scenario of "upload/modify rarely,
> read often"?. Right now I use Gluster as clustered FS - but
> overhead of 1+1 replication in Gluster is just HUGE
>
> Regards,
> Anton.
>
> On Wednesday 03 March 2010, Sage Weil wrote:
> > Hi all,
> >
> > We've pushed out some stable updates to master granches
> > of the git trees and tagged a 0.19.1 stable release.
> > For the server side, the fixes are minor: mainly a
> > problem with /etc/init.d/ceph when /bin/sh is dash
> > (ubuntu) and a rare mds crash. The kernel client fixes
> > are a bit more significant: mainly a bad write(2) return
> > value and a crash when snapshots were deleted.
> >
> > sage
> >
> > -- server side ------------------
> >
> > Sage Weil (8):
> > debian: mount.ceph in /sbin, not /usr/sbin
> > Makefile: include debian/
> > Makefile: fix /sbin hack
> > mds: fix sessionmap decoding
> > init-ceph: don't barf on dash when no command
> > cauthtool: --caps fn alone is a command
> > debian: new release, push, build, publish scripts
> > 0.19.1
> >
> > Yehuda Sadeh (1):
> > automake: fix mount sbin dir when configured with
> > prefix
> >
> >
> > -- kernel client -----------------
> >
> > Alexander Beregalov (1):
> > ceph: move dereference after NULL test
> >
> > Sage Weil (5):
> > ceph: fix handle_forward parsing
> > ceph: reset front len on return to msgpool; BUG on
> > mismatched front iov ceph: set osd request message front
> > length correctly ceph: fix osdmap decoding when pools
> > include (removed) snaps ceph: v0.19.1 stable release
> >
> > Yehuda Sadeh (1):
> > ceph: don't clobber write return value when using
> > O_SYNC
> >
> >
> > ---------------------------------------------------------
> > --------------------- Download Intel® Parallel
> > Studio Eval
> > Try the new software tools for yourself. Speed compiling,
> > find bugs proactively, and fine-tune applications for
> > parallel performance. See why Intel Parallel Studio got
> > high marks during beta. http://p.sf.net/sfu/intel-sw-dev
> > _______________________________________________
> > Ceph-devel mailing list
> > Ceph-devel@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/ceph-devel
> >
>
>
>
>
> ------------------------------------------------------------------------------
> Download Intel® Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Ceph-devel mailing list
> Ceph-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/ceph-devel
>
--
Roland Rabben
Founder & CEO Jotta AS
Cell: +47 90 85 85 39
Phone: +47 21 04 29 00
Email: roland@jotta.no
[-- Attachment #1.2: Type: text/html, Size: 4670 bytes --]
[-- Attachment #2: Type: text/plain, Size: 345 bytes --]
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
[-- Attachment #3: Type: text/plain, Size: 161 bytes --]
_______________________________________________
Ceph-devel mailing list
Ceph-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ceph-devel
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: v0.19.1 stable updates
2010-03-09 11:48 v0.19.1 stable updates Anton
2010-03-09 13:29 ` Roland Rabben
@ 2010-03-10 13:33 ` Anton
2010-03-10 16:43 ` Sage Weil
1 sibling, 1 reply; 6+ messages in thread
From: Anton @ 2010-03-10 13:33 UTC (permalink / raw)
To: ceph-devel; +Cc: Sage Weil, ceph-devel
Considering the silence,
I understand that stuff still in the heavy development, but
0.19 have been named "stable" and I thought that its time to
ask again, but anyway looks like it's time to test. :)
On Tuesday 09 March 2010, Anton wrote:
> Sage, do you think CEPH now is ok (stable enough) to be
> atleast sure that uploaded data will not be corrupted,
> (and could be recovered in the case of fault) - for use
> as simple file store - say in the scenario of
> "upload/modify rarely, read often"?. Right now I use
> Gluster as clustered FS - but overhead of 1+1
> replication in Gluster is just HUGE
>
> Regards,
> Anton.
>
> On Wednesday 03 March 2010, Sage Weil wrote:
> > Hi all,
> >
> > We've pushed out some stable updates to master granches
> > of the git trees and tagged a 0.19.1 stable release.
> > For the server side, the fixes are minor: mainly a
> > problem with /etc/init.d/ceph when /bin/sh is dash
> > (ubuntu) and a rare mds crash. The kernel client
> > fixes are a bit more significant: mainly a bad write(2)
> > return value and a crash when snapshots were deleted.
> >
> > sage
> >
> > -- server side ------------------
> >
> > Sage Weil (8):
> > debian: mount.ceph in /sbin, not /usr/sbin
> > Makefile: include debian/
> > Makefile: fix /sbin hack
> > mds: fix sessionmap decoding
> > init-ceph: don't barf on dash when no command
> > cauthtool: --caps fn alone is a command
> > debian: new release, push, build, publish scripts
> > 0.19.1
> >
> > Yehuda Sadeh (1):
> > automake: fix mount sbin dir when configured with
> > prefix
> >
> >
> > -- kernel client -----------------
> >
> > Alexander Beregalov (1):
> > ceph: move dereference after NULL test
> >
> > Sage Weil (5):
> > ceph: fix handle_forward parsing
> > ceph: reset front len on return to msgpool; BUG
> > on mismatched front iov ceph: set osd request message
> > front length correctly ceph: fix osdmap decoding when
> > pools include (removed) snaps ceph: v0.19.1 stable
> > release
> >
> > Yehuda Sadeh (1):
> > ceph: don't clobber write return value when using
> > O_SYNC
> >
> >
> > -------------------------------------------------------
> >-- --------------------- Download Intel® Parallel
> > Studio Eval
> > Try the new software tools for yourself. Speed
> > compiling, find bugs proactively, and fine-tune
> > applications for parallel performance. See why Intel
> > Parallel Studio got high marks during beta.
> > http://p.sf.net/sfu/intel-sw-dev
> > _______________________________________________
> > Ceph-devel mailing list
> > Ceph-devel@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/ceph-devel
>
> ---------------------------------------------------------
> --------------------- Download Intel® Parallel
> Studio Eval
> Try the new software tools for yourself. Speed compiling,
> find bugs proactively, and fine-tune applications for
> parallel performance. See why Intel Parallel Studio got
> high marks during beta. http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Ceph-devel mailing list
> Ceph-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/ceph-devel
>
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: v0.19.1 stable updates
2010-03-10 13:33 ` Anton
@ 2010-03-10 16:43 ` Sage Weil
0 siblings, 0 replies; 6+ messages in thread
From: Sage Weil @ 2010-03-10 16:43 UTC (permalink / raw)
To: Anton; +Cc: ceph-devel, ceph-devel
Hi Anton,
On Wed, 10 Mar 2010, Anton wrote:
> Considering the silence,
> I understand that stuff still in the heavy development, but
> 0.19 have been named "stable" and I thought that its time to
> ask again, but anyway looks like it's time to test. :)
Sorry for the slow reply, this got lost in my mail yesterday.
> On Tuesday 09 March 2010, Anton wrote:
> > Sage, do you think CEPH now is ok (stable enough) to be
> > atleast sure that uploaded data will not be corrupted,
> > (and could be recovered in the case of fault) - for use
> > as simple file store - say in the scenario of
> > "upload/modify rarely, read often"?. Right now I use
> > Gluster as clustered FS - but overhead of 1+1
> > replication in Gluster is just HUGE
The disk format in 0.19 is stable, which means we will maintain
compatibility with new version going forward so that you won't have to
reformat the file system.
The system itself is still under heavy development, however, and should
not be trusted with important data. The latest code is stable enough that
we would benefit from broader testing on a wider range of hardware and
deployment scenarios, and in a read-mostly environment I don't anticipate
many problems (read: please test!), but you sooner or later you will run
into trouble if you try to use the current code in production.
sage
> >
> > Regards,
> > Anton.
> >
> > On Wednesday 03 March 2010, Sage Weil wrote:
> > > Hi all,
> > >
> > > We've pushed out some stable updates to master granches
> > > of the git trees and tagged a 0.19.1 stable release.
> > > For the server side, the fixes are minor: mainly a
> > > problem with /etc/init.d/ceph when /bin/sh is dash
> > > (ubuntu) and a rare mds crash. The kernel client
> > > fixes are a bit more significant: mainly a bad write(2)
> > > return value and a crash when snapshots were deleted.
> > >
> > > sage
> > >
> > > -- server side ------------------
> > >
> > > Sage Weil (8):
> > > debian: mount.ceph in /sbin, not /usr/sbin
> > > Makefile: include debian/
> > > Makefile: fix /sbin hack
> > > mds: fix sessionmap decoding
> > > init-ceph: don't barf on dash when no command
> > > cauthtool: --caps fn alone is a command
> > > debian: new release, push, build, publish scripts
> > > 0.19.1
> > >
> > > Yehuda Sadeh (1):
> > > automake: fix mount sbin dir when configured with
> > > prefix
> > >
> > >
> > > -- kernel client -----------------
> > >
> > > Alexander Beregalov (1):
> > > ceph: move dereference after NULL test
> > >
> > > Sage Weil (5):
> > > ceph: fix handle_forward parsing
> > > ceph: reset front len on return to msgpool; BUG
> > > on mismatched front iov ceph: set osd request message
> > > front length correctly ceph: fix osdmap decoding when
> > > pools include (removed) snaps ceph: v0.19.1 stable
> > > release
> > >
> > > Yehuda Sadeh (1):
> > > ceph: don't clobber write return value when using
> > > O_SYNC
> > >
> > >
> > > -------------------------------------------------------
> > >-- --------------------- Download Intel® Parallel
> > > Studio Eval
> > > Try the new software tools for yourself. Speed
> > > compiling, find bugs proactively, and fine-tune
> > > applications for parallel performance. See why Intel
> > > Parallel Studio got high marks during beta.
> > > http://p.sf.net/sfu/intel-sw-dev
> > > _______________________________________________
> > > Ceph-devel mailing list
> > > Ceph-devel@lists.sourceforge.net
> > > https://lists.sourceforge.net/lists/listinfo/ceph-devel
> >
> > ---------------------------------------------------------
> > --------------------- Download Intel® Parallel
> > Studio Eval
> > Try the new software tools for yourself. Speed compiling,
> > find bugs proactively, and fine-tune applications for
> > parallel performance. See why Intel Parallel Studio got
> > high marks during beta. http://p.sf.net/sfu/intel-sw-dev
> > _______________________________________________
> > Ceph-devel mailing list
> > Ceph-devel@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/ceph-devel
> >
>
>
>
> ------------------------------------------------------------------------------
> Download Intel® Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Ceph-devel mailing list
> Ceph-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/ceph-devel
>
>
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: v0.19.1 stable updates
2010-03-02 19:19 Sage Weil
@ 2010-03-09 11:26 ` Anton
0 siblings, 0 replies; 6+ messages in thread
From: Anton @ 2010-03-09 11:26 UTC (permalink / raw)
To: ceph-devel; +Cc: Sage Weil, ceph-devel
Sage, do you think CEPH now is ok (stable enough) to be
atleast sure that uploaded data will not be corrupted, (and
could be recovered in the case of fault) - for use as simple
file store - say in the scenario of "upload/modify rarely,
read often"?. Right now I use Gluster as clustered FS - but
overhead of 1+1 replication in Gluster is just HUGE
Regards,
Anton.
On Wednesday 03 March 2010, Sage Weil wrote:
> Hi all,
>
> We've pushed out some stable updates to master granches
> of the git trees and tagged a 0.19.1 stable release.
> For the server side, the fixes are minor: mainly a
> problem with /etc/init.d/ceph when /bin/sh is dash
> (ubuntu) and a rare mds crash. The kernel client fixes
> are a bit more significant: mainly a bad write(2) return
> value and a crash when snapshots were deleted.
>
> sage
>
> -- server side ------------------
>
> Sage Weil (8):
> debian: mount.ceph in /sbin, not /usr/sbin
> Makefile: include debian/
> Makefile: fix /sbin hack
> mds: fix sessionmap decoding
> init-ceph: don't barf on dash when no command
> cauthtool: --caps fn alone is a command
> debian: new release, push, build, publish scripts
> 0.19.1
>
> Yehuda Sadeh (1):
> automake: fix mount sbin dir when configured with
> prefix
>
>
> -- kernel client -----------------
>
> Alexander Beregalov (1):
> ceph: move dereference after NULL test
>
> Sage Weil (5):
> ceph: fix handle_forward parsing
> ceph: reset front len on return to msgpool; BUG on
> mismatched front iov ceph: set osd request message front
> length correctly ceph: fix osdmap decoding when pools
> include (removed) snaps ceph: v0.19.1 stable release
>
> Yehuda Sadeh (1):
> ceph: don't clobber write return value when using
> O_SYNC
>
>
> ---------------------------------------------------------
> --------------------- Download Intel® Parallel
> Studio Eval
> Try the new software tools for yourself. Speed compiling,
> find bugs proactively, and fine-tune applications for
> parallel performance. See why Intel Parallel Studio got
> high marks during beta. http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Ceph-devel mailing list
> Ceph-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/ceph-devel
>
--
Anton V. Gnitko
Technical Director
Eastera Co. Ltd.
+992 43 3770123
+992 37 2270101
http://www.eastera.tj
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
^ permalink raw reply [flat|nested] 6+ messages in thread
* v0.19.1 stable updates
@ 2010-03-02 19:19 Sage Weil
2010-03-09 11:26 ` Anton
0 siblings, 1 reply; 6+ messages in thread
From: Sage Weil @ 2010-03-02 19:19 UTC (permalink / raw)
To: ceph-devel
Hi all,
We've pushed out some stable updates to master granches of the git trees
and tagged a 0.19.1 stable release. For the server side, the fixes are
minor: mainly a problem with /etc/init.d/ceph when /bin/sh is dash
(ubuntu) and a rare mds crash. The kernel client fixes are a bit more
significant: mainly a bad write(2) return value and a crash when snapshots
were deleted.
sage
-- server side ------------------
Sage Weil (8):
debian: mount.ceph in /sbin, not /usr/sbin
Makefile: include debian/
Makefile: fix /sbin hack
mds: fix sessionmap decoding
init-ceph: don't barf on dash when no command
cauthtool: --caps fn alone is a command
debian: new release, push, build, publish scripts
0.19.1
Yehuda Sadeh (1):
automake: fix mount sbin dir when configured with prefix
-- kernel client -----------------
Alexander Beregalov (1):
ceph: move dereference after NULL test
Sage Weil (5):
ceph: fix handle_forward parsing
ceph: reset front len on return to msgpool; BUG on mismatched front iov
ceph: set osd request message front length correctly
ceph: fix osdmap decoding when pools include (removed) snaps
ceph: v0.19.1 stable release
Yehuda Sadeh (1):
ceph: don't clobber write return value when using O_SYNC
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2010-03-10 16:43 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2010-03-09 11:48 v0.19.1 stable updates Anton
2010-03-09 13:29 ` Roland Rabben
2010-03-10 13:33 ` Anton
2010-03-10 16:43 ` Sage Weil
-- strict thread matches above, loose matches on Subject: below --
2010-03-02 19:19 Sage Weil
2010-03-09 11:26 ` Anton
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.