From: Robin Dong san...@taobao.com
Add locks to protect containers and objects when users create/delete
containers or create/delete objects in the same time.
Signed-off-by: Robin Dong san...@taobao.com
---
sheep/http/kv.c| 242 -
From: Robin Dong san...@taobao.com
Using hyper volume and extent structure of onode to store large number of
objects size of which exceed SD_DATA_OBJ_SIZE.
Signed-off-by: Robin Dong san...@taobao.com
---
sheep/http/kv.c| 427 +++--
From: Robin Dong san...@taobao.com
Using hyper volume (size up to 16PB) to store large number of accounts
and containers.
Signed-off-by: Robin Dong san...@taobao.com
---
sheep/http/http.c | 5 +
sheep/http/http.h | 1 +
sheep/http/kv.c| 627
At Wed, 20 Nov 2013 22:04:28 +0800,
Liu Yuan wrote:
On Wed, Nov 20, 2013 at 10:54:02PM +0900, Hitoshi Mitake wrote:
Current dog vdi check cannot work well in a case of inode object
lost. This patch lets the subcommand handle the case correctly.
Signed-off-by: Hitoshi Mitake
From: Liu Yuan namei.u...@gmail.com
Signed-off-by: Liu Yuan namei.u...@gmail.com
Signed-off-by: MORITA Kazutaka morita.kazut...@lab.ntt.co.jp
Conflicts:
sheep/md.c
Conflict was resolved by Hitoshi Mitake.
Signed-off-by: Hitoshi Mitake mitake.hito...@lab.ntt.co.jp
---
sheep/md.c | 1 -
This update only contains trivial changes.
1. remove dead code in logger
2. remove a needless log message in md
3. resolve a build error of sheepfs on Ubuntu 13.01
Basically, users don't have to care about it. It is a preparation for coming big
update.
This patchset contains manual modification
From: MORITA Kazutaka morita.kazut...@lab.ntt.co.jp
The logger process runs with one thread, so no need for
pthread_mutex_lock.
Signed-off-by: MORITA Kazutaka morita.kazut...@lab.ntt.co.jp
Signed-off-by: Liu Yuan namei.u...@gmail.com
---
lib/logger.c | 4
1 file changed, 4 deletions(-)
Because libsheepdog wraps pthread reader writer lock, the flag is
required for building sheepfs too.
Signed-off-by: Hitoshi Mitake mitake.hito...@lab.ntt.co.jp
Signed-off-by: MORITA Kazutaka morita.kazut...@lab.ntt.co.jp
---
sheepfs/Makefile.am | 2 +-
1 file changed, 1 insertion(+), 1
At Tue, 10 Dec 2013 12:36:06 +0800,
Liu Yuan wrote:
On Mon, Dec 09, 2013 at 08:49:45PM +0900, Teruaki Ishizaki wrote:
Previously sheep.log was saved in the Sheepdog store directory,
and used size of Sheepdog store directory included log file size.
So, administrator couldn't do capacity
On Wed, 11 Dec 2013 10:22:05 +0900, Hitoshi Mitake wrote:
At Tue, 10 Dec 2013 12:36:06 +0800,
Liu Yuan wrote:
On Mon, Dec 09, 2013 at 08:49:45PM +0900, Teruaki Ishizaki wrote:
Previously sheep.log was saved in the Sheepdog store directory,
and used size of Sheepdog store directory
On Wed, Dec 11, 2013 at 10:31:54AM +0900, Ryusuke Konishi wrote:
On Wed, 11 Dec 2013 10:22:05 +0900, Hitoshi Mitake wrote:
At Tue, 10 Dec 2013 12:36:06 +0800,
Liu Yuan wrote:
On Mon, Dec 09, 2013 at 08:49:45PM +0900, Teruaki Ishizaki wrote:
Previously sheep.log was saved in the
Hi sheepdog users and developers,
We have a bugtracker on Launchpad:
https://bugs.launchpad.net/sheepdog-project
I'd like you to submit bugs and problems if they seem to be long
living ones. Because mailing lists are not a suitable place for this
purpose. Searching existing bugs from the
Why not adding the check in gateway_forward_request() ? I think we should
check the copy number before send forward the request out.
2013/12/10 Liu Yuan namei.u...@gmail.com
We make sure we write the exact number of copies to honor the promise of
the
redundancy for strict mode. This means
On Wed, Dec 11, 2013 at 10:59:27AM +0800, Robin Dong wrote:
Why not adding the check in gateway_forward_request() ? I think we should
check the copy number before send forward the request out.
I think it is better to check the request as early as possible, no? This is
pointless to run the
2013/12/11 Liu Yuan namei.u...@gmail.com
On Wed, Dec 11, 2013 at 10:59:27AM +0800, Robin Dong wrote:
Why not adding the check in gateway_forward_request() ? I think we should
check the copy number before send forward the request out.
I think it is better to check the request as early as
Reviewed-by: Robin Dong san...@taobao.com
2013/12/10 Liu Yuan namei.u...@gmail.com
Signed-off-by: Liu Yuan namei.u...@gmail.com
---
tests/functional/080 | 37 +
tests/functional/080.out | 14 ++
tests/functional/group |1 +
3
2013/12/11 Liu Yuan namei.u...@gmail.com:
On Wed, Dec 11, 2013 at 10:31:54AM +0900, Ryusuke Konishi wrote:
On Wed, 11 Dec 2013 10:22:05 +0900, Hitoshi Mitake wrote:
At Tue, 10 Dec 2013 12:36:06 +0800,
Liu Yuan wrote:
On Mon, Dec 09, 2013 at 08:49:45PM +0900, Teruaki Ishizaki wrote:
On Tue, Dec 10, 2013 at 06:41:04PM +0800, Robin Dong wrote:
From: Robin Dong san...@taobao.com
Using hyper volume (size up to 16PB) to store large number of accounts
and containers.
I can't apply this set on master. Could you please rebase it? And also please
write a diff what you update
Previously sheep.log was saved in the Sheepdog store directory,
and used size of Sheepdog store directory included log file size.
So, administrator couldn't do capacity planning simply.
This patch adds an option for setting the output directory of sheep.log.
Usage:
$ sheep -L
Previously, sheep had many log options, -l, -F , -d and -L.
This patch merge options to the option '-l'.
Usage:
$ sheep -l dir=log_directory,level=log_level_num,format=log_format
Example:
$ sheep -l dir=/logdir,level=7,format=server
Signed-off-by: Teruaki Ishizaki
On Wed, Dec 11, 2013 at 04:25:45PM +0900, Teruaki Ishizaki wrote:
Previously sheep.log was saved in the Sheepdog store directory,
and used size of Sheepdog store directory included log file size.
So, administrator couldn't do capacity planning simply.
This patch adds an option for setting
On Wed, Dec 11, 2013 at 04:30:15PM +0900, Teruaki Ishizaki wrote:
Previously, sheep had many log options, -l, -F , -d and -L.
This patch merge options to the option '-l'.
Usage:
$ sheep -l dir=log_directory,level=log_level_num,format=log_format
Example:
$ sheep -l
On Tue, Dec 10, 2013 at 11:29:03PM +0900, Hitoshi Mitake wrote:
This update only contains trivial changes.
1. remove dead code in logger
2. remove a needless log message in md
3. resolve a build error of sheepfs on Ubuntu 13.01
Basically, users don't have to care about it. It is a
23 matches
Mail list logo