Re: [sheepdog] [PATCH v4] sheepdog: selectable object size support

(2015/02/02 15:52), Liu Yuan wrote: On Tue, Jan 27, 2015 at 05:35:27PM +0900, Teruaki Ishizaki wrote: Previously, qemu block driver of sheepdog used hard-coded VDI object size. This patch enables users to handle "block_size_shift" value for calculating VDI object size. When you start qemu, you

[sheepdog] [PATCH 3/3] sheep: error massages in sheep_exec_request

normally sheep_exec_request succeeded by retry request. However, always sheep_exec_request shows ERROR. level of logs WARN is desirable. Signed-off-by: Masaki Saeki --- sheep/request.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/sheep/request.c b/sheep/request.c ind

[sheepdog] [PATCH 1/3] sheep: error messages during buffer resize

execute opcode "VDI_STATE_SNAPSHOT_CTL", sheepdog resize buffer if needed, so this situation is not error. Signed-off-by: Masaki Saeki --- sheep/ops.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/sheep/ops.c b/sheep/ops.c index bc2848b..7788460 100644 --- a/sheep/ops.

[sheepdog] [PATCH 0/3] sheep: adjust error loglevel

Some Sheepdog ERROR message information can not be said to be an error. These log, confuse the user unfamiliar to Sheepdog. So, we would like to suggest to reduce the level of some log. Signed-off-by: Masaki Saeki --- sheep/ops.c |2 +- sheep/recovery.c |2 +- sheep/request.c |

[sheepdog] [PATCH 2/3] sheep: error messages in recovery is too frequently

lacking objects during recovery is acceptable in the range of redundancy. So it is not necessary to show all error messages This problem occurs only in EC. because, ec object have not replicated object. Signed-off-by: Masaki Saeki --- sheep/recovery.c |2 +- 1 files changed, 1 insertions(+),

Re: [sheepdog] Redundancy policy via iSCSI

At Tue, 3 Feb 2015 17:17:42 +0800, hujianyang wrote: > > Hi Saeki, > > On 2015/2/3 16:53, Saeki Masaki wrote: > > Hi Hu, > > > > Since Sheepdog has a mechanism that does not place objects in the same > > zone_id. > > Can you try to change ZONE id in each node. > Id Host:Port V

[sheepdog] Build failed in Jenkins: sheepdog-build #623

See Changes: [mitake.hitoshi] deb: escape forward slashes in DAEMON_ARGS [liuyuan] Change tests outputs to be suitable for new cluster info format -- [...truncated 57 lines...] chec

Re: [sheepdog] Redundancy policy via iSCSI

On 2015/2/3 17:19, Bastian Scholz wrote: > Am 2015-02-03 09:54, schrieb hujianyang: >> I'm not clear with this. Just run "sheep /mnt/store/0 -z 0 -p 7000" >> on each host. > > I guess this is your problem. Sheep replicates in different zones, > since you have only one zone defined, sheep dont repl

Re: [sheepdog] Redundancy policy via iSCSI

Am 2015-02-03 09:54, schrieb hujianyang: I'm not clear with this. Just run "sheep /mnt/store/0 -z 0 -p 7000" on each host. I guess this is your problem. Sheep replicates in different zones, since you have only one zone defined, sheep dont replicate the data, it only distribute the data across y

Re: [sheepdog] Redundancy policy via iSCSI

Hi Saeki, On 2015/2/3 16:53, Saeki Masaki wrote: > Hi Hu, > > Since Sheepdog has a mechanism that does not place objects in the same > zone_id. > Can you try to change ZONE id in each node. Id Host:Port V-Nodes Zone 0 130.1.0.147:7000128 0

Re: [sheepdog] Redundancy policy via iSCSI

Hi Hu, Since Sheepdog has a mechanism that does not place objects in the same zone_id. Can you try to change ZONE id in each node. Id Host:Port V-Nodes Zone 0 130.1.0.147:7000128 0 1 130.1.0.148:7000128 0 2 130.1.0.149:7000

Re: [sheepdog] Redundancy policy via iSCSI

Hi Hitoshi, Thanks for your help! Try to answer your questions. On 2015/2/3 16:02, Hitoshi Mitake wrote: > > Hmm, seems strange. For diagnosing, I have some questions: > > 1. Can you see any error messages in log files of sheep? log while performing dd on a formatted filesystem: Feb 04 00:16

Re: [sheepdog] Redundancy policy via iSCSI

At Tue, 03 Feb 2015 17:02:24 +0900, Hitoshi Mitake wrote: > > > Hi Hu, > Thanks for your report! > > At Tue, 3 Feb 2015 15:52:12 +0800, > hujianyang wrote: > > > > Hi Hitoshi, > > > > Sorry for disturb. > > > > I'm testing redundancy policy of sheepdog via iSCSI. I think > > if I create a 1G

Re: [sheepdog] Redundancy policy via iSCSI

Hi Hu, Thanks for your report! At Tue, 3 Feb 2015 15:52:12 +0800, hujianyang wrote: > > Hi Hitoshi, > > Sorry for disturb. > > I'm testing redundancy policy of sheepdog via iSCSI. I think > if I create a 1G v-disk, the total space cost of this device > should be 3*1G under a 3 copies policy. B