Re: [sheepdog] [PATCH v2 0/3] selectable data object size of a VDI

2014-12-15 Thread Hitoshi Mitake
At Mon, 15 Dec 2014 14:30:59 +0900, Hitoshi Mitake wrote: > > At Fri, 12 Dec 2014 21:48:29 +0900, > Teruaki Ishizaki wrote: > > > > Previously users could use 4MB data object size for a VDI > > This patchset enables users to specify data object size of a VDI > > and make a larger VDI without Hype

[sheepdog] Build failed in Jenkins: sheepdog-build #573

2014-12-15 Thread sheepdog-jenkins
See Changes: [mitake.hitoshi] sheepdog proto: add block_size_shift member to sheepdog internal data. [mitake.hitoshi] sheep, dog: add "block_size_shift" option to cluster format command [mitake.hitoshi] sheep, dog: add

[sheepdog] [PATCH v2] sheep: let gateway node exit in a case of gateway only cluster

2014-12-15 Thread Hitoshi Mitake
When a cluster has gateway nodes only, it means the gateway nodes doesn't contribute to I/O of VMs. So this patch simply let them exit and avoid the below recovery issue. Related issue: https://bugs.launchpad.net/sheepdog-project/+bug/1327037 Cc: duron...@qq.com Cc: Yang Zhang <3100100...@zju.edu

Re: [sheepdog] [PATCH] sheep: let gateway node exit in a case of gateway only cluster

2014-12-15 Thread Hitoshi Mitake
At Mon, 15 Dec 2014 23:12:12 +0900, Hitoshi Mitake wrote: > > When a cluster has gateway nodes only, it means the gateway nodes > doesn't contribute to I/O of VMs. So this patch simply let them exit > and avoid the below recovery issue. > > Related issue: > https://bugs.launchpad.net/sheepdog-pro

[sheepdog] [PATCH] sheep: let gateway node exit in a case of gateway only cluster

2014-12-15 Thread Hitoshi Mitake
When a cluster has gateway nodes only, it means the gateway nodes doesn't contribute to I/O of VMs. So this patch simply let them exit and avoid the below recovery issue. Related issue: https://bugs.launchpad.net/sheepdog-project/+bug/1327037 Cc: duron...@qq.com Cc: Yang Zhang <3100100...@zju.edu

[sheepdog] [PATCH RFT 2/4] sheep: construct data structures for representing family tree

2014-12-15 Thread Hitoshi Mitake
This patch adds new data structres which represent VDI family tree. It is different from the existing vdi_state_entry tree. The purpose of these new data structres is efficient tracking of family relation between VDIs. It is required for garbage collecting needless VIDs. Cc: Saeki Masaki Cc: Yuka

[sheepdog] [PATCH RFT 3/4] libsheepdog: provide atomic_clear_bit

2014-12-15 Thread Hitoshi Mitake
The new function is required because the bitmaps (vdi_inuse and vdi_deleted) can be cleared now. Cc: Saeki Masaki Cc: Yuka Kawasaki Signed-off-by: Hitoshi Mitake --- include/bitops.h | 5 + 1 file changed, 5 insertions(+) diff --git a/include/bitops.h b/include/bitops.h index 86a8ac1..6b6

[sheepdog] [PATCH RFT 4/4] sheep: garbage collect needless VIDs

2014-12-15 Thread Hitoshi Mitake
Current sheepdog never recycles VIDs. But it will cause problems e.g. VID space exhaustion, too much garbage inode objects. Keeping deleted inode objects is required because living inodes (snapshots or clones) can point objects of the deleted inodes. So if every member of VDI family is deleted, it

[sheepdog] [PATCH RFT 0/4] garbage collect needless VIDs and inode objects

2014-12-15 Thread Hitoshi Mitake
Current sheepdog never recycles VIDs. But it will cause problems e.g. VID space exhaustion, too much garbage inode objects. Keeping deleted inode objects is required because living inodes (snapshots or clones) can point objects of the deleted inodes. So if every member of VDI family is deleted, it

[sheepdog] [PATCH RFT 1/4] sheep: let vdi state have parent VID

2014-12-15 Thread Hitoshi Mitake
This is a preparation patch. Cc: Saeki Masaki Cc: Yuka Kawasaki Signed-off-by: Hitoshi Mitake --- include/internal_proto.h | 1 + sheep/group.c| 3 ++- sheep/ops.c | 7 --- sheep/plain_store.c | 2 +- sheep/sheep_priv.h | 2 +- sheep/vdi.c