From: MORITA Kazutaka <morita.kazut...@lab.ntt.co.jp>
The current recovery code are not expected to work in multi threads. However,
cluster_enable_recover() can queue a new recovery work even if there is a
running one. To avoid this problem, this patch uses an ordered work queue for
the recovery work.
Signed-off-by: MORITA Kazutaka <morita.kazut...@lab.ntt.co.jp>
---
sheep/sheep.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/sheep/sheep.c b/sheep/sheep.c
index a7e3820..05fced5 100644
--- a/sheep/sheep.c
+++ b/sheep/sheep.c
@@ -338,7 +338,7 @@ static int init_work_queues(void)
sys->gateway_wqueue = init_work_queue("gway", WQ_UNLIMITED);
sys->io_wqueue = init_work_queue("io", WQ_UNLIMITED);
- sys->recovery_wqueue = init_work_queue("rw", WQ_UNLIMITED);
+ sys->recovery_wqueue = init_ordered_work_queue("rw");
sys->deletion_wqueue = init_ordered_work_queue("deletion");
sys->block_wqueue = init_ordered_work_queue("block");
sys->sockfd_wqueue = init_ordered_work_queue("sockfd");
--
1.7.9.5
--
sheepdog mailing list
sheepdog@lists.wpkg.org
http://lists.wpkg.org/mailman/listinfo/sheepdog
