ocfs2: Update default cluster timeouts

Lots of people are having trouble with the default timeouts, which are too
low. These new values are derived from an informal survey taken on
ocfs2-users, as well as data from bug reports. This should reduce the amount
of cluster disconnects and subsequent fencing seen during normal workloads.

Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>

diff --git a/fs/ocfs2/cluster/heartbeat.h b/fs/ocfs2/cluster/heartbeat.h
index 35397dd5ecdbb42c3922cbc9e930081c4c522b87..e511339886b31040e5597695726692b91be07e11 100644 (file)
--- a/fs/ocfs2/cluster/heartbeat.h
+++ b/fs/ocfs2/cluster/heartbeat.h
@@ -35,7 +35,7 @@
 #define O2HB_LIVE_THRESHOLD       2
 /* number of equal samples to be seen as dead */
 extern unsigned int o2hb_dead_threshold;
-#define O2HB_DEFAULT_DEAD_THRESHOLD       7
+#define O2HB_DEFAULT_DEAD_THRESHOLD       31
 /* Otherwise MAX_WRITE_TIMEOUT will be zero... */
 #define O2HB_MIN_DEAD_THRESHOLD          2
 #define O2HB_MAX_WRITE_TIMEOUT_MS (O2HB_REGION_TIMEOUT_MS * (o2hb_dead_threshold - 1))

diff --git a/fs/ocfs2/cluster/tcp.h b/fs/ocfs2/cluster/tcp.h
index da880fc215f075ccc50aac915810409b8efd35e4..f36f66aab3dd788e200ddca6be96546939af03ad 100644 (file)
--- a/fs/ocfs2/cluster/tcp.h
+++ b/fs/ocfs2/cluster/tcp.h
@@ -60,8 +60,8 @@ typedef void (o2net_post_msg_handler_func)(int status, void *data,
 /* same as hb delay, we're waiting for another node to recognize our hb */
 #define O2NET_RECONNECT_DELAY_MS_DEFAULT       2000
 
-#define O2NET_KEEPALIVE_DELAY_MS_DEFAULT       5000
-#define O2NET_IDLE_TIMEOUT_MS_DEFAULT          10000
+#define O2NET_KEEPALIVE_DELAY_MS_DEFAULT       2000
+#define O2NET_IDLE_TIMEOUT_MS_DEFAULT          30000
 
 
 /* TODO: figure this out.... */