diff options
| author | Sunil Mushran <sunil.mushran@oracle.com> | 2007-04-17 13:53:38 -0700 |
|---|---|---|
| committer | Mark Fasheh <mark.fasheh@oracle.com> | 2007-04-26 13:39:02 -0700 |
| commit | bebe6f120b036349f7212205eeaf8248d4820c4b (patch) | |
| tree | 18018b7d54d05743f408c64bad78f08907f3b9f8 | |
| parent | 5d262cc7dd3d47784f8233ad4ec2cc5a08059b71 (diff) | |
ocfs2: Replace panic() with emergency_restart() when fencing
We have noticed panic() hanging leading us to a situation in which
the node, while otherwise dead, is still disk heartbeating. This
leads to a hung cluster as the other nodes are waiting for this
node to stop disk heartbeating. This situation is only resolved
by power resetting the box.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
| -rw-r--r-- | fs/ocfs2/cluster/quorum.c | 5 |
1 files changed, 4 insertions, 1 deletions
diff --git a/fs/ocfs2/cluster/quorum.c b/fs/ocfs2/cluster/quorum.c index 4705d659fe5..bbacf7da48a 100644 --- a/fs/ocfs2/cluster/quorum.c +++ b/fs/ocfs2/cluster/quorum.c @@ -46,6 +46,7 @@ #include <linux/kernel.h> #include <linux/slab.h> #include <linux/workqueue.h> +#include <linux/reboot.h> #include "heartbeat.h" #include "nodemanager.h" @@ -72,7 +73,9 @@ static void o2quo_fence_self(void) /* panic spins with interrupts enabled. with preempt * threads can still schedule, etc, etc */ o2hb_stop_all_regions(); - panic("ocfs2 is very sorry to be fencing this system by panicing\n"); + + printk("ocfs2 is very sorry to be fencing this system by restarting\n"); + emergency_restart(); } /* Indicate that a timeout occured on a hearbeat region write. The |