diff mbox series

[v2] ocfs2: call ocfs2_abort when journal abort

Message ID 20231121122523.1859827-1-srivathsa.d.dara@oracle.com (mailing list archive)
State New
Headers show
Series [v2] ocfs2: call ocfs2_abort when journal abort | expand

Commit Message

Srivathsa Dara Nov. 21, 2023, 12:25 p.m. UTC
From: Ryan Ding <ryan.ding@oracle.com>

Journal can not recover from abort state, so we should panic. Because we
can not release lock resource in this state, other node will hung when it
require a lock owned by this node. So, panic and remaster is a reasonable
choice.

ocfs2_abort() will do above work.

Co-Developed-by: Srivathsa Dara <srivathsa.d.dara@oracle.com>
Signed-off-by: Ryan Ding <ryan.ding@oracle.com>
Signed-off-by: Srivathsa Dara <srivathsa.d.dara@oracle.com>
---
v1-> v2: Remove redundant code and comments related to local mount.
 fs/ocfs2/journal.c | 16 ++++------------
 1 file changed, 4 insertions(+), 12 deletions(-)
diff mbox series

Patch

diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index ce215565d061..52d71e9ded2e 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -14,7 +14,6 @@ 
 #include <linux/kthread.h>
 #include <linux/time.h>
 #include <linux/random.h>
-#include <linux/delay.h>
 #include <linux/writeback.h>
 
 #include <cluster/masklog.h>
@@ -2326,7 +2325,7 @@  static int __ocfs2_wait_on_mount(struct ocfs2_super *osb, int quota)
 
 static int ocfs2_commit_thread(void *arg)
 {
-	int status;
+	int status = 0;
 	struct ocfs2_super *osb = arg;
 	struct ocfs2_journal *journal = osb->journal;
 
@@ -2343,18 +2342,11 @@  static int ocfs2_commit_thread(void *arg)
 
 		status = ocfs2_commit_cache(osb);
 		if (status < 0) {
-			static unsigned long abort_warn_time;
-
-			/* Warn about this once per minute */
-			if (printk_timed_ratelimit(&abort_warn_time, 60*HZ))
-				mlog(ML_ERROR, "status = %d, journal is "
-						"already aborted.\n", status);
 			/*
-			 * After ocfs2_commit_cache() fails, j_num_trans has a
-			 * non-zero value.  Sleep here to avoid a busy-wait
-			 * loop.
+			 * Journal can not recover from abort state, there is
+			 * no need to keep commit cache. So, we should panic.
 			 */
-			msleep_interruptible(1000);
+			ocfs2_abort(osb->sb, "Detected aborted journal");
 		}
 
 		if (kthread_should_stop() && atomic_read(&journal->j_num_trans)){