diff options
| author | Sunil Mushran <sunil.mushran@oracle.com> | 2008-07-15 02:31:10 +0200 |
|---|---|---|
| committer | Mark Fasheh <mfasheh@suse.com> | 2008-08-01 01:21:14 +0200 |
| commit | 539d8264093560b917ee3afe4c7f74e5da09d6a5 (patch) | |
| tree | 1fce83387272c0b2d61bd945769f4984aa5e79ce /fs/ocfs2/journal.h | |
| parent | [PATCH 1/2] ocfs2: Add counter in struct ocfs2_dinode to track journal replays (diff) | |
| download | linux-539d8264093560b917ee3afe4c7f74e5da09d6a5.tar.xz linux-539d8264093560b917ee3afe4c7f74e5da09d6a5.zip | |
[PATCH 2/2] ocfs2: Fix race between mount and recovery
As the fs recovery is asynchronous, there is a small chance that another
node can mount (and thus recover) the slot before the recovery thread
gets to it.
If this happens, the recovery thread will block indefinitely on the
journal/slot lock as that lock will be held for the duration of the mount
(by design) by the node assigned to that slot.
The solution implemented is to keep track of the journal replays using
a recovery generation in the journal inode, which will be incremented by the
thread replaying that journal. The recovery thread, before attempting the
blocking lock on the journal/slot lock, will compare the generation on disk
with what it has cached and skip recovery if it does not match.
This bug appears to have been inadvertently introduced during the mount/umount
vote removal by mainline commit 34d024f84345807bf44163fac84e921513dde323. In the
mount voting scheme, the messaging would indirectly indicate that the slot
was being recovered.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Diffstat (limited to 'fs/ocfs2/journal.h')
| -rw-r--r-- | fs/ocfs2/journal.h | 3 |
1 files changed, 2 insertions, 1 deletions
diff --git a/fs/ocfs2/journal.h b/fs/ocfs2/journal.h index db82be2532ed..2178ebffa05f 100644 --- a/fs/ocfs2/journal.h +++ b/fs/ocfs2/journal.h @@ -161,7 +161,8 @@ int ocfs2_journal_init(struct ocfs2_journal *journal, void ocfs2_journal_shutdown(struct ocfs2_super *osb); int ocfs2_journal_wipe(struct ocfs2_journal *journal, int full); -int ocfs2_journal_load(struct ocfs2_journal *journal, int local); +int ocfs2_journal_load(struct ocfs2_journal *journal, int local, + int replayed); int ocfs2_check_journals_nolocks(struct ocfs2_super *osb); void ocfs2_recovery_thread(struct ocfs2_super *osb, int node_num); |