similar to: [PATCH] ocfs2: unlock rw lock if inode lock failed

Currently, default behavior of O_DIRECT writes was allowing concurrent writing among nodes, no cluster coherency guaranteed (no EX locks was taken), it hurts buffered reads on other nodes by reading stale data from cache. The new mount option introduce a chance to choose two different behaviors for O_DIRECT writes: * coherency=full, as the default value, will disallow concurrent

Hi, I'm resending this patch series. It's rediffed against linux-next branch of Joel's git tree. The first four patches are obvious fixes of deadlocks in quota code and should go in as soon as possible. The other three patches implement lockdep support for OCFS2 cluster locks. So you can have a look whether the code make sence to you and possibly merge them. They should be NOP when

Former logic of ocfs2_file_aio_write() was a bit stricky to unlock the rw_lock and i_alloc_sem, by using some private bits in struct 'iocb' to communite with ocfs2_dio_end_io(), it did work before we introduce the patch of supporting 'coherency=full,buffered' option, since rw_lock and i_alloc_sem were never acquired both at the same time, no mattar we doing buffered or direct IO or

in ocfs2_file_aio_write(), log_exit() could don't log the value which is really returned. this patch fixes it. Signed-off-by: Wengang Wang <wen.gang.wang at oracle.com> --- fs/ocfs2/file.c | 4 +++- 1 files changed, 3 insertions(+), 1 deletions(-) diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 62442e4..a49fa44 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -1918,8

i_alloc_sem has always been a bit of an odd "lock". It''s the only remaining rw_semaphore that can be released by a different thread than the one that locked it, and it''s use case in the core direct I/O code is more like a counter given that the writers already have external serialization. This series removes it in favour of a simpler counter scheme, thus getting rid

Hi, the first four patches in this series fix locking problems in OCFS2 quota code (three of them can lead to potential deadlocks). The fifth patch reorders ip_alloc_sem for directories to be acquired before localalloc locks. Mark would you please merge these? The last two patches implement lockdep annotations for OCFS2 cluster locks. We annotate all the cluster locks except for special ones

Hi, here comes the next version of OCFS2 lockdep support. I've dropped patches with fixes from the series since they were already merged. As Joel suggested, I've simplified the main patch a bit so that we don't have ifdefs around lock declarations and there are also a few other minor improvements. Honza

llseek requires ocfs2 inode lock for updating the file size in SEEK_END. because the file size maybe update on another node. if it not . after call llseek in SEEK_END. the position is old. this bug can be reproduce the following scenario: at first ,we dd a test fileA,the file size is 10k. on NodeA: --------- 1) open the test fileA, lseek the end of file. and print the position. 2) close the test

Hi, "status = -30" means it has encountered EROFS when start transaction. And system panic is because s_mount_opt is set to OCFS2_MOUNT_ERRORS_PANIC in __ocfs2_abort, ideal with OCFS2_MOUNT_ERRORS_PANIC first in ocfs2_handle_error. so I think that it is not reasonable, Therefore, this setting shall be canceled in __ocfs2_abort. thanks ________________________________ zhangguanghui

On 2015/10/14 15:49, Zhangguanghui wrote: > OCFS2 is often used in high-availaibility systems, This patch enhances robustness for the filesystem. > but storage network is unstable?it still triggers a panic? such as ocfs2_start_trans -> __ocfs2_abort ->panic. > The 's_mount_opt' should depend on the mount option set, If errors=continue is set, > mark as a EIO error,

This email is also sent to cluster-devel at redhat.com. Since this issue is about both dlm and ocfs2, I send the email here to look for help from upstream. This is an already known issue. on ocfs2 with user space cluster stack, run the test script from http://people.redhat.com/~teigland/make_panic on the mounted ocfs2 volume from 2 nodes simultaneously, the access to ocfs2 volume on both nodes

On 10/14/2015 03:57 AM, Joseph Qi wrote: > On 2015/10/14 16:45, Zhangguanghui wrote: >> Hi, >> "status = -30" means it has encountered EROFS when start transaction. >> And system panic is because s_mount_opt is set to OCFS2_MOUNT_ERRORS_PANIC in __ocfs2_abort, >> ideal with OCFS2_MOUNT_ERRORS_PANIC first in ocfs2_handle_error. >> so I think that it is

OCFS2 is often used in high-availaibility systems, This patch enhances robustness for the filesystem. but storage network is unstable?it still triggers a panic? such as ocfs2_start_trans -> __ocfs2_abort ->panic. The 's_mount_opt' should depend on the mount option set, If errors=continue is set, mark as a EIO error, change OCFS2_MOUNT_ERRORS_PANIC to OCFS2_MOUNT_ERRORS_CONT in

OCFS2 is often used in high-availaibility systems, This patch enhances robustness for the filesystem. but storage network is unstable?it still triggers a panic? such as ocfs2_start_trans -> __ocfs2_abort ->panic. The 's_mount_opt' should depend on the mount option set, If errors=continue is set, mark as a EIO error, change OCFS2_MOUNT_ERRORS_PANIC to OCFS2_MOUNT_ERRORS_CONT in

For nfs exporting, ocfs2_get_dentry() returns the dentry for fh. ocfs2_get_dentry() may read from disk(when inode not in memory) without any cross cluster lock. this leads to load a stale inode. this patch fixes above problem. solution is that in case of inode is not in memory, we get the cluster lock(PR) of alloc inode where the inode in question is allocated from(this causes node on which

We should call ocfs2_inode_lock_atime instead of ocfs2_inode_lock in ocfs2_file_splice_read like we do in ocfs2_file_aio_read so that we can update atime in splice read if necessary. Signed-off-by: Tao Ma <tao.ma at oracle.com> --- fs/ocfs2/file.c | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 775ac34..95cba48 100644

Hello, I've hit a bug in OCFS2 delete code which results in inodes being left on disk without any links to them. The workload triggering this creates directories on one node and deletes them on another node in the cluster. The inode is not deleted because both nodes bail out from ocfs2_delete_inode() with: Skipping delete of 100405 because it is in use on other nodes The scenario which I

This patch plugs a race between the downconvert thread and an unlock ast message. Specifically, after the downconvert worker has done its task, the dc thread needs to check whether an unlock ast made the downconvert moot. Reported-by: David Teigland <teigland at redhat.com> Signed-off-by: Sunil Mushran <sunil.mushran at oracle.com> Acked-by: Mark Fasheh <mfasheh at sus.com> ---

Resending after implementing review comments.

Resending after implementing review comments.