diff options
author | Xavier Hernandez <xhernandez@datalab.es> | 2014-10-08 09:20:11 +0200 |
---|---|---|
committer | Vijay Bellur <vbellur@redhat.com> | 2014-10-21 11:38:45 -0700 |
commit | 56caf4349c8824fde70783fe404cc6f646dce149 (patch) | |
tree | 02bef10cd75c1422c3ed272d5ecd01edcb404033 /xlators/cluster/ec/src/ec-dir-write.c | |
parent | d57ecca6322a451242f4a2b7b5978de7c8f9088e (diff) |
ec: Fix self-heal issues
Problem: Doing an 'ls' of a directory that has been modified while one
of the bricks was down, sometimes returns the old directory
contents.
Cause: Directories are not marked when they are modified as files are.
The ec xlator balances requests amongst available and healthy
bricks. Since there is no way to detect that a directory is
out of date in one of the bricks, it is used from time to time
to return the directory contents.
Solution: Basically the solution consists in use versioning information
also for directories, however some additional changes have
been necessary.
Changes:
* Use directory versioning:
This required to lock full directory instead of a single entry for
all requests that add or remove entries from it. This is needed to
allow atomic version update. This affects the following fops:
create, mkdir, mknod, link, symlink, rename, unlink, rmdir
Another side effect is that opendir requires to do a previous
lookup to get versioning information and discard out of date
bricks for subsequent readdir(p) calls.
* Restrict directory self-heal:
Till now, when one discrepancy was found in lookup, a self-heal
was automatically started. This caused the versioning information
of a bad directory to be healed instantly, making the original
problem to reapear again.
To solve this, when a missing directory is detected in one or more
bricks on lookup or opendir fops, only a partial self-heal is
performed on it. A partial self-heal basically creates the
directory but does not restore any additional information.
This avoids that an 'ls' could repair the directory and cause the
problem to happen again. With this change, output of 'ls' is
always consistent. However, since the directory has been created
in the brick, this allows any other operation on it (create new
files, for example) to succeed on all bricks and not add additional
work to the self-heal process.
To force a self-heal of a directory, any other operation must be
done on it. For example a getxattr.
With these changes, the correct healing procedure that would avoid
inconsistent directory browsing consists on a post-order traversal
of directoriesi being healed. This way, the directory contents will
be healed before healing the directory itslef.
* Additional changes to fix self-heal errors
- Don't use fop->fd to decide between fd/loc.
open, opendir and create have an fd, but the correct data is in
loc.
- Fix incorrect management of bad bricks per inode/fd.
- Fix incorrect selection of fop's target bricks when there are bad
bricks involved.
- Improved ec_loc_parent() to always return a parent loc as
complete as possible.
Change-Id: Iaf3df174d7857da57d4a87b4a8740a7048b366ad
BUG: 1149726
Signed-off-by: Xavier Hernandez <xhernandez@datalab.es>
Reviewed-on: http://review.gluster.org/8916
Reviewed-by: Dan Lambright <dlambrig@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Diffstat (limited to 'xlators/cluster/ec/src/ec-dir-write.c')
-rw-r--r-- | xlators/cluster/ec/src/ec-dir-write.c | 34 |
1 files changed, 17 insertions, 17 deletions
diff --git a/xlators/cluster/ec/src/ec-dir-write.c b/xlators/cluster/ec/src/ec-dir-write.c index 92012855087..02961acb8bd 100644 --- a/xlators/cluster/ec/src/ec-dir-write.c +++ b/xlators/cluster/ec/src/ec-dir-write.c @@ -210,7 +210,7 @@ int32_t ec_manager_create(ec_fop_data_t * fop, int32_t state) /* Fall through */ case EC_STATE_LOCK: - ec_lock_prepare_entry(fop, &fop->loc[0]); + ec_lock_prepare_entry(fop, &fop->loc[0], 1); ec_lock(fop); return EC_STATE_DISPATCH; @@ -293,7 +293,7 @@ int32_t ec_manager_create(ec_fop_data_t * fop, int32_t state) case -EC_STATE_LOCK_REUSE: case EC_STATE_LOCK_REUSE: - ec_lock_reuse(fop, 1); + ec_lock_reuse(fop); return EC_STATE_UNLOCK; @@ -494,7 +494,7 @@ int32_t ec_manager_link(ec_fop_data_t * fop, int32_t state) // Parent entry of fop->loc[0] should be locked, but I don't // receive enough information to do it (fop->loc[0].parent is // NULL). - ec_lock_prepare_entry(fop, &fop->loc[1]); + ec_lock_prepare_entry(fop, &fop->loc[1], 1); ec_lock(fop); return EC_STATE_GET_SIZE_AND_VERSION; @@ -577,7 +577,7 @@ int32_t ec_manager_link(ec_fop_data_t * fop, int32_t state) case -EC_STATE_LOCK_REUSE: case EC_STATE_LOCK_REUSE: - ec_lock_reuse(fop, 0); + ec_lock_reuse(fop); return EC_STATE_UNLOCK; @@ -765,7 +765,7 @@ int32_t ec_manager_mkdir(ec_fop_data_t * fop, int32_t state) { case EC_STATE_INIT: case EC_STATE_LOCK: - ec_lock_prepare_entry(fop, &fop->loc[0]); + ec_lock_prepare_entry(fop, &fop->loc[0], 1); ec_lock(fop); return EC_STATE_DISPATCH; @@ -837,7 +837,7 @@ int32_t ec_manager_mkdir(ec_fop_data_t * fop, int32_t state) case -EC_STATE_LOCK_REUSE: case EC_STATE_LOCK_REUSE: - ec_lock_reuse(fop, 0); + ec_lock_reuse(fop); return EC_STATE_UNLOCK; @@ -1022,7 +1022,7 @@ int32_t ec_manager_mknod(ec_fop_data_t * fop, int32_t state) { case EC_STATE_INIT: case EC_STATE_LOCK: - ec_lock_prepare_entry(fop, &fop->loc[0]); + ec_lock_prepare_entry(fop, &fop->loc[0], 1); ec_lock(fop); return EC_STATE_DISPATCH; @@ -1094,7 +1094,7 @@ int32_t ec_manager_mknod(ec_fop_data_t * fop, int32_t state) case -EC_STATE_LOCK_REUSE: case EC_STATE_LOCK_REUSE: - ec_lock_reuse(fop, 0); + ec_lock_reuse(fop); return EC_STATE_UNLOCK; @@ -1277,8 +1277,8 @@ int32_t ec_manager_rename(ec_fop_data_t * fop, int32_t state) { case EC_STATE_INIT: case EC_STATE_LOCK: - ec_lock_prepare_entry(fop, &fop->loc[0]); - ec_lock_prepare_entry(fop, &fop->loc[1]); + ec_lock_prepare_entry(fop, &fop->loc[0], 1); + ec_lock_prepare_entry(fop, &fop->loc[1], 1); ec_lock(fop); return EC_STATE_GET_SIZE_AND_VERSION; @@ -1359,7 +1359,7 @@ int32_t ec_manager_rename(ec_fop_data_t * fop, int32_t state) case -EC_STATE_LOCK_REUSE: case EC_STATE_LOCK_REUSE: - ec_lock_reuse(fop, 0); + ec_lock_reuse(fop); return EC_STATE_UNLOCK; @@ -1533,7 +1533,7 @@ int32_t ec_manager_rmdir(ec_fop_data_t * fop, int32_t state) { case EC_STATE_INIT: case EC_STATE_LOCK: - ec_lock_prepare_entry(fop, &fop->loc[0]); + ec_lock_prepare_entry(fop, &fop->loc[0], 1); ec_lock(fop); return EC_STATE_DISPATCH; @@ -1597,7 +1597,7 @@ int32_t ec_manager_rmdir(ec_fop_data_t * fop, int32_t state) case -EC_STATE_LOCK_REUSE: case EC_STATE_LOCK_REUSE: - ec_lock_reuse(fop, 0); + ec_lock_reuse(fop); return EC_STATE_UNLOCK; @@ -1780,7 +1780,7 @@ int32_t ec_manager_symlink(ec_fop_data_t * fop, int32_t state) { case EC_STATE_INIT: case EC_STATE_LOCK: - ec_lock_prepare_entry(fop, &fop->loc[0]); + ec_lock_prepare_entry(fop, &fop->loc[0], 1); ec_lock(fop); return EC_STATE_DISPATCH; @@ -1852,7 +1852,7 @@ int32_t ec_manager_symlink(ec_fop_data_t * fop, int32_t state) case -EC_STATE_LOCK_REUSE: case EC_STATE_LOCK_REUSE: - ec_lock_reuse(fop, 0); + ec_lock_reuse(fop); return EC_STATE_UNLOCK; @@ -2031,7 +2031,7 @@ int32_t ec_manager_unlink(ec_fop_data_t * fop, int32_t state) { case EC_STATE_INIT: case EC_STATE_LOCK: - ec_lock_prepare_entry(fop, &fop->loc[0]); + ec_lock_prepare_entry(fop, &fop->loc[0], 1); ec_lock(fop); return EC_STATE_GET_SIZE_AND_VERSION; @@ -2101,7 +2101,7 @@ int32_t ec_manager_unlink(ec_fop_data_t * fop, int32_t state) case -EC_STATE_LOCK_REUSE: case EC_STATE_LOCK_REUSE: - ec_lock_reuse(fop, 0); + ec_lock_reuse(fop); return EC_STATE_UNLOCK; |