diff options
| author | Ravishankar N <ravishankar@redhat.com> | 2016-02-05 15:10:06 +0530 |
|---|---|---|
| committer | Pranith Kumar Karampuri <pkarampu@redhat.com> | 2016-06-24 04:26:01 -0700 |
| commit | 7a1c1e2904701496968ed14b6d7479fb706c3188 (patch) | |
| tree | 8322bbab49c56e19246ee1bde1e8a67edb058065 /xlators | |
| parent | beaa5d104a3eec1b0b87fac648b028f77fc2e80b (diff) | |
afr:Don't wind reads for files in metadata split-brain
Problem: For a read on a file in metadata split-brain:
1.lookup_done resets event_generation to zero.
2. readv is issued, goes to inode refresh due to mismatching event_gen.
3. After refresh is successful, we update event_generation, data and
metdata readable.
3. We then call afr_read_txn_refresh_done() which in turn calls
afr_inode_get_readable() but doesn't check for EIO. So afr_readv_wind
is called with local->readable (which is populated with data_readable),
thus winding the read to a brick.
4. Also, further parallel reads that come directly go to the wind path
because there is no inode_refresh needed.
Fix:
1.For any afr_read_txn(), readable must be an intersection of data and metadata
readable.
2.Check for EIO in afr_read_txn_refresh_done().
Change-Id: I22dd221fdfaf96d7aced2f474e28ed1337d69f0e
BUG: 1305031
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-on: http://review.gluster.org/13389
Reviewed-by: Ashish Pandey <aspandey@redhat.com>
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Diffstat (limited to 'xlators')
| -rw-r--r-- | xlators/cluster/afr/src/afr-read-txn.c | 14 |
1 files changed, 4 insertions, 10 deletions
diff --git a/xlators/cluster/afr/src/afr-read-txn.c b/xlators/cluster/afr/src/afr-read-txn.c index 32ad6a46d17..74749f029c8 100644 --- a/xlators/cluster/afr/src/afr-read-txn.c +++ b/xlators/cluster/afr/src/afr-read-txn.c @@ -83,7 +83,7 @@ afr_read_txn_refresh_done (call_frame_t *frame, xlator_t *this, int err) &event_generation, local->transaction.type); - if (ret == -1 || !event_generation) + if (ret == -EIO || !event_generation) /* Even after refresh, we don't have a good read subvolume. Time to bail */ AFR_READ_TXN_SET_ERROR_AND_GOTO (-1, EIO, -1, readfn); @@ -218,18 +218,12 @@ afr_read_txn (call_frame_t *frame, xlator_t *this, inode_t *inode, } local->transaction.type = type; - if (local->op == GF_FOP_FSTAT || local->op == GF_FOP_STAT) { - ret = afr_inode_read_subvol_get (inode, this, data, metadata, - &event_generation); - AFR_INTERSECT (local->readable, data, metadata, - priv->child_count); - } else { - ret = afr_inode_read_subvol_type_get (inode, this, local->readable, - &event_generation, type); - } + ret = afr_inode_read_subvol_get (inode, this, data, metadata, + &event_generation); if (ret == -1) /* very first transaction on this inode */ goto refresh; + AFR_INTERSECT (local->readable, data, metadata, priv->child_count); gf_msg_debug (this->name, 0, "%s: generation now vs cached: %d, " "%d", uuid_utoa (inode->gfid), local->event_generation, |