summaryrefslogtreecommitdiffstats
path: root/xlators/storage/posix/src/posix.c
diff options
context:
space:
mode:
authorRavishankar N <ravishankar@redhat.com>2016-04-05 15:16:52 +0530
committerPranith Kumar Karampuri <pkarampu@redhat.com>2016-04-11 04:32:37 -0700
commite1004679563ef17c460f83098983baf105655712 (patch)
tree86b6d63da1aee0131fcc285203f90213f06328d8 /xlators/storage/posix/src/posix.c
parent09c9da3b12da73dd718522fdf7587d3be5a14137 (diff)
arbiter: write performance improvement
Problem: The throughput for a 'dd' workload was much less for arbiter configuration when compared to normal replica-3 volume. There were 2 issues: i)arbiter_writev was using the request dict as response dict while unwinding, leading to incorect GLUSTERFS_WRITE_IS_APPEND and GLUSTERFS_OPEN_FD_COUNT values (=4), leading to immediate post-ops because is_afr_delayed_changelog_post_op_needed() failed due to afr_are_multiple_fds_opened() check. ii) The arbiter code in afr was setting local->transaction.{start and len} =0 to take full file locks. What this meant was even for simultaenous but non-overlapping writevs, afr_transaction_eager_lock_init() was not happening because afr_locals_overlap() always stays true. Consequently is_afr_delayed_changelog_post_op_needed() failed due to local->delayed_post_op not being set. Fix: i) Send appropriate response dict values in arbiter_writev. ii) Modify flock params instead of local->transaction.{start and len} to take full file locks in the transaction. Also changed _fill_writev_xdata() in posix to fill rsp_xdata for whatever key is requested for. Change-Id: I1c5fc5e98aba49ade540bb441a022e65b753432a BUG: 1324004 Signed-off-by: Ravishankar N <ravishankar@redhat.com> Reported-by: Robert Rauch <robert.rauch@gns-systems.de> Reported-by: Russel Purinton <russell.purinton@gmail.com> Reviewed-on: http://review.gluster.org/13906 Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Smoke: Gluster Build System <jenkins@build.gluster.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Diffstat (limited to 'xlators/storage/posix/src/posix.c')
-rw-r--r--xlators/storage/posix/src/posix.c35
1 files changed, 21 insertions, 14 deletions
diff --git a/xlators/storage/posix/src/posix.c b/xlators/storage/posix/src/posix.c
index f72c13d9066..8cce4be1c34 100644
--- a/xlators/storage/posix/src/posix.c
+++ b/xlators/storage/posix/src/posix.c
@@ -3055,28 +3055,35 @@ _fill_writev_xdata (fd_t *fd, dict_t *xdata, xlator_t *this, int is_append)
goto out;
}
- if (!xdata || !dict_get (xdata, GLUSTERFS_OPEN_FD_COUNT))
+ if (!xdata)
goto out;
rsp_xdata = dict_new();
if (!rsp_xdata)
goto out;
- ret = dict_set_uint32 (rsp_xdata, GLUSTERFS_OPEN_FD_COUNT,
- fd->inode->fd_count);
- if (ret < 0) {
- gf_msg (this->name, GF_LOG_WARNING, 0, P_MSG_DICT_SET_FAILED,
- "%s: Failed to set dictionary value for %s",
- uuid_utoa (fd->inode->gfid), GLUSTERFS_OPEN_FD_COUNT);
+ if (dict_get (xdata, GLUSTERFS_OPEN_FD_COUNT)) {
+ ret = dict_set_uint32 (rsp_xdata, GLUSTERFS_OPEN_FD_COUNT,
+ fd->inode->fd_count);
+ if (ret < 0) {
+ gf_msg (this->name, GF_LOG_WARNING, 0,
+ P_MSG_DICT_SET_FAILED, "%s: Failed to set "
+ "dictionary value for %s",
+ uuid_utoa (fd->inode->gfid),
+ GLUSTERFS_OPEN_FD_COUNT);
+ }
}
- ret = dict_set_uint32 (rsp_xdata, GLUSTERFS_WRITE_IS_APPEND,
- is_append);
- if (ret < 0) {
- gf_msg (this->name, GF_LOG_WARNING, 0, P_MSG_DICT_SET_FAILED,
- "%s: Failed to set dictionary value for %s",
- uuid_utoa (fd->inode->gfid),
- GLUSTERFS_WRITE_IS_APPEND);
+ if (dict_get (xdata, GLUSTERFS_WRITE_IS_APPEND)) {
+ ret = dict_set_uint32 (rsp_xdata, GLUSTERFS_WRITE_IS_APPEND,
+ is_append);
+ if (ret < 0) {
+ gf_msg (this->name, GF_LOG_WARNING, 0,
+ P_MSG_DICT_SET_FAILED, "%s: Failed to set "
+ "dictionary value for %s",
+ uuid_utoa (fd->inode->gfid),
+ GLUSTERFS_WRITE_IS_APPEND);
+ }
}
out:
return rsp_xdata;