summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorAshish Pandey <aspandey@redhat.com>2018-03-13 14:03:20 +0530
committerAshish Pandey <aspandey@redhat.com>2018-03-19 14:22:04 +0530
commita644d52a4fe000827020b28736062a54c9a91b44 (patch)
treeb7a8aef058ff42efae61c118143a43e996b49c3f
parent0ec2e903a4e2b9bd9e544bd807ed45af5898e2ab (diff)
cluster/ec: Change default read policy to gfid-hash
Problem: Whenever we read data from file over NFS, NFS reads more data then requested and caches it. Based on the stat information it makes sure that the cached/pre-read data is valid or not. Consider 4 + 2 EC volume and all the bricks are on differnt nodes. In EC, with round-robin read policy, reads are sent on different set of data bricks. This way, it balances the read fops to go on all the bricks and avoid heating UP (overloading) same set of bricks. Due to small difference in clock speed, it is possible that we get minor difference for atime, mtime or ctime for different bricks. That might cause a different stat returned to NFS based on which NFS will discard cached/pre-read data which is actually not changed and could be used. Solution: Change read policy for EC as gfid-hash. That will force all the read to go to same set of bricks. >Change-Id: I825441cc519e94bf3dc3aa0bd4cb7c6ae6392c84 >BUG: 1554743 >Signed-off-by: Ashish Pandey <aspandey@redhat.com> Change-Id: I825441cc519e94bf3dc3aa0bd4cb7c6ae6392c84 BUG: 1557906 Signed-off-by: Ashish Pandey <aspandey@redhat.com>
-rw-r--r--tests/basic/ec/ec-read-policy.t7
-rw-r--r--xlators/cluster/ec/src/ec.c2
2 files changed, 4 insertions, 5 deletions
diff --git a/tests/basic/ec/ec-read-policy.t b/tests/basic/ec/ec-read-policy.t
index e4390aa..fe6fe65 100644
--- a/tests/basic/ec/ec-read-policy.t
+++ b/tests/basic/ec/ec-read-policy.t
@@ -20,10 +20,9 @@ TEST $CLI volume start $V0
TEST glusterfs --direct-io-mode=yes --entry-timeout=0 --attribute-timeout=0 -s $H0 --volfile-id $V0 $M0
EXPECT_WITHIN $CHILD_UP_TIMEOUT "6" ec_child_up_count $V0 0
#TEST volume operations work fine
-EXPECT "round-robin" mount_get_option_value $M0 $V0-disperse-0 read-policy
-TEST $CLI volume set $V0 disperse.read-policy gfid-hash
-EXPECT_WITHIN $CONFIG_UPDATE_TIMEOUT "gfid-hash" mount_get_option_value $M0 $V0-disperse-0 read-policy
-TEST $CLI volume reset $V0 disperse.read-policy
+
+EXPECT "gfid-hash" mount_get_option_value $M0 $V0-disperse-0 read-policy
+TEST $CLI volume set $V0 disperse.read-policy round-robin
EXPECT_WITHIN $CONFIG_UPDATE_TIMEOUT "round-robin" mount_get_option_value $M0 $V0-disperse-0 read-policy
#TEST if the option gives the intended behavior. The way we perform this test
diff --git a/xlators/cluster/ec/src/ec.c b/xlators/cluster/ec/src/ec.c
index f257600..429cac9 100644
--- a/xlators/cluster/ec/src/ec.c
+++ b/xlators/cluster/ec/src/ec.c
@@ -1580,7 +1580,7 @@ struct volume_options options[] =
{ .key = {"read-policy" },
.type = GF_OPTION_TYPE_STR,
.value = {"round-robin", "gfid-hash"},
- .default_value = "round-robin",
+ .default_value = "gfid-hash",
.op_version = {GD_OP_VERSION_3_7_6},
.flags = OPT_FLAG_SETTABLE | OPT_FLAG_CLIENT_OPT | OPT_FLAG_DOC,
.tags = {"disperse"},