You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
82 lines
3.6 KiB
82 lines
3.6 KiB
From d45c64e17e1eb8003ac1086cbd3abea32414c7f9 Mon Sep 17 00:00:00 2001 |
|
From: Raghavendra Gowdappa <rgowdapp@redhat.com> |
|
Date: Tue, 12 Feb 2019 18:33:44 +0530 |
|
Subject: [PATCH 345/346] read-ahead/io-cache: turn off by default |
|
|
|
We've found perf xlators io-cache and read-ahead not adding any |
|
performance improvement. At best read-ahead is redundant due to kernel |
|
read-ahead and at worst io-cache is degrading the performance for |
|
workloads that doesn't involve re-read. Given that VFS already have |
|
both these functionalities, this patch makes these two |
|
translators turned off by default for native fuse mounts. |
|
|
|
For non-native fuse mounts like gfapi (NFS-ganesha/samba) we can have |
|
these xlators on by having custom profiles. |
|
|
|
>Change-Id: Ie7535788909d4c741844473696f001274dc0bb60 |
|
>Signed-off-by: Raghavendra Gowdappa <rgowdapp@redhat.com> |
|
>fixes: bz#1676479 |
|
Upstream fix link: https://review.gluster.org/#/c/glusterfs/+/22203/ |
|
|
|
BUG: 1788656 |
|
Change-Id: Ie7535788909d4c741844473696f001274dc0bb60 |
|
Signed-off-by: Sunil Kumar Acharya <sheggodu@redhat.com> |
|
Reviewed-on: https://code.engineering.redhat.com/gerrit/188967 |
|
Tested-by: RHGS Build Bot <nigelb@redhat.com> |
|
--- |
|
tests/basic/ec/self-heal.t | 2 ++ |
|
tests/basic/glusterd/volfile_server_switch.t | 2 +- |
|
xlators/mgmt/glusterd/src/glusterd-volume-set.c | 4 ++-- |
|
3 files changed, 5 insertions(+), 3 deletions(-) |
|
|
|
diff --git a/tests/basic/ec/self-heal.t b/tests/basic/ec/self-heal.t |
|
index d217559..6329bb6 100644 |
|
--- a/tests/basic/ec/self-heal.t |
|
+++ b/tests/basic/ec/self-heal.t |
|
@@ -131,6 +131,8 @@ TEST $CLI volume create $V0 redundancy 2 $H0:$B0/${V0}{0..5} |
|
TEST $CLI volume set $V0 client-log-level DEBUG |
|
#Write-behind has a bug where lookup can race over write which leads to size mismatch on the mount after a 'cp' |
|
TEST $CLI volume set $V0 performance.write-behind off |
|
+#md-cache can return stale stat due to default timeout being 1 sec |
|
+TEST $CLI volume set $V0 performance.stat-prefetch off |
|
EXPECT "Created" volinfo_field $V0 'Status' |
|
TEST $CLI volume start $V0 |
|
EXPECT_WITHIN $PROCESS_UP_TIMEOUT "Started" volinfo_field $V0 'Status' |
|
diff --git a/tests/basic/glusterd/volfile_server_switch.t b/tests/basic/glusterd/volfile_server_switch.t |
|
index 3090609..e11cfed 100644 |
|
--- a/tests/basic/glusterd/volfile_server_switch.t |
|
+++ b/tests/basic/glusterd/volfile_server_switch.t |
|
@@ -34,7 +34,7 @@ TEST glusterfs --volfile-id=/$V0 --volfile-server=$H1 --volfile-server=$H2 --vol |
|
|
|
TEST kill_glusterd 1 |
|
|
|
-TEST $CLI_2 volume set $V0 performance.io-cache off |
|
+TEST $CLI_2 volume set $V0 performance.write-behind off |
|
|
|
# make sure by this time directory will be created |
|
# TODO: suggest ideal time to wait |
|
diff --git a/xlators/mgmt/glusterd/src/glusterd-volume-set.c b/xlators/mgmt/glusterd/src/glusterd-volume-set.c |
|
index 16601a2..9001b88 100644 |
|
--- a/xlators/mgmt/glusterd/src/glusterd-volume-set.c |
|
+++ b/xlators/mgmt/glusterd/src/glusterd-volume-set.c |
|
@@ -2235,7 +2235,7 @@ struct volopt_map_entry glusterd_volopt_map[] = { |
|
{.key = "performance.read-ahead", |
|
.voltype = "performance/read-ahead", |
|
.option = "!perf", |
|
- .value = "on", |
|
+ .value = "off", |
|
.op_version = 1, |
|
.description = "enable/disable read-ahead translator in the volume.", |
|
.flags = VOLOPT_FLAG_CLIENT_OPT | VOLOPT_FLAG_XLATOR_OPT}, |
|
@@ -2249,7 +2249,7 @@ struct volopt_map_entry glusterd_volopt_map[] = { |
|
{.key = "performance.io-cache", |
|
.voltype = "performance/io-cache", |
|
.option = "!perf", |
|
- .value = "on", |
|
+ .value = "off", |
|
.op_version = 1, |
|
.description = "enable/disable io-cache translator in the volume.", |
|
.flags = VOLOPT_FLAG_CLIENT_OPT}, |
|
-- |
|
1.8.3.1 |
|
|
|
|