summaryrefslogtreecommitdiffstats
path: root/tests/basic
diff options
context:
space:
mode:
authorShreyas Siravara <sshreyas@fb.com>2016-07-16 08:22:43 -0700
committerShreyas Siravara <sshreyas@fb.com>2016-12-14 18:10:30 -0800
commit1ad70fb05cadf14aa5900ba642119e73a3f25c73 (patch)
treef482ac9d24a3f5209a3131ce8c82b0d3622a2f9b /tests/basic
parentb43e301657e81d541174bade1f496caffc1c4aa5 (diff)
write-behind: Allow trickling-writes to be configurable, fix usage of page_size and window_size
Summary: - It adds a configurable option for "trickling-writes". - Makes `__wb_preprocess_winds()` use `wb_inode->window_conf` rather than `page_size`, so that the window-size option is actually respected. - This is a port of D3576122 & D3738605 to 3.8. Test Plan: - Prove test which looks @ brick-level FOPs and ensures that they fall in the right write-size bucket. Reviewed By: rwareing Signature: t1:3576122:1468892648:6923a6a19b18888577ce5173b5c9cb9531f941e7 Change-Id: I379a9f2f0c4768c9052b7e9dd71c5f0469cb2d68 Signed-off-by: Shreyas Siravara <sshreyas@fb.com> Reviewed-on: http://review.gluster.org/16079 NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Kevin Vigor <kvigor@fb.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Diffstat (limited to 'tests/basic')
-rw-r--r--tests/basic/write-behind.t53
1 files changed, 53 insertions, 0 deletions
diff --git a/tests/basic/write-behind.t b/tests/basic/write-behind.t
new file mode 100644
index 00000000000..edad59786af
--- /dev/null
+++ b/tests/basic/write-behind.t
@@ -0,0 +1,53 @@
+#!/bin/bash
+#
+
+. $(dirname $0)/../include.rc
+. $(dirname $0)/../volume.rc
+
+function clear_stats {
+ > /var/lib/glusterfs/stats/glusterfs_d_backends_${V0}0.dump
+}
+
+function got_expected_write_count {
+ expected_size=$1
+ expected_value=$2
+ grep aggr.write_${expected_size} "/var/lib/glusterd/stats/glusterfsd__d_backends_${V0}0.dump" | grep $expected_value
+ if [ $? == 0 ]; then
+ echo "Y";
+ else
+ echo "N";
+ fi
+}
+
+cleanup;
+
+TEST glusterd
+TEST pidof glusterd
+
+TEST $CLI volume create $V0 replica 3 $H0:$B0/${V0}{0,1,2}
+
+# These are needed for our tracking of write sizes
+TEST $CLI volume set $V0 diagnostics.latency-measurement on
+TEST $CLI volume set $V0 diagnostics.count-fop-hits on
+TEST $CLI volume set $V0 diagnostics.stats-dump-interval 2
+
+# Disable this in testing to get deterministic results
+TEST $CLI volume set $V0 performance.write-behind-trickling-writes off
+
+TEST $CLI volume start $V0
+
+sleep 2;
+
+TEST glusterfs -s $H0 --volfile-id $V0 $M0
+
+# Write a 100MB file with a window-size 1MB, we should get 100 writes of 1MB each
+TEST dd if=/dev/zero of=$M0/100mb_file bs=1M count=100
+EXPECT_WITHIN 5 "Y" got_expected_write_count "1mb" 100
+
+TEST $CLI volume set $V0 performance.write-behind-window-size 512KB
+
+# Write a 100MB file with a window-size 512KB, we should get 200 writes of 512KB each
+TEST dd if=/dev/zero of=$M0/100mb_file_2 bs=1M count=100
+EXPECT_WITHIN 5 "Y" got_expected_write_count "512kb" 200
+
+cleanup;