diff options
| author | Dan Lambright <dlambrig@redhat.com> | 2015-11-04 15:33:22 -0500 |
|---|---|---|
| committer | Dan Lambright <dlambrig@redhat.com> | 2015-11-23 04:05:55 -0800 |
| commit | 3b52c71b0ab57a9daaf31bf3dc8563da37927a66 (patch) | |
| tree | 2d47ba199f5cde08b2d9b639670d0ed5c75204b6 /tests/basic/tier/readdir-during-migration.t | |
| parent | f12efd1827077292eba08a109d212a25c62476fe (diff) | |
cluster/tier: readdirp to cold tier only
It is possible a file would get migrated in the middle
of a readdir operation. If there are four subvolumes A,B,C,D,
and if readdir reads them in order and reaches subvol B,
then, if a file is moved from D to A, it will not be included
in the readdir output.
This phenonema has pre-existed in DHT migration but is more
apparent in tiering.
When a file is moved off the hashed subvolume a T file is created.
For tiering, we will make the cold subvolume the hashed subvolume.
This will ensure the creation of a T file. Readdir will not skip T
files in the tier translator.
Making the cold subvolume the hashed subvolume ensures the T
files created on promotions or creates will be less likely to
fill the volume.
Creates still put the data on the hot subvolume.
Change-Id: Ifde557d3d0e94a4570ca9f115adee3db2ee75407
BUG: 1281598
Signed-off-by: Dan Lambright <dlambrig@redhat.com>
Reviewed-on: http://review.gluster.org/12530
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Tested-by: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: N Balachandran <nbalacha@redhat.com>
Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
Diffstat (limited to 'tests/basic/tier/readdir-during-migration.t')
| -rw-r--r-- | tests/basic/tier/readdir-during-migration.t | 64 |
1 files changed, 64 insertions, 0 deletions
diff --git a/tests/basic/tier/readdir-during-migration.t b/tests/basic/tier/readdir-during-migration.t new file mode 100644 index 00000000000..42199c57768 --- /dev/null +++ b/tests/basic/tier/readdir-during-migration.t @@ -0,0 +1,64 @@ +#!/bin/bash + +. $(dirname $0)/../../include.rc +. $(dirname $0)/../../volume.rc +. $(dirname $0)/../../tier.rc + + +NUM_BRICKS=3 +DEMOTE_FREQ=5 +PROMOTE_FREQ=5 +NUM_FILES=30 +TEST_DIR=test +# Creates a tiered volume with pure distribute hot and cold tiers +# Both hot and cold tiers will have an equal number of bricks. + +function create_dist_tier_vol () { + mkdir $B0/cold + mkdir $B0/hot + TEST $CLI volume create $V0 $H0:$B0/cold/${V0}{0..$1} + TEST $CLI volume set $V0 performance.quick-read off + TEST $CLI volume set $V0 performance.io-cache off + TEST $CLI volume start $V0 + TEST $CLI volume attach-tier $V0 $H0:$B0/hot/${V0}{0..$1} + TEST $CLI volume set $V0 cluster.tier-mode test + TEST $CLI volume set $V0 cluster.tier-demote-frequency $DEMOTE_FREQ + TEST $CLI volume set $V0 cluster.tier-promote-frequency $PROMOTE_FREQ + TEST $CLI volume set $V0 cluster.read-freq-threshold 0 + TEST $CLI volume set $V0 cluster.write-freq-threshold 0 +} + +function check_file_count() { + if [ $(ls -1 | wc -l) == $1 ]; then + echo "1" + else + echo "0" + fi +} + +cleanup; + + +TEST glusterd + +#Create and start a tiered volume +create_dist_tier_vol $NUM_BRICKS + +# Mount FUSE +TEST glusterfs -s $H0 --volfile-id $V0 $M0 + +# Create a number of "legacy" files before attaching tier +mkdir $M0/${TEST_DIR} +cd $M0/${TEST_DIR} +TEST create_many_files tfile $NUM_FILES + +EXPECT "1" check_file_count $NUM_FILES + +sleep $DEMOTE_FREQ + +EXPECT "1" check_file_count $NUM_FILES + +cd / + +cleanup; + |
