core: run many bricks within one glusterfsd process

This patch adds support for multiple brick translator stacks running in
a single brick server process.  This reduces our per-brick memory usage
by approximately 3x, and our appetite for TCP ports even more.  It also
creates potential to avoid process/thread thrashing, and to improve QoS
by scheduling more carefully across the bricks, but realizing that
potential will require further work.

Multiplexing is controlled by the "cluster.brick-multiplex" global
option.  By default it's off, and bricks are started in separate
processes as before.  If multiplexing is enabled, then *compatible*
bricks (mostly those with the same transport options) will be started in
the same process.

Backport of:
> Change-Id: I45059454e51d6f4cbb29a4953359c09a408695cb
> BUG: 1385758
> Reviewed-on: https://review.gluster.org/14763

Change-Id: I4bce9080f6c93d50171823298fdf920258317ee8
BUG: 1418091
Signed-off-by: Jeff Darcy <jdarcy@redhat.com>
Reviewed-on: https://review.gluster.org/16496
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com>

21 files changed, 31 insertions, 27 deletions

diff --git a/tests/basic/afr/add-brick-self-heal.t b/tests/basic/afr/add-brick-self-heal.t
index 748d36758e7..a904e22e2a5 100644
--- a/tests/basic/afr/add-brick-self-heal.t
+++ b/tests/basic/afr/add-brick-self-heal.t
@@ -12,7 +12,7 @@ TEST $CLI volume set $V0 cluster.metadata-self-heal off
 TEST $CLI volume set $V0 cluster.entry-self-heal off
 
 TEST $CLI volume set $V0 self-heal-daemon off
-TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 $M0;
+TEST $GFS --volfile-id=$V0 --volfile-server=$H0 $M0;
 
 # Create files
 for i in {1..5}
diff --git a/tests/basic/afr/arbiter-add-brick.t b/tests/basic/afr/arbiter-add-brick.t
index 69e13267ccd..c6fe18cec16 100644
--- a/tests/basic/afr/arbiter-add-brick.t
+++ b/tests/basic/afr/arbiter-add-brick.t
@@ -11,7 +11,7 @@ TEST $CLI volume create $V0 replica 2 $H0:$B0/${V0}{0,1}
 TEST $CLI volume set $V0 performance.stat-prefetch off
 TEST $CLI volume start $V0
 TEST $CLI volume set $V0 self-heal-daemon off
-TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 $M0;
+TEST $GFS --volfile-id=$V0 --volfile-server=$H0 $M0;
 TEST mkdir  $M0/dir1
 TEST dd if=/dev/urandom of=$M0/file1 bs=1024 count=1
 
diff --git a/tests/basic/afr/arbiter-mount.t b/tests/basic/afr/arbiter-mount.t
index 587e808863f..da99096f81f 100644
--- a/tests/basic/afr/arbiter-mount.t
+++ b/tests/basic/afr/arbiter-mount.t
@@ -22,7 +22,7 @@ TEST kill_brick $V0 $H0 $B0/${V0}1
 
 # Doing `mount -t glusterfs $H0:$V0 $M0` fails right away but doesn't work on NetBSD
 # So check that stat <mount> fails instead.
-TEST glusterfs --volfile-id=/$V0 --volfile-server=$H0 $M0
+TEST $GFS --volfile-id=/$V0 --volfile-server=$H0 $M0
 TEST ! stat $M0
 EXPECT_WITHIN $UMOUNT_TIMEOUT "Y" force_umount $M0
 
@@ -34,7 +34,7 @@ EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" brick_up_status $V0 $H0 $B0/${V0}0
 EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" brick_up_status $V0 $H0 $B0/${V0}1
 EXPECT_WITHIN $NFS_EXPORT_TIMEOUT "1" is_nfs_export_available;
 
-TEST glusterfs --volfile-id=/$V0 --volfile-server=$H0 $M0
+TEST $GFS --volfile-id=/$V0 --volfile-server=$H0 $M0
 TEST  stat $M0
 EXPECT_WITHIN $UMOUNT_TIMEOUT "Y" force_umount $M0
 
diff --git a/tests/basic/afr/arbiter-remove-brick.t b/tests/basic/afr/arbiter-remove-brick.t
index 5a6daa95cfd..ec93c8758e4 100644
--- a/tests/basic/afr/arbiter-remove-brick.t
+++ b/tests/basic/afr/arbiter-remove-brick.t
@@ -11,7 +11,7 @@ TEST $CLI volume create $V0 replica 3 arbiter 1  $H0:$B0/${V0}{0,1,2}
 EXPECT "1 x \(2 \+ 1\) = 3" volinfo_field $V0 "Number of Bricks"
 TEST $CLI volume set $V0 performance.stat-prefetch off
 TEST $CLI volume start $V0
-TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 $M0;
+TEST $GFS --volfile-id=$V0 --volfile-server=$H0 $M0;
 
 #syntax check for remove-brick.
 TEST ! $CLI volume remove-brick $V0 replica 2  $H0:$B0/${V0}0 force
diff --git a/tests/basic/afr/arbiter-statfs.t b/tests/basic/afr/arbiter-statfs.t
index 7d136378f11..61cb9e1d04f 100644
--- a/tests/basic/afr/arbiter-statfs.t
+++ b/tests/basic/afr/arbiter-statfs.t
@@ -29,7 +29,7 @@ TEST MOUNT_LOOP $LO3 $B0/${V0}3
 
 TEST $CLI volume create $V0 replica 3 arbiter 1 $H0:$B0/${V0}{1,2,3};
 TEST $CLI volume start $V0
-TEST glusterfs --volfile-server=$H0 --volfile-id=$V0 $M0
+TEST $GFS --volfile-server=$H0 --volfile-id=$V0 $M0
 free_space=$(df -P $M0 | tail -1 | awk '{ print $4}')
 TEST [ $free_space -gt 100000 ]
 TEST force_umount $M0
diff --git a/tests/basic/afr/arbiter.t b/tests/basic/afr/arbiter.t
index 1abc940b095..7c92a9fe6c9 100644
--- a/tests/basic/afr/arbiter.t
+++ b/tests/basic/afr/arbiter.t
@@ -16,7 +16,7 @@ EXPECT 'Started' volinfo_field $V0 'Status'
 EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" brick_up_status $V0 $H0 $B0/${V0}0
 EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" brick_up_status $V0 $H0 $B0/${V0}1
 EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" brick_up_status $V0 $H0 $B0/${V0}2
-TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 --entry-timeout=0 $M0;
+TEST $GFS --volfile-id=$V0 --volfile-server=$H0 $M0;
 TEST ! stat $M0/.meta/graphs/active/$V0-replicate-0/options/arbiter-count
 EXPECT_WITHIN $UMOUNT_TIMEOUT "Y" force_umount $M0
 TEST $CLI volume stop $V0
@@ -42,7 +42,7 @@ EXPECT 'Started' volinfo_field $V0 'Status'
 EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" brick_up_status $V0 $H0 $B0/${V0}0
 EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" brick_up_status $V0 $H0 $B0/${V0}1
 EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" brick_up_status $V0 $H0 $B0/${V0}2
-TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 --attribute-timeout=0 --entry-timeout=0 $M0;
+TEST $GFS --volfile-id=$V0 --volfile-server=$H0 $M0;
 TEST stat $M0/.meta/graphs/active/$V0-replicate-0/options/arbiter-count
 EXPECT "1" cat $M0/.meta/graphs/active/$V0-replicate-0/options/arbiter-count
 
diff --git a/tests/basic/afr/client-side-heal.t b/tests/basic/afr/client-side-heal.t
index d87f4b14063..eba7dc2b3c4 100755
--- a/tests/basic/afr/client-side-heal.t
+++ b/tests/basic/afr/client-side-heal.t
@@ -13,7 +13,7 @@ TEST $CLI volume set $V0 cluster.data-self-heal off
 TEST $CLI volume set $V0 cluster.metadata-self-heal off
 
 TEST $CLI volume start $V0
-TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 --entry-timeout=0 $M0;
+TEST $GFS --volfile-id=$V0 --volfile-server=$H0 $M0;
 echo "some data" > $M0/datafile
 EXPECT 0 echo $?
 TEST touch $M0/mdatafile
@@ -46,11 +46,11 @@ TEST ls $M0/mdatafile
 #To trigger inode refresh for sure, the volume is unmounted and mounted each time.
 #Check that data heal does not happen.
 EXPECT_WITHIN $UMOUNT_TIMEOUT "Y" force_umount $M0
-TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 --entry-timeout=0 $M0;
+TEST $GFS --volfile-id=$V0 --volfile-server=$H0 $M0;
 TEST cat $M0/datafile
 #Check that entry heal does not happen.
 EXPECT_WITHIN $UMOUNT_TIMEOUT "Y" force_umount $M0
-TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 --entry-timeout=0 $M0;
+TEST $GFS --volfile-id=$V0 --volfile-server=$H0 $M0;
 TEST ls $M0/dir
 
 #No heal must have happened
@@ -68,12 +68,12 @@ EXPECT 7 get_pending_heal_count $V0
 #Inode refresh must trigger data and entry heals.
 #To trigger inode refresh for sure, the volume is unmounted and mounted each time.
 EXPECT_WITHIN $UMOUNT_TIMEOUT "Y" force_umount $M0
-TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 --entry-timeout=0 $M0;
+TEST $GFS --volfile-id=$V0 --volfile-server=$H0 $M0;
 TEST cat $M0/datafile
 EXPECT_WITHIN $HEAL_TIMEOUT 6 get_pending_heal_count $V0
 
 EXPECT_WITHIN $UMOUNT_TIMEOUT "Y" force_umount $M0
-TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 --entry-timeout=0 $M0;
+TEST $GFS --volfile-id=$V0 --volfile-server=$H0 $M0;
 TEST ls $M0/dir
 EXPECT 5 get_pending_heal_count $V0
 
diff --git a/tests/basic/afr/data-self-heal.t b/tests/basic/afr/data-self-heal.t
index 5db5d770b6f..0f417b4a0ba 100644
--- a/tests/basic/afr/data-self-heal.t
+++ b/tests/basic/afr/data-self-heal.t
@@ -77,7 +77,7 @@ TEST $CLI volume start $V0
 EXPECT_WITHIN $PROCESS_UP_TIMEOUT "Y" glustershd_up_status
 EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status_in_shd $V0 0
 EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status_in_shd $V0 1
-TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 $M0 --entry-timeout=0 --attribute-timeout=0;
+TEST $GFS --volfile-id=$V0 --volfile-server=$H0 $M0;
 cd $M0
 TEST touch pending-changelog biggest-file-source.txt biggest-file-more-prio-than-changelog.txt same-size-more-prio-to-changelog.txt size-and-witness-same.txt self-accusing-vs-source.txt self-accusing-both.txt self-accusing-vs-innocent.txt self-accusing-bigger-exists.txt size-more-prio-than-self-accused.txt v1-dirty.txt split-brain.txt split-brain-all-dirty.txt split-brain-with-dirty.txt
 
diff --git a/tests/basic/afr/entry-self-heal.t b/tests/basic/afr/entry-self-heal.t
index 337b9c59f84..3c900fdcf9a 100644
--- a/tests/basic/afr/entry-self-heal.t
+++ b/tests/basic/afr/entry-self-heal.t
@@ -81,7 +81,7 @@ TEST $CLI volume set $V0 performance.io-cache off
 TEST $CLI volume set $V0 performance.quick-read off
 TEST $CLI volume start $V0
 
-TEST glusterfs --volfile-id=/$V0 --volfile-server=$H0 $M0 --attribute-timeout=0 --entry-timeout=0 --use-readdirp=no
+TEST $GFS --volfile-id=/$V0 --volfile-server=$H0 --use-readdirp=no $M0
 cd $M0
 #_me_ is dir on which missing entry self-heal happens, _heal is where dir self-heal happens
 #spb is split-brain, fool is all fool
diff --git a/tests/basic/afr/gfid-mismatch.t b/tests/basic/afr/gfid-mismatch.t
index c3399215569..fc15793cf5a 100644
--- a/tests/basic/afr/gfid-mismatch.t
+++ b/tests/basic/afr/gfid-mismatch.t
@@ -13,6 +13,10 @@ TEST $CLI volume set $V0 self-heal-daemon off
 TEST $CLI volume set $V0 stat-prefetch off
 TEST $CLI volume start $V0
 TEST $CLI volume set $V0 cluster.background-self-heal-count 0
+# We can't count on brick0 getting a copy of the file immediately without this,
+# because (especially with multiplexing) it might not have *come up*
+# immediately.
+TEST $CLI volume set $V0 cluster.quorum-type auto
 TEST $GFS --volfile-id=$V0 -s $H0 $M0;
 
 #Test
diff --git a/tests/basic/afr/gfid-self-heal.t b/tests/basic/afr/gfid-self-heal.t
index 0bc53de8a6f..b54edbcae85 100644
--- a/tests/basic/afr/gfid-self-heal.t
+++ b/tests/basic/afr/gfid-self-heal.t
@@ -15,7 +15,7 @@ TEST $CLI volume set $V0 nfs.disable on
 TEST touch $B0/${V0}{0,1}/{1,2,3,4}
 TEST $CLI volume start $V0
 
-TEST glusterfs --volfile-id=/$V0 --volfile-server=$H0 $M0 --attribute-timeout=0 --entry-timeout=0
+TEST $GFS --volfile-id=/$V0 --volfile-server=$H0 $M0
 #Test that readdir returns entries even when no gfids are present
 EXPECT 4 echo $(ls $M0 | grep -v '^\.' | wc -l)
 sleep 2;
diff --git a/tests/basic/afr/heal-quota.t b/tests/basic/afr/heal-quota.t
index 2663906f9d5..96e23363da8 100644
--- a/tests/basic/afr/heal-quota.t
+++ b/tests/basic/afr/heal-quota.t
@@ -13,7 +13,7 @@ TEST $CLI volume create $V0 replica 2 $H0:$B0/${V0}{0,1}
 TEST $CLI volume set $V0 cluster.self-heal-daemon off
 TEST $CLI volume start $V0
 
-TEST glusterfs --attribute-timeout=0 --entry-timeout=0 --volfile-id=/$V0 --volfile-server=$H0 $M0;
+TEST $GFS --volfile-id=/$V0 --volfile-server=$H0 $M0;
 TEST $CLI volume quota $V0 enable
 TEST $CLI volume quota $V0 limit-usage / 10MB
 TEST $CLI volume quota $V0 soft-timeout 0
diff --git a/tests/basic/afr/metadata-self-heal.t b/tests/basic/afr/metadata-self-heal.t
index b88c16a93e1..275aecd2175 100644
--- a/tests/basic/afr/metadata-self-heal.t
+++ b/tests/basic/afr/metadata-self-heal.t
@@ -51,7 +51,7 @@ TEST glusterd
 TEST pidof glusterd
 TEST $CLI volume create $V0 replica 2 $H0:$B0/brick{0,1}
 TEST $CLI volume start $V0
-TEST glusterfs --volfile-id=/$V0 --volfile-server=$H0 $M0 --attribute-timeout=0 --entry-timeout=0
+TEST $GFS --volfile-id=/$V0 --volfile-server=$H0 $M0
 cd $M0
 
 TEST touch a
diff --git a/tests/basic/afr/quorum.t b/tests/basic/afr/quorum.t
index c105290445a..252e25468d7 100644
--- a/tests/basic/afr/quorum.t
+++ b/tests/basic/afr/quorum.t
@@ -19,7 +19,7 @@ TEST $CLI volume set $V0 performance.write-behind off
 TEST $CLI volume set $V0 performance.stat-prefetch off
 TEST $CLI volume set $V0 performance.read-ahead off
 TEST $CLI volume start $V0
-TEST $GFS -s $H0 --volfile-id=$V0 $M0 --direct-io-mode=enable;
+TEST $GFS -s $H0 --volfile-id=$V0 --direct-io-mode=enable $M0;
 
 touch $M0/a
 echo abc > $M0/b
@@ -75,7 +75,7 @@ TEST $CLI volume set $V0 performance.write-behind off
 TEST $CLI volume set $V0 performance.stat-prefetch off
 TEST $CLI volume set $V0 performance.read-ahead off
 TEST $CLI volume start $V0
-TEST $GFS -s $H0 --volfile-id=$V0 $M0 --direct-io-mode=enable;
+TEST $GFS -s $H0 --volfile-id=$V0 --direct-io-mode=enable $M0;
 
 touch $M0/a
 echo abc > $M0/b
diff --git a/tests/basic/afr/replace-brick-self-heal.t b/tests/basic/afr/replace-brick-self-heal.t
index fef671a3875..a8c01a0f377 100644
--- a/tests/basic/afr/replace-brick-self-heal.t
+++ b/tests/basic/afr/replace-brick-self-heal.t
@@ -12,7 +12,7 @@ TEST $CLI volume set $V0 cluster.metadata-self-heal off
 TEST $CLI volume set $V0 cluster.entry-self-heal off
 
 TEST $CLI volume set $V0 self-heal-daemon off
-TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 $M0;
+TEST $GFS --volfile-id=$V0 --volfile-server=$H0 $M0;
 
 # Create files
 for i in {1..5}
diff --git a/tests/basic/afr/root-squash-self-heal.t b/tests/basic/afr/root-squash-self-heal.t
index ff0aa5cecb7..c4fab0a35b2 100644
--- a/tests/basic/afr/root-squash-self-heal.t
+++ b/tests/basic/afr/root-squash-self-heal.t
@@ -12,7 +12,7 @@ TEST $CLI volume set $V0 performance.stat-prefetch off
 TEST $CLI volume set $V0 self-heal-daemon off
 TEST $CLI volume set $V0 server.root-squash on
 TEST $CLI volume start $V0
-TEST glusterfs --volfile-id=/$V0 --volfile-server=$H0 $M0 --attribute-timeout=0 --entry-timeout=0 --no-root-squash=yes --use-readdirp=no
+TEST $GFS --volfile-id=/$V0 --volfile-server=$H0 --no-root-squash=yes --use-readdirp=no $M0
 TEST kill_brick $V0 $H0 $B0/${V0}0
 echo abc > $M0/a
 
diff --git a/tests/basic/afr/self-heald.t b/tests/basic/afr/self-heald.t
index a0906f97cee..24c82777921 100644
--- a/tests/basic/afr/self-heald.t
+++ b/tests/basic/afr/self-heald.t
@@ -50,7 +50,7 @@ TEST $CLI volume set $V0 cluster.background-self-heal-count 0
 TEST $CLI volume set $V0 cluster.eager-lock off
 TEST $CLI volume set $V0 performance.flush-behind off
 TEST $CLI volume start $V0
-TEST glusterfs --volfile-id=/$V0 --volfile-server=$H0 $M0 --attribute-timeout=0 --entry-timeout=0
+TEST $GFS --volfile-id=/$V0 --volfile-server=$H0 $M0
 
 decide_kill=$((`date +"%j"|sed 's/^0*//'` % 2 ))
 
diff --git a/tests/basic/afr/split-brain-favorite-child-policy.t b/tests/basic/afr/split-brain-favorite-child-policy.t
index 3df8e718bf0..0e321c6f095 100644
--- a/tests/basic/afr/split-brain-favorite-child-policy.t
+++ b/tests/basic/afr/split-brain-favorite-child-policy.t
@@ -17,7 +17,7 @@ TEST $CLI volume set $V0 cluster.entry-self-heal off
 TEST $CLI volume set $V0 cluster.data-self-heal off
 TEST $CLI volume set $V0 cluster.metadata-self-heal off
 TEST $CLI volume start $V0
-TEST glusterfs --volfile-id=/$V0 --volfile-server=$H0 $M0 --attribute-timeout=0 --entry-timeout=0
+TEST $GFS --volfile-id=/$V0 --volfile-server=$H0 $M0
 TEST touch $M0/file
 
 ############ Healing using favorite-child-policy = ctime #################
diff --git a/tests/basic/afr/split-brain-heal-info.t b/tests/basic/afr/split-brain-heal-info.t
index eabfbd0880a..66275c57207 100644
--- a/tests/basic/afr/split-brain-heal-info.t
+++ b/tests/basic/afr/split-brain-heal-info.t
@@ -20,7 +20,7 @@ TEST pidof glusterd
 TEST $CLI volume create $V0 replica 2 $H0:$B0/${V0}{0,1}
 TEST $CLI volume start $V0
 TEST $CLI volume set $V0 cluster.self-heal-daemon off
-TEST glusterfs --volfile-id=/$V0 --volfile-server=$H0 $M0 --attribute-timeout=0 --entry-timeout=0
+TEST $GFS --volfile-id=/$V0 --volfile-server=$H0 $M0
 
 TEST mkdir $M0/dspb
 TEST mkdir $M0/mspb
diff --git a/tests/basic/afr/split-brain-healing.t b/tests/basic/afr/split-brain-healing.t
index c66bb5d44df..403d08faab3 100644
--- a/tests/basic/afr/split-brain-healing.t
+++ b/tests/basic/afr/split-brain-healing.t
@@ -35,7 +35,7 @@ TEST $CLI volume set $V0 cluster.data-self-heal off
 TEST $CLI volume set $V0 cluster.metadata-self-heal off
 TEST $CLI volume set $V0 cluster.entry-self-heal off
 TEST $CLI volume start $V0
-TEST glusterfs --volfile-id=/$V0 --volfile-server=$H0 $M0 --attribute-timeout=0 --entry-timeout=0
+TEST $GFS --volfile-id=/$V0 --volfile-server=$H0 $M0
 
 cd $M0
 for i in {1..10}
diff --git a/tests/basic/afr/split-brain-resolution.t b/tests/basic/afr/split-brain-resolution.t
index 84b2cc8db51..e75e15aaa97 100644
--- a/tests/basic/afr/split-brain-resolution.t
+++ b/tests/basic/afr/split-brain-resolution.t
@@ -16,7 +16,7 @@ TEST $CLI volume start $V0
 #Disable self-heal-daemon
 TEST $CLI volume set $V0 cluster.self-heal-daemon off
 
-TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 --entry-timeout=0 $M0;
+TEST $GFS --volfile-id=$V0 --volfile-server=$H0 $M0;
 
 TEST `echo "some-data" > $M0/data-split-brain.txt`
 TEST `echo "some-data" > $M0/metadata-split-brain.txt`