Replace namespace/io-stats/io-threads with 3.6-fb versions

This rolls up multiple patches related to namespace identificaton and
throttling/QoS.  This primarily includes the following, all by Michael
Goulet <mgoulet@fb.com>.

  io-threads: Add weighted round robin queueing by namespace
  https://phabricator.facebook.com/D5615269

  io-threads: Add per-namespaces queue sizes to IO_THREADS_QUEUE_SIZE_KEY
  https://phabricator.facebook.com/D5683162

  io-threads: Implement better slot allocation algorithm
  https://phabricator.facebook.com/D5683186

  io-threads: Only enable weighted queueing on bricks
  https://phabricator.facebook.com/D5700062

  io-threads: Update queue sizes on drain
  https://phabricator.facebook.com/D5704832

  Fix parsing (-1) as default NS weight
  https://phabricator.facebook.com/D5723383

Parts of the following patches have also been applied to satisfy
dependencies.

  io-throttling: Calculate moving averages and throttle offending hosts
  https://phabricator.fb.com/D2516161
  Shreyas Siravara <sshreyas@fb.com>

  Hook up ODS logging for FUSE clients.
  https://phabricator.facebook.com/D3963376
  Kevin Vigor <kvigor@fb.com>

  Add the flag --skip-nfsd-start to skip the NFS daemon stating, even if
  it is enabled
  https://phabricator.facebook.com/D4575368
  Alex Lorca <alexlorca@fb.com>

There are also some "standard" changes: dealing with code that moved,
reindenting to comply with Gluster coding standards, gf_uuid_xxx, etc.

This patch *does* revert some changes which have occurred upstream since
3.6; these will be re-applied as apppropriate on top of this new base.

Change-Id: I69024115da7a60811e5b86beae781d602bdb558d
Signed-off-by: Jeff Darcy <jdarcy@fb.com>

3 files changed, 16 insertions, 4 deletions

diff --git a/tests/basic/afr/durability-off.t b/tests/basic/afr/durability-off.t
index 155ffa09ef0..0c4f470079a 100644
--- a/tests/basic/afr/durability-off.t
+++ b/tests/basic/afr/durability-off.t
@@ -5,6 +5,14 @@
 . $(dirname $0)/../../include.rc
 . $(dirname $0)/../../volume.rc
 
+did_fsync () {
+        local count=$($CLI volume profile $V0 info | grep -w FSYNC | wc -l)
+        if [ "$count" != "0" ]; then
+                echo "Y"
+        else
+                echo "N"
+        fi
+}
 cleanup;
 
 TEST glusterd
@@ -24,7 +32,7 @@ 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 $CLI volume heal $V0
 EXPECT_WITHIN $HEAL_TIMEOUT "0" get_pending_heal_count $V0
-EXPECT "^0$" echo $($CLI volume profile $V0 info | grep -w FSYNC | wc -l)
+EXPECT "N" did_fsync
 
 #Test that fsyncs happen when durability is on
 TEST $CLI volume set $V0 cluster.ensure-durability on
@@ -39,6 +47,6 @@ 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 $CLI volume heal $V0
 EXPECT_WITHIN $HEAL_TIMEOUT "0" get_pending_heal_count $V0
-EXPECT "^2$" echo $($CLI volume profile $V0 info | grep -w FSYNC | wc -l)
+EXPECT "Y" did_fsync
 
 cleanup;
diff --git a/tests/basic/fop-sampling.t b/tests/basic/fop-sampling.t
index e429fd8cb07..a1b3edc3d5d 100644
--- a/tests/basic/fop-sampling.t
+++ b/tests/basic/fop-sampling.t
@@ -13,7 +13,7 @@ function check_path {
         op=$1
         path=$2
         file=$3
-        grep $op $file | awk -F, '{print $11}' | grep $path 2>&1 > /dev/null
+        grep $op $file | awk -F, '{print $12}' | grep $path 2>&1 > /dev/null
         if [ $? -eq 0 ]; then
           echo "Y"
         else
@@ -106,6 +106,8 @@ for dir in "$N0" "$M0"; do
   rm $dir/file2
 done;
 
+read -p "Continue? " nothing
+
 EXPECT_WITHIN 10 "Y" check_path CREATE /file1 $BRICK_SAMPLES
 EXPECT_WITHIN 10 "Y" check_path LOOKUP /file1 $BRICK_SAMPLES
 EXPECT_WITHIN 10 "Y" check_path SETATTR /file1 $BRICK_SAMPLES
diff --git a/tests/basic/stats-dump.t b/tests/basic/stats-dump.t
index af1ad34702b..a188a45eea1 100644
--- a/tests/basic/stats-dump.t
+++ b/tests/basic/stats-dump.t
@@ -39,7 +39,9 @@ FUSE_RET="$?"
 
 # Test that io-stats is getting queue sizes from io-threads
 TEST grep 'queue_size' ${GLUSTERD_WORKDIR}/stats/glusterfs_nfsd_$V0.dump
-TEST ! grep 'queue_size' ${GLUSTERD_WORKDIR}/stats/glusterfsd__d_backends_patchy?.dump
+
+# We should be getting queue sizes on bricks now too.
+TEST grep 'queue_size' ${GLUSTERD_WORKDIR}/stats/glusterfsd__d_backends_patchy?.dump
 
 TEST [ 0 -ne "$BRICK_RET" ]
 TEST [ 0 -ne "$NFSD_RET" ]