glusterfs/tests/cluster.rc
Mohit Agrawal 21c7f7bacc glusterd: socketfile & pidfile related fixes for brick multiplexing feature
Problem: While brick-muliplexing is on after restarting glusterd, CLI is
         not showing pid of all brick processes in all volumes.

Solution: While brick-mux is on all local brick process communicated through one
          UNIX socket but as per current code (glusterd_brick_start) it is trying
          to communicate with separate UNIX socket for each volume which is populated
          based on brick-name and vol-name.Because of multiplexing design only one
          UNIX socket is opened so it is throwing poller error and not able to
          fetch correct status of brick process through cli process.
          To resolve the problem write a new function glusterd_set_socket_filepath_for_mux
          that will call by glusterd_brick_start to validate about the existence of socketpath.
          To avoid the continuous EPOLLERR erros in  logs update socket_connect code.

Test:     To reproduce the issue followed below steps
          1) Create two distributed volumes(dist1 and dist2)
          2) Set cluster.brick-multiplex is on
          3) kill glusterd
          4) run command gluster v status
          After apply the patch it shows correct pid for all volumes

BUG: 1444596
Change-Id: I5d10af69dea0d0ca19511f43870f34295a54a4d2
Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
Reviewed-on: https://review.gluster.org/17101
Smoke: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Prashanth Pai <ppai@redhat.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
2017-05-09 01:30:01 +00:00

182 lines
4.5 KiB
Bash

#!/bin/bash
CLUSTER_PFX="127.1.1"; # ".x" for each glusterd
CLUSTER_COUNT=1; # Just initial definition
function launch_cluster() {
local count=$1;
CLUSTER_COUNT=$count;
define_backends $count;
define_hosts $count;
define_glusterds $count $2;
define_clis $count;
start_glusterds;
}
function define_backends() {
local b;
for i in `seq 1 $count`; do
eval "B$i=$B0/$i";
done
for i in `seq 1 $count`; do
b="B$i";
mkdir -p ${!b}/glusterd;
done
}
function define_glusterds() {
local count=$1;
local h;
local b;
local wopt;
local bopt;
local popt;
for i in `seq 1 $count`; do
b="B$i";
h="H$i";
wopt="management.working-directory=${!b}/glusterd";
bopt="management.transport.socket.bind-address=${!h}";
popt="--pid-file=${!b}/glusterd.pid";
sopt="management.glusterd-sockfile=${!b}/glusterd/gd.sock"
#Get the logdir
logdir=`gluster --print-logdir`
#Fetch the testcases name and prefix the glusterd log with it
logfile=`echo ${0##*/}`_glusterd$i.log
lopt="--log-file=$logdir/$logfile"
if [ "$2" == "-LDEBUG" ]; then
eval "glusterd_$i='glusterd -LDEBUG --xlator-option $wopt --xlator-option $bopt --xlator-option $sopt $lopt $popt'";
eval "glusterd$i='glusterd -LDEBUG --xlator-option $wopt --xlator-option $bopt --xlator-option $sopt $lopt $popt'";
else
eval "glusterd_$i='glusterd --xlator-option $wopt --xlator-option $bopt --xlator-option $sopt $lopt $popt'";
eval "glusterd$i='glusterd --xlator-option $wopt --xlator-option $bopt --xlator-option $sopt $lopt $popt'";
fi
done
}
function start_glusterd() {
local g
local index=$1
g="glusterd_${index}"
${!g}
}
function start_glusterds() {
for i in `seq 1 $CLUSTER_COUNT`; do
start_glusterd $i
done
}
function kill_glusterd() {
local index=$1;
local b;
local pidfile;
b="B$index";
pidfile="${!b}/glusterd.pid";
kill `cat $pidfile`;
}
function kill_node() {
local index=$1;
local h;
h="H$index";
kill -9 $(ps -ef | grep gluster | grep ${!h} | awk '{print $2}');
find $B0/$index/glusterd/vols -name '*.pid' | xargs rm -f
}
function define_hosts() {
local count=$1;
for i in `seq 1 $count`; do
eval "H_$i=${CLUSTER_PFX}.$i"
eval "H$i=${CLUSTER_PFX}.$i";
case $OSTYPE in
NetBSD)
ifconfig lo0 alias ${CLUSTER_PFX}.$i 2>/dev/null
;;
*)
;;
esac
done
}
function define_clis() {
local count=$1;
local h;
for i in `seq 1 $count`; do
b="B$i";
#get the logdir
logdir=`gluster --print-logdir`
#Fetch the testcases name and prefix the cli log with it
logfile=`echo ${0##*/}`_cli$i.log
lopt="--log-file=$logdir/$logfile"
logfile1=`echo ${0##*/}`_cli_$i.log
lopt1="--log-file=$logdir/$logfile1"
eval "CLI_$i='$CLI --glusterd-sock=${!b}/glusterd/gd.sock $lopt'";
eval "CLI$i='$CLI --glusterd-sock=${!b}/glusterd/gd.sock $lopt1'";
done
}
function peer_count() {
$CLI_1 peer status | grep 'Peer in Cluster (Connected)' | wc -l
}
function cluster_rebalance_status_field {
#The rebalance status can be upto 3 words, (ex:'fix-layout in progress'), hence the awk-print $7 thru $9.
#But if the status is less than 3 words, it also prints the next field i.e the run_time_in_secs.(ex:'completed 3.00').
#So we trim the numbers out with `tr`. Finally remove the trailing white spaces with sed. What we get is one of the
#strings in the 'cli_vol_task_status_str' char array of cli-rpc-ops.c
eval \$CLI_$1 volume rebalance $2 status | awk '{print $7,$8,$9}' |sed -n 3p |tr -d '[^0-9+\.]'|sed 's/ *$//g'
}
function cluster_volinfo_field()
{
local vol=$2;
local field=$3;
eval \$CLI_$1 volume info $vol | grep "^$field: " | sed 's/.*: //';
}
function volinfo_field_1()
{
local vol=$1;
local field=$2;
$CLI_1 volume info $vol | grep "^$field: " | sed 's/.*: //';
}
function volinfo_field_2()
{
local vol=$1;
local field=$2;
$CLI_2 volume info $vol | grep "^$field: " | sed 's/.*: //';
}
function brick_up_status_1 {
local vol=$1
local host=$2
local brick=$3
$CLI_1 volume status $vol $host:$brick --xml | sed -ne 's/.*<status>\([01]\)<\/status>/\1/p'
}