forked from pwendell/spark-ec2
-
Notifications
You must be signed in to change notification settings - Fork 4
/
setup.sh
executable file
·117 lines (96 loc) · 3.49 KB
/
setup.sh
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
#!/bin/bash
sudo yum install -y -q pssh
# usage: echo_time_diff name start_time end_time
echo_time_diff () {
local format='%Hh %Mm %Ss'
local diff_secs="$(($3-$2))"
echo "[timing] $1: " "$(date -u -d@"$diff_secs" +"$format")"
}
# Make sure we are in the spark-ec2 directory
pushd /root/spark-ec2 > /dev/null
# Load the environment variables specific to this AMI
source /root/.bash_profile
# Load the cluster variables set by the deploy script
source ec2-variables.sh
# Set hostname based on EC2 private DNS name, so that it is set correctly
# even if the instance is restarted with a different private DNS name
PRIVATE_DNS=`wget -q -O - http://169.254.169.254/latest/meta-data/local-hostname`
PUBLIC_DNS=`wget -q -O - http://169.254.169.254/latest/meta-data/hostname`
hostname $PRIVATE_DNS
echo $PRIVATE_DNS > /etc/hostname
export HOSTNAME=$PRIVATE_DNS # Fix the bash built-in hostname variable too
echo "Setting up Spark on `hostname`..."
# Set up the masters, slaves, etc files based on cluster env variables
echo "$MASTERS" > masters
echo "$SLAVES" > slaves
MASTERS=`cat masters`
NUM_MASTERS=`cat masters | wc -l`
OTHER_MASTERS=`cat masters | sed '1d'`
SLAVES=`cat slaves`
SSH_OPTS="-o StrictHostKeyChecking=no -o ConnectTimeout=5"
if [[ "x$JAVA_HOME" == "x" ]] ; then
echo "Expected JAVA_HOME to be set in .bash_profile!"
exit 1
fi
if [[ `tty` == "not a tty" ]] ; then
echo "Expecting a tty or pty! (use the ssh -t option)."
exit 1
fi
echo "Setting executable permissions on scripts..."
find . -regex "^.+.\(sh\|py\)" | xargs chmod a+x
echo "RSYNC'ing /root/spark-ec2 to other cluster nodes..."
rsync_start_time="$(date +'%s')"
for node in $SLAVES $OTHER_MASTERS; do
echo $node
rsync -e "ssh $SSH_OPTS" -az /root/spark-ec2 $node:/root &
scp $SSH_OPTS ~/.ssh/id_rsa $node:.ssh &
sleep 0.1
done
wait
rsync_end_time="$(date +'%s')"
echo_time_diff "rsync /root/spark-ec2" "$rsync_start_time" "$rsync_end_time"
echo "Running setup-slave on all cluster nodes to mount filesystems, etc..."
setup_slave_start_time="$(date +'%s')"
pssh --inline \
--host "$MASTERS $SLAVES" \
--user root \
--extra-args "-t -t $SSH_OPTS" \
--timeout 0 \
"spark-ec2/setup-slave.sh"
setup_slave_end_time="$(date +'%s')"
echo_time_diff "setup-slave" "$setup_slave_start_time" "$setup_slave_end_time"
# Always include 'scala' module if it's not defined as a work around
# for older versions of the scripts.
if [[ ! $MODULES =~ *scala* ]]; then
MODULES=$(printf "%s\n%s\n" "scala" $MODULES)
fi
# Install / Init module
for module in $MODULES; do
echo "Initializing $module"
module_init_start_time="$(date +'%s')"
if [[ -e $module/init.sh ]]; then
source $module/init.sh
fi
module_init_end_time="$(date +'%s')"
echo_time_diff "$module init" "$module_init_start_time" "$module_init_end_time"
cd /root/spark-ec2 # guard against init.sh changing the cwd
done
# Deploy templates
# TODO: Move configuring templates to a per-module ?
echo "Creating local config files..."
./deploy_templates.py
# Copy spark conf by default
echo "Deploying Spark config files..."
chmod u+x /root/spark/conf/spark-env.sh
/root/spark-ec2/copy-dir /root/spark/conf
# Setup each module
for module in $MODULES; do
echo "Setting up $module"
module_setup_start_time="$(date +'%s')"
source ./$module/setup.sh
sleep 0.1
module_setup_end_time="$(date +'%s')"
echo_time_diff "$module setup" "$module_setup_start_time" "$module_setup_end_time"
cd /root/spark-ec2 # guard against setup.sh changing the cwd
done
popd > /dev/null