blob: dbed877e9fdc104ced22488c186d335076472afb (
plain)
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
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
|
#!/bin/bash
# Copyright (C) 2017 AT&T Intellectual Property. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this code except in compliance
# with the License. You may obtain a copy of the License
# at http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
# implied. See the License for the specific language governing
# permissions and limitations under the License.
# NAME
# check_cluster - check the state of the cluster
#
# USAGE
# check_cluster [-v] [-l] [-t timeout]
# -l do not check localhost first (and restarting the service if necessary)
# -t timeout set how long to wait when accessing the servers
# -v verbose
#
# DESCRIPTION
# Loop through the nodes in the cluster, using pgwget to determine how many are masters, secondaries, down for maintenance, or not up.
# Complain about certain situations.
# If there are multiple masters, and this not the first master in the list, then:
# run pg_ctl_restart
# prevent /ro from returning true
CDF=/opt/app/cdf
if [ -d /opt/app/postgresql-9.5.2 ]
then PGDIR=/opt/app/postgresql-9.5.2
elif [ -d /usr/lib/postgresql/9.6 ]
then PGDIR=/usr/lib/postgresql/9.6
elif [ -d /usr/lib/postgresql/9.5 ]
then PGDIR=/usr/lib/postgresql/9.5
else echo PostgreSQL bin directory not found 1>&2; exit 1
fi
if [ -d /opt/app/postgresql-config-9.5.2/ ]
then CFGDIR=/opt/app/postgresql-config-9.5.2/
else CFGDIR=/opt/app/postgresql-config/
fi
PATH=$PGDIR/bin:$CDF/bin:/opt/app/pgaas/bin:/opt/app/postgresql-prep/bin:$PATH
usage()
{
exec 1>&2
[ $# -gt 0 ] && echo "$@"
echo "Usage: $0 [-v] [-l] [-t timeout] [-d file]"
echo -e " -l do not check localhost first (and restarting the service if necessary)"
echo -e " -t timeout set how long to wait when accessing the servers"
echo -e " -v verbose"
echo -e " -d file duplicate the status output to the given file"
exit 1
}
VERBOSE=false
TIMEOUT=10
TESTLOCAL=:
DFILE=
while getopts d:lt:v c
do
case "$c" in
d ) DFILE=$OPTARG ;;
l ) TESTLOCAL=false ;;
t ) TIMEOUT=$OPTARG ;;
v ) VERBOSE=: ;;
\?) usage ;;
esac
done
shift $(($OPTIND - 1))
# loop through the nodes in the cluster, using pgwget to determine if any are a master. Save in $@
master_count=0
secondary_count=0
total_count=0
down_count=0
maintenance_count=0
MASTERS=
SECONDARIES=
MAINTENANCES=
DOWNS=
MSEP=
SSEP=
BSEP=
DSEP=
HOSTNAME=$(hostname -f)
FOUNDPREVIOUSMASTER=
DOPGCTLSTOP=
if $TESTLOCAL
then
isrw=`pgwget --tries=1 --read-timeout=$TIMEOUT --quiet -O/dev/stdout http://localhost:8000/isrw`
case "$isrw" in
Master | Secondary | Maintenance ) ;;
* )
echo "$(date)|WARNING|RESTARTED|Local iDNS-responder.py not responding. Restarting."
ps -fu postgres | grep "python3 /opt/app/postgresql-prep/bin/iDNS-responder.py" | grep -v grep | awk '{print "kill " $2}' | sh
sleep 10
;;
esac
fi
for i in $(getpropvalue -n pgnodes | sed 's/|/ /g')
do
$VERBOSE && echo -n "Checking $i"
isrw=`pgwget --tries=1 --read-timeout=10 --quiet -O/dev/stdout http://$i:8000/isrw`
$VERBOSE && echo ": $isrw"
case "$isrw" in
Master )
(( master_count = master_count + 1 ))
(( total_count = total_count + 1 ))
MASTERS="$MASTERS$MSEP$i"
MSEP=" "
if [ -z "$FOUNDPREVIOUSMASTER" ]
then FOUNDPREVIOUSMASTER=yes
elif [ -n "$FOUNDPREVIOUSMASTER" -a "$i" = $HOSTNAME ]
then DOPGCTLSTOP=yes
fi
;;
Secondary )
(( secondary_count = secondary_count + 1 ))
(( total_count = total_count + 1 ))
SECONDARIES="$SECONDARIES$SSEP$i"
SSEP=" "
;;
Maintenance )
(( maintenance_count = maintenance_count + 1 ))
(( total_count = total_count + 1 ))
MAINTENANCES="$MAINTENANCES$BSEP$i"
BSEP=" "
;;
* )
DOWNS="$DOWNS$DSEP$i"
DSEP=" "
(( down_count = down_count + 1 ))
(( total_count = total_count + 1 ))
;;
esac
done
(( up_count = master_count + secondary_count ))
date=$(date)
output="$date|INFO|masters=$master_count $MASTERS|secondaries=$secondary_count $SECONDARIES|maintenance=$maintenance_count $MAINTENANCES|down=$down_count $DOWNS|"
echo "$output"
if [ -n "$DFILE" ]
then (umask 022; echo "$output" > $DFILE.tmp && mv $DFILE.tmp $DFILE)
fi
FORCEROOFF=/var/run/postgresql/force-ro-off
if [ $master_count -lt 1 ]
then echo "$date|FATAL|NOMASTER|NO MASTER FOUND|"
elif [ $master_count -gt 1 ]
then echo "$date|FATAL|MULTIPLEMASTERS|TOO MANY MASTERS FOUND|"
if [ -n "$DOPGCTLSTOP" ]
then
touch "$FORCEROOFF"
pg_ctl_stop
fi
fi
if [ -z "$DOPGCTLSTOP" -a -f "$FORCEROOFF" ]
then rm -f "$FORCEROOFF"
fi
if [ $up_count -eq 0 ]
then echo "$date|FATAL|NOREADER|NO SECONDARY FOUND"
fi
if [ $down_count -ne 0 ]
then echo "$date|ERROR|DOWN|One or more systems are down: $DOWNS"
fi
|
