blob: 0cea531fa1554a1b06fa1735427623726022029c (
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
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
|
#!/bin/sh
# script to manage nfs in a clustered environment
start_nfs() {
/bin/mkdir -p $CTDB_VARDIR/state/nfs
/bin/mkdir -p $CTDB_VARDIR/state/statd/ip
startstop_nfs stop
startstop_nfs start
echo 1 > /proc/sys/net/ipv4/tcp_tw_recycle
}
. $CTDB_BASE/functions
service_name="nfs"
service_start="start_nfs"
service_stop="startstop_nfs stop"
service_reconfigure="startstop_nfs restart"
loadconfig
[ "$NFS_SERVER_MODE" != "GANESHA" ] || exit 0
ctdb_start_stop_service
is_ctdb_managed_service || exit 0
case "$1" in
init)
# read statd from persistent database
;;
startup)
ctdb_service_start
mkdir -p $CTDB_VARDIR/state/statd
touch $CTDB_VARDIR/state/statd/update-trigger
;;
shutdown)
ctdb_service_stop
;;
takeip)
ctdb_service_set_reconfigure
;;
releaseip)
ctdb_service_set_reconfigure
;;
monitor)
if ctdb_service_needs_reconfigure ; then
ctdb_service_reconfigure
exit 0
fi
update_tickles 2049
# check that statd responds to rpc requests
# if statd is not running we try to restart it
# we only do this IF we have a rpc.statd command.
# For platforms where rpc.statd does not exist, we skip
# the check completely
p="rpc.statd"
which $p >/dev/null 2>/dev/null && {
if ctdb_check_rpc "STATD" 100024 1 >/dev/null ; then
(service_name="nfs_statd"; ctdb_counter_init)
else
cmd="$p"
cmd="${cmd}${STATD_HOSTNAME:+ -n }${STATD_HOSTNAME}"
cmd="${cmd}${STATD_PORT:+ -p }${STATD_PORT}"
cmd="${cmd}${STATD_OUTGOING_PORT:+ -o }${STATD_OUTGOING_PORT}"
(
service_name="nfs_statd"
ctdb_counter_incr
ctdb_check_counter_limit 10 quiet >/dev/null
) || {
echo "$ctdb_check_rpc_out"
echo "Trying to restart STATD [$cmd]"
$cmd
}
fi
}
# check that NFS responds to rpc requests
[ "$CTDB_NFS_SKIP_KNFSD_ALIVE_CHECK" = "yes" ] || {
if ctdb_check_rpc "NFS" 100003 3 >/dev/null ; then
(service_name="nfs_knfsd"; ctdb_counter_init)
else
(
service_name="nfs_knfsd"
ctdb_counter_incr
ctdb_check_counter_equal 2 || {
echo "Trying to restart NFS service"
startstop_nfs restart >/dev/null 2>&1 &
exit 0
}
ctdb_check_counter_limit 5 quiet >/dev/null
) || {
echo "$ctdb_check_rpc_out"
echo "Trying to restart NFS service"
startstop_nfs restart
exit 1
}
fi
}
# and that its directories are available
[ "$CTDB_NFS_SKIP_SHARE_CHECK" = "yes" ] || {
exportfs | grep -v '^#' | grep '^/' |
sed -e 's/[[:space:]]\+[^[:space:]]*$//' |
ctdb_check_directories
} || exit $?
# check that lockd responds to rpc requests
if ctdb_check_rpc "LOCKD" 100021 1 >/dev/null ; then
(service_name="lockd"; ctdb_counter_init)
else
(
service_name="lockd"
ctdb_counter_incr
ctdb_check_counter_equal 10 || {
echo "Trying to restart NFS lock service"
startstop_nfs restart >/dev/null 2>&1 &
startstop_nfslock restart >/dev/null 2>&1 &
exit 0
}
ctdb_check_counter_limit 15 quiet >/dev/null
) || {
echo "$ctdb_check_rpc_out"
echo "Trying to restart NFS lock service"
startstop_nfs restart
startstop_nfslock restart
exit 1
}
fi
# mount needs special handling since it is sometimes not started
# correctly on RHEL5
if ctdb_check_rpc "MOUNTD" 100005 1 >/dev/null ; then
(service_name="nfs_mountd"; ctdb_counter_init)
else
(
service_name="nfs_mountd"
ctdb_counter_incr
ctdb_check_counter_equal 5 || {
p="rpc.mountd"
cmd="${p}${MOUNTD_PORT:+ -p }${MOUNTD_PORT}"
echo "Trying to restart MOUNTD [${cmd}]"
killall -q -9 $p
$cmd &
exit 0
}
ctdb_check_counter_limit 10 quiet >/dev/null
) || {
echo "$ctdb_check_rpc_out"
p="rpc.mountd"
cmd="${p}${MOUNTD_PORT:+ -p }${MOUNTD_PORT}"
echo "Trying to restart MOUNTD [${cmd}]"
killall -q -9 $p
$cmd &
exit 1
}
fi
# rquotad needs special handling since it is sometimes not started
# correctly on RHEL5
# this is not a critical service so we dont flag the node as unhealthy
ctdb_check_rpc "RQUOTAD" 100011 1 || {
p="rpc.rquotad"
cmd="${p}${RQUOTAD_PORT:+ -p }${RQUOTAD_PORT}"
echo "Trying to restart RQUOTAD [${cmd}]"
killall -q -9 $p
$cmd &
}
# once every 600 seconds, update the statd state database for which
# clients need notifications
LAST_UPDATE=`stat --printf="%Y" $CTDB_VARDIR/state/statd/update-trigger 2>/dev/null`
CURRENT_TIME=`date +"%s"`
[ $CURRENT_TIME -ge $(($LAST_UPDATE + 600)) ] && {
mkdir -p $CTDB_VARDIR/state/statd
touch $CTDB_VARDIR/state/statd/update-trigger
$CTDB_BASE/statd-callout updatelocal &
$CTDB_BASE/statd-callout updateremote &
}
;;
ipreallocated)
# if the ips have been reallocated, we must restart the lockmanager
# across all nodes and ping all statd listeners
[ -x $CTDB_BASE/statd-callout ] && {
$CTDB_BASE/statd-callout notify &
} >/dev/null 2>&1
;;
*)
ctdb_standard_event_handler "$@"
;;
esac
exit 0
|