[ClusterLabs] Cluster stop works after upgrade 1.1.15-rc1

Mon Jun 6 21:03:01 UTC 2016

On 05/24/2016 02:32 AM, Andrey Rogovsky wrote:
> Hi!
> I have 3 nodes cluster in this config:
> 
> # crm configure show
> node 1084754433: a \
> attributes pgsql-data-status="STREAMING|ASYNC" maintenance=off standby=off
> node 1084754434: b \
> attributes pgsql-data-status=LATEST maintenance=off standby=off
> node 1084754435: c \
> attributes pgsql-data-status="STREAMING|ASYNC" maintenance=off
> primitive apache apache \
> params configfile="/etc/apache2/apache2.conf" port=8000
> statusurl="http://localhost:8000/server-status" \
> op monitor interval=2min on-fail=restart \
> op monitor interval=30s on-fail=restart timeout=160s \
> meta is-managed=true
> primitive pgsql-master-ip IPaddr2 \
> params ip=192.168.10.200 nic=peervpn0 \
> op start interval=0s on-fail=restart timeout=60s \
> op monitor interval=10s on-fail=restart timeout=60s \
> op stop interval=0s on-fail=block timeout=60s \
> meta target-role=Started
> primitive pgsqld pgsqlms \
> params bindir="/usr/lib/postgresql/9.4/bin"
> pgdata="/var/lib/postgresql/9.4/main" pghost="/var/run/postgresql"
> recovery_template="/etc/postgresql/9.4/main/recovery.conf.pcmk"
> start_opts="-c config_file=/etc/postgresql/9.4/main/postgresql.conf"
> primary_node=a \
> op start interval=0 on-fail=restart timeout=60s \
> op monitor interval=40s on-fail=restart timeout=60s \
> op promote interval=0s on-fail=restart timeout=60s \
> op demote interval=0s on-fail=stop timeout=120s \
> op monitor interval=30s on-fail=restart role=Master timeout=60s \
> op stop interval=0 on-fail=block timeout=60s \
> op notify interval=0s timeout=60s
> group master pgsql-master-ip
> ms msPostgresql pgsqld \
> meta master-max=1 master-node-max=1 clone-max=3 clone-node-max=1
> target-role=Master notify=true is-managed=true
> clone WebFarm apache \
> meta target-role=Started
> location pgsql-prefers-a msPostgresql role=Master 1000: a
> colocation set_ip inf: master msPostgresql:Master
> order ip_down 0: msPostgresql:demote master:stop symmetrical=false
> order ip_up 0: msPostgresql:promote master:start symmetrical=false
> property cib-bootstrap-options: \
> dc-version=1.1.15-88ac26d \
> cluster-infrastructure=corosync \
> expected-quorum-votes=3 \
> stonith-enabled=false \
> crmd-transition-delay=0 \
> last-lrm-refresh=1461400181 \
> dc-deadtime=2min \
> default-action-timeout=2min \
> have-watchdog=true
> rsc_defaults rsc-options: \
> resource-stickiness=10000 \
> migration-threshold=10
> 
> So, I have master IP 192.168.10.200 which will assigned to
> new msPostgresql master.
> All works fine until I update software to 1.1.15-rc1
> Now IP 192.168.10.200 is not assigned to new msPostgresql master:
> 
> 
> May 24 10:25:45 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (<null>)
> May 24 10:25:45 b attrd[1798]:   notice: Sent delete 914043:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:45 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (<null>)
> May 24 10:25:45 b attrd[1798]:   notice: Sent delete 914046:
> node=1084754434, attr=nodes, id=<n/a>, set=(null), section=status
> May 24 10:25:45 b pgsqlms(pgsqld)[2533]: INFO: pgsql_notify: current
> node LSN: 37B/876EC668
> May 24 10:25:45 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (37B/876EC668)
> May 24 10:25:45 b attrd[1798]:   notice: Sent update 914050:
> lsn_location=37B/876EC668
> May 24 10:25:45 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (c b)
> May 24 10:25:45 b attrd[1798]:   notice: Sent update 914052: nodes=c b
> May 24 10:25:45 b crmd[1800]:   notice: Operation pgsqld_notify_0: ok
> (node=b, call=68635, rc=0, cib-update=0, confirmed=true)
> May 24 10:25:45 b pgsqlms(pgsqld)[2554]: INFO: pgsql_notify: promoting
> instance on node "b"
> May 24 10:25:45 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (<null>)
> May 24 10:25:45 b attrd[1798]:   notice: Sent delete 914056:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:45 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (<null>)
> May 24 10:25:45 b attrd[1798]:   notice: Sent delete 914060:
> node=1084754434, attr=nodes, id=<n/a>, set=(null), section=status
> May 24 10:25:45 b pgsqlms(pgsqld)[2554]: INFO: pgsql_notify: current
> node LSN: 37B/876EC668
> May 24 10:25:45 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (37B/876EC668)
> May 24 10:25:45 b attrd[1798]:   notice: Sent update 914063:
> lsn_location=37B/876EC668
> May 24 10:25:45 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (b c)
> May 24 10:25:45 b attrd[1798]:   notice: Sent update 914065: nodes=b c
> May 24 10:25:45 b crmd[1800]:   notice: Operation pgsqld_notify_0: ok
> (node=b, call=68636, rc=0, cib-update=0, confirmed=true)
> May 24 10:25:45 b postgres[2280]: [2-1] 2016-05-24 10:25:45 MSK FATAL:
>  could not connect to the primary server: could not connect to server:
> No route to host
> May 24 10:25:45 b postgres[2280]: [2-2] #011#011Is the server running on
> host "192.168.10.200" and accepting
> May 24 10:25:45 b postgres[2280]: [2-3] #011#011TCP/IP connections on
> port 5432?
> May 24 10:25:45 b postgres[2280]: [2-4]
> May 24 10:25:45 b pgsqlms(pgsqld)[2577]: INFO: pgsql_notify: promoting
> instance on node "b"
> May 24 10:25:45 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (<null>)
> May 24 10:25:45 b attrd[1798]:   notice: Sent delete 914069:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:45 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (<null>)
> May 24 10:25:45 b attrd[1798]:   notice: Sent delete 914073:
> node=1084754434, attr=nodes, id=<n/a>, set=(null), section=status
> May 24 10:25:45 b pgsqlms(pgsqld)[2577]: INFO: pgsql_notify: current
> node LSN: 37B/876EC668
> May 24 10:25:45 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (37B/876EC668)
> May 24 10:25:45 b attrd[1798]:   notice: Sent update 914076:
> lsn_location=37B/876EC668
> May 24 10:25:45 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (b c)
> May 24 10:25:45 b attrd[1798]:   notice: Sent update 914078: nodes=b c
> May 24 10:25:45 b crmd[1800]:   notice: Operation pgsqld_notify_0: ok
> (node=b, call=68637, rc=0, cib-update=0, confirmed=true)
> May 24 10:25:45 b pgsqlms(pgsqld)[2599]: INFO: pgsql_notify: promoting
> instance on node "b"
> May 24 10:25:45 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (<null>)
> May 24 10:25:45 b attrd[1798]:   notice: Sent delete 914082:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:45 b attrd[1798]:   notice: Sent delete 914084:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:45 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (<null>)
> May 24 10:25:45 b attrd[1798]:   notice: Sent delete 914087:
> node=1084754434, attr=nodes, id=<n/a>, set=(null), section=status
> May 24 10:25:46 b pgsqlms(pgsqld)[2599]: INFO: pgsql_notify: current
> node LSN: 37B/876EC668
> May 24 10:25:46 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (37B/876EC668)
> May 24 10:25:46 b attrd[1798]:   notice: Sent update 914090:
> lsn_location=37B/876EC668
> May 24 10:25:46 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (b c)
> May 24 10:25:46 b attrd[1798]:   notice: Sent update 914092: nodes=b c
> May 24 10:25:46 b crmd[1800]:   notice: Operation pgsqld_notify_0: ok
> (node=b, call=68638, rc=0, cib-update=0, confirmed=true)
> May 24 10:25:46 b pgsqlms(pgsqld)[2623]: INFO: pgsql_notify: promoting
> instance on node "b"
> May 24 10:25:46 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (<null>)
> May 24 10:25:46 b attrd[1798]:   notice: Sent delete 914096:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:46 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (<null>)
> May 24 10:25:46 b attrd[1798]:   notice: Sent delete 914099:
> node=1084754434, attr=nodes, id=<n/a>, set=(null), section=status
> May 24 10:25:46 b pgsqlms(pgsqld)[2623]: INFO: pgsql_notify: current
> node LSN: 37B/876EC668
> May 24 10:25:46 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (37B/876EC668)
> May 24 10:25:46 b attrd[1798]:   notice: Sent update 914103:
> lsn_location=37B/876EC668
> May 24 10:25:46 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (b c)
> May 24 10:25:46 b attrd[1798]:   notice: Sent update 914105: nodes=b c
> May 24 10:25:46 b crmd[1800]:   notice: Operation pgsqld_notify_0: ok
> (node=b, call=68639, rc=0, cib-update=0, confirmed=true)
> May 24 10:25:46 b pgsqlms(pgsqld)[2644]: INFO: pgsql_notify: promoting
> instance on node "b"
> May 24 10:25:46 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (<null>)
> May 24 10:25:46 b attrd[1798]:   notice: Sent delete 914109:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:46 b attrd[1798]:   notice: Sent delete 914111:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:46 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (<null>)
> May 24 10:25:46 b attrd[1798]:   notice: Sent delete 914114:
> node=1084754434, attr=nodes, id=<n/a>, set=(null), section=status
> May 24 10:25:46 b pgsqlms(pgsqld)[2644]: INFO: pgsql_notify: current
> node LSN: 37B/876EC668
> May 24 10:25:46 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (37B/876EC668)
> May 24 10:25:46 b attrd[1798]:   notice: Sent update 914117:
> lsn_location=37B/876EC668
> May 24 10:25:46 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (c b)
> May 24 10:25:46 b attrd[1798]:   notice: Sent update 914119: nodes=c b
> May 24 10:25:46 b crmd[1800]:   notice: Operation pgsqld_notify_0: ok
> (node=b, call=68640, rc=0, cib-update=0, confirmed=true)
> May 24 10:25:46 b pgsqlms(pgsqld)[2665]: INFO: pgsql_notify: promoting
> instance on node "b"
> May 24 10:25:46 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (<null>)
> May 24 10:25:46 b attrd[1798]:   notice: Sent delete 914123:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:46 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (<null>)
> May 24 10:25:46 b attrd[1798]:   notice: Sent delete 914127:
> node=1084754434, attr=nodes, id=<n/a>, set=(null), section=status
> May 24 10:25:46 b pgsqlms(pgsqld)[2665]: INFO: pgsql_notify: current
> node LSN: 37B/876EC668
> May 24 10:25:46 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (37B/876EC668)
> May 24 10:25:46 b attrd[1798]:   notice: Sent update 914130:
> lsn_location=37B/876EC668
> May 24 10:25:46 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (b c)
> May 24 10:25:46 b attrd[1798]:   notice: Sent update 914132: nodes=b c
> May 24 10:25:46 b crmd[1800]:   notice: Operation pgsqld_notify_0: ok
> (node=b, call=68641, rc=0, cib-update=0, confirmed=true)
> May 24 10:25:46 b pgsqlms(pgsqld)[2686]: INFO: pgsql_notify: promoting
> instance on node "b"
> May 24 10:25:46 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (<null>)
> May 24 10:25:46 b attrd[1798]:   notice: Sent delete 914136:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:46 b attrd[1798]:   notice: Sent delete 914138:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:46 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (<null>)
> May 24 10:25:46 b attrd[1798]:   notice: Sent delete 914141:
> node=1084754434, attr=nodes, id=<n/a>, set=(null), section=status
> May 24 10:25:46 b pgsqlms(pgsqld)[2686]: INFO: pgsql_notify: current
> node LSN: 37B/876EC668
> May 24 10:25:46 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (37B/876EC668)
> May 24 10:25:46 b attrd[1798]:   notice: Sent update 914144:
> lsn_location=37B/876EC668
> May 24 10:25:46 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (c b)
> May 24 10:25:46 b attrd[1798]:   notice: Sent update 914146: nodes=c b
> May 24 10:25:46 b crmd[1800]:   notice: Operation pgsqld_notify_0: ok
> (node=b, call=68642, rc=0, cib-update=0, confirmed=true)
> May 24 10:25:47 b pgsqlms(pgsqld)[2707]: INFO: pgsql_notify: promoting
> instance on node "b"
> May 24 10:25:47 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (<null>)
> May 24 10:25:47 b attrd[1798]:   notice: Sent delete 914150:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:47 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (<null>)
> May 24 10:25:47 b attrd[1798]:   notice: Sent delete 914153:
> node=1084754434, attr=nodes, id=<n/a>, set=(null), section=status
> May 24 10:25:47 b pgsqlms(pgsqld)[2707]: INFO: pgsql_notify: current
> node LSN: 37B/876EC668
> May 24 10:25:47 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (37B/876EC668)
> May 24 10:25:47 b attrd[1798]:   notice: Sent update 914157:
> lsn_location=37B/876EC668
> May 24 10:25:47 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (b c)
> May 24 10:25:47 b attrd[1798]:   notice: Sent update 914159: nodes=b c
> May 24 10:25:47 b crmd[1800]:   notice: Operation pgsqld_notify_0: ok
> (node=b, call=68643, rc=0, cib-update=0, confirmed=true)
> May 24 10:25:47 b pgsqlms(pgsqld)[2728]: INFO: pgsql_notify: promoting
> instance on node "b"
> May 24 10:25:47 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (<null>)
> May 24 10:25:47 b attrd[1798]:   notice: Sent delete 914163:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:47 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (<null>)
> May 24 10:25:47 b attrd[1798]:   notice: Sent delete 914167:
> node=1084754434, attr=nodes, id=<n/a>, set=(null), section=status
> May 24 10:25:47 b pgsqlms(pgsqld)[2728]: INFO: pgsql_notify: current
> node LSN: 37B/876EC668
> May 24 10:25:47 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (37B/876EC668)
> May 24 10:25:47 b attrd[1798]:   notice: Sent update 914170:
> lsn_location=37B/876EC668
> May 24 10:25:47 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (b c)
> May 24 10:25:47 b attrd[1798]:   notice: Sent update 914172: nodes=b c
> May 24 10:25:47 b crmd[1800]:   notice: Operation pgsqld_notify_0: ok
> (node=b, call=68644, rc=0, cib-update=0, confirmed=true)
> May 24 10:25:47 b pgsqlms(pgsqld)[2751]: INFO: pgsql_notify: promoting
> instance on node "b"
> May 24 10:25:47 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (<null>)
> May 24 10:25:47 b attrd[1798]:   notice: Sent delete 914178:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:47 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (<null>)
> May 24 10:25:47 b attrd[1798]:   notice: Sent delete 914181:
> node=1084754434, attr=nodes, id=<n/a>, set=(null), section=status
> May 24 10:25:47 b pgsqlms(pgsqld)[2751]: INFO: pgsql_notify: current
> node LSN: 37B/876EC668
> May 24 10:25:47 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (37B/876EC668)
> May 24 10:25:47 b attrd[1798]:   notice: Sent update 914184:
> lsn_location=37B/876EC668
> May 24 10:25:47 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (b c)
> May 24 10:25:47 b attrd[1798]:   notice: Sent update 914186: nodes=b c
> May 24 10:25:47 b crmd[1800]:   notice: Operation pgsqld_notify_0: ok
> (node=b, call=68645, rc=0, cib-update=0, confirmed=true)
> May 24 10:25:47 b pgsqlms(pgsqld)[2774]: INFO: pgsql_notify: promoting
> instance on node "b"
> May 24 10:25:47 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (<null>)
> May 24 10:25:47 b attrd[1798]:   notice: Sent delete 914190:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:47 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (<null>)
> May 24 10:25:47 b attrd[1798]:   notice: Sent delete 914193:
> node=1084754434, attr=nodes, id=<n/a>, set=(null), section=status
> May 24 10:25:47 b pgsqlms(pgsqld)[2774]: INFO: pgsql_notify: current
> node LSN: 37B/876EC668
> May 24 10:25:47 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (37B/876EC668)
> May 24 10:25:47 b attrd[1798]:   notice: Sent update 914197:
> lsn_location=37B/876EC668
> May 24 10:25:47 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (c b)
> May 24 10:25:47 b attrd[1798]:   notice: Sent update 914199: nodes=c b
> May 24 10:25:47 b crmd[1800]:   notice: Operation pgsqld_notify_0: ok
> (node=b, call=68646, rc=0, cib-update=0, confirmed=true)
> May 24 10:25:48 b pgsqlms(pgsqld)[2795]: INFO: pgsql_notify: promoting
> instance on node "b"
> May 24 10:25:48 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (<null>)
> May 24 10:25:48 b attrd[1798]:   notice: Sent delete 914203:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:48 b attrd[1798]:   notice: Sent delete 914205:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:48 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (<null>)
> May 24 10:25:48 b attrd[1798]:   notice: Sent delete 914208:
> node=1084754434, attr=nodes, id=<n/a>, set=(null), section=status
> May 24 10:25:48 b pgsqlms(pgsqld)[2795]: INFO: pgsql_notify: current
> node LSN: 37B/876EC668
> May 24 10:25:48 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (37B/876EC668)
> May 24 10:25:48 b attrd[1798]:   notice: Sent update 914211:
> lsn_location=37B/876EC668
> May 24 10:25:48 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (c b)
> May 24 10:25:48 b attrd[1798]:   notice: Sent update 914213: nodes=c b
> May 24 10:25:48 b crmd[1800]:   notice: Operation pgsqld_notify_0: ok
> (node=b, call=68647, rc=0, cib-update=0, confirmed=true)
> May 24 10:25:48 b pgsqlms(pgsqld)[2816]: INFO: pgsql_notify: promoting
> instance on node "b"
> May 24 10:25:48 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (<null>)
> May 24 10:25:48 b attrd[1798]:   notice: Sent delete 914217:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:48 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (<null>)
> May 24 10:25:48 b attrd[1798]:   notice: Sent delete 914220:
> node=1084754434, attr=nodes, id=<n/a>, set=(null), section=status
> May 24 10:25:48 b pgsqlms(pgsqld)[2816]: INFO: pgsql_notify: current
> node LSN: 37B/876EC668
> May 24 10:25:48 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (37B/876EC668)
> May 24 10:25:48 b attrd[1798]:   notice: Sent update 914224:
> lsn_location=37B/876EC668
> May 24 10:25:48 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (b c)
> May 24 10:25:48 b attrd[1798]:   notice: Sent update 914226: nodes=b c
> May 24 10:25:48 b crmd[1800]:   notice: Operation pgsqld_notify_0: ok
> (node=b, call=68648, rc=0, cib-update=0, confirmed=true)
> May 24 10:25:48 b pgsqlms(pgsqld)[2837]: INFO: pgsql_notify: promoting
> instance on node "b"
> May 24 10:25:48 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (<null>)
> May 24 10:25:48 b attrd[1798]:   notice: Sent delete 914232:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:48 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (<null>)
> May 24 10:25:48 b attrd[1798]:   notice: Sent delete 914235:
> node=1084754434, attr=nodes, id=<n/a>, set=(null), section=status
> May 24 10:25:48 b pgsqlms(pgsqld)[2837]: INFO: pgsql_notify: current
> node LSN: 37B/876EC668
> May 24 10:25:48 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (37B/876EC668)
> May 24 10:25:48 b attrd[1798]:   notice: Sent update 914238:
> lsn_location=37B/876EC668
> May 24 10:25:48 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (b c)
> May 24 10:25:48 b attrd[1798]:   notice: Sent update 914240: nodes=b c
> May 24 10:25:48 b crmd[1800]:   notice: Operation pgsqld_notify_0: ok
> (node=b, call=68649, rc=0, cib-update=0, confirmed=true)
> May 24 10:25:48 b pgsqlms(pgsqld)[2858]: INFO: pgsql_notify: promoting
> instance on node "b"
> May 24 10:25:48 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (<null>)
> May 24 10:25:48 b attrd[1798]:   notice: Sent delete 914244:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:48 b attrd[1798]:   notice: Sent delete 914246:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:48 b postgres[2773]: [2-1] 2016-05-24 10:25:48 MSK FATAL:
>  could not connect to the primary server: could not connect to server:
> No route to host
> May 24 10:25:48 b postgres[2773]: [2-2] #011#011Is the server running on
> host "192.168.10.200" and accepting
> May 24 10:25:48 b postgres[2773]: [2-3] #011#011TCP/IP connections on
> port 5432?
> May 24 10:25:48 b postgres[2773]: [2-4]
> May 24 10:25:48 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (<null>)
> May 24 10:25:48 b attrd[1798]:   notice: Sent delete 914249:
> node=1084754434, attr=nodes, id=<n/a>, set=(null), section=status
> May 24 10:25:48 b pgsqlms(pgsqld)[2858]: INFO: pgsql_notify: current
> node LSN: 37B/876EC668
> May 24 10:25:48 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (37B/876EC668)
> May 24 10:25:48 b attrd[1798]:   notice: Sent update 914252:
> lsn_location=37B/876EC668
> May 24 10:25:48 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (c b)
> May 24 10:25:48 b attrd[1798]:   notice: Sent update 914254: nodes=c b
> May 24 10:25:48 b crmd[1800]:   notice: Operation pgsqld_notify_0: ok
> (node=b, call=68650, rc=0, cib-update=0, confirmed=true)
> May 24 10:25:49 b pgsqlms(pgsqld)[2880]: INFO: pgsql_notify: promoting
> instance on node "b"
> May 24 10:25:49 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (<null>)
> May 24 10:25:49 b attrd[1798]:   notice: Sent delete 914258:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:49 b attrd[1798]:   notice: Sent delete 914260:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:49 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (<null>)
> May 24 10:25:49 b attrd[1798]:   notice: Sent delete 914263:
> node=1084754434, attr=nodes, id=<n/a>, set=(null), section=status
> May 24 10:25:49 b pgsqlms(pgsqld)[2880]: INFO: pgsql_notify: current
> node LSN: 37B/876EC668
> May 24 10:25:49 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (37B/876EC668)
> May 24 10:25:49 b attrd[1798]:   notice: Sent update 914266:
> lsn_location=37B/876EC668
> May 24 10:25:49 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (b c)
> May 24 10:25:49 b attrd[1798]:   notice: Sent update 914268: nodes=b c
> May 24 10:25:49 b crmd[1800]:   notice: Operation pgsqld_notify_0: ok
> (node=b, call=68651, rc=0, cib-update=0, confirmed=true)
> May 24 10:25:49 b pgsqlms(pgsqld)[2901]: INFO: pgsql_notify: promoting
> instance on node "b"
> May 24 10:25:49 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (<null>)
> May 24 10:25:49 b attrd[1798]:   notice: Sent delete 914272:
> node=1084754434, attr=lsn_location, id=<n/a>, set=(null), section=status
> May 24 10:25:49 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (<null>)
> May 24 10:25:49 b attrd[1798]:   notice: Sent delete 914276:
> node=1084754434, attr=nodes, id=<n/a>, set=(null), section=status
> May 24 10:25:49 b pgsqlms(pgsqld)[2901]: INFO: pgsql_notify: current
> node LSN: 37B/876EC668
> May 24 10:25:49 b attrd[1798]:   notice: Sending flush op to all hosts
> for: lsn_location (37B/876EC668)
> May 24 10:25:49 b attrd[1798]:   notice: Sent update 914279:
> lsn_location=37B/876EC668
> May 24 10:25:49 b attrd[1798]:   notice: Sending flush op to all hosts
> for: nodes (b c)
> May 24 10:25:49 b attrd[1798]:   notice: Sent update 914281: nodes=b c
> May 24 10:25:49 b crmd[1800]:   notice: Operation pgsqld_notify_0: ok
> (node=b, call=68652, rc=0, cib-update=0, confirmed=true)
> 
> 
> This looks like RA IPaddr2 is not work correct or pacemaker ignore this
> config and not up on promoted node b.
> 
> How I can resolve this problem?

It looks like the IP address isn't able to be placed because pgsqlms on
b isn't being promoted correctly. The cluster seems to be in a tight
loop of trying to promote it.

Can you attach the output of crm_report from around this time?

crm_report --from "YYYY-M-D H:M:S" --to "YYYY-M-D H:M:S"