[Linux-ha-jp] インターフェースダウンの障害時について

Back to archive index

sekiya_kyuus****@yahoo***** sekiya_kyuus****@yahoo*****
2011年 8月 9日 (火) 16:40:28 JST


$B$O$8$a$^$7$F4XC+$H?=$7$^$9!#(B

$B;HMQ(BOS$B!'(BRHEL5.4
pacemaker$B$N%P!<%8%g%s!'(B1.0.10-1.4

$B8=:_(B2$BBf9=@.$G%/%i%9%?%j%s%0$rAH$s$G$$$^$9!#(B

$BF1;~$K0J2<$N%j%=!<%9$r5/F0$9$k$h$&$K$7$F$$$^$9!#(B
mysql4$B7O(B
mysql5$B7O(B
postgres9$B7O(B

$B$5$i$K>e5-(BDB$B$KBP$7$F0l$D$:$D(BVIP$B$r:n @ .$9$k$h$&$K$7$F$"$j$^$9!#(B

$B0J2<$N$h$&$J9=@.$G$9!#(B
pacemaker
     $B-x(Bmysql4$B$N(BVIP$B!](Bmysql4
     $B-x(Bmysql5$B$N(BVIP$B!](Bmysql5
     $B-x(Bpostgres9$B$N(BVIP$B!](Bpostgres9

db01$B!D%j%=!<%92TF/Cf(B
db02$B!D%j%=!<%9BT5!Cf(B

$B4pK\E*$K$O>e<j$/F0$$$F$$$k$N$G$9$,!"(B
$B%$%s%?!<%U%'!<%9%@%&%s$N>c32;~$K0J2<$N$h$&$J%a%C%;!<%8$,(B
$B1J1s$H%m%0$K=PB3$1$^$9!#(B

$B!Z(Bdb01$B![(B
Aug 09 15:53:47 db01 heartbeat: [2363]: WARN: Rexmit of seq 66672 requested. 3 is max.

$B!Z(Bdb02$B![(B
Aug 09 15:53:47 db02 heartbeat: [4661]: WARN: Gmain_timeout_dispatch: Dispatch function for retransmit request took too long to execute: 20 ms (> 10 ms) (GSource: 0x9822230)

$B%W%m%;%9%@%&%s!"(BVIP$B%@%&%s!"(Bheartbeat$B$N%@%&%s(B
$B$J$I$N>c32;~$K$O$3$N$h$&$J$3$H$K$J$i$J$$$G$9!#(B
$B%$%s%?!<%U%'!<%9%@%&%s;~$N$_$G$9!#(B
($B @ 53N$K$O%$%s%?!<%U%'!<%9%"%C%W;~$G$9$,(B)

$B<B;\$7$?>c32%F%9%H!\I|5l$NN.$l$O0J2<$K$J$j$^$9!#(B
$B-!(Bdb01$B%$%s%?!<%U%'!<%9%@%&%s(B
$B-"(Bdb02$B$K(BVIP$B!\%j%=!<%9$,0\F0$7!"@5>o2TF/(B
$B-#(Bdb01$B%$%s%?!<%U%'!<%9$r%"%C%W(B

$B>e5-<j=g$N-#$r<B;\8e$K%m%0$,=P;O$a$^$9!#(B
($B-"$H-#$N4V$K(Bdb01$B$N(Bheartbeat$B$N:F5/F0$r<B;\$7$F$b7k2L$OF1$8$G$7$?(B)

db01$B$N(Bheartbeat$B$rDd;_$7$F$b(Bdb02$B$N%m%0$O=PB3$1$^$9!#(B
$B$H$$$&$o$1$G!"(Bdb01$B!"(Bdb02$B$N(Bheartbeat$B$r;_$a$J$$$H%m%0$,=PNO$5$l$k>u67$OJQ2=$J$7$G$7$?!#(B
$B#2Bf$H$bDd;_"M5/F08e$O%m%0=PNO$b$H$^$j$^$7$?!#(B

$B$5$i$K!"(Bsar$B%3%^%s%I$r;H$C$F3:Ev$9$k;~4V$N2TF/>pJs$r=P$7$F$_$?$i0J2<$N$h$&$J7k2L$,$G$^$7$?!#(B
$B!Z(Bdb01$B$N7k2L![(B
                  CPU     %user     %nice   %system   %iowait    %steal     %idle
15:50:01          all    106.63      0.00    106.63    106.64    106.64      0.00
               pgpgin/s pgpgout/s   fault/s  majflt/s
15:50:01    432866871482.24 432866871481.63 432866871446.10 432866871482.27

db02$B$NJ}$O @ 5>o$J?tCM$G$7$?!#(B
$B$J$s$+$H$s$G$b$J$$?tCM$,$G$F$7$^$C$F$$$^$7$?!#(B
$B%P%0$J$s$G$7$g$&$+!&!&!&!#(B

<crm$B%3%s%U%#%0(B>
node $id="20ed5d8a-8d4e-409d-a718-4aa3995565b5" db02 \
        attributes standby="off"
node $id="a494a68c-796a-423e-8995-559acdc23502" db01 \
        attributes standby="off"
primitive res_mysql4 ocf:heartbeat:mysql4 \
        op monitor interval="10s"
primitive res_mysql4_vip ocf:heartbeat:IPaddr2 \
        params ip="172.0.0.3" nic="eth0" cidr_netmask="25" \
        op monitor interval="10s"
primitive res_mysql5 ocf:heartbeat:mysql5 \
        op monitor interval="10s"
primitive res_mysql5_vip ocf:heartbeat:IPaddr2 \
        params ip="172.0.0.4" nic="eth0" cidr_netmask="25" \
        op monitor interval="10s"
primitive res_pgsql9 ocf:heartbeat:pgsql9 \
        op monitor interval="10s"
primitive res_pgsql9_vip ocf:heartbeat:IPaddr2 \
        params ip="172.0.0.5" nic="eth0" cidr_netmask="25" \
        op monitor interval="10s"
primitive res_pingd ocf:heartbeat:pingd \
        params name="default_ping_set" host_list="" multiplier="100" dampen="1" \
        op monitor on-fail="block" interval="10s"
group grp_mysql4 res_mysql4_vip res_mysql4 \
        meta target-role="Started"
group grp_mysql5 res_mysql5_vip res_mysql5
group grp_pgsql9 res_pgsql9_vip res_pgsql9 \
        meta target-role="Started"
clone clone_pingd res_pingd
location grp_mysql4_location grp_mysql4 \
        rule $id="grp_mysql4-rule" -inf: not_defined default_ping_set or default_ping_set lt 100
location grp_mysql5_location grp_mysql5 \
        rule $id="grp_mysql5-rule" -inf: not_defined default_ping_set or default_ping_set lt 100
location grp_pgsql9_location grp_pgsql9 \
        rule $id="grp_pgsql9-rule" -inf: not_defined default_ping_set or default_ping_set lt 100
property $id="cib-bootstrap-options" \
        dc-version="1.0.10-da7075976b5ff0bee71074385f8fd02f296ec8a3" \
        cluster-infrastructure="Heartbeat" \
        last-lrm-refresh="1312538205" \
        no-quorum-policy="ignore" \
        stonith-enabled="false"
rsc_defaults $id="rsc-options" \
        resource-stickiness="INFINITY" \
        migration-threshold="3"

<db01 ha.cf>
pacemaker on
node db01
node db02
ucast eth0 172.0.0.2
udpport 694

keepalive 5
warntime 10
deadtime 15
initdead 15

logfile /var/log/heartbeat/ha.log
debug 0
logfacility none

watchdog /dev/watchdog
respawn root /usr/lib64/heartbeat/ifcheckd

<db02 ha.cf>
pacemaker on
node db01
node db02
ucast eth0 172.0.0.1
udpport 694

keepalive 5
warntime 10
deadtime 15
initdead 15

logfile /var/log/heartbeat/ha.log
debug 0
logfacility none

watchdog /dev/watchdog
respawn root /usr/lib64/heartbeat/ifcheckd





Linux-ha-japan メーリングリストの案内
Back to archive index