[pgpool-general-jp: 1177] Re: ヘルスチェックの失敗について

[Tatsuo Ishii]

$B ＠ P0f$H?=$7$^$9!#(B

> $B3^86$H?=$7$^$9!#$*@$OC$K$J$C$F$*$j$^$9!#(B
> 
> PostgreSQL 9.2.4$B$H(Bpgpool-$B-6(B 3.2.5$B$r;H$C$F(B
> $B%Y%s%A%^!<%/B,Dj$r9T$$$^$7$?$,!"(B
> $BB,DjESCf$G%U%'%$%k%*!<%P$,H/@8$7$^$7$?!#(B
> 
> [$BB,Dj4D6-(B]
> $B!!(Bpgpool-$B-6%5!<%P#1Bf(B(master-slave$B%b!<%I$G;HMQ(B)
> $B!!(BPostgreSQL$B$O(BSynchronous Replication$B$N#2Bf9=@.(B
> $B!!%/%i%$%"%s%H#3Bf$+$iIi2Y$r$+$1$kB,Dj(B
> 
> $BB,Dj;~$N(Bpgpool-$B-6%m%0$K$O!"0J2<$N%a%C%;!<%8$,$"$j$^$7$?!#(B
> 
> --
> 
> $B!!(B2013-07-11 13:17:09 ERROR: pid 15380: connect_inet_domain_socket: select() timed out
> $B!!(B2013-07-11 13:17:09 ERROR: pid 15380: make_persistent_db_connection: connection to 192.168.4.21(5432) failed
> $B!!(B2013-07-11 13:30:50 LOG:   pid 13911: connect_inet_domain_socket: select() timed out. retrying...
> $B!!(B2013-07-11 13:30:51 LOG:   pid 13911: connect_inet_domain_socket: select() timed out. retrying...
> $B!!(B2013-07-11 13:36:21 ERROR: pid 15380: connect_inet_domain_socket: select() timed out
> $B!!(B2013-07-11 13:36:21 ERROR: pid 15380: make_persistent_db_connection: connection to 192.168.4.21(5432) failed
> $B!!(B2013-07-11 13:36:21 ERROR: pid 15380: health check failed. 1 th host 192.168.4.21 at port 5432 is down
> $B!!!D(B
> $B!!(B2013-07-11 13:36:21 LOG:   pid 15380: failover: set new primary node: 0
> $B!!(B2013-07-11 13:36:21 LOG:   pid 15380: failover: set new master node: 0
> $B!!(B2013-07-11 13:36:21 LOG:   pid 15380: failover done. shutdown host 192.168.4.21(5432)
> 
> --
> 
> $B%X%k%9%A%'%C%/%W%m%;%9$,<B9T$7$?(Bselect$B$G(B
> timeout$B$,5/$3$C$?$3$H$,860x$G$O$J$$$+$H9M$($F$*$j$^$9!#(B

$B$=$NDL$j$G$9$M!#(B

> $BB,Dj;~$N%X%k%9%A%'%C%/4X78 ＠ _DjCM$O0J2<$G$7$?!#(B
> 
> --
> 
> $B!!(Bhealth_check_period = 10
> $B!!(Bhealth_check_timeout = 60
> $B!!(Bhealth_check_user = 'postgres'
> $B!!(Bhealth_check_password = ''
> $B!!(Bhealth_check_max_retries = 0
> $B!!(Bhealth_check_retry_delay = 1
> 
> --
> 
> $B$=$N8e!"%f!<%6%^%K%e%"%k$N(Bhealth_check_max_retries$B$N5-=R$r;29M$K!"(B
> health_check_max_retries = 3$B$K ＠ _Dj$7$F:FB,Dj$7$?$H$3$m(B
> $B%U%'%$%k%*!<%P$r2sHr$G$-$^$7$?!#(B
> 
> $B<ALd;v9`(B
> --
> $B!&%f!<%6%^%K%e%"%k$N(Bhealth_check_max_retries$B$N5-=R$O0J2<$K$J$C$F$*$j$^$9$,!"(B
> 
> $B!!!V$3$N ＠ _Dj$OF0:n$K$`$i$N$"$k%M%C%H%o!<%/4D6-$K$*$$$F!"%^%9%?$, ＠ 5>o$G$"$k$K$b4X$o$i$:(B
> $B!!!!$?$^$K%X%k%9%A%'%C%/$,<:GT$9$k$3$H$,M=A[$5$l$k>l9g$KM-MQ$G$9!#!W(B
> 
> $B!!K\7o$GH/@8$7$?%X%k%9%A%'%C%/$N<:GT(B($B5Z$S%U%'%$%k%*!<%P(B)$B$O!"(B
> $B!!(B"$B$?$^$K%X%k%9%A%'%C%/$,<:GT$9$k(B"$B%1!<%9$KEv$F$O$^$k$N$G$7$g$&$+!)(B

$B%m%0$r8+$k8B$j!"$=$N%1!<%9$@$H;W$$$^$9!#(B

> $B!!$^$?!"<:GT$9$k%1!<%9$H$7$FB>$K$I$N$h$&$J%1!<%9$,A[Dj$5$l$k$N$+!"65$($FD:$1$J$$$G$7$g$&$+!)(B

$B%X%k%9%A%'%C%/$G$O!"(Bnon blocking socket$B$r;H$C$F(BPostgreSQL$B$K(Bconnect(2)$B$7(B
$B$h$&$H$7$^$9!#$=$N:](Bselect(2)$B$N%?%$%`%"%&%H$r(B1$BIC(B($B%W%m%0%i%`Fb8GDjCM(B)$B$G(B
$B ＠ _Dj$7$F$*$j!"(B

> $B!!(B2013-07-11 13:36:21 ERROR: pid 15380: connect_inet_domain_socket: select() timed out

$B$O!"(Bconnect(2)$B$,(B1$BIC0JFb$K1~Ez$7$J$+$C$?$3$H$r<($7$^$9!#(B
health_check_max_retries = 3$B$H$9$k$H!"$3$N(Bconnect(2)$B$N;n$_$r(B3$B2s$^$G9T$$(B
$B$^$9!#(B3$B2s0JFb$K$&$^$/$$$1$P!"%X%k%9%A%'%C%/$O ＠ .8y$7$?$b$N$H8+$J$5$l$^$9!#(B

connect(2)$B$,(B1$BIC0JFb$K1~Ez$7$J$$M}M3$G$9$,!"$?$^$?$^%M%C%H%o!<%/$,D4;R0-(B
$B$$>l9g$b$"$k$G$7$g$&$7!"(BPostgreSQL$B$NIi2Y$,9b$/$F(Bpostmaster$B$,1~Ez$7$J$+$C(B
$B$?%1!<%9$b9M$($i$l$^$9!#(B
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese: http://www.sraoss.co.jp