Quoting "Tatsuo Ishii" <[email protected]>:
... Sometimes (not always) pgPool hangs after these tests. That means accessing the database through pgPool is not possible any more. Attemps to access the database do not result in an error message but pgPool does not answer (at least not in reasonable time). However accessing the database directly still works perfect. If I restart pgPool everything works fine again.
Following on from my previous post - I've had this occur with PgPool 2.2.2 as previously mentioned.
I didn't have debug mode turned on at the time, but here's an strace for what it's worth. I'll turn on debugging and wait for this to occur again and hopefully provide more useful info.
[pid 29666] <... select resumed> ) = ? ERESTARTNOHAND (To be restarted) [pid 29666] --- SIGCHLD (Child exited) @ 0 (0) ---[pid 29666] rt_sigprocmask(SIG_SETMASK, ~[ILL TRAP ABRT BUS FPE SEGV CONT SYS RTMIN RT_1], NULL, 8) = 0
[pid 29666] write(6, "\0", 1) = 1 [pid 29666] rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0 [pid 29666] sigreturn() = ? (mask now [])[pid 29666] rt_sigprocmask(SIG_SETMASK, ~[ILL TRAP ABRT BUS FPE SEGV CONT SYS RTMIN RT_1], NULL, 8) = 0
[pid 29666] rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0
[pid 29666] select(6, [5], NULL, NULL, {3, 0}) = 1 (in [5], left {3, 0})
[pid 29666] read(5, "\0", 1) = 1
[pid 29666] rt_sigprocmask(SIG_SETMASK, ~[ILL TRAP ABRT BUS FPE SEGV
CONT SYS RTMIN RT_1], NULL, 8) = 0
[pid 29666] wait4(-1, [{WIFEXITED(s) && WEXITSTATUS(s) == 1}],
WNOHANG, NULL) = 2586
[pid 29666] clone(Process 2588 attached (waiting for parent) Process 2588 resumed (parent 29666 ready)child_stack=0, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0xb7fdb718) = 2588
[pid 29666] time(NULL) = 1245308280
[pid 29666] wait4(-1, 0xbf89bf08, WNOHANG, NULL) = 0
[pid 2588] close(5 <unfinished ...>
[pid 29666] rt_sigprocmask(SIG_SETMASK, [], <unfinished ...>
[pid 2588] <... close resumed> ) = 0
[pid 29666] <... rt_sigprocmask resumed> NULL, 8) = 0
[pid 2588] close(6 <unfinished ...>
[pid 29666] select(6, [5], NULL, NULL, {3, 0} <unfinished ...>
[pid 2588] <... close resumed> ) = 0
[pid 2588] rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0
[pid 2588] rt_sigaction(SIGALRM, {SIG_DFL}, {SIG_DFL}, 8) = 0
[pid 2588] rt_sigaction(SIGTERM, {0x804e510, [TERM], SA_RESTART},
{0x804b360, [], SA_RESTART}, 8) = 0
[pid 2588] rt_sigaction(SIGINT, {0x804e510, [INT], SA_RESTART},
{0x804b360, [], SA_RESTART}, 8) = 0
[pid 2588] rt_sigaction(SIGHUP, {0x804d630, [HUP], SA_RESTART},
{0x804adf0, [], SA_RESTART}, 8) = 0
[pid 2588] rt_sigaction(SIGQUIT, {0x804e510, [QUIT], SA_RESTART},
{0x804b360, [], SA_RESTART}, 8) = 0
[pid 2588] rt_sigaction(SIGCHLD, {SIG_DFL}, {0x804aed0, [],
SA_RESTART|SA_NOCLDSTOP}, 8) = 0
[pid 2588] rt_sigaction(SIGUSR1, {0x804e590, [USR1], SA_RESTART},
{0x804af90, [], SA_RESTART}, 8) = 0
[pid 2588] rt_sigaction(SIGUSR2, {0x804d620, [USR2], SA_RESTART},
{0x804ae60, [], SA_RESTART}, 8) = 0
[pid 2588] rt_sigaction(SIGPIPE, {SIG_IGN}, {SIG_IGN}, 8) = 0
[pid 2588] fcntl64(3, F_GETFL) = 0x802 (flags O_RDWR|O_NONBLOCK)
[pid 2588] fcntl64(3, F_SETFL, O_RDWR|O_NONBLOCK) = 0
[pid 2588] fcntl64(4, F_GETFL) = 0x802 (flags O_RDWR|O_NONBLOCK)
[pid 2588] fcntl64(4, F_SETFL, O_RDWR|O_NONBLOCK) = 0
[pid 2588] gettimeofday({1245308280, 191094}, {4294967176, 0}) = 0
[pid 2588] gettimeofday({1245308280, 191283}, NULL) = 0
[pid 2588] select(5, [3 4], NULL, NULL, {600, 0}) = 1 (in [4], left
{599, 493000})
[pid 2587] <... select resumed> ) = 1 (in [4], left {599, 185000})
[pid 2588] gettimeofday({1245308280, 700330}, NULL) = 0
[pid 2588] accept(4, <unfinished ...>
[pid 2587] gettimeofday( <unfinished ...>
[pid 2588] <... accept resumed> 0xbf89b834, [128]) = -1 EAGAIN
(Resource temporarily unavailable)
pgpUMiIxKR9Vh.pgp
Description: PGP Digital Signature
_______________________________________________ Pgpool-general mailing list [email protected] http://pgfoundry.org/mailman/listinfo/pgpool-general
