lseek

Hi all,

what might possibly be wrong with a database/statement what so ever, if
a strace to a postgres process shows this over several ours without
terminating?

lseek(3, 0, SEEK_END) = 0
lseek(5, 0, SEEK_END) = 8192
lseek(6, 0, SEEK_END) = 0
lseek(9, 0, SEEK_END) = 0
lseek(10, 0, SEEK_END) = 0
lseek(11, 0, SEEK_END) = 0
lseek(13, 0, SEEK_END) = 0
lseek(14, 0, SEEK_END) = 0
lseek(15, 0, SEEK_END) = 0
lseek(16, 0, SEEK_END) = 0
lseek(17, 0, SEEK_END) = 0
lseek(18, 0, SEEK_END) = 0
lseek(19, 0, SEEK_END) = 0
lseek(20, 0, SEEK_END) = 0
lseek(21, 0, SEEK_END) = 11214848
lseek(22, 0, SEEK_END) = 0
lseek(23, 0, SEEK_END) = 0
lseek(24, 0, SEEK_END) = 0
lseek(25, 0, SEEK_END) = 0
lseek(26, 0, SEEK_END) = 0
lseek(27, 0, SEEK_END) = 0
lseek(28, 0, SEEK_END) = 0
lseek(29, 0, SEEK_END) = 0
lseek(30, 0, SEEK_END) = 0
lseek(31, 0, SEEK_END) = 0
lseek(32, 0, SEEK_END) = 0
lseek(3, 0, SEEK_END) = 0
lseek(5, 0, SEEK_END) = 8192
lseek(6, 0, SEEK_END) = 0
lseek(9, 0, SEEK_END) = 0
lseek(10, 0, SEEK_END) = 0
lseek(11, 0, SEEK_END) = 0
lseek(13, 0, SEEK_END) = 0
lseek(14, 0, SEEK_END) = 0
lseek(15, 0, SEEK_END) = 0
lseek(16, 0, SEEK_END) = 0
lseek(17, 0, SEEK_END) = 0
lseek(18, 0, SEEK_END) = 0
lseek(19, 0, SEEK_END) = 0
lseek(20, 0, SEEK_END) = 0
lseek(21, 0, SEEK_END) = 11214848

More info: the statement is an INSERT with some huge subselects, running
every night on a PGSQL 8.3.6 on CentOS 4.7 x86_64. In 97% of all
occasions, the job terminates in about 1-2 minutes - however, sometimes
it just hangs. If terminated with kill <pid> and restarted, it always
terminates.

Clueless,

Peter