From @timbunce

Summary of my perl5 (revision 5.0 version 5 subversion 57) configuration​:
Platform​:
osname=aix, osvers=4.3.2.0, archname=aix-thread
uname='aix aasc24 3 4 00ffffff4c00 '
config_args='-desr -Dusethreads -Duseshrplib -Dcc=cc -Dccflags=-DSOCKS -DDEBUGGING -Doptimize=-O2 -Uinstallusrbinperl'
hint=recommended, useposix=true, d_sigaction=define
usethreads=define useperlio=undef d_sfio=undef
use64bits=undef usemultiplicity=undef
Compiler​:
cc='cc_r', optimize='-O2', gccversion=
cppflags='-DSOCKS -DDEBUGGING -D_ALL_SOURCE -D_ANSI_C_SOURCE -D_POSIX_SOURCE -qmaxmem=8192 -DNEED_PTHREAD_INIT -I/usr/local/include'
ccflags ='-DSOCKS -DDEBUGGING -D_ALL_SOURCE -D_ANSI_C_SOURCE -D_POSIX_SOURCE -qmaxmem=8192 -DNEED_PTHREAD_INIT -I/usr/local/include'
stdchar='unsigned char', d_stdstdio=define, usevfork=false
intsize=4, longsize=4, ptrsize=4, doublesize=8
d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=8
alignbytes=8, usemymalloc=n, prototype=define
Linker and Libraries​:
ld='ld', ldflags =' -L/usr/local/lib'
libpth=/usr/local/lib /lib /usr/lib /usr/ccs/lib
libs=-lnsl -lgdbm -ldbm -ldb -ldl -lld -lm -lpthreads -lc_r -lcrypt -lbsd -lPW -lsocks5
libc=, so=a, useshrplib=true, libperl=libperl.a
Dynamic Linking​:
dlsrc=dl_aix.xs, dlext=so, d_dlsymun=undef, ccdlflags='-bE​:perl.exp'
cccdlflags=' ', lddlflags='-bhalt​:4 -bM​:SRE -bI​:$(PERL_INC)/perl.exp -bE​:$(BASEEXT).exp -b noentry -lsocks5 -lpthreads -lc_r -L/usr/local/lib'

t/examp.............ok 104/196

t/examp.............ok 105/196Bad symbol for filehandle at blib/lib/DBD/ExampleP.pm line 321.

The offending line is​:

  closedir($sth->{dbd_datahandle});

The only other relevant lines are​:

  $sth->{dbd_datahandle} = "DBD​::ExampleP​::".++$DBD​::ExampleP​::gensym;
  opendir($sth->{dbd_datahandle}, $dir)

I can't see why the test should fail in this way. Smells to me like
a threaded-perl bug or AIX threaded libc bug.

Please confirm that there's no problem with a non-threaded perl.
Please also do a spot of debugging to see what value is in
$sth->{dbd_datahandle} (etc etc) in case that sheds any light on it.

I'd be grateful if someone could test DBI 1.11 with a threaded 5.005_57
on another platform (I don't have time right now).

Tim.