DBD-Oracle-1.91_1: 91-segv-fork.t seems to hang using Strawberry Perl

[twata1]

I have confiremd with Strawberry Perl 5.32.1 and 5.40.0.

The output of the test using Strawberry Perl 5.40.0 was as follows.

t/00-compile.t ............ ok
#
# Versions for all modules listed in MYMETA.json (including optional ones):
#
# === Configure Requires ===
#
#     Module               Want     Have
#     ------------------- ----- --------
#     DBI                 1.623    1.647
#     ExtUtils::MakeMaker   any  7.73_01
#     perl                5.006 5.040000
#
# === Configure Suggests ===
#
#     Module      Want Have
#     -------- ------- ----
#     JSON::PP 2.27300 4.16
#
# === Build Requires ===
#
#     Module               Want     Have
#     ------------------- ----- --------
#     Config                any 5.040000
#     DBI                 1.623    1.647
#     ExtUtils::MakeMaker   any  7.73_01
#
# === Test Requires ===
#
#     Module                  Want     Have
#     ------------------- -------- --------
#     B                        any     1.89
#     Carp                     any     1.54
#     DBI                    1.623    1.647
#     Data::Dumper             any    2.189
#     Devel::Peek              any     1.34
#     Encode                   any     3.21
#     ExtUtils::MakeMaker      any  7.73_01
#     File::Spec               any     3.90
#     FindBin                  any     1.54
#     IO::Handle               any     1.55
#     IPC::Open3               any     1.22
#     Math::BigInt             any 2.005002
#     Symbol                   any     1.09
#     Test::More               any 1.302210
#     Test::NoWarnings         any     1.06
#     Thread::Semaphore        any     2.13
#     Time::HiRes              any   1.9777
#     lib                      any     0.65
#     perl                5.030003 5.040000
#     strict                   any     1.13
#     threads                  any     2.40
#     threads::shared         1.51     1.69
#     utf8                     any     1.25
#     vars                     any     1.05
#     warnings                 any     1.70
#
# === Test Recommends ===
#
#     Module         Want     Have
#     ---------- -------- --------
#     CPAN::Meta 2.120900 2.150010
#
# === Runtime Requires ===
#
#     Module               Want     Have
#     ---------------- -------- --------
#     DBI                 1.623    1.647
#     Data::Dumper          any    2.189
#     DynaLoader            any     1.56
#     Exporter              any     5.78
#     Test::NoWarnings        1     1.06
#     Test::Simple          0.9 1.302210
#     perl             5.030003 5.040000
#     strict                any     1.13
#     warnings              any     1.70
#
t/00-report-prereqs.t ..... ok
t/00dbdoracletestlib.t .... ok
t/01base.t ................ ok
# OCI client library version: 18.4.0.0
# Database version: Oracle Database 18c Express Edition Release 18.0.0.0.0 - Production Oracle Database 18c Express Edition Releas
e 18.0.0.0.0 - Production
# Version 18.4.0.0.0 Oracle Database 18c Express Edition Release 18.0.0.0.0 - Production 0
t/02versions.t ............ ok
t/05base.t ................ ok
t/10general.t ............. ok
t/12impdata.t ............. ok
t/14threads.t ............. ok
t/15nls.t ................. ok
t/15threads.t ............. ok
t/16cached.t .............. ok
t/16drcp.t ................ ok
t/20select.t .............. ok
t/21nchar.t ............... ok
t/22cset.t ................ ok
t/22nchar_al32utf8.t ...... ok
t/22nchar_utf8.t .......... ok
t/23wide_db.t ............. ok
t/23wide_db_8bit.t ........ ok
t/23wide_db_al32utf8.t .... ok
t/24implicit_utf8.t ....... ok
t/25plsql.t ............... ok
t/26exe_array.t ........... ok
t/28array_bind.t .......... ok
t/30long.t ................ ok
t/31lob.t ................. ok
t/31lob_extended.t ........ ok
t/32xmltype.t ............. ok
t/34pres_lobs.t ........... ok
t/36lob_leak.t ............ ok
t/38taf.t ................. ok
#
# Oracle SERVER version: 18.0.0.0.0
# Oracle CLIENT version: 18.4.0.0
t/39attr.t ................ ok
t/40ph_type.t ............. ok
Can't determine open_cursors from v$parameter, so using default
t/50cursor.t .............. ok
t/51scroll.t .............. ok
t/55nested.t .............. ok
t/56embbeded.t ............ ok
t/58object.t .............. ok
t/60reauth.t .............. ok
t/70meta.t ................ ok
t/80ora_charset.t ......... ok
Can't locate object method "new" via package "Thread::Queue" (perhaps you forgot to load "Thread::Queue"?) at t/90-segv-threads.t
line 303.
# Tests were run but no plan was declared and done_testing() was not seen.
# Looks like your test exited with 255 just after 2.
t/90-segv-threads.t .......
Dubious, test returned 255 (wstat 65280, 0xff00)
All 2 subtests passed
No such signal: SIGUSR1 at t/91-segv-fork.t line 255.
No such signal: SIGUSR2 at t/91-segv-fork.t line 256.
No such signal: SIGUSR1 at t/91-segv-fork.t line 255.
No such signal: SIGUSR2 at t/91-segv-fork.t line 256.
No such signal: SIGUSR1 at t/91-segv-fork.t line 255.
No such signal: SIGUSR2 at t/91-segv-fork.t line 256.
No such signal: SIGUSR1 at t/91-segv-fork.t line 255.
No such signal: SIGUSR2 at t/91-segv-fork.t line 256.
No such signal: SIGUSR1 at t/91-segv-fork.t line 214.
Unrecognized signal name "USR1" at t/91-segv-fork.t line 228.
        (in cleanup) Unrecognized signal name "USR2" at t/91-segv-fork.t line 153.
Unrecognized signal name "USR2" at t/91-segv-fork.t line 153.
END failed--call queue aborted.
# Tests were run but no plan was declared and done_testing() was not seen.
# Looks like your test exited with 25 just after 12.
(hangs)

Thank you,

[ancientwizard]

Well shucks! Sorry it's the UNIX/Linux person in me. Apparently, Windows doesn't have an equivalent USR1 & USR2 signals. I'm using them as a simple means for the parent and children tap on each other's shoulders. The AI says you might be able to safely use "BREAK" and "HUP" as replacements for the message passing. If the test detects its Windows, then alters the behavior. I don't have windows (with Perl and Oracle) so I can't test even if I made the change.

If you can't come up with the change. I'll try later (make a guess) and have you test it. A tactical fix would be to skip the test on Windows.

Please note the test is incomplete. I've mentioned another bug and it is related to forking and interrupt handling. I'll definitely be making change to this test to reproduce the issue outside of my in-office environment.

[djzort]

I think for now just skip these tests on windows and have a comment asking someone to help with an appropriate alternative.

[djzort]

Github actions does support windows, perhaps someone can send in a pull request to get windows testing going?

[twata1]

My thoughts on t/90-segv-threads.t and test t/91-segv-fork.t are as follows.
I believe that there are other uses for this module that do not use threads or forking.
Would this idea be acceptable?

• If environment variable AUTHOR_TESTING is set, make them testable.
• Remove use 5.030003; from those tests.
• For t/91-segv-fork.t, skip for windows.

Thank you,

[ancientwizard]

My thoughts on t/90-segv-threads.t and test t/91-segv-fork.t are as follows. I believe that there are other uses for this module that do not use threads or forking. Would this idea be acceptable?

• If environment variable AUTHOR_TESTING is set, make them testable.
• Remove use 5.030003; from those tests.
• For t/91-segv-fork.t, skip for windows.

Thank you,

I've already removed the perl version stuff. Some how it didn't make it into last pull. It will get there during the next. While others may not use threads knowing that stuff isn't broken is a good thing. We my yet still add the author testing but as I have proven just because the user isn't using threads oracle creates them.

I also made a patch for the USR1 and USR2 siginal so it should work on windows with the update

[djzort]

Check out https://metacpan.org/release/ZARQUON/DBD-Oracle-1.91_2

[twata1]

Thank you for handling this issue.

Still t/91-segv-fork.t appears to hang testing DBD-Oracle-1.91_2 on windows.
The output of the test using Strawberry Perl 5.40.0 was as follows.

t/00-compile.t ............ ok
#
# Versions for all modules listed in MYMETA.json (including optional ones):
#
# === Configure Requires ===
#
#     Module               Want     Have
#     ------------------- ----- --------
#     DBI                 1.623    1.647
#     ExtUtils::MakeMaker   any  7.73_01
#     perl                5.006 5.040000
#
# === Configure Suggests ===
#
#     Module      Want Have
#     -------- ------- ----
#     JSON::PP 2.27300 4.16
#
# === Build Requires ===
#
#     Module               Want     Have
#     ------------------- ----- --------
#     Config                any 5.040000
#     DBI                 1.623    1.647
#     ExtUtils::MakeMaker   any  7.73_01
#
# === Test Requires ===
#
#     Module               Want     Have
#     ------------------- ----- --------
#     B                     any     1.89
#     Carp                  any     1.54
#     DBI                 1.623    1.647
#     Data::Dumper          any    2.189
#     Devel::Peek           any     1.34
#     Encode                any     3.21
#     ExtUtils::MakeMaker   any  7.73_01
#     File::Spec            any     3.90
#     FindBin               any     1.54
#     IO::Handle            any     1.55
#     IPC::Open3            any     1.22
#     Math::BigInt          any 2.005002
#     POSIX                 any     2.20
#     Symbol                any     1.09
#     Test::More            any 1.302210
#     Test::NoWarnings      any     1.06
#     Thread::Queue         any     3.14
#     Thread::Semaphore     any     2.13
#     Time::HiRes           any   1.9777
#     lib                   any     0.65
#     perl                5.008 5.040000
#     strict                any     1.13
#     threads               any     2.40
#     threads::shared      1.51     1.69
#     utf8                  any     1.25
#     vars                  any     1.05
#     warnings              any     1.70
#
# === Test Recommends ===
#
#     Module         Want     Have
#     ---------- -------- --------
#     CPAN::Meta 2.120900 2.150010
#
# === Runtime Requires ===
#
#     Module            Want     Have
#     ---------------- ----- --------
#     DBI              1.623    1.647
#     Data::Dumper       any    2.189
#     DynaLoader         any     1.56
#     Exporter           any     5.78
#     Test::NoWarnings     1     1.06
#     Test::Simple       0.9 1.302210
#     perl             5.008 5.040000
#     strict             any     1.13
#     warnings           any     1.70
#
t/00-report-prereqs.t ..... ok
t/00dbdoracletestlib.t .... ok
t/01base.t ................ ok
# OCI client library version: 18.4.0.0
# Database version: Oracle Database 18c Express Edition Release 18.0.0.0.0 - Production Oracle Database 18c Express Edition Releas
e 18.0.0.0.0 - Production
# Version 18.4.0.0.0 Oracle Database 18c Express Edition Release 18.0.0.0.0 - Production 0
t/02versions.t ............ ok
t/05base.t ................ ok
t/10general.t ............. ok
t/12impdata.t ............. ok
t/14threads.t ............. ok
t/15nls.t ................. ok
t/15threads.t ............. ok
t/16cached.t .............. ok
t/16drcp.t ................ ok
t/20select.t .............. ok
t/21nchar.t ............... ok
t/22cset.t ................ ok
t/22nchar_al32utf8.t ...... ok
t/22nchar_utf8.t .......... ok
t/23wide_db.t ............. ok
t/23wide_db_8bit.t ........ ok
t/23wide_db_al32utf8.t .... ok
t/24implicit_utf8.t ....... ok
t/25plsql.t ............... ok
t/26exe_array.t ........... ok
t/28array_bind.t .......... ok
t/30long.t ................ ok
t/31lob.t ................. ok
t/31lob_extended.t ........ ok
t/32xmltype.t ............. ok
t/34pres_lobs.t ........... ok
t/36lob_leak.t ............ ok
t/38taf.t ................. ok
#
# Oracle SERVER version: 18.0.0.0.0
# Oracle CLIENT version: 18.4.0.0
t/39attr.t ................ ok
t/40ph_type.t ............. ok
Can't determine open_cursors from v$parameter, so using default
t/50cursor.t .............. ok
t/51scroll.t .............. ok
t/55nested.t .............. ok
t/56embbeded.t ............ ok
t/58object.t .............. ok
t/60reauth.t .............. ok
t/70meta.t ................ ok
t/80ora_charset.t ......... ok
t/90-segv-threads.t ....... ok
No such signal: SIGUSR1 at t/91-segv-fork.t line 254.
No such signal: SIGUSR2 at t/91-segv-fork.t line 255.
No such signal: SIGUSR1 at t/91-segv-fork.t line 254.
No such signal: SIGUSR2 at t/91-segv-fork.t line 255.
No such signal: SIGUSR1 at t/91-segv-fork.t line 254.
No such signal: SIGUSR2 at t/91-segv-fork.t line 255.
No such signal: SIGUSR1 at t/91-segv-fork.t line 254.
No such signal: SIGUSR2 at t/91-segv-fork.t line 255.
No such signal: SIGUSR1 at t/91-segv-fork.t line 213.
Unrecognized signal name "USR1" at t/91-segv-fork.t line 227.
        (in cleanup) Unrecognized signal name "USR2" at t/91-segv-fork.t line 152.
Unrecognized signal name "USR2" at t/91-segv-fork.t line 152.
END failed--call queue aborted.
# Tests were run but no plan was declared and done_testing() was not seen.
# Looks like your test exited with 25 just after 12.
(hangs)

bonus:
Results-92-segv-fork.t-on-Strawberry-Perl-5.40.0.log.txt

Thank you,

[ancientwizard]

Thank you for handling this issue.

Still t/91-segv-fork.t appears to hang testing DBD-Oracle-1.91_2 on windows. The output of the test using Strawberry Perl 5.40.0 was as follows.

bonus: Results-92-segv-fork.t-on-Strawberry-Perl-5.40.0.log.txt

Thank you,

This output suggests you don't have the updated test. I have a feeling that some changes failed to make it to master during a pull. I have run access a couple of lost changes in the last few days. Staring at git diffs isn't fun. When I next have a look see where they get off too.