Skip Menu |

This queue is for tickets about the Net-LDAP-Server-Test CPAN distribution.

Report information
The Basics
Id: 83905
Status: resolved
Worked: 20 min
Priority: 0/
Queue: Net-LDAP-Server-Test

People
Owner: karman [...] cpan.org
Requestors: chisel [...] chizography.net
Cc:
AdminCc:

Bug Information
Severity: Important
Broken in: 0.17
Fixed in: (no value)



Subject: Test suite freezes: Unable to listen on port 10636: Address already in use
This is blocking the installation of the module, which is blocking the testing of a fix for another module [https://rt.cpan.org/Ticket/Display.html?id=83900] This has happened every time I've tried to install the module: Show quoted text
---- cut here ---- Net-LDAP-Server-Test-0.17/lib/Net/LDAP/Server/Test.pm CPAN: File::Temp loaded ok (v0.22) CPAN: Parse::CPAN::Meta loaded ok (v1.4404) CPAN: CPAN::Meta loaded ok (v2.120921) CPAN: Module::CoreList loaded ok (v2.49_02) CPAN.pm: Going to build K/KA/KARMAN/Net-LDAP-Server-Test-0.17.tar.gz Checking if your kit is complete... Looks good Writing Makefile for Net::LDAP::Server::Test Writing MYMETA.yml and MYMETA.json cp lib/Net/LDAP/Server/Test.pm blib/lib/Net/LDAP/Server/Test.pm Manifying blib/man3/Net::LDAP::Server::Test.3 KARMAN/Net-LDAP-Server-Test-0.17.tar.gz /usr/bin/make -j 3 -j 3 -- OK Running make test PERL_DL_NONLAZY=1 /opt/xt/xt-perl/bin/perl "-MExtUtils::Command::MM" "-e" "test_harness(0, 'blib/lib', 'blib/arch')" t/*.t t/00-load.t ........... 1/1 # Testing Net::LDAP::Server::Test 0.17, Perl 5.014002, /opt/xt/xt- perl/bin/perl t/00-load.t ........... ok t/01-ldap.t ........... 1/12 # stop() server waitpid(7244, 0) took too long Unable to listen on port 10636: Address already in use at /tmp/.cpan/build/Net-LDAP- Server-Test-0.17-5k9t6t/blib/lib/Net/LDAP/Server/Test.pm line 964.
---- cut here ----
This has been a consistent but un-repeatable issue since the initial release of this module and there are several workarounds but no fix. What OS/platform are you using? Are you using any iptables-like firewall? Does fork() work otherwise normally for you in this environment?
On Tue Mar 12 18:00:58 2013, KARMAN wrote: Show quoted text
> This has been a consistent but un-repeatable issue since the initial > release of this module and > there are several workarounds but no fix. > > What OS/platform are you using? Are you using any iptables-like > firewall? Does fork() work > otherwise normally for you in this environment?
I'm using: ➔ cat /etc/redhat-release CentOS release 5.8 (Final) ➔ uname -a Linux fulcrum-chz.dave.net-a-porter.com 2.6.18-308.24.1.el5 #1 SMP Tue Dec 4 17:43:34 EST 2012 x86_64 x86_64 x86_64 GNU/Linux I believe our jenkins build server is Centos5 too. fork() has never shown any signs of misbehaving. We rely on it moderately heavily for daemons for our apps. There are some iptables rules on the box, but they're all listed as ACCEPT no DROP or REJECT rules in the output from 'iptables --list'
Ok, I have the same OS in front of me, with no problems. What Perl version?
On Wed Mar 13 09:27:10 2013, KARMAN wrote: Show quoted text
> Ok, I have the same OS in front of me, with no problems. What Perl version?
Nevermind, I see Perl 5.014002 in the test output.
I have added a diagnostic warning and upped the kill signal in https://github.com/karpet/net-ldap-server- test/commit/e3f8e52b93e5a0341e0d12ec86ce463a93e90ed8 please check that out. Even if it does not fix the issue I would like to see the output.
Here's what I get checking out the repo from scratch: [5330][c.wright@fulcrum-chz:development]➔ git clone git://github.com/karpet/net-ldap- server-test.git Cloning into net-ldap-server-test... remote: Counting objects: 327, done. remote: Compressing objects: 100% (113/113), done. remote: Total 327 (delta 109), reused 292 (delta 74) Receiving objects: 100% (327/327), 46.53 KiB, done. Resolving deltas: 100% (109/109), done. [5331][c.wright@fulcrum-chz:development]➔ cd net-ldap-server-test/ [5332][c.wright@fulcrum-chz:net-ldap-server-test][master]➔ perl Makefile.PL Checking if your kit is complete... Looks good Writing Makefile for Net::LDAP::Server::Test Writing MYMETA.yml and MYMETA.json [5333][c.wright@fulcrum-chz:net-ldap-server-test][master⚡]➔ make test cp lib/Net/LDAP/Server/Test.pm blib/lib/Net/LDAP/Server/Test.pm PERL_DL_NONLAZY=1 /opt/xt/xt-perl/bin/perl "-MExtUtils::Command::MM" "-e" "test_harness(0, 'blib/lib', 'blib/arch')" t/*.t t/00-load.t ........... 1/1 # Testing Net::LDAP::Server::Test 0.17, Perl 5.016003, /opt/xt/xt- perl/bin/perl t/00-load.t ........... ok t/01-ldap.t ........... 1/12 # stop() server waitpid(4894, 0) took too long kill(9,4894) returned 1 Unable to listen on port 10636: Address already in use at /home/c.wright/development/net- ldap-server-test/blib/lib/Net/LDAP/Server/Test.pm line 964.
On Wed Mar 13 13:39:45 2013, CHISEL wrote: Show quoted text
> Here's what I get checking out the repo from scratch: >
thanks for the quick test. I just pushed some more diagnostics to github. Please git pull and try again.
This one passed (I just did a git pull); there's a slight pause - presumably because of the new 'tried loop'. I've run it a few times with no failure. [5337][c.wright@fulcrum-chz:net-ldap-server-test][master⚡]➔ make test cp lib/Net/LDAP/Server/Test.pm blib/lib/Net/LDAP/Server/Test.pm PERL_DL_NONLAZY=1 /opt/xt/xt-perl/bin/perl "-MExtUtils::Command::MM" "-e" "test_harness(0, 'blib/lib', 'blib/arch')" t/*.t t/00-load.t ........... 1/1 # Testing Net::LDAP::Server::Test 0.17, Perl 5.016003, /opt/xt/xt- perl/bin/perl t/00-load.t ........... ok t/01-ldap.t ........... 4/12 # stop() server waitpid(20194, 0) took too long kill(9,20194) returned 1 Waiting for port to close... t/01-ldap.t ........... ok t/02-ad.t ............. ok t/03-socket.t ......... ok t/04attributes.t ...... ok t/05-scope.t .......... ok t/06-no-such-entry.t .. ok t/07-error-codes.t .... # LDAP_SUCCESS==0 # LDAP_NO_SUCH_OBJECT==32 # LDAP_CONTROL_PAGED==1.2.840.113556.1.4.319 # LDAP_OPERATIONS_ERROR==1 # LDAP_UNWILLING_TO_PERFORM==53 # LDAP_ALREADY_EXISTS==68 # LDAP_TYPE_OR_VALUE_EXISTS==20 # LDAP_NO_SUCH_ATTRIBUTE==16 t/07-error-codes.t .... ok t/boilerplate.t ....... ok t/pod-coverage.t ...... ok t/pod.t ............... ok All tests successful. Files=11, Tests=84, 5 wallclock secs ( 0.06 usr 0.02 sys + 1.18 cusr 0.27 csys = 1.53 CPU) Result: PASS [5338][c.wright@fulcrum-chz:net-ldap-server-test][master⚡]➔ git tree -n 1 * 8f52aca [18 hours ago] <Peter Karman> (HEAD, origin/master, origin/HEAD, master) RT #83905: add more tests for whether port closes in stop() [5339][c.wright@fulcrum-chz:net-ldap-server-test][master⚡]➔
On Thu Mar 14 07:46:28 2013, CHISEL wrote: Show quoted text
> This one passed (I just did a git pull); there's a slight pause - > presumably because of the new > 'tried loop'. > I've run it a few times with no failure. >
Thanks. 0.18 has been pushed to cpan.
That's great news. Thank you for a quick resolution.