#855931 ruby-eventmachine: FTBFS: E: Build killed with signal TERM after 150 minutes of inactivity

Package:
src:ruby-eventmachine
Source:
ruby-eventmachine
Submitter:
Lucas Nussbaum
Date:
2025-08-17 17:48:09 UTC
Severity:
important
Tags:
#855931#5
Date:
2017-02-23 13:28:20 UTC
From:
To:
Hi,

During a rebuild of all packages in stretch (in a stretch chroot, not a
sid chroot), your package failed to build on amd64.

Relevant part (hopefully):
http://aws-logs.debian.net/2017/02/21/ruby-eventmachine_1.0.7-4_testing.log

A list of current common problems and possible solutions is available at
http://wiki.debian.org/qa.debian.org/FTBFS . You're welcome to contribute!

About the archive rebuild: The rebuild was done on EC2 VM instances from
Amazon Web Services, using a clean, minimal and up-to-date chroot. Every
failed build was retried once to eliminate random failures.

#855931#10
Date:
2017-03-04 07:19:22 UTC
From:
To:
I could not reproduce it on  a freshly created stretch chroot with sbuild.

sbuild -d stretch ruby-eventmachine_1.0.7-4.dsc

#855931#19
Date:
2017-03-06 12:12:12 UTC
From:
To:
Hi,

Well, I just gave it another try, and I could reproduce it...

Lucas

#855931#24
Date:
2017-03-08 08:46:50 UTC
From:
To:
ruby-eventmachine is one the packages whose tests will go crazy on EC2,
probably due to some weirdness in how exactly EC2 VM's are configured.
Most likely something kernel-related.

#855931#35
Date:
2022-12-31 14:03:13 UTC
From:
To:
retitle 855931 ruby-eventmachine: FTBFS due to hanging flaky tests
found 855931 1.3~pre20220315-df4ab006-3
retitle 998097 ruby-eventmachine: FTBFS on IPV6-only buildds
thanks

Hi,

Also, this package has had it's share of FTBFS bugs due to failing
tests, so I wonder if this should be fixed with wack-a-mole or if it's
better to disable the test suite. I tried to make sense of the different
failures to fix bug 998097, but there have been so many tries to fix or
disable several tests that I couldn't make sense of it in the time I
spent on it. I have the feeling that some of the timeout issues are due
to the test suite not properly cleaning up if one test fails and than
hanging later on.

For documentation purposes, at least I got this from archeology (all but
test_kb.rb pass on my system):
  disabled_tests= [
+  # has a tast that (sometimes?) seems to hang
    'tests/test_exc.rb',
+  # some tests requiring network access
    'tests/test_get_sock_opt.rb',
+  # several tests need internet (google.com)
    'tests/test_httpclient2.rb',
+  # several tests need internet (google.com)
    'tests/test_httpclient.rb',
+  # some tests requiring network access
    'tests/test_idle_connection.rb',
+  # mips sometimes too slow
    'tests/test_inactivity_timeout.rb',
+  # error: terminate called after throwing an instance of
'std::runtime_error' what(): unable to add new descriptor: Operation not
permitted
    'tests/test_kb.rb',
+  # one test needs internet (192.0.2.0)
    'tests/test_pending_connect_timeout.rb',
+  # several tests need internet
    'tests/test_resolver.rb',
+  # some tests requiring network access
    'tests/test_set_sock_opt.rb',
+  # one test needs internet (192.0.2.0)
    'tests/test_unbind_reason.rb',
  ]

https://buildd.debian.org/status/fetch.php?pkg=ruby-eventmachine&arch=amd64&ver=1.3%7Epre20220315-df4ab006-3&stamp=1665953644&raw=0

TestConnectionCount:
   test_idle_connection_count:				.: (0.000488)
   test_idle_connection_count_epoll:			.: (0.000377)
   test_idle_connection_count_kqueue:			.: (0.000320)
   test_num_close_scheduled:				.: (0.000794)
   test_with_some_connections:				.: (0.001581)
TestConnectionWrite:
   test_with_naughty_callback:				.: (0.000710)
TestDefer:
   test_defers:
E: Build killed with signal TERM after 150 minutes of inactivity


https://buildd.debian.org/status/fetch.php?pkg=ruby-eventmachine&arch=i386&ver=1.3%7Epre20201020-b50c135-4%2Bb1&stamp=1635553347&raw=0

TestConnectionCount:
   test_idle_connection_count:				.: (0.000292)
   test_idle_connection_count_epoll:			.: (0.000344)
   test_idle_connection_count_kqueue:			.: (0.000234)
   test_num_close_scheduled:				.: (0.000539)
   test_with_some_connections:				.: (0.001199)
TestConnectionWrite:
   test_with_naughty_callback:				.: (0.000396)
TestDefer:
   test_defers:
E: Build killed with signal TERM after 150 minutes of inactivity


https://buildd.debian.org/status/fetch.php?pkg=ruby-eventmachine&arch=s390x&ver=1.2.7-1%7Eexp1&stamp=1546537666&raw=0

TestSockOpt:
   test_get_sock_opt:					.: (0.000343)
   test_set_sock_opt:					.: (0.000254)
TestSomeExceptions:
   test_a:						.: (0.000130)
   test_b:						.: (0.000072)
   test_exception_on_unbind:
E: Build killed with signal TERM after 150 minutes of inactivity


Non release architectures:
alpha seems to suffer a lot from this, one example

https://buildd.debian.org/status/fetch.php?pkg=ruby-eventmachine&arch=alpha&ver=1.0.7-4.2&stamp=1544037015&raw=0


TestConnectionCount:
   test_idle_connection_count:				.: (0.000977)
   test_num_close_scheduled:				.: (15.075295)
   test_with_some_connections:
E: Build killed with signal TERM after 150 minutes of inactivity


https://buildd.debian.org/status/fetch.php?pkg=ruby-eventmachine&arch=hppa&ver=1.3%7Epre20220315-df4ab006-3&stamp=1666020944&raw=0

   test_attach_pipe:					E
===============================================================================
Error: test_attach_pipe(TestAttach): Errno::E111: Unknown error 111 -
connect(2) for "127.0.0.1" port 486688568
/<<PKGBUILDDIR>>/tests/em_test_helper.rb:51:in `initialize'
/<<PKGBUILDDIR>>/tests/em_test_helper.rb:51:in `new'
/<<PKGBUILDDIR>>/tests/em_test_helper.rb:51:in `port_in_use?'
/<<PKGBUILDDIR>>/tests/em_test_helper.rb:62:in `next_port'
/<<PKGBUILDDIR>>/tests/test_attach.rb:32:in `setup'
===============================================================================/<<PKGBUILDDIR>>/tests/test_attach.rb:38:
warning: global variable `$r' not initialized
/<<PKGBUILDDIR>>/tests/test_attach.rb:38: warning: global variable `$w'
not initialized

: (19373.112091)
   test_attach_server:
E: Build killed with signal TERM after 600 minutes of inactivity


https://buildd.debian.org/status/fetch.php?pkg=ruby-eventmachine&arch=sparc64&ver=1.3%7Epre2019019-g8f7e128-1%7Eexp2&stamp=1548244484&raw=0

===============================================================================
Error: test_any_with_external_client_tlsv1_2(TestSSLProtocols):
OpenSSL::SSL::SSLError: SSL_connect SYSCALL returned=5 errno=0
state=SSLv3/TLS write finished
/<<PKGBUILDDIR>>/tests/test_ssl_protocols.rb:128:in `connect'
/<<PKGBUILDDIR>>/tests/test_ssl_protocols.rb:128:in `block (2 levels) in
external_client'
/<<PKGBUILDDIR>>/debian/ruby-eventmachine/usr/lib/ruby/vendor_ruby/eventmachine.rb:1078:in
`block in spawn_threadpool'
===============================================================================
E
===============================================================================
Error: test_any_with_external_client_tlsv1_2(TestSSLProtocols):
OpenSSL::SSL::SSLError: SSL_connect SYSCALL returned=5 errno=0
state=SSLv3/TLS write finished
/<<PKGBUILDDIR>>/tests/test_ssl_protocols.rb:128:in `connect'
/<<PKGBUILDDIR>>/tests/test_ssl_protocols.rb:128:in `block (2 levels) in
external_client'
/<<PKGBUILDDIR>>/debian/ruby-eventmachine/usr/lib/ruby/vendor_ruby/eventmachine.rb:1078:in
`block in spawn_threadpool'
===============================================================================
: (0.112230)
   test_any_with_external_client_tlsv1_3:
E: Build killed with signal TERM after 600 minutes of inactivity


Paul