2:03 a.m.
https://bugzilla.redhat.com/show_bug.cgi?id=1100158
Bug ID: 1100158
Summary: perl-Server-Starter-0.17 FTBFS sometimes because of
races in tests
Product: Fedora
Version: rawhide
Component: perl-Server-Starter
Assignee: rc040203(a)freenet.de
Reporter: ppisar(a)redhat.com
QA Contact: extras-qa(a)fedoraproject.org
CC: perl-devel(a)lists.fedoraproject.org,
rc040203(a)freenet.de
External Bug ID: CPAN 73711
perl-Server-Starter-0.17 does not pass tests sometimes:
[...]
# matches '(?^m:^FOO=foo-value2$)'
received TERM, sending TERM to all workers:30528,30535
worker 30528 died, status:0
worker 30535 died, status:0
exiting
[Test::TCP] Child process does not block(PID: 30526, PPID: 30525) at
/builddir/build/BUILD/Server-Starter-0.17/inc/Test/TCP.pm line 89.
# Looks like you failed 1 test of 4.
t/07-envdir.t ........
Dubious, test returned 1 (wstat 256, 0x100)
Failed 1/4 subtests
Test Summary Report
-------------------
t/06-autorestart.t (Wstat: 512 Tests: 28 Failed: 2)
Failed tests: 6-7
Non-zero exit status: 2
t/07-envdir.t (Wstat: 256 Tests: 4 Failed: 1)
Failed test: 3
Non-zero exit status: 1
Files=8, Tests=103, 137 wallclock secs ( 0.06 usr 0.00 sys + 1.50 cusr 0.25
csys = 1.81 CPU)
Result: FAIL
This happens especially if the build host is busy and delays hardcoded in the
tests are not enough. E.g. t/07-envdir.t:
my $restart = sub {
sleep 1;
kill "HUP", $server_pid;
sleep 2;
};
If one removes these sleep()s, then the tests fails all the time. This is known
to the upstream <https://rt.cpan.org/Public/Bug/Display.html?id=73711>;.
--
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug
https://bugzilla.redhat.com/token.cgi?t=WfPEMPxPbb&a=cc_unsubscribe
10:38 a.m.
New subject: [Bug 1100158] perl-Server-Starter-0.17 FTBFS sometimes because of races in tests
https://bugzilla.redhat.com/show_bug.cgi?id=1100158
Petr Pisar <ppisar(a)redhat.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |ASSIGNED
Assignee|rc040203(a)freenet.de |ppisar(a)redhat.com
--
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug
https://bugzilla.redhat.com/token.cgi?t=oQPWS4tzBo&a=cc_unsubscribe
5:30 a.m.
New subject: [Bug 1100158] perl-Server-Starter-0.17 FTBFS sometimes because of races in tests
https://bugzilla.redhat.com/show_bug.cgi?id=1100158
Petr Pisar <ppisar(a)redhat.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |CLOSED
Fixed In Version| |perl-Server-Starter-0.17-3.
| |fc21
Resolution|--- |RAWHIDE
Last Closed| |2014-06-17 06:30:17
--- Comment #1 from Petr Pisar <ppisar(a)redhat.com> ---
I fixed the t/07-envdir.t test.
--
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug
https://bugzilla.redhat.com/token.cgi?t=60ytWky9wC&a=cc_unsubscribe
8:52 a.m.
New subject: [Bug 1100158] perl-Server-Starter-0.17 FTBFS sometimes because of races in tests
https://bugzilla.redhat.com/show_bug.cgi?id=1100158
Petr Pisar <ppisar(a)redhat.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|CLOSED |ASSIGNED
Resolution|RAWHIDE |---
Keywords| |Reopened
--- Comment #2 from Petr Pisar <ppisar(a)redhat.com> ---
There is another race in t/06-autorestart.t:
# Failed test 'status after auto-restart'
# at t/06-autorestart.t line 55.
# '1:26986
# 2:26992
# '
# doesn't match '(?^s:^2:\d+\n$)'
new worker is now running, sending USR1 to old workers:26992
sleep 1 secs
killing old workers
received TERM, sending TERM to all workers:26992,26997
worker 26992 died, status:15
worker 26997 died, status:0
exiting
[Test::TCP] Child process does not block(PID: 26985, PPID: 26967) at
/builddir/build/BUILD/Server-Starter-0.17/inc/Test/TCP.pm line 89.
# Looks like you failed 1 test of 28.
t/06-autorestart.t ...
Dubious, test returned 1 (wstat 256, 0x100)
Failed
--
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug
https://bugzilla.redhat.com/token.cgi?t=WdkjPQuJPC&a=cc_unsubscribe
8:13 a.m.
New subject: [Bug 1100158] perl-Server-Starter-0.17 FTBFS sometimes because of races in tests
https://bugzilla.redhat.com/show_bug.cgi?id=1100158
--- Comment #3 from Petr Pisar <ppisar(a)redhat.com> ---
The t/06-autorestart.t is full of sleeps racing with 4s auto_restart_interval.
--
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug
https://bugzilla.redhat.com/token.cgi?t=fVcNgyq8sK&a=cc_unsubscribe
8:55 a.m.
New subject: [Bug 1100158] perl-Server-Starter-0.17 FTBFS sometimes because of races in tests
https://bugzilla.redhat.com/show_bug.cgi?id=1100158
Petr Pisar <ppisar(a)redhat.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |CLOSED
Fixed In Version|perl-Server-Starter-0.17-3. |perl-Server-Starter-0.17-4.
|fc21 |fc22
Resolution|--- |RAWHIDE
Last Closed|2014-06-17 06:30:17 |2014-07-10 09:55:34
--- Comment #4 from Petr Pisar <ppisar(a)redhat.com> ---
I fixed the 5/06-autorestart.t test.
--
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug
https://bugzilla.redhat.com/token.cgi?t=OR2LDz5J2i&a=cc_unsubscribe
5:44 a.m.
New subject: [Bug 1100158] perl-Server-Starter-0.17 FTBFS sometimes because of races in tests
https://bugzilla.redhat.com/show_bug.cgi?id=1100158
Petr Pisar <ppisar(a)redhat.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|CLOSED |ASSIGNED
Resolution|RAWHIDE |---
--- Comment #5 from Petr Pisar <ppisar(a)redhat.com> ---
The is another race in t/01-starter.t:
# Failed test 'status after restart'
# at t/01-starter.t line 53.
# '1:22344
# 2:22354
# '
# doesn't match '(?^s:^2:\d+\n$)'
received TERM, sending TERM to all workers:22354
worker 22354 died, status:0
exiting
[Test::TCP] Child process does not block(PID: 22343, PPID: 22342) at
/builddir/build/BUILD/Server-Starter-0.17/inc/Test/TCP.pm line 89.
start_server (pid:22356) starting now...
starting new worker 22357
received HUP (num_old_workers=0)
spawning a new worker (num_old_workers=0)
starting new worker 22358
new worker is now running, sending USR1 to old workers:22357
sleep 0 secs
killing old workers
# Failed test 'status during restart'
# at t/01-starter.t line 51.
# '1:22357
# '
# doesn't match '(?^s:^1:\d+\n2:\d+$)'
# Failed test 'status after restart'
# at t/01-starter.t line 53.
# '1:22357
# 2:22358
# '
# doesn't match '(?^s:^2:\d+\n$)'
received TERM, sending TERM to all workers:22357,22358
worker 22357 died, status:15
worker 22358 died, status:0
exiting
[Test::TCP] Child process does not block(PID: 22356, PPID: 22342) at
/builddir/build/BUILD/Server-Starter-0.17/inc/Test/TCP.pm line 89.
# Looks like you failed 3 tests of 28.
t/01-starter.t .......
Dubious, test returned 3 (wstat 768, 0x300)
Failed
--
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug
https://bugzilla.redhat.com/token.cgi?t=iUwg1cuOJc&a=cc_unsubscribe
6:15 a.m.
New subject: [Bug 1100158] perl-Server-Starter-0.17 FTBFS sometimes because of races in tests
https://bugzilla.redhat.com/show_bug.cgi?id=1100158
Petr Pisar <ppisar(a)redhat.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |CLOSED
Fixed In Version|perl-Server-Starter-0.17-4. |perl-Server-Starter-0.17-5.
|fc22 |fc22
Resolution|--- |RAWHIDE
Last Closed|2014-07-10 09:55:34 |2014-08-08 07:15:40
--- Comment #6 from Petr Pisar <ppisar(a)redhat.com> ---
perl-Server-Starter-0.17-5.fc22 fixes the t/01-starter.t. I hope this is
everything.
--
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug
https://bugzilla.redhat.com/token.cgi?t=vZoAbdebkL&a=cc_unsubscribe
6:55 a.m.
New subject: [Bug 1100158] perl-Server-Starter-0.17 FTBFS sometimes because of races in tests
https://bugzilla.redhat.com/show_bug.cgi?id=1100158
Petr Pisar <ppisar(a)redhat.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|CLOSED |ASSIGNED
Resolution|RAWHIDE |---
--- Comment #7 from Petr Pisar <ppisar(a)redhat.com> ---
There is another race in t/05-killolddelay.t:
# Failed test 'status after restart'
# at t/05-killolddelay.t line 58.
# '1:19612
# 2:19621
# '
# doesn't match '(?^s:^2:\d+\n$)'
received TERM, sending TERM to all workers:19621
worker 19621 died, status:0
exiting
[Test::TCP] Child process does not block(PID: 19611, PPID: 19610) at
/builddir/build/BUILD/Server-Starter-0.17/inc/Test/TCP.pm line 89.
start_server (pid:20006) starting now...
starting new worker 20007
received HUP (num_old_workers=0)
spawning a new worker (num_old_workers=0)
starting new worker 20010
new worker is now running, sending USR1 to old workers:20007
sleep 3 secs
killing old workers
old worker 20007 died, status:0
received TERM, sending TERM to all workers:20010
worker 20010 died, status:0
exiting
[Test::TCP] Child process does not block(PID: 20006, PPID: 19610) at
/builddir/build/BUILD/Server-Starter-0.17/inc/Test/TCP.pm line 89.
# Looks like you failed 1 test of 28.
t/05-killolddelay.t ..
Dubious, test returned 1 (wstat 256, 0x100)
Failed
--
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug
https://bugzilla.redhat.com/token.cgi?t=sRMJCPMOBJ&a=cc_unsubscribe
7:32 a.m.
New subject: [Bug 1100158] perl-Server-Starter-0.17 FTBFS sometimes because of races in tests
https://bugzilla.redhat.com/show_bug.cgi?id=1100158
Petr Pisar <ppisar(a)redhat.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |CLOSED
Fixed In Version|perl-Server-Starter-0.17-5. |perl-Server-Starter-0.17-6.
|fc22 |fc22
Resolution|--- |RAWHIDE
Last Closed|2014-08-08 07:15:40 |2014-08-21 08:32:08
--
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug
https://bugzilla.redhat.com/token.cgi?t=yVyhgkY9Bm&a=cc_unsubscribe
10:52 a.m.
New subject: [Bug 1100158] perl-Server-Starter-0.17 FTBFS sometimes because of races in tests
https://bugzilla.redhat.com/show_bug.cgi?id=1100158
Petr Pisar <ppisar(a)redhat.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|CLOSED |ASSIGNED
Resolution|RAWHIDE |---
--- Comment #8 from Petr Pisar <ppisar(a)redhat.com> ---
There another race. If t/05-killolddelay-echod.pl is not fast enough with
writing received signal name (e.g. by adding "sleep 2;" in the beginning of a
signal handler), the t/06-autorestart.t tests fails:
t/05-killolddelay.t .. ok
start_server (pid:17417) starting now...
starting new worker 17418
autorestart triggered (interval=4)
spawning a new worker (num_old_workers=0)
starting new worker 17421
new worker is now running, sending TERM to old workers:17418
sleep 1 secs
killing old workers
old worker 17418 died, status:0
autorestart triggered (interval=4)
spawning a new worker (num_old_workers=0)
starting new worker 17431
# Server changed from <1:17418
# > to <2:17421
# >
# Failed test 'signal sent on hup'
# at t/06-autorestart.t line 63.
# got: ''
# expected: 'TERM'
new worker is now running, sending TERM to old workers:17421
--
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug
https://bugzilla.redhat.com/token.cgi?t=ksJRxjQRsI&a=cc_unsubscribe
2:40 a.m.
New subject: [Bug 1100158] perl-Server-Starter-0.17 FTBFS sometimes because of races in tests
https://bugzilla.redhat.com/show_bug.cgi?id=1100158
Petr Pisar <ppisar(a)redhat.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |CLOSED
Fixed In Version|perl-Server-Starter-0.17-6. |perl-Server-Starter-0.17-8.
|fc22 |fc22
Resolution|--- |RAWHIDE
Last Closed|2014-08-21 08:32:08 |2014-09-09 03:40:55
--- Comment #9 from Petr Pisar <ppisar(a)redhat.com> ---
Fixed with perl-Server-Starter-0.17-8.fc22.
--
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug
https://bugzilla.redhat.com/token.cgi?t=KZflVWnMxn&a=cc_unsubscribe
3550
days inactive
3660
days old
perl-devel@lists.fedoraproject.org
11 comments
1 participants
participants (1)
-
Red Hat Bugzilla